Upload folder using huggingface_hub
Browse filesThis view is limited to 50 files because it contains too many changes.
See raw diff
- .gitattributes +21 -0
- checkpoint-19200/README.md +202 -0
- checkpoint-19200/adapter_config.json +34 -0
- checkpoint-19200/adapter_model.safetensors +3 -0
- checkpoint-19200/rng_state.pth +3 -0
- checkpoint-19200/scheduler.pt +3 -0
- checkpoint-19200/special_tokens_map.json +34 -0
- checkpoint-19200/tokenizer.json +3 -0
- checkpoint-19200/tokenizer.model +3 -0
- checkpoint-19200/tokenizer_config.json +71 -0
- checkpoint-19200/trainer_state.json +1941 -0
- checkpoint-19200/training_args.bin +3 -0
- checkpoint-20000/README.md +202 -0
- checkpoint-20000/adapter_config.json +34 -0
- checkpoint-20000/adapter_model.safetensors +3 -0
- checkpoint-20000/rng_state.pth +3 -0
- checkpoint-20000/scheduler.pt +3 -0
- checkpoint-20000/special_tokens_map.json +34 -0
- checkpoint-20000/tokenizer.json +3 -0
- checkpoint-20000/tokenizer.model +3 -0
- checkpoint-20000/tokenizer_config.json +71 -0
- checkpoint-20000/trainer_state.json +2021 -0
- checkpoint-20000/training_args.bin +3 -0
- checkpoint-20800/README.md +202 -0
- checkpoint-20800/adapter_config.json +34 -0
- checkpoint-20800/adapter_model.safetensors +3 -0
- checkpoint-20800/rng_state.pth +3 -0
- checkpoint-20800/scheduler.pt +3 -0
- checkpoint-20800/special_tokens_map.json +34 -0
- checkpoint-20800/tokenizer.json +3 -0
- checkpoint-20800/tokenizer.model +3 -0
- checkpoint-20800/tokenizer_config.json +71 -0
- checkpoint-20800/trainer_state.json +2101 -0
- checkpoint-20800/training_args.bin +3 -0
- checkpoint-21600/README.md +202 -0
- checkpoint-21600/adapter_config.json +34 -0
- checkpoint-21600/adapter_model.safetensors +3 -0
- checkpoint-21600/rng_state.pth +3 -0
- checkpoint-21600/scheduler.pt +3 -0
- checkpoint-21600/special_tokens_map.json +34 -0
- checkpoint-21600/tokenizer.json +3 -0
- checkpoint-21600/tokenizer.model +3 -0
- checkpoint-21600/tokenizer_config.json +71 -0
- checkpoint-21600/trainer_state.json +2181 -0
- checkpoint-21600/training_args.bin +3 -0
- checkpoint-22400/README.md +202 -0
- checkpoint-22400/adapter_config.json +34 -0
- checkpoint-22400/adapter_model.safetensors +3 -0
- checkpoint-22400/rng_state.pth +3 -0
- checkpoint-22400/scheduler.pt +3 -0
.gitattributes
CHANGED
@@ -33,3 +33,24 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
|
|
33 |
*.zip filter=lfs diff=lfs merge=lfs -text
|
34 |
*.zst filter=lfs diff=lfs merge=lfs -text
|
35 |
*tfevents* filter=lfs diff=lfs merge=lfs -text
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
33 |
*.zip filter=lfs diff=lfs merge=lfs -text
|
34 |
*.zst filter=lfs diff=lfs merge=lfs -text
|
35 |
*tfevents* filter=lfs diff=lfs merge=lfs -text
|
36 |
+
checkpoint-19200/tokenizer.json filter=lfs diff=lfs merge=lfs -text
|
37 |
+
checkpoint-20000/tokenizer.json filter=lfs diff=lfs merge=lfs -text
|
38 |
+
checkpoint-20800/tokenizer.json filter=lfs diff=lfs merge=lfs -text
|
39 |
+
checkpoint-21600/tokenizer.json filter=lfs diff=lfs merge=lfs -text
|
40 |
+
checkpoint-22400/tokenizer.json filter=lfs diff=lfs merge=lfs -text
|
41 |
+
checkpoint-23200/tokenizer.json filter=lfs diff=lfs merge=lfs -text
|
42 |
+
checkpoint-24000/tokenizer.json filter=lfs diff=lfs merge=lfs -text
|
43 |
+
checkpoint-24800/tokenizer.json filter=lfs diff=lfs merge=lfs -text
|
44 |
+
checkpoint-25600/tokenizer.json filter=lfs diff=lfs merge=lfs -text
|
45 |
+
checkpoint-26400/tokenizer.json filter=lfs diff=lfs merge=lfs -text
|
46 |
+
checkpoint-27200/tokenizer.json filter=lfs diff=lfs merge=lfs -text
|
47 |
+
checkpoint-28000/tokenizer.json filter=lfs diff=lfs merge=lfs -text
|
48 |
+
checkpoint-28800/tokenizer.json filter=lfs diff=lfs merge=lfs -text
|
49 |
+
checkpoint-29600/tokenizer.json filter=lfs diff=lfs merge=lfs -text
|
50 |
+
checkpoint-30400/tokenizer.json filter=lfs diff=lfs merge=lfs -text
|
51 |
+
checkpoint-31200/tokenizer.json filter=lfs diff=lfs merge=lfs -text
|
52 |
+
checkpoint-32000/tokenizer.json filter=lfs diff=lfs merge=lfs -text
|
53 |
+
checkpoint-32800/tokenizer.json filter=lfs diff=lfs merge=lfs -text
|
54 |
+
checkpoint-33600/tokenizer.json filter=lfs diff=lfs merge=lfs -text
|
55 |
+
checkpoint-34400/tokenizer.json filter=lfs diff=lfs merge=lfs -text
|
56 |
+
checkpoint-epoch-4/tokenizer.json filter=lfs diff=lfs merge=lfs -text
|
checkpoint-19200/README.md
ADDED
@@ -0,0 +1,202 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
---
|
2 |
+
library_name: peft
|
3 |
+
base_model: unsloth/gemma-2b-bnb-4bit
|
4 |
+
---
|
5 |
+
|
6 |
+
# Model Card for Model ID
|
7 |
+
|
8 |
+
<!-- Provide a quick summary of what the model is/does. -->
|
9 |
+
|
10 |
+
|
11 |
+
|
12 |
+
## Model Details
|
13 |
+
|
14 |
+
### Model Description
|
15 |
+
|
16 |
+
<!-- Provide a longer summary of what this model is. -->
|
17 |
+
|
18 |
+
|
19 |
+
|
20 |
+
- **Developed by:** [More Information Needed]
|
21 |
+
- **Funded by [optional]:** [More Information Needed]
|
22 |
+
- **Shared by [optional]:** [More Information Needed]
|
23 |
+
- **Model type:** [More Information Needed]
|
24 |
+
- **Language(s) (NLP):** [More Information Needed]
|
25 |
+
- **License:** [More Information Needed]
|
26 |
+
- **Finetuned from model [optional]:** [More Information Needed]
|
27 |
+
|
28 |
+
### Model Sources [optional]
|
29 |
+
|
30 |
+
<!-- Provide the basic links for the model. -->
|
31 |
+
|
32 |
+
- **Repository:** [More Information Needed]
|
33 |
+
- **Paper [optional]:** [More Information Needed]
|
34 |
+
- **Demo [optional]:** [More Information Needed]
|
35 |
+
|
36 |
+
## Uses
|
37 |
+
|
38 |
+
<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
|
39 |
+
|
40 |
+
### Direct Use
|
41 |
+
|
42 |
+
<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
|
43 |
+
|
44 |
+
[More Information Needed]
|
45 |
+
|
46 |
+
### Downstream Use [optional]
|
47 |
+
|
48 |
+
<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
|
49 |
+
|
50 |
+
[More Information Needed]
|
51 |
+
|
52 |
+
### Out-of-Scope Use
|
53 |
+
|
54 |
+
<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
|
55 |
+
|
56 |
+
[More Information Needed]
|
57 |
+
|
58 |
+
## Bias, Risks, and Limitations
|
59 |
+
|
60 |
+
<!-- This section is meant to convey both technical and sociotechnical limitations. -->
|
61 |
+
|
62 |
+
[More Information Needed]
|
63 |
+
|
64 |
+
### Recommendations
|
65 |
+
|
66 |
+
<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
|
67 |
+
|
68 |
+
Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
|
69 |
+
|
70 |
+
## How to Get Started with the Model
|
71 |
+
|
72 |
+
Use the code below to get started with the model.
|
73 |
+
|
74 |
+
[More Information Needed]
|
75 |
+
|
76 |
+
## Training Details
|
77 |
+
|
78 |
+
### Training Data
|
79 |
+
|
80 |
+
<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
|
81 |
+
|
82 |
+
[More Information Needed]
|
83 |
+
|
84 |
+
### Training Procedure
|
85 |
+
|
86 |
+
<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
|
87 |
+
|
88 |
+
#### Preprocessing [optional]
|
89 |
+
|
90 |
+
[More Information Needed]
|
91 |
+
|
92 |
+
|
93 |
+
#### Training Hyperparameters
|
94 |
+
|
95 |
+
- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
|
96 |
+
|
97 |
+
#### Speeds, Sizes, Times [optional]
|
98 |
+
|
99 |
+
<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
|
100 |
+
|
101 |
+
[More Information Needed]
|
102 |
+
|
103 |
+
## Evaluation
|
104 |
+
|
105 |
+
<!-- This section describes the evaluation protocols and provides the results. -->
|
106 |
+
|
107 |
+
### Testing Data, Factors & Metrics
|
108 |
+
|
109 |
+
#### Testing Data
|
110 |
+
|
111 |
+
<!-- This should link to a Dataset Card if possible. -->
|
112 |
+
|
113 |
+
[More Information Needed]
|
114 |
+
|
115 |
+
#### Factors
|
116 |
+
|
117 |
+
<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
|
118 |
+
|
119 |
+
[More Information Needed]
|
120 |
+
|
121 |
+
#### Metrics
|
122 |
+
|
123 |
+
<!-- These are the evaluation metrics being used, ideally with a description of why. -->
|
124 |
+
|
125 |
+
[More Information Needed]
|
126 |
+
|
127 |
+
### Results
|
128 |
+
|
129 |
+
[More Information Needed]
|
130 |
+
|
131 |
+
#### Summary
|
132 |
+
|
133 |
+
|
134 |
+
|
135 |
+
## Model Examination [optional]
|
136 |
+
|
137 |
+
<!-- Relevant interpretability work for the model goes here -->
|
138 |
+
|
139 |
+
[More Information Needed]
|
140 |
+
|
141 |
+
## Environmental Impact
|
142 |
+
|
143 |
+
<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
|
144 |
+
|
145 |
+
Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
|
146 |
+
|
147 |
+
- **Hardware Type:** [More Information Needed]
|
148 |
+
- **Hours used:** [More Information Needed]
|
149 |
+
- **Cloud Provider:** [More Information Needed]
|
150 |
+
- **Compute Region:** [More Information Needed]
|
151 |
+
- **Carbon Emitted:** [More Information Needed]
|
152 |
+
|
153 |
+
## Technical Specifications [optional]
|
154 |
+
|
155 |
+
### Model Architecture and Objective
|
156 |
+
|
157 |
+
[More Information Needed]
|
158 |
+
|
159 |
+
### Compute Infrastructure
|
160 |
+
|
161 |
+
[More Information Needed]
|
162 |
+
|
163 |
+
#### Hardware
|
164 |
+
|
165 |
+
[More Information Needed]
|
166 |
+
|
167 |
+
#### Software
|
168 |
+
|
169 |
+
[More Information Needed]
|
170 |
+
|
171 |
+
## Citation [optional]
|
172 |
+
|
173 |
+
<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
|
174 |
+
|
175 |
+
**BibTeX:**
|
176 |
+
|
177 |
+
[More Information Needed]
|
178 |
+
|
179 |
+
**APA:**
|
180 |
+
|
181 |
+
[More Information Needed]
|
182 |
+
|
183 |
+
## Glossary [optional]
|
184 |
+
|
185 |
+
<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
|
186 |
+
|
187 |
+
[More Information Needed]
|
188 |
+
|
189 |
+
## More Information [optional]
|
190 |
+
|
191 |
+
[More Information Needed]
|
192 |
+
|
193 |
+
## Model Card Authors [optional]
|
194 |
+
|
195 |
+
[More Information Needed]
|
196 |
+
|
197 |
+
## Model Card Contact
|
198 |
+
|
199 |
+
[More Information Needed]
|
200 |
+
### Framework versions
|
201 |
+
|
202 |
+
- PEFT 0.10.0
|
checkpoint-19200/adapter_config.json
ADDED
@@ -0,0 +1,34 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"alpha_pattern": {},
|
3 |
+
"auto_mapping": null,
|
4 |
+
"base_model_name_or_path": "unsloth/gemma-2b-bnb-4bit",
|
5 |
+
"bias": "none",
|
6 |
+
"fan_in_fan_out": false,
|
7 |
+
"inference_mode": true,
|
8 |
+
"init_lora_weights": true,
|
9 |
+
"layer_replication": null,
|
10 |
+
"layers_pattern": null,
|
11 |
+
"layers_to_transform": null,
|
12 |
+
"loftq_config": {},
|
13 |
+
"lora_alpha": 32,
|
14 |
+
"lora_dropout": 0,
|
15 |
+
"megatron_config": null,
|
16 |
+
"megatron_core": "megatron.core",
|
17 |
+
"modules_to_save": null,
|
18 |
+
"peft_type": "LORA",
|
19 |
+
"r": 64,
|
20 |
+
"rank_pattern": {},
|
21 |
+
"revision": "unsloth",
|
22 |
+
"target_modules": [
|
23 |
+
"v_proj",
|
24 |
+
"up_proj",
|
25 |
+
"gate_proj",
|
26 |
+
"o_proj",
|
27 |
+
"k_proj",
|
28 |
+
"down_proj",
|
29 |
+
"q_proj"
|
30 |
+
],
|
31 |
+
"task_type": "CAUSAL_LM",
|
32 |
+
"use_dora": false,
|
33 |
+
"use_rslora": false
|
34 |
+
}
|
checkpoint-19200/adapter_model.safetensors
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:285fd80afeed0c673cd24ce99ab039194c8fa120f6a1db2001e02bf9571beb6d
|
3 |
+
size 313820248
|
checkpoint-19200/rng_state.pth
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:7e9c79e040cd1d90c9235f5ea004ce98bb1df30785570b9fdcf2a9e29948f896
|
3 |
+
size 14244
|
checkpoint-19200/scheduler.pt
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:a3348413c5565e84f72b5de03d822ed1866796c840ac5265b9b52f983f212aba
|
3 |
+
size 1064
|
checkpoint-19200/special_tokens_map.json
ADDED
@@ -0,0 +1,34 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"additional_special_tokens": [
|
3 |
+
"<start_of_turn>",
|
4 |
+
"<end_of_turn>"
|
5 |
+
],
|
6 |
+
"bos_token": {
|
7 |
+
"content": "<bos>",
|
8 |
+
"lstrip": false,
|
9 |
+
"normalized": false,
|
10 |
+
"rstrip": false,
|
11 |
+
"single_word": false
|
12 |
+
},
|
13 |
+
"eos_token": {
|
14 |
+
"content": "<eos>",
|
15 |
+
"lstrip": false,
|
16 |
+
"normalized": false,
|
17 |
+
"rstrip": false,
|
18 |
+
"single_word": false
|
19 |
+
},
|
20 |
+
"pad_token": {
|
21 |
+
"content": "<pad>",
|
22 |
+
"lstrip": false,
|
23 |
+
"normalized": false,
|
24 |
+
"rstrip": false,
|
25 |
+
"single_word": false
|
26 |
+
},
|
27 |
+
"unk_token": {
|
28 |
+
"content": "<unk>",
|
29 |
+
"lstrip": false,
|
30 |
+
"normalized": false,
|
31 |
+
"rstrip": false,
|
32 |
+
"single_word": false
|
33 |
+
}
|
34 |
+
}
|
checkpoint-19200/tokenizer.json
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:05e97791a5e007260de1db7e1692e53150e08cea481e2bf25435553380c147ee
|
3 |
+
size 17477929
|
checkpoint-19200/tokenizer.model
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:61a7b147390c64585d6c3543dd6fc636906c9af3865a5548f27f31aee1d4c8e2
|
3 |
+
size 4241003
|
checkpoint-19200/tokenizer_config.json
ADDED
@@ -0,0 +1,71 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"add_bos_token": true,
|
3 |
+
"add_eos_token": false,
|
4 |
+
"added_tokens_decoder": {
|
5 |
+
"0": {
|
6 |
+
"content": "<pad>",
|
7 |
+
"lstrip": false,
|
8 |
+
"normalized": false,
|
9 |
+
"rstrip": false,
|
10 |
+
"single_word": false,
|
11 |
+
"special": true
|
12 |
+
},
|
13 |
+
"1": {
|
14 |
+
"content": "<eos>",
|
15 |
+
"lstrip": false,
|
16 |
+
"normalized": false,
|
17 |
+
"rstrip": false,
|
18 |
+
"single_word": false,
|
19 |
+
"special": true
|
20 |
+
},
|
21 |
+
"2": {
|
22 |
+
"content": "<bos>",
|
23 |
+
"lstrip": false,
|
24 |
+
"normalized": false,
|
25 |
+
"rstrip": false,
|
26 |
+
"single_word": false,
|
27 |
+
"special": true
|
28 |
+
},
|
29 |
+
"3": {
|
30 |
+
"content": "<unk>",
|
31 |
+
"lstrip": false,
|
32 |
+
"normalized": false,
|
33 |
+
"rstrip": false,
|
34 |
+
"single_word": false,
|
35 |
+
"special": true
|
36 |
+
},
|
37 |
+
"106": {
|
38 |
+
"content": "<start_of_turn>",
|
39 |
+
"lstrip": false,
|
40 |
+
"normalized": false,
|
41 |
+
"rstrip": false,
|
42 |
+
"single_word": false,
|
43 |
+
"special": true
|
44 |
+
},
|
45 |
+
"107": {
|
46 |
+
"content": "<end_of_turn>",
|
47 |
+
"lstrip": false,
|
48 |
+
"normalized": false,
|
49 |
+
"rstrip": false,
|
50 |
+
"single_word": false,
|
51 |
+
"special": true
|
52 |
+
}
|
53 |
+
},
|
54 |
+
"additional_special_tokens": [
|
55 |
+
"<start_of_turn>",
|
56 |
+
"<end_of_turn>"
|
57 |
+
],
|
58 |
+
"bos_token": "<bos>",
|
59 |
+
"chat_template": "{{ bos_token }}{% if messages[0]['role'] == 'system' %}{{ raise_exception('System role not supported') }}{% endif %}{% for message in messages %}{% if (message['role'] == 'user') != (loop.index0 % 2 == 0) %}{{ raise_exception('Conversation roles must alternate user/assistant/user/assistant/...') }}{% endif %}{% if (message['role'] == 'assistant') %}{% set role = 'model' %}{% else %}{% set role = message['role'] %}{% endif %}{{ '<start_of_turn>' + role + '\n' + message['content'] | trim + '<end_of_turn>\n' }}{% endfor %}{% if add_generation_prompt %}{{'<start_of_turn>model\n'}}{% endif %}",
|
60 |
+
"clean_up_tokenization_spaces": false,
|
61 |
+
"eos_token": "<eos>",
|
62 |
+
"legacy": null,
|
63 |
+
"model_max_length": 8192,
|
64 |
+
"pad_token": "<pad>",
|
65 |
+
"padding_side": "right",
|
66 |
+
"sp_model_kwargs": {},
|
67 |
+
"spaces_between_special_tokens": false,
|
68 |
+
"tokenizer_class": "GemmaTokenizer",
|
69 |
+
"unk_token": "<unk>",
|
70 |
+
"use_default_system_prompt": false
|
71 |
+
}
|
checkpoint-19200/trainer_state.json
ADDED
@@ -0,0 +1,1941 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"best_metric": null,
|
3 |
+
"best_model_checkpoint": null,
|
4 |
+
"epoch": 2.187161815799966,
|
5 |
+
"eval_steps": 200,
|
6 |
+
"global_step": 19200,
|
7 |
+
"is_hyper_param_search": false,
|
8 |
+
"is_local_process_zero": true,
|
9 |
+
"is_world_process_zero": true,
|
10 |
+
"log_history": [
|
11 |
+
{
|
12 |
+
"epoch": 0.02,
|
13 |
+
"grad_norm": 0.1643640249967575,
|
14 |
+
"learning_rate": 0.0001988891104338166,
|
15 |
+
"loss": 1.7673,
|
16 |
+
"step": 200
|
17 |
+
},
|
18 |
+
{
|
19 |
+
"epoch": 0.02,
|
20 |
+
"eval_bertscore": 0.7312520742416382,
|
21 |
+
"eval_loss": 1.7944419384002686,
|
22 |
+
"eval_rouge1": 0.645726048132668,
|
23 |
+
"eval_rouge2": 0.342840307585653,
|
24 |
+
"eval_rougeL": 0.5174784271125388,
|
25 |
+
"eval_rougeLsum": 0.6359911842715976,
|
26 |
+
"eval_runtime": 67.7968,
|
27 |
+
"eval_samples_per_second": 0.147,
|
28 |
+
"eval_steps_per_second": 0.074,
|
29 |
+
"step": 200
|
30 |
+
},
|
31 |
+
{
|
32 |
+
"epoch": 0.05,
|
33 |
+
"grad_norm": 0.17238478362560272,
|
34 |
+
"learning_rate": 0.00019774973651978238,
|
35 |
+
"loss": 1.6985,
|
36 |
+
"step": 400
|
37 |
+
},
|
38 |
+
{
|
39 |
+
"epoch": 0.05,
|
40 |
+
"eval_bertscore": 0.733666718006134,
|
41 |
+
"eval_loss": 1.7791178226470947,
|
42 |
+
"eval_rouge1": 0.6540909153028596,
|
43 |
+
"eval_rouge2": 0.3548819059818129,
|
44 |
+
"eval_rougeL": 0.527257232694246,
|
45 |
+
"eval_rougeLsum": 0.6442799950994005,
|
46 |
+
"eval_runtime": 15.1267,
|
47 |
+
"eval_samples_per_second": 0.661,
|
48 |
+
"eval_steps_per_second": 0.331,
|
49 |
+
"step": 400
|
50 |
+
},
|
51 |
+
{
|
52 |
+
"epoch": 0.07,
|
53 |
+
"grad_norm": 0.19368696212768555,
|
54 |
+
"learning_rate": 0.00019661036260574814,
|
55 |
+
"loss": 1.6962,
|
56 |
+
"step": 600
|
57 |
+
},
|
58 |
+
{
|
59 |
+
"epoch": 0.07,
|
60 |
+
"eval_bertscore": 0.7339462041854858,
|
61 |
+
"eval_loss": 1.7609882354736328,
|
62 |
+
"eval_rouge1": 0.6384337329686338,
|
63 |
+
"eval_rouge2": 0.3415514270662107,
|
64 |
+
"eval_rougeL": 0.51206080148464,
|
65 |
+
"eval_rougeLsum": 0.6261968614666548,
|
66 |
+
"eval_runtime": 15.2068,
|
67 |
+
"eval_samples_per_second": 0.658,
|
68 |
+
"eval_steps_per_second": 0.329,
|
69 |
+
"step": 600
|
70 |
+
},
|
71 |
+
{
|
72 |
+
"epoch": 0.09,
|
73 |
+
"grad_norm": 0.18629203736782074,
|
74 |
+
"learning_rate": 0.00019547098869171392,
|
75 |
+
"loss": 1.6825,
|
76 |
+
"step": 800
|
77 |
+
},
|
78 |
+
{
|
79 |
+
"epoch": 0.09,
|
80 |
+
"eval_bertscore": 0.7363594174385071,
|
81 |
+
"eval_loss": 1.7610784769058228,
|
82 |
+
"eval_rouge1": 0.6461624591922237,
|
83 |
+
"eval_rouge2": 0.3477371388439609,
|
84 |
+
"eval_rougeL": 0.5187429174752844,
|
85 |
+
"eval_rougeLsum": 0.6361089823008282,
|
86 |
+
"eval_runtime": 15.173,
|
87 |
+
"eval_samples_per_second": 0.659,
|
88 |
+
"eval_steps_per_second": 0.33,
|
89 |
+
"step": 800
|
90 |
+
},
|
91 |
+
{
|
92 |
+
"epoch": 0.11,
|
93 |
+
"grad_norm": 0.1799013316631317,
|
94 |
+
"learning_rate": 0.00019433161477767967,
|
95 |
+
"loss": 1.6848,
|
96 |
+
"step": 1000
|
97 |
+
},
|
98 |
+
{
|
99 |
+
"epoch": 0.11,
|
100 |
+
"eval_bertscore": 0.7334067225456238,
|
101 |
+
"eval_loss": 1.7576347589492798,
|
102 |
+
"eval_rouge1": 0.6345119236349537,
|
103 |
+
"eval_rouge2": 0.3422519149071803,
|
104 |
+
"eval_rougeL": 0.5111983101326238,
|
105 |
+
"eval_rougeLsum": 0.6244653120436832,
|
106 |
+
"eval_runtime": 15.2847,
|
107 |
+
"eval_samples_per_second": 0.654,
|
108 |
+
"eval_steps_per_second": 0.327,
|
109 |
+
"step": 1000
|
110 |
+
},
|
111 |
+
{
|
112 |
+
"epoch": 0.14,
|
113 |
+
"grad_norm": 0.22036150097846985,
|
114 |
+
"learning_rate": 0.00019319224086364545,
|
115 |
+
"loss": 1.6714,
|
116 |
+
"step": 1200
|
117 |
+
},
|
118 |
+
{
|
119 |
+
"epoch": 0.14,
|
120 |
+
"eval_bertscore": 0.7323788404464722,
|
121 |
+
"eval_loss": 1.7521806955337524,
|
122 |
+
"eval_rouge1": 0.6452540184557478,
|
123 |
+
"eval_rouge2": 0.3465145726476423,
|
124 |
+
"eval_rougeL": 0.516711757588783,
|
125 |
+
"eval_rougeLsum": 0.6341049885677059,
|
126 |
+
"eval_runtime": 15.1247,
|
127 |
+
"eval_samples_per_second": 0.661,
|
128 |
+
"eval_steps_per_second": 0.331,
|
129 |
+
"step": 1200
|
130 |
+
},
|
131 |
+
{
|
132 |
+
"epoch": 0.16,
|
133 |
+
"grad_norm": 0.21381086111068726,
|
134 |
+
"learning_rate": 0.0001920528669496112,
|
135 |
+
"loss": 1.6669,
|
136 |
+
"step": 1400
|
137 |
+
},
|
138 |
+
{
|
139 |
+
"epoch": 0.16,
|
140 |
+
"eval_bertscore": 0.7313202619552612,
|
141 |
+
"eval_loss": 1.7520482540130615,
|
142 |
+
"eval_rouge1": 0.6397526546254797,
|
143 |
+
"eval_rouge2": 0.3452671288110514,
|
144 |
+
"eval_rougeL": 0.5176580626678706,
|
145 |
+
"eval_rougeLsum": 0.6296746647539768,
|
146 |
+
"eval_runtime": 15.183,
|
147 |
+
"eval_samples_per_second": 0.659,
|
148 |
+
"eval_steps_per_second": 0.329,
|
149 |
+
"step": 1400
|
150 |
+
},
|
151 |
+
{
|
152 |
+
"epoch": 0.18,
|
153 |
+
"grad_norm": 0.20332874357700348,
|
154 |
+
"learning_rate": 0.00019091349303557696,
|
155 |
+
"loss": 1.671,
|
156 |
+
"step": 1600
|
157 |
+
},
|
158 |
+
{
|
159 |
+
"epoch": 0.18,
|
160 |
+
"eval_bertscore": 0.7349230647087097,
|
161 |
+
"eval_loss": 1.7473630905151367,
|
162 |
+
"eval_rouge1": 0.637439872504459,
|
163 |
+
"eval_rouge2": 0.34307164454056094,
|
164 |
+
"eval_rougeL": 0.5129717676228565,
|
165 |
+
"eval_rougeLsum": 0.6272190896182391,
|
166 |
+
"eval_runtime": 15.5672,
|
167 |
+
"eval_samples_per_second": 0.642,
|
168 |
+
"eval_steps_per_second": 0.321,
|
169 |
+
"step": 1600
|
170 |
+
},
|
171 |
+
{
|
172 |
+
"epoch": 0.21,
|
173 |
+
"grad_norm": 0.2025599479675293,
|
174 |
+
"learning_rate": 0.00018977411912154274,
|
175 |
+
"loss": 1.6721,
|
176 |
+
"step": 1800
|
177 |
+
},
|
178 |
+
{
|
179 |
+
"epoch": 0.21,
|
180 |
+
"eval_bertscore": 0.7357184290885925,
|
181 |
+
"eval_loss": 1.7516342401504517,
|
182 |
+
"eval_rouge1": 0.6387615819926658,
|
183 |
+
"eval_rouge2": 0.34366787517105574,
|
184 |
+
"eval_rougeL": 0.5129026911770751,
|
185 |
+
"eval_rougeLsum": 0.6289314118258257,
|
186 |
+
"eval_runtime": 15.9574,
|
187 |
+
"eval_samples_per_second": 0.627,
|
188 |
+
"eval_steps_per_second": 0.313,
|
189 |
+
"step": 1800
|
190 |
+
},
|
191 |
+
{
|
192 |
+
"epoch": 0.23,
|
193 |
+
"grad_norm": 0.20457112789154053,
|
194 |
+
"learning_rate": 0.0001886347452075085,
|
195 |
+
"loss": 1.671,
|
196 |
+
"step": 2000
|
197 |
+
},
|
198 |
+
{
|
199 |
+
"epoch": 0.23,
|
200 |
+
"eval_bertscore": 0.733718752861023,
|
201 |
+
"eval_loss": 1.7501707077026367,
|
202 |
+
"eval_rouge1": 0.6346207681220664,
|
203 |
+
"eval_rouge2": 0.33748369437614106,
|
204 |
+
"eval_rougeL": 0.5085159047705141,
|
205 |
+
"eval_rougeLsum": 0.6239953154441167,
|
206 |
+
"eval_runtime": 15.0863,
|
207 |
+
"eval_samples_per_second": 0.663,
|
208 |
+
"eval_steps_per_second": 0.331,
|
209 |
+
"step": 2000
|
210 |
+
},
|
211 |
+
{
|
212 |
+
"epoch": 0.25,
|
213 |
+
"grad_norm": 0.22552740573883057,
|
214 |
+
"learning_rate": 0.00018749537129347424,
|
215 |
+
"loss": 1.6496,
|
216 |
+
"step": 2200
|
217 |
+
},
|
218 |
+
{
|
219 |
+
"epoch": 0.25,
|
220 |
+
"eval_bertscore": 0.7368552684783936,
|
221 |
+
"eval_loss": 1.7437107563018799,
|
222 |
+
"eval_rouge1": 0.6490756387878311,
|
223 |
+
"eval_rouge2": 0.3448817738175175,
|
224 |
+
"eval_rougeL": 0.5235187045706321,
|
225 |
+
"eval_rougeLsum": 0.6377780857890332,
|
226 |
+
"eval_runtime": 15.07,
|
227 |
+
"eval_samples_per_second": 0.664,
|
228 |
+
"eval_steps_per_second": 0.332,
|
229 |
+
"step": 2200
|
230 |
+
},
|
231 |
+
{
|
232 |
+
"epoch": 0.27,
|
233 |
+
"grad_norm": 0.22573673725128174,
|
234 |
+
"learning_rate": 0.00018635599737944,
|
235 |
+
"loss": 1.6629,
|
236 |
+
"step": 2400
|
237 |
+
},
|
238 |
+
{
|
239 |
+
"epoch": 0.27,
|
240 |
+
"eval_bertscore": 0.7314499616622925,
|
241 |
+
"eval_loss": 1.7462828159332275,
|
242 |
+
"eval_rouge1": 0.6511482369678803,
|
243 |
+
"eval_rouge2": 0.34632544827771805,
|
244 |
+
"eval_rougeL": 0.5212417191003778,
|
245 |
+
"eval_rougeLsum": 0.6415391907940229,
|
246 |
+
"eval_runtime": 15.2078,
|
247 |
+
"eval_samples_per_second": 0.658,
|
248 |
+
"eval_steps_per_second": 0.329,
|
249 |
+
"step": 2400
|
250 |
+
},
|
251 |
+
{
|
252 |
+
"epoch": 0.3,
|
253 |
+
"grad_norm": 0.26426687836647034,
|
254 |
+
"learning_rate": 0.00018521662346540575,
|
255 |
+
"loss": 1.6644,
|
256 |
+
"step": 2600
|
257 |
+
},
|
258 |
+
{
|
259 |
+
"epoch": 0.3,
|
260 |
+
"eval_bertscore": 0.7363359928131104,
|
261 |
+
"eval_loss": 1.7505037784576416,
|
262 |
+
"eval_rouge1": 0.6498296552335481,
|
263 |
+
"eval_rouge2": 0.34873833589761183,
|
264 |
+
"eval_rougeL": 0.5194028620820592,
|
265 |
+
"eval_rougeLsum": 0.6404603087578984,
|
266 |
+
"eval_runtime": 14.8403,
|
267 |
+
"eval_samples_per_second": 0.674,
|
268 |
+
"eval_steps_per_second": 0.337,
|
269 |
+
"step": 2600
|
270 |
+
},
|
271 |
+
{
|
272 |
+
"epoch": 0.32,
|
273 |
+
"grad_norm": 0.20142091810703278,
|
274 |
+
"learning_rate": 0.00018407724955137153,
|
275 |
+
"loss": 1.6535,
|
276 |
+
"step": 2800
|
277 |
+
},
|
278 |
+
{
|
279 |
+
"epoch": 0.32,
|
280 |
+
"eval_bertscore": 0.7304679155349731,
|
281 |
+
"eval_loss": 1.7511039972305298,
|
282 |
+
"eval_rouge1": 0.6475130585388738,
|
283 |
+
"eval_rouge2": 0.34648331046897884,
|
284 |
+
"eval_rougeL": 0.5218042284020985,
|
285 |
+
"eval_rougeLsum": 0.6382749834402862,
|
286 |
+
"eval_runtime": 15.0162,
|
287 |
+
"eval_samples_per_second": 0.666,
|
288 |
+
"eval_steps_per_second": 0.333,
|
289 |
+
"step": 2800
|
290 |
+
},
|
291 |
+
{
|
292 |
+
"epoch": 0.34,
|
293 |
+
"grad_norm": 0.23283220827579498,
|
294 |
+
"learning_rate": 0.00018293787563733728,
|
295 |
+
"loss": 1.6477,
|
296 |
+
"step": 3000
|
297 |
+
},
|
298 |
+
{
|
299 |
+
"epoch": 0.34,
|
300 |
+
"eval_bertscore": 0.7327049374580383,
|
301 |
+
"eval_loss": 1.7461665868759155,
|
302 |
+
"eval_rouge1": 0.6309349586871908,
|
303 |
+
"eval_rouge2": 0.3387882478990309,
|
304 |
+
"eval_rougeL": 0.5042059192403674,
|
305 |
+
"eval_rougeLsum": 0.6210432469674847,
|
306 |
+
"eval_runtime": 15.463,
|
307 |
+
"eval_samples_per_second": 0.647,
|
308 |
+
"eval_steps_per_second": 0.323,
|
309 |
+
"step": 3000
|
310 |
+
},
|
311 |
+
{
|
312 |
+
"epoch": 0.36,
|
313 |
+
"grad_norm": 0.21316750347614288,
|
314 |
+
"learning_rate": 0.00018179850172330306,
|
315 |
+
"loss": 1.6614,
|
316 |
+
"step": 3200
|
317 |
+
},
|
318 |
+
{
|
319 |
+
"epoch": 0.36,
|
320 |
+
"eval_bertscore": 0.7314620018005371,
|
321 |
+
"eval_loss": 1.7468239068984985,
|
322 |
+
"eval_rouge1": 0.6480904534152265,
|
323 |
+
"eval_rouge2": 0.3479530168963481,
|
324 |
+
"eval_rougeL": 0.5193148273848067,
|
325 |
+
"eval_rougeLsum": 0.6366010767207634,
|
326 |
+
"eval_runtime": 15.0381,
|
327 |
+
"eval_samples_per_second": 0.665,
|
328 |
+
"eval_steps_per_second": 0.332,
|
329 |
+
"step": 3200
|
330 |
+
},
|
331 |
+
{
|
332 |
+
"epoch": 0.39,
|
333 |
+
"grad_norm": 0.26080408692359924,
|
334 |
+
"learning_rate": 0.00018065912780926882,
|
335 |
+
"loss": 1.6591,
|
336 |
+
"step": 3400
|
337 |
+
},
|
338 |
+
{
|
339 |
+
"epoch": 0.39,
|
340 |
+
"eval_bertscore": 0.7327477335929871,
|
341 |
+
"eval_loss": 1.7442594766616821,
|
342 |
+
"eval_rouge1": 0.6424613378037144,
|
343 |
+
"eval_rouge2": 0.34731770322974903,
|
344 |
+
"eval_rougeL": 0.5160705879794565,
|
345 |
+
"eval_rougeLsum": 0.6327006420281607,
|
346 |
+
"eval_runtime": 15.5373,
|
347 |
+
"eval_samples_per_second": 0.644,
|
348 |
+
"eval_steps_per_second": 0.322,
|
349 |
+
"step": 3400
|
350 |
+
},
|
351 |
+
{
|
352 |
+
"epoch": 0.41,
|
353 |
+
"grad_norm": 0.23274216055870056,
|
354 |
+
"learning_rate": 0.0001795197538952346,
|
355 |
+
"loss": 1.6613,
|
356 |
+
"step": 3600
|
357 |
+
},
|
358 |
+
{
|
359 |
+
"epoch": 0.41,
|
360 |
+
"eval_bertscore": 0.736956000328064,
|
361 |
+
"eval_loss": 1.7429373264312744,
|
362 |
+
"eval_rouge1": 0.6514574160666677,
|
363 |
+
"eval_rouge2": 0.3556199242231646,
|
364 |
+
"eval_rougeL": 0.5249726237675663,
|
365 |
+
"eval_rougeLsum": 0.6406261097623661,
|
366 |
+
"eval_runtime": 14.9415,
|
367 |
+
"eval_samples_per_second": 0.669,
|
368 |
+
"eval_steps_per_second": 0.335,
|
369 |
+
"step": 3600
|
370 |
+
},
|
371 |
+
{
|
372 |
+
"epoch": 0.43,
|
373 |
+
"grad_norm": 0.23616766929626465,
|
374 |
+
"learning_rate": 0.00017838037998120035,
|
375 |
+
"loss": 1.6479,
|
376 |
+
"step": 3800
|
377 |
+
},
|
378 |
+
{
|
379 |
+
"epoch": 0.43,
|
380 |
+
"eval_bertscore": 0.7349627614021301,
|
381 |
+
"eval_loss": 1.7420669794082642,
|
382 |
+
"eval_rouge1": 0.655851684949526,
|
383 |
+
"eval_rouge2": 0.35254590691865084,
|
384 |
+
"eval_rougeL": 0.5248980956621441,
|
385 |
+
"eval_rougeLsum": 0.6449637270581419,
|
386 |
+
"eval_runtime": 15.4178,
|
387 |
+
"eval_samples_per_second": 0.649,
|
388 |
+
"eval_steps_per_second": 0.324,
|
389 |
+
"step": 3800
|
390 |
+
},
|
391 |
+
{
|
392 |
+
"epoch": 0.46,
|
393 |
+
"grad_norm": 0.23260319232940674,
|
394 |
+
"learning_rate": 0.0001772410060671661,
|
395 |
+
"loss": 1.6569,
|
396 |
+
"step": 4000
|
397 |
+
},
|
398 |
+
{
|
399 |
+
"epoch": 0.46,
|
400 |
+
"eval_bertscore": 0.7332885265350342,
|
401 |
+
"eval_loss": 1.7401313781738281,
|
402 |
+
"eval_rouge1": 0.6669483634140105,
|
403 |
+
"eval_rouge2": 0.35873988835161297,
|
404 |
+
"eval_rougeL": 0.5343868725007427,
|
405 |
+
"eval_rougeLsum": 0.6555353134690931,
|
406 |
+
"eval_runtime": 15.0822,
|
407 |
+
"eval_samples_per_second": 0.663,
|
408 |
+
"eval_steps_per_second": 0.332,
|
409 |
+
"step": 4000
|
410 |
+
},
|
411 |
+
{
|
412 |
+
"epoch": 0.48,
|
413 |
+
"grad_norm": 0.2366473525762558,
|
414 |
+
"learning_rate": 0.00017610163215313186,
|
415 |
+
"loss": 1.6599,
|
416 |
+
"step": 4200
|
417 |
+
},
|
418 |
+
{
|
419 |
+
"epoch": 0.48,
|
420 |
+
"eval_bertscore": 0.7335314750671387,
|
421 |
+
"eval_loss": 1.7385823726654053,
|
422 |
+
"eval_rouge1": 0.6559297063578133,
|
423 |
+
"eval_rouge2": 0.35483499789990636,
|
424 |
+
"eval_rougeL": 0.5297939800986089,
|
425 |
+
"eval_rougeLsum": 0.6454544372491222,
|
426 |
+
"eval_runtime": 15.1029,
|
427 |
+
"eval_samples_per_second": 0.662,
|
428 |
+
"eval_steps_per_second": 0.331,
|
429 |
+
"step": 4200
|
430 |
+
},
|
431 |
+
{
|
432 |
+
"epoch": 0.5,
|
433 |
+
"grad_norm": 0.20628753304481506,
|
434 |
+
"learning_rate": 0.0001749622582390976,
|
435 |
+
"loss": 1.6454,
|
436 |
+
"step": 4400
|
437 |
+
},
|
438 |
+
{
|
439 |
+
"epoch": 0.5,
|
440 |
+
"eval_bertscore": 0.7342795133590698,
|
441 |
+
"eval_loss": 1.7422058582305908,
|
442 |
+
"eval_rouge1": 0.660746519614568,
|
443 |
+
"eval_rouge2": 0.3633965895561597,
|
444 |
+
"eval_rougeL": 0.5369036980876734,
|
445 |
+
"eval_rougeLsum": 0.650338328328998,
|
446 |
+
"eval_runtime": 15.4434,
|
447 |
+
"eval_samples_per_second": 0.648,
|
448 |
+
"eval_steps_per_second": 0.324,
|
449 |
+
"step": 4400
|
450 |
+
},
|
451 |
+
{
|
452 |
+
"epoch": 0.52,
|
453 |
+
"grad_norm": 0.2239149957895279,
|
454 |
+
"learning_rate": 0.0001738228843250634,
|
455 |
+
"loss": 1.6594,
|
456 |
+
"step": 4600
|
457 |
+
},
|
458 |
+
{
|
459 |
+
"epoch": 0.52,
|
460 |
+
"eval_bertscore": 0.7313543558120728,
|
461 |
+
"eval_loss": 1.740854263305664,
|
462 |
+
"eval_rouge1": 0.6591645132619427,
|
463 |
+
"eval_rouge2": 0.35766117432431743,
|
464 |
+
"eval_rougeL": 0.532710255034635,
|
465 |
+
"eval_rougeLsum": 0.6479428185884644,
|
466 |
+
"eval_runtime": 14.9436,
|
467 |
+
"eval_samples_per_second": 0.669,
|
468 |
+
"eval_steps_per_second": 0.335,
|
469 |
+
"step": 4600
|
470 |
+
},
|
471 |
+
{
|
472 |
+
"epoch": 0.55,
|
473 |
+
"grad_norm": 0.24808338284492493,
|
474 |
+
"learning_rate": 0.00017268351041102914,
|
475 |
+
"loss": 1.6604,
|
476 |
+
"step": 4800
|
477 |
+
},
|
478 |
+
{
|
479 |
+
"epoch": 0.55,
|
480 |
+
"eval_bertscore": 0.7333321571350098,
|
481 |
+
"eval_loss": 1.7385585308074951,
|
482 |
+
"eval_rouge1": 0.6532115808232871,
|
483 |
+
"eval_rouge2": 0.35333788022501567,
|
484 |
+
"eval_rougeL": 0.5284071547874328,
|
485 |
+
"eval_rougeLsum": 0.6410472452797623,
|
486 |
+
"eval_runtime": 15.0277,
|
487 |
+
"eval_samples_per_second": 0.665,
|
488 |
+
"eval_steps_per_second": 0.333,
|
489 |
+
"step": 4800
|
490 |
+
},
|
491 |
+
{
|
492 |
+
"epoch": 0.57,
|
493 |
+
"grad_norm": 0.2555364966392517,
|
494 |
+
"learning_rate": 0.0001715441364969949,
|
495 |
+
"loss": 1.6493,
|
496 |
+
"step": 5000
|
497 |
+
},
|
498 |
+
{
|
499 |
+
"epoch": 0.57,
|
500 |
+
"eval_bertscore": 0.7318152189254761,
|
501 |
+
"eval_loss": 1.7357494831085205,
|
502 |
+
"eval_rouge1": 0.6476755890502586,
|
503 |
+
"eval_rouge2": 0.35312778275949164,
|
504 |
+
"eval_rougeL": 0.5227601228049905,
|
505 |
+
"eval_rougeLsum": 0.6371331138372852,
|
506 |
+
"eval_runtime": 14.8807,
|
507 |
+
"eval_samples_per_second": 0.672,
|
508 |
+
"eval_steps_per_second": 0.336,
|
509 |
+
"step": 5000
|
510 |
+
},
|
511 |
+
{
|
512 |
+
"epoch": 0.59,
|
513 |
+
"grad_norm": 0.21518155932426453,
|
514 |
+
"learning_rate": 0.00017040476258296068,
|
515 |
+
"loss": 1.644,
|
516 |
+
"step": 5200
|
517 |
+
},
|
518 |
+
{
|
519 |
+
"epoch": 0.59,
|
520 |
+
"eval_bertscore": 0.734805703163147,
|
521 |
+
"eval_loss": 1.74032723903656,
|
522 |
+
"eval_rouge1": 0.6476813733451636,
|
523 |
+
"eval_rouge2": 0.3509259728617576,
|
524 |
+
"eval_rougeL": 0.5221334872800274,
|
525 |
+
"eval_rougeLsum": 0.636892384667733,
|
526 |
+
"eval_runtime": 15.1271,
|
527 |
+
"eval_samples_per_second": 0.661,
|
528 |
+
"eval_steps_per_second": 0.331,
|
529 |
+
"step": 5200
|
530 |
+
},
|
531 |
+
{
|
532 |
+
"epoch": 0.62,
|
533 |
+
"grad_norm": 0.26086658239364624,
|
534 |
+
"learning_rate": 0.00016926538866892643,
|
535 |
+
"loss": 1.6449,
|
536 |
+
"step": 5400
|
537 |
+
},
|
538 |
+
{
|
539 |
+
"epoch": 0.62,
|
540 |
+
"eval_bertscore": 0.7339995503425598,
|
541 |
+
"eval_loss": 1.7338205575942993,
|
542 |
+
"eval_rouge1": 0.6416889902864902,
|
543 |
+
"eval_rouge2": 0.3479045880347737,
|
544 |
+
"eval_rougeL": 0.5160577838468976,
|
545 |
+
"eval_rougeLsum": 0.6317983411796093,
|
546 |
+
"eval_runtime": 14.9992,
|
547 |
+
"eval_samples_per_second": 0.667,
|
548 |
+
"eval_steps_per_second": 0.333,
|
549 |
+
"step": 5400
|
550 |
+
},
|
551 |
+
{
|
552 |
+
"epoch": 0.64,
|
553 |
+
"grad_norm": 0.25449469685554504,
|
554 |
+
"learning_rate": 0.0001681260147548922,
|
555 |
+
"loss": 1.6299,
|
556 |
+
"step": 5600
|
557 |
+
},
|
558 |
+
{
|
559 |
+
"epoch": 0.64,
|
560 |
+
"eval_bertscore": 0.7306328415870667,
|
561 |
+
"eval_loss": 1.7369228601455688,
|
562 |
+
"eval_rouge1": 0.6390760905985684,
|
563 |
+
"eval_rouge2": 0.3409328272828699,
|
564 |
+
"eval_rougeL": 0.5111832543685331,
|
565 |
+
"eval_rougeLsum": 0.6285753423407665,
|
566 |
+
"eval_runtime": 15.483,
|
567 |
+
"eval_samples_per_second": 0.646,
|
568 |
+
"eval_steps_per_second": 0.323,
|
569 |
+
"step": 5600
|
570 |
+
},
|
571 |
+
{
|
572 |
+
"epoch": 0.66,
|
573 |
+
"grad_norm": 0.24706102907657623,
|
574 |
+
"learning_rate": 0.00016698664084085796,
|
575 |
+
"loss": 1.6374,
|
576 |
+
"step": 5800
|
577 |
+
},
|
578 |
+
{
|
579 |
+
"epoch": 0.66,
|
580 |
+
"eval_bertscore": 0.732075572013855,
|
581 |
+
"eval_loss": 1.7343876361846924,
|
582 |
+
"eval_rouge1": 0.6378821977913272,
|
583 |
+
"eval_rouge2": 0.34619427775171585,
|
584 |
+
"eval_rougeL": 0.5120186953041237,
|
585 |
+
"eval_rougeLsum": 0.6284056323839109,
|
586 |
+
"eval_runtime": 15.7341,
|
587 |
+
"eval_samples_per_second": 0.636,
|
588 |
+
"eval_steps_per_second": 0.318,
|
589 |
+
"step": 5800
|
590 |
+
},
|
591 |
+
{
|
592 |
+
"epoch": 0.68,
|
593 |
+
"grad_norm": 0.24373260140419006,
|
594 |
+
"learning_rate": 0.00016584726692682372,
|
595 |
+
"loss": 1.6427,
|
596 |
+
"step": 6000
|
597 |
+
},
|
598 |
+
{
|
599 |
+
"epoch": 0.68,
|
600 |
+
"eval_bertscore": 0.7350374460220337,
|
601 |
+
"eval_loss": 1.729591965675354,
|
602 |
+
"eval_rouge1": 0.6516545226356616,
|
603 |
+
"eval_rouge2": 0.35485762033878543,
|
604 |
+
"eval_rougeL": 0.5249054193354852,
|
605 |
+
"eval_rougeLsum": 0.6411016821651583,
|
606 |
+
"eval_runtime": 15.5199,
|
607 |
+
"eval_samples_per_second": 0.644,
|
608 |
+
"eval_steps_per_second": 0.322,
|
609 |
+
"step": 6000
|
610 |
+
},
|
611 |
+
{
|
612 |
+
"epoch": 0.71,
|
613 |
+
"grad_norm": 0.24616578221321106,
|
614 |
+
"learning_rate": 0.0001647078930127895,
|
615 |
+
"loss": 1.6296,
|
616 |
+
"step": 6200
|
617 |
+
},
|
618 |
+
{
|
619 |
+
"epoch": 0.71,
|
620 |
+
"eval_bertscore": 0.7335461378097534,
|
621 |
+
"eval_loss": 1.7302274703979492,
|
622 |
+
"eval_rouge1": 0.6531411706717427,
|
623 |
+
"eval_rouge2": 0.35003053174601517,
|
624 |
+
"eval_rougeL": 0.5212483686089053,
|
625 |
+
"eval_rougeLsum": 0.6438454124825417,
|
626 |
+
"eval_runtime": 15.3058,
|
627 |
+
"eval_samples_per_second": 0.653,
|
628 |
+
"eval_steps_per_second": 0.327,
|
629 |
+
"step": 6200
|
630 |
+
},
|
631 |
+
{
|
632 |
+
"epoch": 0.73,
|
633 |
+
"grad_norm": 0.24500492215156555,
|
634 |
+
"learning_rate": 0.00016356851909875522,
|
635 |
+
"loss": 1.6251,
|
636 |
+
"step": 6400
|
637 |
+
},
|
638 |
+
{
|
639 |
+
"epoch": 0.73,
|
640 |
+
"eval_bertscore": 0.7347471714019775,
|
641 |
+
"eval_loss": 1.7292228937149048,
|
642 |
+
"eval_rouge1": 0.6565322485502556,
|
643 |
+
"eval_rouge2": 0.35887540291607073,
|
644 |
+
"eval_rougeL": 0.5284326132878907,
|
645 |
+
"eval_rougeLsum": 0.6469750895866724,
|
646 |
+
"eval_runtime": 14.9708,
|
647 |
+
"eval_samples_per_second": 0.668,
|
648 |
+
"eval_steps_per_second": 0.334,
|
649 |
+
"step": 6400
|
650 |
+
},
|
651 |
+
{
|
652 |
+
"epoch": 0.75,
|
653 |
+
"grad_norm": 0.26575228571891785,
|
654 |
+
"learning_rate": 0.000162429145184721,
|
655 |
+
"loss": 1.6389,
|
656 |
+
"step": 6600
|
657 |
+
},
|
658 |
+
{
|
659 |
+
"epoch": 0.75,
|
660 |
+
"eval_bertscore": 0.7319897413253784,
|
661 |
+
"eval_loss": 1.7287580966949463,
|
662 |
+
"eval_rouge1": 0.6458608801881298,
|
663 |
+
"eval_rouge2": 0.3503480901452204,
|
664 |
+
"eval_rougeL": 0.519626708150005,
|
665 |
+
"eval_rougeLsum": 0.6362405734928169,
|
666 |
+
"eval_runtime": 15.3509,
|
667 |
+
"eval_samples_per_second": 0.651,
|
668 |
+
"eval_steps_per_second": 0.326,
|
669 |
+
"step": 6600
|
670 |
+
},
|
671 |
+
{
|
672 |
+
"epoch": 0.77,
|
673 |
+
"grad_norm": 0.27144965529441833,
|
674 |
+
"learning_rate": 0.00016128977127068676,
|
675 |
+
"loss": 1.6476,
|
676 |
+
"step": 6800
|
677 |
+
},
|
678 |
+
{
|
679 |
+
"epoch": 0.77,
|
680 |
+
"eval_bertscore": 0.7392772436141968,
|
681 |
+
"eval_loss": 1.7257907390594482,
|
682 |
+
"eval_rouge1": 0.6543238897579965,
|
683 |
+
"eval_rouge2": 0.3606726049451984,
|
684 |
+
"eval_rougeL": 0.5317585887753791,
|
685 |
+
"eval_rougeLsum": 0.6452028420624081,
|
686 |
+
"eval_runtime": 15.0268,
|
687 |
+
"eval_samples_per_second": 0.665,
|
688 |
+
"eval_steps_per_second": 0.333,
|
689 |
+
"step": 6800
|
690 |
+
},
|
691 |
+
{
|
692 |
+
"epoch": 0.8,
|
693 |
+
"grad_norm": 0.2579711079597473,
|
694 |
+
"learning_rate": 0.00016015039735665254,
|
695 |
+
"loss": 1.6316,
|
696 |
+
"step": 7000
|
697 |
+
},
|
698 |
+
{
|
699 |
+
"epoch": 0.8,
|
700 |
+
"eval_bertscore": 0.7375612854957581,
|
701 |
+
"eval_loss": 1.7296981811523438,
|
702 |
+
"eval_rouge1": 0.658906725408624,
|
703 |
+
"eval_rouge2": 0.35825094165644866,
|
704 |
+
"eval_rougeL": 0.5323299193377959,
|
705 |
+
"eval_rougeLsum": 0.6500364347290426,
|
706 |
+
"eval_runtime": 14.9244,
|
707 |
+
"eval_samples_per_second": 0.67,
|
708 |
+
"eval_steps_per_second": 0.335,
|
709 |
+
"step": 7000
|
710 |
+
},
|
711 |
+
{
|
712 |
+
"epoch": 0.82,
|
713 |
+
"grad_norm": 0.2589207589626312,
|
714 |
+
"learning_rate": 0.0001590110234426183,
|
715 |
+
"loss": 1.6432,
|
716 |
+
"step": 7200
|
717 |
+
},
|
718 |
+
{
|
719 |
+
"epoch": 0.82,
|
720 |
+
"eval_bertscore": 0.735145092010498,
|
721 |
+
"eval_loss": 1.72720205783844,
|
722 |
+
"eval_rouge1": 0.6678646250245518,
|
723 |
+
"eval_rouge2": 0.36332843150846983,
|
724 |
+
"eval_rougeL": 0.537576430733886,
|
725 |
+
"eval_rougeLsum": 0.6579789388660506,
|
726 |
+
"eval_runtime": 14.9067,
|
727 |
+
"eval_samples_per_second": 0.671,
|
728 |
+
"eval_steps_per_second": 0.335,
|
729 |
+
"step": 7200
|
730 |
+
},
|
731 |
+
{
|
732 |
+
"epoch": 0.84,
|
733 |
+
"grad_norm": 0.26652559638023376,
|
734 |
+
"learning_rate": 0.00015787164952858404,
|
735 |
+
"loss": 1.6488,
|
736 |
+
"step": 7400
|
737 |
+
},
|
738 |
+
{
|
739 |
+
"epoch": 0.84,
|
740 |
+
"eval_bertscore": 0.7320755124092102,
|
741 |
+
"eval_loss": 1.7282931804656982,
|
742 |
+
"eval_rouge1": 0.6325633297780734,
|
743 |
+
"eval_rouge2": 0.34505856555703185,
|
744 |
+
"eval_rougeL": 0.5100006743383693,
|
745 |
+
"eval_rougeLsum": 0.6230385336341938,
|
746 |
+
"eval_runtime": 14.9075,
|
747 |
+
"eval_samples_per_second": 0.671,
|
748 |
+
"eval_steps_per_second": 0.335,
|
749 |
+
"step": 7400
|
750 |
+
},
|
751 |
+
{
|
752 |
+
"epoch": 0.87,
|
753 |
+
"grad_norm": 0.27353721857070923,
|
754 |
+
"learning_rate": 0.00015673227561454982,
|
755 |
+
"loss": 1.6486,
|
756 |
+
"step": 7600
|
757 |
+
},
|
758 |
+
{
|
759 |
+
"epoch": 0.87,
|
760 |
+
"eval_bertscore": 0.7367390394210815,
|
761 |
+
"eval_loss": 1.7290785312652588,
|
762 |
+
"eval_rouge1": 0.639487116874423,
|
763 |
+
"eval_rouge2": 0.3466574229736927,
|
764 |
+
"eval_rougeL": 0.515038120249177,
|
765 |
+
"eval_rougeLsum": 0.6301157215372983,
|
766 |
+
"eval_runtime": 15.0876,
|
767 |
+
"eval_samples_per_second": 0.663,
|
768 |
+
"eval_steps_per_second": 0.331,
|
769 |
+
"step": 7600
|
770 |
+
},
|
771 |
+
{
|
772 |
+
"epoch": 0.89,
|
773 |
+
"grad_norm": 0.24777938425540924,
|
774 |
+
"learning_rate": 0.00015559290170051558,
|
775 |
+
"loss": 1.6271,
|
776 |
+
"step": 7800
|
777 |
+
},
|
778 |
+
{
|
779 |
+
"epoch": 0.89,
|
780 |
+
"eval_bertscore": 0.735866904258728,
|
781 |
+
"eval_loss": 1.7264015674591064,
|
782 |
+
"eval_rouge1": 0.64939597901302,
|
783 |
+
"eval_rouge2": 0.3554282813944538,
|
784 |
+
"eval_rougeL": 0.5247953329477759,
|
785 |
+
"eval_rougeLsum": 0.6405524812915908,
|
786 |
+
"eval_runtime": 14.8892,
|
787 |
+
"eval_samples_per_second": 0.672,
|
788 |
+
"eval_steps_per_second": 0.336,
|
789 |
+
"step": 7800
|
790 |
+
},
|
791 |
+
{
|
792 |
+
"epoch": 0.91,
|
793 |
+
"grad_norm": 0.2703794538974762,
|
794 |
+
"learning_rate": 0.00015445352778648136,
|
795 |
+
"loss": 1.6415,
|
796 |
+
"step": 8000
|
797 |
+
},
|
798 |
+
{
|
799 |
+
"epoch": 0.91,
|
800 |
+
"eval_bertscore": 0.7325771450996399,
|
801 |
+
"eval_loss": 1.7271970510482788,
|
802 |
+
"eval_rouge1": 0.6659432894288253,
|
803 |
+
"eval_rouge2": 0.35962933912652617,
|
804 |
+
"eval_rougeL": 0.5385420432813512,
|
805 |
+
"eval_rougeLsum": 0.6557027031484046,
|
806 |
+
"eval_runtime": 14.8319,
|
807 |
+
"eval_samples_per_second": 0.674,
|
808 |
+
"eval_steps_per_second": 0.337,
|
809 |
+
"step": 8000
|
810 |
+
},
|
811 |
+
{
|
812 |
+
"epoch": 0.93,
|
813 |
+
"grad_norm": 0.28753793239593506,
|
814 |
+
"learning_rate": 0.0001533141538724471,
|
815 |
+
"loss": 1.6239,
|
816 |
+
"step": 8200
|
817 |
+
},
|
818 |
+
{
|
819 |
+
"epoch": 0.93,
|
820 |
+
"eval_bertscore": 0.7350013852119446,
|
821 |
+
"eval_loss": 1.7266199588775635,
|
822 |
+
"eval_rouge1": 0.6549282561414593,
|
823 |
+
"eval_rouge2": 0.35694530595734475,
|
824 |
+
"eval_rougeL": 0.5301601006964574,
|
825 |
+
"eval_rougeLsum": 0.6441779306137909,
|
826 |
+
"eval_runtime": 14.9005,
|
827 |
+
"eval_samples_per_second": 0.671,
|
828 |
+
"eval_steps_per_second": 0.336,
|
829 |
+
"step": 8200
|
830 |
+
},
|
831 |
+
{
|
832 |
+
"epoch": 0.96,
|
833 |
+
"grad_norm": 0.23870150744915009,
|
834 |
+
"learning_rate": 0.00015217477995841286,
|
835 |
+
"loss": 1.6293,
|
836 |
+
"step": 8400
|
837 |
+
},
|
838 |
+
{
|
839 |
+
"epoch": 0.96,
|
840 |
+
"eval_bertscore": 0.7271261811256409,
|
841 |
+
"eval_loss": 1.7256368398666382,
|
842 |
+
"eval_rouge1": 0.6515513829901936,
|
843 |
+
"eval_rouge2": 0.35217616104918836,
|
844 |
+
"eval_rougeL": 0.5236553509227138,
|
845 |
+
"eval_rougeLsum": 0.6411473505324752,
|
846 |
+
"eval_runtime": 14.9938,
|
847 |
+
"eval_samples_per_second": 0.667,
|
848 |
+
"eval_steps_per_second": 0.333,
|
849 |
+
"step": 8400
|
850 |
+
},
|
851 |
+
{
|
852 |
+
"epoch": 0.98,
|
853 |
+
"grad_norm": 0.28276997804641724,
|
854 |
+
"learning_rate": 0.00015103540604437861,
|
855 |
+
"loss": 1.6242,
|
856 |
+
"step": 8600
|
857 |
+
},
|
858 |
+
{
|
859 |
+
"epoch": 0.98,
|
860 |
+
"eval_bertscore": 0.7347334027290344,
|
861 |
+
"eval_loss": 1.717627763748169,
|
862 |
+
"eval_rouge1": 0.6350112495847634,
|
863 |
+
"eval_rouge2": 0.3477570751550898,
|
864 |
+
"eval_rougeL": 0.5146616989899861,
|
865 |
+
"eval_rougeLsum": 0.6246669376525157,
|
866 |
+
"eval_runtime": 15.7032,
|
867 |
+
"eval_samples_per_second": 0.637,
|
868 |
+
"eval_steps_per_second": 0.318,
|
869 |
+
"step": 8600
|
870 |
+
},
|
871 |
+
{
|
872 |
+
"epoch": 1.0,
|
873 |
+
"grad_norm": 0.24915842711925507,
|
874 |
+
"learning_rate": 0.00014989603213034437,
|
875 |
+
"loss": 1.6245,
|
876 |
+
"step": 8800
|
877 |
+
},
|
878 |
+
{
|
879 |
+
"epoch": 1.0,
|
880 |
+
"eval_bertscore": 0.7313701510429382,
|
881 |
+
"eval_loss": 1.7292964458465576,
|
882 |
+
"eval_rouge1": 0.6479528105367669,
|
883 |
+
"eval_rouge2": 0.35020983244262877,
|
884 |
+
"eval_rougeL": 0.5200907337780047,
|
885 |
+
"eval_rougeLsum": 0.6372896614836894,
|
886 |
+
"eval_runtime": 15.0043,
|
887 |
+
"eval_samples_per_second": 0.666,
|
888 |
+
"eval_steps_per_second": 0.333,
|
889 |
+
"step": 8800
|
890 |
+
},
|
891 |
+
{
|
892 |
+
"epoch": 1.03,
|
893 |
+
"grad_norm": 0.24036027491092682,
|
894 |
+
"learning_rate": 0.00014875665821631015,
|
895 |
+
"loss": 1.5364,
|
896 |
+
"step": 9000
|
897 |
+
},
|
898 |
+
{
|
899 |
+
"epoch": 1.03,
|
900 |
+
"eval_bertscore": 0.7327737808227539,
|
901 |
+
"eval_loss": 1.7339435815811157,
|
902 |
+
"eval_rouge1": 0.6568524178922349,
|
903 |
+
"eval_rouge2": 0.35560270713543163,
|
904 |
+
"eval_rougeL": 0.5310443670833082,
|
905 |
+
"eval_rougeLsum": 0.6480993679097387,
|
906 |
+
"eval_runtime": 14.9303,
|
907 |
+
"eval_samples_per_second": 0.67,
|
908 |
+
"eval_steps_per_second": 0.335,
|
909 |
+
"step": 9000
|
910 |
+
},
|
911 |
+
{
|
912 |
+
"epoch": 1.05,
|
913 |
+
"grad_norm": 0.2729027271270752,
|
914 |
+
"learning_rate": 0.0001476172843022759,
|
915 |
+
"loss": 1.5182,
|
916 |
+
"step": 9200
|
917 |
+
},
|
918 |
+
{
|
919 |
+
"epoch": 1.05,
|
920 |
+
"eval_bertscore": 0.7334672212600708,
|
921 |
+
"eval_loss": 1.739061713218689,
|
922 |
+
"eval_rouge1": 0.6552962187824329,
|
923 |
+
"eval_rouge2": 0.35210314124279196,
|
924 |
+
"eval_rougeL": 0.5272039052368354,
|
925 |
+
"eval_rougeLsum": 0.6437473533492806,
|
926 |
+
"eval_runtime": 15.5418,
|
927 |
+
"eval_samples_per_second": 0.643,
|
928 |
+
"eval_steps_per_second": 0.322,
|
929 |
+
"step": 9200
|
930 |
+
},
|
931 |
+
{
|
932 |
+
"epoch": 1.07,
|
933 |
+
"grad_norm": 0.2909716069698334,
|
934 |
+
"learning_rate": 0.00014647791038824168,
|
935 |
+
"loss": 1.5276,
|
936 |
+
"step": 9400
|
937 |
+
},
|
938 |
+
{
|
939 |
+
"epoch": 1.07,
|
940 |
+
"eval_bertscore": 0.7288902997970581,
|
941 |
+
"eval_loss": 1.736944556236267,
|
942 |
+
"eval_rouge1": 0.6533271685598301,
|
943 |
+
"eval_rouge2": 0.35279315321532184,
|
944 |
+
"eval_rougeL": 0.5262688234671329,
|
945 |
+
"eval_rougeLsum": 0.6424084937151033,
|
946 |
+
"eval_runtime": 15.1115,
|
947 |
+
"eval_samples_per_second": 0.662,
|
948 |
+
"eval_steps_per_second": 0.331,
|
949 |
+
"step": 9400
|
950 |
+
},
|
951 |
+
{
|
952 |
+
"epoch": 1.09,
|
953 |
+
"grad_norm": 0.3035859763622284,
|
954 |
+
"learning_rate": 0.00014533853647420743,
|
955 |
+
"loss": 1.5445,
|
956 |
+
"step": 9600
|
957 |
+
},
|
958 |
+
{
|
959 |
+
"epoch": 1.09,
|
960 |
+
"eval_bertscore": 0.7308284044265747,
|
961 |
+
"eval_loss": 1.737762689590454,
|
962 |
+
"eval_rouge1": 0.6619725777891359,
|
963 |
+
"eval_rouge2": 0.3611963714506864,
|
964 |
+
"eval_rougeL": 0.5363802967084452,
|
965 |
+
"eval_rougeLsum": 0.6516690557971352,
|
966 |
+
"eval_runtime": 14.9932,
|
967 |
+
"eval_samples_per_second": 0.667,
|
968 |
+
"eval_steps_per_second": 0.333,
|
969 |
+
"step": 9600
|
970 |
+
},
|
971 |
+
{
|
972 |
+
"epoch": 1.12,
|
973 |
+
"grad_norm": 0.26574915647506714,
|
974 |
+
"learning_rate": 0.0001441991625601732,
|
975 |
+
"loss": 1.5342,
|
976 |
+
"step": 9800
|
977 |
+
},
|
978 |
+
{
|
979 |
+
"epoch": 1.12,
|
980 |
+
"eval_bertscore": 0.7328712344169617,
|
981 |
+
"eval_loss": 1.7393991947174072,
|
982 |
+
"eval_rouge1": 0.6856504003396884,
|
983 |
+
"eval_rouge2": 0.3761098841062477,
|
984 |
+
"eval_rougeL": 0.555477293163325,
|
985 |
+
"eval_rougeLsum": 0.6757574283262289,
|
986 |
+
"eval_runtime": 14.8121,
|
987 |
+
"eval_samples_per_second": 0.675,
|
988 |
+
"eval_steps_per_second": 0.338,
|
989 |
+
"step": 9800
|
990 |
+
},
|
991 |
+
{
|
992 |
+
"epoch": 1.14,
|
993 |
+
"grad_norm": 0.315468430519104,
|
994 |
+
"learning_rate": 0.00014305978864613897,
|
995 |
+
"loss": 1.543,
|
996 |
+
"step": 10000
|
997 |
+
},
|
998 |
+
{
|
999 |
+
"epoch": 1.14,
|
1000 |
+
"eval_bertscore": 0.7349387407302856,
|
1001 |
+
"eval_loss": 1.7352710962295532,
|
1002 |
+
"eval_rouge1": 0.6749953128982036,
|
1003 |
+
"eval_rouge2": 0.3720385250530084,
|
1004 |
+
"eval_rougeL": 0.5472261566474382,
|
1005 |
+
"eval_rougeLsum": 0.6657643539219252,
|
1006 |
+
"eval_runtime": 14.909,
|
1007 |
+
"eval_samples_per_second": 0.671,
|
1008 |
+
"eval_steps_per_second": 0.335,
|
1009 |
+
"step": 10000
|
1010 |
+
},
|
1011 |
+
{
|
1012 |
+
"epoch": 1.16,
|
1013 |
+
"grad_norm": 0.29815027117729187,
|
1014 |
+
"learning_rate": 0.00014192041473210472,
|
1015 |
+
"loss": 1.5547,
|
1016 |
+
"step": 10200
|
1017 |
+
},
|
1018 |
+
{
|
1019 |
+
"epoch": 1.16,
|
1020 |
+
"eval_bertscore": 0.7359883189201355,
|
1021 |
+
"eval_loss": 1.7269136905670166,
|
1022 |
+
"eval_rouge1": 0.6561141614088863,
|
1023 |
+
"eval_rouge2": 0.3606175666303814,
|
1024 |
+
"eval_rougeL": 0.5302270771032793,
|
1025 |
+
"eval_rougeLsum": 0.6446912079521883,
|
1026 |
+
"eval_runtime": 14.9527,
|
1027 |
+
"eval_samples_per_second": 0.669,
|
1028 |
+
"eval_steps_per_second": 0.334,
|
1029 |
+
"step": 10200
|
1030 |
+
},
|
1031 |
+
{
|
1032 |
+
"epoch": 1.18,
|
1033 |
+
"grad_norm": 0.3595702350139618,
|
1034 |
+
"learning_rate": 0.00014078104081807047,
|
1035 |
+
"loss": 1.5567,
|
1036 |
+
"step": 10400
|
1037 |
+
},
|
1038 |
+
{
|
1039 |
+
"epoch": 1.18,
|
1040 |
+
"eval_bertscore": 0.7328116297721863,
|
1041 |
+
"eval_loss": 1.7341644763946533,
|
1042 |
+
"eval_rouge1": 0.6420332714823549,
|
1043 |
+
"eval_rouge2": 0.35094864549032,
|
1044 |
+
"eval_rougeL": 0.5179556761398367,
|
1045 |
+
"eval_rougeLsum": 0.631987397226809,
|
1046 |
+
"eval_runtime": 15.1438,
|
1047 |
+
"eval_samples_per_second": 0.66,
|
1048 |
+
"eval_steps_per_second": 0.33,
|
1049 |
+
"step": 10400
|
1050 |
+
},
|
1051 |
+
{
|
1052 |
+
"epoch": 1.21,
|
1053 |
+
"grad_norm": 0.2718666195869446,
|
1054 |
+
"learning_rate": 0.00013964166690403623,
|
1055 |
+
"loss": 1.5408,
|
1056 |
+
"step": 10600
|
1057 |
+
},
|
1058 |
+
{
|
1059 |
+
"epoch": 1.21,
|
1060 |
+
"eval_bertscore": 0.7337731122970581,
|
1061 |
+
"eval_loss": 1.7330901622772217,
|
1062 |
+
"eval_rouge1": 0.661681342484687,
|
1063 |
+
"eval_rouge2": 0.3626833509973693,
|
1064 |
+
"eval_rougeL": 0.5329424447373774,
|
1065 |
+
"eval_rougeLsum": 0.6519750177144633,
|
1066 |
+
"eval_runtime": 14.8142,
|
1067 |
+
"eval_samples_per_second": 0.675,
|
1068 |
+
"eval_steps_per_second": 0.338,
|
1069 |
+
"step": 10600
|
1070 |
+
},
|
1071 |
+
{
|
1072 |
+
"epoch": 1.23,
|
1073 |
+
"grad_norm": 0.29183274507522583,
|
1074 |
+
"learning_rate": 0.00013850229299000198,
|
1075 |
+
"loss": 1.5422,
|
1076 |
+
"step": 10800
|
1077 |
+
},
|
1078 |
+
{
|
1079 |
+
"epoch": 1.23,
|
1080 |
+
"eval_bertscore": 0.7331051230430603,
|
1081 |
+
"eval_loss": 1.7297636270523071,
|
1082 |
+
"eval_rouge1": 0.6655497978238063,
|
1083 |
+
"eval_rouge2": 0.3614235788441926,
|
1084 |
+
"eval_rougeL": 0.5327210061667442,
|
1085 |
+
"eval_rougeLsum": 0.6548836840483913,
|
1086 |
+
"eval_runtime": 15.1359,
|
1087 |
+
"eval_samples_per_second": 0.661,
|
1088 |
+
"eval_steps_per_second": 0.33,
|
1089 |
+
"step": 10800
|
1090 |
+
},
|
1091 |
+
{
|
1092 |
+
"epoch": 1.25,
|
1093 |
+
"grad_norm": 0.30979740619659424,
|
1094 |
+
"learning_rate": 0.00013736291907596776,
|
1095 |
+
"loss": 1.5372,
|
1096 |
+
"step": 11000
|
1097 |
+
},
|
1098 |
+
{
|
1099 |
+
"epoch": 1.25,
|
1100 |
+
"eval_bertscore": 0.7312799692153931,
|
1101 |
+
"eval_loss": 1.732444167137146,
|
1102 |
+
"eval_rouge1": 0.6568292865033993,
|
1103 |
+
"eval_rouge2": 0.35876682221562006,
|
1104 |
+
"eval_rougeL": 0.5300878844981931,
|
1105 |
+
"eval_rougeLsum": 0.6461751645858989,
|
1106 |
+
"eval_runtime": 14.819,
|
1107 |
+
"eval_samples_per_second": 0.675,
|
1108 |
+
"eval_steps_per_second": 0.337,
|
1109 |
+
"step": 11000
|
1110 |
+
},
|
1111 |
+
{
|
1112 |
+
"epoch": 1.28,
|
1113 |
+
"grad_norm": 0.31343138217926025,
|
1114 |
+
"learning_rate": 0.0001362235451619335,
|
1115 |
+
"loss": 1.5301,
|
1116 |
+
"step": 11200
|
1117 |
+
},
|
1118 |
+
{
|
1119 |
+
"epoch": 1.28,
|
1120 |
+
"eval_bertscore": 0.7317885160446167,
|
1121 |
+
"eval_loss": 1.7358499765396118,
|
1122 |
+
"eval_rouge1": 0.6548673097943329,
|
1123 |
+
"eval_rouge2": 0.3609116081432997,
|
1124 |
+
"eval_rougeL": 0.5279887650752133,
|
1125 |
+
"eval_rougeLsum": 0.6466232329097188,
|
1126 |
+
"eval_runtime": 14.8259,
|
1127 |
+
"eval_samples_per_second": 0.674,
|
1128 |
+
"eval_steps_per_second": 0.337,
|
1129 |
+
"step": 11200
|
1130 |
+
},
|
1131 |
+
{
|
1132 |
+
"epoch": 1.3,
|
1133 |
+
"grad_norm": 0.36181533336639404,
|
1134 |
+
"learning_rate": 0.0001350841712478993,
|
1135 |
+
"loss": 1.5421,
|
1136 |
+
"step": 11400
|
1137 |
+
},
|
1138 |
+
{
|
1139 |
+
"epoch": 1.3,
|
1140 |
+
"eval_bertscore": 0.7316756248474121,
|
1141 |
+
"eval_loss": 1.7282969951629639,
|
1142 |
+
"eval_rouge1": 0.6551882964480251,
|
1143 |
+
"eval_rouge2": 0.3580708921400697,
|
1144 |
+
"eval_rougeL": 0.5255367305995147,
|
1145 |
+
"eval_rougeLsum": 0.6449192953008009,
|
1146 |
+
"eval_runtime": 14.8816,
|
1147 |
+
"eval_samples_per_second": 0.672,
|
1148 |
+
"eval_steps_per_second": 0.336,
|
1149 |
+
"step": 11400
|
1150 |
+
},
|
1151 |
+
{
|
1152 |
+
"epoch": 1.32,
|
1153 |
+
"grad_norm": 0.30600836873054504,
|
1154 |
+
"learning_rate": 0.00013394479733386505,
|
1155 |
+
"loss": 1.5538,
|
1156 |
+
"step": 11600
|
1157 |
+
},
|
1158 |
+
{
|
1159 |
+
"epoch": 1.32,
|
1160 |
+
"eval_bertscore": 0.7311854362487793,
|
1161 |
+
"eval_loss": 1.7313562631607056,
|
1162 |
+
"eval_rouge1": 0.6592751199424156,
|
1163 |
+
"eval_rouge2": 0.35802855072854206,
|
1164 |
+
"eval_rougeL": 0.5297288176377084,
|
1165 |
+
"eval_rougeLsum": 0.6489455314962717,
|
1166 |
+
"eval_runtime": 15.0693,
|
1167 |
+
"eval_samples_per_second": 0.664,
|
1168 |
+
"eval_steps_per_second": 0.332,
|
1169 |
+
"step": 11600
|
1170 |
+
},
|
1171 |
+
{
|
1172 |
+
"epoch": 1.34,
|
1173 |
+
"grad_norm": 0.29904893040657043,
|
1174 |
+
"learning_rate": 0.0001328054234198308,
|
1175 |
+
"loss": 1.5328,
|
1176 |
+
"step": 11800
|
1177 |
+
},
|
1178 |
+
{
|
1179 |
+
"epoch": 1.34,
|
1180 |
+
"eval_bertscore": 0.7312635183334351,
|
1181 |
+
"eval_loss": 1.7318429946899414,
|
1182 |
+
"eval_rouge1": 0.6577169369077195,
|
1183 |
+
"eval_rouge2": 0.3582474830918887,
|
1184 |
+
"eval_rougeL": 0.5314990647771975,
|
1185 |
+
"eval_rougeLsum": 0.6454785220479866,
|
1186 |
+
"eval_runtime": 15.0235,
|
1187 |
+
"eval_samples_per_second": 0.666,
|
1188 |
+
"eval_steps_per_second": 0.333,
|
1189 |
+
"step": 11800
|
1190 |
+
},
|
1191 |
+
{
|
1192 |
+
"epoch": 1.37,
|
1193 |
+
"grad_norm": 0.3025416433811188,
|
1194 |
+
"learning_rate": 0.00013166604950579658,
|
1195 |
+
"loss": 1.5349,
|
1196 |
+
"step": 12000
|
1197 |
+
},
|
1198 |
+
{
|
1199 |
+
"epoch": 1.37,
|
1200 |
+
"eval_bertscore": 0.7325812578201294,
|
1201 |
+
"eval_loss": 1.7309118509292603,
|
1202 |
+
"eval_rouge1": 0.6629133074951261,
|
1203 |
+
"eval_rouge2": 0.3678158453940578,
|
1204 |
+
"eval_rougeL": 0.5380936907276155,
|
1205 |
+
"eval_rougeLsum": 0.654883061928214,
|
1206 |
+
"eval_runtime": 14.7666,
|
1207 |
+
"eval_samples_per_second": 0.677,
|
1208 |
+
"eval_steps_per_second": 0.339,
|
1209 |
+
"step": 12000
|
1210 |
+
},
|
1211 |
+
{
|
1212 |
+
"epoch": 1.39,
|
1213 |
+
"grad_norm": 0.34982389211654663,
|
1214 |
+
"learning_rate": 0.00013052667559176233,
|
1215 |
+
"loss": 1.5513,
|
1216 |
+
"step": 12200
|
1217 |
+
},
|
1218 |
+
{
|
1219 |
+
"epoch": 1.39,
|
1220 |
+
"eval_bertscore": 0.7340582609176636,
|
1221 |
+
"eval_loss": 1.7363474369049072,
|
1222 |
+
"eval_rouge1": 0.6555817937418287,
|
1223 |
+
"eval_rouge2": 0.35630500078358396,
|
1224 |
+
"eval_rougeL": 0.5272412353478366,
|
1225 |
+
"eval_rougeLsum": 0.6445837479327643,
|
1226 |
+
"eval_runtime": 14.9664,
|
1227 |
+
"eval_samples_per_second": 0.668,
|
1228 |
+
"eval_steps_per_second": 0.334,
|
1229 |
+
"step": 12200
|
1230 |
+
},
|
1231 |
+
{
|
1232 |
+
"epoch": 1.41,
|
1233 |
+
"grad_norm": 0.35809043049812317,
|
1234 |
+
"learning_rate": 0.0001293873016777281,
|
1235 |
+
"loss": 1.5444,
|
1236 |
+
"step": 12400
|
1237 |
+
},
|
1238 |
+
{
|
1239 |
+
"epoch": 1.41,
|
1240 |
+
"eval_bertscore": 0.7334069013595581,
|
1241 |
+
"eval_loss": 1.7328729629516602,
|
1242 |
+
"eval_rouge1": 0.6617971237669364,
|
1243 |
+
"eval_rouge2": 0.35951260376512423,
|
1244 |
+
"eval_rougeL": 0.5345512305507059,
|
1245 |
+
"eval_rougeLsum": 0.648363531132752,
|
1246 |
+
"eval_runtime": 15.2146,
|
1247 |
+
"eval_samples_per_second": 0.657,
|
1248 |
+
"eval_steps_per_second": 0.329,
|
1249 |
+
"step": 12400
|
1250 |
+
},
|
1251 |
+
{
|
1252 |
+
"epoch": 1.44,
|
1253 |
+
"grad_norm": 0.2954196631908417,
|
1254 |
+
"learning_rate": 0.00012824792776369387,
|
1255 |
+
"loss": 1.5406,
|
1256 |
+
"step": 12600
|
1257 |
+
},
|
1258 |
+
{
|
1259 |
+
"epoch": 1.44,
|
1260 |
+
"eval_bertscore": 0.7321678400039673,
|
1261 |
+
"eval_loss": 1.7335160970687866,
|
1262 |
+
"eval_rouge1": 0.6573625593086756,
|
1263 |
+
"eval_rouge2": 0.36210525247389347,
|
1264 |
+
"eval_rougeL": 0.5379361120230158,
|
1265 |
+
"eval_rougeLsum": 0.6459787883452857,
|
1266 |
+
"eval_runtime": 14.8942,
|
1267 |
+
"eval_samples_per_second": 0.671,
|
1268 |
+
"eval_steps_per_second": 0.336,
|
1269 |
+
"step": 12600
|
1270 |
+
},
|
1271 |
+
{
|
1272 |
+
"epoch": 1.46,
|
1273 |
+
"grad_norm": 0.32190588116645813,
|
1274 |
+
"learning_rate": 0.00012710855384965962,
|
1275 |
+
"loss": 1.5491,
|
1276 |
+
"step": 12800
|
1277 |
+
},
|
1278 |
+
{
|
1279 |
+
"epoch": 1.46,
|
1280 |
+
"eval_bertscore": 0.7346011400222778,
|
1281 |
+
"eval_loss": 1.7364966869354248,
|
1282 |
+
"eval_rouge1": 0.6481210247390559,
|
1283 |
+
"eval_rouge2": 0.3521173896017687,
|
1284 |
+
"eval_rougeL": 0.5240500581372636,
|
1285 |
+
"eval_rougeLsum": 0.63706442433335,
|
1286 |
+
"eval_runtime": 14.923,
|
1287 |
+
"eval_samples_per_second": 0.67,
|
1288 |
+
"eval_steps_per_second": 0.335,
|
1289 |
+
"step": 12800
|
1290 |
+
},
|
1291 |
+
{
|
1292 |
+
"epoch": 1.48,
|
1293 |
+
"grad_norm": 0.33323267102241516,
|
1294 |
+
"learning_rate": 0.00012596917993562537,
|
1295 |
+
"loss": 1.5596,
|
1296 |
+
"step": 13000
|
1297 |
+
},
|
1298 |
+
{
|
1299 |
+
"epoch": 1.48,
|
1300 |
+
"eval_bertscore": 0.7331587076187134,
|
1301 |
+
"eval_loss": 1.7332260608673096,
|
1302 |
+
"eval_rouge1": 0.6561257401878793,
|
1303 |
+
"eval_rouge2": 0.3548063723792664,
|
1304 |
+
"eval_rougeL": 0.527807776001489,
|
1305 |
+
"eval_rougeLsum": 0.6451911907984706,
|
1306 |
+
"eval_runtime": 15.4703,
|
1307 |
+
"eval_samples_per_second": 0.646,
|
1308 |
+
"eval_steps_per_second": 0.323,
|
1309 |
+
"step": 13000
|
1310 |
+
},
|
1311 |
+
{
|
1312 |
+
"epoch": 1.5,
|
1313 |
+
"grad_norm": 0.3564057946205139,
|
1314 |
+
"learning_rate": 0.00012482980602159113,
|
1315 |
+
"loss": 1.5261,
|
1316 |
+
"step": 13200
|
1317 |
+
},
|
1318 |
+
{
|
1319 |
+
"epoch": 1.5,
|
1320 |
+
"eval_bertscore": 0.7306063771247864,
|
1321 |
+
"eval_loss": 1.7368810176849365,
|
1322 |
+
"eval_rouge1": 0.637722723890071,
|
1323 |
+
"eval_rouge2": 0.3455358728458236,
|
1324 |
+
"eval_rougeL": 0.5136372690435154,
|
1325 |
+
"eval_rougeLsum": 0.6273570573595115,
|
1326 |
+
"eval_runtime": 15.3533,
|
1327 |
+
"eval_samples_per_second": 0.651,
|
1328 |
+
"eval_steps_per_second": 0.326,
|
1329 |
+
"step": 13200
|
1330 |
+
},
|
1331 |
+
{
|
1332 |
+
"epoch": 1.53,
|
1333 |
+
"grad_norm": 0.29219934344291687,
|
1334 |
+
"learning_rate": 0.0001236904321075569,
|
1335 |
+
"loss": 1.519,
|
1336 |
+
"step": 13400
|
1337 |
+
},
|
1338 |
+
{
|
1339 |
+
"epoch": 1.53,
|
1340 |
+
"eval_bertscore": 0.7338696122169495,
|
1341 |
+
"eval_loss": 1.734724998474121,
|
1342 |
+
"eval_rouge1": 0.6442107446420164,
|
1343 |
+
"eval_rouge2": 0.3494748457109431,
|
1344 |
+
"eval_rougeL": 0.5207483892007314,
|
1345 |
+
"eval_rougeLsum": 0.632886404907802,
|
1346 |
+
"eval_runtime": 15.3077,
|
1347 |
+
"eval_samples_per_second": 0.653,
|
1348 |
+
"eval_steps_per_second": 0.327,
|
1349 |
+
"step": 13400
|
1350 |
+
},
|
1351 |
+
{
|
1352 |
+
"epoch": 1.55,
|
1353 |
+
"grad_norm": 0.34681758284568787,
|
1354 |
+
"learning_rate": 0.00012255105819352266,
|
1355 |
+
"loss": 1.5419,
|
1356 |
+
"step": 13600
|
1357 |
+
},
|
1358 |
+
{
|
1359 |
+
"epoch": 1.55,
|
1360 |
+
"eval_bertscore": 0.7350045442581177,
|
1361 |
+
"eval_loss": 1.7329858541488647,
|
1362 |
+
"eval_rouge1": 0.6606839869796519,
|
1363 |
+
"eval_rouge2": 0.362188561160822,
|
1364 |
+
"eval_rougeL": 0.5342033818317451,
|
1365 |
+
"eval_rougeLsum": 0.6493340000068861,
|
1366 |
+
"eval_runtime": 15.44,
|
1367 |
+
"eval_samples_per_second": 0.648,
|
1368 |
+
"eval_steps_per_second": 0.324,
|
1369 |
+
"step": 13600
|
1370 |
+
},
|
1371 |
+
{
|
1372 |
+
"epoch": 1.57,
|
1373 |
+
"grad_norm": 0.3043666481971741,
|
1374 |
+
"learning_rate": 0.00012141168427948844,
|
1375 |
+
"loss": 1.5402,
|
1376 |
+
"step": 13800
|
1377 |
+
},
|
1378 |
+
{
|
1379 |
+
"epoch": 1.57,
|
1380 |
+
"eval_bertscore": 0.7363221645355225,
|
1381 |
+
"eval_loss": 1.7308530807495117,
|
1382 |
+
"eval_rouge1": 0.6638252384356028,
|
1383 |
+
"eval_rouge2": 0.3643237697892826,
|
1384 |
+
"eval_rougeL": 0.5403775887381331,
|
1385 |
+
"eval_rougeLsum": 0.6537260000827279,
|
1386 |
+
"eval_runtime": 14.7668,
|
1387 |
+
"eval_samples_per_second": 0.677,
|
1388 |
+
"eval_steps_per_second": 0.339,
|
1389 |
+
"step": 13800
|
1390 |
+
},
|
1391 |
+
{
|
1392 |
+
"epoch": 1.59,
|
1393 |
+
"grad_norm": 0.4073585867881775,
|
1394 |
+
"learning_rate": 0.00012027231036545419,
|
1395 |
+
"loss": 1.5256,
|
1396 |
+
"step": 14000
|
1397 |
+
},
|
1398 |
+
{
|
1399 |
+
"epoch": 1.59,
|
1400 |
+
"eval_bertscore": 0.7310279607772827,
|
1401 |
+
"eval_loss": 1.7326784133911133,
|
1402 |
+
"eval_rouge1": 0.6609594314120198,
|
1403 |
+
"eval_rouge2": 0.3601530714440473,
|
1404 |
+
"eval_rougeL": 0.5344452687135626,
|
1405 |
+
"eval_rougeLsum": 0.6480936554342305,
|
1406 |
+
"eval_runtime": 14.8565,
|
1407 |
+
"eval_samples_per_second": 0.673,
|
1408 |
+
"eval_steps_per_second": 0.337,
|
1409 |
+
"step": 14000
|
1410 |
+
},
|
1411 |
+
{
|
1412 |
+
"epoch": 1.62,
|
1413 |
+
"grad_norm": 0.3211813271045685,
|
1414 |
+
"learning_rate": 0.00011913293645141995,
|
1415 |
+
"loss": 1.5366,
|
1416 |
+
"step": 14200
|
1417 |
+
},
|
1418 |
+
{
|
1419 |
+
"epoch": 1.62,
|
1420 |
+
"eval_bertscore": 0.7356667518615723,
|
1421 |
+
"eval_loss": 1.7280094623565674,
|
1422 |
+
"eval_rouge1": 0.6519353227375031,
|
1423 |
+
"eval_rouge2": 0.3587025716186173,
|
1424 |
+
"eval_rougeL": 0.5306356200586075,
|
1425 |
+
"eval_rougeLsum": 0.6408870347994059,
|
1426 |
+
"eval_runtime": 14.9264,
|
1427 |
+
"eval_samples_per_second": 0.67,
|
1428 |
+
"eval_steps_per_second": 0.335,
|
1429 |
+
"step": 14200
|
1430 |
+
},
|
1431 |
+
{
|
1432 |
+
"epoch": 1.64,
|
1433 |
+
"grad_norm": 0.32776832580566406,
|
1434 |
+
"learning_rate": 0.00011799356253738571,
|
1435 |
+
"loss": 1.5504,
|
1436 |
+
"step": 14400
|
1437 |
+
},
|
1438 |
+
{
|
1439 |
+
"epoch": 1.64,
|
1440 |
+
"eval_bertscore": 0.7331353425979614,
|
1441 |
+
"eval_loss": 1.7308950424194336,
|
1442 |
+
"eval_rouge1": 0.6627702292814652,
|
1443 |
+
"eval_rouge2": 0.36117793957379707,
|
1444 |
+
"eval_rougeL": 0.5369305446079228,
|
1445 |
+
"eval_rougeLsum": 0.6516924083980089,
|
1446 |
+
"eval_runtime": 16.0138,
|
1447 |
+
"eval_samples_per_second": 0.624,
|
1448 |
+
"eval_steps_per_second": 0.312,
|
1449 |
+
"step": 14400
|
1450 |
+
},
|
1451 |
+
{
|
1452 |
+
"epoch": 1.66,
|
1453 |
+
"grad_norm": 0.3209726810455322,
|
1454 |
+
"learning_rate": 0.00011685418862335147,
|
1455 |
+
"loss": 1.5473,
|
1456 |
+
"step": 14600
|
1457 |
+
},
|
1458 |
+
{
|
1459 |
+
"epoch": 1.66,
|
1460 |
+
"eval_bertscore": 0.732498824596405,
|
1461 |
+
"eval_loss": 1.7328402996063232,
|
1462 |
+
"eval_rouge1": 0.6482679740803596,
|
1463 |
+
"eval_rouge2": 0.3538726087405498,
|
1464 |
+
"eval_rougeL": 0.5267677183598017,
|
1465 |
+
"eval_rougeLsum": 0.6366529460029322,
|
1466 |
+
"eval_runtime": 15.0195,
|
1467 |
+
"eval_samples_per_second": 0.666,
|
1468 |
+
"eval_steps_per_second": 0.333,
|
1469 |
+
"step": 14600
|
1470 |
+
},
|
1471 |
+
{
|
1472 |
+
"epoch": 1.69,
|
1473 |
+
"grad_norm": 0.3174591064453125,
|
1474 |
+
"learning_rate": 0.00011571481470931725,
|
1475 |
+
"loss": 1.5568,
|
1476 |
+
"step": 14800
|
1477 |
+
},
|
1478 |
+
{
|
1479 |
+
"epoch": 1.69,
|
1480 |
+
"eval_bertscore": 0.7335298657417297,
|
1481 |
+
"eval_loss": 1.7310253381729126,
|
1482 |
+
"eval_rouge1": 0.6560468577439627,
|
1483 |
+
"eval_rouge2": 0.36039371229175,
|
1484 |
+
"eval_rougeL": 0.5318708569729291,
|
1485 |
+
"eval_rougeLsum": 0.6444857558837042,
|
1486 |
+
"eval_runtime": 14.9774,
|
1487 |
+
"eval_samples_per_second": 0.668,
|
1488 |
+
"eval_steps_per_second": 0.334,
|
1489 |
+
"step": 14800
|
1490 |
+
},
|
1491 |
+
{
|
1492 |
+
"epoch": 1.71,
|
1493 |
+
"grad_norm": 0.2936408817768097,
|
1494 |
+
"learning_rate": 0.000114575440795283,
|
1495 |
+
"loss": 1.5345,
|
1496 |
+
"step": 15000
|
1497 |
+
},
|
1498 |
+
{
|
1499 |
+
"epoch": 1.71,
|
1500 |
+
"eval_bertscore": 0.7322725057601929,
|
1501 |
+
"eval_loss": 1.7270629405975342,
|
1502 |
+
"eval_rouge1": 0.6387060930656672,
|
1503 |
+
"eval_rouge2": 0.3480508127989137,
|
1504 |
+
"eval_rougeL": 0.5148670834213287,
|
1505 |
+
"eval_rougeLsum": 0.6273654952601909,
|
1506 |
+
"eval_runtime": 15.6687,
|
1507 |
+
"eval_samples_per_second": 0.638,
|
1508 |
+
"eval_steps_per_second": 0.319,
|
1509 |
+
"step": 15000
|
1510 |
+
},
|
1511 |
+
{
|
1512 |
+
"epoch": 1.73,
|
1513 |
+
"grad_norm": 0.32960689067840576,
|
1514 |
+
"learning_rate": 0.00011343606688124875,
|
1515 |
+
"loss": 1.5362,
|
1516 |
+
"step": 15200
|
1517 |
+
},
|
1518 |
+
{
|
1519 |
+
"epoch": 1.73,
|
1520 |
+
"eval_bertscore": 0.7337037920951843,
|
1521 |
+
"eval_loss": 1.7287395000457764,
|
1522 |
+
"eval_rouge1": 0.6476816970229771,
|
1523 |
+
"eval_rouge2": 0.3532248216683249,
|
1524 |
+
"eval_rougeL": 0.5253136618838716,
|
1525 |
+
"eval_rougeLsum": 0.6347493764394183,
|
1526 |
+
"eval_runtime": 15.0045,
|
1527 |
+
"eval_samples_per_second": 0.666,
|
1528 |
+
"eval_steps_per_second": 0.333,
|
1529 |
+
"step": 15200
|
1530 |
+
},
|
1531 |
+
{
|
1532 |
+
"epoch": 1.75,
|
1533 |
+
"grad_norm": 0.33265602588653564,
|
1534 |
+
"learning_rate": 0.00011229669296721452,
|
1535 |
+
"loss": 1.5215,
|
1536 |
+
"step": 15400
|
1537 |
+
},
|
1538 |
+
{
|
1539 |
+
"epoch": 1.75,
|
1540 |
+
"eval_bertscore": 0.7330806851387024,
|
1541 |
+
"eval_loss": 1.7265052795410156,
|
1542 |
+
"eval_rouge1": 0.6529393512177359,
|
1543 |
+
"eval_rouge2": 0.36182153145062224,
|
1544 |
+
"eval_rougeL": 0.5317061134915853,
|
1545 |
+
"eval_rougeLsum": 0.6413066299251913,
|
1546 |
+
"eval_runtime": 15.0256,
|
1547 |
+
"eval_samples_per_second": 0.666,
|
1548 |
+
"eval_steps_per_second": 0.333,
|
1549 |
+
"step": 15400
|
1550 |
+
},
|
1551 |
+
{
|
1552 |
+
"epoch": 1.78,
|
1553 |
+
"grad_norm": 0.3436201512813568,
|
1554 |
+
"learning_rate": 0.00011115731905318027,
|
1555 |
+
"loss": 1.539,
|
1556 |
+
"step": 15600
|
1557 |
+
},
|
1558 |
+
{
|
1559 |
+
"epoch": 1.78,
|
1560 |
+
"eval_bertscore": 0.7335551977157593,
|
1561 |
+
"eval_loss": 1.7254730463027954,
|
1562 |
+
"eval_rouge1": 0.6388518781767971,
|
1563 |
+
"eval_rouge2": 0.3501853846588857,
|
1564 |
+
"eval_rougeL": 0.5196828245794569,
|
1565 |
+
"eval_rougeLsum": 0.629333993884722,
|
1566 |
+
"eval_runtime": 15.2595,
|
1567 |
+
"eval_samples_per_second": 0.655,
|
1568 |
+
"eval_steps_per_second": 0.328,
|
1569 |
+
"step": 15600
|
1570 |
+
},
|
1571 |
+
{
|
1572 |
+
"epoch": 1.8,
|
1573 |
+
"grad_norm": 0.3428190350532532,
|
1574 |
+
"learning_rate": 0.00011001794513914605,
|
1575 |
+
"loss": 1.5273,
|
1576 |
+
"step": 15800
|
1577 |
+
},
|
1578 |
+
{
|
1579 |
+
"epoch": 1.8,
|
1580 |
+
"eval_bertscore": 0.7331770658493042,
|
1581 |
+
"eval_loss": 1.7286018133163452,
|
1582 |
+
"eval_rouge1": 0.6581941047310954,
|
1583 |
+
"eval_rouge2": 0.36277983926897583,
|
1584 |
+
"eval_rougeL": 0.5336464680120501,
|
1585 |
+
"eval_rougeLsum": 0.6489239720278894,
|
1586 |
+
"eval_runtime": 14.8252,
|
1587 |
+
"eval_samples_per_second": 0.675,
|
1588 |
+
"eval_steps_per_second": 0.337,
|
1589 |
+
"step": 15800
|
1590 |
+
},
|
1591 |
+
{
|
1592 |
+
"epoch": 1.82,
|
1593 |
+
"grad_norm": 0.363164484500885,
|
1594 |
+
"learning_rate": 0.0001088785712251118,
|
1595 |
+
"loss": 1.5445,
|
1596 |
+
"step": 16000
|
1597 |
+
},
|
1598 |
+
{
|
1599 |
+
"epoch": 1.82,
|
1600 |
+
"eval_bertscore": 0.7377282977104187,
|
1601 |
+
"eval_loss": 1.7363064289093018,
|
1602 |
+
"eval_rouge1": 0.6547011011872876,
|
1603 |
+
"eval_rouge2": 0.3553220826957326,
|
1604 |
+
"eval_rougeL": 0.5256073814411315,
|
1605 |
+
"eval_rougeLsum": 0.6420095316923398,
|
1606 |
+
"eval_runtime": 14.8599,
|
1607 |
+
"eval_samples_per_second": 0.673,
|
1608 |
+
"eval_steps_per_second": 0.336,
|
1609 |
+
"step": 16000
|
1610 |
+
},
|
1611 |
+
{
|
1612 |
+
"epoch": 1.85,
|
1613 |
+
"grad_norm": 0.3098333775997162,
|
1614 |
+
"learning_rate": 0.00010773919731107757,
|
1615 |
+
"loss": 1.5319,
|
1616 |
+
"step": 16200
|
1617 |
+
},
|
1618 |
+
{
|
1619 |
+
"epoch": 1.85,
|
1620 |
+
"eval_bertscore": 0.7324053645133972,
|
1621 |
+
"eval_loss": 1.7284066677093506,
|
1622 |
+
"eval_rouge1": 0.6477379941950916,
|
1623 |
+
"eval_rouge2": 0.3535918140554809,
|
1624 |
+
"eval_rougeL": 0.5226838544730126,
|
1625 |
+
"eval_rougeLsum": 0.6373271915355557,
|
1626 |
+
"eval_runtime": 14.9318,
|
1627 |
+
"eval_samples_per_second": 0.67,
|
1628 |
+
"eval_steps_per_second": 0.335,
|
1629 |
+
"step": 16200
|
1630 |
+
},
|
1631 |
+
{
|
1632 |
+
"epoch": 1.87,
|
1633 |
+
"grad_norm": 0.3637208938598633,
|
1634 |
+
"learning_rate": 0.00010659982339704332,
|
1635 |
+
"loss": 1.5442,
|
1636 |
+
"step": 16400
|
1637 |
+
},
|
1638 |
+
{
|
1639 |
+
"epoch": 1.87,
|
1640 |
+
"eval_bertscore": 0.7347462773323059,
|
1641 |
+
"eval_loss": 1.7252963781356812,
|
1642 |
+
"eval_rouge1": 0.6494449840449819,
|
1643 |
+
"eval_rouge2": 0.3586550050575282,
|
1644 |
+
"eval_rougeL": 0.5275675395159809,
|
1645 |
+
"eval_rougeLsum": 0.6396738714026391,
|
1646 |
+
"eval_runtime": 15.2218,
|
1647 |
+
"eval_samples_per_second": 0.657,
|
1648 |
+
"eval_steps_per_second": 0.328,
|
1649 |
+
"step": 16400
|
1650 |
+
},
|
1651 |
+
{
|
1652 |
+
"epoch": 1.89,
|
1653 |
+
"grad_norm": 0.35197457671165466,
|
1654 |
+
"learning_rate": 0.00010546044948300908,
|
1655 |
+
"loss": 1.5131,
|
1656 |
+
"step": 16600
|
1657 |
+
},
|
1658 |
+
{
|
1659 |
+
"epoch": 1.89,
|
1660 |
+
"eval_bertscore": 0.7329785227775574,
|
1661 |
+
"eval_loss": 1.7285687923431396,
|
1662 |
+
"eval_rouge1": 0.6582047811143328,
|
1663 |
+
"eval_rouge2": 0.3637700686094697,
|
1664 |
+
"eval_rougeL": 0.5355021948480279,
|
1665 |
+
"eval_rougeLsum": 0.6483245595148677,
|
1666 |
+
"eval_runtime": 14.8772,
|
1667 |
+
"eval_samples_per_second": 0.672,
|
1668 |
+
"eval_steps_per_second": 0.336,
|
1669 |
+
"step": 16600
|
1670 |
+
},
|
1671 |
+
{
|
1672 |
+
"epoch": 1.91,
|
1673 |
+
"grad_norm": 0.3406757116317749,
|
1674 |
+
"learning_rate": 0.00010432107556897486,
|
1675 |
+
"loss": 1.5394,
|
1676 |
+
"step": 16800
|
1677 |
+
},
|
1678 |
+
{
|
1679 |
+
"epoch": 1.91,
|
1680 |
+
"eval_bertscore": 0.7345961332321167,
|
1681 |
+
"eval_loss": 1.7324028015136719,
|
1682 |
+
"eval_rouge1": 0.6408293615351552,
|
1683 |
+
"eval_rouge2": 0.3520120690778129,
|
1684 |
+
"eval_rougeL": 0.5145218014745592,
|
1685 |
+
"eval_rougeLsum": 0.6297802607384266,
|
1686 |
+
"eval_runtime": 15.1044,
|
1687 |
+
"eval_samples_per_second": 0.662,
|
1688 |
+
"eval_steps_per_second": 0.331,
|
1689 |
+
"step": 16800
|
1690 |
+
},
|
1691 |
+
{
|
1692 |
+
"epoch": 1.94,
|
1693 |
+
"grad_norm": 0.3417683243751526,
|
1694 |
+
"learning_rate": 0.00010318170165494061,
|
1695 |
+
"loss": 1.526,
|
1696 |
+
"step": 17000
|
1697 |
+
},
|
1698 |
+
{
|
1699 |
+
"epoch": 1.94,
|
1700 |
+
"eval_bertscore": 0.735752522945404,
|
1701 |
+
"eval_loss": 1.7288110256195068,
|
1702 |
+
"eval_rouge1": 0.641158513352794,
|
1703 |
+
"eval_rouge2": 0.3544166440855814,
|
1704 |
+
"eval_rougeL": 0.5215201980495414,
|
1705 |
+
"eval_rougeLsum": 0.630550065494593,
|
1706 |
+
"eval_runtime": 15.0797,
|
1707 |
+
"eval_samples_per_second": 0.663,
|
1708 |
+
"eval_steps_per_second": 0.332,
|
1709 |
+
"step": 17000
|
1710 |
+
},
|
1711 |
+
{
|
1712 |
+
"epoch": 1.96,
|
1713 |
+
"grad_norm": 0.3256611227989197,
|
1714 |
+
"learning_rate": 0.00010204232774090639,
|
1715 |
+
"loss": 1.5484,
|
1716 |
+
"step": 17200
|
1717 |
+
},
|
1718 |
+
{
|
1719 |
+
"epoch": 1.96,
|
1720 |
+
"eval_bertscore": 0.7356327772140503,
|
1721 |
+
"eval_loss": 1.7305186986923218,
|
1722 |
+
"eval_rouge1": 0.6400269226515611,
|
1723 |
+
"eval_rouge2": 0.3502884634173268,
|
1724 |
+
"eval_rougeL": 0.517312321281175,
|
1725 |
+
"eval_rougeLsum": 0.6284556997614409,
|
1726 |
+
"eval_runtime": 15.4097,
|
1727 |
+
"eval_samples_per_second": 0.649,
|
1728 |
+
"eval_steps_per_second": 0.324,
|
1729 |
+
"step": 17200
|
1730 |
+
},
|
1731 |
+
{
|
1732 |
+
"epoch": 1.98,
|
1733 |
+
"grad_norm": 0.4035187363624573,
|
1734 |
+
"learning_rate": 0.00010090295382687213,
|
1735 |
+
"loss": 1.5261,
|
1736 |
+
"step": 17400
|
1737 |
+
},
|
1738 |
+
{
|
1739 |
+
"epoch": 1.98,
|
1740 |
+
"eval_bertscore": 0.7339992523193359,
|
1741 |
+
"eval_loss": 1.7282793521881104,
|
1742 |
+
"eval_rouge1": 0.6335770390416183,
|
1743 |
+
"eval_rouge2": 0.34592404578075897,
|
1744 |
+
"eval_rougeL": 0.5109045259792113,
|
1745 |
+
"eval_rougeLsum": 0.6218413683710426,
|
1746 |
+
"eval_runtime": 15.1959,
|
1747 |
+
"eval_samples_per_second": 0.658,
|
1748 |
+
"eval_steps_per_second": 0.329,
|
1749 |
+
"step": 17400
|
1750 |
+
},
|
1751 |
+
{
|
1752 |
+
"epoch": 2.0,
|
1753 |
+
"grad_norm": 0.34987062215805054,
|
1754 |
+
"learning_rate": 9.97635799128379e-05,
|
1755 |
+
"loss": 1.5199,
|
1756 |
+
"step": 17600
|
1757 |
+
},
|
1758 |
+
{
|
1759 |
+
"epoch": 2.0,
|
1760 |
+
"eval_bertscore": 0.7326329946517944,
|
1761 |
+
"eval_loss": 1.7544715404510498,
|
1762 |
+
"eval_rouge1": 0.6451558045750205,
|
1763 |
+
"eval_rouge2": 0.35565806935653943,
|
1764 |
+
"eval_rougeL": 0.5217034865840529,
|
1765 |
+
"eval_rougeLsum": 0.6329869715356753,
|
1766 |
+
"eval_runtime": 15.0351,
|
1767 |
+
"eval_samples_per_second": 0.665,
|
1768 |
+
"eval_steps_per_second": 0.333,
|
1769 |
+
"step": 17600
|
1770 |
+
},
|
1771 |
+
{
|
1772 |
+
"epoch": 2.03,
|
1773 |
+
"grad_norm": 0.37184038758277893,
|
1774 |
+
"learning_rate": 9.862420599880366e-05,
|
1775 |
+
"loss": 1.41,
|
1776 |
+
"step": 17800
|
1777 |
+
},
|
1778 |
+
{
|
1779 |
+
"epoch": 2.03,
|
1780 |
+
"eval_bertscore": 0.7315141558647156,
|
1781 |
+
"eval_loss": 1.7585878372192383,
|
1782 |
+
"eval_rouge1": 0.6469319193583706,
|
1783 |
+
"eval_rouge2": 0.3514447211469598,
|
1784 |
+
"eval_rougeL": 0.524755857688278,
|
1785 |
+
"eval_rougeLsum": 0.6350164781858667,
|
1786 |
+
"eval_runtime": 14.9583,
|
1787 |
+
"eval_samples_per_second": 0.669,
|
1788 |
+
"eval_steps_per_second": 0.334,
|
1789 |
+
"step": 17800
|
1790 |
+
},
|
1791 |
+
{
|
1792 |
+
"epoch": 2.05,
|
1793 |
+
"grad_norm": 0.3812776803970337,
|
1794 |
+
"learning_rate": 9.748483208476943e-05,
|
1795 |
+
"loss": 1.4132,
|
1796 |
+
"step": 18000
|
1797 |
+
},
|
1798 |
+
{
|
1799 |
+
"epoch": 2.05,
|
1800 |
+
"eval_bertscore": 0.7335561513900757,
|
1801 |
+
"eval_loss": 1.764611840248108,
|
1802 |
+
"eval_rouge1": 0.6381916780473581,
|
1803 |
+
"eval_rouge2": 0.3482510604092539,
|
1804 |
+
"eval_rougeL": 0.5162105225823392,
|
1805 |
+
"eval_rougeLsum": 0.627150245441782,
|
1806 |
+
"eval_runtime": 15.8515,
|
1807 |
+
"eval_samples_per_second": 0.631,
|
1808 |
+
"eval_steps_per_second": 0.315,
|
1809 |
+
"step": 18000
|
1810 |
+
},
|
1811 |
+
{
|
1812 |
+
"epoch": 2.07,
|
1813 |
+
"grad_norm": 0.45525220036506653,
|
1814 |
+
"learning_rate": 9.634545817073518e-05,
|
1815 |
+
"loss": 1.4,
|
1816 |
+
"step": 18200
|
1817 |
+
},
|
1818 |
+
{
|
1819 |
+
"epoch": 2.07,
|
1820 |
+
"eval_bertscore": 0.73627769947052,
|
1821 |
+
"eval_loss": 1.7585163116455078,
|
1822 |
+
"eval_rouge1": 0.6670097658027134,
|
1823 |
+
"eval_rouge2": 0.3658295359911405,
|
1824 |
+
"eval_rougeL": 0.5429667657900548,
|
1825 |
+
"eval_rougeLsum": 0.6543501745791419,
|
1826 |
+
"eval_runtime": 15.0301,
|
1827 |
+
"eval_samples_per_second": 0.665,
|
1828 |
+
"eval_steps_per_second": 0.333,
|
1829 |
+
"step": 18200
|
1830 |
+
},
|
1831 |
+
{
|
1832 |
+
"epoch": 2.1,
|
1833 |
+
"grad_norm": 0.37322184443473816,
|
1834 |
+
"learning_rate": 9.520608425670095e-05,
|
1835 |
+
"loss": 1.4293,
|
1836 |
+
"step": 18400
|
1837 |
+
},
|
1838 |
+
{
|
1839 |
+
"epoch": 2.1,
|
1840 |
+
"eval_bertscore": 0.730435848236084,
|
1841 |
+
"eval_loss": 1.764052391052246,
|
1842 |
+
"eval_rouge1": 0.6640215078213034,
|
1843 |
+
"eval_rouge2": 0.3625932287322054,
|
1844 |
+
"eval_rougeL": 0.5379978391335138,
|
1845 |
+
"eval_rougeLsum": 0.6542054656293199,
|
1846 |
+
"eval_runtime": 15.0762,
|
1847 |
+
"eval_samples_per_second": 0.663,
|
1848 |
+
"eval_steps_per_second": 0.332,
|
1849 |
+
"step": 18400
|
1850 |
+
},
|
1851 |
+
{
|
1852 |
+
"epoch": 2.12,
|
1853 |
+
"grad_norm": 0.4260891079902649,
|
1854 |
+
"learning_rate": 9.40667103426667e-05,
|
1855 |
+
"loss": 1.4077,
|
1856 |
+
"step": 18600
|
1857 |
+
},
|
1858 |
+
{
|
1859 |
+
"epoch": 2.12,
|
1860 |
+
"eval_bertscore": 0.7309869527816772,
|
1861 |
+
"eval_loss": 1.762108564376831,
|
1862 |
+
"eval_rouge1": 0.6571171081737958,
|
1863 |
+
"eval_rouge2": 0.35780421333141865,
|
1864 |
+
"eval_rougeL": 0.5320129270967632,
|
1865 |
+
"eval_rougeLsum": 0.64587787409523,
|
1866 |
+
"eval_runtime": 14.9004,
|
1867 |
+
"eval_samples_per_second": 0.671,
|
1868 |
+
"eval_steps_per_second": 0.336,
|
1869 |
+
"step": 18600
|
1870 |
+
},
|
1871 |
+
{
|
1872 |
+
"epoch": 2.14,
|
1873 |
+
"grad_norm": 0.39479926228523254,
|
1874 |
+
"learning_rate": 9.292733642863247e-05,
|
1875 |
+
"loss": 1.4165,
|
1876 |
+
"step": 18800
|
1877 |
+
},
|
1878 |
+
{
|
1879 |
+
"epoch": 2.14,
|
1880 |
+
"eval_bertscore": 0.7324444651603699,
|
1881 |
+
"eval_loss": 1.7607113122940063,
|
1882 |
+
"eval_rouge1": 0.6628398862884018,
|
1883 |
+
"eval_rouge2": 0.3627259806721216,
|
1884 |
+
"eval_rougeL": 0.5366106483832656,
|
1885 |
+
"eval_rougeLsum": 0.6528364858807157,
|
1886 |
+
"eval_runtime": 15.5766,
|
1887 |
+
"eval_samples_per_second": 0.642,
|
1888 |
+
"eval_steps_per_second": 0.321,
|
1889 |
+
"step": 18800
|
1890 |
+
},
|
1891 |
+
{
|
1892 |
+
"epoch": 2.16,
|
1893 |
+
"grad_norm": 0.39267703890800476,
|
1894 |
+
"learning_rate": 9.178796251459824e-05,
|
1895 |
+
"loss": 1.4123,
|
1896 |
+
"step": 19000
|
1897 |
+
},
|
1898 |
+
{
|
1899 |
+
"epoch": 2.16,
|
1900 |
+
"eval_bertscore": 0.7298994064331055,
|
1901 |
+
"eval_loss": 1.7668545246124268,
|
1902 |
+
"eval_rouge1": 0.6490850022857569,
|
1903 |
+
"eval_rouge2": 0.3532323419511264,
|
1904 |
+
"eval_rougeL": 0.5212823000193295,
|
1905 |
+
"eval_rougeLsum": 0.636442724466695,
|
1906 |
+
"eval_runtime": 14.9094,
|
1907 |
+
"eval_samples_per_second": 0.671,
|
1908 |
+
"eval_steps_per_second": 0.335,
|
1909 |
+
"step": 19000
|
1910 |
+
},
|
1911 |
+
{
|
1912 |
+
"epoch": 2.19,
|
1913 |
+
"grad_norm": 0.38221287727355957,
|
1914 |
+
"learning_rate": 9.0648588600564e-05,
|
1915 |
+
"loss": 1.401,
|
1916 |
+
"step": 19200
|
1917 |
+
},
|
1918 |
+
{
|
1919 |
+
"epoch": 2.19,
|
1920 |
+
"eval_bertscore": 0.7316875457763672,
|
1921 |
+
"eval_loss": 1.764147400856018,
|
1922 |
+
"eval_rouge1": 0.6490326710625849,
|
1923 |
+
"eval_rouge2": 0.3510351037900723,
|
1924 |
+
"eval_rougeL": 0.5239165028795836,
|
1925 |
+
"eval_rougeLsum": 0.6373687316421427,
|
1926 |
+
"eval_runtime": 15.1192,
|
1927 |
+
"eval_samples_per_second": 0.661,
|
1928 |
+
"eval_steps_per_second": 0.331,
|
1929 |
+
"step": 19200
|
1930 |
+
}
|
1931 |
+
],
|
1932 |
+
"logging_steps": 200,
|
1933 |
+
"max_steps": 35112,
|
1934 |
+
"num_input_tokens_seen": 0,
|
1935 |
+
"num_train_epochs": 4,
|
1936 |
+
"save_steps": 800,
|
1937 |
+
"total_flos": 1.9452726542216356e+18,
|
1938 |
+
"train_batch_size": 2,
|
1939 |
+
"trial_name": null,
|
1940 |
+
"trial_params": null
|
1941 |
+
}
|
checkpoint-19200/training_args.bin
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:d576743d4294cfbcae212243377437f73cabbcded2df9da7f73e6e39ecb38f56
|
3 |
+
size 5048
|
checkpoint-20000/README.md
ADDED
@@ -0,0 +1,202 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
---
|
2 |
+
library_name: peft
|
3 |
+
base_model: unsloth/gemma-2b-bnb-4bit
|
4 |
+
---
|
5 |
+
|
6 |
+
# Model Card for Model ID
|
7 |
+
|
8 |
+
<!-- Provide a quick summary of what the model is/does. -->
|
9 |
+
|
10 |
+
|
11 |
+
|
12 |
+
## Model Details
|
13 |
+
|
14 |
+
### Model Description
|
15 |
+
|
16 |
+
<!-- Provide a longer summary of what this model is. -->
|
17 |
+
|
18 |
+
|
19 |
+
|
20 |
+
- **Developed by:** [More Information Needed]
|
21 |
+
- **Funded by [optional]:** [More Information Needed]
|
22 |
+
- **Shared by [optional]:** [More Information Needed]
|
23 |
+
- **Model type:** [More Information Needed]
|
24 |
+
- **Language(s) (NLP):** [More Information Needed]
|
25 |
+
- **License:** [More Information Needed]
|
26 |
+
- **Finetuned from model [optional]:** [More Information Needed]
|
27 |
+
|
28 |
+
### Model Sources [optional]
|
29 |
+
|
30 |
+
<!-- Provide the basic links for the model. -->
|
31 |
+
|
32 |
+
- **Repository:** [More Information Needed]
|
33 |
+
- **Paper [optional]:** [More Information Needed]
|
34 |
+
- **Demo [optional]:** [More Information Needed]
|
35 |
+
|
36 |
+
## Uses
|
37 |
+
|
38 |
+
<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
|
39 |
+
|
40 |
+
### Direct Use
|
41 |
+
|
42 |
+
<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
|
43 |
+
|
44 |
+
[More Information Needed]
|
45 |
+
|
46 |
+
### Downstream Use [optional]
|
47 |
+
|
48 |
+
<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
|
49 |
+
|
50 |
+
[More Information Needed]
|
51 |
+
|
52 |
+
### Out-of-Scope Use
|
53 |
+
|
54 |
+
<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
|
55 |
+
|
56 |
+
[More Information Needed]
|
57 |
+
|
58 |
+
## Bias, Risks, and Limitations
|
59 |
+
|
60 |
+
<!-- This section is meant to convey both technical and sociotechnical limitations. -->
|
61 |
+
|
62 |
+
[More Information Needed]
|
63 |
+
|
64 |
+
### Recommendations
|
65 |
+
|
66 |
+
<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
|
67 |
+
|
68 |
+
Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
|
69 |
+
|
70 |
+
## How to Get Started with the Model
|
71 |
+
|
72 |
+
Use the code below to get started with the model.
|
73 |
+
|
74 |
+
[More Information Needed]
|
75 |
+
|
76 |
+
## Training Details
|
77 |
+
|
78 |
+
### Training Data
|
79 |
+
|
80 |
+
<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
|
81 |
+
|
82 |
+
[More Information Needed]
|
83 |
+
|
84 |
+
### Training Procedure
|
85 |
+
|
86 |
+
<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
|
87 |
+
|
88 |
+
#### Preprocessing [optional]
|
89 |
+
|
90 |
+
[More Information Needed]
|
91 |
+
|
92 |
+
|
93 |
+
#### Training Hyperparameters
|
94 |
+
|
95 |
+
- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
|
96 |
+
|
97 |
+
#### Speeds, Sizes, Times [optional]
|
98 |
+
|
99 |
+
<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
|
100 |
+
|
101 |
+
[More Information Needed]
|
102 |
+
|
103 |
+
## Evaluation
|
104 |
+
|
105 |
+
<!-- This section describes the evaluation protocols and provides the results. -->
|
106 |
+
|
107 |
+
### Testing Data, Factors & Metrics
|
108 |
+
|
109 |
+
#### Testing Data
|
110 |
+
|
111 |
+
<!-- This should link to a Dataset Card if possible. -->
|
112 |
+
|
113 |
+
[More Information Needed]
|
114 |
+
|
115 |
+
#### Factors
|
116 |
+
|
117 |
+
<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
|
118 |
+
|
119 |
+
[More Information Needed]
|
120 |
+
|
121 |
+
#### Metrics
|
122 |
+
|
123 |
+
<!-- These are the evaluation metrics being used, ideally with a description of why. -->
|
124 |
+
|
125 |
+
[More Information Needed]
|
126 |
+
|
127 |
+
### Results
|
128 |
+
|
129 |
+
[More Information Needed]
|
130 |
+
|
131 |
+
#### Summary
|
132 |
+
|
133 |
+
|
134 |
+
|
135 |
+
## Model Examination [optional]
|
136 |
+
|
137 |
+
<!-- Relevant interpretability work for the model goes here -->
|
138 |
+
|
139 |
+
[More Information Needed]
|
140 |
+
|
141 |
+
## Environmental Impact
|
142 |
+
|
143 |
+
<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
|
144 |
+
|
145 |
+
Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
|
146 |
+
|
147 |
+
- **Hardware Type:** [More Information Needed]
|
148 |
+
- **Hours used:** [More Information Needed]
|
149 |
+
- **Cloud Provider:** [More Information Needed]
|
150 |
+
- **Compute Region:** [More Information Needed]
|
151 |
+
- **Carbon Emitted:** [More Information Needed]
|
152 |
+
|
153 |
+
## Technical Specifications [optional]
|
154 |
+
|
155 |
+
### Model Architecture and Objective
|
156 |
+
|
157 |
+
[More Information Needed]
|
158 |
+
|
159 |
+
### Compute Infrastructure
|
160 |
+
|
161 |
+
[More Information Needed]
|
162 |
+
|
163 |
+
#### Hardware
|
164 |
+
|
165 |
+
[More Information Needed]
|
166 |
+
|
167 |
+
#### Software
|
168 |
+
|
169 |
+
[More Information Needed]
|
170 |
+
|
171 |
+
## Citation [optional]
|
172 |
+
|
173 |
+
<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
|
174 |
+
|
175 |
+
**BibTeX:**
|
176 |
+
|
177 |
+
[More Information Needed]
|
178 |
+
|
179 |
+
**APA:**
|
180 |
+
|
181 |
+
[More Information Needed]
|
182 |
+
|
183 |
+
## Glossary [optional]
|
184 |
+
|
185 |
+
<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
|
186 |
+
|
187 |
+
[More Information Needed]
|
188 |
+
|
189 |
+
## More Information [optional]
|
190 |
+
|
191 |
+
[More Information Needed]
|
192 |
+
|
193 |
+
## Model Card Authors [optional]
|
194 |
+
|
195 |
+
[More Information Needed]
|
196 |
+
|
197 |
+
## Model Card Contact
|
198 |
+
|
199 |
+
[More Information Needed]
|
200 |
+
### Framework versions
|
201 |
+
|
202 |
+
- PEFT 0.10.0
|
checkpoint-20000/adapter_config.json
ADDED
@@ -0,0 +1,34 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"alpha_pattern": {},
|
3 |
+
"auto_mapping": null,
|
4 |
+
"base_model_name_or_path": "unsloth/gemma-2b-bnb-4bit",
|
5 |
+
"bias": "none",
|
6 |
+
"fan_in_fan_out": false,
|
7 |
+
"inference_mode": true,
|
8 |
+
"init_lora_weights": true,
|
9 |
+
"layer_replication": null,
|
10 |
+
"layers_pattern": null,
|
11 |
+
"layers_to_transform": null,
|
12 |
+
"loftq_config": {},
|
13 |
+
"lora_alpha": 32,
|
14 |
+
"lora_dropout": 0,
|
15 |
+
"megatron_config": null,
|
16 |
+
"megatron_core": "megatron.core",
|
17 |
+
"modules_to_save": null,
|
18 |
+
"peft_type": "LORA",
|
19 |
+
"r": 64,
|
20 |
+
"rank_pattern": {},
|
21 |
+
"revision": "unsloth",
|
22 |
+
"target_modules": [
|
23 |
+
"v_proj",
|
24 |
+
"up_proj",
|
25 |
+
"gate_proj",
|
26 |
+
"o_proj",
|
27 |
+
"k_proj",
|
28 |
+
"down_proj",
|
29 |
+
"q_proj"
|
30 |
+
],
|
31 |
+
"task_type": "CAUSAL_LM",
|
32 |
+
"use_dora": false,
|
33 |
+
"use_rslora": false
|
34 |
+
}
|
checkpoint-20000/adapter_model.safetensors
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:ba64a48660d96e6ae210c02e7b193ff3ddad9dba223c1a63500a488145c438cb
|
3 |
+
size 313820248
|
checkpoint-20000/rng_state.pth
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:0fba149b3295ba204666aba65c69a1092c2683c7409ae96a94ad417bde27841b
|
3 |
+
size 14244
|
checkpoint-20000/scheduler.pt
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:de935c76c515679cf99a388f4ac63df38c3b321d9fb6d4b28cb9bf7338ecb1af
|
3 |
+
size 1064
|
checkpoint-20000/special_tokens_map.json
ADDED
@@ -0,0 +1,34 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"additional_special_tokens": [
|
3 |
+
"<start_of_turn>",
|
4 |
+
"<end_of_turn>"
|
5 |
+
],
|
6 |
+
"bos_token": {
|
7 |
+
"content": "<bos>",
|
8 |
+
"lstrip": false,
|
9 |
+
"normalized": false,
|
10 |
+
"rstrip": false,
|
11 |
+
"single_word": false
|
12 |
+
},
|
13 |
+
"eos_token": {
|
14 |
+
"content": "<eos>",
|
15 |
+
"lstrip": false,
|
16 |
+
"normalized": false,
|
17 |
+
"rstrip": false,
|
18 |
+
"single_word": false
|
19 |
+
},
|
20 |
+
"pad_token": {
|
21 |
+
"content": "<pad>",
|
22 |
+
"lstrip": false,
|
23 |
+
"normalized": false,
|
24 |
+
"rstrip": false,
|
25 |
+
"single_word": false
|
26 |
+
},
|
27 |
+
"unk_token": {
|
28 |
+
"content": "<unk>",
|
29 |
+
"lstrip": false,
|
30 |
+
"normalized": false,
|
31 |
+
"rstrip": false,
|
32 |
+
"single_word": false
|
33 |
+
}
|
34 |
+
}
|
checkpoint-20000/tokenizer.json
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:05e97791a5e007260de1db7e1692e53150e08cea481e2bf25435553380c147ee
|
3 |
+
size 17477929
|
checkpoint-20000/tokenizer.model
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:61a7b147390c64585d6c3543dd6fc636906c9af3865a5548f27f31aee1d4c8e2
|
3 |
+
size 4241003
|
checkpoint-20000/tokenizer_config.json
ADDED
@@ -0,0 +1,71 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"add_bos_token": true,
|
3 |
+
"add_eos_token": false,
|
4 |
+
"added_tokens_decoder": {
|
5 |
+
"0": {
|
6 |
+
"content": "<pad>",
|
7 |
+
"lstrip": false,
|
8 |
+
"normalized": false,
|
9 |
+
"rstrip": false,
|
10 |
+
"single_word": false,
|
11 |
+
"special": true
|
12 |
+
},
|
13 |
+
"1": {
|
14 |
+
"content": "<eos>",
|
15 |
+
"lstrip": false,
|
16 |
+
"normalized": false,
|
17 |
+
"rstrip": false,
|
18 |
+
"single_word": false,
|
19 |
+
"special": true
|
20 |
+
},
|
21 |
+
"2": {
|
22 |
+
"content": "<bos>",
|
23 |
+
"lstrip": false,
|
24 |
+
"normalized": false,
|
25 |
+
"rstrip": false,
|
26 |
+
"single_word": false,
|
27 |
+
"special": true
|
28 |
+
},
|
29 |
+
"3": {
|
30 |
+
"content": "<unk>",
|
31 |
+
"lstrip": false,
|
32 |
+
"normalized": false,
|
33 |
+
"rstrip": false,
|
34 |
+
"single_word": false,
|
35 |
+
"special": true
|
36 |
+
},
|
37 |
+
"106": {
|
38 |
+
"content": "<start_of_turn>",
|
39 |
+
"lstrip": false,
|
40 |
+
"normalized": false,
|
41 |
+
"rstrip": false,
|
42 |
+
"single_word": false,
|
43 |
+
"special": true
|
44 |
+
},
|
45 |
+
"107": {
|
46 |
+
"content": "<end_of_turn>",
|
47 |
+
"lstrip": false,
|
48 |
+
"normalized": false,
|
49 |
+
"rstrip": false,
|
50 |
+
"single_word": false,
|
51 |
+
"special": true
|
52 |
+
}
|
53 |
+
},
|
54 |
+
"additional_special_tokens": [
|
55 |
+
"<start_of_turn>",
|
56 |
+
"<end_of_turn>"
|
57 |
+
],
|
58 |
+
"bos_token": "<bos>",
|
59 |
+
"chat_template": "{{ bos_token }}{% if messages[0]['role'] == 'system' %}{{ raise_exception('System role not supported') }}{% endif %}{% for message in messages %}{% if (message['role'] == 'user') != (loop.index0 % 2 == 0) %}{{ raise_exception('Conversation roles must alternate user/assistant/user/assistant/...') }}{% endif %}{% if (message['role'] == 'assistant') %}{% set role = 'model' %}{% else %}{% set role = message['role'] %}{% endif %}{{ '<start_of_turn>' + role + '\n' + message['content'] | trim + '<end_of_turn>\n' }}{% endfor %}{% if add_generation_prompt %}{{'<start_of_turn>model\n'}}{% endif %}",
|
60 |
+
"clean_up_tokenization_spaces": false,
|
61 |
+
"eos_token": "<eos>",
|
62 |
+
"legacy": null,
|
63 |
+
"model_max_length": 8192,
|
64 |
+
"pad_token": "<pad>",
|
65 |
+
"padding_side": "right",
|
66 |
+
"sp_model_kwargs": {},
|
67 |
+
"spaces_between_special_tokens": false,
|
68 |
+
"tokenizer_class": "GemmaTokenizer",
|
69 |
+
"unk_token": "<unk>",
|
70 |
+
"use_default_system_prompt": false
|
71 |
+
}
|
checkpoint-20000/trainer_state.json
ADDED
@@ -0,0 +1,2021 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"best_metric": null,
|
3 |
+
"best_model_checkpoint": null,
|
4 |
+
"epoch": 2.2782935581249646,
|
5 |
+
"eval_steps": 200,
|
6 |
+
"global_step": 20000,
|
7 |
+
"is_hyper_param_search": false,
|
8 |
+
"is_local_process_zero": true,
|
9 |
+
"is_world_process_zero": true,
|
10 |
+
"log_history": [
|
11 |
+
{
|
12 |
+
"epoch": 0.02,
|
13 |
+
"grad_norm": 0.1643640249967575,
|
14 |
+
"learning_rate": 0.0001988891104338166,
|
15 |
+
"loss": 1.7673,
|
16 |
+
"step": 200
|
17 |
+
},
|
18 |
+
{
|
19 |
+
"epoch": 0.02,
|
20 |
+
"eval_bertscore": 0.7312520742416382,
|
21 |
+
"eval_loss": 1.7944419384002686,
|
22 |
+
"eval_rouge1": 0.645726048132668,
|
23 |
+
"eval_rouge2": 0.342840307585653,
|
24 |
+
"eval_rougeL": 0.5174784271125388,
|
25 |
+
"eval_rougeLsum": 0.6359911842715976,
|
26 |
+
"eval_runtime": 67.7968,
|
27 |
+
"eval_samples_per_second": 0.147,
|
28 |
+
"eval_steps_per_second": 0.074,
|
29 |
+
"step": 200
|
30 |
+
},
|
31 |
+
{
|
32 |
+
"epoch": 0.05,
|
33 |
+
"grad_norm": 0.17238478362560272,
|
34 |
+
"learning_rate": 0.00019774973651978238,
|
35 |
+
"loss": 1.6985,
|
36 |
+
"step": 400
|
37 |
+
},
|
38 |
+
{
|
39 |
+
"epoch": 0.05,
|
40 |
+
"eval_bertscore": 0.733666718006134,
|
41 |
+
"eval_loss": 1.7791178226470947,
|
42 |
+
"eval_rouge1": 0.6540909153028596,
|
43 |
+
"eval_rouge2": 0.3548819059818129,
|
44 |
+
"eval_rougeL": 0.527257232694246,
|
45 |
+
"eval_rougeLsum": 0.6442799950994005,
|
46 |
+
"eval_runtime": 15.1267,
|
47 |
+
"eval_samples_per_second": 0.661,
|
48 |
+
"eval_steps_per_second": 0.331,
|
49 |
+
"step": 400
|
50 |
+
},
|
51 |
+
{
|
52 |
+
"epoch": 0.07,
|
53 |
+
"grad_norm": 0.19368696212768555,
|
54 |
+
"learning_rate": 0.00019661036260574814,
|
55 |
+
"loss": 1.6962,
|
56 |
+
"step": 600
|
57 |
+
},
|
58 |
+
{
|
59 |
+
"epoch": 0.07,
|
60 |
+
"eval_bertscore": 0.7339462041854858,
|
61 |
+
"eval_loss": 1.7609882354736328,
|
62 |
+
"eval_rouge1": 0.6384337329686338,
|
63 |
+
"eval_rouge2": 0.3415514270662107,
|
64 |
+
"eval_rougeL": 0.51206080148464,
|
65 |
+
"eval_rougeLsum": 0.6261968614666548,
|
66 |
+
"eval_runtime": 15.2068,
|
67 |
+
"eval_samples_per_second": 0.658,
|
68 |
+
"eval_steps_per_second": 0.329,
|
69 |
+
"step": 600
|
70 |
+
},
|
71 |
+
{
|
72 |
+
"epoch": 0.09,
|
73 |
+
"grad_norm": 0.18629203736782074,
|
74 |
+
"learning_rate": 0.00019547098869171392,
|
75 |
+
"loss": 1.6825,
|
76 |
+
"step": 800
|
77 |
+
},
|
78 |
+
{
|
79 |
+
"epoch": 0.09,
|
80 |
+
"eval_bertscore": 0.7363594174385071,
|
81 |
+
"eval_loss": 1.7610784769058228,
|
82 |
+
"eval_rouge1": 0.6461624591922237,
|
83 |
+
"eval_rouge2": 0.3477371388439609,
|
84 |
+
"eval_rougeL": 0.5187429174752844,
|
85 |
+
"eval_rougeLsum": 0.6361089823008282,
|
86 |
+
"eval_runtime": 15.173,
|
87 |
+
"eval_samples_per_second": 0.659,
|
88 |
+
"eval_steps_per_second": 0.33,
|
89 |
+
"step": 800
|
90 |
+
},
|
91 |
+
{
|
92 |
+
"epoch": 0.11,
|
93 |
+
"grad_norm": 0.1799013316631317,
|
94 |
+
"learning_rate": 0.00019433161477767967,
|
95 |
+
"loss": 1.6848,
|
96 |
+
"step": 1000
|
97 |
+
},
|
98 |
+
{
|
99 |
+
"epoch": 0.11,
|
100 |
+
"eval_bertscore": 0.7334067225456238,
|
101 |
+
"eval_loss": 1.7576347589492798,
|
102 |
+
"eval_rouge1": 0.6345119236349537,
|
103 |
+
"eval_rouge2": 0.3422519149071803,
|
104 |
+
"eval_rougeL": 0.5111983101326238,
|
105 |
+
"eval_rougeLsum": 0.6244653120436832,
|
106 |
+
"eval_runtime": 15.2847,
|
107 |
+
"eval_samples_per_second": 0.654,
|
108 |
+
"eval_steps_per_second": 0.327,
|
109 |
+
"step": 1000
|
110 |
+
},
|
111 |
+
{
|
112 |
+
"epoch": 0.14,
|
113 |
+
"grad_norm": 0.22036150097846985,
|
114 |
+
"learning_rate": 0.00019319224086364545,
|
115 |
+
"loss": 1.6714,
|
116 |
+
"step": 1200
|
117 |
+
},
|
118 |
+
{
|
119 |
+
"epoch": 0.14,
|
120 |
+
"eval_bertscore": 0.7323788404464722,
|
121 |
+
"eval_loss": 1.7521806955337524,
|
122 |
+
"eval_rouge1": 0.6452540184557478,
|
123 |
+
"eval_rouge2": 0.3465145726476423,
|
124 |
+
"eval_rougeL": 0.516711757588783,
|
125 |
+
"eval_rougeLsum": 0.6341049885677059,
|
126 |
+
"eval_runtime": 15.1247,
|
127 |
+
"eval_samples_per_second": 0.661,
|
128 |
+
"eval_steps_per_second": 0.331,
|
129 |
+
"step": 1200
|
130 |
+
},
|
131 |
+
{
|
132 |
+
"epoch": 0.16,
|
133 |
+
"grad_norm": 0.21381086111068726,
|
134 |
+
"learning_rate": 0.0001920528669496112,
|
135 |
+
"loss": 1.6669,
|
136 |
+
"step": 1400
|
137 |
+
},
|
138 |
+
{
|
139 |
+
"epoch": 0.16,
|
140 |
+
"eval_bertscore": 0.7313202619552612,
|
141 |
+
"eval_loss": 1.7520482540130615,
|
142 |
+
"eval_rouge1": 0.6397526546254797,
|
143 |
+
"eval_rouge2": 0.3452671288110514,
|
144 |
+
"eval_rougeL": 0.5176580626678706,
|
145 |
+
"eval_rougeLsum": 0.6296746647539768,
|
146 |
+
"eval_runtime": 15.183,
|
147 |
+
"eval_samples_per_second": 0.659,
|
148 |
+
"eval_steps_per_second": 0.329,
|
149 |
+
"step": 1400
|
150 |
+
},
|
151 |
+
{
|
152 |
+
"epoch": 0.18,
|
153 |
+
"grad_norm": 0.20332874357700348,
|
154 |
+
"learning_rate": 0.00019091349303557696,
|
155 |
+
"loss": 1.671,
|
156 |
+
"step": 1600
|
157 |
+
},
|
158 |
+
{
|
159 |
+
"epoch": 0.18,
|
160 |
+
"eval_bertscore": 0.7349230647087097,
|
161 |
+
"eval_loss": 1.7473630905151367,
|
162 |
+
"eval_rouge1": 0.637439872504459,
|
163 |
+
"eval_rouge2": 0.34307164454056094,
|
164 |
+
"eval_rougeL": 0.5129717676228565,
|
165 |
+
"eval_rougeLsum": 0.6272190896182391,
|
166 |
+
"eval_runtime": 15.5672,
|
167 |
+
"eval_samples_per_second": 0.642,
|
168 |
+
"eval_steps_per_second": 0.321,
|
169 |
+
"step": 1600
|
170 |
+
},
|
171 |
+
{
|
172 |
+
"epoch": 0.21,
|
173 |
+
"grad_norm": 0.2025599479675293,
|
174 |
+
"learning_rate": 0.00018977411912154274,
|
175 |
+
"loss": 1.6721,
|
176 |
+
"step": 1800
|
177 |
+
},
|
178 |
+
{
|
179 |
+
"epoch": 0.21,
|
180 |
+
"eval_bertscore": 0.7357184290885925,
|
181 |
+
"eval_loss": 1.7516342401504517,
|
182 |
+
"eval_rouge1": 0.6387615819926658,
|
183 |
+
"eval_rouge2": 0.34366787517105574,
|
184 |
+
"eval_rougeL": 0.5129026911770751,
|
185 |
+
"eval_rougeLsum": 0.6289314118258257,
|
186 |
+
"eval_runtime": 15.9574,
|
187 |
+
"eval_samples_per_second": 0.627,
|
188 |
+
"eval_steps_per_second": 0.313,
|
189 |
+
"step": 1800
|
190 |
+
},
|
191 |
+
{
|
192 |
+
"epoch": 0.23,
|
193 |
+
"grad_norm": 0.20457112789154053,
|
194 |
+
"learning_rate": 0.0001886347452075085,
|
195 |
+
"loss": 1.671,
|
196 |
+
"step": 2000
|
197 |
+
},
|
198 |
+
{
|
199 |
+
"epoch": 0.23,
|
200 |
+
"eval_bertscore": 0.733718752861023,
|
201 |
+
"eval_loss": 1.7501707077026367,
|
202 |
+
"eval_rouge1": 0.6346207681220664,
|
203 |
+
"eval_rouge2": 0.33748369437614106,
|
204 |
+
"eval_rougeL": 0.5085159047705141,
|
205 |
+
"eval_rougeLsum": 0.6239953154441167,
|
206 |
+
"eval_runtime": 15.0863,
|
207 |
+
"eval_samples_per_second": 0.663,
|
208 |
+
"eval_steps_per_second": 0.331,
|
209 |
+
"step": 2000
|
210 |
+
},
|
211 |
+
{
|
212 |
+
"epoch": 0.25,
|
213 |
+
"grad_norm": 0.22552740573883057,
|
214 |
+
"learning_rate": 0.00018749537129347424,
|
215 |
+
"loss": 1.6496,
|
216 |
+
"step": 2200
|
217 |
+
},
|
218 |
+
{
|
219 |
+
"epoch": 0.25,
|
220 |
+
"eval_bertscore": 0.7368552684783936,
|
221 |
+
"eval_loss": 1.7437107563018799,
|
222 |
+
"eval_rouge1": 0.6490756387878311,
|
223 |
+
"eval_rouge2": 0.3448817738175175,
|
224 |
+
"eval_rougeL": 0.5235187045706321,
|
225 |
+
"eval_rougeLsum": 0.6377780857890332,
|
226 |
+
"eval_runtime": 15.07,
|
227 |
+
"eval_samples_per_second": 0.664,
|
228 |
+
"eval_steps_per_second": 0.332,
|
229 |
+
"step": 2200
|
230 |
+
},
|
231 |
+
{
|
232 |
+
"epoch": 0.27,
|
233 |
+
"grad_norm": 0.22573673725128174,
|
234 |
+
"learning_rate": 0.00018635599737944,
|
235 |
+
"loss": 1.6629,
|
236 |
+
"step": 2400
|
237 |
+
},
|
238 |
+
{
|
239 |
+
"epoch": 0.27,
|
240 |
+
"eval_bertscore": 0.7314499616622925,
|
241 |
+
"eval_loss": 1.7462828159332275,
|
242 |
+
"eval_rouge1": 0.6511482369678803,
|
243 |
+
"eval_rouge2": 0.34632544827771805,
|
244 |
+
"eval_rougeL": 0.5212417191003778,
|
245 |
+
"eval_rougeLsum": 0.6415391907940229,
|
246 |
+
"eval_runtime": 15.2078,
|
247 |
+
"eval_samples_per_second": 0.658,
|
248 |
+
"eval_steps_per_second": 0.329,
|
249 |
+
"step": 2400
|
250 |
+
},
|
251 |
+
{
|
252 |
+
"epoch": 0.3,
|
253 |
+
"grad_norm": 0.26426687836647034,
|
254 |
+
"learning_rate": 0.00018521662346540575,
|
255 |
+
"loss": 1.6644,
|
256 |
+
"step": 2600
|
257 |
+
},
|
258 |
+
{
|
259 |
+
"epoch": 0.3,
|
260 |
+
"eval_bertscore": 0.7363359928131104,
|
261 |
+
"eval_loss": 1.7505037784576416,
|
262 |
+
"eval_rouge1": 0.6498296552335481,
|
263 |
+
"eval_rouge2": 0.34873833589761183,
|
264 |
+
"eval_rougeL": 0.5194028620820592,
|
265 |
+
"eval_rougeLsum": 0.6404603087578984,
|
266 |
+
"eval_runtime": 14.8403,
|
267 |
+
"eval_samples_per_second": 0.674,
|
268 |
+
"eval_steps_per_second": 0.337,
|
269 |
+
"step": 2600
|
270 |
+
},
|
271 |
+
{
|
272 |
+
"epoch": 0.32,
|
273 |
+
"grad_norm": 0.20142091810703278,
|
274 |
+
"learning_rate": 0.00018407724955137153,
|
275 |
+
"loss": 1.6535,
|
276 |
+
"step": 2800
|
277 |
+
},
|
278 |
+
{
|
279 |
+
"epoch": 0.32,
|
280 |
+
"eval_bertscore": 0.7304679155349731,
|
281 |
+
"eval_loss": 1.7511039972305298,
|
282 |
+
"eval_rouge1": 0.6475130585388738,
|
283 |
+
"eval_rouge2": 0.34648331046897884,
|
284 |
+
"eval_rougeL": 0.5218042284020985,
|
285 |
+
"eval_rougeLsum": 0.6382749834402862,
|
286 |
+
"eval_runtime": 15.0162,
|
287 |
+
"eval_samples_per_second": 0.666,
|
288 |
+
"eval_steps_per_second": 0.333,
|
289 |
+
"step": 2800
|
290 |
+
},
|
291 |
+
{
|
292 |
+
"epoch": 0.34,
|
293 |
+
"grad_norm": 0.23283220827579498,
|
294 |
+
"learning_rate": 0.00018293787563733728,
|
295 |
+
"loss": 1.6477,
|
296 |
+
"step": 3000
|
297 |
+
},
|
298 |
+
{
|
299 |
+
"epoch": 0.34,
|
300 |
+
"eval_bertscore": 0.7327049374580383,
|
301 |
+
"eval_loss": 1.7461665868759155,
|
302 |
+
"eval_rouge1": 0.6309349586871908,
|
303 |
+
"eval_rouge2": 0.3387882478990309,
|
304 |
+
"eval_rougeL": 0.5042059192403674,
|
305 |
+
"eval_rougeLsum": 0.6210432469674847,
|
306 |
+
"eval_runtime": 15.463,
|
307 |
+
"eval_samples_per_second": 0.647,
|
308 |
+
"eval_steps_per_second": 0.323,
|
309 |
+
"step": 3000
|
310 |
+
},
|
311 |
+
{
|
312 |
+
"epoch": 0.36,
|
313 |
+
"grad_norm": 0.21316750347614288,
|
314 |
+
"learning_rate": 0.00018179850172330306,
|
315 |
+
"loss": 1.6614,
|
316 |
+
"step": 3200
|
317 |
+
},
|
318 |
+
{
|
319 |
+
"epoch": 0.36,
|
320 |
+
"eval_bertscore": 0.7314620018005371,
|
321 |
+
"eval_loss": 1.7468239068984985,
|
322 |
+
"eval_rouge1": 0.6480904534152265,
|
323 |
+
"eval_rouge2": 0.3479530168963481,
|
324 |
+
"eval_rougeL": 0.5193148273848067,
|
325 |
+
"eval_rougeLsum": 0.6366010767207634,
|
326 |
+
"eval_runtime": 15.0381,
|
327 |
+
"eval_samples_per_second": 0.665,
|
328 |
+
"eval_steps_per_second": 0.332,
|
329 |
+
"step": 3200
|
330 |
+
},
|
331 |
+
{
|
332 |
+
"epoch": 0.39,
|
333 |
+
"grad_norm": 0.26080408692359924,
|
334 |
+
"learning_rate": 0.00018065912780926882,
|
335 |
+
"loss": 1.6591,
|
336 |
+
"step": 3400
|
337 |
+
},
|
338 |
+
{
|
339 |
+
"epoch": 0.39,
|
340 |
+
"eval_bertscore": 0.7327477335929871,
|
341 |
+
"eval_loss": 1.7442594766616821,
|
342 |
+
"eval_rouge1": 0.6424613378037144,
|
343 |
+
"eval_rouge2": 0.34731770322974903,
|
344 |
+
"eval_rougeL": 0.5160705879794565,
|
345 |
+
"eval_rougeLsum": 0.6327006420281607,
|
346 |
+
"eval_runtime": 15.5373,
|
347 |
+
"eval_samples_per_second": 0.644,
|
348 |
+
"eval_steps_per_second": 0.322,
|
349 |
+
"step": 3400
|
350 |
+
},
|
351 |
+
{
|
352 |
+
"epoch": 0.41,
|
353 |
+
"grad_norm": 0.23274216055870056,
|
354 |
+
"learning_rate": 0.0001795197538952346,
|
355 |
+
"loss": 1.6613,
|
356 |
+
"step": 3600
|
357 |
+
},
|
358 |
+
{
|
359 |
+
"epoch": 0.41,
|
360 |
+
"eval_bertscore": 0.736956000328064,
|
361 |
+
"eval_loss": 1.7429373264312744,
|
362 |
+
"eval_rouge1": 0.6514574160666677,
|
363 |
+
"eval_rouge2": 0.3556199242231646,
|
364 |
+
"eval_rougeL": 0.5249726237675663,
|
365 |
+
"eval_rougeLsum": 0.6406261097623661,
|
366 |
+
"eval_runtime": 14.9415,
|
367 |
+
"eval_samples_per_second": 0.669,
|
368 |
+
"eval_steps_per_second": 0.335,
|
369 |
+
"step": 3600
|
370 |
+
},
|
371 |
+
{
|
372 |
+
"epoch": 0.43,
|
373 |
+
"grad_norm": 0.23616766929626465,
|
374 |
+
"learning_rate": 0.00017838037998120035,
|
375 |
+
"loss": 1.6479,
|
376 |
+
"step": 3800
|
377 |
+
},
|
378 |
+
{
|
379 |
+
"epoch": 0.43,
|
380 |
+
"eval_bertscore": 0.7349627614021301,
|
381 |
+
"eval_loss": 1.7420669794082642,
|
382 |
+
"eval_rouge1": 0.655851684949526,
|
383 |
+
"eval_rouge2": 0.35254590691865084,
|
384 |
+
"eval_rougeL": 0.5248980956621441,
|
385 |
+
"eval_rougeLsum": 0.6449637270581419,
|
386 |
+
"eval_runtime": 15.4178,
|
387 |
+
"eval_samples_per_second": 0.649,
|
388 |
+
"eval_steps_per_second": 0.324,
|
389 |
+
"step": 3800
|
390 |
+
},
|
391 |
+
{
|
392 |
+
"epoch": 0.46,
|
393 |
+
"grad_norm": 0.23260319232940674,
|
394 |
+
"learning_rate": 0.0001772410060671661,
|
395 |
+
"loss": 1.6569,
|
396 |
+
"step": 4000
|
397 |
+
},
|
398 |
+
{
|
399 |
+
"epoch": 0.46,
|
400 |
+
"eval_bertscore": 0.7332885265350342,
|
401 |
+
"eval_loss": 1.7401313781738281,
|
402 |
+
"eval_rouge1": 0.6669483634140105,
|
403 |
+
"eval_rouge2": 0.35873988835161297,
|
404 |
+
"eval_rougeL": 0.5343868725007427,
|
405 |
+
"eval_rougeLsum": 0.6555353134690931,
|
406 |
+
"eval_runtime": 15.0822,
|
407 |
+
"eval_samples_per_second": 0.663,
|
408 |
+
"eval_steps_per_second": 0.332,
|
409 |
+
"step": 4000
|
410 |
+
},
|
411 |
+
{
|
412 |
+
"epoch": 0.48,
|
413 |
+
"grad_norm": 0.2366473525762558,
|
414 |
+
"learning_rate": 0.00017610163215313186,
|
415 |
+
"loss": 1.6599,
|
416 |
+
"step": 4200
|
417 |
+
},
|
418 |
+
{
|
419 |
+
"epoch": 0.48,
|
420 |
+
"eval_bertscore": 0.7335314750671387,
|
421 |
+
"eval_loss": 1.7385823726654053,
|
422 |
+
"eval_rouge1": 0.6559297063578133,
|
423 |
+
"eval_rouge2": 0.35483499789990636,
|
424 |
+
"eval_rougeL": 0.5297939800986089,
|
425 |
+
"eval_rougeLsum": 0.6454544372491222,
|
426 |
+
"eval_runtime": 15.1029,
|
427 |
+
"eval_samples_per_second": 0.662,
|
428 |
+
"eval_steps_per_second": 0.331,
|
429 |
+
"step": 4200
|
430 |
+
},
|
431 |
+
{
|
432 |
+
"epoch": 0.5,
|
433 |
+
"grad_norm": 0.20628753304481506,
|
434 |
+
"learning_rate": 0.0001749622582390976,
|
435 |
+
"loss": 1.6454,
|
436 |
+
"step": 4400
|
437 |
+
},
|
438 |
+
{
|
439 |
+
"epoch": 0.5,
|
440 |
+
"eval_bertscore": 0.7342795133590698,
|
441 |
+
"eval_loss": 1.7422058582305908,
|
442 |
+
"eval_rouge1": 0.660746519614568,
|
443 |
+
"eval_rouge2": 0.3633965895561597,
|
444 |
+
"eval_rougeL": 0.5369036980876734,
|
445 |
+
"eval_rougeLsum": 0.650338328328998,
|
446 |
+
"eval_runtime": 15.4434,
|
447 |
+
"eval_samples_per_second": 0.648,
|
448 |
+
"eval_steps_per_second": 0.324,
|
449 |
+
"step": 4400
|
450 |
+
},
|
451 |
+
{
|
452 |
+
"epoch": 0.52,
|
453 |
+
"grad_norm": 0.2239149957895279,
|
454 |
+
"learning_rate": 0.0001738228843250634,
|
455 |
+
"loss": 1.6594,
|
456 |
+
"step": 4600
|
457 |
+
},
|
458 |
+
{
|
459 |
+
"epoch": 0.52,
|
460 |
+
"eval_bertscore": 0.7313543558120728,
|
461 |
+
"eval_loss": 1.740854263305664,
|
462 |
+
"eval_rouge1": 0.6591645132619427,
|
463 |
+
"eval_rouge2": 0.35766117432431743,
|
464 |
+
"eval_rougeL": 0.532710255034635,
|
465 |
+
"eval_rougeLsum": 0.6479428185884644,
|
466 |
+
"eval_runtime": 14.9436,
|
467 |
+
"eval_samples_per_second": 0.669,
|
468 |
+
"eval_steps_per_second": 0.335,
|
469 |
+
"step": 4600
|
470 |
+
},
|
471 |
+
{
|
472 |
+
"epoch": 0.55,
|
473 |
+
"grad_norm": 0.24808338284492493,
|
474 |
+
"learning_rate": 0.00017268351041102914,
|
475 |
+
"loss": 1.6604,
|
476 |
+
"step": 4800
|
477 |
+
},
|
478 |
+
{
|
479 |
+
"epoch": 0.55,
|
480 |
+
"eval_bertscore": 0.7333321571350098,
|
481 |
+
"eval_loss": 1.7385585308074951,
|
482 |
+
"eval_rouge1": 0.6532115808232871,
|
483 |
+
"eval_rouge2": 0.35333788022501567,
|
484 |
+
"eval_rougeL": 0.5284071547874328,
|
485 |
+
"eval_rougeLsum": 0.6410472452797623,
|
486 |
+
"eval_runtime": 15.0277,
|
487 |
+
"eval_samples_per_second": 0.665,
|
488 |
+
"eval_steps_per_second": 0.333,
|
489 |
+
"step": 4800
|
490 |
+
},
|
491 |
+
{
|
492 |
+
"epoch": 0.57,
|
493 |
+
"grad_norm": 0.2555364966392517,
|
494 |
+
"learning_rate": 0.0001715441364969949,
|
495 |
+
"loss": 1.6493,
|
496 |
+
"step": 5000
|
497 |
+
},
|
498 |
+
{
|
499 |
+
"epoch": 0.57,
|
500 |
+
"eval_bertscore": 0.7318152189254761,
|
501 |
+
"eval_loss": 1.7357494831085205,
|
502 |
+
"eval_rouge1": 0.6476755890502586,
|
503 |
+
"eval_rouge2": 0.35312778275949164,
|
504 |
+
"eval_rougeL": 0.5227601228049905,
|
505 |
+
"eval_rougeLsum": 0.6371331138372852,
|
506 |
+
"eval_runtime": 14.8807,
|
507 |
+
"eval_samples_per_second": 0.672,
|
508 |
+
"eval_steps_per_second": 0.336,
|
509 |
+
"step": 5000
|
510 |
+
},
|
511 |
+
{
|
512 |
+
"epoch": 0.59,
|
513 |
+
"grad_norm": 0.21518155932426453,
|
514 |
+
"learning_rate": 0.00017040476258296068,
|
515 |
+
"loss": 1.644,
|
516 |
+
"step": 5200
|
517 |
+
},
|
518 |
+
{
|
519 |
+
"epoch": 0.59,
|
520 |
+
"eval_bertscore": 0.734805703163147,
|
521 |
+
"eval_loss": 1.74032723903656,
|
522 |
+
"eval_rouge1": 0.6476813733451636,
|
523 |
+
"eval_rouge2": 0.3509259728617576,
|
524 |
+
"eval_rougeL": 0.5221334872800274,
|
525 |
+
"eval_rougeLsum": 0.636892384667733,
|
526 |
+
"eval_runtime": 15.1271,
|
527 |
+
"eval_samples_per_second": 0.661,
|
528 |
+
"eval_steps_per_second": 0.331,
|
529 |
+
"step": 5200
|
530 |
+
},
|
531 |
+
{
|
532 |
+
"epoch": 0.62,
|
533 |
+
"grad_norm": 0.26086658239364624,
|
534 |
+
"learning_rate": 0.00016926538866892643,
|
535 |
+
"loss": 1.6449,
|
536 |
+
"step": 5400
|
537 |
+
},
|
538 |
+
{
|
539 |
+
"epoch": 0.62,
|
540 |
+
"eval_bertscore": 0.7339995503425598,
|
541 |
+
"eval_loss": 1.7338205575942993,
|
542 |
+
"eval_rouge1": 0.6416889902864902,
|
543 |
+
"eval_rouge2": 0.3479045880347737,
|
544 |
+
"eval_rougeL": 0.5160577838468976,
|
545 |
+
"eval_rougeLsum": 0.6317983411796093,
|
546 |
+
"eval_runtime": 14.9992,
|
547 |
+
"eval_samples_per_second": 0.667,
|
548 |
+
"eval_steps_per_second": 0.333,
|
549 |
+
"step": 5400
|
550 |
+
},
|
551 |
+
{
|
552 |
+
"epoch": 0.64,
|
553 |
+
"grad_norm": 0.25449469685554504,
|
554 |
+
"learning_rate": 0.0001681260147548922,
|
555 |
+
"loss": 1.6299,
|
556 |
+
"step": 5600
|
557 |
+
},
|
558 |
+
{
|
559 |
+
"epoch": 0.64,
|
560 |
+
"eval_bertscore": 0.7306328415870667,
|
561 |
+
"eval_loss": 1.7369228601455688,
|
562 |
+
"eval_rouge1": 0.6390760905985684,
|
563 |
+
"eval_rouge2": 0.3409328272828699,
|
564 |
+
"eval_rougeL": 0.5111832543685331,
|
565 |
+
"eval_rougeLsum": 0.6285753423407665,
|
566 |
+
"eval_runtime": 15.483,
|
567 |
+
"eval_samples_per_second": 0.646,
|
568 |
+
"eval_steps_per_second": 0.323,
|
569 |
+
"step": 5600
|
570 |
+
},
|
571 |
+
{
|
572 |
+
"epoch": 0.66,
|
573 |
+
"grad_norm": 0.24706102907657623,
|
574 |
+
"learning_rate": 0.00016698664084085796,
|
575 |
+
"loss": 1.6374,
|
576 |
+
"step": 5800
|
577 |
+
},
|
578 |
+
{
|
579 |
+
"epoch": 0.66,
|
580 |
+
"eval_bertscore": 0.732075572013855,
|
581 |
+
"eval_loss": 1.7343876361846924,
|
582 |
+
"eval_rouge1": 0.6378821977913272,
|
583 |
+
"eval_rouge2": 0.34619427775171585,
|
584 |
+
"eval_rougeL": 0.5120186953041237,
|
585 |
+
"eval_rougeLsum": 0.6284056323839109,
|
586 |
+
"eval_runtime": 15.7341,
|
587 |
+
"eval_samples_per_second": 0.636,
|
588 |
+
"eval_steps_per_second": 0.318,
|
589 |
+
"step": 5800
|
590 |
+
},
|
591 |
+
{
|
592 |
+
"epoch": 0.68,
|
593 |
+
"grad_norm": 0.24373260140419006,
|
594 |
+
"learning_rate": 0.00016584726692682372,
|
595 |
+
"loss": 1.6427,
|
596 |
+
"step": 6000
|
597 |
+
},
|
598 |
+
{
|
599 |
+
"epoch": 0.68,
|
600 |
+
"eval_bertscore": 0.7350374460220337,
|
601 |
+
"eval_loss": 1.729591965675354,
|
602 |
+
"eval_rouge1": 0.6516545226356616,
|
603 |
+
"eval_rouge2": 0.35485762033878543,
|
604 |
+
"eval_rougeL": 0.5249054193354852,
|
605 |
+
"eval_rougeLsum": 0.6411016821651583,
|
606 |
+
"eval_runtime": 15.5199,
|
607 |
+
"eval_samples_per_second": 0.644,
|
608 |
+
"eval_steps_per_second": 0.322,
|
609 |
+
"step": 6000
|
610 |
+
},
|
611 |
+
{
|
612 |
+
"epoch": 0.71,
|
613 |
+
"grad_norm": 0.24616578221321106,
|
614 |
+
"learning_rate": 0.0001647078930127895,
|
615 |
+
"loss": 1.6296,
|
616 |
+
"step": 6200
|
617 |
+
},
|
618 |
+
{
|
619 |
+
"epoch": 0.71,
|
620 |
+
"eval_bertscore": 0.7335461378097534,
|
621 |
+
"eval_loss": 1.7302274703979492,
|
622 |
+
"eval_rouge1": 0.6531411706717427,
|
623 |
+
"eval_rouge2": 0.35003053174601517,
|
624 |
+
"eval_rougeL": 0.5212483686089053,
|
625 |
+
"eval_rougeLsum": 0.6438454124825417,
|
626 |
+
"eval_runtime": 15.3058,
|
627 |
+
"eval_samples_per_second": 0.653,
|
628 |
+
"eval_steps_per_second": 0.327,
|
629 |
+
"step": 6200
|
630 |
+
},
|
631 |
+
{
|
632 |
+
"epoch": 0.73,
|
633 |
+
"grad_norm": 0.24500492215156555,
|
634 |
+
"learning_rate": 0.00016356851909875522,
|
635 |
+
"loss": 1.6251,
|
636 |
+
"step": 6400
|
637 |
+
},
|
638 |
+
{
|
639 |
+
"epoch": 0.73,
|
640 |
+
"eval_bertscore": 0.7347471714019775,
|
641 |
+
"eval_loss": 1.7292228937149048,
|
642 |
+
"eval_rouge1": 0.6565322485502556,
|
643 |
+
"eval_rouge2": 0.35887540291607073,
|
644 |
+
"eval_rougeL": 0.5284326132878907,
|
645 |
+
"eval_rougeLsum": 0.6469750895866724,
|
646 |
+
"eval_runtime": 14.9708,
|
647 |
+
"eval_samples_per_second": 0.668,
|
648 |
+
"eval_steps_per_second": 0.334,
|
649 |
+
"step": 6400
|
650 |
+
},
|
651 |
+
{
|
652 |
+
"epoch": 0.75,
|
653 |
+
"grad_norm": 0.26575228571891785,
|
654 |
+
"learning_rate": 0.000162429145184721,
|
655 |
+
"loss": 1.6389,
|
656 |
+
"step": 6600
|
657 |
+
},
|
658 |
+
{
|
659 |
+
"epoch": 0.75,
|
660 |
+
"eval_bertscore": 0.7319897413253784,
|
661 |
+
"eval_loss": 1.7287580966949463,
|
662 |
+
"eval_rouge1": 0.6458608801881298,
|
663 |
+
"eval_rouge2": 0.3503480901452204,
|
664 |
+
"eval_rougeL": 0.519626708150005,
|
665 |
+
"eval_rougeLsum": 0.6362405734928169,
|
666 |
+
"eval_runtime": 15.3509,
|
667 |
+
"eval_samples_per_second": 0.651,
|
668 |
+
"eval_steps_per_second": 0.326,
|
669 |
+
"step": 6600
|
670 |
+
},
|
671 |
+
{
|
672 |
+
"epoch": 0.77,
|
673 |
+
"grad_norm": 0.27144965529441833,
|
674 |
+
"learning_rate": 0.00016128977127068676,
|
675 |
+
"loss": 1.6476,
|
676 |
+
"step": 6800
|
677 |
+
},
|
678 |
+
{
|
679 |
+
"epoch": 0.77,
|
680 |
+
"eval_bertscore": 0.7392772436141968,
|
681 |
+
"eval_loss": 1.7257907390594482,
|
682 |
+
"eval_rouge1": 0.6543238897579965,
|
683 |
+
"eval_rouge2": 0.3606726049451984,
|
684 |
+
"eval_rougeL": 0.5317585887753791,
|
685 |
+
"eval_rougeLsum": 0.6452028420624081,
|
686 |
+
"eval_runtime": 15.0268,
|
687 |
+
"eval_samples_per_second": 0.665,
|
688 |
+
"eval_steps_per_second": 0.333,
|
689 |
+
"step": 6800
|
690 |
+
},
|
691 |
+
{
|
692 |
+
"epoch": 0.8,
|
693 |
+
"grad_norm": 0.2579711079597473,
|
694 |
+
"learning_rate": 0.00016015039735665254,
|
695 |
+
"loss": 1.6316,
|
696 |
+
"step": 7000
|
697 |
+
},
|
698 |
+
{
|
699 |
+
"epoch": 0.8,
|
700 |
+
"eval_bertscore": 0.7375612854957581,
|
701 |
+
"eval_loss": 1.7296981811523438,
|
702 |
+
"eval_rouge1": 0.658906725408624,
|
703 |
+
"eval_rouge2": 0.35825094165644866,
|
704 |
+
"eval_rougeL": 0.5323299193377959,
|
705 |
+
"eval_rougeLsum": 0.6500364347290426,
|
706 |
+
"eval_runtime": 14.9244,
|
707 |
+
"eval_samples_per_second": 0.67,
|
708 |
+
"eval_steps_per_second": 0.335,
|
709 |
+
"step": 7000
|
710 |
+
},
|
711 |
+
{
|
712 |
+
"epoch": 0.82,
|
713 |
+
"grad_norm": 0.2589207589626312,
|
714 |
+
"learning_rate": 0.0001590110234426183,
|
715 |
+
"loss": 1.6432,
|
716 |
+
"step": 7200
|
717 |
+
},
|
718 |
+
{
|
719 |
+
"epoch": 0.82,
|
720 |
+
"eval_bertscore": 0.735145092010498,
|
721 |
+
"eval_loss": 1.72720205783844,
|
722 |
+
"eval_rouge1": 0.6678646250245518,
|
723 |
+
"eval_rouge2": 0.36332843150846983,
|
724 |
+
"eval_rougeL": 0.537576430733886,
|
725 |
+
"eval_rougeLsum": 0.6579789388660506,
|
726 |
+
"eval_runtime": 14.9067,
|
727 |
+
"eval_samples_per_second": 0.671,
|
728 |
+
"eval_steps_per_second": 0.335,
|
729 |
+
"step": 7200
|
730 |
+
},
|
731 |
+
{
|
732 |
+
"epoch": 0.84,
|
733 |
+
"grad_norm": 0.26652559638023376,
|
734 |
+
"learning_rate": 0.00015787164952858404,
|
735 |
+
"loss": 1.6488,
|
736 |
+
"step": 7400
|
737 |
+
},
|
738 |
+
{
|
739 |
+
"epoch": 0.84,
|
740 |
+
"eval_bertscore": 0.7320755124092102,
|
741 |
+
"eval_loss": 1.7282931804656982,
|
742 |
+
"eval_rouge1": 0.6325633297780734,
|
743 |
+
"eval_rouge2": 0.34505856555703185,
|
744 |
+
"eval_rougeL": 0.5100006743383693,
|
745 |
+
"eval_rougeLsum": 0.6230385336341938,
|
746 |
+
"eval_runtime": 14.9075,
|
747 |
+
"eval_samples_per_second": 0.671,
|
748 |
+
"eval_steps_per_second": 0.335,
|
749 |
+
"step": 7400
|
750 |
+
},
|
751 |
+
{
|
752 |
+
"epoch": 0.87,
|
753 |
+
"grad_norm": 0.27353721857070923,
|
754 |
+
"learning_rate": 0.00015673227561454982,
|
755 |
+
"loss": 1.6486,
|
756 |
+
"step": 7600
|
757 |
+
},
|
758 |
+
{
|
759 |
+
"epoch": 0.87,
|
760 |
+
"eval_bertscore": 0.7367390394210815,
|
761 |
+
"eval_loss": 1.7290785312652588,
|
762 |
+
"eval_rouge1": 0.639487116874423,
|
763 |
+
"eval_rouge2": 0.3466574229736927,
|
764 |
+
"eval_rougeL": 0.515038120249177,
|
765 |
+
"eval_rougeLsum": 0.6301157215372983,
|
766 |
+
"eval_runtime": 15.0876,
|
767 |
+
"eval_samples_per_second": 0.663,
|
768 |
+
"eval_steps_per_second": 0.331,
|
769 |
+
"step": 7600
|
770 |
+
},
|
771 |
+
{
|
772 |
+
"epoch": 0.89,
|
773 |
+
"grad_norm": 0.24777938425540924,
|
774 |
+
"learning_rate": 0.00015559290170051558,
|
775 |
+
"loss": 1.6271,
|
776 |
+
"step": 7800
|
777 |
+
},
|
778 |
+
{
|
779 |
+
"epoch": 0.89,
|
780 |
+
"eval_bertscore": 0.735866904258728,
|
781 |
+
"eval_loss": 1.7264015674591064,
|
782 |
+
"eval_rouge1": 0.64939597901302,
|
783 |
+
"eval_rouge2": 0.3554282813944538,
|
784 |
+
"eval_rougeL": 0.5247953329477759,
|
785 |
+
"eval_rougeLsum": 0.6405524812915908,
|
786 |
+
"eval_runtime": 14.8892,
|
787 |
+
"eval_samples_per_second": 0.672,
|
788 |
+
"eval_steps_per_second": 0.336,
|
789 |
+
"step": 7800
|
790 |
+
},
|
791 |
+
{
|
792 |
+
"epoch": 0.91,
|
793 |
+
"grad_norm": 0.2703794538974762,
|
794 |
+
"learning_rate": 0.00015445352778648136,
|
795 |
+
"loss": 1.6415,
|
796 |
+
"step": 8000
|
797 |
+
},
|
798 |
+
{
|
799 |
+
"epoch": 0.91,
|
800 |
+
"eval_bertscore": 0.7325771450996399,
|
801 |
+
"eval_loss": 1.7271970510482788,
|
802 |
+
"eval_rouge1": 0.6659432894288253,
|
803 |
+
"eval_rouge2": 0.35962933912652617,
|
804 |
+
"eval_rougeL": 0.5385420432813512,
|
805 |
+
"eval_rougeLsum": 0.6557027031484046,
|
806 |
+
"eval_runtime": 14.8319,
|
807 |
+
"eval_samples_per_second": 0.674,
|
808 |
+
"eval_steps_per_second": 0.337,
|
809 |
+
"step": 8000
|
810 |
+
},
|
811 |
+
{
|
812 |
+
"epoch": 0.93,
|
813 |
+
"grad_norm": 0.28753793239593506,
|
814 |
+
"learning_rate": 0.0001533141538724471,
|
815 |
+
"loss": 1.6239,
|
816 |
+
"step": 8200
|
817 |
+
},
|
818 |
+
{
|
819 |
+
"epoch": 0.93,
|
820 |
+
"eval_bertscore": 0.7350013852119446,
|
821 |
+
"eval_loss": 1.7266199588775635,
|
822 |
+
"eval_rouge1": 0.6549282561414593,
|
823 |
+
"eval_rouge2": 0.35694530595734475,
|
824 |
+
"eval_rougeL": 0.5301601006964574,
|
825 |
+
"eval_rougeLsum": 0.6441779306137909,
|
826 |
+
"eval_runtime": 14.9005,
|
827 |
+
"eval_samples_per_second": 0.671,
|
828 |
+
"eval_steps_per_second": 0.336,
|
829 |
+
"step": 8200
|
830 |
+
},
|
831 |
+
{
|
832 |
+
"epoch": 0.96,
|
833 |
+
"grad_norm": 0.23870150744915009,
|
834 |
+
"learning_rate": 0.00015217477995841286,
|
835 |
+
"loss": 1.6293,
|
836 |
+
"step": 8400
|
837 |
+
},
|
838 |
+
{
|
839 |
+
"epoch": 0.96,
|
840 |
+
"eval_bertscore": 0.7271261811256409,
|
841 |
+
"eval_loss": 1.7256368398666382,
|
842 |
+
"eval_rouge1": 0.6515513829901936,
|
843 |
+
"eval_rouge2": 0.35217616104918836,
|
844 |
+
"eval_rougeL": 0.5236553509227138,
|
845 |
+
"eval_rougeLsum": 0.6411473505324752,
|
846 |
+
"eval_runtime": 14.9938,
|
847 |
+
"eval_samples_per_second": 0.667,
|
848 |
+
"eval_steps_per_second": 0.333,
|
849 |
+
"step": 8400
|
850 |
+
},
|
851 |
+
{
|
852 |
+
"epoch": 0.98,
|
853 |
+
"grad_norm": 0.28276997804641724,
|
854 |
+
"learning_rate": 0.00015103540604437861,
|
855 |
+
"loss": 1.6242,
|
856 |
+
"step": 8600
|
857 |
+
},
|
858 |
+
{
|
859 |
+
"epoch": 0.98,
|
860 |
+
"eval_bertscore": 0.7347334027290344,
|
861 |
+
"eval_loss": 1.717627763748169,
|
862 |
+
"eval_rouge1": 0.6350112495847634,
|
863 |
+
"eval_rouge2": 0.3477570751550898,
|
864 |
+
"eval_rougeL": 0.5146616989899861,
|
865 |
+
"eval_rougeLsum": 0.6246669376525157,
|
866 |
+
"eval_runtime": 15.7032,
|
867 |
+
"eval_samples_per_second": 0.637,
|
868 |
+
"eval_steps_per_second": 0.318,
|
869 |
+
"step": 8600
|
870 |
+
},
|
871 |
+
{
|
872 |
+
"epoch": 1.0,
|
873 |
+
"grad_norm": 0.24915842711925507,
|
874 |
+
"learning_rate": 0.00014989603213034437,
|
875 |
+
"loss": 1.6245,
|
876 |
+
"step": 8800
|
877 |
+
},
|
878 |
+
{
|
879 |
+
"epoch": 1.0,
|
880 |
+
"eval_bertscore": 0.7313701510429382,
|
881 |
+
"eval_loss": 1.7292964458465576,
|
882 |
+
"eval_rouge1": 0.6479528105367669,
|
883 |
+
"eval_rouge2": 0.35020983244262877,
|
884 |
+
"eval_rougeL": 0.5200907337780047,
|
885 |
+
"eval_rougeLsum": 0.6372896614836894,
|
886 |
+
"eval_runtime": 15.0043,
|
887 |
+
"eval_samples_per_second": 0.666,
|
888 |
+
"eval_steps_per_second": 0.333,
|
889 |
+
"step": 8800
|
890 |
+
},
|
891 |
+
{
|
892 |
+
"epoch": 1.03,
|
893 |
+
"grad_norm": 0.24036027491092682,
|
894 |
+
"learning_rate": 0.00014875665821631015,
|
895 |
+
"loss": 1.5364,
|
896 |
+
"step": 9000
|
897 |
+
},
|
898 |
+
{
|
899 |
+
"epoch": 1.03,
|
900 |
+
"eval_bertscore": 0.7327737808227539,
|
901 |
+
"eval_loss": 1.7339435815811157,
|
902 |
+
"eval_rouge1": 0.6568524178922349,
|
903 |
+
"eval_rouge2": 0.35560270713543163,
|
904 |
+
"eval_rougeL": 0.5310443670833082,
|
905 |
+
"eval_rougeLsum": 0.6480993679097387,
|
906 |
+
"eval_runtime": 14.9303,
|
907 |
+
"eval_samples_per_second": 0.67,
|
908 |
+
"eval_steps_per_second": 0.335,
|
909 |
+
"step": 9000
|
910 |
+
},
|
911 |
+
{
|
912 |
+
"epoch": 1.05,
|
913 |
+
"grad_norm": 0.2729027271270752,
|
914 |
+
"learning_rate": 0.0001476172843022759,
|
915 |
+
"loss": 1.5182,
|
916 |
+
"step": 9200
|
917 |
+
},
|
918 |
+
{
|
919 |
+
"epoch": 1.05,
|
920 |
+
"eval_bertscore": 0.7334672212600708,
|
921 |
+
"eval_loss": 1.739061713218689,
|
922 |
+
"eval_rouge1": 0.6552962187824329,
|
923 |
+
"eval_rouge2": 0.35210314124279196,
|
924 |
+
"eval_rougeL": 0.5272039052368354,
|
925 |
+
"eval_rougeLsum": 0.6437473533492806,
|
926 |
+
"eval_runtime": 15.5418,
|
927 |
+
"eval_samples_per_second": 0.643,
|
928 |
+
"eval_steps_per_second": 0.322,
|
929 |
+
"step": 9200
|
930 |
+
},
|
931 |
+
{
|
932 |
+
"epoch": 1.07,
|
933 |
+
"grad_norm": 0.2909716069698334,
|
934 |
+
"learning_rate": 0.00014647791038824168,
|
935 |
+
"loss": 1.5276,
|
936 |
+
"step": 9400
|
937 |
+
},
|
938 |
+
{
|
939 |
+
"epoch": 1.07,
|
940 |
+
"eval_bertscore": 0.7288902997970581,
|
941 |
+
"eval_loss": 1.736944556236267,
|
942 |
+
"eval_rouge1": 0.6533271685598301,
|
943 |
+
"eval_rouge2": 0.35279315321532184,
|
944 |
+
"eval_rougeL": 0.5262688234671329,
|
945 |
+
"eval_rougeLsum": 0.6424084937151033,
|
946 |
+
"eval_runtime": 15.1115,
|
947 |
+
"eval_samples_per_second": 0.662,
|
948 |
+
"eval_steps_per_second": 0.331,
|
949 |
+
"step": 9400
|
950 |
+
},
|
951 |
+
{
|
952 |
+
"epoch": 1.09,
|
953 |
+
"grad_norm": 0.3035859763622284,
|
954 |
+
"learning_rate": 0.00014533853647420743,
|
955 |
+
"loss": 1.5445,
|
956 |
+
"step": 9600
|
957 |
+
},
|
958 |
+
{
|
959 |
+
"epoch": 1.09,
|
960 |
+
"eval_bertscore": 0.7308284044265747,
|
961 |
+
"eval_loss": 1.737762689590454,
|
962 |
+
"eval_rouge1": 0.6619725777891359,
|
963 |
+
"eval_rouge2": 0.3611963714506864,
|
964 |
+
"eval_rougeL": 0.5363802967084452,
|
965 |
+
"eval_rougeLsum": 0.6516690557971352,
|
966 |
+
"eval_runtime": 14.9932,
|
967 |
+
"eval_samples_per_second": 0.667,
|
968 |
+
"eval_steps_per_second": 0.333,
|
969 |
+
"step": 9600
|
970 |
+
},
|
971 |
+
{
|
972 |
+
"epoch": 1.12,
|
973 |
+
"grad_norm": 0.26574915647506714,
|
974 |
+
"learning_rate": 0.0001441991625601732,
|
975 |
+
"loss": 1.5342,
|
976 |
+
"step": 9800
|
977 |
+
},
|
978 |
+
{
|
979 |
+
"epoch": 1.12,
|
980 |
+
"eval_bertscore": 0.7328712344169617,
|
981 |
+
"eval_loss": 1.7393991947174072,
|
982 |
+
"eval_rouge1": 0.6856504003396884,
|
983 |
+
"eval_rouge2": 0.3761098841062477,
|
984 |
+
"eval_rougeL": 0.555477293163325,
|
985 |
+
"eval_rougeLsum": 0.6757574283262289,
|
986 |
+
"eval_runtime": 14.8121,
|
987 |
+
"eval_samples_per_second": 0.675,
|
988 |
+
"eval_steps_per_second": 0.338,
|
989 |
+
"step": 9800
|
990 |
+
},
|
991 |
+
{
|
992 |
+
"epoch": 1.14,
|
993 |
+
"grad_norm": 0.315468430519104,
|
994 |
+
"learning_rate": 0.00014305978864613897,
|
995 |
+
"loss": 1.543,
|
996 |
+
"step": 10000
|
997 |
+
},
|
998 |
+
{
|
999 |
+
"epoch": 1.14,
|
1000 |
+
"eval_bertscore": 0.7349387407302856,
|
1001 |
+
"eval_loss": 1.7352710962295532,
|
1002 |
+
"eval_rouge1": 0.6749953128982036,
|
1003 |
+
"eval_rouge2": 0.3720385250530084,
|
1004 |
+
"eval_rougeL": 0.5472261566474382,
|
1005 |
+
"eval_rougeLsum": 0.6657643539219252,
|
1006 |
+
"eval_runtime": 14.909,
|
1007 |
+
"eval_samples_per_second": 0.671,
|
1008 |
+
"eval_steps_per_second": 0.335,
|
1009 |
+
"step": 10000
|
1010 |
+
},
|
1011 |
+
{
|
1012 |
+
"epoch": 1.16,
|
1013 |
+
"grad_norm": 0.29815027117729187,
|
1014 |
+
"learning_rate": 0.00014192041473210472,
|
1015 |
+
"loss": 1.5547,
|
1016 |
+
"step": 10200
|
1017 |
+
},
|
1018 |
+
{
|
1019 |
+
"epoch": 1.16,
|
1020 |
+
"eval_bertscore": 0.7359883189201355,
|
1021 |
+
"eval_loss": 1.7269136905670166,
|
1022 |
+
"eval_rouge1": 0.6561141614088863,
|
1023 |
+
"eval_rouge2": 0.3606175666303814,
|
1024 |
+
"eval_rougeL": 0.5302270771032793,
|
1025 |
+
"eval_rougeLsum": 0.6446912079521883,
|
1026 |
+
"eval_runtime": 14.9527,
|
1027 |
+
"eval_samples_per_second": 0.669,
|
1028 |
+
"eval_steps_per_second": 0.334,
|
1029 |
+
"step": 10200
|
1030 |
+
},
|
1031 |
+
{
|
1032 |
+
"epoch": 1.18,
|
1033 |
+
"grad_norm": 0.3595702350139618,
|
1034 |
+
"learning_rate": 0.00014078104081807047,
|
1035 |
+
"loss": 1.5567,
|
1036 |
+
"step": 10400
|
1037 |
+
},
|
1038 |
+
{
|
1039 |
+
"epoch": 1.18,
|
1040 |
+
"eval_bertscore": 0.7328116297721863,
|
1041 |
+
"eval_loss": 1.7341644763946533,
|
1042 |
+
"eval_rouge1": 0.6420332714823549,
|
1043 |
+
"eval_rouge2": 0.35094864549032,
|
1044 |
+
"eval_rougeL": 0.5179556761398367,
|
1045 |
+
"eval_rougeLsum": 0.631987397226809,
|
1046 |
+
"eval_runtime": 15.1438,
|
1047 |
+
"eval_samples_per_second": 0.66,
|
1048 |
+
"eval_steps_per_second": 0.33,
|
1049 |
+
"step": 10400
|
1050 |
+
},
|
1051 |
+
{
|
1052 |
+
"epoch": 1.21,
|
1053 |
+
"grad_norm": 0.2718666195869446,
|
1054 |
+
"learning_rate": 0.00013964166690403623,
|
1055 |
+
"loss": 1.5408,
|
1056 |
+
"step": 10600
|
1057 |
+
},
|
1058 |
+
{
|
1059 |
+
"epoch": 1.21,
|
1060 |
+
"eval_bertscore": 0.7337731122970581,
|
1061 |
+
"eval_loss": 1.7330901622772217,
|
1062 |
+
"eval_rouge1": 0.661681342484687,
|
1063 |
+
"eval_rouge2": 0.3626833509973693,
|
1064 |
+
"eval_rougeL": 0.5329424447373774,
|
1065 |
+
"eval_rougeLsum": 0.6519750177144633,
|
1066 |
+
"eval_runtime": 14.8142,
|
1067 |
+
"eval_samples_per_second": 0.675,
|
1068 |
+
"eval_steps_per_second": 0.338,
|
1069 |
+
"step": 10600
|
1070 |
+
},
|
1071 |
+
{
|
1072 |
+
"epoch": 1.23,
|
1073 |
+
"grad_norm": 0.29183274507522583,
|
1074 |
+
"learning_rate": 0.00013850229299000198,
|
1075 |
+
"loss": 1.5422,
|
1076 |
+
"step": 10800
|
1077 |
+
},
|
1078 |
+
{
|
1079 |
+
"epoch": 1.23,
|
1080 |
+
"eval_bertscore": 0.7331051230430603,
|
1081 |
+
"eval_loss": 1.7297636270523071,
|
1082 |
+
"eval_rouge1": 0.6655497978238063,
|
1083 |
+
"eval_rouge2": 0.3614235788441926,
|
1084 |
+
"eval_rougeL": 0.5327210061667442,
|
1085 |
+
"eval_rougeLsum": 0.6548836840483913,
|
1086 |
+
"eval_runtime": 15.1359,
|
1087 |
+
"eval_samples_per_second": 0.661,
|
1088 |
+
"eval_steps_per_second": 0.33,
|
1089 |
+
"step": 10800
|
1090 |
+
},
|
1091 |
+
{
|
1092 |
+
"epoch": 1.25,
|
1093 |
+
"grad_norm": 0.30979740619659424,
|
1094 |
+
"learning_rate": 0.00013736291907596776,
|
1095 |
+
"loss": 1.5372,
|
1096 |
+
"step": 11000
|
1097 |
+
},
|
1098 |
+
{
|
1099 |
+
"epoch": 1.25,
|
1100 |
+
"eval_bertscore": 0.7312799692153931,
|
1101 |
+
"eval_loss": 1.732444167137146,
|
1102 |
+
"eval_rouge1": 0.6568292865033993,
|
1103 |
+
"eval_rouge2": 0.35876682221562006,
|
1104 |
+
"eval_rougeL": 0.5300878844981931,
|
1105 |
+
"eval_rougeLsum": 0.6461751645858989,
|
1106 |
+
"eval_runtime": 14.819,
|
1107 |
+
"eval_samples_per_second": 0.675,
|
1108 |
+
"eval_steps_per_second": 0.337,
|
1109 |
+
"step": 11000
|
1110 |
+
},
|
1111 |
+
{
|
1112 |
+
"epoch": 1.28,
|
1113 |
+
"grad_norm": 0.31343138217926025,
|
1114 |
+
"learning_rate": 0.0001362235451619335,
|
1115 |
+
"loss": 1.5301,
|
1116 |
+
"step": 11200
|
1117 |
+
},
|
1118 |
+
{
|
1119 |
+
"epoch": 1.28,
|
1120 |
+
"eval_bertscore": 0.7317885160446167,
|
1121 |
+
"eval_loss": 1.7358499765396118,
|
1122 |
+
"eval_rouge1": 0.6548673097943329,
|
1123 |
+
"eval_rouge2": 0.3609116081432997,
|
1124 |
+
"eval_rougeL": 0.5279887650752133,
|
1125 |
+
"eval_rougeLsum": 0.6466232329097188,
|
1126 |
+
"eval_runtime": 14.8259,
|
1127 |
+
"eval_samples_per_second": 0.674,
|
1128 |
+
"eval_steps_per_second": 0.337,
|
1129 |
+
"step": 11200
|
1130 |
+
},
|
1131 |
+
{
|
1132 |
+
"epoch": 1.3,
|
1133 |
+
"grad_norm": 0.36181533336639404,
|
1134 |
+
"learning_rate": 0.0001350841712478993,
|
1135 |
+
"loss": 1.5421,
|
1136 |
+
"step": 11400
|
1137 |
+
},
|
1138 |
+
{
|
1139 |
+
"epoch": 1.3,
|
1140 |
+
"eval_bertscore": 0.7316756248474121,
|
1141 |
+
"eval_loss": 1.7282969951629639,
|
1142 |
+
"eval_rouge1": 0.6551882964480251,
|
1143 |
+
"eval_rouge2": 0.3580708921400697,
|
1144 |
+
"eval_rougeL": 0.5255367305995147,
|
1145 |
+
"eval_rougeLsum": 0.6449192953008009,
|
1146 |
+
"eval_runtime": 14.8816,
|
1147 |
+
"eval_samples_per_second": 0.672,
|
1148 |
+
"eval_steps_per_second": 0.336,
|
1149 |
+
"step": 11400
|
1150 |
+
},
|
1151 |
+
{
|
1152 |
+
"epoch": 1.32,
|
1153 |
+
"grad_norm": 0.30600836873054504,
|
1154 |
+
"learning_rate": 0.00013394479733386505,
|
1155 |
+
"loss": 1.5538,
|
1156 |
+
"step": 11600
|
1157 |
+
},
|
1158 |
+
{
|
1159 |
+
"epoch": 1.32,
|
1160 |
+
"eval_bertscore": 0.7311854362487793,
|
1161 |
+
"eval_loss": 1.7313562631607056,
|
1162 |
+
"eval_rouge1": 0.6592751199424156,
|
1163 |
+
"eval_rouge2": 0.35802855072854206,
|
1164 |
+
"eval_rougeL": 0.5297288176377084,
|
1165 |
+
"eval_rougeLsum": 0.6489455314962717,
|
1166 |
+
"eval_runtime": 15.0693,
|
1167 |
+
"eval_samples_per_second": 0.664,
|
1168 |
+
"eval_steps_per_second": 0.332,
|
1169 |
+
"step": 11600
|
1170 |
+
},
|
1171 |
+
{
|
1172 |
+
"epoch": 1.34,
|
1173 |
+
"grad_norm": 0.29904893040657043,
|
1174 |
+
"learning_rate": 0.0001328054234198308,
|
1175 |
+
"loss": 1.5328,
|
1176 |
+
"step": 11800
|
1177 |
+
},
|
1178 |
+
{
|
1179 |
+
"epoch": 1.34,
|
1180 |
+
"eval_bertscore": 0.7312635183334351,
|
1181 |
+
"eval_loss": 1.7318429946899414,
|
1182 |
+
"eval_rouge1": 0.6577169369077195,
|
1183 |
+
"eval_rouge2": 0.3582474830918887,
|
1184 |
+
"eval_rougeL": 0.5314990647771975,
|
1185 |
+
"eval_rougeLsum": 0.6454785220479866,
|
1186 |
+
"eval_runtime": 15.0235,
|
1187 |
+
"eval_samples_per_second": 0.666,
|
1188 |
+
"eval_steps_per_second": 0.333,
|
1189 |
+
"step": 11800
|
1190 |
+
},
|
1191 |
+
{
|
1192 |
+
"epoch": 1.37,
|
1193 |
+
"grad_norm": 0.3025416433811188,
|
1194 |
+
"learning_rate": 0.00013166604950579658,
|
1195 |
+
"loss": 1.5349,
|
1196 |
+
"step": 12000
|
1197 |
+
},
|
1198 |
+
{
|
1199 |
+
"epoch": 1.37,
|
1200 |
+
"eval_bertscore": 0.7325812578201294,
|
1201 |
+
"eval_loss": 1.7309118509292603,
|
1202 |
+
"eval_rouge1": 0.6629133074951261,
|
1203 |
+
"eval_rouge2": 0.3678158453940578,
|
1204 |
+
"eval_rougeL": 0.5380936907276155,
|
1205 |
+
"eval_rougeLsum": 0.654883061928214,
|
1206 |
+
"eval_runtime": 14.7666,
|
1207 |
+
"eval_samples_per_second": 0.677,
|
1208 |
+
"eval_steps_per_second": 0.339,
|
1209 |
+
"step": 12000
|
1210 |
+
},
|
1211 |
+
{
|
1212 |
+
"epoch": 1.39,
|
1213 |
+
"grad_norm": 0.34982389211654663,
|
1214 |
+
"learning_rate": 0.00013052667559176233,
|
1215 |
+
"loss": 1.5513,
|
1216 |
+
"step": 12200
|
1217 |
+
},
|
1218 |
+
{
|
1219 |
+
"epoch": 1.39,
|
1220 |
+
"eval_bertscore": 0.7340582609176636,
|
1221 |
+
"eval_loss": 1.7363474369049072,
|
1222 |
+
"eval_rouge1": 0.6555817937418287,
|
1223 |
+
"eval_rouge2": 0.35630500078358396,
|
1224 |
+
"eval_rougeL": 0.5272412353478366,
|
1225 |
+
"eval_rougeLsum": 0.6445837479327643,
|
1226 |
+
"eval_runtime": 14.9664,
|
1227 |
+
"eval_samples_per_second": 0.668,
|
1228 |
+
"eval_steps_per_second": 0.334,
|
1229 |
+
"step": 12200
|
1230 |
+
},
|
1231 |
+
{
|
1232 |
+
"epoch": 1.41,
|
1233 |
+
"grad_norm": 0.35809043049812317,
|
1234 |
+
"learning_rate": 0.0001293873016777281,
|
1235 |
+
"loss": 1.5444,
|
1236 |
+
"step": 12400
|
1237 |
+
},
|
1238 |
+
{
|
1239 |
+
"epoch": 1.41,
|
1240 |
+
"eval_bertscore": 0.7334069013595581,
|
1241 |
+
"eval_loss": 1.7328729629516602,
|
1242 |
+
"eval_rouge1": 0.6617971237669364,
|
1243 |
+
"eval_rouge2": 0.35951260376512423,
|
1244 |
+
"eval_rougeL": 0.5345512305507059,
|
1245 |
+
"eval_rougeLsum": 0.648363531132752,
|
1246 |
+
"eval_runtime": 15.2146,
|
1247 |
+
"eval_samples_per_second": 0.657,
|
1248 |
+
"eval_steps_per_second": 0.329,
|
1249 |
+
"step": 12400
|
1250 |
+
},
|
1251 |
+
{
|
1252 |
+
"epoch": 1.44,
|
1253 |
+
"grad_norm": 0.2954196631908417,
|
1254 |
+
"learning_rate": 0.00012824792776369387,
|
1255 |
+
"loss": 1.5406,
|
1256 |
+
"step": 12600
|
1257 |
+
},
|
1258 |
+
{
|
1259 |
+
"epoch": 1.44,
|
1260 |
+
"eval_bertscore": 0.7321678400039673,
|
1261 |
+
"eval_loss": 1.7335160970687866,
|
1262 |
+
"eval_rouge1": 0.6573625593086756,
|
1263 |
+
"eval_rouge2": 0.36210525247389347,
|
1264 |
+
"eval_rougeL": 0.5379361120230158,
|
1265 |
+
"eval_rougeLsum": 0.6459787883452857,
|
1266 |
+
"eval_runtime": 14.8942,
|
1267 |
+
"eval_samples_per_second": 0.671,
|
1268 |
+
"eval_steps_per_second": 0.336,
|
1269 |
+
"step": 12600
|
1270 |
+
},
|
1271 |
+
{
|
1272 |
+
"epoch": 1.46,
|
1273 |
+
"grad_norm": 0.32190588116645813,
|
1274 |
+
"learning_rate": 0.00012710855384965962,
|
1275 |
+
"loss": 1.5491,
|
1276 |
+
"step": 12800
|
1277 |
+
},
|
1278 |
+
{
|
1279 |
+
"epoch": 1.46,
|
1280 |
+
"eval_bertscore": 0.7346011400222778,
|
1281 |
+
"eval_loss": 1.7364966869354248,
|
1282 |
+
"eval_rouge1": 0.6481210247390559,
|
1283 |
+
"eval_rouge2": 0.3521173896017687,
|
1284 |
+
"eval_rougeL": 0.5240500581372636,
|
1285 |
+
"eval_rougeLsum": 0.63706442433335,
|
1286 |
+
"eval_runtime": 14.923,
|
1287 |
+
"eval_samples_per_second": 0.67,
|
1288 |
+
"eval_steps_per_second": 0.335,
|
1289 |
+
"step": 12800
|
1290 |
+
},
|
1291 |
+
{
|
1292 |
+
"epoch": 1.48,
|
1293 |
+
"grad_norm": 0.33323267102241516,
|
1294 |
+
"learning_rate": 0.00012596917993562537,
|
1295 |
+
"loss": 1.5596,
|
1296 |
+
"step": 13000
|
1297 |
+
},
|
1298 |
+
{
|
1299 |
+
"epoch": 1.48,
|
1300 |
+
"eval_bertscore": 0.7331587076187134,
|
1301 |
+
"eval_loss": 1.7332260608673096,
|
1302 |
+
"eval_rouge1": 0.6561257401878793,
|
1303 |
+
"eval_rouge2": 0.3548063723792664,
|
1304 |
+
"eval_rougeL": 0.527807776001489,
|
1305 |
+
"eval_rougeLsum": 0.6451911907984706,
|
1306 |
+
"eval_runtime": 15.4703,
|
1307 |
+
"eval_samples_per_second": 0.646,
|
1308 |
+
"eval_steps_per_second": 0.323,
|
1309 |
+
"step": 13000
|
1310 |
+
},
|
1311 |
+
{
|
1312 |
+
"epoch": 1.5,
|
1313 |
+
"grad_norm": 0.3564057946205139,
|
1314 |
+
"learning_rate": 0.00012482980602159113,
|
1315 |
+
"loss": 1.5261,
|
1316 |
+
"step": 13200
|
1317 |
+
},
|
1318 |
+
{
|
1319 |
+
"epoch": 1.5,
|
1320 |
+
"eval_bertscore": 0.7306063771247864,
|
1321 |
+
"eval_loss": 1.7368810176849365,
|
1322 |
+
"eval_rouge1": 0.637722723890071,
|
1323 |
+
"eval_rouge2": 0.3455358728458236,
|
1324 |
+
"eval_rougeL": 0.5136372690435154,
|
1325 |
+
"eval_rougeLsum": 0.6273570573595115,
|
1326 |
+
"eval_runtime": 15.3533,
|
1327 |
+
"eval_samples_per_second": 0.651,
|
1328 |
+
"eval_steps_per_second": 0.326,
|
1329 |
+
"step": 13200
|
1330 |
+
},
|
1331 |
+
{
|
1332 |
+
"epoch": 1.53,
|
1333 |
+
"grad_norm": 0.29219934344291687,
|
1334 |
+
"learning_rate": 0.0001236904321075569,
|
1335 |
+
"loss": 1.519,
|
1336 |
+
"step": 13400
|
1337 |
+
},
|
1338 |
+
{
|
1339 |
+
"epoch": 1.53,
|
1340 |
+
"eval_bertscore": 0.7338696122169495,
|
1341 |
+
"eval_loss": 1.734724998474121,
|
1342 |
+
"eval_rouge1": 0.6442107446420164,
|
1343 |
+
"eval_rouge2": 0.3494748457109431,
|
1344 |
+
"eval_rougeL": 0.5207483892007314,
|
1345 |
+
"eval_rougeLsum": 0.632886404907802,
|
1346 |
+
"eval_runtime": 15.3077,
|
1347 |
+
"eval_samples_per_second": 0.653,
|
1348 |
+
"eval_steps_per_second": 0.327,
|
1349 |
+
"step": 13400
|
1350 |
+
},
|
1351 |
+
{
|
1352 |
+
"epoch": 1.55,
|
1353 |
+
"grad_norm": 0.34681758284568787,
|
1354 |
+
"learning_rate": 0.00012255105819352266,
|
1355 |
+
"loss": 1.5419,
|
1356 |
+
"step": 13600
|
1357 |
+
},
|
1358 |
+
{
|
1359 |
+
"epoch": 1.55,
|
1360 |
+
"eval_bertscore": 0.7350045442581177,
|
1361 |
+
"eval_loss": 1.7329858541488647,
|
1362 |
+
"eval_rouge1": 0.6606839869796519,
|
1363 |
+
"eval_rouge2": 0.362188561160822,
|
1364 |
+
"eval_rougeL": 0.5342033818317451,
|
1365 |
+
"eval_rougeLsum": 0.6493340000068861,
|
1366 |
+
"eval_runtime": 15.44,
|
1367 |
+
"eval_samples_per_second": 0.648,
|
1368 |
+
"eval_steps_per_second": 0.324,
|
1369 |
+
"step": 13600
|
1370 |
+
},
|
1371 |
+
{
|
1372 |
+
"epoch": 1.57,
|
1373 |
+
"grad_norm": 0.3043666481971741,
|
1374 |
+
"learning_rate": 0.00012141168427948844,
|
1375 |
+
"loss": 1.5402,
|
1376 |
+
"step": 13800
|
1377 |
+
},
|
1378 |
+
{
|
1379 |
+
"epoch": 1.57,
|
1380 |
+
"eval_bertscore": 0.7363221645355225,
|
1381 |
+
"eval_loss": 1.7308530807495117,
|
1382 |
+
"eval_rouge1": 0.6638252384356028,
|
1383 |
+
"eval_rouge2": 0.3643237697892826,
|
1384 |
+
"eval_rougeL": 0.5403775887381331,
|
1385 |
+
"eval_rougeLsum": 0.6537260000827279,
|
1386 |
+
"eval_runtime": 14.7668,
|
1387 |
+
"eval_samples_per_second": 0.677,
|
1388 |
+
"eval_steps_per_second": 0.339,
|
1389 |
+
"step": 13800
|
1390 |
+
},
|
1391 |
+
{
|
1392 |
+
"epoch": 1.59,
|
1393 |
+
"grad_norm": 0.4073585867881775,
|
1394 |
+
"learning_rate": 0.00012027231036545419,
|
1395 |
+
"loss": 1.5256,
|
1396 |
+
"step": 14000
|
1397 |
+
},
|
1398 |
+
{
|
1399 |
+
"epoch": 1.59,
|
1400 |
+
"eval_bertscore": 0.7310279607772827,
|
1401 |
+
"eval_loss": 1.7326784133911133,
|
1402 |
+
"eval_rouge1": 0.6609594314120198,
|
1403 |
+
"eval_rouge2": 0.3601530714440473,
|
1404 |
+
"eval_rougeL": 0.5344452687135626,
|
1405 |
+
"eval_rougeLsum": 0.6480936554342305,
|
1406 |
+
"eval_runtime": 14.8565,
|
1407 |
+
"eval_samples_per_second": 0.673,
|
1408 |
+
"eval_steps_per_second": 0.337,
|
1409 |
+
"step": 14000
|
1410 |
+
},
|
1411 |
+
{
|
1412 |
+
"epoch": 1.62,
|
1413 |
+
"grad_norm": 0.3211813271045685,
|
1414 |
+
"learning_rate": 0.00011913293645141995,
|
1415 |
+
"loss": 1.5366,
|
1416 |
+
"step": 14200
|
1417 |
+
},
|
1418 |
+
{
|
1419 |
+
"epoch": 1.62,
|
1420 |
+
"eval_bertscore": 0.7356667518615723,
|
1421 |
+
"eval_loss": 1.7280094623565674,
|
1422 |
+
"eval_rouge1": 0.6519353227375031,
|
1423 |
+
"eval_rouge2": 0.3587025716186173,
|
1424 |
+
"eval_rougeL": 0.5306356200586075,
|
1425 |
+
"eval_rougeLsum": 0.6408870347994059,
|
1426 |
+
"eval_runtime": 14.9264,
|
1427 |
+
"eval_samples_per_second": 0.67,
|
1428 |
+
"eval_steps_per_second": 0.335,
|
1429 |
+
"step": 14200
|
1430 |
+
},
|
1431 |
+
{
|
1432 |
+
"epoch": 1.64,
|
1433 |
+
"grad_norm": 0.32776832580566406,
|
1434 |
+
"learning_rate": 0.00011799356253738571,
|
1435 |
+
"loss": 1.5504,
|
1436 |
+
"step": 14400
|
1437 |
+
},
|
1438 |
+
{
|
1439 |
+
"epoch": 1.64,
|
1440 |
+
"eval_bertscore": 0.7331353425979614,
|
1441 |
+
"eval_loss": 1.7308950424194336,
|
1442 |
+
"eval_rouge1": 0.6627702292814652,
|
1443 |
+
"eval_rouge2": 0.36117793957379707,
|
1444 |
+
"eval_rougeL": 0.5369305446079228,
|
1445 |
+
"eval_rougeLsum": 0.6516924083980089,
|
1446 |
+
"eval_runtime": 16.0138,
|
1447 |
+
"eval_samples_per_second": 0.624,
|
1448 |
+
"eval_steps_per_second": 0.312,
|
1449 |
+
"step": 14400
|
1450 |
+
},
|
1451 |
+
{
|
1452 |
+
"epoch": 1.66,
|
1453 |
+
"grad_norm": 0.3209726810455322,
|
1454 |
+
"learning_rate": 0.00011685418862335147,
|
1455 |
+
"loss": 1.5473,
|
1456 |
+
"step": 14600
|
1457 |
+
},
|
1458 |
+
{
|
1459 |
+
"epoch": 1.66,
|
1460 |
+
"eval_bertscore": 0.732498824596405,
|
1461 |
+
"eval_loss": 1.7328402996063232,
|
1462 |
+
"eval_rouge1": 0.6482679740803596,
|
1463 |
+
"eval_rouge2": 0.3538726087405498,
|
1464 |
+
"eval_rougeL": 0.5267677183598017,
|
1465 |
+
"eval_rougeLsum": 0.6366529460029322,
|
1466 |
+
"eval_runtime": 15.0195,
|
1467 |
+
"eval_samples_per_second": 0.666,
|
1468 |
+
"eval_steps_per_second": 0.333,
|
1469 |
+
"step": 14600
|
1470 |
+
},
|
1471 |
+
{
|
1472 |
+
"epoch": 1.69,
|
1473 |
+
"grad_norm": 0.3174591064453125,
|
1474 |
+
"learning_rate": 0.00011571481470931725,
|
1475 |
+
"loss": 1.5568,
|
1476 |
+
"step": 14800
|
1477 |
+
},
|
1478 |
+
{
|
1479 |
+
"epoch": 1.69,
|
1480 |
+
"eval_bertscore": 0.7335298657417297,
|
1481 |
+
"eval_loss": 1.7310253381729126,
|
1482 |
+
"eval_rouge1": 0.6560468577439627,
|
1483 |
+
"eval_rouge2": 0.36039371229175,
|
1484 |
+
"eval_rougeL": 0.5318708569729291,
|
1485 |
+
"eval_rougeLsum": 0.6444857558837042,
|
1486 |
+
"eval_runtime": 14.9774,
|
1487 |
+
"eval_samples_per_second": 0.668,
|
1488 |
+
"eval_steps_per_second": 0.334,
|
1489 |
+
"step": 14800
|
1490 |
+
},
|
1491 |
+
{
|
1492 |
+
"epoch": 1.71,
|
1493 |
+
"grad_norm": 0.2936408817768097,
|
1494 |
+
"learning_rate": 0.000114575440795283,
|
1495 |
+
"loss": 1.5345,
|
1496 |
+
"step": 15000
|
1497 |
+
},
|
1498 |
+
{
|
1499 |
+
"epoch": 1.71,
|
1500 |
+
"eval_bertscore": 0.7322725057601929,
|
1501 |
+
"eval_loss": 1.7270629405975342,
|
1502 |
+
"eval_rouge1": 0.6387060930656672,
|
1503 |
+
"eval_rouge2": 0.3480508127989137,
|
1504 |
+
"eval_rougeL": 0.5148670834213287,
|
1505 |
+
"eval_rougeLsum": 0.6273654952601909,
|
1506 |
+
"eval_runtime": 15.6687,
|
1507 |
+
"eval_samples_per_second": 0.638,
|
1508 |
+
"eval_steps_per_second": 0.319,
|
1509 |
+
"step": 15000
|
1510 |
+
},
|
1511 |
+
{
|
1512 |
+
"epoch": 1.73,
|
1513 |
+
"grad_norm": 0.32960689067840576,
|
1514 |
+
"learning_rate": 0.00011343606688124875,
|
1515 |
+
"loss": 1.5362,
|
1516 |
+
"step": 15200
|
1517 |
+
},
|
1518 |
+
{
|
1519 |
+
"epoch": 1.73,
|
1520 |
+
"eval_bertscore": 0.7337037920951843,
|
1521 |
+
"eval_loss": 1.7287395000457764,
|
1522 |
+
"eval_rouge1": 0.6476816970229771,
|
1523 |
+
"eval_rouge2": 0.3532248216683249,
|
1524 |
+
"eval_rougeL": 0.5253136618838716,
|
1525 |
+
"eval_rougeLsum": 0.6347493764394183,
|
1526 |
+
"eval_runtime": 15.0045,
|
1527 |
+
"eval_samples_per_second": 0.666,
|
1528 |
+
"eval_steps_per_second": 0.333,
|
1529 |
+
"step": 15200
|
1530 |
+
},
|
1531 |
+
{
|
1532 |
+
"epoch": 1.75,
|
1533 |
+
"grad_norm": 0.33265602588653564,
|
1534 |
+
"learning_rate": 0.00011229669296721452,
|
1535 |
+
"loss": 1.5215,
|
1536 |
+
"step": 15400
|
1537 |
+
},
|
1538 |
+
{
|
1539 |
+
"epoch": 1.75,
|
1540 |
+
"eval_bertscore": 0.7330806851387024,
|
1541 |
+
"eval_loss": 1.7265052795410156,
|
1542 |
+
"eval_rouge1": 0.6529393512177359,
|
1543 |
+
"eval_rouge2": 0.36182153145062224,
|
1544 |
+
"eval_rougeL": 0.5317061134915853,
|
1545 |
+
"eval_rougeLsum": 0.6413066299251913,
|
1546 |
+
"eval_runtime": 15.0256,
|
1547 |
+
"eval_samples_per_second": 0.666,
|
1548 |
+
"eval_steps_per_second": 0.333,
|
1549 |
+
"step": 15400
|
1550 |
+
},
|
1551 |
+
{
|
1552 |
+
"epoch": 1.78,
|
1553 |
+
"grad_norm": 0.3436201512813568,
|
1554 |
+
"learning_rate": 0.00011115731905318027,
|
1555 |
+
"loss": 1.539,
|
1556 |
+
"step": 15600
|
1557 |
+
},
|
1558 |
+
{
|
1559 |
+
"epoch": 1.78,
|
1560 |
+
"eval_bertscore": 0.7335551977157593,
|
1561 |
+
"eval_loss": 1.7254730463027954,
|
1562 |
+
"eval_rouge1": 0.6388518781767971,
|
1563 |
+
"eval_rouge2": 0.3501853846588857,
|
1564 |
+
"eval_rougeL": 0.5196828245794569,
|
1565 |
+
"eval_rougeLsum": 0.629333993884722,
|
1566 |
+
"eval_runtime": 15.2595,
|
1567 |
+
"eval_samples_per_second": 0.655,
|
1568 |
+
"eval_steps_per_second": 0.328,
|
1569 |
+
"step": 15600
|
1570 |
+
},
|
1571 |
+
{
|
1572 |
+
"epoch": 1.8,
|
1573 |
+
"grad_norm": 0.3428190350532532,
|
1574 |
+
"learning_rate": 0.00011001794513914605,
|
1575 |
+
"loss": 1.5273,
|
1576 |
+
"step": 15800
|
1577 |
+
},
|
1578 |
+
{
|
1579 |
+
"epoch": 1.8,
|
1580 |
+
"eval_bertscore": 0.7331770658493042,
|
1581 |
+
"eval_loss": 1.7286018133163452,
|
1582 |
+
"eval_rouge1": 0.6581941047310954,
|
1583 |
+
"eval_rouge2": 0.36277983926897583,
|
1584 |
+
"eval_rougeL": 0.5336464680120501,
|
1585 |
+
"eval_rougeLsum": 0.6489239720278894,
|
1586 |
+
"eval_runtime": 14.8252,
|
1587 |
+
"eval_samples_per_second": 0.675,
|
1588 |
+
"eval_steps_per_second": 0.337,
|
1589 |
+
"step": 15800
|
1590 |
+
},
|
1591 |
+
{
|
1592 |
+
"epoch": 1.82,
|
1593 |
+
"grad_norm": 0.363164484500885,
|
1594 |
+
"learning_rate": 0.0001088785712251118,
|
1595 |
+
"loss": 1.5445,
|
1596 |
+
"step": 16000
|
1597 |
+
},
|
1598 |
+
{
|
1599 |
+
"epoch": 1.82,
|
1600 |
+
"eval_bertscore": 0.7377282977104187,
|
1601 |
+
"eval_loss": 1.7363064289093018,
|
1602 |
+
"eval_rouge1": 0.6547011011872876,
|
1603 |
+
"eval_rouge2": 0.3553220826957326,
|
1604 |
+
"eval_rougeL": 0.5256073814411315,
|
1605 |
+
"eval_rougeLsum": 0.6420095316923398,
|
1606 |
+
"eval_runtime": 14.8599,
|
1607 |
+
"eval_samples_per_second": 0.673,
|
1608 |
+
"eval_steps_per_second": 0.336,
|
1609 |
+
"step": 16000
|
1610 |
+
},
|
1611 |
+
{
|
1612 |
+
"epoch": 1.85,
|
1613 |
+
"grad_norm": 0.3098333775997162,
|
1614 |
+
"learning_rate": 0.00010773919731107757,
|
1615 |
+
"loss": 1.5319,
|
1616 |
+
"step": 16200
|
1617 |
+
},
|
1618 |
+
{
|
1619 |
+
"epoch": 1.85,
|
1620 |
+
"eval_bertscore": 0.7324053645133972,
|
1621 |
+
"eval_loss": 1.7284066677093506,
|
1622 |
+
"eval_rouge1": 0.6477379941950916,
|
1623 |
+
"eval_rouge2": 0.3535918140554809,
|
1624 |
+
"eval_rougeL": 0.5226838544730126,
|
1625 |
+
"eval_rougeLsum": 0.6373271915355557,
|
1626 |
+
"eval_runtime": 14.9318,
|
1627 |
+
"eval_samples_per_second": 0.67,
|
1628 |
+
"eval_steps_per_second": 0.335,
|
1629 |
+
"step": 16200
|
1630 |
+
},
|
1631 |
+
{
|
1632 |
+
"epoch": 1.87,
|
1633 |
+
"grad_norm": 0.3637208938598633,
|
1634 |
+
"learning_rate": 0.00010659982339704332,
|
1635 |
+
"loss": 1.5442,
|
1636 |
+
"step": 16400
|
1637 |
+
},
|
1638 |
+
{
|
1639 |
+
"epoch": 1.87,
|
1640 |
+
"eval_bertscore": 0.7347462773323059,
|
1641 |
+
"eval_loss": 1.7252963781356812,
|
1642 |
+
"eval_rouge1": 0.6494449840449819,
|
1643 |
+
"eval_rouge2": 0.3586550050575282,
|
1644 |
+
"eval_rougeL": 0.5275675395159809,
|
1645 |
+
"eval_rougeLsum": 0.6396738714026391,
|
1646 |
+
"eval_runtime": 15.2218,
|
1647 |
+
"eval_samples_per_second": 0.657,
|
1648 |
+
"eval_steps_per_second": 0.328,
|
1649 |
+
"step": 16400
|
1650 |
+
},
|
1651 |
+
{
|
1652 |
+
"epoch": 1.89,
|
1653 |
+
"grad_norm": 0.35197457671165466,
|
1654 |
+
"learning_rate": 0.00010546044948300908,
|
1655 |
+
"loss": 1.5131,
|
1656 |
+
"step": 16600
|
1657 |
+
},
|
1658 |
+
{
|
1659 |
+
"epoch": 1.89,
|
1660 |
+
"eval_bertscore": 0.7329785227775574,
|
1661 |
+
"eval_loss": 1.7285687923431396,
|
1662 |
+
"eval_rouge1": 0.6582047811143328,
|
1663 |
+
"eval_rouge2": 0.3637700686094697,
|
1664 |
+
"eval_rougeL": 0.5355021948480279,
|
1665 |
+
"eval_rougeLsum": 0.6483245595148677,
|
1666 |
+
"eval_runtime": 14.8772,
|
1667 |
+
"eval_samples_per_second": 0.672,
|
1668 |
+
"eval_steps_per_second": 0.336,
|
1669 |
+
"step": 16600
|
1670 |
+
},
|
1671 |
+
{
|
1672 |
+
"epoch": 1.91,
|
1673 |
+
"grad_norm": 0.3406757116317749,
|
1674 |
+
"learning_rate": 0.00010432107556897486,
|
1675 |
+
"loss": 1.5394,
|
1676 |
+
"step": 16800
|
1677 |
+
},
|
1678 |
+
{
|
1679 |
+
"epoch": 1.91,
|
1680 |
+
"eval_bertscore": 0.7345961332321167,
|
1681 |
+
"eval_loss": 1.7324028015136719,
|
1682 |
+
"eval_rouge1": 0.6408293615351552,
|
1683 |
+
"eval_rouge2": 0.3520120690778129,
|
1684 |
+
"eval_rougeL": 0.5145218014745592,
|
1685 |
+
"eval_rougeLsum": 0.6297802607384266,
|
1686 |
+
"eval_runtime": 15.1044,
|
1687 |
+
"eval_samples_per_second": 0.662,
|
1688 |
+
"eval_steps_per_second": 0.331,
|
1689 |
+
"step": 16800
|
1690 |
+
},
|
1691 |
+
{
|
1692 |
+
"epoch": 1.94,
|
1693 |
+
"grad_norm": 0.3417683243751526,
|
1694 |
+
"learning_rate": 0.00010318170165494061,
|
1695 |
+
"loss": 1.526,
|
1696 |
+
"step": 17000
|
1697 |
+
},
|
1698 |
+
{
|
1699 |
+
"epoch": 1.94,
|
1700 |
+
"eval_bertscore": 0.735752522945404,
|
1701 |
+
"eval_loss": 1.7288110256195068,
|
1702 |
+
"eval_rouge1": 0.641158513352794,
|
1703 |
+
"eval_rouge2": 0.3544166440855814,
|
1704 |
+
"eval_rougeL": 0.5215201980495414,
|
1705 |
+
"eval_rougeLsum": 0.630550065494593,
|
1706 |
+
"eval_runtime": 15.0797,
|
1707 |
+
"eval_samples_per_second": 0.663,
|
1708 |
+
"eval_steps_per_second": 0.332,
|
1709 |
+
"step": 17000
|
1710 |
+
},
|
1711 |
+
{
|
1712 |
+
"epoch": 1.96,
|
1713 |
+
"grad_norm": 0.3256611227989197,
|
1714 |
+
"learning_rate": 0.00010204232774090639,
|
1715 |
+
"loss": 1.5484,
|
1716 |
+
"step": 17200
|
1717 |
+
},
|
1718 |
+
{
|
1719 |
+
"epoch": 1.96,
|
1720 |
+
"eval_bertscore": 0.7356327772140503,
|
1721 |
+
"eval_loss": 1.7305186986923218,
|
1722 |
+
"eval_rouge1": 0.6400269226515611,
|
1723 |
+
"eval_rouge2": 0.3502884634173268,
|
1724 |
+
"eval_rougeL": 0.517312321281175,
|
1725 |
+
"eval_rougeLsum": 0.6284556997614409,
|
1726 |
+
"eval_runtime": 15.4097,
|
1727 |
+
"eval_samples_per_second": 0.649,
|
1728 |
+
"eval_steps_per_second": 0.324,
|
1729 |
+
"step": 17200
|
1730 |
+
},
|
1731 |
+
{
|
1732 |
+
"epoch": 1.98,
|
1733 |
+
"grad_norm": 0.4035187363624573,
|
1734 |
+
"learning_rate": 0.00010090295382687213,
|
1735 |
+
"loss": 1.5261,
|
1736 |
+
"step": 17400
|
1737 |
+
},
|
1738 |
+
{
|
1739 |
+
"epoch": 1.98,
|
1740 |
+
"eval_bertscore": 0.7339992523193359,
|
1741 |
+
"eval_loss": 1.7282793521881104,
|
1742 |
+
"eval_rouge1": 0.6335770390416183,
|
1743 |
+
"eval_rouge2": 0.34592404578075897,
|
1744 |
+
"eval_rougeL": 0.5109045259792113,
|
1745 |
+
"eval_rougeLsum": 0.6218413683710426,
|
1746 |
+
"eval_runtime": 15.1959,
|
1747 |
+
"eval_samples_per_second": 0.658,
|
1748 |
+
"eval_steps_per_second": 0.329,
|
1749 |
+
"step": 17400
|
1750 |
+
},
|
1751 |
+
{
|
1752 |
+
"epoch": 2.0,
|
1753 |
+
"grad_norm": 0.34987062215805054,
|
1754 |
+
"learning_rate": 9.97635799128379e-05,
|
1755 |
+
"loss": 1.5199,
|
1756 |
+
"step": 17600
|
1757 |
+
},
|
1758 |
+
{
|
1759 |
+
"epoch": 2.0,
|
1760 |
+
"eval_bertscore": 0.7326329946517944,
|
1761 |
+
"eval_loss": 1.7544715404510498,
|
1762 |
+
"eval_rouge1": 0.6451558045750205,
|
1763 |
+
"eval_rouge2": 0.35565806935653943,
|
1764 |
+
"eval_rougeL": 0.5217034865840529,
|
1765 |
+
"eval_rougeLsum": 0.6329869715356753,
|
1766 |
+
"eval_runtime": 15.0351,
|
1767 |
+
"eval_samples_per_second": 0.665,
|
1768 |
+
"eval_steps_per_second": 0.333,
|
1769 |
+
"step": 17600
|
1770 |
+
},
|
1771 |
+
{
|
1772 |
+
"epoch": 2.03,
|
1773 |
+
"grad_norm": 0.37184038758277893,
|
1774 |
+
"learning_rate": 9.862420599880366e-05,
|
1775 |
+
"loss": 1.41,
|
1776 |
+
"step": 17800
|
1777 |
+
},
|
1778 |
+
{
|
1779 |
+
"epoch": 2.03,
|
1780 |
+
"eval_bertscore": 0.7315141558647156,
|
1781 |
+
"eval_loss": 1.7585878372192383,
|
1782 |
+
"eval_rouge1": 0.6469319193583706,
|
1783 |
+
"eval_rouge2": 0.3514447211469598,
|
1784 |
+
"eval_rougeL": 0.524755857688278,
|
1785 |
+
"eval_rougeLsum": 0.6350164781858667,
|
1786 |
+
"eval_runtime": 14.9583,
|
1787 |
+
"eval_samples_per_second": 0.669,
|
1788 |
+
"eval_steps_per_second": 0.334,
|
1789 |
+
"step": 17800
|
1790 |
+
},
|
1791 |
+
{
|
1792 |
+
"epoch": 2.05,
|
1793 |
+
"grad_norm": 0.3812776803970337,
|
1794 |
+
"learning_rate": 9.748483208476943e-05,
|
1795 |
+
"loss": 1.4132,
|
1796 |
+
"step": 18000
|
1797 |
+
},
|
1798 |
+
{
|
1799 |
+
"epoch": 2.05,
|
1800 |
+
"eval_bertscore": 0.7335561513900757,
|
1801 |
+
"eval_loss": 1.764611840248108,
|
1802 |
+
"eval_rouge1": 0.6381916780473581,
|
1803 |
+
"eval_rouge2": 0.3482510604092539,
|
1804 |
+
"eval_rougeL": 0.5162105225823392,
|
1805 |
+
"eval_rougeLsum": 0.627150245441782,
|
1806 |
+
"eval_runtime": 15.8515,
|
1807 |
+
"eval_samples_per_second": 0.631,
|
1808 |
+
"eval_steps_per_second": 0.315,
|
1809 |
+
"step": 18000
|
1810 |
+
},
|
1811 |
+
{
|
1812 |
+
"epoch": 2.07,
|
1813 |
+
"grad_norm": 0.45525220036506653,
|
1814 |
+
"learning_rate": 9.634545817073518e-05,
|
1815 |
+
"loss": 1.4,
|
1816 |
+
"step": 18200
|
1817 |
+
},
|
1818 |
+
{
|
1819 |
+
"epoch": 2.07,
|
1820 |
+
"eval_bertscore": 0.73627769947052,
|
1821 |
+
"eval_loss": 1.7585163116455078,
|
1822 |
+
"eval_rouge1": 0.6670097658027134,
|
1823 |
+
"eval_rouge2": 0.3658295359911405,
|
1824 |
+
"eval_rougeL": 0.5429667657900548,
|
1825 |
+
"eval_rougeLsum": 0.6543501745791419,
|
1826 |
+
"eval_runtime": 15.0301,
|
1827 |
+
"eval_samples_per_second": 0.665,
|
1828 |
+
"eval_steps_per_second": 0.333,
|
1829 |
+
"step": 18200
|
1830 |
+
},
|
1831 |
+
{
|
1832 |
+
"epoch": 2.1,
|
1833 |
+
"grad_norm": 0.37322184443473816,
|
1834 |
+
"learning_rate": 9.520608425670095e-05,
|
1835 |
+
"loss": 1.4293,
|
1836 |
+
"step": 18400
|
1837 |
+
},
|
1838 |
+
{
|
1839 |
+
"epoch": 2.1,
|
1840 |
+
"eval_bertscore": 0.730435848236084,
|
1841 |
+
"eval_loss": 1.764052391052246,
|
1842 |
+
"eval_rouge1": 0.6640215078213034,
|
1843 |
+
"eval_rouge2": 0.3625932287322054,
|
1844 |
+
"eval_rougeL": 0.5379978391335138,
|
1845 |
+
"eval_rougeLsum": 0.6542054656293199,
|
1846 |
+
"eval_runtime": 15.0762,
|
1847 |
+
"eval_samples_per_second": 0.663,
|
1848 |
+
"eval_steps_per_second": 0.332,
|
1849 |
+
"step": 18400
|
1850 |
+
},
|
1851 |
+
{
|
1852 |
+
"epoch": 2.12,
|
1853 |
+
"grad_norm": 0.4260891079902649,
|
1854 |
+
"learning_rate": 9.40667103426667e-05,
|
1855 |
+
"loss": 1.4077,
|
1856 |
+
"step": 18600
|
1857 |
+
},
|
1858 |
+
{
|
1859 |
+
"epoch": 2.12,
|
1860 |
+
"eval_bertscore": 0.7309869527816772,
|
1861 |
+
"eval_loss": 1.762108564376831,
|
1862 |
+
"eval_rouge1": 0.6571171081737958,
|
1863 |
+
"eval_rouge2": 0.35780421333141865,
|
1864 |
+
"eval_rougeL": 0.5320129270967632,
|
1865 |
+
"eval_rougeLsum": 0.64587787409523,
|
1866 |
+
"eval_runtime": 14.9004,
|
1867 |
+
"eval_samples_per_second": 0.671,
|
1868 |
+
"eval_steps_per_second": 0.336,
|
1869 |
+
"step": 18600
|
1870 |
+
},
|
1871 |
+
{
|
1872 |
+
"epoch": 2.14,
|
1873 |
+
"grad_norm": 0.39479926228523254,
|
1874 |
+
"learning_rate": 9.292733642863247e-05,
|
1875 |
+
"loss": 1.4165,
|
1876 |
+
"step": 18800
|
1877 |
+
},
|
1878 |
+
{
|
1879 |
+
"epoch": 2.14,
|
1880 |
+
"eval_bertscore": 0.7324444651603699,
|
1881 |
+
"eval_loss": 1.7607113122940063,
|
1882 |
+
"eval_rouge1": 0.6628398862884018,
|
1883 |
+
"eval_rouge2": 0.3627259806721216,
|
1884 |
+
"eval_rougeL": 0.5366106483832656,
|
1885 |
+
"eval_rougeLsum": 0.6528364858807157,
|
1886 |
+
"eval_runtime": 15.5766,
|
1887 |
+
"eval_samples_per_second": 0.642,
|
1888 |
+
"eval_steps_per_second": 0.321,
|
1889 |
+
"step": 18800
|
1890 |
+
},
|
1891 |
+
{
|
1892 |
+
"epoch": 2.16,
|
1893 |
+
"grad_norm": 0.39267703890800476,
|
1894 |
+
"learning_rate": 9.178796251459824e-05,
|
1895 |
+
"loss": 1.4123,
|
1896 |
+
"step": 19000
|
1897 |
+
},
|
1898 |
+
{
|
1899 |
+
"epoch": 2.16,
|
1900 |
+
"eval_bertscore": 0.7298994064331055,
|
1901 |
+
"eval_loss": 1.7668545246124268,
|
1902 |
+
"eval_rouge1": 0.6490850022857569,
|
1903 |
+
"eval_rouge2": 0.3532323419511264,
|
1904 |
+
"eval_rougeL": 0.5212823000193295,
|
1905 |
+
"eval_rougeLsum": 0.636442724466695,
|
1906 |
+
"eval_runtime": 14.9094,
|
1907 |
+
"eval_samples_per_second": 0.671,
|
1908 |
+
"eval_steps_per_second": 0.335,
|
1909 |
+
"step": 19000
|
1910 |
+
},
|
1911 |
+
{
|
1912 |
+
"epoch": 2.19,
|
1913 |
+
"grad_norm": 0.38221287727355957,
|
1914 |
+
"learning_rate": 9.0648588600564e-05,
|
1915 |
+
"loss": 1.401,
|
1916 |
+
"step": 19200
|
1917 |
+
},
|
1918 |
+
{
|
1919 |
+
"epoch": 2.19,
|
1920 |
+
"eval_bertscore": 0.7316875457763672,
|
1921 |
+
"eval_loss": 1.764147400856018,
|
1922 |
+
"eval_rouge1": 0.6490326710625849,
|
1923 |
+
"eval_rouge2": 0.3510351037900723,
|
1924 |
+
"eval_rougeL": 0.5239165028795836,
|
1925 |
+
"eval_rougeLsum": 0.6373687316421427,
|
1926 |
+
"eval_runtime": 15.1192,
|
1927 |
+
"eval_samples_per_second": 0.661,
|
1928 |
+
"eval_steps_per_second": 0.331,
|
1929 |
+
"step": 19200
|
1930 |
+
},
|
1931 |
+
{
|
1932 |
+
"epoch": 2.21,
|
1933 |
+
"grad_norm": 0.3653150200843811,
|
1934 |
+
"learning_rate": 8.950921468652976e-05,
|
1935 |
+
"loss": 1.4109,
|
1936 |
+
"step": 19400
|
1937 |
+
},
|
1938 |
+
{
|
1939 |
+
"epoch": 2.21,
|
1940 |
+
"eval_bertscore": 0.7348155975341797,
|
1941 |
+
"eval_loss": 1.7640550136566162,
|
1942 |
+
"eval_rouge1": 0.6462152873276823,
|
1943 |
+
"eval_rouge2": 0.3483599145461069,
|
1944 |
+
"eval_rougeL": 0.5193372430687719,
|
1945 |
+
"eval_rougeLsum": 0.6334254357511564,
|
1946 |
+
"eval_runtime": 14.9291,
|
1947 |
+
"eval_samples_per_second": 0.67,
|
1948 |
+
"eval_steps_per_second": 0.335,
|
1949 |
+
"step": 19400
|
1950 |
+
},
|
1951 |
+
{
|
1952 |
+
"epoch": 2.23,
|
1953 |
+
"grad_norm": 0.38049009442329407,
|
1954 |
+
"learning_rate": 8.836984077249551e-05,
|
1955 |
+
"loss": 1.4189,
|
1956 |
+
"step": 19600
|
1957 |
+
},
|
1958 |
+
{
|
1959 |
+
"epoch": 2.23,
|
1960 |
+
"eval_bertscore": 0.7357938885688782,
|
1961 |
+
"eval_loss": 1.7696326971054077,
|
1962 |
+
"eval_rouge1": 0.6377276221057538,
|
1963 |
+
"eval_rouge2": 0.3455397190390045,
|
1964 |
+
"eval_rougeL": 0.5118069428064842,
|
1965 |
+
"eval_rougeLsum": 0.6264501633078481,
|
1966 |
+
"eval_runtime": 14.8653,
|
1967 |
+
"eval_samples_per_second": 0.673,
|
1968 |
+
"eval_steps_per_second": 0.336,
|
1969 |
+
"step": 19600
|
1970 |
+
},
|
1971 |
+
{
|
1972 |
+
"epoch": 2.26,
|
1973 |
+
"grad_norm": 0.42111098766326904,
|
1974 |
+
"learning_rate": 8.723046685846128e-05,
|
1975 |
+
"loss": 1.4152,
|
1976 |
+
"step": 19800
|
1977 |
+
},
|
1978 |
+
{
|
1979 |
+
"epoch": 2.26,
|
1980 |
+
"eval_bertscore": 0.7339056134223938,
|
1981 |
+
"eval_loss": 1.7658218145370483,
|
1982 |
+
"eval_rouge1": 0.6494820372695989,
|
1983 |
+
"eval_rouge2": 0.34691658128805236,
|
1984 |
+
"eval_rougeL": 0.5193228965163086,
|
1985 |
+
"eval_rougeLsum": 0.6365347065687565,
|
1986 |
+
"eval_runtime": 15.3562,
|
1987 |
+
"eval_samples_per_second": 0.651,
|
1988 |
+
"eval_steps_per_second": 0.326,
|
1989 |
+
"step": 19800
|
1990 |
+
},
|
1991 |
+
{
|
1992 |
+
"epoch": 2.28,
|
1993 |
+
"grad_norm": 0.4452258050441742,
|
1994 |
+
"learning_rate": 8.609109294442704e-05,
|
1995 |
+
"loss": 1.4101,
|
1996 |
+
"step": 20000
|
1997 |
+
},
|
1998 |
+
{
|
1999 |
+
"epoch": 2.28,
|
2000 |
+
"eval_bertscore": 0.7296434640884399,
|
2001 |
+
"eval_loss": 1.7714240550994873,
|
2002 |
+
"eval_rouge1": 0.6565600824405751,
|
2003 |
+
"eval_rouge2": 0.3533618655201594,
|
2004 |
+
"eval_rougeL": 0.5263318202066467,
|
2005 |
+
"eval_rougeLsum": 0.6444964824298407,
|
2006 |
+
"eval_runtime": 14.8578,
|
2007 |
+
"eval_samples_per_second": 0.673,
|
2008 |
+
"eval_steps_per_second": 0.337,
|
2009 |
+
"step": 20000
|
2010 |
+
}
|
2011 |
+
],
|
2012 |
+
"logging_steps": 200,
|
2013 |
+
"max_steps": 35112,
|
2014 |
+
"num_input_tokens_seen": 0,
|
2015 |
+
"num_train_epochs": 4,
|
2016 |
+
"save_steps": 800,
|
2017 |
+
"total_flos": 2.0270434657179034e+18,
|
2018 |
+
"train_batch_size": 2,
|
2019 |
+
"trial_name": null,
|
2020 |
+
"trial_params": null
|
2021 |
+
}
|
checkpoint-20000/training_args.bin
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:d576743d4294cfbcae212243377437f73cabbcded2df9da7f73e6e39ecb38f56
|
3 |
+
size 5048
|
checkpoint-20800/README.md
ADDED
@@ -0,0 +1,202 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
---
|
2 |
+
library_name: peft
|
3 |
+
base_model: unsloth/gemma-2b-bnb-4bit
|
4 |
+
---
|
5 |
+
|
6 |
+
# Model Card for Model ID
|
7 |
+
|
8 |
+
<!-- Provide a quick summary of what the model is/does. -->
|
9 |
+
|
10 |
+
|
11 |
+
|
12 |
+
## Model Details
|
13 |
+
|
14 |
+
### Model Description
|
15 |
+
|
16 |
+
<!-- Provide a longer summary of what this model is. -->
|
17 |
+
|
18 |
+
|
19 |
+
|
20 |
+
- **Developed by:** [More Information Needed]
|
21 |
+
- **Funded by [optional]:** [More Information Needed]
|
22 |
+
- **Shared by [optional]:** [More Information Needed]
|
23 |
+
- **Model type:** [More Information Needed]
|
24 |
+
- **Language(s) (NLP):** [More Information Needed]
|
25 |
+
- **License:** [More Information Needed]
|
26 |
+
- **Finetuned from model [optional]:** [More Information Needed]
|
27 |
+
|
28 |
+
### Model Sources [optional]
|
29 |
+
|
30 |
+
<!-- Provide the basic links for the model. -->
|
31 |
+
|
32 |
+
- **Repository:** [More Information Needed]
|
33 |
+
- **Paper [optional]:** [More Information Needed]
|
34 |
+
- **Demo [optional]:** [More Information Needed]
|
35 |
+
|
36 |
+
## Uses
|
37 |
+
|
38 |
+
<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
|
39 |
+
|
40 |
+
### Direct Use
|
41 |
+
|
42 |
+
<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
|
43 |
+
|
44 |
+
[More Information Needed]
|
45 |
+
|
46 |
+
### Downstream Use [optional]
|
47 |
+
|
48 |
+
<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
|
49 |
+
|
50 |
+
[More Information Needed]
|
51 |
+
|
52 |
+
### Out-of-Scope Use
|
53 |
+
|
54 |
+
<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
|
55 |
+
|
56 |
+
[More Information Needed]
|
57 |
+
|
58 |
+
## Bias, Risks, and Limitations
|
59 |
+
|
60 |
+
<!-- This section is meant to convey both technical and sociotechnical limitations. -->
|
61 |
+
|
62 |
+
[More Information Needed]
|
63 |
+
|
64 |
+
### Recommendations
|
65 |
+
|
66 |
+
<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
|
67 |
+
|
68 |
+
Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
|
69 |
+
|
70 |
+
## How to Get Started with the Model
|
71 |
+
|
72 |
+
Use the code below to get started with the model.
|
73 |
+
|
74 |
+
[More Information Needed]
|
75 |
+
|
76 |
+
## Training Details
|
77 |
+
|
78 |
+
### Training Data
|
79 |
+
|
80 |
+
<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
|
81 |
+
|
82 |
+
[More Information Needed]
|
83 |
+
|
84 |
+
### Training Procedure
|
85 |
+
|
86 |
+
<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
|
87 |
+
|
88 |
+
#### Preprocessing [optional]
|
89 |
+
|
90 |
+
[More Information Needed]
|
91 |
+
|
92 |
+
|
93 |
+
#### Training Hyperparameters
|
94 |
+
|
95 |
+
- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
|
96 |
+
|
97 |
+
#### Speeds, Sizes, Times [optional]
|
98 |
+
|
99 |
+
<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
|
100 |
+
|
101 |
+
[More Information Needed]
|
102 |
+
|
103 |
+
## Evaluation
|
104 |
+
|
105 |
+
<!-- This section describes the evaluation protocols and provides the results. -->
|
106 |
+
|
107 |
+
### Testing Data, Factors & Metrics
|
108 |
+
|
109 |
+
#### Testing Data
|
110 |
+
|
111 |
+
<!-- This should link to a Dataset Card if possible. -->
|
112 |
+
|
113 |
+
[More Information Needed]
|
114 |
+
|
115 |
+
#### Factors
|
116 |
+
|
117 |
+
<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
|
118 |
+
|
119 |
+
[More Information Needed]
|
120 |
+
|
121 |
+
#### Metrics
|
122 |
+
|
123 |
+
<!-- These are the evaluation metrics being used, ideally with a description of why. -->
|
124 |
+
|
125 |
+
[More Information Needed]
|
126 |
+
|
127 |
+
### Results
|
128 |
+
|
129 |
+
[More Information Needed]
|
130 |
+
|
131 |
+
#### Summary
|
132 |
+
|
133 |
+
|
134 |
+
|
135 |
+
## Model Examination [optional]
|
136 |
+
|
137 |
+
<!-- Relevant interpretability work for the model goes here -->
|
138 |
+
|
139 |
+
[More Information Needed]
|
140 |
+
|
141 |
+
## Environmental Impact
|
142 |
+
|
143 |
+
<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
|
144 |
+
|
145 |
+
Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
|
146 |
+
|
147 |
+
- **Hardware Type:** [More Information Needed]
|
148 |
+
- **Hours used:** [More Information Needed]
|
149 |
+
- **Cloud Provider:** [More Information Needed]
|
150 |
+
- **Compute Region:** [More Information Needed]
|
151 |
+
- **Carbon Emitted:** [More Information Needed]
|
152 |
+
|
153 |
+
## Technical Specifications [optional]
|
154 |
+
|
155 |
+
### Model Architecture and Objective
|
156 |
+
|
157 |
+
[More Information Needed]
|
158 |
+
|
159 |
+
### Compute Infrastructure
|
160 |
+
|
161 |
+
[More Information Needed]
|
162 |
+
|
163 |
+
#### Hardware
|
164 |
+
|
165 |
+
[More Information Needed]
|
166 |
+
|
167 |
+
#### Software
|
168 |
+
|
169 |
+
[More Information Needed]
|
170 |
+
|
171 |
+
## Citation [optional]
|
172 |
+
|
173 |
+
<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
|
174 |
+
|
175 |
+
**BibTeX:**
|
176 |
+
|
177 |
+
[More Information Needed]
|
178 |
+
|
179 |
+
**APA:**
|
180 |
+
|
181 |
+
[More Information Needed]
|
182 |
+
|
183 |
+
## Glossary [optional]
|
184 |
+
|
185 |
+
<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
|
186 |
+
|
187 |
+
[More Information Needed]
|
188 |
+
|
189 |
+
## More Information [optional]
|
190 |
+
|
191 |
+
[More Information Needed]
|
192 |
+
|
193 |
+
## Model Card Authors [optional]
|
194 |
+
|
195 |
+
[More Information Needed]
|
196 |
+
|
197 |
+
## Model Card Contact
|
198 |
+
|
199 |
+
[More Information Needed]
|
200 |
+
### Framework versions
|
201 |
+
|
202 |
+
- PEFT 0.10.0
|
checkpoint-20800/adapter_config.json
ADDED
@@ -0,0 +1,34 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"alpha_pattern": {},
|
3 |
+
"auto_mapping": null,
|
4 |
+
"base_model_name_or_path": "unsloth/gemma-2b-bnb-4bit",
|
5 |
+
"bias": "none",
|
6 |
+
"fan_in_fan_out": false,
|
7 |
+
"inference_mode": true,
|
8 |
+
"init_lora_weights": true,
|
9 |
+
"layer_replication": null,
|
10 |
+
"layers_pattern": null,
|
11 |
+
"layers_to_transform": null,
|
12 |
+
"loftq_config": {},
|
13 |
+
"lora_alpha": 32,
|
14 |
+
"lora_dropout": 0,
|
15 |
+
"megatron_config": null,
|
16 |
+
"megatron_core": "megatron.core",
|
17 |
+
"modules_to_save": null,
|
18 |
+
"peft_type": "LORA",
|
19 |
+
"r": 64,
|
20 |
+
"rank_pattern": {},
|
21 |
+
"revision": "unsloth",
|
22 |
+
"target_modules": [
|
23 |
+
"v_proj",
|
24 |
+
"up_proj",
|
25 |
+
"gate_proj",
|
26 |
+
"o_proj",
|
27 |
+
"k_proj",
|
28 |
+
"down_proj",
|
29 |
+
"q_proj"
|
30 |
+
],
|
31 |
+
"task_type": "CAUSAL_LM",
|
32 |
+
"use_dora": false,
|
33 |
+
"use_rslora": false
|
34 |
+
}
|
checkpoint-20800/adapter_model.safetensors
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:8bbe4df1701bf265adab036c3e96cb82f9cb29bffaf953f7cc1cc5f43da0a96a
|
3 |
+
size 313820248
|
checkpoint-20800/rng_state.pth
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:769c9d116d716759c38c18535106ad97215ee71bcb7213d14aa01a6e2eeb4af3
|
3 |
+
size 14244
|
checkpoint-20800/scheduler.pt
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:c63a677b50085086e66ea55fbe01bc8715f3747ab6b695c5ef7d013aa1811797
|
3 |
+
size 1064
|
checkpoint-20800/special_tokens_map.json
ADDED
@@ -0,0 +1,34 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"additional_special_tokens": [
|
3 |
+
"<start_of_turn>",
|
4 |
+
"<end_of_turn>"
|
5 |
+
],
|
6 |
+
"bos_token": {
|
7 |
+
"content": "<bos>",
|
8 |
+
"lstrip": false,
|
9 |
+
"normalized": false,
|
10 |
+
"rstrip": false,
|
11 |
+
"single_word": false
|
12 |
+
},
|
13 |
+
"eos_token": {
|
14 |
+
"content": "<eos>",
|
15 |
+
"lstrip": false,
|
16 |
+
"normalized": false,
|
17 |
+
"rstrip": false,
|
18 |
+
"single_word": false
|
19 |
+
},
|
20 |
+
"pad_token": {
|
21 |
+
"content": "<pad>",
|
22 |
+
"lstrip": false,
|
23 |
+
"normalized": false,
|
24 |
+
"rstrip": false,
|
25 |
+
"single_word": false
|
26 |
+
},
|
27 |
+
"unk_token": {
|
28 |
+
"content": "<unk>",
|
29 |
+
"lstrip": false,
|
30 |
+
"normalized": false,
|
31 |
+
"rstrip": false,
|
32 |
+
"single_word": false
|
33 |
+
}
|
34 |
+
}
|
checkpoint-20800/tokenizer.json
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:05e97791a5e007260de1db7e1692e53150e08cea481e2bf25435553380c147ee
|
3 |
+
size 17477929
|
checkpoint-20800/tokenizer.model
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:61a7b147390c64585d6c3543dd6fc636906c9af3865a5548f27f31aee1d4c8e2
|
3 |
+
size 4241003
|
checkpoint-20800/tokenizer_config.json
ADDED
@@ -0,0 +1,71 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"add_bos_token": true,
|
3 |
+
"add_eos_token": false,
|
4 |
+
"added_tokens_decoder": {
|
5 |
+
"0": {
|
6 |
+
"content": "<pad>",
|
7 |
+
"lstrip": false,
|
8 |
+
"normalized": false,
|
9 |
+
"rstrip": false,
|
10 |
+
"single_word": false,
|
11 |
+
"special": true
|
12 |
+
},
|
13 |
+
"1": {
|
14 |
+
"content": "<eos>",
|
15 |
+
"lstrip": false,
|
16 |
+
"normalized": false,
|
17 |
+
"rstrip": false,
|
18 |
+
"single_word": false,
|
19 |
+
"special": true
|
20 |
+
},
|
21 |
+
"2": {
|
22 |
+
"content": "<bos>",
|
23 |
+
"lstrip": false,
|
24 |
+
"normalized": false,
|
25 |
+
"rstrip": false,
|
26 |
+
"single_word": false,
|
27 |
+
"special": true
|
28 |
+
},
|
29 |
+
"3": {
|
30 |
+
"content": "<unk>",
|
31 |
+
"lstrip": false,
|
32 |
+
"normalized": false,
|
33 |
+
"rstrip": false,
|
34 |
+
"single_word": false,
|
35 |
+
"special": true
|
36 |
+
},
|
37 |
+
"106": {
|
38 |
+
"content": "<start_of_turn>",
|
39 |
+
"lstrip": false,
|
40 |
+
"normalized": false,
|
41 |
+
"rstrip": false,
|
42 |
+
"single_word": false,
|
43 |
+
"special": true
|
44 |
+
},
|
45 |
+
"107": {
|
46 |
+
"content": "<end_of_turn>",
|
47 |
+
"lstrip": false,
|
48 |
+
"normalized": false,
|
49 |
+
"rstrip": false,
|
50 |
+
"single_word": false,
|
51 |
+
"special": true
|
52 |
+
}
|
53 |
+
},
|
54 |
+
"additional_special_tokens": [
|
55 |
+
"<start_of_turn>",
|
56 |
+
"<end_of_turn>"
|
57 |
+
],
|
58 |
+
"bos_token": "<bos>",
|
59 |
+
"chat_template": "{{ bos_token }}{% if messages[0]['role'] == 'system' %}{{ raise_exception('System role not supported') }}{% endif %}{% for message in messages %}{% if (message['role'] == 'user') != (loop.index0 % 2 == 0) %}{{ raise_exception('Conversation roles must alternate user/assistant/user/assistant/...') }}{% endif %}{% if (message['role'] == 'assistant') %}{% set role = 'model' %}{% else %}{% set role = message['role'] %}{% endif %}{{ '<start_of_turn>' + role + '\n' + message['content'] | trim + '<end_of_turn>\n' }}{% endfor %}{% if add_generation_prompt %}{{'<start_of_turn>model\n'}}{% endif %}",
|
60 |
+
"clean_up_tokenization_spaces": false,
|
61 |
+
"eos_token": "<eos>",
|
62 |
+
"legacy": null,
|
63 |
+
"model_max_length": 8192,
|
64 |
+
"pad_token": "<pad>",
|
65 |
+
"padding_side": "right",
|
66 |
+
"sp_model_kwargs": {},
|
67 |
+
"spaces_between_special_tokens": false,
|
68 |
+
"tokenizer_class": "GemmaTokenizer",
|
69 |
+
"unk_token": "<unk>",
|
70 |
+
"use_default_system_prompt": false
|
71 |
+
}
|
checkpoint-20800/trainer_state.json
ADDED
@@ -0,0 +1,2101 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"best_metric": null,
|
3 |
+
"best_model_checkpoint": null,
|
4 |
+
"epoch": 2.369425300449963,
|
5 |
+
"eval_steps": 200,
|
6 |
+
"global_step": 20800,
|
7 |
+
"is_hyper_param_search": false,
|
8 |
+
"is_local_process_zero": true,
|
9 |
+
"is_world_process_zero": true,
|
10 |
+
"log_history": [
|
11 |
+
{
|
12 |
+
"epoch": 0.02,
|
13 |
+
"grad_norm": 0.1643640249967575,
|
14 |
+
"learning_rate": 0.0001988891104338166,
|
15 |
+
"loss": 1.7673,
|
16 |
+
"step": 200
|
17 |
+
},
|
18 |
+
{
|
19 |
+
"epoch": 0.02,
|
20 |
+
"eval_bertscore": 0.7312520742416382,
|
21 |
+
"eval_loss": 1.7944419384002686,
|
22 |
+
"eval_rouge1": 0.645726048132668,
|
23 |
+
"eval_rouge2": 0.342840307585653,
|
24 |
+
"eval_rougeL": 0.5174784271125388,
|
25 |
+
"eval_rougeLsum": 0.6359911842715976,
|
26 |
+
"eval_runtime": 67.7968,
|
27 |
+
"eval_samples_per_second": 0.147,
|
28 |
+
"eval_steps_per_second": 0.074,
|
29 |
+
"step": 200
|
30 |
+
},
|
31 |
+
{
|
32 |
+
"epoch": 0.05,
|
33 |
+
"grad_norm": 0.17238478362560272,
|
34 |
+
"learning_rate": 0.00019774973651978238,
|
35 |
+
"loss": 1.6985,
|
36 |
+
"step": 400
|
37 |
+
},
|
38 |
+
{
|
39 |
+
"epoch": 0.05,
|
40 |
+
"eval_bertscore": 0.733666718006134,
|
41 |
+
"eval_loss": 1.7791178226470947,
|
42 |
+
"eval_rouge1": 0.6540909153028596,
|
43 |
+
"eval_rouge2": 0.3548819059818129,
|
44 |
+
"eval_rougeL": 0.527257232694246,
|
45 |
+
"eval_rougeLsum": 0.6442799950994005,
|
46 |
+
"eval_runtime": 15.1267,
|
47 |
+
"eval_samples_per_second": 0.661,
|
48 |
+
"eval_steps_per_second": 0.331,
|
49 |
+
"step": 400
|
50 |
+
},
|
51 |
+
{
|
52 |
+
"epoch": 0.07,
|
53 |
+
"grad_norm": 0.19368696212768555,
|
54 |
+
"learning_rate": 0.00019661036260574814,
|
55 |
+
"loss": 1.6962,
|
56 |
+
"step": 600
|
57 |
+
},
|
58 |
+
{
|
59 |
+
"epoch": 0.07,
|
60 |
+
"eval_bertscore": 0.7339462041854858,
|
61 |
+
"eval_loss": 1.7609882354736328,
|
62 |
+
"eval_rouge1": 0.6384337329686338,
|
63 |
+
"eval_rouge2": 0.3415514270662107,
|
64 |
+
"eval_rougeL": 0.51206080148464,
|
65 |
+
"eval_rougeLsum": 0.6261968614666548,
|
66 |
+
"eval_runtime": 15.2068,
|
67 |
+
"eval_samples_per_second": 0.658,
|
68 |
+
"eval_steps_per_second": 0.329,
|
69 |
+
"step": 600
|
70 |
+
},
|
71 |
+
{
|
72 |
+
"epoch": 0.09,
|
73 |
+
"grad_norm": 0.18629203736782074,
|
74 |
+
"learning_rate": 0.00019547098869171392,
|
75 |
+
"loss": 1.6825,
|
76 |
+
"step": 800
|
77 |
+
},
|
78 |
+
{
|
79 |
+
"epoch": 0.09,
|
80 |
+
"eval_bertscore": 0.7363594174385071,
|
81 |
+
"eval_loss": 1.7610784769058228,
|
82 |
+
"eval_rouge1": 0.6461624591922237,
|
83 |
+
"eval_rouge2": 0.3477371388439609,
|
84 |
+
"eval_rougeL": 0.5187429174752844,
|
85 |
+
"eval_rougeLsum": 0.6361089823008282,
|
86 |
+
"eval_runtime": 15.173,
|
87 |
+
"eval_samples_per_second": 0.659,
|
88 |
+
"eval_steps_per_second": 0.33,
|
89 |
+
"step": 800
|
90 |
+
},
|
91 |
+
{
|
92 |
+
"epoch": 0.11,
|
93 |
+
"grad_norm": 0.1799013316631317,
|
94 |
+
"learning_rate": 0.00019433161477767967,
|
95 |
+
"loss": 1.6848,
|
96 |
+
"step": 1000
|
97 |
+
},
|
98 |
+
{
|
99 |
+
"epoch": 0.11,
|
100 |
+
"eval_bertscore": 0.7334067225456238,
|
101 |
+
"eval_loss": 1.7576347589492798,
|
102 |
+
"eval_rouge1": 0.6345119236349537,
|
103 |
+
"eval_rouge2": 0.3422519149071803,
|
104 |
+
"eval_rougeL": 0.5111983101326238,
|
105 |
+
"eval_rougeLsum": 0.6244653120436832,
|
106 |
+
"eval_runtime": 15.2847,
|
107 |
+
"eval_samples_per_second": 0.654,
|
108 |
+
"eval_steps_per_second": 0.327,
|
109 |
+
"step": 1000
|
110 |
+
},
|
111 |
+
{
|
112 |
+
"epoch": 0.14,
|
113 |
+
"grad_norm": 0.22036150097846985,
|
114 |
+
"learning_rate": 0.00019319224086364545,
|
115 |
+
"loss": 1.6714,
|
116 |
+
"step": 1200
|
117 |
+
},
|
118 |
+
{
|
119 |
+
"epoch": 0.14,
|
120 |
+
"eval_bertscore": 0.7323788404464722,
|
121 |
+
"eval_loss": 1.7521806955337524,
|
122 |
+
"eval_rouge1": 0.6452540184557478,
|
123 |
+
"eval_rouge2": 0.3465145726476423,
|
124 |
+
"eval_rougeL": 0.516711757588783,
|
125 |
+
"eval_rougeLsum": 0.6341049885677059,
|
126 |
+
"eval_runtime": 15.1247,
|
127 |
+
"eval_samples_per_second": 0.661,
|
128 |
+
"eval_steps_per_second": 0.331,
|
129 |
+
"step": 1200
|
130 |
+
},
|
131 |
+
{
|
132 |
+
"epoch": 0.16,
|
133 |
+
"grad_norm": 0.21381086111068726,
|
134 |
+
"learning_rate": 0.0001920528669496112,
|
135 |
+
"loss": 1.6669,
|
136 |
+
"step": 1400
|
137 |
+
},
|
138 |
+
{
|
139 |
+
"epoch": 0.16,
|
140 |
+
"eval_bertscore": 0.7313202619552612,
|
141 |
+
"eval_loss": 1.7520482540130615,
|
142 |
+
"eval_rouge1": 0.6397526546254797,
|
143 |
+
"eval_rouge2": 0.3452671288110514,
|
144 |
+
"eval_rougeL": 0.5176580626678706,
|
145 |
+
"eval_rougeLsum": 0.6296746647539768,
|
146 |
+
"eval_runtime": 15.183,
|
147 |
+
"eval_samples_per_second": 0.659,
|
148 |
+
"eval_steps_per_second": 0.329,
|
149 |
+
"step": 1400
|
150 |
+
},
|
151 |
+
{
|
152 |
+
"epoch": 0.18,
|
153 |
+
"grad_norm": 0.20332874357700348,
|
154 |
+
"learning_rate": 0.00019091349303557696,
|
155 |
+
"loss": 1.671,
|
156 |
+
"step": 1600
|
157 |
+
},
|
158 |
+
{
|
159 |
+
"epoch": 0.18,
|
160 |
+
"eval_bertscore": 0.7349230647087097,
|
161 |
+
"eval_loss": 1.7473630905151367,
|
162 |
+
"eval_rouge1": 0.637439872504459,
|
163 |
+
"eval_rouge2": 0.34307164454056094,
|
164 |
+
"eval_rougeL": 0.5129717676228565,
|
165 |
+
"eval_rougeLsum": 0.6272190896182391,
|
166 |
+
"eval_runtime": 15.5672,
|
167 |
+
"eval_samples_per_second": 0.642,
|
168 |
+
"eval_steps_per_second": 0.321,
|
169 |
+
"step": 1600
|
170 |
+
},
|
171 |
+
{
|
172 |
+
"epoch": 0.21,
|
173 |
+
"grad_norm": 0.2025599479675293,
|
174 |
+
"learning_rate": 0.00018977411912154274,
|
175 |
+
"loss": 1.6721,
|
176 |
+
"step": 1800
|
177 |
+
},
|
178 |
+
{
|
179 |
+
"epoch": 0.21,
|
180 |
+
"eval_bertscore": 0.7357184290885925,
|
181 |
+
"eval_loss": 1.7516342401504517,
|
182 |
+
"eval_rouge1": 0.6387615819926658,
|
183 |
+
"eval_rouge2": 0.34366787517105574,
|
184 |
+
"eval_rougeL": 0.5129026911770751,
|
185 |
+
"eval_rougeLsum": 0.6289314118258257,
|
186 |
+
"eval_runtime": 15.9574,
|
187 |
+
"eval_samples_per_second": 0.627,
|
188 |
+
"eval_steps_per_second": 0.313,
|
189 |
+
"step": 1800
|
190 |
+
},
|
191 |
+
{
|
192 |
+
"epoch": 0.23,
|
193 |
+
"grad_norm": 0.20457112789154053,
|
194 |
+
"learning_rate": 0.0001886347452075085,
|
195 |
+
"loss": 1.671,
|
196 |
+
"step": 2000
|
197 |
+
},
|
198 |
+
{
|
199 |
+
"epoch": 0.23,
|
200 |
+
"eval_bertscore": 0.733718752861023,
|
201 |
+
"eval_loss": 1.7501707077026367,
|
202 |
+
"eval_rouge1": 0.6346207681220664,
|
203 |
+
"eval_rouge2": 0.33748369437614106,
|
204 |
+
"eval_rougeL": 0.5085159047705141,
|
205 |
+
"eval_rougeLsum": 0.6239953154441167,
|
206 |
+
"eval_runtime": 15.0863,
|
207 |
+
"eval_samples_per_second": 0.663,
|
208 |
+
"eval_steps_per_second": 0.331,
|
209 |
+
"step": 2000
|
210 |
+
},
|
211 |
+
{
|
212 |
+
"epoch": 0.25,
|
213 |
+
"grad_norm": 0.22552740573883057,
|
214 |
+
"learning_rate": 0.00018749537129347424,
|
215 |
+
"loss": 1.6496,
|
216 |
+
"step": 2200
|
217 |
+
},
|
218 |
+
{
|
219 |
+
"epoch": 0.25,
|
220 |
+
"eval_bertscore": 0.7368552684783936,
|
221 |
+
"eval_loss": 1.7437107563018799,
|
222 |
+
"eval_rouge1": 0.6490756387878311,
|
223 |
+
"eval_rouge2": 0.3448817738175175,
|
224 |
+
"eval_rougeL": 0.5235187045706321,
|
225 |
+
"eval_rougeLsum": 0.6377780857890332,
|
226 |
+
"eval_runtime": 15.07,
|
227 |
+
"eval_samples_per_second": 0.664,
|
228 |
+
"eval_steps_per_second": 0.332,
|
229 |
+
"step": 2200
|
230 |
+
},
|
231 |
+
{
|
232 |
+
"epoch": 0.27,
|
233 |
+
"grad_norm": 0.22573673725128174,
|
234 |
+
"learning_rate": 0.00018635599737944,
|
235 |
+
"loss": 1.6629,
|
236 |
+
"step": 2400
|
237 |
+
},
|
238 |
+
{
|
239 |
+
"epoch": 0.27,
|
240 |
+
"eval_bertscore": 0.7314499616622925,
|
241 |
+
"eval_loss": 1.7462828159332275,
|
242 |
+
"eval_rouge1": 0.6511482369678803,
|
243 |
+
"eval_rouge2": 0.34632544827771805,
|
244 |
+
"eval_rougeL": 0.5212417191003778,
|
245 |
+
"eval_rougeLsum": 0.6415391907940229,
|
246 |
+
"eval_runtime": 15.2078,
|
247 |
+
"eval_samples_per_second": 0.658,
|
248 |
+
"eval_steps_per_second": 0.329,
|
249 |
+
"step": 2400
|
250 |
+
},
|
251 |
+
{
|
252 |
+
"epoch": 0.3,
|
253 |
+
"grad_norm": 0.26426687836647034,
|
254 |
+
"learning_rate": 0.00018521662346540575,
|
255 |
+
"loss": 1.6644,
|
256 |
+
"step": 2600
|
257 |
+
},
|
258 |
+
{
|
259 |
+
"epoch": 0.3,
|
260 |
+
"eval_bertscore": 0.7363359928131104,
|
261 |
+
"eval_loss": 1.7505037784576416,
|
262 |
+
"eval_rouge1": 0.6498296552335481,
|
263 |
+
"eval_rouge2": 0.34873833589761183,
|
264 |
+
"eval_rougeL": 0.5194028620820592,
|
265 |
+
"eval_rougeLsum": 0.6404603087578984,
|
266 |
+
"eval_runtime": 14.8403,
|
267 |
+
"eval_samples_per_second": 0.674,
|
268 |
+
"eval_steps_per_second": 0.337,
|
269 |
+
"step": 2600
|
270 |
+
},
|
271 |
+
{
|
272 |
+
"epoch": 0.32,
|
273 |
+
"grad_norm": 0.20142091810703278,
|
274 |
+
"learning_rate": 0.00018407724955137153,
|
275 |
+
"loss": 1.6535,
|
276 |
+
"step": 2800
|
277 |
+
},
|
278 |
+
{
|
279 |
+
"epoch": 0.32,
|
280 |
+
"eval_bertscore": 0.7304679155349731,
|
281 |
+
"eval_loss": 1.7511039972305298,
|
282 |
+
"eval_rouge1": 0.6475130585388738,
|
283 |
+
"eval_rouge2": 0.34648331046897884,
|
284 |
+
"eval_rougeL": 0.5218042284020985,
|
285 |
+
"eval_rougeLsum": 0.6382749834402862,
|
286 |
+
"eval_runtime": 15.0162,
|
287 |
+
"eval_samples_per_second": 0.666,
|
288 |
+
"eval_steps_per_second": 0.333,
|
289 |
+
"step": 2800
|
290 |
+
},
|
291 |
+
{
|
292 |
+
"epoch": 0.34,
|
293 |
+
"grad_norm": 0.23283220827579498,
|
294 |
+
"learning_rate": 0.00018293787563733728,
|
295 |
+
"loss": 1.6477,
|
296 |
+
"step": 3000
|
297 |
+
},
|
298 |
+
{
|
299 |
+
"epoch": 0.34,
|
300 |
+
"eval_bertscore": 0.7327049374580383,
|
301 |
+
"eval_loss": 1.7461665868759155,
|
302 |
+
"eval_rouge1": 0.6309349586871908,
|
303 |
+
"eval_rouge2": 0.3387882478990309,
|
304 |
+
"eval_rougeL": 0.5042059192403674,
|
305 |
+
"eval_rougeLsum": 0.6210432469674847,
|
306 |
+
"eval_runtime": 15.463,
|
307 |
+
"eval_samples_per_second": 0.647,
|
308 |
+
"eval_steps_per_second": 0.323,
|
309 |
+
"step": 3000
|
310 |
+
},
|
311 |
+
{
|
312 |
+
"epoch": 0.36,
|
313 |
+
"grad_norm": 0.21316750347614288,
|
314 |
+
"learning_rate": 0.00018179850172330306,
|
315 |
+
"loss": 1.6614,
|
316 |
+
"step": 3200
|
317 |
+
},
|
318 |
+
{
|
319 |
+
"epoch": 0.36,
|
320 |
+
"eval_bertscore": 0.7314620018005371,
|
321 |
+
"eval_loss": 1.7468239068984985,
|
322 |
+
"eval_rouge1": 0.6480904534152265,
|
323 |
+
"eval_rouge2": 0.3479530168963481,
|
324 |
+
"eval_rougeL": 0.5193148273848067,
|
325 |
+
"eval_rougeLsum": 0.6366010767207634,
|
326 |
+
"eval_runtime": 15.0381,
|
327 |
+
"eval_samples_per_second": 0.665,
|
328 |
+
"eval_steps_per_second": 0.332,
|
329 |
+
"step": 3200
|
330 |
+
},
|
331 |
+
{
|
332 |
+
"epoch": 0.39,
|
333 |
+
"grad_norm": 0.26080408692359924,
|
334 |
+
"learning_rate": 0.00018065912780926882,
|
335 |
+
"loss": 1.6591,
|
336 |
+
"step": 3400
|
337 |
+
},
|
338 |
+
{
|
339 |
+
"epoch": 0.39,
|
340 |
+
"eval_bertscore": 0.7327477335929871,
|
341 |
+
"eval_loss": 1.7442594766616821,
|
342 |
+
"eval_rouge1": 0.6424613378037144,
|
343 |
+
"eval_rouge2": 0.34731770322974903,
|
344 |
+
"eval_rougeL": 0.5160705879794565,
|
345 |
+
"eval_rougeLsum": 0.6327006420281607,
|
346 |
+
"eval_runtime": 15.5373,
|
347 |
+
"eval_samples_per_second": 0.644,
|
348 |
+
"eval_steps_per_second": 0.322,
|
349 |
+
"step": 3400
|
350 |
+
},
|
351 |
+
{
|
352 |
+
"epoch": 0.41,
|
353 |
+
"grad_norm": 0.23274216055870056,
|
354 |
+
"learning_rate": 0.0001795197538952346,
|
355 |
+
"loss": 1.6613,
|
356 |
+
"step": 3600
|
357 |
+
},
|
358 |
+
{
|
359 |
+
"epoch": 0.41,
|
360 |
+
"eval_bertscore": 0.736956000328064,
|
361 |
+
"eval_loss": 1.7429373264312744,
|
362 |
+
"eval_rouge1": 0.6514574160666677,
|
363 |
+
"eval_rouge2": 0.3556199242231646,
|
364 |
+
"eval_rougeL": 0.5249726237675663,
|
365 |
+
"eval_rougeLsum": 0.6406261097623661,
|
366 |
+
"eval_runtime": 14.9415,
|
367 |
+
"eval_samples_per_second": 0.669,
|
368 |
+
"eval_steps_per_second": 0.335,
|
369 |
+
"step": 3600
|
370 |
+
},
|
371 |
+
{
|
372 |
+
"epoch": 0.43,
|
373 |
+
"grad_norm": 0.23616766929626465,
|
374 |
+
"learning_rate": 0.00017838037998120035,
|
375 |
+
"loss": 1.6479,
|
376 |
+
"step": 3800
|
377 |
+
},
|
378 |
+
{
|
379 |
+
"epoch": 0.43,
|
380 |
+
"eval_bertscore": 0.7349627614021301,
|
381 |
+
"eval_loss": 1.7420669794082642,
|
382 |
+
"eval_rouge1": 0.655851684949526,
|
383 |
+
"eval_rouge2": 0.35254590691865084,
|
384 |
+
"eval_rougeL": 0.5248980956621441,
|
385 |
+
"eval_rougeLsum": 0.6449637270581419,
|
386 |
+
"eval_runtime": 15.4178,
|
387 |
+
"eval_samples_per_second": 0.649,
|
388 |
+
"eval_steps_per_second": 0.324,
|
389 |
+
"step": 3800
|
390 |
+
},
|
391 |
+
{
|
392 |
+
"epoch": 0.46,
|
393 |
+
"grad_norm": 0.23260319232940674,
|
394 |
+
"learning_rate": 0.0001772410060671661,
|
395 |
+
"loss": 1.6569,
|
396 |
+
"step": 4000
|
397 |
+
},
|
398 |
+
{
|
399 |
+
"epoch": 0.46,
|
400 |
+
"eval_bertscore": 0.7332885265350342,
|
401 |
+
"eval_loss": 1.7401313781738281,
|
402 |
+
"eval_rouge1": 0.6669483634140105,
|
403 |
+
"eval_rouge2": 0.35873988835161297,
|
404 |
+
"eval_rougeL": 0.5343868725007427,
|
405 |
+
"eval_rougeLsum": 0.6555353134690931,
|
406 |
+
"eval_runtime": 15.0822,
|
407 |
+
"eval_samples_per_second": 0.663,
|
408 |
+
"eval_steps_per_second": 0.332,
|
409 |
+
"step": 4000
|
410 |
+
},
|
411 |
+
{
|
412 |
+
"epoch": 0.48,
|
413 |
+
"grad_norm": 0.2366473525762558,
|
414 |
+
"learning_rate": 0.00017610163215313186,
|
415 |
+
"loss": 1.6599,
|
416 |
+
"step": 4200
|
417 |
+
},
|
418 |
+
{
|
419 |
+
"epoch": 0.48,
|
420 |
+
"eval_bertscore": 0.7335314750671387,
|
421 |
+
"eval_loss": 1.7385823726654053,
|
422 |
+
"eval_rouge1": 0.6559297063578133,
|
423 |
+
"eval_rouge2": 0.35483499789990636,
|
424 |
+
"eval_rougeL": 0.5297939800986089,
|
425 |
+
"eval_rougeLsum": 0.6454544372491222,
|
426 |
+
"eval_runtime": 15.1029,
|
427 |
+
"eval_samples_per_second": 0.662,
|
428 |
+
"eval_steps_per_second": 0.331,
|
429 |
+
"step": 4200
|
430 |
+
},
|
431 |
+
{
|
432 |
+
"epoch": 0.5,
|
433 |
+
"grad_norm": 0.20628753304481506,
|
434 |
+
"learning_rate": 0.0001749622582390976,
|
435 |
+
"loss": 1.6454,
|
436 |
+
"step": 4400
|
437 |
+
},
|
438 |
+
{
|
439 |
+
"epoch": 0.5,
|
440 |
+
"eval_bertscore": 0.7342795133590698,
|
441 |
+
"eval_loss": 1.7422058582305908,
|
442 |
+
"eval_rouge1": 0.660746519614568,
|
443 |
+
"eval_rouge2": 0.3633965895561597,
|
444 |
+
"eval_rougeL": 0.5369036980876734,
|
445 |
+
"eval_rougeLsum": 0.650338328328998,
|
446 |
+
"eval_runtime": 15.4434,
|
447 |
+
"eval_samples_per_second": 0.648,
|
448 |
+
"eval_steps_per_second": 0.324,
|
449 |
+
"step": 4400
|
450 |
+
},
|
451 |
+
{
|
452 |
+
"epoch": 0.52,
|
453 |
+
"grad_norm": 0.2239149957895279,
|
454 |
+
"learning_rate": 0.0001738228843250634,
|
455 |
+
"loss": 1.6594,
|
456 |
+
"step": 4600
|
457 |
+
},
|
458 |
+
{
|
459 |
+
"epoch": 0.52,
|
460 |
+
"eval_bertscore": 0.7313543558120728,
|
461 |
+
"eval_loss": 1.740854263305664,
|
462 |
+
"eval_rouge1": 0.6591645132619427,
|
463 |
+
"eval_rouge2": 0.35766117432431743,
|
464 |
+
"eval_rougeL": 0.532710255034635,
|
465 |
+
"eval_rougeLsum": 0.6479428185884644,
|
466 |
+
"eval_runtime": 14.9436,
|
467 |
+
"eval_samples_per_second": 0.669,
|
468 |
+
"eval_steps_per_second": 0.335,
|
469 |
+
"step": 4600
|
470 |
+
},
|
471 |
+
{
|
472 |
+
"epoch": 0.55,
|
473 |
+
"grad_norm": 0.24808338284492493,
|
474 |
+
"learning_rate": 0.00017268351041102914,
|
475 |
+
"loss": 1.6604,
|
476 |
+
"step": 4800
|
477 |
+
},
|
478 |
+
{
|
479 |
+
"epoch": 0.55,
|
480 |
+
"eval_bertscore": 0.7333321571350098,
|
481 |
+
"eval_loss": 1.7385585308074951,
|
482 |
+
"eval_rouge1": 0.6532115808232871,
|
483 |
+
"eval_rouge2": 0.35333788022501567,
|
484 |
+
"eval_rougeL": 0.5284071547874328,
|
485 |
+
"eval_rougeLsum": 0.6410472452797623,
|
486 |
+
"eval_runtime": 15.0277,
|
487 |
+
"eval_samples_per_second": 0.665,
|
488 |
+
"eval_steps_per_second": 0.333,
|
489 |
+
"step": 4800
|
490 |
+
},
|
491 |
+
{
|
492 |
+
"epoch": 0.57,
|
493 |
+
"grad_norm": 0.2555364966392517,
|
494 |
+
"learning_rate": 0.0001715441364969949,
|
495 |
+
"loss": 1.6493,
|
496 |
+
"step": 5000
|
497 |
+
},
|
498 |
+
{
|
499 |
+
"epoch": 0.57,
|
500 |
+
"eval_bertscore": 0.7318152189254761,
|
501 |
+
"eval_loss": 1.7357494831085205,
|
502 |
+
"eval_rouge1": 0.6476755890502586,
|
503 |
+
"eval_rouge2": 0.35312778275949164,
|
504 |
+
"eval_rougeL": 0.5227601228049905,
|
505 |
+
"eval_rougeLsum": 0.6371331138372852,
|
506 |
+
"eval_runtime": 14.8807,
|
507 |
+
"eval_samples_per_second": 0.672,
|
508 |
+
"eval_steps_per_second": 0.336,
|
509 |
+
"step": 5000
|
510 |
+
},
|
511 |
+
{
|
512 |
+
"epoch": 0.59,
|
513 |
+
"grad_norm": 0.21518155932426453,
|
514 |
+
"learning_rate": 0.00017040476258296068,
|
515 |
+
"loss": 1.644,
|
516 |
+
"step": 5200
|
517 |
+
},
|
518 |
+
{
|
519 |
+
"epoch": 0.59,
|
520 |
+
"eval_bertscore": 0.734805703163147,
|
521 |
+
"eval_loss": 1.74032723903656,
|
522 |
+
"eval_rouge1": 0.6476813733451636,
|
523 |
+
"eval_rouge2": 0.3509259728617576,
|
524 |
+
"eval_rougeL": 0.5221334872800274,
|
525 |
+
"eval_rougeLsum": 0.636892384667733,
|
526 |
+
"eval_runtime": 15.1271,
|
527 |
+
"eval_samples_per_second": 0.661,
|
528 |
+
"eval_steps_per_second": 0.331,
|
529 |
+
"step": 5200
|
530 |
+
},
|
531 |
+
{
|
532 |
+
"epoch": 0.62,
|
533 |
+
"grad_norm": 0.26086658239364624,
|
534 |
+
"learning_rate": 0.00016926538866892643,
|
535 |
+
"loss": 1.6449,
|
536 |
+
"step": 5400
|
537 |
+
},
|
538 |
+
{
|
539 |
+
"epoch": 0.62,
|
540 |
+
"eval_bertscore": 0.7339995503425598,
|
541 |
+
"eval_loss": 1.7338205575942993,
|
542 |
+
"eval_rouge1": 0.6416889902864902,
|
543 |
+
"eval_rouge2": 0.3479045880347737,
|
544 |
+
"eval_rougeL": 0.5160577838468976,
|
545 |
+
"eval_rougeLsum": 0.6317983411796093,
|
546 |
+
"eval_runtime": 14.9992,
|
547 |
+
"eval_samples_per_second": 0.667,
|
548 |
+
"eval_steps_per_second": 0.333,
|
549 |
+
"step": 5400
|
550 |
+
},
|
551 |
+
{
|
552 |
+
"epoch": 0.64,
|
553 |
+
"grad_norm": 0.25449469685554504,
|
554 |
+
"learning_rate": 0.0001681260147548922,
|
555 |
+
"loss": 1.6299,
|
556 |
+
"step": 5600
|
557 |
+
},
|
558 |
+
{
|
559 |
+
"epoch": 0.64,
|
560 |
+
"eval_bertscore": 0.7306328415870667,
|
561 |
+
"eval_loss": 1.7369228601455688,
|
562 |
+
"eval_rouge1": 0.6390760905985684,
|
563 |
+
"eval_rouge2": 0.3409328272828699,
|
564 |
+
"eval_rougeL": 0.5111832543685331,
|
565 |
+
"eval_rougeLsum": 0.6285753423407665,
|
566 |
+
"eval_runtime": 15.483,
|
567 |
+
"eval_samples_per_second": 0.646,
|
568 |
+
"eval_steps_per_second": 0.323,
|
569 |
+
"step": 5600
|
570 |
+
},
|
571 |
+
{
|
572 |
+
"epoch": 0.66,
|
573 |
+
"grad_norm": 0.24706102907657623,
|
574 |
+
"learning_rate": 0.00016698664084085796,
|
575 |
+
"loss": 1.6374,
|
576 |
+
"step": 5800
|
577 |
+
},
|
578 |
+
{
|
579 |
+
"epoch": 0.66,
|
580 |
+
"eval_bertscore": 0.732075572013855,
|
581 |
+
"eval_loss": 1.7343876361846924,
|
582 |
+
"eval_rouge1": 0.6378821977913272,
|
583 |
+
"eval_rouge2": 0.34619427775171585,
|
584 |
+
"eval_rougeL": 0.5120186953041237,
|
585 |
+
"eval_rougeLsum": 0.6284056323839109,
|
586 |
+
"eval_runtime": 15.7341,
|
587 |
+
"eval_samples_per_second": 0.636,
|
588 |
+
"eval_steps_per_second": 0.318,
|
589 |
+
"step": 5800
|
590 |
+
},
|
591 |
+
{
|
592 |
+
"epoch": 0.68,
|
593 |
+
"grad_norm": 0.24373260140419006,
|
594 |
+
"learning_rate": 0.00016584726692682372,
|
595 |
+
"loss": 1.6427,
|
596 |
+
"step": 6000
|
597 |
+
},
|
598 |
+
{
|
599 |
+
"epoch": 0.68,
|
600 |
+
"eval_bertscore": 0.7350374460220337,
|
601 |
+
"eval_loss": 1.729591965675354,
|
602 |
+
"eval_rouge1": 0.6516545226356616,
|
603 |
+
"eval_rouge2": 0.35485762033878543,
|
604 |
+
"eval_rougeL": 0.5249054193354852,
|
605 |
+
"eval_rougeLsum": 0.6411016821651583,
|
606 |
+
"eval_runtime": 15.5199,
|
607 |
+
"eval_samples_per_second": 0.644,
|
608 |
+
"eval_steps_per_second": 0.322,
|
609 |
+
"step": 6000
|
610 |
+
},
|
611 |
+
{
|
612 |
+
"epoch": 0.71,
|
613 |
+
"grad_norm": 0.24616578221321106,
|
614 |
+
"learning_rate": 0.0001647078930127895,
|
615 |
+
"loss": 1.6296,
|
616 |
+
"step": 6200
|
617 |
+
},
|
618 |
+
{
|
619 |
+
"epoch": 0.71,
|
620 |
+
"eval_bertscore": 0.7335461378097534,
|
621 |
+
"eval_loss": 1.7302274703979492,
|
622 |
+
"eval_rouge1": 0.6531411706717427,
|
623 |
+
"eval_rouge2": 0.35003053174601517,
|
624 |
+
"eval_rougeL": 0.5212483686089053,
|
625 |
+
"eval_rougeLsum": 0.6438454124825417,
|
626 |
+
"eval_runtime": 15.3058,
|
627 |
+
"eval_samples_per_second": 0.653,
|
628 |
+
"eval_steps_per_second": 0.327,
|
629 |
+
"step": 6200
|
630 |
+
},
|
631 |
+
{
|
632 |
+
"epoch": 0.73,
|
633 |
+
"grad_norm": 0.24500492215156555,
|
634 |
+
"learning_rate": 0.00016356851909875522,
|
635 |
+
"loss": 1.6251,
|
636 |
+
"step": 6400
|
637 |
+
},
|
638 |
+
{
|
639 |
+
"epoch": 0.73,
|
640 |
+
"eval_bertscore": 0.7347471714019775,
|
641 |
+
"eval_loss": 1.7292228937149048,
|
642 |
+
"eval_rouge1": 0.6565322485502556,
|
643 |
+
"eval_rouge2": 0.35887540291607073,
|
644 |
+
"eval_rougeL": 0.5284326132878907,
|
645 |
+
"eval_rougeLsum": 0.6469750895866724,
|
646 |
+
"eval_runtime": 14.9708,
|
647 |
+
"eval_samples_per_second": 0.668,
|
648 |
+
"eval_steps_per_second": 0.334,
|
649 |
+
"step": 6400
|
650 |
+
},
|
651 |
+
{
|
652 |
+
"epoch": 0.75,
|
653 |
+
"grad_norm": 0.26575228571891785,
|
654 |
+
"learning_rate": 0.000162429145184721,
|
655 |
+
"loss": 1.6389,
|
656 |
+
"step": 6600
|
657 |
+
},
|
658 |
+
{
|
659 |
+
"epoch": 0.75,
|
660 |
+
"eval_bertscore": 0.7319897413253784,
|
661 |
+
"eval_loss": 1.7287580966949463,
|
662 |
+
"eval_rouge1": 0.6458608801881298,
|
663 |
+
"eval_rouge2": 0.3503480901452204,
|
664 |
+
"eval_rougeL": 0.519626708150005,
|
665 |
+
"eval_rougeLsum": 0.6362405734928169,
|
666 |
+
"eval_runtime": 15.3509,
|
667 |
+
"eval_samples_per_second": 0.651,
|
668 |
+
"eval_steps_per_second": 0.326,
|
669 |
+
"step": 6600
|
670 |
+
},
|
671 |
+
{
|
672 |
+
"epoch": 0.77,
|
673 |
+
"grad_norm": 0.27144965529441833,
|
674 |
+
"learning_rate": 0.00016128977127068676,
|
675 |
+
"loss": 1.6476,
|
676 |
+
"step": 6800
|
677 |
+
},
|
678 |
+
{
|
679 |
+
"epoch": 0.77,
|
680 |
+
"eval_bertscore": 0.7392772436141968,
|
681 |
+
"eval_loss": 1.7257907390594482,
|
682 |
+
"eval_rouge1": 0.6543238897579965,
|
683 |
+
"eval_rouge2": 0.3606726049451984,
|
684 |
+
"eval_rougeL": 0.5317585887753791,
|
685 |
+
"eval_rougeLsum": 0.6452028420624081,
|
686 |
+
"eval_runtime": 15.0268,
|
687 |
+
"eval_samples_per_second": 0.665,
|
688 |
+
"eval_steps_per_second": 0.333,
|
689 |
+
"step": 6800
|
690 |
+
},
|
691 |
+
{
|
692 |
+
"epoch": 0.8,
|
693 |
+
"grad_norm": 0.2579711079597473,
|
694 |
+
"learning_rate": 0.00016015039735665254,
|
695 |
+
"loss": 1.6316,
|
696 |
+
"step": 7000
|
697 |
+
},
|
698 |
+
{
|
699 |
+
"epoch": 0.8,
|
700 |
+
"eval_bertscore": 0.7375612854957581,
|
701 |
+
"eval_loss": 1.7296981811523438,
|
702 |
+
"eval_rouge1": 0.658906725408624,
|
703 |
+
"eval_rouge2": 0.35825094165644866,
|
704 |
+
"eval_rougeL": 0.5323299193377959,
|
705 |
+
"eval_rougeLsum": 0.6500364347290426,
|
706 |
+
"eval_runtime": 14.9244,
|
707 |
+
"eval_samples_per_second": 0.67,
|
708 |
+
"eval_steps_per_second": 0.335,
|
709 |
+
"step": 7000
|
710 |
+
},
|
711 |
+
{
|
712 |
+
"epoch": 0.82,
|
713 |
+
"grad_norm": 0.2589207589626312,
|
714 |
+
"learning_rate": 0.0001590110234426183,
|
715 |
+
"loss": 1.6432,
|
716 |
+
"step": 7200
|
717 |
+
},
|
718 |
+
{
|
719 |
+
"epoch": 0.82,
|
720 |
+
"eval_bertscore": 0.735145092010498,
|
721 |
+
"eval_loss": 1.72720205783844,
|
722 |
+
"eval_rouge1": 0.6678646250245518,
|
723 |
+
"eval_rouge2": 0.36332843150846983,
|
724 |
+
"eval_rougeL": 0.537576430733886,
|
725 |
+
"eval_rougeLsum": 0.6579789388660506,
|
726 |
+
"eval_runtime": 14.9067,
|
727 |
+
"eval_samples_per_second": 0.671,
|
728 |
+
"eval_steps_per_second": 0.335,
|
729 |
+
"step": 7200
|
730 |
+
},
|
731 |
+
{
|
732 |
+
"epoch": 0.84,
|
733 |
+
"grad_norm": 0.26652559638023376,
|
734 |
+
"learning_rate": 0.00015787164952858404,
|
735 |
+
"loss": 1.6488,
|
736 |
+
"step": 7400
|
737 |
+
},
|
738 |
+
{
|
739 |
+
"epoch": 0.84,
|
740 |
+
"eval_bertscore": 0.7320755124092102,
|
741 |
+
"eval_loss": 1.7282931804656982,
|
742 |
+
"eval_rouge1": 0.6325633297780734,
|
743 |
+
"eval_rouge2": 0.34505856555703185,
|
744 |
+
"eval_rougeL": 0.5100006743383693,
|
745 |
+
"eval_rougeLsum": 0.6230385336341938,
|
746 |
+
"eval_runtime": 14.9075,
|
747 |
+
"eval_samples_per_second": 0.671,
|
748 |
+
"eval_steps_per_second": 0.335,
|
749 |
+
"step": 7400
|
750 |
+
},
|
751 |
+
{
|
752 |
+
"epoch": 0.87,
|
753 |
+
"grad_norm": 0.27353721857070923,
|
754 |
+
"learning_rate": 0.00015673227561454982,
|
755 |
+
"loss": 1.6486,
|
756 |
+
"step": 7600
|
757 |
+
},
|
758 |
+
{
|
759 |
+
"epoch": 0.87,
|
760 |
+
"eval_bertscore": 0.7367390394210815,
|
761 |
+
"eval_loss": 1.7290785312652588,
|
762 |
+
"eval_rouge1": 0.639487116874423,
|
763 |
+
"eval_rouge2": 0.3466574229736927,
|
764 |
+
"eval_rougeL": 0.515038120249177,
|
765 |
+
"eval_rougeLsum": 0.6301157215372983,
|
766 |
+
"eval_runtime": 15.0876,
|
767 |
+
"eval_samples_per_second": 0.663,
|
768 |
+
"eval_steps_per_second": 0.331,
|
769 |
+
"step": 7600
|
770 |
+
},
|
771 |
+
{
|
772 |
+
"epoch": 0.89,
|
773 |
+
"grad_norm": 0.24777938425540924,
|
774 |
+
"learning_rate": 0.00015559290170051558,
|
775 |
+
"loss": 1.6271,
|
776 |
+
"step": 7800
|
777 |
+
},
|
778 |
+
{
|
779 |
+
"epoch": 0.89,
|
780 |
+
"eval_bertscore": 0.735866904258728,
|
781 |
+
"eval_loss": 1.7264015674591064,
|
782 |
+
"eval_rouge1": 0.64939597901302,
|
783 |
+
"eval_rouge2": 0.3554282813944538,
|
784 |
+
"eval_rougeL": 0.5247953329477759,
|
785 |
+
"eval_rougeLsum": 0.6405524812915908,
|
786 |
+
"eval_runtime": 14.8892,
|
787 |
+
"eval_samples_per_second": 0.672,
|
788 |
+
"eval_steps_per_second": 0.336,
|
789 |
+
"step": 7800
|
790 |
+
},
|
791 |
+
{
|
792 |
+
"epoch": 0.91,
|
793 |
+
"grad_norm": 0.2703794538974762,
|
794 |
+
"learning_rate": 0.00015445352778648136,
|
795 |
+
"loss": 1.6415,
|
796 |
+
"step": 8000
|
797 |
+
},
|
798 |
+
{
|
799 |
+
"epoch": 0.91,
|
800 |
+
"eval_bertscore": 0.7325771450996399,
|
801 |
+
"eval_loss": 1.7271970510482788,
|
802 |
+
"eval_rouge1": 0.6659432894288253,
|
803 |
+
"eval_rouge2": 0.35962933912652617,
|
804 |
+
"eval_rougeL": 0.5385420432813512,
|
805 |
+
"eval_rougeLsum": 0.6557027031484046,
|
806 |
+
"eval_runtime": 14.8319,
|
807 |
+
"eval_samples_per_second": 0.674,
|
808 |
+
"eval_steps_per_second": 0.337,
|
809 |
+
"step": 8000
|
810 |
+
},
|
811 |
+
{
|
812 |
+
"epoch": 0.93,
|
813 |
+
"grad_norm": 0.28753793239593506,
|
814 |
+
"learning_rate": 0.0001533141538724471,
|
815 |
+
"loss": 1.6239,
|
816 |
+
"step": 8200
|
817 |
+
},
|
818 |
+
{
|
819 |
+
"epoch": 0.93,
|
820 |
+
"eval_bertscore": 0.7350013852119446,
|
821 |
+
"eval_loss": 1.7266199588775635,
|
822 |
+
"eval_rouge1": 0.6549282561414593,
|
823 |
+
"eval_rouge2": 0.35694530595734475,
|
824 |
+
"eval_rougeL": 0.5301601006964574,
|
825 |
+
"eval_rougeLsum": 0.6441779306137909,
|
826 |
+
"eval_runtime": 14.9005,
|
827 |
+
"eval_samples_per_second": 0.671,
|
828 |
+
"eval_steps_per_second": 0.336,
|
829 |
+
"step": 8200
|
830 |
+
},
|
831 |
+
{
|
832 |
+
"epoch": 0.96,
|
833 |
+
"grad_norm": 0.23870150744915009,
|
834 |
+
"learning_rate": 0.00015217477995841286,
|
835 |
+
"loss": 1.6293,
|
836 |
+
"step": 8400
|
837 |
+
},
|
838 |
+
{
|
839 |
+
"epoch": 0.96,
|
840 |
+
"eval_bertscore": 0.7271261811256409,
|
841 |
+
"eval_loss": 1.7256368398666382,
|
842 |
+
"eval_rouge1": 0.6515513829901936,
|
843 |
+
"eval_rouge2": 0.35217616104918836,
|
844 |
+
"eval_rougeL": 0.5236553509227138,
|
845 |
+
"eval_rougeLsum": 0.6411473505324752,
|
846 |
+
"eval_runtime": 14.9938,
|
847 |
+
"eval_samples_per_second": 0.667,
|
848 |
+
"eval_steps_per_second": 0.333,
|
849 |
+
"step": 8400
|
850 |
+
},
|
851 |
+
{
|
852 |
+
"epoch": 0.98,
|
853 |
+
"grad_norm": 0.28276997804641724,
|
854 |
+
"learning_rate": 0.00015103540604437861,
|
855 |
+
"loss": 1.6242,
|
856 |
+
"step": 8600
|
857 |
+
},
|
858 |
+
{
|
859 |
+
"epoch": 0.98,
|
860 |
+
"eval_bertscore": 0.7347334027290344,
|
861 |
+
"eval_loss": 1.717627763748169,
|
862 |
+
"eval_rouge1": 0.6350112495847634,
|
863 |
+
"eval_rouge2": 0.3477570751550898,
|
864 |
+
"eval_rougeL": 0.5146616989899861,
|
865 |
+
"eval_rougeLsum": 0.6246669376525157,
|
866 |
+
"eval_runtime": 15.7032,
|
867 |
+
"eval_samples_per_second": 0.637,
|
868 |
+
"eval_steps_per_second": 0.318,
|
869 |
+
"step": 8600
|
870 |
+
},
|
871 |
+
{
|
872 |
+
"epoch": 1.0,
|
873 |
+
"grad_norm": 0.24915842711925507,
|
874 |
+
"learning_rate": 0.00014989603213034437,
|
875 |
+
"loss": 1.6245,
|
876 |
+
"step": 8800
|
877 |
+
},
|
878 |
+
{
|
879 |
+
"epoch": 1.0,
|
880 |
+
"eval_bertscore": 0.7313701510429382,
|
881 |
+
"eval_loss": 1.7292964458465576,
|
882 |
+
"eval_rouge1": 0.6479528105367669,
|
883 |
+
"eval_rouge2": 0.35020983244262877,
|
884 |
+
"eval_rougeL": 0.5200907337780047,
|
885 |
+
"eval_rougeLsum": 0.6372896614836894,
|
886 |
+
"eval_runtime": 15.0043,
|
887 |
+
"eval_samples_per_second": 0.666,
|
888 |
+
"eval_steps_per_second": 0.333,
|
889 |
+
"step": 8800
|
890 |
+
},
|
891 |
+
{
|
892 |
+
"epoch": 1.03,
|
893 |
+
"grad_norm": 0.24036027491092682,
|
894 |
+
"learning_rate": 0.00014875665821631015,
|
895 |
+
"loss": 1.5364,
|
896 |
+
"step": 9000
|
897 |
+
},
|
898 |
+
{
|
899 |
+
"epoch": 1.03,
|
900 |
+
"eval_bertscore": 0.7327737808227539,
|
901 |
+
"eval_loss": 1.7339435815811157,
|
902 |
+
"eval_rouge1": 0.6568524178922349,
|
903 |
+
"eval_rouge2": 0.35560270713543163,
|
904 |
+
"eval_rougeL": 0.5310443670833082,
|
905 |
+
"eval_rougeLsum": 0.6480993679097387,
|
906 |
+
"eval_runtime": 14.9303,
|
907 |
+
"eval_samples_per_second": 0.67,
|
908 |
+
"eval_steps_per_second": 0.335,
|
909 |
+
"step": 9000
|
910 |
+
},
|
911 |
+
{
|
912 |
+
"epoch": 1.05,
|
913 |
+
"grad_norm": 0.2729027271270752,
|
914 |
+
"learning_rate": 0.0001476172843022759,
|
915 |
+
"loss": 1.5182,
|
916 |
+
"step": 9200
|
917 |
+
},
|
918 |
+
{
|
919 |
+
"epoch": 1.05,
|
920 |
+
"eval_bertscore": 0.7334672212600708,
|
921 |
+
"eval_loss": 1.739061713218689,
|
922 |
+
"eval_rouge1": 0.6552962187824329,
|
923 |
+
"eval_rouge2": 0.35210314124279196,
|
924 |
+
"eval_rougeL": 0.5272039052368354,
|
925 |
+
"eval_rougeLsum": 0.6437473533492806,
|
926 |
+
"eval_runtime": 15.5418,
|
927 |
+
"eval_samples_per_second": 0.643,
|
928 |
+
"eval_steps_per_second": 0.322,
|
929 |
+
"step": 9200
|
930 |
+
},
|
931 |
+
{
|
932 |
+
"epoch": 1.07,
|
933 |
+
"grad_norm": 0.2909716069698334,
|
934 |
+
"learning_rate": 0.00014647791038824168,
|
935 |
+
"loss": 1.5276,
|
936 |
+
"step": 9400
|
937 |
+
},
|
938 |
+
{
|
939 |
+
"epoch": 1.07,
|
940 |
+
"eval_bertscore": 0.7288902997970581,
|
941 |
+
"eval_loss": 1.736944556236267,
|
942 |
+
"eval_rouge1": 0.6533271685598301,
|
943 |
+
"eval_rouge2": 0.35279315321532184,
|
944 |
+
"eval_rougeL": 0.5262688234671329,
|
945 |
+
"eval_rougeLsum": 0.6424084937151033,
|
946 |
+
"eval_runtime": 15.1115,
|
947 |
+
"eval_samples_per_second": 0.662,
|
948 |
+
"eval_steps_per_second": 0.331,
|
949 |
+
"step": 9400
|
950 |
+
},
|
951 |
+
{
|
952 |
+
"epoch": 1.09,
|
953 |
+
"grad_norm": 0.3035859763622284,
|
954 |
+
"learning_rate": 0.00014533853647420743,
|
955 |
+
"loss": 1.5445,
|
956 |
+
"step": 9600
|
957 |
+
},
|
958 |
+
{
|
959 |
+
"epoch": 1.09,
|
960 |
+
"eval_bertscore": 0.7308284044265747,
|
961 |
+
"eval_loss": 1.737762689590454,
|
962 |
+
"eval_rouge1": 0.6619725777891359,
|
963 |
+
"eval_rouge2": 0.3611963714506864,
|
964 |
+
"eval_rougeL": 0.5363802967084452,
|
965 |
+
"eval_rougeLsum": 0.6516690557971352,
|
966 |
+
"eval_runtime": 14.9932,
|
967 |
+
"eval_samples_per_second": 0.667,
|
968 |
+
"eval_steps_per_second": 0.333,
|
969 |
+
"step": 9600
|
970 |
+
},
|
971 |
+
{
|
972 |
+
"epoch": 1.12,
|
973 |
+
"grad_norm": 0.26574915647506714,
|
974 |
+
"learning_rate": 0.0001441991625601732,
|
975 |
+
"loss": 1.5342,
|
976 |
+
"step": 9800
|
977 |
+
},
|
978 |
+
{
|
979 |
+
"epoch": 1.12,
|
980 |
+
"eval_bertscore": 0.7328712344169617,
|
981 |
+
"eval_loss": 1.7393991947174072,
|
982 |
+
"eval_rouge1": 0.6856504003396884,
|
983 |
+
"eval_rouge2": 0.3761098841062477,
|
984 |
+
"eval_rougeL": 0.555477293163325,
|
985 |
+
"eval_rougeLsum": 0.6757574283262289,
|
986 |
+
"eval_runtime": 14.8121,
|
987 |
+
"eval_samples_per_second": 0.675,
|
988 |
+
"eval_steps_per_second": 0.338,
|
989 |
+
"step": 9800
|
990 |
+
},
|
991 |
+
{
|
992 |
+
"epoch": 1.14,
|
993 |
+
"grad_norm": 0.315468430519104,
|
994 |
+
"learning_rate": 0.00014305978864613897,
|
995 |
+
"loss": 1.543,
|
996 |
+
"step": 10000
|
997 |
+
},
|
998 |
+
{
|
999 |
+
"epoch": 1.14,
|
1000 |
+
"eval_bertscore": 0.7349387407302856,
|
1001 |
+
"eval_loss": 1.7352710962295532,
|
1002 |
+
"eval_rouge1": 0.6749953128982036,
|
1003 |
+
"eval_rouge2": 0.3720385250530084,
|
1004 |
+
"eval_rougeL": 0.5472261566474382,
|
1005 |
+
"eval_rougeLsum": 0.6657643539219252,
|
1006 |
+
"eval_runtime": 14.909,
|
1007 |
+
"eval_samples_per_second": 0.671,
|
1008 |
+
"eval_steps_per_second": 0.335,
|
1009 |
+
"step": 10000
|
1010 |
+
},
|
1011 |
+
{
|
1012 |
+
"epoch": 1.16,
|
1013 |
+
"grad_norm": 0.29815027117729187,
|
1014 |
+
"learning_rate": 0.00014192041473210472,
|
1015 |
+
"loss": 1.5547,
|
1016 |
+
"step": 10200
|
1017 |
+
},
|
1018 |
+
{
|
1019 |
+
"epoch": 1.16,
|
1020 |
+
"eval_bertscore": 0.7359883189201355,
|
1021 |
+
"eval_loss": 1.7269136905670166,
|
1022 |
+
"eval_rouge1": 0.6561141614088863,
|
1023 |
+
"eval_rouge2": 0.3606175666303814,
|
1024 |
+
"eval_rougeL": 0.5302270771032793,
|
1025 |
+
"eval_rougeLsum": 0.6446912079521883,
|
1026 |
+
"eval_runtime": 14.9527,
|
1027 |
+
"eval_samples_per_second": 0.669,
|
1028 |
+
"eval_steps_per_second": 0.334,
|
1029 |
+
"step": 10200
|
1030 |
+
},
|
1031 |
+
{
|
1032 |
+
"epoch": 1.18,
|
1033 |
+
"grad_norm": 0.3595702350139618,
|
1034 |
+
"learning_rate": 0.00014078104081807047,
|
1035 |
+
"loss": 1.5567,
|
1036 |
+
"step": 10400
|
1037 |
+
},
|
1038 |
+
{
|
1039 |
+
"epoch": 1.18,
|
1040 |
+
"eval_bertscore": 0.7328116297721863,
|
1041 |
+
"eval_loss": 1.7341644763946533,
|
1042 |
+
"eval_rouge1": 0.6420332714823549,
|
1043 |
+
"eval_rouge2": 0.35094864549032,
|
1044 |
+
"eval_rougeL": 0.5179556761398367,
|
1045 |
+
"eval_rougeLsum": 0.631987397226809,
|
1046 |
+
"eval_runtime": 15.1438,
|
1047 |
+
"eval_samples_per_second": 0.66,
|
1048 |
+
"eval_steps_per_second": 0.33,
|
1049 |
+
"step": 10400
|
1050 |
+
},
|
1051 |
+
{
|
1052 |
+
"epoch": 1.21,
|
1053 |
+
"grad_norm": 0.2718666195869446,
|
1054 |
+
"learning_rate": 0.00013964166690403623,
|
1055 |
+
"loss": 1.5408,
|
1056 |
+
"step": 10600
|
1057 |
+
},
|
1058 |
+
{
|
1059 |
+
"epoch": 1.21,
|
1060 |
+
"eval_bertscore": 0.7337731122970581,
|
1061 |
+
"eval_loss": 1.7330901622772217,
|
1062 |
+
"eval_rouge1": 0.661681342484687,
|
1063 |
+
"eval_rouge2": 0.3626833509973693,
|
1064 |
+
"eval_rougeL": 0.5329424447373774,
|
1065 |
+
"eval_rougeLsum": 0.6519750177144633,
|
1066 |
+
"eval_runtime": 14.8142,
|
1067 |
+
"eval_samples_per_second": 0.675,
|
1068 |
+
"eval_steps_per_second": 0.338,
|
1069 |
+
"step": 10600
|
1070 |
+
},
|
1071 |
+
{
|
1072 |
+
"epoch": 1.23,
|
1073 |
+
"grad_norm": 0.29183274507522583,
|
1074 |
+
"learning_rate": 0.00013850229299000198,
|
1075 |
+
"loss": 1.5422,
|
1076 |
+
"step": 10800
|
1077 |
+
},
|
1078 |
+
{
|
1079 |
+
"epoch": 1.23,
|
1080 |
+
"eval_bertscore": 0.7331051230430603,
|
1081 |
+
"eval_loss": 1.7297636270523071,
|
1082 |
+
"eval_rouge1": 0.6655497978238063,
|
1083 |
+
"eval_rouge2": 0.3614235788441926,
|
1084 |
+
"eval_rougeL": 0.5327210061667442,
|
1085 |
+
"eval_rougeLsum": 0.6548836840483913,
|
1086 |
+
"eval_runtime": 15.1359,
|
1087 |
+
"eval_samples_per_second": 0.661,
|
1088 |
+
"eval_steps_per_second": 0.33,
|
1089 |
+
"step": 10800
|
1090 |
+
},
|
1091 |
+
{
|
1092 |
+
"epoch": 1.25,
|
1093 |
+
"grad_norm": 0.30979740619659424,
|
1094 |
+
"learning_rate": 0.00013736291907596776,
|
1095 |
+
"loss": 1.5372,
|
1096 |
+
"step": 11000
|
1097 |
+
},
|
1098 |
+
{
|
1099 |
+
"epoch": 1.25,
|
1100 |
+
"eval_bertscore": 0.7312799692153931,
|
1101 |
+
"eval_loss": 1.732444167137146,
|
1102 |
+
"eval_rouge1": 0.6568292865033993,
|
1103 |
+
"eval_rouge2": 0.35876682221562006,
|
1104 |
+
"eval_rougeL": 0.5300878844981931,
|
1105 |
+
"eval_rougeLsum": 0.6461751645858989,
|
1106 |
+
"eval_runtime": 14.819,
|
1107 |
+
"eval_samples_per_second": 0.675,
|
1108 |
+
"eval_steps_per_second": 0.337,
|
1109 |
+
"step": 11000
|
1110 |
+
},
|
1111 |
+
{
|
1112 |
+
"epoch": 1.28,
|
1113 |
+
"grad_norm": 0.31343138217926025,
|
1114 |
+
"learning_rate": 0.0001362235451619335,
|
1115 |
+
"loss": 1.5301,
|
1116 |
+
"step": 11200
|
1117 |
+
},
|
1118 |
+
{
|
1119 |
+
"epoch": 1.28,
|
1120 |
+
"eval_bertscore": 0.7317885160446167,
|
1121 |
+
"eval_loss": 1.7358499765396118,
|
1122 |
+
"eval_rouge1": 0.6548673097943329,
|
1123 |
+
"eval_rouge2": 0.3609116081432997,
|
1124 |
+
"eval_rougeL": 0.5279887650752133,
|
1125 |
+
"eval_rougeLsum": 0.6466232329097188,
|
1126 |
+
"eval_runtime": 14.8259,
|
1127 |
+
"eval_samples_per_second": 0.674,
|
1128 |
+
"eval_steps_per_second": 0.337,
|
1129 |
+
"step": 11200
|
1130 |
+
},
|
1131 |
+
{
|
1132 |
+
"epoch": 1.3,
|
1133 |
+
"grad_norm": 0.36181533336639404,
|
1134 |
+
"learning_rate": 0.0001350841712478993,
|
1135 |
+
"loss": 1.5421,
|
1136 |
+
"step": 11400
|
1137 |
+
},
|
1138 |
+
{
|
1139 |
+
"epoch": 1.3,
|
1140 |
+
"eval_bertscore": 0.7316756248474121,
|
1141 |
+
"eval_loss": 1.7282969951629639,
|
1142 |
+
"eval_rouge1": 0.6551882964480251,
|
1143 |
+
"eval_rouge2": 0.3580708921400697,
|
1144 |
+
"eval_rougeL": 0.5255367305995147,
|
1145 |
+
"eval_rougeLsum": 0.6449192953008009,
|
1146 |
+
"eval_runtime": 14.8816,
|
1147 |
+
"eval_samples_per_second": 0.672,
|
1148 |
+
"eval_steps_per_second": 0.336,
|
1149 |
+
"step": 11400
|
1150 |
+
},
|
1151 |
+
{
|
1152 |
+
"epoch": 1.32,
|
1153 |
+
"grad_norm": 0.30600836873054504,
|
1154 |
+
"learning_rate": 0.00013394479733386505,
|
1155 |
+
"loss": 1.5538,
|
1156 |
+
"step": 11600
|
1157 |
+
},
|
1158 |
+
{
|
1159 |
+
"epoch": 1.32,
|
1160 |
+
"eval_bertscore": 0.7311854362487793,
|
1161 |
+
"eval_loss": 1.7313562631607056,
|
1162 |
+
"eval_rouge1": 0.6592751199424156,
|
1163 |
+
"eval_rouge2": 0.35802855072854206,
|
1164 |
+
"eval_rougeL": 0.5297288176377084,
|
1165 |
+
"eval_rougeLsum": 0.6489455314962717,
|
1166 |
+
"eval_runtime": 15.0693,
|
1167 |
+
"eval_samples_per_second": 0.664,
|
1168 |
+
"eval_steps_per_second": 0.332,
|
1169 |
+
"step": 11600
|
1170 |
+
},
|
1171 |
+
{
|
1172 |
+
"epoch": 1.34,
|
1173 |
+
"grad_norm": 0.29904893040657043,
|
1174 |
+
"learning_rate": 0.0001328054234198308,
|
1175 |
+
"loss": 1.5328,
|
1176 |
+
"step": 11800
|
1177 |
+
},
|
1178 |
+
{
|
1179 |
+
"epoch": 1.34,
|
1180 |
+
"eval_bertscore": 0.7312635183334351,
|
1181 |
+
"eval_loss": 1.7318429946899414,
|
1182 |
+
"eval_rouge1": 0.6577169369077195,
|
1183 |
+
"eval_rouge2": 0.3582474830918887,
|
1184 |
+
"eval_rougeL": 0.5314990647771975,
|
1185 |
+
"eval_rougeLsum": 0.6454785220479866,
|
1186 |
+
"eval_runtime": 15.0235,
|
1187 |
+
"eval_samples_per_second": 0.666,
|
1188 |
+
"eval_steps_per_second": 0.333,
|
1189 |
+
"step": 11800
|
1190 |
+
},
|
1191 |
+
{
|
1192 |
+
"epoch": 1.37,
|
1193 |
+
"grad_norm": 0.3025416433811188,
|
1194 |
+
"learning_rate": 0.00013166604950579658,
|
1195 |
+
"loss": 1.5349,
|
1196 |
+
"step": 12000
|
1197 |
+
},
|
1198 |
+
{
|
1199 |
+
"epoch": 1.37,
|
1200 |
+
"eval_bertscore": 0.7325812578201294,
|
1201 |
+
"eval_loss": 1.7309118509292603,
|
1202 |
+
"eval_rouge1": 0.6629133074951261,
|
1203 |
+
"eval_rouge2": 0.3678158453940578,
|
1204 |
+
"eval_rougeL": 0.5380936907276155,
|
1205 |
+
"eval_rougeLsum": 0.654883061928214,
|
1206 |
+
"eval_runtime": 14.7666,
|
1207 |
+
"eval_samples_per_second": 0.677,
|
1208 |
+
"eval_steps_per_second": 0.339,
|
1209 |
+
"step": 12000
|
1210 |
+
},
|
1211 |
+
{
|
1212 |
+
"epoch": 1.39,
|
1213 |
+
"grad_norm": 0.34982389211654663,
|
1214 |
+
"learning_rate": 0.00013052667559176233,
|
1215 |
+
"loss": 1.5513,
|
1216 |
+
"step": 12200
|
1217 |
+
},
|
1218 |
+
{
|
1219 |
+
"epoch": 1.39,
|
1220 |
+
"eval_bertscore": 0.7340582609176636,
|
1221 |
+
"eval_loss": 1.7363474369049072,
|
1222 |
+
"eval_rouge1": 0.6555817937418287,
|
1223 |
+
"eval_rouge2": 0.35630500078358396,
|
1224 |
+
"eval_rougeL": 0.5272412353478366,
|
1225 |
+
"eval_rougeLsum": 0.6445837479327643,
|
1226 |
+
"eval_runtime": 14.9664,
|
1227 |
+
"eval_samples_per_second": 0.668,
|
1228 |
+
"eval_steps_per_second": 0.334,
|
1229 |
+
"step": 12200
|
1230 |
+
},
|
1231 |
+
{
|
1232 |
+
"epoch": 1.41,
|
1233 |
+
"grad_norm": 0.35809043049812317,
|
1234 |
+
"learning_rate": 0.0001293873016777281,
|
1235 |
+
"loss": 1.5444,
|
1236 |
+
"step": 12400
|
1237 |
+
},
|
1238 |
+
{
|
1239 |
+
"epoch": 1.41,
|
1240 |
+
"eval_bertscore": 0.7334069013595581,
|
1241 |
+
"eval_loss": 1.7328729629516602,
|
1242 |
+
"eval_rouge1": 0.6617971237669364,
|
1243 |
+
"eval_rouge2": 0.35951260376512423,
|
1244 |
+
"eval_rougeL": 0.5345512305507059,
|
1245 |
+
"eval_rougeLsum": 0.648363531132752,
|
1246 |
+
"eval_runtime": 15.2146,
|
1247 |
+
"eval_samples_per_second": 0.657,
|
1248 |
+
"eval_steps_per_second": 0.329,
|
1249 |
+
"step": 12400
|
1250 |
+
},
|
1251 |
+
{
|
1252 |
+
"epoch": 1.44,
|
1253 |
+
"grad_norm": 0.2954196631908417,
|
1254 |
+
"learning_rate": 0.00012824792776369387,
|
1255 |
+
"loss": 1.5406,
|
1256 |
+
"step": 12600
|
1257 |
+
},
|
1258 |
+
{
|
1259 |
+
"epoch": 1.44,
|
1260 |
+
"eval_bertscore": 0.7321678400039673,
|
1261 |
+
"eval_loss": 1.7335160970687866,
|
1262 |
+
"eval_rouge1": 0.6573625593086756,
|
1263 |
+
"eval_rouge2": 0.36210525247389347,
|
1264 |
+
"eval_rougeL": 0.5379361120230158,
|
1265 |
+
"eval_rougeLsum": 0.6459787883452857,
|
1266 |
+
"eval_runtime": 14.8942,
|
1267 |
+
"eval_samples_per_second": 0.671,
|
1268 |
+
"eval_steps_per_second": 0.336,
|
1269 |
+
"step": 12600
|
1270 |
+
},
|
1271 |
+
{
|
1272 |
+
"epoch": 1.46,
|
1273 |
+
"grad_norm": 0.32190588116645813,
|
1274 |
+
"learning_rate": 0.00012710855384965962,
|
1275 |
+
"loss": 1.5491,
|
1276 |
+
"step": 12800
|
1277 |
+
},
|
1278 |
+
{
|
1279 |
+
"epoch": 1.46,
|
1280 |
+
"eval_bertscore": 0.7346011400222778,
|
1281 |
+
"eval_loss": 1.7364966869354248,
|
1282 |
+
"eval_rouge1": 0.6481210247390559,
|
1283 |
+
"eval_rouge2": 0.3521173896017687,
|
1284 |
+
"eval_rougeL": 0.5240500581372636,
|
1285 |
+
"eval_rougeLsum": 0.63706442433335,
|
1286 |
+
"eval_runtime": 14.923,
|
1287 |
+
"eval_samples_per_second": 0.67,
|
1288 |
+
"eval_steps_per_second": 0.335,
|
1289 |
+
"step": 12800
|
1290 |
+
},
|
1291 |
+
{
|
1292 |
+
"epoch": 1.48,
|
1293 |
+
"grad_norm": 0.33323267102241516,
|
1294 |
+
"learning_rate": 0.00012596917993562537,
|
1295 |
+
"loss": 1.5596,
|
1296 |
+
"step": 13000
|
1297 |
+
},
|
1298 |
+
{
|
1299 |
+
"epoch": 1.48,
|
1300 |
+
"eval_bertscore": 0.7331587076187134,
|
1301 |
+
"eval_loss": 1.7332260608673096,
|
1302 |
+
"eval_rouge1": 0.6561257401878793,
|
1303 |
+
"eval_rouge2": 0.3548063723792664,
|
1304 |
+
"eval_rougeL": 0.527807776001489,
|
1305 |
+
"eval_rougeLsum": 0.6451911907984706,
|
1306 |
+
"eval_runtime": 15.4703,
|
1307 |
+
"eval_samples_per_second": 0.646,
|
1308 |
+
"eval_steps_per_second": 0.323,
|
1309 |
+
"step": 13000
|
1310 |
+
},
|
1311 |
+
{
|
1312 |
+
"epoch": 1.5,
|
1313 |
+
"grad_norm": 0.3564057946205139,
|
1314 |
+
"learning_rate": 0.00012482980602159113,
|
1315 |
+
"loss": 1.5261,
|
1316 |
+
"step": 13200
|
1317 |
+
},
|
1318 |
+
{
|
1319 |
+
"epoch": 1.5,
|
1320 |
+
"eval_bertscore": 0.7306063771247864,
|
1321 |
+
"eval_loss": 1.7368810176849365,
|
1322 |
+
"eval_rouge1": 0.637722723890071,
|
1323 |
+
"eval_rouge2": 0.3455358728458236,
|
1324 |
+
"eval_rougeL": 0.5136372690435154,
|
1325 |
+
"eval_rougeLsum": 0.6273570573595115,
|
1326 |
+
"eval_runtime": 15.3533,
|
1327 |
+
"eval_samples_per_second": 0.651,
|
1328 |
+
"eval_steps_per_second": 0.326,
|
1329 |
+
"step": 13200
|
1330 |
+
},
|
1331 |
+
{
|
1332 |
+
"epoch": 1.53,
|
1333 |
+
"grad_norm": 0.29219934344291687,
|
1334 |
+
"learning_rate": 0.0001236904321075569,
|
1335 |
+
"loss": 1.519,
|
1336 |
+
"step": 13400
|
1337 |
+
},
|
1338 |
+
{
|
1339 |
+
"epoch": 1.53,
|
1340 |
+
"eval_bertscore": 0.7338696122169495,
|
1341 |
+
"eval_loss": 1.734724998474121,
|
1342 |
+
"eval_rouge1": 0.6442107446420164,
|
1343 |
+
"eval_rouge2": 0.3494748457109431,
|
1344 |
+
"eval_rougeL": 0.5207483892007314,
|
1345 |
+
"eval_rougeLsum": 0.632886404907802,
|
1346 |
+
"eval_runtime": 15.3077,
|
1347 |
+
"eval_samples_per_second": 0.653,
|
1348 |
+
"eval_steps_per_second": 0.327,
|
1349 |
+
"step": 13400
|
1350 |
+
},
|
1351 |
+
{
|
1352 |
+
"epoch": 1.55,
|
1353 |
+
"grad_norm": 0.34681758284568787,
|
1354 |
+
"learning_rate": 0.00012255105819352266,
|
1355 |
+
"loss": 1.5419,
|
1356 |
+
"step": 13600
|
1357 |
+
},
|
1358 |
+
{
|
1359 |
+
"epoch": 1.55,
|
1360 |
+
"eval_bertscore": 0.7350045442581177,
|
1361 |
+
"eval_loss": 1.7329858541488647,
|
1362 |
+
"eval_rouge1": 0.6606839869796519,
|
1363 |
+
"eval_rouge2": 0.362188561160822,
|
1364 |
+
"eval_rougeL": 0.5342033818317451,
|
1365 |
+
"eval_rougeLsum": 0.6493340000068861,
|
1366 |
+
"eval_runtime": 15.44,
|
1367 |
+
"eval_samples_per_second": 0.648,
|
1368 |
+
"eval_steps_per_second": 0.324,
|
1369 |
+
"step": 13600
|
1370 |
+
},
|
1371 |
+
{
|
1372 |
+
"epoch": 1.57,
|
1373 |
+
"grad_norm": 0.3043666481971741,
|
1374 |
+
"learning_rate": 0.00012141168427948844,
|
1375 |
+
"loss": 1.5402,
|
1376 |
+
"step": 13800
|
1377 |
+
},
|
1378 |
+
{
|
1379 |
+
"epoch": 1.57,
|
1380 |
+
"eval_bertscore": 0.7363221645355225,
|
1381 |
+
"eval_loss": 1.7308530807495117,
|
1382 |
+
"eval_rouge1": 0.6638252384356028,
|
1383 |
+
"eval_rouge2": 0.3643237697892826,
|
1384 |
+
"eval_rougeL": 0.5403775887381331,
|
1385 |
+
"eval_rougeLsum": 0.6537260000827279,
|
1386 |
+
"eval_runtime": 14.7668,
|
1387 |
+
"eval_samples_per_second": 0.677,
|
1388 |
+
"eval_steps_per_second": 0.339,
|
1389 |
+
"step": 13800
|
1390 |
+
},
|
1391 |
+
{
|
1392 |
+
"epoch": 1.59,
|
1393 |
+
"grad_norm": 0.4073585867881775,
|
1394 |
+
"learning_rate": 0.00012027231036545419,
|
1395 |
+
"loss": 1.5256,
|
1396 |
+
"step": 14000
|
1397 |
+
},
|
1398 |
+
{
|
1399 |
+
"epoch": 1.59,
|
1400 |
+
"eval_bertscore": 0.7310279607772827,
|
1401 |
+
"eval_loss": 1.7326784133911133,
|
1402 |
+
"eval_rouge1": 0.6609594314120198,
|
1403 |
+
"eval_rouge2": 0.3601530714440473,
|
1404 |
+
"eval_rougeL": 0.5344452687135626,
|
1405 |
+
"eval_rougeLsum": 0.6480936554342305,
|
1406 |
+
"eval_runtime": 14.8565,
|
1407 |
+
"eval_samples_per_second": 0.673,
|
1408 |
+
"eval_steps_per_second": 0.337,
|
1409 |
+
"step": 14000
|
1410 |
+
},
|
1411 |
+
{
|
1412 |
+
"epoch": 1.62,
|
1413 |
+
"grad_norm": 0.3211813271045685,
|
1414 |
+
"learning_rate": 0.00011913293645141995,
|
1415 |
+
"loss": 1.5366,
|
1416 |
+
"step": 14200
|
1417 |
+
},
|
1418 |
+
{
|
1419 |
+
"epoch": 1.62,
|
1420 |
+
"eval_bertscore": 0.7356667518615723,
|
1421 |
+
"eval_loss": 1.7280094623565674,
|
1422 |
+
"eval_rouge1": 0.6519353227375031,
|
1423 |
+
"eval_rouge2": 0.3587025716186173,
|
1424 |
+
"eval_rougeL": 0.5306356200586075,
|
1425 |
+
"eval_rougeLsum": 0.6408870347994059,
|
1426 |
+
"eval_runtime": 14.9264,
|
1427 |
+
"eval_samples_per_second": 0.67,
|
1428 |
+
"eval_steps_per_second": 0.335,
|
1429 |
+
"step": 14200
|
1430 |
+
},
|
1431 |
+
{
|
1432 |
+
"epoch": 1.64,
|
1433 |
+
"grad_norm": 0.32776832580566406,
|
1434 |
+
"learning_rate": 0.00011799356253738571,
|
1435 |
+
"loss": 1.5504,
|
1436 |
+
"step": 14400
|
1437 |
+
},
|
1438 |
+
{
|
1439 |
+
"epoch": 1.64,
|
1440 |
+
"eval_bertscore": 0.7331353425979614,
|
1441 |
+
"eval_loss": 1.7308950424194336,
|
1442 |
+
"eval_rouge1": 0.6627702292814652,
|
1443 |
+
"eval_rouge2": 0.36117793957379707,
|
1444 |
+
"eval_rougeL": 0.5369305446079228,
|
1445 |
+
"eval_rougeLsum": 0.6516924083980089,
|
1446 |
+
"eval_runtime": 16.0138,
|
1447 |
+
"eval_samples_per_second": 0.624,
|
1448 |
+
"eval_steps_per_second": 0.312,
|
1449 |
+
"step": 14400
|
1450 |
+
},
|
1451 |
+
{
|
1452 |
+
"epoch": 1.66,
|
1453 |
+
"grad_norm": 0.3209726810455322,
|
1454 |
+
"learning_rate": 0.00011685418862335147,
|
1455 |
+
"loss": 1.5473,
|
1456 |
+
"step": 14600
|
1457 |
+
},
|
1458 |
+
{
|
1459 |
+
"epoch": 1.66,
|
1460 |
+
"eval_bertscore": 0.732498824596405,
|
1461 |
+
"eval_loss": 1.7328402996063232,
|
1462 |
+
"eval_rouge1": 0.6482679740803596,
|
1463 |
+
"eval_rouge2": 0.3538726087405498,
|
1464 |
+
"eval_rougeL": 0.5267677183598017,
|
1465 |
+
"eval_rougeLsum": 0.6366529460029322,
|
1466 |
+
"eval_runtime": 15.0195,
|
1467 |
+
"eval_samples_per_second": 0.666,
|
1468 |
+
"eval_steps_per_second": 0.333,
|
1469 |
+
"step": 14600
|
1470 |
+
},
|
1471 |
+
{
|
1472 |
+
"epoch": 1.69,
|
1473 |
+
"grad_norm": 0.3174591064453125,
|
1474 |
+
"learning_rate": 0.00011571481470931725,
|
1475 |
+
"loss": 1.5568,
|
1476 |
+
"step": 14800
|
1477 |
+
},
|
1478 |
+
{
|
1479 |
+
"epoch": 1.69,
|
1480 |
+
"eval_bertscore": 0.7335298657417297,
|
1481 |
+
"eval_loss": 1.7310253381729126,
|
1482 |
+
"eval_rouge1": 0.6560468577439627,
|
1483 |
+
"eval_rouge2": 0.36039371229175,
|
1484 |
+
"eval_rougeL": 0.5318708569729291,
|
1485 |
+
"eval_rougeLsum": 0.6444857558837042,
|
1486 |
+
"eval_runtime": 14.9774,
|
1487 |
+
"eval_samples_per_second": 0.668,
|
1488 |
+
"eval_steps_per_second": 0.334,
|
1489 |
+
"step": 14800
|
1490 |
+
},
|
1491 |
+
{
|
1492 |
+
"epoch": 1.71,
|
1493 |
+
"grad_norm": 0.2936408817768097,
|
1494 |
+
"learning_rate": 0.000114575440795283,
|
1495 |
+
"loss": 1.5345,
|
1496 |
+
"step": 15000
|
1497 |
+
},
|
1498 |
+
{
|
1499 |
+
"epoch": 1.71,
|
1500 |
+
"eval_bertscore": 0.7322725057601929,
|
1501 |
+
"eval_loss": 1.7270629405975342,
|
1502 |
+
"eval_rouge1": 0.6387060930656672,
|
1503 |
+
"eval_rouge2": 0.3480508127989137,
|
1504 |
+
"eval_rougeL": 0.5148670834213287,
|
1505 |
+
"eval_rougeLsum": 0.6273654952601909,
|
1506 |
+
"eval_runtime": 15.6687,
|
1507 |
+
"eval_samples_per_second": 0.638,
|
1508 |
+
"eval_steps_per_second": 0.319,
|
1509 |
+
"step": 15000
|
1510 |
+
},
|
1511 |
+
{
|
1512 |
+
"epoch": 1.73,
|
1513 |
+
"grad_norm": 0.32960689067840576,
|
1514 |
+
"learning_rate": 0.00011343606688124875,
|
1515 |
+
"loss": 1.5362,
|
1516 |
+
"step": 15200
|
1517 |
+
},
|
1518 |
+
{
|
1519 |
+
"epoch": 1.73,
|
1520 |
+
"eval_bertscore": 0.7337037920951843,
|
1521 |
+
"eval_loss": 1.7287395000457764,
|
1522 |
+
"eval_rouge1": 0.6476816970229771,
|
1523 |
+
"eval_rouge2": 0.3532248216683249,
|
1524 |
+
"eval_rougeL": 0.5253136618838716,
|
1525 |
+
"eval_rougeLsum": 0.6347493764394183,
|
1526 |
+
"eval_runtime": 15.0045,
|
1527 |
+
"eval_samples_per_second": 0.666,
|
1528 |
+
"eval_steps_per_second": 0.333,
|
1529 |
+
"step": 15200
|
1530 |
+
},
|
1531 |
+
{
|
1532 |
+
"epoch": 1.75,
|
1533 |
+
"grad_norm": 0.33265602588653564,
|
1534 |
+
"learning_rate": 0.00011229669296721452,
|
1535 |
+
"loss": 1.5215,
|
1536 |
+
"step": 15400
|
1537 |
+
},
|
1538 |
+
{
|
1539 |
+
"epoch": 1.75,
|
1540 |
+
"eval_bertscore": 0.7330806851387024,
|
1541 |
+
"eval_loss": 1.7265052795410156,
|
1542 |
+
"eval_rouge1": 0.6529393512177359,
|
1543 |
+
"eval_rouge2": 0.36182153145062224,
|
1544 |
+
"eval_rougeL": 0.5317061134915853,
|
1545 |
+
"eval_rougeLsum": 0.6413066299251913,
|
1546 |
+
"eval_runtime": 15.0256,
|
1547 |
+
"eval_samples_per_second": 0.666,
|
1548 |
+
"eval_steps_per_second": 0.333,
|
1549 |
+
"step": 15400
|
1550 |
+
},
|
1551 |
+
{
|
1552 |
+
"epoch": 1.78,
|
1553 |
+
"grad_norm": 0.3436201512813568,
|
1554 |
+
"learning_rate": 0.00011115731905318027,
|
1555 |
+
"loss": 1.539,
|
1556 |
+
"step": 15600
|
1557 |
+
},
|
1558 |
+
{
|
1559 |
+
"epoch": 1.78,
|
1560 |
+
"eval_bertscore": 0.7335551977157593,
|
1561 |
+
"eval_loss": 1.7254730463027954,
|
1562 |
+
"eval_rouge1": 0.6388518781767971,
|
1563 |
+
"eval_rouge2": 0.3501853846588857,
|
1564 |
+
"eval_rougeL": 0.5196828245794569,
|
1565 |
+
"eval_rougeLsum": 0.629333993884722,
|
1566 |
+
"eval_runtime": 15.2595,
|
1567 |
+
"eval_samples_per_second": 0.655,
|
1568 |
+
"eval_steps_per_second": 0.328,
|
1569 |
+
"step": 15600
|
1570 |
+
},
|
1571 |
+
{
|
1572 |
+
"epoch": 1.8,
|
1573 |
+
"grad_norm": 0.3428190350532532,
|
1574 |
+
"learning_rate": 0.00011001794513914605,
|
1575 |
+
"loss": 1.5273,
|
1576 |
+
"step": 15800
|
1577 |
+
},
|
1578 |
+
{
|
1579 |
+
"epoch": 1.8,
|
1580 |
+
"eval_bertscore": 0.7331770658493042,
|
1581 |
+
"eval_loss": 1.7286018133163452,
|
1582 |
+
"eval_rouge1": 0.6581941047310954,
|
1583 |
+
"eval_rouge2": 0.36277983926897583,
|
1584 |
+
"eval_rougeL": 0.5336464680120501,
|
1585 |
+
"eval_rougeLsum": 0.6489239720278894,
|
1586 |
+
"eval_runtime": 14.8252,
|
1587 |
+
"eval_samples_per_second": 0.675,
|
1588 |
+
"eval_steps_per_second": 0.337,
|
1589 |
+
"step": 15800
|
1590 |
+
},
|
1591 |
+
{
|
1592 |
+
"epoch": 1.82,
|
1593 |
+
"grad_norm": 0.363164484500885,
|
1594 |
+
"learning_rate": 0.0001088785712251118,
|
1595 |
+
"loss": 1.5445,
|
1596 |
+
"step": 16000
|
1597 |
+
},
|
1598 |
+
{
|
1599 |
+
"epoch": 1.82,
|
1600 |
+
"eval_bertscore": 0.7377282977104187,
|
1601 |
+
"eval_loss": 1.7363064289093018,
|
1602 |
+
"eval_rouge1": 0.6547011011872876,
|
1603 |
+
"eval_rouge2": 0.3553220826957326,
|
1604 |
+
"eval_rougeL": 0.5256073814411315,
|
1605 |
+
"eval_rougeLsum": 0.6420095316923398,
|
1606 |
+
"eval_runtime": 14.8599,
|
1607 |
+
"eval_samples_per_second": 0.673,
|
1608 |
+
"eval_steps_per_second": 0.336,
|
1609 |
+
"step": 16000
|
1610 |
+
},
|
1611 |
+
{
|
1612 |
+
"epoch": 1.85,
|
1613 |
+
"grad_norm": 0.3098333775997162,
|
1614 |
+
"learning_rate": 0.00010773919731107757,
|
1615 |
+
"loss": 1.5319,
|
1616 |
+
"step": 16200
|
1617 |
+
},
|
1618 |
+
{
|
1619 |
+
"epoch": 1.85,
|
1620 |
+
"eval_bertscore": 0.7324053645133972,
|
1621 |
+
"eval_loss": 1.7284066677093506,
|
1622 |
+
"eval_rouge1": 0.6477379941950916,
|
1623 |
+
"eval_rouge2": 0.3535918140554809,
|
1624 |
+
"eval_rougeL": 0.5226838544730126,
|
1625 |
+
"eval_rougeLsum": 0.6373271915355557,
|
1626 |
+
"eval_runtime": 14.9318,
|
1627 |
+
"eval_samples_per_second": 0.67,
|
1628 |
+
"eval_steps_per_second": 0.335,
|
1629 |
+
"step": 16200
|
1630 |
+
},
|
1631 |
+
{
|
1632 |
+
"epoch": 1.87,
|
1633 |
+
"grad_norm": 0.3637208938598633,
|
1634 |
+
"learning_rate": 0.00010659982339704332,
|
1635 |
+
"loss": 1.5442,
|
1636 |
+
"step": 16400
|
1637 |
+
},
|
1638 |
+
{
|
1639 |
+
"epoch": 1.87,
|
1640 |
+
"eval_bertscore": 0.7347462773323059,
|
1641 |
+
"eval_loss": 1.7252963781356812,
|
1642 |
+
"eval_rouge1": 0.6494449840449819,
|
1643 |
+
"eval_rouge2": 0.3586550050575282,
|
1644 |
+
"eval_rougeL": 0.5275675395159809,
|
1645 |
+
"eval_rougeLsum": 0.6396738714026391,
|
1646 |
+
"eval_runtime": 15.2218,
|
1647 |
+
"eval_samples_per_second": 0.657,
|
1648 |
+
"eval_steps_per_second": 0.328,
|
1649 |
+
"step": 16400
|
1650 |
+
},
|
1651 |
+
{
|
1652 |
+
"epoch": 1.89,
|
1653 |
+
"grad_norm": 0.35197457671165466,
|
1654 |
+
"learning_rate": 0.00010546044948300908,
|
1655 |
+
"loss": 1.5131,
|
1656 |
+
"step": 16600
|
1657 |
+
},
|
1658 |
+
{
|
1659 |
+
"epoch": 1.89,
|
1660 |
+
"eval_bertscore": 0.7329785227775574,
|
1661 |
+
"eval_loss": 1.7285687923431396,
|
1662 |
+
"eval_rouge1": 0.6582047811143328,
|
1663 |
+
"eval_rouge2": 0.3637700686094697,
|
1664 |
+
"eval_rougeL": 0.5355021948480279,
|
1665 |
+
"eval_rougeLsum": 0.6483245595148677,
|
1666 |
+
"eval_runtime": 14.8772,
|
1667 |
+
"eval_samples_per_second": 0.672,
|
1668 |
+
"eval_steps_per_second": 0.336,
|
1669 |
+
"step": 16600
|
1670 |
+
},
|
1671 |
+
{
|
1672 |
+
"epoch": 1.91,
|
1673 |
+
"grad_norm": 0.3406757116317749,
|
1674 |
+
"learning_rate": 0.00010432107556897486,
|
1675 |
+
"loss": 1.5394,
|
1676 |
+
"step": 16800
|
1677 |
+
},
|
1678 |
+
{
|
1679 |
+
"epoch": 1.91,
|
1680 |
+
"eval_bertscore": 0.7345961332321167,
|
1681 |
+
"eval_loss": 1.7324028015136719,
|
1682 |
+
"eval_rouge1": 0.6408293615351552,
|
1683 |
+
"eval_rouge2": 0.3520120690778129,
|
1684 |
+
"eval_rougeL": 0.5145218014745592,
|
1685 |
+
"eval_rougeLsum": 0.6297802607384266,
|
1686 |
+
"eval_runtime": 15.1044,
|
1687 |
+
"eval_samples_per_second": 0.662,
|
1688 |
+
"eval_steps_per_second": 0.331,
|
1689 |
+
"step": 16800
|
1690 |
+
},
|
1691 |
+
{
|
1692 |
+
"epoch": 1.94,
|
1693 |
+
"grad_norm": 0.3417683243751526,
|
1694 |
+
"learning_rate": 0.00010318170165494061,
|
1695 |
+
"loss": 1.526,
|
1696 |
+
"step": 17000
|
1697 |
+
},
|
1698 |
+
{
|
1699 |
+
"epoch": 1.94,
|
1700 |
+
"eval_bertscore": 0.735752522945404,
|
1701 |
+
"eval_loss": 1.7288110256195068,
|
1702 |
+
"eval_rouge1": 0.641158513352794,
|
1703 |
+
"eval_rouge2": 0.3544166440855814,
|
1704 |
+
"eval_rougeL": 0.5215201980495414,
|
1705 |
+
"eval_rougeLsum": 0.630550065494593,
|
1706 |
+
"eval_runtime": 15.0797,
|
1707 |
+
"eval_samples_per_second": 0.663,
|
1708 |
+
"eval_steps_per_second": 0.332,
|
1709 |
+
"step": 17000
|
1710 |
+
},
|
1711 |
+
{
|
1712 |
+
"epoch": 1.96,
|
1713 |
+
"grad_norm": 0.3256611227989197,
|
1714 |
+
"learning_rate": 0.00010204232774090639,
|
1715 |
+
"loss": 1.5484,
|
1716 |
+
"step": 17200
|
1717 |
+
},
|
1718 |
+
{
|
1719 |
+
"epoch": 1.96,
|
1720 |
+
"eval_bertscore": 0.7356327772140503,
|
1721 |
+
"eval_loss": 1.7305186986923218,
|
1722 |
+
"eval_rouge1": 0.6400269226515611,
|
1723 |
+
"eval_rouge2": 0.3502884634173268,
|
1724 |
+
"eval_rougeL": 0.517312321281175,
|
1725 |
+
"eval_rougeLsum": 0.6284556997614409,
|
1726 |
+
"eval_runtime": 15.4097,
|
1727 |
+
"eval_samples_per_second": 0.649,
|
1728 |
+
"eval_steps_per_second": 0.324,
|
1729 |
+
"step": 17200
|
1730 |
+
},
|
1731 |
+
{
|
1732 |
+
"epoch": 1.98,
|
1733 |
+
"grad_norm": 0.4035187363624573,
|
1734 |
+
"learning_rate": 0.00010090295382687213,
|
1735 |
+
"loss": 1.5261,
|
1736 |
+
"step": 17400
|
1737 |
+
},
|
1738 |
+
{
|
1739 |
+
"epoch": 1.98,
|
1740 |
+
"eval_bertscore": 0.7339992523193359,
|
1741 |
+
"eval_loss": 1.7282793521881104,
|
1742 |
+
"eval_rouge1": 0.6335770390416183,
|
1743 |
+
"eval_rouge2": 0.34592404578075897,
|
1744 |
+
"eval_rougeL": 0.5109045259792113,
|
1745 |
+
"eval_rougeLsum": 0.6218413683710426,
|
1746 |
+
"eval_runtime": 15.1959,
|
1747 |
+
"eval_samples_per_second": 0.658,
|
1748 |
+
"eval_steps_per_second": 0.329,
|
1749 |
+
"step": 17400
|
1750 |
+
},
|
1751 |
+
{
|
1752 |
+
"epoch": 2.0,
|
1753 |
+
"grad_norm": 0.34987062215805054,
|
1754 |
+
"learning_rate": 9.97635799128379e-05,
|
1755 |
+
"loss": 1.5199,
|
1756 |
+
"step": 17600
|
1757 |
+
},
|
1758 |
+
{
|
1759 |
+
"epoch": 2.0,
|
1760 |
+
"eval_bertscore": 0.7326329946517944,
|
1761 |
+
"eval_loss": 1.7544715404510498,
|
1762 |
+
"eval_rouge1": 0.6451558045750205,
|
1763 |
+
"eval_rouge2": 0.35565806935653943,
|
1764 |
+
"eval_rougeL": 0.5217034865840529,
|
1765 |
+
"eval_rougeLsum": 0.6329869715356753,
|
1766 |
+
"eval_runtime": 15.0351,
|
1767 |
+
"eval_samples_per_second": 0.665,
|
1768 |
+
"eval_steps_per_second": 0.333,
|
1769 |
+
"step": 17600
|
1770 |
+
},
|
1771 |
+
{
|
1772 |
+
"epoch": 2.03,
|
1773 |
+
"grad_norm": 0.37184038758277893,
|
1774 |
+
"learning_rate": 9.862420599880366e-05,
|
1775 |
+
"loss": 1.41,
|
1776 |
+
"step": 17800
|
1777 |
+
},
|
1778 |
+
{
|
1779 |
+
"epoch": 2.03,
|
1780 |
+
"eval_bertscore": 0.7315141558647156,
|
1781 |
+
"eval_loss": 1.7585878372192383,
|
1782 |
+
"eval_rouge1": 0.6469319193583706,
|
1783 |
+
"eval_rouge2": 0.3514447211469598,
|
1784 |
+
"eval_rougeL": 0.524755857688278,
|
1785 |
+
"eval_rougeLsum": 0.6350164781858667,
|
1786 |
+
"eval_runtime": 14.9583,
|
1787 |
+
"eval_samples_per_second": 0.669,
|
1788 |
+
"eval_steps_per_second": 0.334,
|
1789 |
+
"step": 17800
|
1790 |
+
},
|
1791 |
+
{
|
1792 |
+
"epoch": 2.05,
|
1793 |
+
"grad_norm": 0.3812776803970337,
|
1794 |
+
"learning_rate": 9.748483208476943e-05,
|
1795 |
+
"loss": 1.4132,
|
1796 |
+
"step": 18000
|
1797 |
+
},
|
1798 |
+
{
|
1799 |
+
"epoch": 2.05,
|
1800 |
+
"eval_bertscore": 0.7335561513900757,
|
1801 |
+
"eval_loss": 1.764611840248108,
|
1802 |
+
"eval_rouge1": 0.6381916780473581,
|
1803 |
+
"eval_rouge2": 0.3482510604092539,
|
1804 |
+
"eval_rougeL": 0.5162105225823392,
|
1805 |
+
"eval_rougeLsum": 0.627150245441782,
|
1806 |
+
"eval_runtime": 15.8515,
|
1807 |
+
"eval_samples_per_second": 0.631,
|
1808 |
+
"eval_steps_per_second": 0.315,
|
1809 |
+
"step": 18000
|
1810 |
+
},
|
1811 |
+
{
|
1812 |
+
"epoch": 2.07,
|
1813 |
+
"grad_norm": 0.45525220036506653,
|
1814 |
+
"learning_rate": 9.634545817073518e-05,
|
1815 |
+
"loss": 1.4,
|
1816 |
+
"step": 18200
|
1817 |
+
},
|
1818 |
+
{
|
1819 |
+
"epoch": 2.07,
|
1820 |
+
"eval_bertscore": 0.73627769947052,
|
1821 |
+
"eval_loss": 1.7585163116455078,
|
1822 |
+
"eval_rouge1": 0.6670097658027134,
|
1823 |
+
"eval_rouge2": 0.3658295359911405,
|
1824 |
+
"eval_rougeL": 0.5429667657900548,
|
1825 |
+
"eval_rougeLsum": 0.6543501745791419,
|
1826 |
+
"eval_runtime": 15.0301,
|
1827 |
+
"eval_samples_per_second": 0.665,
|
1828 |
+
"eval_steps_per_second": 0.333,
|
1829 |
+
"step": 18200
|
1830 |
+
},
|
1831 |
+
{
|
1832 |
+
"epoch": 2.1,
|
1833 |
+
"grad_norm": 0.37322184443473816,
|
1834 |
+
"learning_rate": 9.520608425670095e-05,
|
1835 |
+
"loss": 1.4293,
|
1836 |
+
"step": 18400
|
1837 |
+
},
|
1838 |
+
{
|
1839 |
+
"epoch": 2.1,
|
1840 |
+
"eval_bertscore": 0.730435848236084,
|
1841 |
+
"eval_loss": 1.764052391052246,
|
1842 |
+
"eval_rouge1": 0.6640215078213034,
|
1843 |
+
"eval_rouge2": 0.3625932287322054,
|
1844 |
+
"eval_rougeL": 0.5379978391335138,
|
1845 |
+
"eval_rougeLsum": 0.6542054656293199,
|
1846 |
+
"eval_runtime": 15.0762,
|
1847 |
+
"eval_samples_per_second": 0.663,
|
1848 |
+
"eval_steps_per_second": 0.332,
|
1849 |
+
"step": 18400
|
1850 |
+
},
|
1851 |
+
{
|
1852 |
+
"epoch": 2.12,
|
1853 |
+
"grad_norm": 0.4260891079902649,
|
1854 |
+
"learning_rate": 9.40667103426667e-05,
|
1855 |
+
"loss": 1.4077,
|
1856 |
+
"step": 18600
|
1857 |
+
},
|
1858 |
+
{
|
1859 |
+
"epoch": 2.12,
|
1860 |
+
"eval_bertscore": 0.7309869527816772,
|
1861 |
+
"eval_loss": 1.762108564376831,
|
1862 |
+
"eval_rouge1": 0.6571171081737958,
|
1863 |
+
"eval_rouge2": 0.35780421333141865,
|
1864 |
+
"eval_rougeL": 0.5320129270967632,
|
1865 |
+
"eval_rougeLsum": 0.64587787409523,
|
1866 |
+
"eval_runtime": 14.9004,
|
1867 |
+
"eval_samples_per_second": 0.671,
|
1868 |
+
"eval_steps_per_second": 0.336,
|
1869 |
+
"step": 18600
|
1870 |
+
},
|
1871 |
+
{
|
1872 |
+
"epoch": 2.14,
|
1873 |
+
"grad_norm": 0.39479926228523254,
|
1874 |
+
"learning_rate": 9.292733642863247e-05,
|
1875 |
+
"loss": 1.4165,
|
1876 |
+
"step": 18800
|
1877 |
+
},
|
1878 |
+
{
|
1879 |
+
"epoch": 2.14,
|
1880 |
+
"eval_bertscore": 0.7324444651603699,
|
1881 |
+
"eval_loss": 1.7607113122940063,
|
1882 |
+
"eval_rouge1": 0.6628398862884018,
|
1883 |
+
"eval_rouge2": 0.3627259806721216,
|
1884 |
+
"eval_rougeL": 0.5366106483832656,
|
1885 |
+
"eval_rougeLsum": 0.6528364858807157,
|
1886 |
+
"eval_runtime": 15.5766,
|
1887 |
+
"eval_samples_per_second": 0.642,
|
1888 |
+
"eval_steps_per_second": 0.321,
|
1889 |
+
"step": 18800
|
1890 |
+
},
|
1891 |
+
{
|
1892 |
+
"epoch": 2.16,
|
1893 |
+
"grad_norm": 0.39267703890800476,
|
1894 |
+
"learning_rate": 9.178796251459824e-05,
|
1895 |
+
"loss": 1.4123,
|
1896 |
+
"step": 19000
|
1897 |
+
},
|
1898 |
+
{
|
1899 |
+
"epoch": 2.16,
|
1900 |
+
"eval_bertscore": 0.7298994064331055,
|
1901 |
+
"eval_loss": 1.7668545246124268,
|
1902 |
+
"eval_rouge1": 0.6490850022857569,
|
1903 |
+
"eval_rouge2": 0.3532323419511264,
|
1904 |
+
"eval_rougeL": 0.5212823000193295,
|
1905 |
+
"eval_rougeLsum": 0.636442724466695,
|
1906 |
+
"eval_runtime": 14.9094,
|
1907 |
+
"eval_samples_per_second": 0.671,
|
1908 |
+
"eval_steps_per_second": 0.335,
|
1909 |
+
"step": 19000
|
1910 |
+
},
|
1911 |
+
{
|
1912 |
+
"epoch": 2.19,
|
1913 |
+
"grad_norm": 0.38221287727355957,
|
1914 |
+
"learning_rate": 9.0648588600564e-05,
|
1915 |
+
"loss": 1.401,
|
1916 |
+
"step": 19200
|
1917 |
+
},
|
1918 |
+
{
|
1919 |
+
"epoch": 2.19,
|
1920 |
+
"eval_bertscore": 0.7316875457763672,
|
1921 |
+
"eval_loss": 1.764147400856018,
|
1922 |
+
"eval_rouge1": 0.6490326710625849,
|
1923 |
+
"eval_rouge2": 0.3510351037900723,
|
1924 |
+
"eval_rougeL": 0.5239165028795836,
|
1925 |
+
"eval_rougeLsum": 0.6373687316421427,
|
1926 |
+
"eval_runtime": 15.1192,
|
1927 |
+
"eval_samples_per_second": 0.661,
|
1928 |
+
"eval_steps_per_second": 0.331,
|
1929 |
+
"step": 19200
|
1930 |
+
},
|
1931 |
+
{
|
1932 |
+
"epoch": 2.21,
|
1933 |
+
"grad_norm": 0.3653150200843811,
|
1934 |
+
"learning_rate": 8.950921468652976e-05,
|
1935 |
+
"loss": 1.4109,
|
1936 |
+
"step": 19400
|
1937 |
+
},
|
1938 |
+
{
|
1939 |
+
"epoch": 2.21,
|
1940 |
+
"eval_bertscore": 0.7348155975341797,
|
1941 |
+
"eval_loss": 1.7640550136566162,
|
1942 |
+
"eval_rouge1": 0.6462152873276823,
|
1943 |
+
"eval_rouge2": 0.3483599145461069,
|
1944 |
+
"eval_rougeL": 0.5193372430687719,
|
1945 |
+
"eval_rougeLsum": 0.6334254357511564,
|
1946 |
+
"eval_runtime": 14.9291,
|
1947 |
+
"eval_samples_per_second": 0.67,
|
1948 |
+
"eval_steps_per_second": 0.335,
|
1949 |
+
"step": 19400
|
1950 |
+
},
|
1951 |
+
{
|
1952 |
+
"epoch": 2.23,
|
1953 |
+
"grad_norm": 0.38049009442329407,
|
1954 |
+
"learning_rate": 8.836984077249551e-05,
|
1955 |
+
"loss": 1.4189,
|
1956 |
+
"step": 19600
|
1957 |
+
},
|
1958 |
+
{
|
1959 |
+
"epoch": 2.23,
|
1960 |
+
"eval_bertscore": 0.7357938885688782,
|
1961 |
+
"eval_loss": 1.7696326971054077,
|
1962 |
+
"eval_rouge1": 0.6377276221057538,
|
1963 |
+
"eval_rouge2": 0.3455397190390045,
|
1964 |
+
"eval_rougeL": 0.5118069428064842,
|
1965 |
+
"eval_rougeLsum": 0.6264501633078481,
|
1966 |
+
"eval_runtime": 14.8653,
|
1967 |
+
"eval_samples_per_second": 0.673,
|
1968 |
+
"eval_steps_per_second": 0.336,
|
1969 |
+
"step": 19600
|
1970 |
+
},
|
1971 |
+
{
|
1972 |
+
"epoch": 2.26,
|
1973 |
+
"grad_norm": 0.42111098766326904,
|
1974 |
+
"learning_rate": 8.723046685846128e-05,
|
1975 |
+
"loss": 1.4152,
|
1976 |
+
"step": 19800
|
1977 |
+
},
|
1978 |
+
{
|
1979 |
+
"epoch": 2.26,
|
1980 |
+
"eval_bertscore": 0.7339056134223938,
|
1981 |
+
"eval_loss": 1.7658218145370483,
|
1982 |
+
"eval_rouge1": 0.6494820372695989,
|
1983 |
+
"eval_rouge2": 0.34691658128805236,
|
1984 |
+
"eval_rougeL": 0.5193228965163086,
|
1985 |
+
"eval_rougeLsum": 0.6365347065687565,
|
1986 |
+
"eval_runtime": 15.3562,
|
1987 |
+
"eval_samples_per_second": 0.651,
|
1988 |
+
"eval_steps_per_second": 0.326,
|
1989 |
+
"step": 19800
|
1990 |
+
},
|
1991 |
+
{
|
1992 |
+
"epoch": 2.28,
|
1993 |
+
"grad_norm": 0.4452258050441742,
|
1994 |
+
"learning_rate": 8.609109294442704e-05,
|
1995 |
+
"loss": 1.4101,
|
1996 |
+
"step": 20000
|
1997 |
+
},
|
1998 |
+
{
|
1999 |
+
"epoch": 2.28,
|
2000 |
+
"eval_bertscore": 0.7296434640884399,
|
2001 |
+
"eval_loss": 1.7714240550994873,
|
2002 |
+
"eval_rouge1": 0.6565600824405751,
|
2003 |
+
"eval_rouge2": 0.3533618655201594,
|
2004 |
+
"eval_rougeL": 0.5263318202066467,
|
2005 |
+
"eval_rougeLsum": 0.6444964824298407,
|
2006 |
+
"eval_runtime": 14.8578,
|
2007 |
+
"eval_samples_per_second": 0.673,
|
2008 |
+
"eval_steps_per_second": 0.337,
|
2009 |
+
"step": 20000
|
2010 |
+
},
|
2011 |
+
{
|
2012 |
+
"epoch": 2.3,
|
2013 |
+
"grad_norm": 0.4030967652797699,
|
2014 |
+
"learning_rate": 8.495171903039281e-05,
|
2015 |
+
"loss": 1.4049,
|
2016 |
+
"step": 20200
|
2017 |
+
},
|
2018 |
+
{
|
2019 |
+
"epoch": 2.3,
|
2020 |
+
"eval_bertscore": 0.7307097315788269,
|
2021 |
+
"eval_loss": 1.774444580078125,
|
2022 |
+
"eval_rouge1": 0.6517204836155526,
|
2023 |
+
"eval_rouge2": 0.3521339653276223,
|
2024 |
+
"eval_rougeL": 0.5223211728244184,
|
2025 |
+
"eval_rougeLsum": 0.6398710531932736,
|
2026 |
+
"eval_runtime": 15.8543,
|
2027 |
+
"eval_samples_per_second": 0.631,
|
2028 |
+
"eval_steps_per_second": 0.315,
|
2029 |
+
"step": 20200
|
2030 |
+
},
|
2031 |
+
{
|
2032 |
+
"epoch": 2.32,
|
2033 |
+
"grad_norm": 0.33409813046455383,
|
2034 |
+
"learning_rate": 8.381234511635858e-05,
|
2035 |
+
"loss": 1.4243,
|
2036 |
+
"step": 20400
|
2037 |
+
},
|
2038 |
+
{
|
2039 |
+
"epoch": 2.32,
|
2040 |
+
"eval_bertscore": 0.7312101721763611,
|
2041 |
+
"eval_loss": 1.7654094696044922,
|
2042 |
+
"eval_rouge1": 0.6607249126293291,
|
2043 |
+
"eval_rouge2": 0.3545993249716188,
|
2044 |
+
"eval_rougeL": 0.5320161007986739,
|
2045 |
+
"eval_rougeLsum": 0.6503315335963733,
|
2046 |
+
"eval_runtime": 14.8739,
|
2047 |
+
"eval_samples_per_second": 0.672,
|
2048 |
+
"eval_steps_per_second": 0.336,
|
2049 |
+
"step": 20400
|
2050 |
+
},
|
2051 |
+
{
|
2052 |
+
"epoch": 2.35,
|
2053 |
+
"grad_norm": 0.4044789671897888,
|
2054 |
+
"learning_rate": 8.267297120232433e-05,
|
2055 |
+
"loss": 1.413,
|
2056 |
+
"step": 20600
|
2057 |
+
},
|
2058 |
+
{
|
2059 |
+
"epoch": 2.35,
|
2060 |
+
"eval_bertscore": 0.7342169880867004,
|
2061 |
+
"eval_loss": 1.769879937171936,
|
2062 |
+
"eval_rouge1": 0.6442777880355144,
|
2063 |
+
"eval_rouge2": 0.35006080708477183,
|
2064 |
+
"eval_rougeL": 0.5218799478770955,
|
2065 |
+
"eval_rougeLsum": 0.6332700294558067,
|
2066 |
+
"eval_runtime": 14.9089,
|
2067 |
+
"eval_samples_per_second": 0.671,
|
2068 |
+
"eval_steps_per_second": 0.335,
|
2069 |
+
"step": 20600
|
2070 |
+
},
|
2071 |
+
{
|
2072 |
+
"epoch": 2.37,
|
2073 |
+
"grad_norm": 0.39801183342933655,
|
2074 |
+
"learning_rate": 8.153359728829008e-05,
|
2075 |
+
"loss": 1.4177,
|
2076 |
+
"step": 20800
|
2077 |
+
},
|
2078 |
+
{
|
2079 |
+
"epoch": 2.37,
|
2080 |
+
"eval_bertscore": 0.7343758344650269,
|
2081 |
+
"eval_loss": 1.7737929821014404,
|
2082 |
+
"eval_rouge1": 0.6495678172205896,
|
2083 |
+
"eval_rouge2": 0.3505195734345703,
|
2084 |
+
"eval_rougeL": 0.5263025592812188,
|
2085 |
+
"eval_rougeLsum": 0.6390057749428748,
|
2086 |
+
"eval_runtime": 15.2148,
|
2087 |
+
"eval_samples_per_second": 0.657,
|
2088 |
+
"eval_steps_per_second": 0.329,
|
2089 |
+
"step": 20800
|
2090 |
+
}
|
2091 |
+
],
|
2092 |
+
"logging_steps": 200,
|
2093 |
+
"max_steps": 35112,
|
2094 |
+
"num_input_tokens_seen": 0,
|
2095 |
+
"num_train_epochs": 4,
|
2096 |
+
"save_steps": 800,
|
2097 |
+
"total_flos": 2.1077602489057567e+18,
|
2098 |
+
"train_batch_size": 2,
|
2099 |
+
"trial_name": null,
|
2100 |
+
"trial_params": null
|
2101 |
+
}
|
checkpoint-20800/training_args.bin
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:d576743d4294cfbcae212243377437f73cabbcded2df9da7f73e6e39ecb38f56
|
3 |
+
size 5048
|
checkpoint-21600/README.md
ADDED
@@ -0,0 +1,202 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
---
|
2 |
+
library_name: peft
|
3 |
+
base_model: unsloth/gemma-2b-bnb-4bit
|
4 |
+
---
|
5 |
+
|
6 |
+
# Model Card for Model ID
|
7 |
+
|
8 |
+
<!-- Provide a quick summary of what the model is/does. -->
|
9 |
+
|
10 |
+
|
11 |
+
|
12 |
+
## Model Details
|
13 |
+
|
14 |
+
### Model Description
|
15 |
+
|
16 |
+
<!-- Provide a longer summary of what this model is. -->
|
17 |
+
|
18 |
+
|
19 |
+
|
20 |
+
- **Developed by:** [More Information Needed]
|
21 |
+
- **Funded by [optional]:** [More Information Needed]
|
22 |
+
- **Shared by [optional]:** [More Information Needed]
|
23 |
+
- **Model type:** [More Information Needed]
|
24 |
+
- **Language(s) (NLP):** [More Information Needed]
|
25 |
+
- **License:** [More Information Needed]
|
26 |
+
- **Finetuned from model [optional]:** [More Information Needed]
|
27 |
+
|
28 |
+
### Model Sources [optional]
|
29 |
+
|
30 |
+
<!-- Provide the basic links for the model. -->
|
31 |
+
|
32 |
+
- **Repository:** [More Information Needed]
|
33 |
+
- **Paper [optional]:** [More Information Needed]
|
34 |
+
- **Demo [optional]:** [More Information Needed]
|
35 |
+
|
36 |
+
## Uses
|
37 |
+
|
38 |
+
<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
|
39 |
+
|
40 |
+
### Direct Use
|
41 |
+
|
42 |
+
<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
|
43 |
+
|
44 |
+
[More Information Needed]
|
45 |
+
|
46 |
+
### Downstream Use [optional]
|
47 |
+
|
48 |
+
<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
|
49 |
+
|
50 |
+
[More Information Needed]
|
51 |
+
|
52 |
+
### Out-of-Scope Use
|
53 |
+
|
54 |
+
<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
|
55 |
+
|
56 |
+
[More Information Needed]
|
57 |
+
|
58 |
+
## Bias, Risks, and Limitations
|
59 |
+
|
60 |
+
<!-- This section is meant to convey both technical and sociotechnical limitations. -->
|
61 |
+
|
62 |
+
[More Information Needed]
|
63 |
+
|
64 |
+
### Recommendations
|
65 |
+
|
66 |
+
<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
|
67 |
+
|
68 |
+
Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
|
69 |
+
|
70 |
+
## How to Get Started with the Model
|
71 |
+
|
72 |
+
Use the code below to get started with the model.
|
73 |
+
|
74 |
+
[More Information Needed]
|
75 |
+
|
76 |
+
## Training Details
|
77 |
+
|
78 |
+
### Training Data
|
79 |
+
|
80 |
+
<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
|
81 |
+
|
82 |
+
[More Information Needed]
|
83 |
+
|
84 |
+
### Training Procedure
|
85 |
+
|
86 |
+
<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
|
87 |
+
|
88 |
+
#### Preprocessing [optional]
|
89 |
+
|
90 |
+
[More Information Needed]
|
91 |
+
|
92 |
+
|
93 |
+
#### Training Hyperparameters
|
94 |
+
|
95 |
+
- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
|
96 |
+
|
97 |
+
#### Speeds, Sizes, Times [optional]
|
98 |
+
|
99 |
+
<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
|
100 |
+
|
101 |
+
[More Information Needed]
|
102 |
+
|
103 |
+
## Evaluation
|
104 |
+
|
105 |
+
<!-- This section describes the evaluation protocols and provides the results. -->
|
106 |
+
|
107 |
+
### Testing Data, Factors & Metrics
|
108 |
+
|
109 |
+
#### Testing Data
|
110 |
+
|
111 |
+
<!-- This should link to a Dataset Card if possible. -->
|
112 |
+
|
113 |
+
[More Information Needed]
|
114 |
+
|
115 |
+
#### Factors
|
116 |
+
|
117 |
+
<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
|
118 |
+
|
119 |
+
[More Information Needed]
|
120 |
+
|
121 |
+
#### Metrics
|
122 |
+
|
123 |
+
<!-- These are the evaluation metrics being used, ideally with a description of why. -->
|
124 |
+
|
125 |
+
[More Information Needed]
|
126 |
+
|
127 |
+
### Results
|
128 |
+
|
129 |
+
[More Information Needed]
|
130 |
+
|
131 |
+
#### Summary
|
132 |
+
|
133 |
+
|
134 |
+
|
135 |
+
## Model Examination [optional]
|
136 |
+
|
137 |
+
<!-- Relevant interpretability work for the model goes here -->
|
138 |
+
|
139 |
+
[More Information Needed]
|
140 |
+
|
141 |
+
## Environmental Impact
|
142 |
+
|
143 |
+
<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
|
144 |
+
|
145 |
+
Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
|
146 |
+
|
147 |
+
- **Hardware Type:** [More Information Needed]
|
148 |
+
- **Hours used:** [More Information Needed]
|
149 |
+
- **Cloud Provider:** [More Information Needed]
|
150 |
+
- **Compute Region:** [More Information Needed]
|
151 |
+
- **Carbon Emitted:** [More Information Needed]
|
152 |
+
|
153 |
+
## Technical Specifications [optional]
|
154 |
+
|
155 |
+
### Model Architecture and Objective
|
156 |
+
|
157 |
+
[More Information Needed]
|
158 |
+
|
159 |
+
### Compute Infrastructure
|
160 |
+
|
161 |
+
[More Information Needed]
|
162 |
+
|
163 |
+
#### Hardware
|
164 |
+
|
165 |
+
[More Information Needed]
|
166 |
+
|
167 |
+
#### Software
|
168 |
+
|
169 |
+
[More Information Needed]
|
170 |
+
|
171 |
+
## Citation [optional]
|
172 |
+
|
173 |
+
<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
|
174 |
+
|
175 |
+
**BibTeX:**
|
176 |
+
|
177 |
+
[More Information Needed]
|
178 |
+
|
179 |
+
**APA:**
|
180 |
+
|
181 |
+
[More Information Needed]
|
182 |
+
|
183 |
+
## Glossary [optional]
|
184 |
+
|
185 |
+
<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
|
186 |
+
|
187 |
+
[More Information Needed]
|
188 |
+
|
189 |
+
## More Information [optional]
|
190 |
+
|
191 |
+
[More Information Needed]
|
192 |
+
|
193 |
+
## Model Card Authors [optional]
|
194 |
+
|
195 |
+
[More Information Needed]
|
196 |
+
|
197 |
+
## Model Card Contact
|
198 |
+
|
199 |
+
[More Information Needed]
|
200 |
+
### Framework versions
|
201 |
+
|
202 |
+
- PEFT 0.10.0
|
checkpoint-21600/adapter_config.json
ADDED
@@ -0,0 +1,34 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"alpha_pattern": {},
|
3 |
+
"auto_mapping": null,
|
4 |
+
"base_model_name_or_path": "unsloth/gemma-2b-bnb-4bit",
|
5 |
+
"bias": "none",
|
6 |
+
"fan_in_fan_out": false,
|
7 |
+
"inference_mode": true,
|
8 |
+
"init_lora_weights": true,
|
9 |
+
"layer_replication": null,
|
10 |
+
"layers_pattern": null,
|
11 |
+
"layers_to_transform": null,
|
12 |
+
"loftq_config": {},
|
13 |
+
"lora_alpha": 32,
|
14 |
+
"lora_dropout": 0,
|
15 |
+
"megatron_config": null,
|
16 |
+
"megatron_core": "megatron.core",
|
17 |
+
"modules_to_save": null,
|
18 |
+
"peft_type": "LORA",
|
19 |
+
"r": 64,
|
20 |
+
"rank_pattern": {},
|
21 |
+
"revision": "unsloth",
|
22 |
+
"target_modules": [
|
23 |
+
"v_proj",
|
24 |
+
"up_proj",
|
25 |
+
"gate_proj",
|
26 |
+
"o_proj",
|
27 |
+
"k_proj",
|
28 |
+
"down_proj",
|
29 |
+
"q_proj"
|
30 |
+
],
|
31 |
+
"task_type": "CAUSAL_LM",
|
32 |
+
"use_dora": false,
|
33 |
+
"use_rslora": false
|
34 |
+
}
|
checkpoint-21600/adapter_model.safetensors
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:5a9fcd274c0981e09f7a56fc7df572e5f0b95bc58848f8bf7531f5054e18c99c
|
3 |
+
size 313820248
|
checkpoint-21600/rng_state.pth
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:6a56b97f12072d267caf15b1f0d1101714ba6380a7c60c5a83e68808dc36d773
|
3 |
+
size 14244
|
checkpoint-21600/scheduler.pt
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:a4ca451b22b6741c0b9a90d4a0705be6a2e71a2b099ab0a034a9a7bd72141ef9
|
3 |
+
size 1064
|
checkpoint-21600/special_tokens_map.json
ADDED
@@ -0,0 +1,34 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"additional_special_tokens": [
|
3 |
+
"<start_of_turn>",
|
4 |
+
"<end_of_turn>"
|
5 |
+
],
|
6 |
+
"bos_token": {
|
7 |
+
"content": "<bos>",
|
8 |
+
"lstrip": false,
|
9 |
+
"normalized": false,
|
10 |
+
"rstrip": false,
|
11 |
+
"single_word": false
|
12 |
+
},
|
13 |
+
"eos_token": {
|
14 |
+
"content": "<eos>",
|
15 |
+
"lstrip": false,
|
16 |
+
"normalized": false,
|
17 |
+
"rstrip": false,
|
18 |
+
"single_word": false
|
19 |
+
},
|
20 |
+
"pad_token": {
|
21 |
+
"content": "<pad>",
|
22 |
+
"lstrip": false,
|
23 |
+
"normalized": false,
|
24 |
+
"rstrip": false,
|
25 |
+
"single_word": false
|
26 |
+
},
|
27 |
+
"unk_token": {
|
28 |
+
"content": "<unk>",
|
29 |
+
"lstrip": false,
|
30 |
+
"normalized": false,
|
31 |
+
"rstrip": false,
|
32 |
+
"single_word": false
|
33 |
+
}
|
34 |
+
}
|
checkpoint-21600/tokenizer.json
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:05e97791a5e007260de1db7e1692e53150e08cea481e2bf25435553380c147ee
|
3 |
+
size 17477929
|
checkpoint-21600/tokenizer.model
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:61a7b147390c64585d6c3543dd6fc636906c9af3865a5548f27f31aee1d4c8e2
|
3 |
+
size 4241003
|
checkpoint-21600/tokenizer_config.json
ADDED
@@ -0,0 +1,71 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"add_bos_token": true,
|
3 |
+
"add_eos_token": false,
|
4 |
+
"added_tokens_decoder": {
|
5 |
+
"0": {
|
6 |
+
"content": "<pad>",
|
7 |
+
"lstrip": false,
|
8 |
+
"normalized": false,
|
9 |
+
"rstrip": false,
|
10 |
+
"single_word": false,
|
11 |
+
"special": true
|
12 |
+
},
|
13 |
+
"1": {
|
14 |
+
"content": "<eos>",
|
15 |
+
"lstrip": false,
|
16 |
+
"normalized": false,
|
17 |
+
"rstrip": false,
|
18 |
+
"single_word": false,
|
19 |
+
"special": true
|
20 |
+
},
|
21 |
+
"2": {
|
22 |
+
"content": "<bos>",
|
23 |
+
"lstrip": false,
|
24 |
+
"normalized": false,
|
25 |
+
"rstrip": false,
|
26 |
+
"single_word": false,
|
27 |
+
"special": true
|
28 |
+
},
|
29 |
+
"3": {
|
30 |
+
"content": "<unk>",
|
31 |
+
"lstrip": false,
|
32 |
+
"normalized": false,
|
33 |
+
"rstrip": false,
|
34 |
+
"single_word": false,
|
35 |
+
"special": true
|
36 |
+
},
|
37 |
+
"106": {
|
38 |
+
"content": "<start_of_turn>",
|
39 |
+
"lstrip": false,
|
40 |
+
"normalized": false,
|
41 |
+
"rstrip": false,
|
42 |
+
"single_word": false,
|
43 |
+
"special": true
|
44 |
+
},
|
45 |
+
"107": {
|
46 |
+
"content": "<end_of_turn>",
|
47 |
+
"lstrip": false,
|
48 |
+
"normalized": false,
|
49 |
+
"rstrip": false,
|
50 |
+
"single_word": false,
|
51 |
+
"special": true
|
52 |
+
}
|
53 |
+
},
|
54 |
+
"additional_special_tokens": [
|
55 |
+
"<start_of_turn>",
|
56 |
+
"<end_of_turn>"
|
57 |
+
],
|
58 |
+
"bos_token": "<bos>",
|
59 |
+
"chat_template": "{{ bos_token }}{% if messages[0]['role'] == 'system' %}{{ raise_exception('System role not supported') }}{% endif %}{% for message in messages %}{% if (message['role'] == 'user') != (loop.index0 % 2 == 0) %}{{ raise_exception('Conversation roles must alternate user/assistant/user/assistant/...') }}{% endif %}{% if (message['role'] == 'assistant') %}{% set role = 'model' %}{% else %}{% set role = message['role'] %}{% endif %}{{ '<start_of_turn>' + role + '\n' + message['content'] | trim + '<end_of_turn>\n' }}{% endfor %}{% if add_generation_prompt %}{{'<start_of_turn>model\n'}}{% endif %}",
|
60 |
+
"clean_up_tokenization_spaces": false,
|
61 |
+
"eos_token": "<eos>",
|
62 |
+
"legacy": null,
|
63 |
+
"model_max_length": 8192,
|
64 |
+
"pad_token": "<pad>",
|
65 |
+
"padding_side": "right",
|
66 |
+
"sp_model_kwargs": {},
|
67 |
+
"spaces_between_special_tokens": false,
|
68 |
+
"tokenizer_class": "GemmaTokenizer",
|
69 |
+
"unk_token": "<unk>",
|
70 |
+
"use_default_system_prompt": false
|
71 |
+
}
|
checkpoint-21600/trainer_state.json
ADDED
@@ -0,0 +1,2181 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"best_metric": null,
|
3 |
+
"best_model_checkpoint": null,
|
4 |
+
"epoch": 2.4605570427749615,
|
5 |
+
"eval_steps": 200,
|
6 |
+
"global_step": 21600,
|
7 |
+
"is_hyper_param_search": false,
|
8 |
+
"is_local_process_zero": true,
|
9 |
+
"is_world_process_zero": true,
|
10 |
+
"log_history": [
|
11 |
+
{
|
12 |
+
"epoch": 0.02,
|
13 |
+
"grad_norm": 0.1643640249967575,
|
14 |
+
"learning_rate": 0.0001988891104338166,
|
15 |
+
"loss": 1.7673,
|
16 |
+
"step": 200
|
17 |
+
},
|
18 |
+
{
|
19 |
+
"epoch": 0.02,
|
20 |
+
"eval_bertscore": 0.7312520742416382,
|
21 |
+
"eval_loss": 1.7944419384002686,
|
22 |
+
"eval_rouge1": 0.645726048132668,
|
23 |
+
"eval_rouge2": 0.342840307585653,
|
24 |
+
"eval_rougeL": 0.5174784271125388,
|
25 |
+
"eval_rougeLsum": 0.6359911842715976,
|
26 |
+
"eval_runtime": 67.7968,
|
27 |
+
"eval_samples_per_second": 0.147,
|
28 |
+
"eval_steps_per_second": 0.074,
|
29 |
+
"step": 200
|
30 |
+
},
|
31 |
+
{
|
32 |
+
"epoch": 0.05,
|
33 |
+
"grad_norm": 0.17238478362560272,
|
34 |
+
"learning_rate": 0.00019774973651978238,
|
35 |
+
"loss": 1.6985,
|
36 |
+
"step": 400
|
37 |
+
},
|
38 |
+
{
|
39 |
+
"epoch": 0.05,
|
40 |
+
"eval_bertscore": 0.733666718006134,
|
41 |
+
"eval_loss": 1.7791178226470947,
|
42 |
+
"eval_rouge1": 0.6540909153028596,
|
43 |
+
"eval_rouge2": 0.3548819059818129,
|
44 |
+
"eval_rougeL": 0.527257232694246,
|
45 |
+
"eval_rougeLsum": 0.6442799950994005,
|
46 |
+
"eval_runtime": 15.1267,
|
47 |
+
"eval_samples_per_second": 0.661,
|
48 |
+
"eval_steps_per_second": 0.331,
|
49 |
+
"step": 400
|
50 |
+
},
|
51 |
+
{
|
52 |
+
"epoch": 0.07,
|
53 |
+
"grad_norm": 0.19368696212768555,
|
54 |
+
"learning_rate": 0.00019661036260574814,
|
55 |
+
"loss": 1.6962,
|
56 |
+
"step": 600
|
57 |
+
},
|
58 |
+
{
|
59 |
+
"epoch": 0.07,
|
60 |
+
"eval_bertscore": 0.7339462041854858,
|
61 |
+
"eval_loss": 1.7609882354736328,
|
62 |
+
"eval_rouge1": 0.6384337329686338,
|
63 |
+
"eval_rouge2": 0.3415514270662107,
|
64 |
+
"eval_rougeL": 0.51206080148464,
|
65 |
+
"eval_rougeLsum": 0.6261968614666548,
|
66 |
+
"eval_runtime": 15.2068,
|
67 |
+
"eval_samples_per_second": 0.658,
|
68 |
+
"eval_steps_per_second": 0.329,
|
69 |
+
"step": 600
|
70 |
+
},
|
71 |
+
{
|
72 |
+
"epoch": 0.09,
|
73 |
+
"grad_norm": 0.18629203736782074,
|
74 |
+
"learning_rate": 0.00019547098869171392,
|
75 |
+
"loss": 1.6825,
|
76 |
+
"step": 800
|
77 |
+
},
|
78 |
+
{
|
79 |
+
"epoch": 0.09,
|
80 |
+
"eval_bertscore": 0.7363594174385071,
|
81 |
+
"eval_loss": 1.7610784769058228,
|
82 |
+
"eval_rouge1": 0.6461624591922237,
|
83 |
+
"eval_rouge2": 0.3477371388439609,
|
84 |
+
"eval_rougeL": 0.5187429174752844,
|
85 |
+
"eval_rougeLsum": 0.6361089823008282,
|
86 |
+
"eval_runtime": 15.173,
|
87 |
+
"eval_samples_per_second": 0.659,
|
88 |
+
"eval_steps_per_second": 0.33,
|
89 |
+
"step": 800
|
90 |
+
},
|
91 |
+
{
|
92 |
+
"epoch": 0.11,
|
93 |
+
"grad_norm": 0.1799013316631317,
|
94 |
+
"learning_rate": 0.00019433161477767967,
|
95 |
+
"loss": 1.6848,
|
96 |
+
"step": 1000
|
97 |
+
},
|
98 |
+
{
|
99 |
+
"epoch": 0.11,
|
100 |
+
"eval_bertscore": 0.7334067225456238,
|
101 |
+
"eval_loss": 1.7576347589492798,
|
102 |
+
"eval_rouge1": 0.6345119236349537,
|
103 |
+
"eval_rouge2": 0.3422519149071803,
|
104 |
+
"eval_rougeL": 0.5111983101326238,
|
105 |
+
"eval_rougeLsum": 0.6244653120436832,
|
106 |
+
"eval_runtime": 15.2847,
|
107 |
+
"eval_samples_per_second": 0.654,
|
108 |
+
"eval_steps_per_second": 0.327,
|
109 |
+
"step": 1000
|
110 |
+
},
|
111 |
+
{
|
112 |
+
"epoch": 0.14,
|
113 |
+
"grad_norm": 0.22036150097846985,
|
114 |
+
"learning_rate": 0.00019319224086364545,
|
115 |
+
"loss": 1.6714,
|
116 |
+
"step": 1200
|
117 |
+
},
|
118 |
+
{
|
119 |
+
"epoch": 0.14,
|
120 |
+
"eval_bertscore": 0.7323788404464722,
|
121 |
+
"eval_loss": 1.7521806955337524,
|
122 |
+
"eval_rouge1": 0.6452540184557478,
|
123 |
+
"eval_rouge2": 0.3465145726476423,
|
124 |
+
"eval_rougeL": 0.516711757588783,
|
125 |
+
"eval_rougeLsum": 0.6341049885677059,
|
126 |
+
"eval_runtime": 15.1247,
|
127 |
+
"eval_samples_per_second": 0.661,
|
128 |
+
"eval_steps_per_second": 0.331,
|
129 |
+
"step": 1200
|
130 |
+
},
|
131 |
+
{
|
132 |
+
"epoch": 0.16,
|
133 |
+
"grad_norm": 0.21381086111068726,
|
134 |
+
"learning_rate": 0.0001920528669496112,
|
135 |
+
"loss": 1.6669,
|
136 |
+
"step": 1400
|
137 |
+
},
|
138 |
+
{
|
139 |
+
"epoch": 0.16,
|
140 |
+
"eval_bertscore": 0.7313202619552612,
|
141 |
+
"eval_loss": 1.7520482540130615,
|
142 |
+
"eval_rouge1": 0.6397526546254797,
|
143 |
+
"eval_rouge2": 0.3452671288110514,
|
144 |
+
"eval_rougeL": 0.5176580626678706,
|
145 |
+
"eval_rougeLsum": 0.6296746647539768,
|
146 |
+
"eval_runtime": 15.183,
|
147 |
+
"eval_samples_per_second": 0.659,
|
148 |
+
"eval_steps_per_second": 0.329,
|
149 |
+
"step": 1400
|
150 |
+
},
|
151 |
+
{
|
152 |
+
"epoch": 0.18,
|
153 |
+
"grad_norm": 0.20332874357700348,
|
154 |
+
"learning_rate": 0.00019091349303557696,
|
155 |
+
"loss": 1.671,
|
156 |
+
"step": 1600
|
157 |
+
},
|
158 |
+
{
|
159 |
+
"epoch": 0.18,
|
160 |
+
"eval_bertscore": 0.7349230647087097,
|
161 |
+
"eval_loss": 1.7473630905151367,
|
162 |
+
"eval_rouge1": 0.637439872504459,
|
163 |
+
"eval_rouge2": 0.34307164454056094,
|
164 |
+
"eval_rougeL": 0.5129717676228565,
|
165 |
+
"eval_rougeLsum": 0.6272190896182391,
|
166 |
+
"eval_runtime": 15.5672,
|
167 |
+
"eval_samples_per_second": 0.642,
|
168 |
+
"eval_steps_per_second": 0.321,
|
169 |
+
"step": 1600
|
170 |
+
},
|
171 |
+
{
|
172 |
+
"epoch": 0.21,
|
173 |
+
"grad_norm": 0.2025599479675293,
|
174 |
+
"learning_rate": 0.00018977411912154274,
|
175 |
+
"loss": 1.6721,
|
176 |
+
"step": 1800
|
177 |
+
},
|
178 |
+
{
|
179 |
+
"epoch": 0.21,
|
180 |
+
"eval_bertscore": 0.7357184290885925,
|
181 |
+
"eval_loss": 1.7516342401504517,
|
182 |
+
"eval_rouge1": 0.6387615819926658,
|
183 |
+
"eval_rouge2": 0.34366787517105574,
|
184 |
+
"eval_rougeL": 0.5129026911770751,
|
185 |
+
"eval_rougeLsum": 0.6289314118258257,
|
186 |
+
"eval_runtime": 15.9574,
|
187 |
+
"eval_samples_per_second": 0.627,
|
188 |
+
"eval_steps_per_second": 0.313,
|
189 |
+
"step": 1800
|
190 |
+
},
|
191 |
+
{
|
192 |
+
"epoch": 0.23,
|
193 |
+
"grad_norm": 0.20457112789154053,
|
194 |
+
"learning_rate": 0.0001886347452075085,
|
195 |
+
"loss": 1.671,
|
196 |
+
"step": 2000
|
197 |
+
},
|
198 |
+
{
|
199 |
+
"epoch": 0.23,
|
200 |
+
"eval_bertscore": 0.733718752861023,
|
201 |
+
"eval_loss": 1.7501707077026367,
|
202 |
+
"eval_rouge1": 0.6346207681220664,
|
203 |
+
"eval_rouge2": 0.33748369437614106,
|
204 |
+
"eval_rougeL": 0.5085159047705141,
|
205 |
+
"eval_rougeLsum": 0.6239953154441167,
|
206 |
+
"eval_runtime": 15.0863,
|
207 |
+
"eval_samples_per_second": 0.663,
|
208 |
+
"eval_steps_per_second": 0.331,
|
209 |
+
"step": 2000
|
210 |
+
},
|
211 |
+
{
|
212 |
+
"epoch": 0.25,
|
213 |
+
"grad_norm": 0.22552740573883057,
|
214 |
+
"learning_rate": 0.00018749537129347424,
|
215 |
+
"loss": 1.6496,
|
216 |
+
"step": 2200
|
217 |
+
},
|
218 |
+
{
|
219 |
+
"epoch": 0.25,
|
220 |
+
"eval_bertscore": 0.7368552684783936,
|
221 |
+
"eval_loss": 1.7437107563018799,
|
222 |
+
"eval_rouge1": 0.6490756387878311,
|
223 |
+
"eval_rouge2": 0.3448817738175175,
|
224 |
+
"eval_rougeL": 0.5235187045706321,
|
225 |
+
"eval_rougeLsum": 0.6377780857890332,
|
226 |
+
"eval_runtime": 15.07,
|
227 |
+
"eval_samples_per_second": 0.664,
|
228 |
+
"eval_steps_per_second": 0.332,
|
229 |
+
"step": 2200
|
230 |
+
},
|
231 |
+
{
|
232 |
+
"epoch": 0.27,
|
233 |
+
"grad_norm": 0.22573673725128174,
|
234 |
+
"learning_rate": 0.00018635599737944,
|
235 |
+
"loss": 1.6629,
|
236 |
+
"step": 2400
|
237 |
+
},
|
238 |
+
{
|
239 |
+
"epoch": 0.27,
|
240 |
+
"eval_bertscore": 0.7314499616622925,
|
241 |
+
"eval_loss": 1.7462828159332275,
|
242 |
+
"eval_rouge1": 0.6511482369678803,
|
243 |
+
"eval_rouge2": 0.34632544827771805,
|
244 |
+
"eval_rougeL": 0.5212417191003778,
|
245 |
+
"eval_rougeLsum": 0.6415391907940229,
|
246 |
+
"eval_runtime": 15.2078,
|
247 |
+
"eval_samples_per_second": 0.658,
|
248 |
+
"eval_steps_per_second": 0.329,
|
249 |
+
"step": 2400
|
250 |
+
},
|
251 |
+
{
|
252 |
+
"epoch": 0.3,
|
253 |
+
"grad_norm": 0.26426687836647034,
|
254 |
+
"learning_rate": 0.00018521662346540575,
|
255 |
+
"loss": 1.6644,
|
256 |
+
"step": 2600
|
257 |
+
},
|
258 |
+
{
|
259 |
+
"epoch": 0.3,
|
260 |
+
"eval_bertscore": 0.7363359928131104,
|
261 |
+
"eval_loss": 1.7505037784576416,
|
262 |
+
"eval_rouge1": 0.6498296552335481,
|
263 |
+
"eval_rouge2": 0.34873833589761183,
|
264 |
+
"eval_rougeL": 0.5194028620820592,
|
265 |
+
"eval_rougeLsum": 0.6404603087578984,
|
266 |
+
"eval_runtime": 14.8403,
|
267 |
+
"eval_samples_per_second": 0.674,
|
268 |
+
"eval_steps_per_second": 0.337,
|
269 |
+
"step": 2600
|
270 |
+
},
|
271 |
+
{
|
272 |
+
"epoch": 0.32,
|
273 |
+
"grad_norm": 0.20142091810703278,
|
274 |
+
"learning_rate": 0.00018407724955137153,
|
275 |
+
"loss": 1.6535,
|
276 |
+
"step": 2800
|
277 |
+
},
|
278 |
+
{
|
279 |
+
"epoch": 0.32,
|
280 |
+
"eval_bertscore": 0.7304679155349731,
|
281 |
+
"eval_loss": 1.7511039972305298,
|
282 |
+
"eval_rouge1": 0.6475130585388738,
|
283 |
+
"eval_rouge2": 0.34648331046897884,
|
284 |
+
"eval_rougeL": 0.5218042284020985,
|
285 |
+
"eval_rougeLsum": 0.6382749834402862,
|
286 |
+
"eval_runtime": 15.0162,
|
287 |
+
"eval_samples_per_second": 0.666,
|
288 |
+
"eval_steps_per_second": 0.333,
|
289 |
+
"step": 2800
|
290 |
+
},
|
291 |
+
{
|
292 |
+
"epoch": 0.34,
|
293 |
+
"grad_norm": 0.23283220827579498,
|
294 |
+
"learning_rate": 0.00018293787563733728,
|
295 |
+
"loss": 1.6477,
|
296 |
+
"step": 3000
|
297 |
+
},
|
298 |
+
{
|
299 |
+
"epoch": 0.34,
|
300 |
+
"eval_bertscore": 0.7327049374580383,
|
301 |
+
"eval_loss": 1.7461665868759155,
|
302 |
+
"eval_rouge1": 0.6309349586871908,
|
303 |
+
"eval_rouge2": 0.3387882478990309,
|
304 |
+
"eval_rougeL": 0.5042059192403674,
|
305 |
+
"eval_rougeLsum": 0.6210432469674847,
|
306 |
+
"eval_runtime": 15.463,
|
307 |
+
"eval_samples_per_second": 0.647,
|
308 |
+
"eval_steps_per_second": 0.323,
|
309 |
+
"step": 3000
|
310 |
+
},
|
311 |
+
{
|
312 |
+
"epoch": 0.36,
|
313 |
+
"grad_norm": 0.21316750347614288,
|
314 |
+
"learning_rate": 0.00018179850172330306,
|
315 |
+
"loss": 1.6614,
|
316 |
+
"step": 3200
|
317 |
+
},
|
318 |
+
{
|
319 |
+
"epoch": 0.36,
|
320 |
+
"eval_bertscore": 0.7314620018005371,
|
321 |
+
"eval_loss": 1.7468239068984985,
|
322 |
+
"eval_rouge1": 0.6480904534152265,
|
323 |
+
"eval_rouge2": 0.3479530168963481,
|
324 |
+
"eval_rougeL": 0.5193148273848067,
|
325 |
+
"eval_rougeLsum": 0.6366010767207634,
|
326 |
+
"eval_runtime": 15.0381,
|
327 |
+
"eval_samples_per_second": 0.665,
|
328 |
+
"eval_steps_per_second": 0.332,
|
329 |
+
"step": 3200
|
330 |
+
},
|
331 |
+
{
|
332 |
+
"epoch": 0.39,
|
333 |
+
"grad_norm": 0.26080408692359924,
|
334 |
+
"learning_rate": 0.00018065912780926882,
|
335 |
+
"loss": 1.6591,
|
336 |
+
"step": 3400
|
337 |
+
},
|
338 |
+
{
|
339 |
+
"epoch": 0.39,
|
340 |
+
"eval_bertscore": 0.7327477335929871,
|
341 |
+
"eval_loss": 1.7442594766616821,
|
342 |
+
"eval_rouge1": 0.6424613378037144,
|
343 |
+
"eval_rouge2": 0.34731770322974903,
|
344 |
+
"eval_rougeL": 0.5160705879794565,
|
345 |
+
"eval_rougeLsum": 0.6327006420281607,
|
346 |
+
"eval_runtime": 15.5373,
|
347 |
+
"eval_samples_per_second": 0.644,
|
348 |
+
"eval_steps_per_second": 0.322,
|
349 |
+
"step": 3400
|
350 |
+
},
|
351 |
+
{
|
352 |
+
"epoch": 0.41,
|
353 |
+
"grad_norm": 0.23274216055870056,
|
354 |
+
"learning_rate": 0.0001795197538952346,
|
355 |
+
"loss": 1.6613,
|
356 |
+
"step": 3600
|
357 |
+
},
|
358 |
+
{
|
359 |
+
"epoch": 0.41,
|
360 |
+
"eval_bertscore": 0.736956000328064,
|
361 |
+
"eval_loss": 1.7429373264312744,
|
362 |
+
"eval_rouge1": 0.6514574160666677,
|
363 |
+
"eval_rouge2": 0.3556199242231646,
|
364 |
+
"eval_rougeL": 0.5249726237675663,
|
365 |
+
"eval_rougeLsum": 0.6406261097623661,
|
366 |
+
"eval_runtime": 14.9415,
|
367 |
+
"eval_samples_per_second": 0.669,
|
368 |
+
"eval_steps_per_second": 0.335,
|
369 |
+
"step": 3600
|
370 |
+
},
|
371 |
+
{
|
372 |
+
"epoch": 0.43,
|
373 |
+
"grad_norm": 0.23616766929626465,
|
374 |
+
"learning_rate": 0.00017838037998120035,
|
375 |
+
"loss": 1.6479,
|
376 |
+
"step": 3800
|
377 |
+
},
|
378 |
+
{
|
379 |
+
"epoch": 0.43,
|
380 |
+
"eval_bertscore": 0.7349627614021301,
|
381 |
+
"eval_loss": 1.7420669794082642,
|
382 |
+
"eval_rouge1": 0.655851684949526,
|
383 |
+
"eval_rouge2": 0.35254590691865084,
|
384 |
+
"eval_rougeL": 0.5248980956621441,
|
385 |
+
"eval_rougeLsum": 0.6449637270581419,
|
386 |
+
"eval_runtime": 15.4178,
|
387 |
+
"eval_samples_per_second": 0.649,
|
388 |
+
"eval_steps_per_second": 0.324,
|
389 |
+
"step": 3800
|
390 |
+
},
|
391 |
+
{
|
392 |
+
"epoch": 0.46,
|
393 |
+
"grad_norm": 0.23260319232940674,
|
394 |
+
"learning_rate": 0.0001772410060671661,
|
395 |
+
"loss": 1.6569,
|
396 |
+
"step": 4000
|
397 |
+
},
|
398 |
+
{
|
399 |
+
"epoch": 0.46,
|
400 |
+
"eval_bertscore": 0.7332885265350342,
|
401 |
+
"eval_loss": 1.7401313781738281,
|
402 |
+
"eval_rouge1": 0.6669483634140105,
|
403 |
+
"eval_rouge2": 0.35873988835161297,
|
404 |
+
"eval_rougeL": 0.5343868725007427,
|
405 |
+
"eval_rougeLsum": 0.6555353134690931,
|
406 |
+
"eval_runtime": 15.0822,
|
407 |
+
"eval_samples_per_second": 0.663,
|
408 |
+
"eval_steps_per_second": 0.332,
|
409 |
+
"step": 4000
|
410 |
+
},
|
411 |
+
{
|
412 |
+
"epoch": 0.48,
|
413 |
+
"grad_norm": 0.2366473525762558,
|
414 |
+
"learning_rate": 0.00017610163215313186,
|
415 |
+
"loss": 1.6599,
|
416 |
+
"step": 4200
|
417 |
+
},
|
418 |
+
{
|
419 |
+
"epoch": 0.48,
|
420 |
+
"eval_bertscore": 0.7335314750671387,
|
421 |
+
"eval_loss": 1.7385823726654053,
|
422 |
+
"eval_rouge1": 0.6559297063578133,
|
423 |
+
"eval_rouge2": 0.35483499789990636,
|
424 |
+
"eval_rougeL": 0.5297939800986089,
|
425 |
+
"eval_rougeLsum": 0.6454544372491222,
|
426 |
+
"eval_runtime": 15.1029,
|
427 |
+
"eval_samples_per_second": 0.662,
|
428 |
+
"eval_steps_per_second": 0.331,
|
429 |
+
"step": 4200
|
430 |
+
},
|
431 |
+
{
|
432 |
+
"epoch": 0.5,
|
433 |
+
"grad_norm": 0.20628753304481506,
|
434 |
+
"learning_rate": 0.0001749622582390976,
|
435 |
+
"loss": 1.6454,
|
436 |
+
"step": 4400
|
437 |
+
},
|
438 |
+
{
|
439 |
+
"epoch": 0.5,
|
440 |
+
"eval_bertscore": 0.7342795133590698,
|
441 |
+
"eval_loss": 1.7422058582305908,
|
442 |
+
"eval_rouge1": 0.660746519614568,
|
443 |
+
"eval_rouge2": 0.3633965895561597,
|
444 |
+
"eval_rougeL": 0.5369036980876734,
|
445 |
+
"eval_rougeLsum": 0.650338328328998,
|
446 |
+
"eval_runtime": 15.4434,
|
447 |
+
"eval_samples_per_second": 0.648,
|
448 |
+
"eval_steps_per_second": 0.324,
|
449 |
+
"step": 4400
|
450 |
+
},
|
451 |
+
{
|
452 |
+
"epoch": 0.52,
|
453 |
+
"grad_norm": 0.2239149957895279,
|
454 |
+
"learning_rate": 0.0001738228843250634,
|
455 |
+
"loss": 1.6594,
|
456 |
+
"step": 4600
|
457 |
+
},
|
458 |
+
{
|
459 |
+
"epoch": 0.52,
|
460 |
+
"eval_bertscore": 0.7313543558120728,
|
461 |
+
"eval_loss": 1.740854263305664,
|
462 |
+
"eval_rouge1": 0.6591645132619427,
|
463 |
+
"eval_rouge2": 0.35766117432431743,
|
464 |
+
"eval_rougeL": 0.532710255034635,
|
465 |
+
"eval_rougeLsum": 0.6479428185884644,
|
466 |
+
"eval_runtime": 14.9436,
|
467 |
+
"eval_samples_per_second": 0.669,
|
468 |
+
"eval_steps_per_second": 0.335,
|
469 |
+
"step": 4600
|
470 |
+
},
|
471 |
+
{
|
472 |
+
"epoch": 0.55,
|
473 |
+
"grad_norm": 0.24808338284492493,
|
474 |
+
"learning_rate": 0.00017268351041102914,
|
475 |
+
"loss": 1.6604,
|
476 |
+
"step": 4800
|
477 |
+
},
|
478 |
+
{
|
479 |
+
"epoch": 0.55,
|
480 |
+
"eval_bertscore": 0.7333321571350098,
|
481 |
+
"eval_loss": 1.7385585308074951,
|
482 |
+
"eval_rouge1": 0.6532115808232871,
|
483 |
+
"eval_rouge2": 0.35333788022501567,
|
484 |
+
"eval_rougeL": 0.5284071547874328,
|
485 |
+
"eval_rougeLsum": 0.6410472452797623,
|
486 |
+
"eval_runtime": 15.0277,
|
487 |
+
"eval_samples_per_second": 0.665,
|
488 |
+
"eval_steps_per_second": 0.333,
|
489 |
+
"step": 4800
|
490 |
+
},
|
491 |
+
{
|
492 |
+
"epoch": 0.57,
|
493 |
+
"grad_norm": 0.2555364966392517,
|
494 |
+
"learning_rate": 0.0001715441364969949,
|
495 |
+
"loss": 1.6493,
|
496 |
+
"step": 5000
|
497 |
+
},
|
498 |
+
{
|
499 |
+
"epoch": 0.57,
|
500 |
+
"eval_bertscore": 0.7318152189254761,
|
501 |
+
"eval_loss": 1.7357494831085205,
|
502 |
+
"eval_rouge1": 0.6476755890502586,
|
503 |
+
"eval_rouge2": 0.35312778275949164,
|
504 |
+
"eval_rougeL": 0.5227601228049905,
|
505 |
+
"eval_rougeLsum": 0.6371331138372852,
|
506 |
+
"eval_runtime": 14.8807,
|
507 |
+
"eval_samples_per_second": 0.672,
|
508 |
+
"eval_steps_per_second": 0.336,
|
509 |
+
"step": 5000
|
510 |
+
},
|
511 |
+
{
|
512 |
+
"epoch": 0.59,
|
513 |
+
"grad_norm": 0.21518155932426453,
|
514 |
+
"learning_rate": 0.00017040476258296068,
|
515 |
+
"loss": 1.644,
|
516 |
+
"step": 5200
|
517 |
+
},
|
518 |
+
{
|
519 |
+
"epoch": 0.59,
|
520 |
+
"eval_bertscore": 0.734805703163147,
|
521 |
+
"eval_loss": 1.74032723903656,
|
522 |
+
"eval_rouge1": 0.6476813733451636,
|
523 |
+
"eval_rouge2": 0.3509259728617576,
|
524 |
+
"eval_rougeL": 0.5221334872800274,
|
525 |
+
"eval_rougeLsum": 0.636892384667733,
|
526 |
+
"eval_runtime": 15.1271,
|
527 |
+
"eval_samples_per_second": 0.661,
|
528 |
+
"eval_steps_per_second": 0.331,
|
529 |
+
"step": 5200
|
530 |
+
},
|
531 |
+
{
|
532 |
+
"epoch": 0.62,
|
533 |
+
"grad_norm": 0.26086658239364624,
|
534 |
+
"learning_rate": 0.00016926538866892643,
|
535 |
+
"loss": 1.6449,
|
536 |
+
"step": 5400
|
537 |
+
},
|
538 |
+
{
|
539 |
+
"epoch": 0.62,
|
540 |
+
"eval_bertscore": 0.7339995503425598,
|
541 |
+
"eval_loss": 1.7338205575942993,
|
542 |
+
"eval_rouge1": 0.6416889902864902,
|
543 |
+
"eval_rouge2": 0.3479045880347737,
|
544 |
+
"eval_rougeL": 0.5160577838468976,
|
545 |
+
"eval_rougeLsum": 0.6317983411796093,
|
546 |
+
"eval_runtime": 14.9992,
|
547 |
+
"eval_samples_per_second": 0.667,
|
548 |
+
"eval_steps_per_second": 0.333,
|
549 |
+
"step": 5400
|
550 |
+
},
|
551 |
+
{
|
552 |
+
"epoch": 0.64,
|
553 |
+
"grad_norm": 0.25449469685554504,
|
554 |
+
"learning_rate": 0.0001681260147548922,
|
555 |
+
"loss": 1.6299,
|
556 |
+
"step": 5600
|
557 |
+
},
|
558 |
+
{
|
559 |
+
"epoch": 0.64,
|
560 |
+
"eval_bertscore": 0.7306328415870667,
|
561 |
+
"eval_loss": 1.7369228601455688,
|
562 |
+
"eval_rouge1": 0.6390760905985684,
|
563 |
+
"eval_rouge2": 0.3409328272828699,
|
564 |
+
"eval_rougeL": 0.5111832543685331,
|
565 |
+
"eval_rougeLsum": 0.6285753423407665,
|
566 |
+
"eval_runtime": 15.483,
|
567 |
+
"eval_samples_per_second": 0.646,
|
568 |
+
"eval_steps_per_second": 0.323,
|
569 |
+
"step": 5600
|
570 |
+
},
|
571 |
+
{
|
572 |
+
"epoch": 0.66,
|
573 |
+
"grad_norm": 0.24706102907657623,
|
574 |
+
"learning_rate": 0.00016698664084085796,
|
575 |
+
"loss": 1.6374,
|
576 |
+
"step": 5800
|
577 |
+
},
|
578 |
+
{
|
579 |
+
"epoch": 0.66,
|
580 |
+
"eval_bertscore": 0.732075572013855,
|
581 |
+
"eval_loss": 1.7343876361846924,
|
582 |
+
"eval_rouge1": 0.6378821977913272,
|
583 |
+
"eval_rouge2": 0.34619427775171585,
|
584 |
+
"eval_rougeL": 0.5120186953041237,
|
585 |
+
"eval_rougeLsum": 0.6284056323839109,
|
586 |
+
"eval_runtime": 15.7341,
|
587 |
+
"eval_samples_per_second": 0.636,
|
588 |
+
"eval_steps_per_second": 0.318,
|
589 |
+
"step": 5800
|
590 |
+
},
|
591 |
+
{
|
592 |
+
"epoch": 0.68,
|
593 |
+
"grad_norm": 0.24373260140419006,
|
594 |
+
"learning_rate": 0.00016584726692682372,
|
595 |
+
"loss": 1.6427,
|
596 |
+
"step": 6000
|
597 |
+
},
|
598 |
+
{
|
599 |
+
"epoch": 0.68,
|
600 |
+
"eval_bertscore": 0.7350374460220337,
|
601 |
+
"eval_loss": 1.729591965675354,
|
602 |
+
"eval_rouge1": 0.6516545226356616,
|
603 |
+
"eval_rouge2": 0.35485762033878543,
|
604 |
+
"eval_rougeL": 0.5249054193354852,
|
605 |
+
"eval_rougeLsum": 0.6411016821651583,
|
606 |
+
"eval_runtime": 15.5199,
|
607 |
+
"eval_samples_per_second": 0.644,
|
608 |
+
"eval_steps_per_second": 0.322,
|
609 |
+
"step": 6000
|
610 |
+
},
|
611 |
+
{
|
612 |
+
"epoch": 0.71,
|
613 |
+
"grad_norm": 0.24616578221321106,
|
614 |
+
"learning_rate": 0.0001647078930127895,
|
615 |
+
"loss": 1.6296,
|
616 |
+
"step": 6200
|
617 |
+
},
|
618 |
+
{
|
619 |
+
"epoch": 0.71,
|
620 |
+
"eval_bertscore": 0.7335461378097534,
|
621 |
+
"eval_loss": 1.7302274703979492,
|
622 |
+
"eval_rouge1": 0.6531411706717427,
|
623 |
+
"eval_rouge2": 0.35003053174601517,
|
624 |
+
"eval_rougeL": 0.5212483686089053,
|
625 |
+
"eval_rougeLsum": 0.6438454124825417,
|
626 |
+
"eval_runtime": 15.3058,
|
627 |
+
"eval_samples_per_second": 0.653,
|
628 |
+
"eval_steps_per_second": 0.327,
|
629 |
+
"step": 6200
|
630 |
+
},
|
631 |
+
{
|
632 |
+
"epoch": 0.73,
|
633 |
+
"grad_norm": 0.24500492215156555,
|
634 |
+
"learning_rate": 0.00016356851909875522,
|
635 |
+
"loss": 1.6251,
|
636 |
+
"step": 6400
|
637 |
+
},
|
638 |
+
{
|
639 |
+
"epoch": 0.73,
|
640 |
+
"eval_bertscore": 0.7347471714019775,
|
641 |
+
"eval_loss": 1.7292228937149048,
|
642 |
+
"eval_rouge1": 0.6565322485502556,
|
643 |
+
"eval_rouge2": 0.35887540291607073,
|
644 |
+
"eval_rougeL": 0.5284326132878907,
|
645 |
+
"eval_rougeLsum": 0.6469750895866724,
|
646 |
+
"eval_runtime": 14.9708,
|
647 |
+
"eval_samples_per_second": 0.668,
|
648 |
+
"eval_steps_per_second": 0.334,
|
649 |
+
"step": 6400
|
650 |
+
},
|
651 |
+
{
|
652 |
+
"epoch": 0.75,
|
653 |
+
"grad_norm": 0.26575228571891785,
|
654 |
+
"learning_rate": 0.000162429145184721,
|
655 |
+
"loss": 1.6389,
|
656 |
+
"step": 6600
|
657 |
+
},
|
658 |
+
{
|
659 |
+
"epoch": 0.75,
|
660 |
+
"eval_bertscore": 0.7319897413253784,
|
661 |
+
"eval_loss": 1.7287580966949463,
|
662 |
+
"eval_rouge1": 0.6458608801881298,
|
663 |
+
"eval_rouge2": 0.3503480901452204,
|
664 |
+
"eval_rougeL": 0.519626708150005,
|
665 |
+
"eval_rougeLsum": 0.6362405734928169,
|
666 |
+
"eval_runtime": 15.3509,
|
667 |
+
"eval_samples_per_second": 0.651,
|
668 |
+
"eval_steps_per_second": 0.326,
|
669 |
+
"step": 6600
|
670 |
+
},
|
671 |
+
{
|
672 |
+
"epoch": 0.77,
|
673 |
+
"grad_norm": 0.27144965529441833,
|
674 |
+
"learning_rate": 0.00016128977127068676,
|
675 |
+
"loss": 1.6476,
|
676 |
+
"step": 6800
|
677 |
+
},
|
678 |
+
{
|
679 |
+
"epoch": 0.77,
|
680 |
+
"eval_bertscore": 0.7392772436141968,
|
681 |
+
"eval_loss": 1.7257907390594482,
|
682 |
+
"eval_rouge1": 0.6543238897579965,
|
683 |
+
"eval_rouge2": 0.3606726049451984,
|
684 |
+
"eval_rougeL": 0.5317585887753791,
|
685 |
+
"eval_rougeLsum": 0.6452028420624081,
|
686 |
+
"eval_runtime": 15.0268,
|
687 |
+
"eval_samples_per_second": 0.665,
|
688 |
+
"eval_steps_per_second": 0.333,
|
689 |
+
"step": 6800
|
690 |
+
},
|
691 |
+
{
|
692 |
+
"epoch": 0.8,
|
693 |
+
"grad_norm": 0.2579711079597473,
|
694 |
+
"learning_rate": 0.00016015039735665254,
|
695 |
+
"loss": 1.6316,
|
696 |
+
"step": 7000
|
697 |
+
},
|
698 |
+
{
|
699 |
+
"epoch": 0.8,
|
700 |
+
"eval_bertscore": 0.7375612854957581,
|
701 |
+
"eval_loss": 1.7296981811523438,
|
702 |
+
"eval_rouge1": 0.658906725408624,
|
703 |
+
"eval_rouge2": 0.35825094165644866,
|
704 |
+
"eval_rougeL": 0.5323299193377959,
|
705 |
+
"eval_rougeLsum": 0.6500364347290426,
|
706 |
+
"eval_runtime": 14.9244,
|
707 |
+
"eval_samples_per_second": 0.67,
|
708 |
+
"eval_steps_per_second": 0.335,
|
709 |
+
"step": 7000
|
710 |
+
},
|
711 |
+
{
|
712 |
+
"epoch": 0.82,
|
713 |
+
"grad_norm": 0.2589207589626312,
|
714 |
+
"learning_rate": 0.0001590110234426183,
|
715 |
+
"loss": 1.6432,
|
716 |
+
"step": 7200
|
717 |
+
},
|
718 |
+
{
|
719 |
+
"epoch": 0.82,
|
720 |
+
"eval_bertscore": 0.735145092010498,
|
721 |
+
"eval_loss": 1.72720205783844,
|
722 |
+
"eval_rouge1": 0.6678646250245518,
|
723 |
+
"eval_rouge2": 0.36332843150846983,
|
724 |
+
"eval_rougeL": 0.537576430733886,
|
725 |
+
"eval_rougeLsum": 0.6579789388660506,
|
726 |
+
"eval_runtime": 14.9067,
|
727 |
+
"eval_samples_per_second": 0.671,
|
728 |
+
"eval_steps_per_second": 0.335,
|
729 |
+
"step": 7200
|
730 |
+
},
|
731 |
+
{
|
732 |
+
"epoch": 0.84,
|
733 |
+
"grad_norm": 0.26652559638023376,
|
734 |
+
"learning_rate": 0.00015787164952858404,
|
735 |
+
"loss": 1.6488,
|
736 |
+
"step": 7400
|
737 |
+
},
|
738 |
+
{
|
739 |
+
"epoch": 0.84,
|
740 |
+
"eval_bertscore": 0.7320755124092102,
|
741 |
+
"eval_loss": 1.7282931804656982,
|
742 |
+
"eval_rouge1": 0.6325633297780734,
|
743 |
+
"eval_rouge2": 0.34505856555703185,
|
744 |
+
"eval_rougeL": 0.5100006743383693,
|
745 |
+
"eval_rougeLsum": 0.6230385336341938,
|
746 |
+
"eval_runtime": 14.9075,
|
747 |
+
"eval_samples_per_second": 0.671,
|
748 |
+
"eval_steps_per_second": 0.335,
|
749 |
+
"step": 7400
|
750 |
+
},
|
751 |
+
{
|
752 |
+
"epoch": 0.87,
|
753 |
+
"grad_norm": 0.27353721857070923,
|
754 |
+
"learning_rate": 0.00015673227561454982,
|
755 |
+
"loss": 1.6486,
|
756 |
+
"step": 7600
|
757 |
+
},
|
758 |
+
{
|
759 |
+
"epoch": 0.87,
|
760 |
+
"eval_bertscore": 0.7367390394210815,
|
761 |
+
"eval_loss": 1.7290785312652588,
|
762 |
+
"eval_rouge1": 0.639487116874423,
|
763 |
+
"eval_rouge2": 0.3466574229736927,
|
764 |
+
"eval_rougeL": 0.515038120249177,
|
765 |
+
"eval_rougeLsum": 0.6301157215372983,
|
766 |
+
"eval_runtime": 15.0876,
|
767 |
+
"eval_samples_per_second": 0.663,
|
768 |
+
"eval_steps_per_second": 0.331,
|
769 |
+
"step": 7600
|
770 |
+
},
|
771 |
+
{
|
772 |
+
"epoch": 0.89,
|
773 |
+
"grad_norm": 0.24777938425540924,
|
774 |
+
"learning_rate": 0.00015559290170051558,
|
775 |
+
"loss": 1.6271,
|
776 |
+
"step": 7800
|
777 |
+
},
|
778 |
+
{
|
779 |
+
"epoch": 0.89,
|
780 |
+
"eval_bertscore": 0.735866904258728,
|
781 |
+
"eval_loss": 1.7264015674591064,
|
782 |
+
"eval_rouge1": 0.64939597901302,
|
783 |
+
"eval_rouge2": 0.3554282813944538,
|
784 |
+
"eval_rougeL": 0.5247953329477759,
|
785 |
+
"eval_rougeLsum": 0.6405524812915908,
|
786 |
+
"eval_runtime": 14.8892,
|
787 |
+
"eval_samples_per_second": 0.672,
|
788 |
+
"eval_steps_per_second": 0.336,
|
789 |
+
"step": 7800
|
790 |
+
},
|
791 |
+
{
|
792 |
+
"epoch": 0.91,
|
793 |
+
"grad_norm": 0.2703794538974762,
|
794 |
+
"learning_rate": 0.00015445352778648136,
|
795 |
+
"loss": 1.6415,
|
796 |
+
"step": 8000
|
797 |
+
},
|
798 |
+
{
|
799 |
+
"epoch": 0.91,
|
800 |
+
"eval_bertscore": 0.7325771450996399,
|
801 |
+
"eval_loss": 1.7271970510482788,
|
802 |
+
"eval_rouge1": 0.6659432894288253,
|
803 |
+
"eval_rouge2": 0.35962933912652617,
|
804 |
+
"eval_rougeL": 0.5385420432813512,
|
805 |
+
"eval_rougeLsum": 0.6557027031484046,
|
806 |
+
"eval_runtime": 14.8319,
|
807 |
+
"eval_samples_per_second": 0.674,
|
808 |
+
"eval_steps_per_second": 0.337,
|
809 |
+
"step": 8000
|
810 |
+
},
|
811 |
+
{
|
812 |
+
"epoch": 0.93,
|
813 |
+
"grad_norm": 0.28753793239593506,
|
814 |
+
"learning_rate": 0.0001533141538724471,
|
815 |
+
"loss": 1.6239,
|
816 |
+
"step": 8200
|
817 |
+
},
|
818 |
+
{
|
819 |
+
"epoch": 0.93,
|
820 |
+
"eval_bertscore": 0.7350013852119446,
|
821 |
+
"eval_loss": 1.7266199588775635,
|
822 |
+
"eval_rouge1": 0.6549282561414593,
|
823 |
+
"eval_rouge2": 0.35694530595734475,
|
824 |
+
"eval_rougeL": 0.5301601006964574,
|
825 |
+
"eval_rougeLsum": 0.6441779306137909,
|
826 |
+
"eval_runtime": 14.9005,
|
827 |
+
"eval_samples_per_second": 0.671,
|
828 |
+
"eval_steps_per_second": 0.336,
|
829 |
+
"step": 8200
|
830 |
+
},
|
831 |
+
{
|
832 |
+
"epoch": 0.96,
|
833 |
+
"grad_norm": 0.23870150744915009,
|
834 |
+
"learning_rate": 0.00015217477995841286,
|
835 |
+
"loss": 1.6293,
|
836 |
+
"step": 8400
|
837 |
+
},
|
838 |
+
{
|
839 |
+
"epoch": 0.96,
|
840 |
+
"eval_bertscore": 0.7271261811256409,
|
841 |
+
"eval_loss": 1.7256368398666382,
|
842 |
+
"eval_rouge1": 0.6515513829901936,
|
843 |
+
"eval_rouge2": 0.35217616104918836,
|
844 |
+
"eval_rougeL": 0.5236553509227138,
|
845 |
+
"eval_rougeLsum": 0.6411473505324752,
|
846 |
+
"eval_runtime": 14.9938,
|
847 |
+
"eval_samples_per_second": 0.667,
|
848 |
+
"eval_steps_per_second": 0.333,
|
849 |
+
"step": 8400
|
850 |
+
},
|
851 |
+
{
|
852 |
+
"epoch": 0.98,
|
853 |
+
"grad_norm": 0.28276997804641724,
|
854 |
+
"learning_rate": 0.00015103540604437861,
|
855 |
+
"loss": 1.6242,
|
856 |
+
"step": 8600
|
857 |
+
},
|
858 |
+
{
|
859 |
+
"epoch": 0.98,
|
860 |
+
"eval_bertscore": 0.7347334027290344,
|
861 |
+
"eval_loss": 1.717627763748169,
|
862 |
+
"eval_rouge1": 0.6350112495847634,
|
863 |
+
"eval_rouge2": 0.3477570751550898,
|
864 |
+
"eval_rougeL": 0.5146616989899861,
|
865 |
+
"eval_rougeLsum": 0.6246669376525157,
|
866 |
+
"eval_runtime": 15.7032,
|
867 |
+
"eval_samples_per_second": 0.637,
|
868 |
+
"eval_steps_per_second": 0.318,
|
869 |
+
"step": 8600
|
870 |
+
},
|
871 |
+
{
|
872 |
+
"epoch": 1.0,
|
873 |
+
"grad_norm": 0.24915842711925507,
|
874 |
+
"learning_rate": 0.00014989603213034437,
|
875 |
+
"loss": 1.6245,
|
876 |
+
"step": 8800
|
877 |
+
},
|
878 |
+
{
|
879 |
+
"epoch": 1.0,
|
880 |
+
"eval_bertscore": 0.7313701510429382,
|
881 |
+
"eval_loss": 1.7292964458465576,
|
882 |
+
"eval_rouge1": 0.6479528105367669,
|
883 |
+
"eval_rouge2": 0.35020983244262877,
|
884 |
+
"eval_rougeL": 0.5200907337780047,
|
885 |
+
"eval_rougeLsum": 0.6372896614836894,
|
886 |
+
"eval_runtime": 15.0043,
|
887 |
+
"eval_samples_per_second": 0.666,
|
888 |
+
"eval_steps_per_second": 0.333,
|
889 |
+
"step": 8800
|
890 |
+
},
|
891 |
+
{
|
892 |
+
"epoch": 1.03,
|
893 |
+
"grad_norm": 0.24036027491092682,
|
894 |
+
"learning_rate": 0.00014875665821631015,
|
895 |
+
"loss": 1.5364,
|
896 |
+
"step": 9000
|
897 |
+
},
|
898 |
+
{
|
899 |
+
"epoch": 1.03,
|
900 |
+
"eval_bertscore": 0.7327737808227539,
|
901 |
+
"eval_loss": 1.7339435815811157,
|
902 |
+
"eval_rouge1": 0.6568524178922349,
|
903 |
+
"eval_rouge2": 0.35560270713543163,
|
904 |
+
"eval_rougeL": 0.5310443670833082,
|
905 |
+
"eval_rougeLsum": 0.6480993679097387,
|
906 |
+
"eval_runtime": 14.9303,
|
907 |
+
"eval_samples_per_second": 0.67,
|
908 |
+
"eval_steps_per_second": 0.335,
|
909 |
+
"step": 9000
|
910 |
+
},
|
911 |
+
{
|
912 |
+
"epoch": 1.05,
|
913 |
+
"grad_norm": 0.2729027271270752,
|
914 |
+
"learning_rate": 0.0001476172843022759,
|
915 |
+
"loss": 1.5182,
|
916 |
+
"step": 9200
|
917 |
+
},
|
918 |
+
{
|
919 |
+
"epoch": 1.05,
|
920 |
+
"eval_bertscore": 0.7334672212600708,
|
921 |
+
"eval_loss": 1.739061713218689,
|
922 |
+
"eval_rouge1": 0.6552962187824329,
|
923 |
+
"eval_rouge2": 0.35210314124279196,
|
924 |
+
"eval_rougeL": 0.5272039052368354,
|
925 |
+
"eval_rougeLsum": 0.6437473533492806,
|
926 |
+
"eval_runtime": 15.5418,
|
927 |
+
"eval_samples_per_second": 0.643,
|
928 |
+
"eval_steps_per_second": 0.322,
|
929 |
+
"step": 9200
|
930 |
+
},
|
931 |
+
{
|
932 |
+
"epoch": 1.07,
|
933 |
+
"grad_norm": 0.2909716069698334,
|
934 |
+
"learning_rate": 0.00014647791038824168,
|
935 |
+
"loss": 1.5276,
|
936 |
+
"step": 9400
|
937 |
+
},
|
938 |
+
{
|
939 |
+
"epoch": 1.07,
|
940 |
+
"eval_bertscore": 0.7288902997970581,
|
941 |
+
"eval_loss": 1.736944556236267,
|
942 |
+
"eval_rouge1": 0.6533271685598301,
|
943 |
+
"eval_rouge2": 0.35279315321532184,
|
944 |
+
"eval_rougeL": 0.5262688234671329,
|
945 |
+
"eval_rougeLsum": 0.6424084937151033,
|
946 |
+
"eval_runtime": 15.1115,
|
947 |
+
"eval_samples_per_second": 0.662,
|
948 |
+
"eval_steps_per_second": 0.331,
|
949 |
+
"step": 9400
|
950 |
+
},
|
951 |
+
{
|
952 |
+
"epoch": 1.09,
|
953 |
+
"grad_norm": 0.3035859763622284,
|
954 |
+
"learning_rate": 0.00014533853647420743,
|
955 |
+
"loss": 1.5445,
|
956 |
+
"step": 9600
|
957 |
+
},
|
958 |
+
{
|
959 |
+
"epoch": 1.09,
|
960 |
+
"eval_bertscore": 0.7308284044265747,
|
961 |
+
"eval_loss": 1.737762689590454,
|
962 |
+
"eval_rouge1": 0.6619725777891359,
|
963 |
+
"eval_rouge2": 0.3611963714506864,
|
964 |
+
"eval_rougeL": 0.5363802967084452,
|
965 |
+
"eval_rougeLsum": 0.6516690557971352,
|
966 |
+
"eval_runtime": 14.9932,
|
967 |
+
"eval_samples_per_second": 0.667,
|
968 |
+
"eval_steps_per_second": 0.333,
|
969 |
+
"step": 9600
|
970 |
+
},
|
971 |
+
{
|
972 |
+
"epoch": 1.12,
|
973 |
+
"grad_norm": 0.26574915647506714,
|
974 |
+
"learning_rate": 0.0001441991625601732,
|
975 |
+
"loss": 1.5342,
|
976 |
+
"step": 9800
|
977 |
+
},
|
978 |
+
{
|
979 |
+
"epoch": 1.12,
|
980 |
+
"eval_bertscore": 0.7328712344169617,
|
981 |
+
"eval_loss": 1.7393991947174072,
|
982 |
+
"eval_rouge1": 0.6856504003396884,
|
983 |
+
"eval_rouge2": 0.3761098841062477,
|
984 |
+
"eval_rougeL": 0.555477293163325,
|
985 |
+
"eval_rougeLsum": 0.6757574283262289,
|
986 |
+
"eval_runtime": 14.8121,
|
987 |
+
"eval_samples_per_second": 0.675,
|
988 |
+
"eval_steps_per_second": 0.338,
|
989 |
+
"step": 9800
|
990 |
+
},
|
991 |
+
{
|
992 |
+
"epoch": 1.14,
|
993 |
+
"grad_norm": 0.315468430519104,
|
994 |
+
"learning_rate": 0.00014305978864613897,
|
995 |
+
"loss": 1.543,
|
996 |
+
"step": 10000
|
997 |
+
},
|
998 |
+
{
|
999 |
+
"epoch": 1.14,
|
1000 |
+
"eval_bertscore": 0.7349387407302856,
|
1001 |
+
"eval_loss": 1.7352710962295532,
|
1002 |
+
"eval_rouge1": 0.6749953128982036,
|
1003 |
+
"eval_rouge2": 0.3720385250530084,
|
1004 |
+
"eval_rougeL": 0.5472261566474382,
|
1005 |
+
"eval_rougeLsum": 0.6657643539219252,
|
1006 |
+
"eval_runtime": 14.909,
|
1007 |
+
"eval_samples_per_second": 0.671,
|
1008 |
+
"eval_steps_per_second": 0.335,
|
1009 |
+
"step": 10000
|
1010 |
+
},
|
1011 |
+
{
|
1012 |
+
"epoch": 1.16,
|
1013 |
+
"grad_norm": 0.29815027117729187,
|
1014 |
+
"learning_rate": 0.00014192041473210472,
|
1015 |
+
"loss": 1.5547,
|
1016 |
+
"step": 10200
|
1017 |
+
},
|
1018 |
+
{
|
1019 |
+
"epoch": 1.16,
|
1020 |
+
"eval_bertscore": 0.7359883189201355,
|
1021 |
+
"eval_loss": 1.7269136905670166,
|
1022 |
+
"eval_rouge1": 0.6561141614088863,
|
1023 |
+
"eval_rouge2": 0.3606175666303814,
|
1024 |
+
"eval_rougeL": 0.5302270771032793,
|
1025 |
+
"eval_rougeLsum": 0.6446912079521883,
|
1026 |
+
"eval_runtime": 14.9527,
|
1027 |
+
"eval_samples_per_second": 0.669,
|
1028 |
+
"eval_steps_per_second": 0.334,
|
1029 |
+
"step": 10200
|
1030 |
+
},
|
1031 |
+
{
|
1032 |
+
"epoch": 1.18,
|
1033 |
+
"grad_norm": 0.3595702350139618,
|
1034 |
+
"learning_rate": 0.00014078104081807047,
|
1035 |
+
"loss": 1.5567,
|
1036 |
+
"step": 10400
|
1037 |
+
},
|
1038 |
+
{
|
1039 |
+
"epoch": 1.18,
|
1040 |
+
"eval_bertscore": 0.7328116297721863,
|
1041 |
+
"eval_loss": 1.7341644763946533,
|
1042 |
+
"eval_rouge1": 0.6420332714823549,
|
1043 |
+
"eval_rouge2": 0.35094864549032,
|
1044 |
+
"eval_rougeL": 0.5179556761398367,
|
1045 |
+
"eval_rougeLsum": 0.631987397226809,
|
1046 |
+
"eval_runtime": 15.1438,
|
1047 |
+
"eval_samples_per_second": 0.66,
|
1048 |
+
"eval_steps_per_second": 0.33,
|
1049 |
+
"step": 10400
|
1050 |
+
},
|
1051 |
+
{
|
1052 |
+
"epoch": 1.21,
|
1053 |
+
"grad_norm": 0.2718666195869446,
|
1054 |
+
"learning_rate": 0.00013964166690403623,
|
1055 |
+
"loss": 1.5408,
|
1056 |
+
"step": 10600
|
1057 |
+
},
|
1058 |
+
{
|
1059 |
+
"epoch": 1.21,
|
1060 |
+
"eval_bertscore": 0.7337731122970581,
|
1061 |
+
"eval_loss": 1.7330901622772217,
|
1062 |
+
"eval_rouge1": 0.661681342484687,
|
1063 |
+
"eval_rouge2": 0.3626833509973693,
|
1064 |
+
"eval_rougeL": 0.5329424447373774,
|
1065 |
+
"eval_rougeLsum": 0.6519750177144633,
|
1066 |
+
"eval_runtime": 14.8142,
|
1067 |
+
"eval_samples_per_second": 0.675,
|
1068 |
+
"eval_steps_per_second": 0.338,
|
1069 |
+
"step": 10600
|
1070 |
+
},
|
1071 |
+
{
|
1072 |
+
"epoch": 1.23,
|
1073 |
+
"grad_norm": 0.29183274507522583,
|
1074 |
+
"learning_rate": 0.00013850229299000198,
|
1075 |
+
"loss": 1.5422,
|
1076 |
+
"step": 10800
|
1077 |
+
},
|
1078 |
+
{
|
1079 |
+
"epoch": 1.23,
|
1080 |
+
"eval_bertscore": 0.7331051230430603,
|
1081 |
+
"eval_loss": 1.7297636270523071,
|
1082 |
+
"eval_rouge1": 0.6655497978238063,
|
1083 |
+
"eval_rouge2": 0.3614235788441926,
|
1084 |
+
"eval_rougeL": 0.5327210061667442,
|
1085 |
+
"eval_rougeLsum": 0.6548836840483913,
|
1086 |
+
"eval_runtime": 15.1359,
|
1087 |
+
"eval_samples_per_second": 0.661,
|
1088 |
+
"eval_steps_per_second": 0.33,
|
1089 |
+
"step": 10800
|
1090 |
+
},
|
1091 |
+
{
|
1092 |
+
"epoch": 1.25,
|
1093 |
+
"grad_norm": 0.30979740619659424,
|
1094 |
+
"learning_rate": 0.00013736291907596776,
|
1095 |
+
"loss": 1.5372,
|
1096 |
+
"step": 11000
|
1097 |
+
},
|
1098 |
+
{
|
1099 |
+
"epoch": 1.25,
|
1100 |
+
"eval_bertscore": 0.7312799692153931,
|
1101 |
+
"eval_loss": 1.732444167137146,
|
1102 |
+
"eval_rouge1": 0.6568292865033993,
|
1103 |
+
"eval_rouge2": 0.35876682221562006,
|
1104 |
+
"eval_rougeL": 0.5300878844981931,
|
1105 |
+
"eval_rougeLsum": 0.6461751645858989,
|
1106 |
+
"eval_runtime": 14.819,
|
1107 |
+
"eval_samples_per_second": 0.675,
|
1108 |
+
"eval_steps_per_second": 0.337,
|
1109 |
+
"step": 11000
|
1110 |
+
},
|
1111 |
+
{
|
1112 |
+
"epoch": 1.28,
|
1113 |
+
"grad_norm": 0.31343138217926025,
|
1114 |
+
"learning_rate": 0.0001362235451619335,
|
1115 |
+
"loss": 1.5301,
|
1116 |
+
"step": 11200
|
1117 |
+
},
|
1118 |
+
{
|
1119 |
+
"epoch": 1.28,
|
1120 |
+
"eval_bertscore": 0.7317885160446167,
|
1121 |
+
"eval_loss": 1.7358499765396118,
|
1122 |
+
"eval_rouge1": 0.6548673097943329,
|
1123 |
+
"eval_rouge2": 0.3609116081432997,
|
1124 |
+
"eval_rougeL": 0.5279887650752133,
|
1125 |
+
"eval_rougeLsum": 0.6466232329097188,
|
1126 |
+
"eval_runtime": 14.8259,
|
1127 |
+
"eval_samples_per_second": 0.674,
|
1128 |
+
"eval_steps_per_second": 0.337,
|
1129 |
+
"step": 11200
|
1130 |
+
},
|
1131 |
+
{
|
1132 |
+
"epoch": 1.3,
|
1133 |
+
"grad_norm": 0.36181533336639404,
|
1134 |
+
"learning_rate": 0.0001350841712478993,
|
1135 |
+
"loss": 1.5421,
|
1136 |
+
"step": 11400
|
1137 |
+
},
|
1138 |
+
{
|
1139 |
+
"epoch": 1.3,
|
1140 |
+
"eval_bertscore": 0.7316756248474121,
|
1141 |
+
"eval_loss": 1.7282969951629639,
|
1142 |
+
"eval_rouge1": 0.6551882964480251,
|
1143 |
+
"eval_rouge2": 0.3580708921400697,
|
1144 |
+
"eval_rougeL": 0.5255367305995147,
|
1145 |
+
"eval_rougeLsum": 0.6449192953008009,
|
1146 |
+
"eval_runtime": 14.8816,
|
1147 |
+
"eval_samples_per_second": 0.672,
|
1148 |
+
"eval_steps_per_second": 0.336,
|
1149 |
+
"step": 11400
|
1150 |
+
},
|
1151 |
+
{
|
1152 |
+
"epoch": 1.32,
|
1153 |
+
"grad_norm": 0.30600836873054504,
|
1154 |
+
"learning_rate": 0.00013394479733386505,
|
1155 |
+
"loss": 1.5538,
|
1156 |
+
"step": 11600
|
1157 |
+
},
|
1158 |
+
{
|
1159 |
+
"epoch": 1.32,
|
1160 |
+
"eval_bertscore": 0.7311854362487793,
|
1161 |
+
"eval_loss": 1.7313562631607056,
|
1162 |
+
"eval_rouge1": 0.6592751199424156,
|
1163 |
+
"eval_rouge2": 0.35802855072854206,
|
1164 |
+
"eval_rougeL": 0.5297288176377084,
|
1165 |
+
"eval_rougeLsum": 0.6489455314962717,
|
1166 |
+
"eval_runtime": 15.0693,
|
1167 |
+
"eval_samples_per_second": 0.664,
|
1168 |
+
"eval_steps_per_second": 0.332,
|
1169 |
+
"step": 11600
|
1170 |
+
},
|
1171 |
+
{
|
1172 |
+
"epoch": 1.34,
|
1173 |
+
"grad_norm": 0.29904893040657043,
|
1174 |
+
"learning_rate": 0.0001328054234198308,
|
1175 |
+
"loss": 1.5328,
|
1176 |
+
"step": 11800
|
1177 |
+
},
|
1178 |
+
{
|
1179 |
+
"epoch": 1.34,
|
1180 |
+
"eval_bertscore": 0.7312635183334351,
|
1181 |
+
"eval_loss": 1.7318429946899414,
|
1182 |
+
"eval_rouge1": 0.6577169369077195,
|
1183 |
+
"eval_rouge2": 0.3582474830918887,
|
1184 |
+
"eval_rougeL": 0.5314990647771975,
|
1185 |
+
"eval_rougeLsum": 0.6454785220479866,
|
1186 |
+
"eval_runtime": 15.0235,
|
1187 |
+
"eval_samples_per_second": 0.666,
|
1188 |
+
"eval_steps_per_second": 0.333,
|
1189 |
+
"step": 11800
|
1190 |
+
},
|
1191 |
+
{
|
1192 |
+
"epoch": 1.37,
|
1193 |
+
"grad_norm": 0.3025416433811188,
|
1194 |
+
"learning_rate": 0.00013166604950579658,
|
1195 |
+
"loss": 1.5349,
|
1196 |
+
"step": 12000
|
1197 |
+
},
|
1198 |
+
{
|
1199 |
+
"epoch": 1.37,
|
1200 |
+
"eval_bertscore": 0.7325812578201294,
|
1201 |
+
"eval_loss": 1.7309118509292603,
|
1202 |
+
"eval_rouge1": 0.6629133074951261,
|
1203 |
+
"eval_rouge2": 0.3678158453940578,
|
1204 |
+
"eval_rougeL": 0.5380936907276155,
|
1205 |
+
"eval_rougeLsum": 0.654883061928214,
|
1206 |
+
"eval_runtime": 14.7666,
|
1207 |
+
"eval_samples_per_second": 0.677,
|
1208 |
+
"eval_steps_per_second": 0.339,
|
1209 |
+
"step": 12000
|
1210 |
+
},
|
1211 |
+
{
|
1212 |
+
"epoch": 1.39,
|
1213 |
+
"grad_norm": 0.34982389211654663,
|
1214 |
+
"learning_rate": 0.00013052667559176233,
|
1215 |
+
"loss": 1.5513,
|
1216 |
+
"step": 12200
|
1217 |
+
},
|
1218 |
+
{
|
1219 |
+
"epoch": 1.39,
|
1220 |
+
"eval_bertscore": 0.7340582609176636,
|
1221 |
+
"eval_loss": 1.7363474369049072,
|
1222 |
+
"eval_rouge1": 0.6555817937418287,
|
1223 |
+
"eval_rouge2": 0.35630500078358396,
|
1224 |
+
"eval_rougeL": 0.5272412353478366,
|
1225 |
+
"eval_rougeLsum": 0.6445837479327643,
|
1226 |
+
"eval_runtime": 14.9664,
|
1227 |
+
"eval_samples_per_second": 0.668,
|
1228 |
+
"eval_steps_per_second": 0.334,
|
1229 |
+
"step": 12200
|
1230 |
+
},
|
1231 |
+
{
|
1232 |
+
"epoch": 1.41,
|
1233 |
+
"grad_norm": 0.35809043049812317,
|
1234 |
+
"learning_rate": 0.0001293873016777281,
|
1235 |
+
"loss": 1.5444,
|
1236 |
+
"step": 12400
|
1237 |
+
},
|
1238 |
+
{
|
1239 |
+
"epoch": 1.41,
|
1240 |
+
"eval_bertscore": 0.7334069013595581,
|
1241 |
+
"eval_loss": 1.7328729629516602,
|
1242 |
+
"eval_rouge1": 0.6617971237669364,
|
1243 |
+
"eval_rouge2": 0.35951260376512423,
|
1244 |
+
"eval_rougeL": 0.5345512305507059,
|
1245 |
+
"eval_rougeLsum": 0.648363531132752,
|
1246 |
+
"eval_runtime": 15.2146,
|
1247 |
+
"eval_samples_per_second": 0.657,
|
1248 |
+
"eval_steps_per_second": 0.329,
|
1249 |
+
"step": 12400
|
1250 |
+
},
|
1251 |
+
{
|
1252 |
+
"epoch": 1.44,
|
1253 |
+
"grad_norm": 0.2954196631908417,
|
1254 |
+
"learning_rate": 0.00012824792776369387,
|
1255 |
+
"loss": 1.5406,
|
1256 |
+
"step": 12600
|
1257 |
+
},
|
1258 |
+
{
|
1259 |
+
"epoch": 1.44,
|
1260 |
+
"eval_bertscore": 0.7321678400039673,
|
1261 |
+
"eval_loss": 1.7335160970687866,
|
1262 |
+
"eval_rouge1": 0.6573625593086756,
|
1263 |
+
"eval_rouge2": 0.36210525247389347,
|
1264 |
+
"eval_rougeL": 0.5379361120230158,
|
1265 |
+
"eval_rougeLsum": 0.6459787883452857,
|
1266 |
+
"eval_runtime": 14.8942,
|
1267 |
+
"eval_samples_per_second": 0.671,
|
1268 |
+
"eval_steps_per_second": 0.336,
|
1269 |
+
"step": 12600
|
1270 |
+
},
|
1271 |
+
{
|
1272 |
+
"epoch": 1.46,
|
1273 |
+
"grad_norm": 0.32190588116645813,
|
1274 |
+
"learning_rate": 0.00012710855384965962,
|
1275 |
+
"loss": 1.5491,
|
1276 |
+
"step": 12800
|
1277 |
+
},
|
1278 |
+
{
|
1279 |
+
"epoch": 1.46,
|
1280 |
+
"eval_bertscore": 0.7346011400222778,
|
1281 |
+
"eval_loss": 1.7364966869354248,
|
1282 |
+
"eval_rouge1": 0.6481210247390559,
|
1283 |
+
"eval_rouge2": 0.3521173896017687,
|
1284 |
+
"eval_rougeL": 0.5240500581372636,
|
1285 |
+
"eval_rougeLsum": 0.63706442433335,
|
1286 |
+
"eval_runtime": 14.923,
|
1287 |
+
"eval_samples_per_second": 0.67,
|
1288 |
+
"eval_steps_per_second": 0.335,
|
1289 |
+
"step": 12800
|
1290 |
+
},
|
1291 |
+
{
|
1292 |
+
"epoch": 1.48,
|
1293 |
+
"grad_norm": 0.33323267102241516,
|
1294 |
+
"learning_rate": 0.00012596917993562537,
|
1295 |
+
"loss": 1.5596,
|
1296 |
+
"step": 13000
|
1297 |
+
},
|
1298 |
+
{
|
1299 |
+
"epoch": 1.48,
|
1300 |
+
"eval_bertscore": 0.7331587076187134,
|
1301 |
+
"eval_loss": 1.7332260608673096,
|
1302 |
+
"eval_rouge1": 0.6561257401878793,
|
1303 |
+
"eval_rouge2": 0.3548063723792664,
|
1304 |
+
"eval_rougeL": 0.527807776001489,
|
1305 |
+
"eval_rougeLsum": 0.6451911907984706,
|
1306 |
+
"eval_runtime": 15.4703,
|
1307 |
+
"eval_samples_per_second": 0.646,
|
1308 |
+
"eval_steps_per_second": 0.323,
|
1309 |
+
"step": 13000
|
1310 |
+
},
|
1311 |
+
{
|
1312 |
+
"epoch": 1.5,
|
1313 |
+
"grad_norm": 0.3564057946205139,
|
1314 |
+
"learning_rate": 0.00012482980602159113,
|
1315 |
+
"loss": 1.5261,
|
1316 |
+
"step": 13200
|
1317 |
+
},
|
1318 |
+
{
|
1319 |
+
"epoch": 1.5,
|
1320 |
+
"eval_bertscore": 0.7306063771247864,
|
1321 |
+
"eval_loss": 1.7368810176849365,
|
1322 |
+
"eval_rouge1": 0.637722723890071,
|
1323 |
+
"eval_rouge2": 0.3455358728458236,
|
1324 |
+
"eval_rougeL": 0.5136372690435154,
|
1325 |
+
"eval_rougeLsum": 0.6273570573595115,
|
1326 |
+
"eval_runtime": 15.3533,
|
1327 |
+
"eval_samples_per_second": 0.651,
|
1328 |
+
"eval_steps_per_second": 0.326,
|
1329 |
+
"step": 13200
|
1330 |
+
},
|
1331 |
+
{
|
1332 |
+
"epoch": 1.53,
|
1333 |
+
"grad_norm": 0.29219934344291687,
|
1334 |
+
"learning_rate": 0.0001236904321075569,
|
1335 |
+
"loss": 1.519,
|
1336 |
+
"step": 13400
|
1337 |
+
},
|
1338 |
+
{
|
1339 |
+
"epoch": 1.53,
|
1340 |
+
"eval_bertscore": 0.7338696122169495,
|
1341 |
+
"eval_loss": 1.734724998474121,
|
1342 |
+
"eval_rouge1": 0.6442107446420164,
|
1343 |
+
"eval_rouge2": 0.3494748457109431,
|
1344 |
+
"eval_rougeL": 0.5207483892007314,
|
1345 |
+
"eval_rougeLsum": 0.632886404907802,
|
1346 |
+
"eval_runtime": 15.3077,
|
1347 |
+
"eval_samples_per_second": 0.653,
|
1348 |
+
"eval_steps_per_second": 0.327,
|
1349 |
+
"step": 13400
|
1350 |
+
},
|
1351 |
+
{
|
1352 |
+
"epoch": 1.55,
|
1353 |
+
"grad_norm": 0.34681758284568787,
|
1354 |
+
"learning_rate": 0.00012255105819352266,
|
1355 |
+
"loss": 1.5419,
|
1356 |
+
"step": 13600
|
1357 |
+
},
|
1358 |
+
{
|
1359 |
+
"epoch": 1.55,
|
1360 |
+
"eval_bertscore": 0.7350045442581177,
|
1361 |
+
"eval_loss": 1.7329858541488647,
|
1362 |
+
"eval_rouge1": 0.6606839869796519,
|
1363 |
+
"eval_rouge2": 0.362188561160822,
|
1364 |
+
"eval_rougeL": 0.5342033818317451,
|
1365 |
+
"eval_rougeLsum": 0.6493340000068861,
|
1366 |
+
"eval_runtime": 15.44,
|
1367 |
+
"eval_samples_per_second": 0.648,
|
1368 |
+
"eval_steps_per_second": 0.324,
|
1369 |
+
"step": 13600
|
1370 |
+
},
|
1371 |
+
{
|
1372 |
+
"epoch": 1.57,
|
1373 |
+
"grad_norm": 0.3043666481971741,
|
1374 |
+
"learning_rate": 0.00012141168427948844,
|
1375 |
+
"loss": 1.5402,
|
1376 |
+
"step": 13800
|
1377 |
+
},
|
1378 |
+
{
|
1379 |
+
"epoch": 1.57,
|
1380 |
+
"eval_bertscore": 0.7363221645355225,
|
1381 |
+
"eval_loss": 1.7308530807495117,
|
1382 |
+
"eval_rouge1": 0.6638252384356028,
|
1383 |
+
"eval_rouge2": 0.3643237697892826,
|
1384 |
+
"eval_rougeL": 0.5403775887381331,
|
1385 |
+
"eval_rougeLsum": 0.6537260000827279,
|
1386 |
+
"eval_runtime": 14.7668,
|
1387 |
+
"eval_samples_per_second": 0.677,
|
1388 |
+
"eval_steps_per_second": 0.339,
|
1389 |
+
"step": 13800
|
1390 |
+
},
|
1391 |
+
{
|
1392 |
+
"epoch": 1.59,
|
1393 |
+
"grad_norm": 0.4073585867881775,
|
1394 |
+
"learning_rate": 0.00012027231036545419,
|
1395 |
+
"loss": 1.5256,
|
1396 |
+
"step": 14000
|
1397 |
+
},
|
1398 |
+
{
|
1399 |
+
"epoch": 1.59,
|
1400 |
+
"eval_bertscore": 0.7310279607772827,
|
1401 |
+
"eval_loss": 1.7326784133911133,
|
1402 |
+
"eval_rouge1": 0.6609594314120198,
|
1403 |
+
"eval_rouge2": 0.3601530714440473,
|
1404 |
+
"eval_rougeL": 0.5344452687135626,
|
1405 |
+
"eval_rougeLsum": 0.6480936554342305,
|
1406 |
+
"eval_runtime": 14.8565,
|
1407 |
+
"eval_samples_per_second": 0.673,
|
1408 |
+
"eval_steps_per_second": 0.337,
|
1409 |
+
"step": 14000
|
1410 |
+
},
|
1411 |
+
{
|
1412 |
+
"epoch": 1.62,
|
1413 |
+
"grad_norm": 0.3211813271045685,
|
1414 |
+
"learning_rate": 0.00011913293645141995,
|
1415 |
+
"loss": 1.5366,
|
1416 |
+
"step": 14200
|
1417 |
+
},
|
1418 |
+
{
|
1419 |
+
"epoch": 1.62,
|
1420 |
+
"eval_bertscore": 0.7356667518615723,
|
1421 |
+
"eval_loss": 1.7280094623565674,
|
1422 |
+
"eval_rouge1": 0.6519353227375031,
|
1423 |
+
"eval_rouge2": 0.3587025716186173,
|
1424 |
+
"eval_rougeL": 0.5306356200586075,
|
1425 |
+
"eval_rougeLsum": 0.6408870347994059,
|
1426 |
+
"eval_runtime": 14.9264,
|
1427 |
+
"eval_samples_per_second": 0.67,
|
1428 |
+
"eval_steps_per_second": 0.335,
|
1429 |
+
"step": 14200
|
1430 |
+
},
|
1431 |
+
{
|
1432 |
+
"epoch": 1.64,
|
1433 |
+
"grad_norm": 0.32776832580566406,
|
1434 |
+
"learning_rate": 0.00011799356253738571,
|
1435 |
+
"loss": 1.5504,
|
1436 |
+
"step": 14400
|
1437 |
+
},
|
1438 |
+
{
|
1439 |
+
"epoch": 1.64,
|
1440 |
+
"eval_bertscore": 0.7331353425979614,
|
1441 |
+
"eval_loss": 1.7308950424194336,
|
1442 |
+
"eval_rouge1": 0.6627702292814652,
|
1443 |
+
"eval_rouge2": 0.36117793957379707,
|
1444 |
+
"eval_rougeL": 0.5369305446079228,
|
1445 |
+
"eval_rougeLsum": 0.6516924083980089,
|
1446 |
+
"eval_runtime": 16.0138,
|
1447 |
+
"eval_samples_per_second": 0.624,
|
1448 |
+
"eval_steps_per_second": 0.312,
|
1449 |
+
"step": 14400
|
1450 |
+
},
|
1451 |
+
{
|
1452 |
+
"epoch": 1.66,
|
1453 |
+
"grad_norm": 0.3209726810455322,
|
1454 |
+
"learning_rate": 0.00011685418862335147,
|
1455 |
+
"loss": 1.5473,
|
1456 |
+
"step": 14600
|
1457 |
+
},
|
1458 |
+
{
|
1459 |
+
"epoch": 1.66,
|
1460 |
+
"eval_bertscore": 0.732498824596405,
|
1461 |
+
"eval_loss": 1.7328402996063232,
|
1462 |
+
"eval_rouge1": 0.6482679740803596,
|
1463 |
+
"eval_rouge2": 0.3538726087405498,
|
1464 |
+
"eval_rougeL": 0.5267677183598017,
|
1465 |
+
"eval_rougeLsum": 0.6366529460029322,
|
1466 |
+
"eval_runtime": 15.0195,
|
1467 |
+
"eval_samples_per_second": 0.666,
|
1468 |
+
"eval_steps_per_second": 0.333,
|
1469 |
+
"step": 14600
|
1470 |
+
},
|
1471 |
+
{
|
1472 |
+
"epoch": 1.69,
|
1473 |
+
"grad_norm": 0.3174591064453125,
|
1474 |
+
"learning_rate": 0.00011571481470931725,
|
1475 |
+
"loss": 1.5568,
|
1476 |
+
"step": 14800
|
1477 |
+
},
|
1478 |
+
{
|
1479 |
+
"epoch": 1.69,
|
1480 |
+
"eval_bertscore": 0.7335298657417297,
|
1481 |
+
"eval_loss": 1.7310253381729126,
|
1482 |
+
"eval_rouge1": 0.6560468577439627,
|
1483 |
+
"eval_rouge2": 0.36039371229175,
|
1484 |
+
"eval_rougeL": 0.5318708569729291,
|
1485 |
+
"eval_rougeLsum": 0.6444857558837042,
|
1486 |
+
"eval_runtime": 14.9774,
|
1487 |
+
"eval_samples_per_second": 0.668,
|
1488 |
+
"eval_steps_per_second": 0.334,
|
1489 |
+
"step": 14800
|
1490 |
+
},
|
1491 |
+
{
|
1492 |
+
"epoch": 1.71,
|
1493 |
+
"grad_norm": 0.2936408817768097,
|
1494 |
+
"learning_rate": 0.000114575440795283,
|
1495 |
+
"loss": 1.5345,
|
1496 |
+
"step": 15000
|
1497 |
+
},
|
1498 |
+
{
|
1499 |
+
"epoch": 1.71,
|
1500 |
+
"eval_bertscore": 0.7322725057601929,
|
1501 |
+
"eval_loss": 1.7270629405975342,
|
1502 |
+
"eval_rouge1": 0.6387060930656672,
|
1503 |
+
"eval_rouge2": 0.3480508127989137,
|
1504 |
+
"eval_rougeL": 0.5148670834213287,
|
1505 |
+
"eval_rougeLsum": 0.6273654952601909,
|
1506 |
+
"eval_runtime": 15.6687,
|
1507 |
+
"eval_samples_per_second": 0.638,
|
1508 |
+
"eval_steps_per_second": 0.319,
|
1509 |
+
"step": 15000
|
1510 |
+
},
|
1511 |
+
{
|
1512 |
+
"epoch": 1.73,
|
1513 |
+
"grad_norm": 0.32960689067840576,
|
1514 |
+
"learning_rate": 0.00011343606688124875,
|
1515 |
+
"loss": 1.5362,
|
1516 |
+
"step": 15200
|
1517 |
+
},
|
1518 |
+
{
|
1519 |
+
"epoch": 1.73,
|
1520 |
+
"eval_bertscore": 0.7337037920951843,
|
1521 |
+
"eval_loss": 1.7287395000457764,
|
1522 |
+
"eval_rouge1": 0.6476816970229771,
|
1523 |
+
"eval_rouge2": 0.3532248216683249,
|
1524 |
+
"eval_rougeL": 0.5253136618838716,
|
1525 |
+
"eval_rougeLsum": 0.6347493764394183,
|
1526 |
+
"eval_runtime": 15.0045,
|
1527 |
+
"eval_samples_per_second": 0.666,
|
1528 |
+
"eval_steps_per_second": 0.333,
|
1529 |
+
"step": 15200
|
1530 |
+
},
|
1531 |
+
{
|
1532 |
+
"epoch": 1.75,
|
1533 |
+
"grad_norm": 0.33265602588653564,
|
1534 |
+
"learning_rate": 0.00011229669296721452,
|
1535 |
+
"loss": 1.5215,
|
1536 |
+
"step": 15400
|
1537 |
+
},
|
1538 |
+
{
|
1539 |
+
"epoch": 1.75,
|
1540 |
+
"eval_bertscore": 0.7330806851387024,
|
1541 |
+
"eval_loss": 1.7265052795410156,
|
1542 |
+
"eval_rouge1": 0.6529393512177359,
|
1543 |
+
"eval_rouge2": 0.36182153145062224,
|
1544 |
+
"eval_rougeL": 0.5317061134915853,
|
1545 |
+
"eval_rougeLsum": 0.6413066299251913,
|
1546 |
+
"eval_runtime": 15.0256,
|
1547 |
+
"eval_samples_per_second": 0.666,
|
1548 |
+
"eval_steps_per_second": 0.333,
|
1549 |
+
"step": 15400
|
1550 |
+
},
|
1551 |
+
{
|
1552 |
+
"epoch": 1.78,
|
1553 |
+
"grad_norm": 0.3436201512813568,
|
1554 |
+
"learning_rate": 0.00011115731905318027,
|
1555 |
+
"loss": 1.539,
|
1556 |
+
"step": 15600
|
1557 |
+
},
|
1558 |
+
{
|
1559 |
+
"epoch": 1.78,
|
1560 |
+
"eval_bertscore": 0.7335551977157593,
|
1561 |
+
"eval_loss": 1.7254730463027954,
|
1562 |
+
"eval_rouge1": 0.6388518781767971,
|
1563 |
+
"eval_rouge2": 0.3501853846588857,
|
1564 |
+
"eval_rougeL": 0.5196828245794569,
|
1565 |
+
"eval_rougeLsum": 0.629333993884722,
|
1566 |
+
"eval_runtime": 15.2595,
|
1567 |
+
"eval_samples_per_second": 0.655,
|
1568 |
+
"eval_steps_per_second": 0.328,
|
1569 |
+
"step": 15600
|
1570 |
+
},
|
1571 |
+
{
|
1572 |
+
"epoch": 1.8,
|
1573 |
+
"grad_norm": 0.3428190350532532,
|
1574 |
+
"learning_rate": 0.00011001794513914605,
|
1575 |
+
"loss": 1.5273,
|
1576 |
+
"step": 15800
|
1577 |
+
},
|
1578 |
+
{
|
1579 |
+
"epoch": 1.8,
|
1580 |
+
"eval_bertscore": 0.7331770658493042,
|
1581 |
+
"eval_loss": 1.7286018133163452,
|
1582 |
+
"eval_rouge1": 0.6581941047310954,
|
1583 |
+
"eval_rouge2": 0.36277983926897583,
|
1584 |
+
"eval_rougeL": 0.5336464680120501,
|
1585 |
+
"eval_rougeLsum": 0.6489239720278894,
|
1586 |
+
"eval_runtime": 14.8252,
|
1587 |
+
"eval_samples_per_second": 0.675,
|
1588 |
+
"eval_steps_per_second": 0.337,
|
1589 |
+
"step": 15800
|
1590 |
+
},
|
1591 |
+
{
|
1592 |
+
"epoch": 1.82,
|
1593 |
+
"grad_norm": 0.363164484500885,
|
1594 |
+
"learning_rate": 0.0001088785712251118,
|
1595 |
+
"loss": 1.5445,
|
1596 |
+
"step": 16000
|
1597 |
+
},
|
1598 |
+
{
|
1599 |
+
"epoch": 1.82,
|
1600 |
+
"eval_bertscore": 0.7377282977104187,
|
1601 |
+
"eval_loss": 1.7363064289093018,
|
1602 |
+
"eval_rouge1": 0.6547011011872876,
|
1603 |
+
"eval_rouge2": 0.3553220826957326,
|
1604 |
+
"eval_rougeL": 0.5256073814411315,
|
1605 |
+
"eval_rougeLsum": 0.6420095316923398,
|
1606 |
+
"eval_runtime": 14.8599,
|
1607 |
+
"eval_samples_per_second": 0.673,
|
1608 |
+
"eval_steps_per_second": 0.336,
|
1609 |
+
"step": 16000
|
1610 |
+
},
|
1611 |
+
{
|
1612 |
+
"epoch": 1.85,
|
1613 |
+
"grad_norm": 0.3098333775997162,
|
1614 |
+
"learning_rate": 0.00010773919731107757,
|
1615 |
+
"loss": 1.5319,
|
1616 |
+
"step": 16200
|
1617 |
+
},
|
1618 |
+
{
|
1619 |
+
"epoch": 1.85,
|
1620 |
+
"eval_bertscore": 0.7324053645133972,
|
1621 |
+
"eval_loss": 1.7284066677093506,
|
1622 |
+
"eval_rouge1": 0.6477379941950916,
|
1623 |
+
"eval_rouge2": 0.3535918140554809,
|
1624 |
+
"eval_rougeL": 0.5226838544730126,
|
1625 |
+
"eval_rougeLsum": 0.6373271915355557,
|
1626 |
+
"eval_runtime": 14.9318,
|
1627 |
+
"eval_samples_per_second": 0.67,
|
1628 |
+
"eval_steps_per_second": 0.335,
|
1629 |
+
"step": 16200
|
1630 |
+
},
|
1631 |
+
{
|
1632 |
+
"epoch": 1.87,
|
1633 |
+
"grad_norm": 0.3637208938598633,
|
1634 |
+
"learning_rate": 0.00010659982339704332,
|
1635 |
+
"loss": 1.5442,
|
1636 |
+
"step": 16400
|
1637 |
+
},
|
1638 |
+
{
|
1639 |
+
"epoch": 1.87,
|
1640 |
+
"eval_bertscore": 0.7347462773323059,
|
1641 |
+
"eval_loss": 1.7252963781356812,
|
1642 |
+
"eval_rouge1": 0.6494449840449819,
|
1643 |
+
"eval_rouge2": 0.3586550050575282,
|
1644 |
+
"eval_rougeL": 0.5275675395159809,
|
1645 |
+
"eval_rougeLsum": 0.6396738714026391,
|
1646 |
+
"eval_runtime": 15.2218,
|
1647 |
+
"eval_samples_per_second": 0.657,
|
1648 |
+
"eval_steps_per_second": 0.328,
|
1649 |
+
"step": 16400
|
1650 |
+
},
|
1651 |
+
{
|
1652 |
+
"epoch": 1.89,
|
1653 |
+
"grad_norm": 0.35197457671165466,
|
1654 |
+
"learning_rate": 0.00010546044948300908,
|
1655 |
+
"loss": 1.5131,
|
1656 |
+
"step": 16600
|
1657 |
+
},
|
1658 |
+
{
|
1659 |
+
"epoch": 1.89,
|
1660 |
+
"eval_bertscore": 0.7329785227775574,
|
1661 |
+
"eval_loss": 1.7285687923431396,
|
1662 |
+
"eval_rouge1": 0.6582047811143328,
|
1663 |
+
"eval_rouge2": 0.3637700686094697,
|
1664 |
+
"eval_rougeL": 0.5355021948480279,
|
1665 |
+
"eval_rougeLsum": 0.6483245595148677,
|
1666 |
+
"eval_runtime": 14.8772,
|
1667 |
+
"eval_samples_per_second": 0.672,
|
1668 |
+
"eval_steps_per_second": 0.336,
|
1669 |
+
"step": 16600
|
1670 |
+
},
|
1671 |
+
{
|
1672 |
+
"epoch": 1.91,
|
1673 |
+
"grad_norm": 0.3406757116317749,
|
1674 |
+
"learning_rate": 0.00010432107556897486,
|
1675 |
+
"loss": 1.5394,
|
1676 |
+
"step": 16800
|
1677 |
+
},
|
1678 |
+
{
|
1679 |
+
"epoch": 1.91,
|
1680 |
+
"eval_bertscore": 0.7345961332321167,
|
1681 |
+
"eval_loss": 1.7324028015136719,
|
1682 |
+
"eval_rouge1": 0.6408293615351552,
|
1683 |
+
"eval_rouge2": 0.3520120690778129,
|
1684 |
+
"eval_rougeL": 0.5145218014745592,
|
1685 |
+
"eval_rougeLsum": 0.6297802607384266,
|
1686 |
+
"eval_runtime": 15.1044,
|
1687 |
+
"eval_samples_per_second": 0.662,
|
1688 |
+
"eval_steps_per_second": 0.331,
|
1689 |
+
"step": 16800
|
1690 |
+
},
|
1691 |
+
{
|
1692 |
+
"epoch": 1.94,
|
1693 |
+
"grad_norm": 0.3417683243751526,
|
1694 |
+
"learning_rate": 0.00010318170165494061,
|
1695 |
+
"loss": 1.526,
|
1696 |
+
"step": 17000
|
1697 |
+
},
|
1698 |
+
{
|
1699 |
+
"epoch": 1.94,
|
1700 |
+
"eval_bertscore": 0.735752522945404,
|
1701 |
+
"eval_loss": 1.7288110256195068,
|
1702 |
+
"eval_rouge1": 0.641158513352794,
|
1703 |
+
"eval_rouge2": 0.3544166440855814,
|
1704 |
+
"eval_rougeL": 0.5215201980495414,
|
1705 |
+
"eval_rougeLsum": 0.630550065494593,
|
1706 |
+
"eval_runtime": 15.0797,
|
1707 |
+
"eval_samples_per_second": 0.663,
|
1708 |
+
"eval_steps_per_second": 0.332,
|
1709 |
+
"step": 17000
|
1710 |
+
},
|
1711 |
+
{
|
1712 |
+
"epoch": 1.96,
|
1713 |
+
"grad_norm": 0.3256611227989197,
|
1714 |
+
"learning_rate": 0.00010204232774090639,
|
1715 |
+
"loss": 1.5484,
|
1716 |
+
"step": 17200
|
1717 |
+
},
|
1718 |
+
{
|
1719 |
+
"epoch": 1.96,
|
1720 |
+
"eval_bertscore": 0.7356327772140503,
|
1721 |
+
"eval_loss": 1.7305186986923218,
|
1722 |
+
"eval_rouge1": 0.6400269226515611,
|
1723 |
+
"eval_rouge2": 0.3502884634173268,
|
1724 |
+
"eval_rougeL": 0.517312321281175,
|
1725 |
+
"eval_rougeLsum": 0.6284556997614409,
|
1726 |
+
"eval_runtime": 15.4097,
|
1727 |
+
"eval_samples_per_second": 0.649,
|
1728 |
+
"eval_steps_per_second": 0.324,
|
1729 |
+
"step": 17200
|
1730 |
+
},
|
1731 |
+
{
|
1732 |
+
"epoch": 1.98,
|
1733 |
+
"grad_norm": 0.4035187363624573,
|
1734 |
+
"learning_rate": 0.00010090295382687213,
|
1735 |
+
"loss": 1.5261,
|
1736 |
+
"step": 17400
|
1737 |
+
},
|
1738 |
+
{
|
1739 |
+
"epoch": 1.98,
|
1740 |
+
"eval_bertscore": 0.7339992523193359,
|
1741 |
+
"eval_loss": 1.7282793521881104,
|
1742 |
+
"eval_rouge1": 0.6335770390416183,
|
1743 |
+
"eval_rouge2": 0.34592404578075897,
|
1744 |
+
"eval_rougeL": 0.5109045259792113,
|
1745 |
+
"eval_rougeLsum": 0.6218413683710426,
|
1746 |
+
"eval_runtime": 15.1959,
|
1747 |
+
"eval_samples_per_second": 0.658,
|
1748 |
+
"eval_steps_per_second": 0.329,
|
1749 |
+
"step": 17400
|
1750 |
+
},
|
1751 |
+
{
|
1752 |
+
"epoch": 2.0,
|
1753 |
+
"grad_norm": 0.34987062215805054,
|
1754 |
+
"learning_rate": 9.97635799128379e-05,
|
1755 |
+
"loss": 1.5199,
|
1756 |
+
"step": 17600
|
1757 |
+
},
|
1758 |
+
{
|
1759 |
+
"epoch": 2.0,
|
1760 |
+
"eval_bertscore": 0.7326329946517944,
|
1761 |
+
"eval_loss": 1.7544715404510498,
|
1762 |
+
"eval_rouge1": 0.6451558045750205,
|
1763 |
+
"eval_rouge2": 0.35565806935653943,
|
1764 |
+
"eval_rougeL": 0.5217034865840529,
|
1765 |
+
"eval_rougeLsum": 0.6329869715356753,
|
1766 |
+
"eval_runtime": 15.0351,
|
1767 |
+
"eval_samples_per_second": 0.665,
|
1768 |
+
"eval_steps_per_second": 0.333,
|
1769 |
+
"step": 17600
|
1770 |
+
},
|
1771 |
+
{
|
1772 |
+
"epoch": 2.03,
|
1773 |
+
"grad_norm": 0.37184038758277893,
|
1774 |
+
"learning_rate": 9.862420599880366e-05,
|
1775 |
+
"loss": 1.41,
|
1776 |
+
"step": 17800
|
1777 |
+
},
|
1778 |
+
{
|
1779 |
+
"epoch": 2.03,
|
1780 |
+
"eval_bertscore": 0.7315141558647156,
|
1781 |
+
"eval_loss": 1.7585878372192383,
|
1782 |
+
"eval_rouge1": 0.6469319193583706,
|
1783 |
+
"eval_rouge2": 0.3514447211469598,
|
1784 |
+
"eval_rougeL": 0.524755857688278,
|
1785 |
+
"eval_rougeLsum": 0.6350164781858667,
|
1786 |
+
"eval_runtime": 14.9583,
|
1787 |
+
"eval_samples_per_second": 0.669,
|
1788 |
+
"eval_steps_per_second": 0.334,
|
1789 |
+
"step": 17800
|
1790 |
+
},
|
1791 |
+
{
|
1792 |
+
"epoch": 2.05,
|
1793 |
+
"grad_norm": 0.3812776803970337,
|
1794 |
+
"learning_rate": 9.748483208476943e-05,
|
1795 |
+
"loss": 1.4132,
|
1796 |
+
"step": 18000
|
1797 |
+
},
|
1798 |
+
{
|
1799 |
+
"epoch": 2.05,
|
1800 |
+
"eval_bertscore": 0.7335561513900757,
|
1801 |
+
"eval_loss": 1.764611840248108,
|
1802 |
+
"eval_rouge1": 0.6381916780473581,
|
1803 |
+
"eval_rouge2": 0.3482510604092539,
|
1804 |
+
"eval_rougeL": 0.5162105225823392,
|
1805 |
+
"eval_rougeLsum": 0.627150245441782,
|
1806 |
+
"eval_runtime": 15.8515,
|
1807 |
+
"eval_samples_per_second": 0.631,
|
1808 |
+
"eval_steps_per_second": 0.315,
|
1809 |
+
"step": 18000
|
1810 |
+
},
|
1811 |
+
{
|
1812 |
+
"epoch": 2.07,
|
1813 |
+
"grad_norm": 0.45525220036506653,
|
1814 |
+
"learning_rate": 9.634545817073518e-05,
|
1815 |
+
"loss": 1.4,
|
1816 |
+
"step": 18200
|
1817 |
+
},
|
1818 |
+
{
|
1819 |
+
"epoch": 2.07,
|
1820 |
+
"eval_bertscore": 0.73627769947052,
|
1821 |
+
"eval_loss": 1.7585163116455078,
|
1822 |
+
"eval_rouge1": 0.6670097658027134,
|
1823 |
+
"eval_rouge2": 0.3658295359911405,
|
1824 |
+
"eval_rougeL": 0.5429667657900548,
|
1825 |
+
"eval_rougeLsum": 0.6543501745791419,
|
1826 |
+
"eval_runtime": 15.0301,
|
1827 |
+
"eval_samples_per_second": 0.665,
|
1828 |
+
"eval_steps_per_second": 0.333,
|
1829 |
+
"step": 18200
|
1830 |
+
},
|
1831 |
+
{
|
1832 |
+
"epoch": 2.1,
|
1833 |
+
"grad_norm": 0.37322184443473816,
|
1834 |
+
"learning_rate": 9.520608425670095e-05,
|
1835 |
+
"loss": 1.4293,
|
1836 |
+
"step": 18400
|
1837 |
+
},
|
1838 |
+
{
|
1839 |
+
"epoch": 2.1,
|
1840 |
+
"eval_bertscore": 0.730435848236084,
|
1841 |
+
"eval_loss": 1.764052391052246,
|
1842 |
+
"eval_rouge1": 0.6640215078213034,
|
1843 |
+
"eval_rouge2": 0.3625932287322054,
|
1844 |
+
"eval_rougeL": 0.5379978391335138,
|
1845 |
+
"eval_rougeLsum": 0.6542054656293199,
|
1846 |
+
"eval_runtime": 15.0762,
|
1847 |
+
"eval_samples_per_second": 0.663,
|
1848 |
+
"eval_steps_per_second": 0.332,
|
1849 |
+
"step": 18400
|
1850 |
+
},
|
1851 |
+
{
|
1852 |
+
"epoch": 2.12,
|
1853 |
+
"grad_norm": 0.4260891079902649,
|
1854 |
+
"learning_rate": 9.40667103426667e-05,
|
1855 |
+
"loss": 1.4077,
|
1856 |
+
"step": 18600
|
1857 |
+
},
|
1858 |
+
{
|
1859 |
+
"epoch": 2.12,
|
1860 |
+
"eval_bertscore": 0.7309869527816772,
|
1861 |
+
"eval_loss": 1.762108564376831,
|
1862 |
+
"eval_rouge1": 0.6571171081737958,
|
1863 |
+
"eval_rouge2": 0.35780421333141865,
|
1864 |
+
"eval_rougeL": 0.5320129270967632,
|
1865 |
+
"eval_rougeLsum": 0.64587787409523,
|
1866 |
+
"eval_runtime": 14.9004,
|
1867 |
+
"eval_samples_per_second": 0.671,
|
1868 |
+
"eval_steps_per_second": 0.336,
|
1869 |
+
"step": 18600
|
1870 |
+
},
|
1871 |
+
{
|
1872 |
+
"epoch": 2.14,
|
1873 |
+
"grad_norm": 0.39479926228523254,
|
1874 |
+
"learning_rate": 9.292733642863247e-05,
|
1875 |
+
"loss": 1.4165,
|
1876 |
+
"step": 18800
|
1877 |
+
},
|
1878 |
+
{
|
1879 |
+
"epoch": 2.14,
|
1880 |
+
"eval_bertscore": 0.7324444651603699,
|
1881 |
+
"eval_loss": 1.7607113122940063,
|
1882 |
+
"eval_rouge1": 0.6628398862884018,
|
1883 |
+
"eval_rouge2": 0.3627259806721216,
|
1884 |
+
"eval_rougeL": 0.5366106483832656,
|
1885 |
+
"eval_rougeLsum": 0.6528364858807157,
|
1886 |
+
"eval_runtime": 15.5766,
|
1887 |
+
"eval_samples_per_second": 0.642,
|
1888 |
+
"eval_steps_per_second": 0.321,
|
1889 |
+
"step": 18800
|
1890 |
+
},
|
1891 |
+
{
|
1892 |
+
"epoch": 2.16,
|
1893 |
+
"grad_norm": 0.39267703890800476,
|
1894 |
+
"learning_rate": 9.178796251459824e-05,
|
1895 |
+
"loss": 1.4123,
|
1896 |
+
"step": 19000
|
1897 |
+
},
|
1898 |
+
{
|
1899 |
+
"epoch": 2.16,
|
1900 |
+
"eval_bertscore": 0.7298994064331055,
|
1901 |
+
"eval_loss": 1.7668545246124268,
|
1902 |
+
"eval_rouge1": 0.6490850022857569,
|
1903 |
+
"eval_rouge2": 0.3532323419511264,
|
1904 |
+
"eval_rougeL": 0.5212823000193295,
|
1905 |
+
"eval_rougeLsum": 0.636442724466695,
|
1906 |
+
"eval_runtime": 14.9094,
|
1907 |
+
"eval_samples_per_second": 0.671,
|
1908 |
+
"eval_steps_per_second": 0.335,
|
1909 |
+
"step": 19000
|
1910 |
+
},
|
1911 |
+
{
|
1912 |
+
"epoch": 2.19,
|
1913 |
+
"grad_norm": 0.38221287727355957,
|
1914 |
+
"learning_rate": 9.0648588600564e-05,
|
1915 |
+
"loss": 1.401,
|
1916 |
+
"step": 19200
|
1917 |
+
},
|
1918 |
+
{
|
1919 |
+
"epoch": 2.19,
|
1920 |
+
"eval_bertscore": 0.7316875457763672,
|
1921 |
+
"eval_loss": 1.764147400856018,
|
1922 |
+
"eval_rouge1": 0.6490326710625849,
|
1923 |
+
"eval_rouge2": 0.3510351037900723,
|
1924 |
+
"eval_rougeL": 0.5239165028795836,
|
1925 |
+
"eval_rougeLsum": 0.6373687316421427,
|
1926 |
+
"eval_runtime": 15.1192,
|
1927 |
+
"eval_samples_per_second": 0.661,
|
1928 |
+
"eval_steps_per_second": 0.331,
|
1929 |
+
"step": 19200
|
1930 |
+
},
|
1931 |
+
{
|
1932 |
+
"epoch": 2.21,
|
1933 |
+
"grad_norm": 0.3653150200843811,
|
1934 |
+
"learning_rate": 8.950921468652976e-05,
|
1935 |
+
"loss": 1.4109,
|
1936 |
+
"step": 19400
|
1937 |
+
},
|
1938 |
+
{
|
1939 |
+
"epoch": 2.21,
|
1940 |
+
"eval_bertscore": 0.7348155975341797,
|
1941 |
+
"eval_loss": 1.7640550136566162,
|
1942 |
+
"eval_rouge1": 0.6462152873276823,
|
1943 |
+
"eval_rouge2": 0.3483599145461069,
|
1944 |
+
"eval_rougeL": 0.5193372430687719,
|
1945 |
+
"eval_rougeLsum": 0.6334254357511564,
|
1946 |
+
"eval_runtime": 14.9291,
|
1947 |
+
"eval_samples_per_second": 0.67,
|
1948 |
+
"eval_steps_per_second": 0.335,
|
1949 |
+
"step": 19400
|
1950 |
+
},
|
1951 |
+
{
|
1952 |
+
"epoch": 2.23,
|
1953 |
+
"grad_norm": 0.38049009442329407,
|
1954 |
+
"learning_rate": 8.836984077249551e-05,
|
1955 |
+
"loss": 1.4189,
|
1956 |
+
"step": 19600
|
1957 |
+
},
|
1958 |
+
{
|
1959 |
+
"epoch": 2.23,
|
1960 |
+
"eval_bertscore": 0.7357938885688782,
|
1961 |
+
"eval_loss": 1.7696326971054077,
|
1962 |
+
"eval_rouge1": 0.6377276221057538,
|
1963 |
+
"eval_rouge2": 0.3455397190390045,
|
1964 |
+
"eval_rougeL": 0.5118069428064842,
|
1965 |
+
"eval_rougeLsum": 0.6264501633078481,
|
1966 |
+
"eval_runtime": 14.8653,
|
1967 |
+
"eval_samples_per_second": 0.673,
|
1968 |
+
"eval_steps_per_second": 0.336,
|
1969 |
+
"step": 19600
|
1970 |
+
},
|
1971 |
+
{
|
1972 |
+
"epoch": 2.26,
|
1973 |
+
"grad_norm": 0.42111098766326904,
|
1974 |
+
"learning_rate": 8.723046685846128e-05,
|
1975 |
+
"loss": 1.4152,
|
1976 |
+
"step": 19800
|
1977 |
+
},
|
1978 |
+
{
|
1979 |
+
"epoch": 2.26,
|
1980 |
+
"eval_bertscore": 0.7339056134223938,
|
1981 |
+
"eval_loss": 1.7658218145370483,
|
1982 |
+
"eval_rouge1": 0.6494820372695989,
|
1983 |
+
"eval_rouge2": 0.34691658128805236,
|
1984 |
+
"eval_rougeL": 0.5193228965163086,
|
1985 |
+
"eval_rougeLsum": 0.6365347065687565,
|
1986 |
+
"eval_runtime": 15.3562,
|
1987 |
+
"eval_samples_per_second": 0.651,
|
1988 |
+
"eval_steps_per_second": 0.326,
|
1989 |
+
"step": 19800
|
1990 |
+
},
|
1991 |
+
{
|
1992 |
+
"epoch": 2.28,
|
1993 |
+
"grad_norm": 0.4452258050441742,
|
1994 |
+
"learning_rate": 8.609109294442704e-05,
|
1995 |
+
"loss": 1.4101,
|
1996 |
+
"step": 20000
|
1997 |
+
},
|
1998 |
+
{
|
1999 |
+
"epoch": 2.28,
|
2000 |
+
"eval_bertscore": 0.7296434640884399,
|
2001 |
+
"eval_loss": 1.7714240550994873,
|
2002 |
+
"eval_rouge1": 0.6565600824405751,
|
2003 |
+
"eval_rouge2": 0.3533618655201594,
|
2004 |
+
"eval_rougeL": 0.5263318202066467,
|
2005 |
+
"eval_rougeLsum": 0.6444964824298407,
|
2006 |
+
"eval_runtime": 14.8578,
|
2007 |
+
"eval_samples_per_second": 0.673,
|
2008 |
+
"eval_steps_per_second": 0.337,
|
2009 |
+
"step": 20000
|
2010 |
+
},
|
2011 |
+
{
|
2012 |
+
"epoch": 2.3,
|
2013 |
+
"grad_norm": 0.4030967652797699,
|
2014 |
+
"learning_rate": 8.495171903039281e-05,
|
2015 |
+
"loss": 1.4049,
|
2016 |
+
"step": 20200
|
2017 |
+
},
|
2018 |
+
{
|
2019 |
+
"epoch": 2.3,
|
2020 |
+
"eval_bertscore": 0.7307097315788269,
|
2021 |
+
"eval_loss": 1.774444580078125,
|
2022 |
+
"eval_rouge1": 0.6517204836155526,
|
2023 |
+
"eval_rouge2": 0.3521339653276223,
|
2024 |
+
"eval_rougeL": 0.5223211728244184,
|
2025 |
+
"eval_rougeLsum": 0.6398710531932736,
|
2026 |
+
"eval_runtime": 15.8543,
|
2027 |
+
"eval_samples_per_second": 0.631,
|
2028 |
+
"eval_steps_per_second": 0.315,
|
2029 |
+
"step": 20200
|
2030 |
+
},
|
2031 |
+
{
|
2032 |
+
"epoch": 2.32,
|
2033 |
+
"grad_norm": 0.33409813046455383,
|
2034 |
+
"learning_rate": 8.381234511635858e-05,
|
2035 |
+
"loss": 1.4243,
|
2036 |
+
"step": 20400
|
2037 |
+
},
|
2038 |
+
{
|
2039 |
+
"epoch": 2.32,
|
2040 |
+
"eval_bertscore": 0.7312101721763611,
|
2041 |
+
"eval_loss": 1.7654094696044922,
|
2042 |
+
"eval_rouge1": 0.6607249126293291,
|
2043 |
+
"eval_rouge2": 0.3545993249716188,
|
2044 |
+
"eval_rougeL": 0.5320161007986739,
|
2045 |
+
"eval_rougeLsum": 0.6503315335963733,
|
2046 |
+
"eval_runtime": 14.8739,
|
2047 |
+
"eval_samples_per_second": 0.672,
|
2048 |
+
"eval_steps_per_second": 0.336,
|
2049 |
+
"step": 20400
|
2050 |
+
},
|
2051 |
+
{
|
2052 |
+
"epoch": 2.35,
|
2053 |
+
"grad_norm": 0.4044789671897888,
|
2054 |
+
"learning_rate": 8.267297120232433e-05,
|
2055 |
+
"loss": 1.413,
|
2056 |
+
"step": 20600
|
2057 |
+
},
|
2058 |
+
{
|
2059 |
+
"epoch": 2.35,
|
2060 |
+
"eval_bertscore": 0.7342169880867004,
|
2061 |
+
"eval_loss": 1.769879937171936,
|
2062 |
+
"eval_rouge1": 0.6442777880355144,
|
2063 |
+
"eval_rouge2": 0.35006080708477183,
|
2064 |
+
"eval_rougeL": 0.5218799478770955,
|
2065 |
+
"eval_rougeLsum": 0.6332700294558067,
|
2066 |
+
"eval_runtime": 14.9089,
|
2067 |
+
"eval_samples_per_second": 0.671,
|
2068 |
+
"eval_steps_per_second": 0.335,
|
2069 |
+
"step": 20600
|
2070 |
+
},
|
2071 |
+
{
|
2072 |
+
"epoch": 2.37,
|
2073 |
+
"grad_norm": 0.39801183342933655,
|
2074 |
+
"learning_rate": 8.153359728829008e-05,
|
2075 |
+
"loss": 1.4177,
|
2076 |
+
"step": 20800
|
2077 |
+
},
|
2078 |
+
{
|
2079 |
+
"epoch": 2.37,
|
2080 |
+
"eval_bertscore": 0.7343758344650269,
|
2081 |
+
"eval_loss": 1.7737929821014404,
|
2082 |
+
"eval_rouge1": 0.6495678172205896,
|
2083 |
+
"eval_rouge2": 0.3505195734345703,
|
2084 |
+
"eval_rougeL": 0.5263025592812188,
|
2085 |
+
"eval_rougeLsum": 0.6390057749428748,
|
2086 |
+
"eval_runtime": 15.2148,
|
2087 |
+
"eval_samples_per_second": 0.657,
|
2088 |
+
"eval_steps_per_second": 0.329,
|
2089 |
+
"step": 20800
|
2090 |
+
},
|
2091 |
+
{
|
2092 |
+
"epoch": 2.39,
|
2093 |
+
"grad_norm": 0.36868759989738464,
|
2094 |
+
"learning_rate": 8.039422337425585e-05,
|
2095 |
+
"loss": 1.421,
|
2096 |
+
"step": 21000
|
2097 |
+
},
|
2098 |
+
{
|
2099 |
+
"epoch": 2.39,
|
2100 |
+
"eval_bertscore": 0.7333502173423767,
|
2101 |
+
"eval_loss": 1.7708820104599,
|
2102 |
+
"eval_rouge1": 0.656319412860679,
|
2103 |
+
"eval_rouge2": 0.3557406341135577,
|
2104 |
+
"eval_rougeL": 0.5293456110466322,
|
2105 |
+
"eval_rougeLsum": 0.6421819358163285,
|
2106 |
+
"eval_runtime": 14.9091,
|
2107 |
+
"eval_samples_per_second": 0.671,
|
2108 |
+
"eval_steps_per_second": 0.335,
|
2109 |
+
"step": 21000
|
2110 |
+
},
|
2111 |
+
{
|
2112 |
+
"epoch": 2.41,
|
2113 |
+
"grad_norm": 0.46111443638801575,
|
2114 |
+
"learning_rate": 7.925484946022162e-05,
|
2115 |
+
"loss": 1.4102,
|
2116 |
+
"step": 21200
|
2117 |
+
},
|
2118 |
+
{
|
2119 |
+
"epoch": 2.41,
|
2120 |
+
"eval_bertscore": 0.736262321472168,
|
2121 |
+
"eval_loss": 1.768972635269165,
|
2122 |
+
"eval_rouge1": 0.6582574071278393,
|
2123 |
+
"eval_rouge2": 0.3557625250443591,
|
2124 |
+
"eval_rougeL": 0.5322500342922363,
|
2125 |
+
"eval_rougeLsum": 0.646623827844921,
|
2126 |
+
"eval_runtime": 15.4088,
|
2127 |
+
"eval_samples_per_second": 0.649,
|
2128 |
+
"eval_steps_per_second": 0.324,
|
2129 |
+
"step": 21200
|
2130 |
+
},
|
2131 |
+
{
|
2132 |
+
"epoch": 2.44,
|
2133 |
+
"grad_norm": 0.41794517636299133,
|
2134 |
+
"learning_rate": 7.811547554618738e-05,
|
2135 |
+
"loss": 1.4231,
|
2136 |
+
"step": 21400
|
2137 |
+
},
|
2138 |
+
{
|
2139 |
+
"epoch": 2.44,
|
2140 |
+
"eval_bertscore": 0.7349900603294373,
|
2141 |
+
"eval_loss": 1.7673609256744385,
|
2142 |
+
"eval_rouge1": 0.6599777278993147,
|
2143 |
+
"eval_rouge2": 0.35744569380532043,
|
2144 |
+
"eval_rougeL": 0.5359850821835463,
|
2145 |
+
"eval_rougeLsum": 0.6469206354455653,
|
2146 |
+
"eval_runtime": 14.8837,
|
2147 |
+
"eval_samples_per_second": 0.672,
|
2148 |
+
"eval_steps_per_second": 0.336,
|
2149 |
+
"step": 21400
|
2150 |
+
},
|
2151 |
+
{
|
2152 |
+
"epoch": 2.46,
|
2153 |
+
"grad_norm": 0.3874039351940155,
|
2154 |
+
"learning_rate": 7.697610163215314e-05,
|
2155 |
+
"loss": 1.4158,
|
2156 |
+
"step": 21600
|
2157 |
+
},
|
2158 |
+
{
|
2159 |
+
"epoch": 2.46,
|
2160 |
+
"eval_bertscore": 0.7362676858901978,
|
2161 |
+
"eval_loss": 1.764347791671753,
|
2162 |
+
"eval_rouge1": 0.6576168971663054,
|
2163 |
+
"eval_rouge2": 0.36010190798950537,
|
2164 |
+
"eval_rougeL": 0.5365592740576962,
|
2165 |
+
"eval_rougeLsum": 0.6455601225938818,
|
2166 |
+
"eval_runtime": 15.4178,
|
2167 |
+
"eval_samples_per_second": 0.649,
|
2168 |
+
"eval_steps_per_second": 0.324,
|
2169 |
+
"step": 21600
|
2170 |
+
}
|
2171 |
+
],
|
2172 |
+
"logging_steps": 200,
|
2173 |
+
"max_steps": 35112,
|
2174 |
+
"num_input_tokens_seen": 0,
|
2175 |
+
"num_train_epochs": 4,
|
2176 |
+
"save_steps": 800,
|
2177 |
+
"total_flos": 2.1880581377217454e+18,
|
2178 |
+
"train_batch_size": 2,
|
2179 |
+
"trial_name": null,
|
2180 |
+
"trial_params": null
|
2181 |
+
}
|
checkpoint-21600/training_args.bin
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:d576743d4294cfbcae212243377437f73cabbcded2df9da7f73e6e39ecb38f56
|
3 |
+
size 5048
|
checkpoint-22400/README.md
ADDED
@@ -0,0 +1,202 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
---
|
2 |
+
library_name: peft
|
3 |
+
base_model: unsloth/gemma-2b-bnb-4bit
|
4 |
+
---
|
5 |
+
|
6 |
+
# Model Card for Model ID
|
7 |
+
|
8 |
+
<!-- Provide a quick summary of what the model is/does. -->
|
9 |
+
|
10 |
+
|
11 |
+
|
12 |
+
## Model Details
|
13 |
+
|
14 |
+
### Model Description
|
15 |
+
|
16 |
+
<!-- Provide a longer summary of what this model is. -->
|
17 |
+
|
18 |
+
|
19 |
+
|
20 |
+
- **Developed by:** [More Information Needed]
|
21 |
+
- **Funded by [optional]:** [More Information Needed]
|
22 |
+
- **Shared by [optional]:** [More Information Needed]
|
23 |
+
- **Model type:** [More Information Needed]
|
24 |
+
- **Language(s) (NLP):** [More Information Needed]
|
25 |
+
- **License:** [More Information Needed]
|
26 |
+
- **Finetuned from model [optional]:** [More Information Needed]
|
27 |
+
|
28 |
+
### Model Sources [optional]
|
29 |
+
|
30 |
+
<!-- Provide the basic links for the model. -->
|
31 |
+
|
32 |
+
- **Repository:** [More Information Needed]
|
33 |
+
- **Paper [optional]:** [More Information Needed]
|
34 |
+
- **Demo [optional]:** [More Information Needed]
|
35 |
+
|
36 |
+
## Uses
|
37 |
+
|
38 |
+
<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
|
39 |
+
|
40 |
+
### Direct Use
|
41 |
+
|
42 |
+
<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
|
43 |
+
|
44 |
+
[More Information Needed]
|
45 |
+
|
46 |
+
### Downstream Use [optional]
|
47 |
+
|
48 |
+
<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
|
49 |
+
|
50 |
+
[More Information Needed]
|
51 |
+
|
52 |
+
### Out-of-Scope Use
|
53 |
+
|
54 |
+
<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
|
55 |
+
|
56 |
+
[More Information Needed]
|
57 |
+
|
58 |
+
## Bias, Risks, and Limitations
|
59 |
+
|
60 |
+
<!-- This section is meant to convey both technical and sociotechnical limitations. -->
|
61 |
+
|
62 |
+
[More Information Needed]
|
63 |
+
|
64 |
+
### Recommendations
|
65 |
+
|
66 |
+
<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
|
67 |
+
|
68 |
+
Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
|
69 |
+
|
70 |
+
## How to Get Started with the Model
|
71 |
+
|
72 |
+
Use the code below to get started with the model.
|
73 |
+
|
74 |
+
[More Information Needed]
|
75 |
+
|
76 |
+
## Training Details
|
77 |
+
|
78 |
+
### Training Data
|
79 |
+
|
80 |
+
<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
|
81 |
+
|
82 |
+
[More Information Needed]
|
83 |
+
|
84 |
+
### Training Procedure
|
85 |
+
|
86 |
+
<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
|
87 |
+
|
88 |
+
#### Preprocessing [optional]
|
89 |
+
|
90 |
+
[More Information Needed]
|
91 |
+
|
92 |
+
|
93 |
+
#### Training Hyperparameters
|
94 |
+
|
95 |
+
- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
|
96 |
+
|
97 |
+
#### Speeds, Sizes, Times [optional]
|
98 |
+
|
99 |
+
<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
|
100 |
+
|
101 |
+
[More Information Needed]
|
102 |
+
|
103 |
+
## Evaluation
|
104 |
+
|
105 |
+
<!-- This section describes the evaluation protocols and provides the results. -->
|
106 |
+
|
107 |
+
### Testing Data, Factors & Metrics
|
108 |
+
|
109 |
+
#### Testing Data
|
110 |
+
|
111 |
+
<!-- This should link to a Dataset Card if possible. -->
|
112 |
+
|
113 |
+
[More Information Needed]
|
114 |
+
|
115 |
+
#### Factors
|
116 |
+
|
117 |
+
<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
|
118 |
+
|
119 |
+
[More Information Needed]
|
120 |
+
|
121 |
+
#### Metrics
|
122 |
+
|
123 |
+
<!-- These are the evaluation metrics being used, ideally with a description of why. -->
|
124 |
+
|
125 |
+
[More Information Needed]
|
126 |
+
|
127 |
+
### Results
|
128 |
+
|
129 |
+
[More Information Needed]
|
130 |
+
|
131 |
+
#### Summary
|
132 |
+
|
133 |
+
|
134 |
+
|
135 |
+
## Model Examination [optional]
|
136 |
+
|
137 |
+
<!-- Relevant interpretability work for the model goes here -->
|
138 |
+
|
139 |
+
[More Information Needed]
|
140 |
+
|
141 |
+
## Environmental Impact
|
142 |
+
|
143 |
+
<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
|
144 |
+
|
145 |
+
Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
|
146 |
+
|
147 |
+
- **Hardware Type:** [More Information Needed]
|
148 |
+
- **Hours used:** [More Information Needed]
|
149 |
+
- **Cloud Provider:** [More Information Needed]
|
150 |
+
- **Compute Region:** [More Information Needed]
|
151 |
+
- **Carbon Emitted:** [More Information Needed]
|
152 |
+
|
153 |
+
## Technical Specifications [optional]
|
154 |
+
|
155 |
+
### Model Architecture and Objective
|
156 |
+
|
157 |
+
[More Information Needed]
|
158 |
+
|
159 |
+
### Compute Infrastructure
|
160 |
+
|
161 |
+
[More Information Needed]
|
162 |
+
|
163 |
+
#### Hardware
|
164 |
+
|
165 |
+
[More Information Needed]
|
166 |
+
|
167 |
+
#### Software
|
168 |
+
|
169 |
+
[More Information Needed]
|
170 |
+
|
171 |
+
## Citation [optional]
|
172 |
+
|
173 |
+
<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
|
174 |
+
|
175 |
+
**BibTeX:**
|
176 |
+
|
177 |
+
[More Information Needed]
|
178 |
+
|
179 |
+
**APA:**
|
180 |
+
|
181 |
+
[More Information Needed]
|
182 |
+
|
183 |
+
## Glossary [optional]
|
184 |
+
|
185 |
+
<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
|
186 |
+
|
187 |
+
[More Information Needed]
|
188 |
+
|
189 |
+
## More Information [optional]
|
190 |
+
|
191 |
+
[More Information Needed]
|
192 |
+
|
193 |
+
## Model Card Authors [optional]
|
194 |
+
|
195 |
+
[More Information Needed]
|
196 |
+
|
197 |
+
## Model Card Contact
|
198 |
+
|
199 |
+
[More Information Needed]
|
200 |
+
### Framework versions
|
201 |
+
|
202 |
+
- PEFT 0.10.0
|
checkpoint-22400/adapter_config.json
ADDED
@@ -0,0 +1,34 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"alpha_pattern": {},
|
3 |
+
"auto_mapping": null,
|
4 |
+
"base_model_name_or_path": "unsloth/gemma-2b-bnb-4bit",
|
5 |
+
"bias": "none",
|
6 |
+
"fan_in_fan_out": false,
|
7 |
+
"inference_mode": true,
|
8 |
+
"init_lora_weights": true,
|
9 |
+
"layer_replication": null,
|
10 |
+
"layers_pattern": null,
|
11 |
+
"layers_to_transform": null,
|
12 |
+
"loftq_config": {},
|
13 |
+
"lora_alpha": 32,
|
14 |
+
"lora_dropout": 0,
|
15 |
+
"megatron_config": null,
|
16 |
+
"megatron_core": "megatron.core",
|
17 |
+
"modules_to_save": null,
|
18 |
+
"peft_type": "LORA",
|
19 |
+
"r": 64,
|
20 |
+
"rank_pattern": {},
|
21 |
+
"revision": "unsloth",
|
22 |
+
"target_modules": [
|
23 |
+
"v_proj",
|
24 |
+
"up_proj",
|
25 |
+
"gate_proj",
|
26 |
+
"o_proj",
|
27 |
+
"k_proj",
|
28 |
+
"down_proj",
|
29 |
+
"q_proj"
|
30 |
+
],
|
31 |
+
"task_type": "CAUSAL_LM",
|
32 |
+
"use_dora": false,
|
33 |
+
"use_rslora": false
|
34 |
+
}
|
checkpoint-22400/adapter_model.safetensors
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:cd6015a3bec35ad1dc604e30fb01a9c051ef9b3e36304464382ec41c027196bb
|
3 |
+
size 313820248
|
checkpoint-22400/rng_state.pth
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:51466a5ddd7196f688cac7aee47b26b078df13365d4372200c01a591729462c5
|
3 |
+
size 14244
|
checkpoint-22400/scheduler.pt
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:512b506129b7457fd633b55ef740589dea9c8726485f78ed74b0523c9944ff7f
|
3 |
+
size 1064
|