Hugging Face's logo

emendes3
/
llava_13b_neighborhood_synthetic

PEFT
Safetensors
llava_llama
Generated from Trainer
4-bit precision
bitsandbytes
Model card Files Community
emendes3 commited on
Commit
f00557f
•
1 Parent(s): a76de30

Model save

Browse files
Files changed (14) hide show
  1. README.md +3 -11
  2. adapter_config.json +6 -6
  3. adapter_model.safetensors +1 -1
  4. config.json +63 -0
  5. non_lora_trainables.bin +3 -0
  6. num_examples=200/llava-v1.5-13b_1.0/README.md +202 -0
  7. num_examples=200/llava-v1.5-13b_1.0/adapter_config.json +34 -0
  8. num_examples=200/llava-v1.5-13b_1.0/adapter_model.safetensors +3 -0
  9. num_examples=200/llava-v1.5-13b_1.0/special_tokens_map.json +24 -0
  10. num_examples=200/llava-v1.5-13b_1.0/tokenizer.model +3 -0
  11. num_examples=200/llava-v1.5-13b_1.0/tokenizer_config.json +42 -0
  12. num_examples=200/llava-v1.5-13b_1.0/training_args.bin +3 -0
  13. trainer_state.json +3430 -0
  14. training_args.bin +1 -1
README.md CHANGED
@@ -1,27 +1,19 @@
1
  ---
2
  library_name: peft
3
  tags:
4
- - liuhaotian/llava-v1.5-13b_20.0
5
  - generated_from_trainer
6
  base_model: liuhaotian/llava-v1.5-13b
7
  model-index:
8
- - name: liuhaotian/llava-v1.5-13b_20.0
9
  results: []
10
  ---
11
 
12
  <!-- This model card has been generated automatically according to the information the Trainer had access to. You
13
  should probably proofread and complete it, then remove this comment. -->
14
 
15
- # liuhaotian/llava-v1.5-13b_20.0
16
 
17
- This model is a fine-tuned version of [liuhaotian/llava-v1.5-13b_20.0](https://huggingface.co/liuhaotian/llava-v1.5-13b_20.0) on an unknown dataset.
18
- It achieves the following results on the evaluation set:
19
- - eval_loss: 0.0141
20
- - eval_runtime: 48.0124
21
- - eval_samples_per_second: 8.852
22
- - eval_steps_per_second: 0.562
23
- - epoch: 19.0
24
- - step: 513
25
 
26
  ## Model description
27
 
 
1
  ---
2
  library_name: peft
3
  tags:
 
4
  - generated_from_trainer
5
  base_model: liuhaotian/llava-v1.5-13b
6
  model-index:
7
+ - name: llava_13b_neighborhood_synthetic
8
  results: []
9
  ---
10
 
11
  <!-- This model card has been generated automatically according to the information the Trainer had access to. You
12
  should probably proofread and complete it, then remove this comment. -->
13
 
14
+ # llava_13b_neighborhood_synthetic
15
 
16
+ This model is a fine-tuned version of [liuhaotian/llava-v1.5-13b](https://huggingface.co/liuhaotian/llava-v1.5-13b) on an unknown dataset.
 
 
 
 
 
 
 
17
 
18
  ## Model description
19
 
adapter_config.json CHANGED
@@ -20,13 +20,13 @@
20
  "rank_pattern": {},
21
  "revision": null,
22
  "target_modules": [
23
- "k_proj",
24
- "v_proj",
25
- "up_proj",
26
- "gate_proj",
27
- "q_proj",
28
  "o_proj",
29
- "down_proj"
 
 
 
 
 
30
  ],
31
  "task_type": "CAUSAL_LM",
32
  "use_dora": false,
 
20
  "rank_pattern": {},
21
  "revision": null,
22
  "target_modules": [
 
 
 
 
 
23
  "o_proj",
24
+ "down_proj",
25
+ "q_proj",
26
+ "gate_proj",
27
+ "up_proj",
28
+ "k_proj",
29
+ "v_proj"
30
  ],
31
  "task_type": "CAUSAL_LM",
32
  "use_dora": false,
adapter_model.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:ace1a4dff71d7789df6b2906e8e8a24f3b24f7db2d8361fcecff4597df0855f6
3
  size 1001466944
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b5a8eb12dab76e380f3c42b18f6319f30c1adb8df79d1b454b1da4e3b4ea1333
3
  size 1001466944
config.json ADDED
@@ -0,0 +1,63 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "liuhaotian/llava-v1.5-13b",
3
+ "architectures": [
4
+ "LlavaLlamaForCausalLM"
5
+ ],
6
+ "attention_bias": false,
7
+ "attention_dropout": 0.0,
8
+ "bos_token_id": 1,
9
+ "eos_token_id": 2,
10
+ "freeze_mm_mlp_adapter": false,
11
+ "freeze_mm_vision_resampler": false,
12
+ "hidden_act": "silu",
13
+ "hidden_size": 5120,
14
+ "image_aspect_ratio": "pad",
15
+ "initializer_range": 0.02,
16
+ "intermediate_size": 13824,
17
+ "max_length": 4096,
18
+ "max_position_embeddings": 4096,
19
+ "mm_hidden_size": 1024,
20
+ "mm_patch_merge_type": "flat",
21
+ "mm_projector_lr": 2e-05,
22
+ "mm_projector_type": "mlp2x_gelu",
23
+ "mm_resampler_type": null,
24
+ "mm_use_im_patch_token": false,
25
+ "mm_use_im_start_end": false,
26
+ "mm_vision_select_feature": "patch",
27
+ "mm_vision_select_layer": -2,
28
+ "mm_vision_tower": "openai/clip-vit-large-patch14-336",
29
+ "model_type": "llava_llama",
30
+ "num_attention_heads": 40,
31
+ "num_hidden_layers": 40,
32
+ "num_key_value_heads": 40,
33
+ "pad_token_id": 0,
34
+ "pretraining_tp": 1,
35
+ "quantization_config": {
36
+ "bnb_4bit_compute_dtype": "bfloat16",
37
+ "bnb_4bit_quant_type": "nf4",
38
+ "bnb_4bit_use_double_quant": true,
39
+ "llm_int8_enable_fp32_cpu_offload": false,
40
+ "llm_int8_has_fp16_weight": false,
41
+ "llm_int8_skip_modules": [
42
+ "mm_projector"
43
+ ],
44
+ "llm_int8_threshold": 6.0,
45
+ "load_in_4bit": true,
46
+ "load_in_8bit": false,
47
+ "quant_method": "bitsandbytes"
48
+ },
49
+ "rms_norm_eps": 1e-05,
50
+ "rope_scaling": null,
51
+ "rope_theta": 10000.0,
52
+ "tie_word_embeddings": false,
53
+ "tokenizer_model_max_length": 2048,
54
+ "tokenizer_padding_side": "right",
55
+ "torch_dtype": "bfloat16",
56
+ "transformers_version": "4.37.2",
57
+ "tune_mm_mlp_adapter": false,
58
+ "tune_mm_vision_resampler": false,
59
+ "unfreeze_mm_vision_tower": false,
60
+ "use_cache": true,
61
+ "use_mm_proj": true,
62
+ "vocab_size": 32000
63
+ }
non_lora_trainables.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d68ddfe6ce0358ed97665d43ab246ef5884bd28da1ad884819ce894242a22ee8
3
+ size 62937264
num_examples=200/llava-v1.5-13b_1.0/README.md ADDED
@@ -0,0 +1,202 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ library_name: peft
3
+ base_model: liuhaotian/llava-v1.5-13b
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
num_examples=200/llava-v1.5-13b_1.0/adapter_config.json ADDED
@@ -0,0 +1,34 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "alpha_pattern": {},
3
+ "auto_mapping": null,
4
+ "base_model_name_or_path": "liuhaotian/llava-v1.5-13b",
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": 256,
14
+ "lora_dropout": 0.05,
15
+ "megatron_config": null,
16
+ "megatron_core": "megatron.core",
17
+ "modules_to_save": null,
18
+ "peft_type": "LORA",
19
+ "r": 128,
20
+ "rank_pattern": {},
21
+ "revision": null,
22
+ "target_modules": [
23
+ "o_proj",
24
+ "down_proj",
25
+ "q_proj",
26
+ "gate_proj",
27
+ "up_proj",
28
+ "k_proj",
29
+ "v_proj"
30
+ ],
31
+ "task_type": "CAUSAL_LM",
32
+ "use_dora": false,
33
+ "use_rslora": false
34
+ }
num_examples=200/llava-v1.5-13b_1.0/adapter_model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b5a8eb12dab76e380f3c42b18f6319f30c1adb8df79d1b454b1da4e3b4ea1333
3
+ size 1001466944
num_examples=200/llava-v1.5-13b_1.0/special_tokens_map.json ADDED
@@ -0,0 +1,24 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "bos_token": {
3
+ "content": "<s>",
4
+ "lstrip": false,
5
+ "normalized": false,
6
+ "rstrip": false,
7
+ "single_word": false
8
+ },
9
+ "eos_token": {
10
+ "content": "</s>",
11
+ "lstrip": false,
12
+ "normalized": false,
13
+ "rstrip": false,
14
+ "single_word": false
15
+ },
16
+ "pad_token": "<unk>",
17
+ "unk_token": {
18
+ "content": "<unk>",
19
+ "lstrip": false,
20
+ "normalized": false,
21
+ "rstrip": false,
22
+ "single_word": false
23
+ }
24
+ }
num_examples=200/llava-v1.5-13b_1.0/tokenizer.model ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:9e556afd44213b6bd1be2b850ebbbd98f5481437a8021afaf58ee7fb1818d347
3
+ size 499723
num_examples=200/llava-v1.5-13b_1.0/tokenizer_config.json ADDED
@@ -0,0 +1,42 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "add_bos_token": true,
3
+ "add_eos_token": false,
4
+ "added_tokens_decoder": {
5
+ "0": {
6
+ "content": "<unk>",
7
+ "lstrip": false,
8
+ "normalized": false,
9
+ "rstrip": false,
10
+ "single_word": false,
11
+ "special": true
12
+ },
13
+ "1": {
14
+ "content": "<s>",
15
+ "lstrip": false,
16
+ "normalized": false,
17
+ "rstrip": false,
18
+ "single_word": false,
19
+ "special": true
20
+ },
21
+ "2": {
22
+ "content": "</s>",
23
+ "lstrip": false,
24
+ "normalized": false,
25
+ "rstrip": false,
26
+ "single_word": false,
27
+ "special": true
28
+ }
29
+ },
30
+ "bos_token": "<s>",
31
+ "clean_up_tokenization_spaces": false,
32
+ "eos_token": "</s>",
33
+ "legacy": false,
34
+ "model_max_length": 2048,
35
+ "pad_token": "<unk>",
36
+ "padding_side": "right",
37
+ "sp_model_kwargs": {},
38
+ "spaces_between_special_tokens": false,
39
+ "tokenizer_class": "LlamaTokenizer",
40
+ "unk_token": "<unk>",
41
+ "use_default_system_prompt": false
42
+ }
num_examples=200/llava-v1.5-13b_1.0/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:044e9b2d025bc9ec7774e2eb37aa6fee7a0384b1e20dec78d3770c07373e38a2
3
+ size 6840
trainer_state.json ADDED
@@ -0,0 +1,3430 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": null,
3
+ "best_model_checkpoint": null,
4
+ "epoch": 20.0,
5
+ "eval_steps": 500,
6
+ "global_step": 540,
7
+ "is_hyper_param_search": false,
8
+ "is_local_process_zero": true,
9
+ "is_world_process_zero": true,
10
+ "log_history": [
11
+ {
12
+ "epoch": 0.04,
13
+ "learning_rate": 0.0,
14
+ "loss": 1.5775,
15
+ "step": 1
16
+ },
17
+ {
18
+ "epoch": 0.07,
19
+ "learning_rate": 4.8930108423645205e-05,
20
+ "loss": 1.1159,
21
+ "step": 2
22
+ },
23
+ {
24
+ "epoch": 0.11,
25
+ "learning_rate": 7.755238700769803e-05,
26
+ "loss": 1.0365,
27
+ "step": 3
28
+ },
29
+ {
30
+ "epoch": 0.15,
31
+ "learning_rate": 9.786021684729041e-05,
32
+ "loss": 1.086,
33
+ "step": 4
34
+ },
35
+ {
36
+ "epoch": 0.19,
37
+ "learning_rate": 0.00011361219343474658,
38
+ "loss": 1.1609,
39
+ "step": 5
40
+ },
41
+ {
42
+ "epoch": 0.22,
43
+ "learning_rate": 0.00012648249543134322,
44
+ "loss": 0.9921,
45
+ "step": 6
46
+ },
47
+ {
48
+ "epoch": 0.26,
49
+ "learning_rate": 0.0001373641807199326,
50
+ "loss": 1.0796,
51
+ "step": 7
52
+ },
53
+ {
54
+ "epoch": 0.3,
55
+ "learning_rate": 0.00014679032527093562,
56
+ "loss": 0.9986,
57
+ "step": 8
58
+ },
59
+ {
60
+ "epoch": 0.33,
61
+ "learning_rate": 0.00015510477401539606,
62
+ "loss": 0.952,
63
+ "step": 9
64
+ },
65
+ {
66
+ "epoch": 0.37,
67
+ "learning_rate": 0.0001625423018583918,
68
+ "loss": 0.8872,
69
+ "step": 10
70
+ },
71
+ {
72
+ "epoch": 0.41,
73
+ "learning_rate": 0.0001692703641841094,
74
+ "loss": 0.8786,
75
+ "step": 11
76
+ },
77
+ {
78
+ "epoch": 0.44,
79
+ "learning_rate": 0.00017541260385498843,
80
+ "loss": 0.7808,
81
+ "step": 12
82
+ },
83
+ {
84
+ "epoch": 0.48,
85
+ "learning_rate": 0.00018106291662380673,
86
+ "loss": 0.814,
87
+ "step": 13
88
+ },
89
+ {
90
+ "epoch": 0.52,
91
+ "learning_rate": 0.0001862942891435778,
92
+ "loss": 0.7429,
93
+ "step": 14
94
+ },
95
+ {
96
+ "epoch": 0.56,
97
+ "learning_rate": 0.0001911645804424446,
98
+ "loss": 0.7052,
99
+ "step": 15
100
+ },
101
+ {
102
+ "epoch": 0.59,
103
+ "learning_rate": 0.00019572043369458082,
104
+ "loss": 0.7136,
105
+ "step": 16
106
+ },
107
+ {
108
+ "epoch": 0.63,
109
+ "learning_rate": 0.0002,
110
+ "loss": 0.53,
111
+ "step": 17
112
+ },
113
+ {
114
+ "epoch": 0.67,
115
+ "learning_rate": 0.0002,
116
+ "loss": 0.5256,
117
+ "step": 18
118
+ },
119
+ {
120
+ "epoch": 0.7,
121
+ "learning_rate": 0.0002,
122
+ "loss": 0.607,
123
+ "step": 19
124
+ },
125
+ {
126
+ "epoch": 0.74,
127
+ "learning_rate": 0.0002,
128
+ "loss": 0.497,
129
+ "step": 20
130
+ },
131
+ {
132
+ "epoch": 0.78,
133
+ "learning_rate": 0.0002,
134
+ "loss": 0.3626,
135
+ "step": 21
136
+ },
137
+ {
138
+ "epoch": 0.81,
139
+ "learning_rate": 0.0002,
140
+ "loss": 0.5296,
141
+ "step": 22
142
+ },
143
+ {
144
+ "epoch": 0.85,
145
+ "learning_rate": 0.0002,
146
+ "loss": 0.3902,
147
+ "step": 23
148
+ },
149
+ {
150
+ "epoch": 0.89,
151
+ "learning_rate": 0.0002,
152
+ "loss": 0.2852,
153
+ "step": 24
154
+ },
155
+ {
156
+ "epoch": 0.93,
157
+ "learning_rate": 0.0002,
158
+ "loss": 0.3997,
159
+ "step": 25
160
+ },
161
+ {
162
+ "epoch": 0.96,
163
+ "learning_rate": 0.0002,
164
+ "loss": 0.2443,
165
+ "step": 26
166
+ },
167
+ {
168
+ "epoch": 1.0,
169
+ "learning_rate": 0.0002,
170
+ "loss": 0.2707,
171
+ "step": 27
172
+ },
173
+ {
174
+ "epoch": 1.0,
175
+ "eval_loss": 0.1848859339952469,
176
+ "eval_runtime": 47.8434,
177
+ "eval_samples_per_second": 8.883,
178
+ "eval_steps_per_second": 0.564,
179
+ "step": 27
180
+ },
181
+ {
182
+ "epoch": 1.04,
183
+ "learning_rate": 0.0002,
184
+ "loss": 0.1057,
185
+ "step": 28
186
+ },
187
+ {
188
+ "epoch": 1.07,
189
+ "learning_rate": 0.0002,
190
+ "loss": 0.0915,
191
+ "step": 29
192
+ },
193
+ {
194
+ "epoch": 1.11,
195
+ "learning_rate": 0.0002,
196
+ "loss": 0.1582,
197
+ "step": 30
198
+ },
199
+ {
200
+ "epoch": 1.15,
201
+ "learning_rate": 0.0002,
202
+ "loss": 0.1896,
203
+ "step": 31
204
+ },
205
+ {
206
+ "epoch": 1.19,
207
+ "learning_rate": 0.0002,
208
+ "loss": 0.1535,
209
+ "step": 32
210
+ },
211
+ {
212
+ "epoch": 1.22,
213
+ "learning_rate": 0.0002,
214
+ "loss": 0.1184,
215
+ "step": 33
216
+ },
217
+ {
218
+ "epoch": 1.26,
219
+ "learning_rate": 0.0002,
220
+ "loss": 0.156,
221
+ "step": 34
222
+ },
223
+ {
224
+ "epoch": 1.3,
225
+ "learning_rate": 0.0002,
226
+ "loss": 0.154,
227
+ "step": 35
228
+ },
229
+ {
230
+ "epoch": 1.33,
231
+ "learning_rate": 0.0002,
232
+ "loss": 0.0889,
233
+ "step": 36
234
+ },
235
+ {
236
+ "epoch": 1.37,
237
+ "learning_rate": 0.0002,
238
+ "loss": 0.0977,
239
+ "step": 37
240
+ },
241
+ {
242
+ "epoch": 1.41,
243
+ "learning_rate": 0.0002,
244
+ "loss": 0.1077,
245
+ "step": 38
246
+ },
247
+ {
248
+ "epoch": 1.44,
249
+ "learning_rate": 0.0002,
250
+ "loss": 0.1463,
251
+ "step": 39
252
+ },
253
+ {
254
+ "epoch": 1.48,
255
+ "learning_rate": 0.0002,
256
+ "loss": 0.1963,
257
+ "step": 40
258
+ },
259
+ {
260
+ "epoch": 1.52,
261
+ "learning_rate": 0.0002,
262
+ "loss": 0.1263,
263
+ "step": 41
264
+ },
265
+ {
266
+ "epoch": 1.56,
267
+ "learning_rate": 0.0002,
268
+ "loss": 0.0733,
269
+ "step": 42
270
+ },
271
+ {
272
+ "epoch": 1.59,
273
+ "learning_rate": 0.0002,
274
+ "loss": 0.1453,
275
+ "step": 43
276
+ },
277
+ {
278
+ "epoch": 1.63,
279
+ "learning_rate": 0.0002,
280
+ "loss": 0.0345,
281
+ "step": 44
282
+ },
283
+ {
284
+ "epoch": 1.67,
285
+ "learning_rate": 0.0002,
286
+ "loss": 0.125,
287
+ "step": 45
288
+ },
289
+ {
290
+ "epoch": 1.7,
291
+ "learning_rate": 0.0002,
292
+ "loss": 0.0515,
293
+ "step": 46
294
+ },
295
+ {
296
+ "epoch": 1.74,
297
+ "learning_rate": 0.0002,
298
+ "loss": 0.0361,
299
+ "step": 47
300
+ },
301
+ {
302
+ "epoch": 1.78,
303
+ "learning_rate": 0.0002,
304
+ "loss": 0.1165,
305
+ "step": 48
306
+ },
307
+ {
308
+ "epoch": 1.81,
309
+ "learning_rate": 0.0002,
310
+ "loss": 0.0376,
311
+ "step": 49
312
+ },
313
+ {
314
+ "epoch": 1.85,
315
+ "learning_rate": 0.0002,
316
+ "loss": 0.0424,
317
+ "step": 50
318
+ },
319
+ {
320
+ "epoch": 1.89,
321
+ "learning_rate": 0.0002,
322
+ "loss": 0.0507,
323
+ "step": 51
324
+ },
325
+ {
326
+ "epoch": 1.93,
327
+ "learning_rate": 0.0002,
328
+ "loss": 0.0993,
329
+ "step": 52
330
+ },
331
+ {
332
+ "epoch": 1.96,
333
+ "learning_rate": 0.0002,
334
+ "loss": 0.0838,
335
+ "step": 53
336
+ },
337
+ {
338
+ "epoch": 2.0,
339
+ "learning_rate": 0.0002,
340
+ "loss": 0.0687,
341
+ "step": 54
342
+ },
343
+ {
344
+ "epoch": 2.0,
345
+ "eval_loss": 0.041033171117305756,
346
+ "eval_runtime": 47.992,
347
+ "eval_samples_per_second": 8.856,
348
+ "eval_steps_per_second": 0.563,
349
+ "step": 54
350
+ },
351
+ {
352
+ "epoch": 2.04,
353
+ "learning_rate": 0.0002,
354
+ "loss": 0.0265,
355
+ "step": 55
356
+ },
357
+ {
358
+ "epoch": 2.07,
359
+ "learning_rate": 0.0002,
360
+ "loss": 0.0478,
361
+ "step": 56
362
+ },
363
+ {
364
+ "epoch": 2.11,
365
+ "learning_rate": 0.0002,
366
+ "loss": 0.0317,
367
+ "step": 57
368
+ },
369
+ {
370
+ "epoch": 2.15,
371
+ "learning_rate": 0.0002,
372
+ "loss": 0.0233,
373
+ "step": 58
374
+ },
375
+ {
376
+ "epoch": 2.19,
377
+ "learning_rate": 0.0002,
378
+ "loss": 0.0298,
379
+ "step": 59
380
+ },
381
+ {
382
+ "epoch": 2.22,
383
+ "learning_rate": 0.0002,
384
+ "loss": 0.0169,
385
+ "step": 60
386
+ },
387
+ {
388
+ "epoch": 2.26,
389
+ "learning_rate": 0.0002,
390
+ "loss": 0.0293,
391
+ "step": 61
392
+ },
393
+ {
394
+ "epoch": 2.3,
395
+ "learning_rate": 0.0002,
396
+ "loss": 0.0231,
397
+ "step": 62
398
+ },
399
+ {
400
+ "epoch": 2.33,
401
+ "learning_rate": 0.0002,
402
+ "loss": 0.022,
403
+ "step": 63
404
+ },
405
+ {
406
+ "epoch": 2.37,
407
+ "learning_rate": 0.0002,
408
+ "loss": 0.0369,
409
+ "step": 64
410
+ },
411
+ {
412
+ "epoch": 2.41,
413
+ "learning_rate": 0.0002,
414
+ "loss": 0.0055,
415
+ "step": 65
416
+ },
417
+ {
418
+ "epoch": 2.44,
419
+ "learning_rate": 0.0002,
420
+ "loss": 0.012,
421
+ "step": 66
422
+ },
423
+ {
424
+ "epoch": 2.48,
425
+ "learning_rate": 0.0002,
426
+ "loss": 0.0197,
427
+ "step": 67
428
+ },
429
+ {
430
+ "epoch": 2.52,
431
+ "learning_rate": 0.0002,
432
+ "loss": 0.0275,
433
+ "step": 68
434
+ },
435
+ {
436
+ "epoch": 2.56,
437
+ "learning_rate": 0.0002,
438
+ "loss": 0.0411,
439
+ "step": 69
440
+ },
441
+ {
442
+ "epoch": 2.59,
443
+ "learning_rate": 0.0002,
444
+ "loss": 0.028,
445
+ "step": 70
446
+ },
447
+ {
448
+ "epoch": 2.63,
449
+ "learning_rate": 0.0002,
450
+ "loss": 0.0084,
451
+ "step": 71
452
+ },
453
+ {
454
+ "epoch": 2.67,
455
+ "learning_rate": 0.0002,
456
+ "loss": 0.0315,
457
+ "step": 72
458
+ },
459
+ {
460
+ "epoch": 2.7,
461
+ "learning_rate": 0.0002,
462
+ "loss": 0.0127,
463
+ "step": 73
464
+ },
465
+ {
466
+ "epoch": 2.74,
467
+ "learning_rate": 0.0002,
468
+ "loss": 0.0132,
469
+ "step": 74
470
+ },
471
+ {
472
+ "epoch": 2.78,
473
+ "learning_rate": 0.0002,
474
+ "loss": 0.0121,
475
+ "step": 75
476
+ },
477
+ {
478
+ "epoch": 2.81,
479
+ "learning_rate": 0.0002,
480
+ "loss": 0.0246,
481
+ "step": 76
482
+ },
483
+ {
484
+ "epoch": 2.85,
485
+ "learning_rate": 0.0002,
486
+ "loss": 0.0243,
487
+ "step": 77
488
+ },
489
+ {
490
+ "epoch": 2.89,
491
+ "learning_rate": 0.0002,
492
+ "loss": 0.0189,
493
+ "step": 78
494
+ },
495
+ {
496
+ "epoch": 2.93,
497
+ "learning_rate": 0.0002,
498
+ "loss": 0.0093,
499
+ "step": 79
500
+ },
501
+ {
502
+ "epoch": 2.96,
503
+ "learning_rate": 0.0002,
504
+ "loss": 0.0196,
505
+ "step": 80
506
+ },
507
+ {
508
+ "epoch": 3.0,
509
+ "learning_rate": 0.0002,
510
+ "loss": 0.0128,
511
+ "step": 81
512
+ },
513
+ {
514
+ "epoch": 3.0,
515
+ "eval_loss": 0.01152033545076847,
516
+ "eval_runtime": 47.9951,
517
+ "eval_samples_per_second": 8.855,
518
+ "eval_steps_per_second": 0.563,
519
+ "step": 81
520
+ },
521
+ {
522
+ "epoch": 3.04,
523
+ "learning_rate": 0.0002,
524
+ "loss": 0.0071,
525
+ "step": 82
526
+ },
527
+ {
528
+ "epoch": 3.07,
529
+ "learning_rate": 0.0002,
530
+ "loss": 0.01,
531
+ "step": 83
532
+ },
533
+ {
534
+ "epoch": 3.11,
535
+ "learning_rate": 0.0002,
536
+ "loss": 0.0067,
537
+ "step": 84
538
+ },
539
+ {
540
+ "epoch": 3.15,
541
+ "learning_rate": 0.0002,
542
+ "loss": 0.0128,
543
+ "step": 85
544
+ },
545
+ {
546
+ "epoch": 3.19,
547
+ "learning_rate": 0.0002,
548
+ "loss": 0.0058,
549
+ "step": 86
550
+ },
551
+ {
552
+ "epoch": 3.22,
553
+ "learning_rate": 0.0002,
554
+ "loss": 0.0096,
555
+ "step": 87
556
+ },
557
+ {
558
+ "epoch": 3.26,
559
+ "learning_rate": 0.0002,
560
+ "loss": 0.0039,
561
+ "step": 88
562
+ },
563
+ {
564
+ "epoch": 3.3,
565
+ "learning_rate": 0.0002,
566
+ "loss": 0.0083,
567
+ "step": 89
568
+ },
569
+ {
570
+ "epoch": 3.33,
571
+ "learning_rate": 0.0002,
572
+ "loss": 0.0096,
573
+ "step": 90
574
+ },
575
+ {
576
+ "epoch": 3.37,
577
+ "learning_rate": 0.0002,
578
+ "loss": 0.007,
579
+ "step": 91
580
+ },
581
+ {
582
+ "epoch": 3.41,
583
+ "learning_rate": 0.0002,
584
+ "loss": 0.0102,
585
+ "step": 92
586
+ },
587
+ {
588
+ "epoch": 3.44,
589
+ "learning_rate": 0.0002,
590
+ "loss": 0.0168,
591
+ "step": 93
592
+ },
593
+ {
594
+ "epoch": 3.48,
595
+ "learning_rate": 0.0002,
596
+ "loss": 0.0137,
597
+ "step": 94
598
+ },
599
+ {
600
+ "epoch": 3.52,
601
+ "learning_rate": 0.0002,
602
+ "loss": 0.0105,
603
+ "step": 95
604
+ },
605
+ {
606
+ "epoch": 3.56,
607
+ "learning_rate": 0.0002,
608
+ "loss": 0.0068,
609
+ "step": 96
610
+ },
611
+ {
612
+ "epoch": 3.59,
613
+ "learning_rate": 0.0002,
614
+ "loss": 0.0079,
615
+ "step": 97
616
+ },
617
+ {
618
+ "epoch": 3.63,
619
+ "learning_rate": 0.0002,
620
+ "loss": 0.0075,
621
+ "step": 98
622
+ },
623
+ {
624
+ "epoch": 3.67,
625
+ "learning_rate": 0.0002,
626
+ "loss": 0.003,
627
+ "step": 99
628
+ },
629
+ {
630
+ "epoch": 3.7,
631
+ "learning_rate": 0.0002,
632
+ "loss": 0.0103,
633
+ "step": 100
634
+ },
635
+ {
636
+ "epoch": 3.74,
637
+ "learning_rate": 0.0002,
638
+ "loss": 0.0078,
639
+ "step": 101
640
+ },
641
+ {
642
+ "epoch": 3.78,
643
+ "learning_rate": 0.0002,
644
+ "loss": 0.0135,
645
+ "step": 102
646
+ },
647
+ {
648
+ "epoch": 3.81,
649
+ "learning_rate": 0.0002,
650
+ "loss": 0.0139,
651
+ "step": 103
652
+ },
653
+ {
654
+ "epoch": 3.85,
655
+ "learning_rate": 0.0002,
656
+ "loss": 0.0097,
657
+ "step": 104
658
+ },
659
+ {
660
+ "epoch": 3.89,
661
+ "learning_rate": 0.0002,
662
+ "loss": 0.0096,
663
+ "step": 105
664
+ },
665
+ {
666
+ "epoch": 3.93,
667
+ "learning_rate": 0.0002,
668
+ "loss": 0.0072,
669
+ "step": 106
670
+ },
671
+ {
672
+ "epoch": 3.96,
673
+ "learning_rate": 0.0002,
674
+ "loss": 0.0085,
675
+ "step": 107
676
+ },
677
+ {
678
+ "epoch": 4.0,
679
+ "learning_rate": 0.0002,
680
+ "loss": 0.0137,
681
+ "step": 108
682
+ },
683
+ {
684
+ "epoch": 4.0,
685
+ "eval_loss": 0.010391350835561752,
686
+ "eval_runtime": 47.9023,
687
+ "eval_samples_per_second": 8.872,
688
+ "eval_steps_per_second": 0.564,
689
+ "step": 108
690
+ },
691
+ {
692
+ "epoch": 4.04,
693
+ "learning_rate": 0.0002,
694
+ "loss": 0.0101,
695
+ "step": 109
696
+ },
697
+ {
698
+ "epoch": 4.07,
699
+ "learning_rate": 0.0002,
700
+ "loss": 0.005,
701
+ "step": 110
702
+ },
703
+ {
704
+ "epoch": 4.11,
705
+ "learning_rate": 0.0002,
706
+ "loss": 0.0045,
707
+ "step": 111
708
+ },
709
+ {
710
+ "epoch": 4.15,
711
+ "learning_rate": 0.0002,
712
+ "loss": 0.0092,
713
+ "step": 112
714
+ },
715
+ {
716
+ "epoch": 4.19,
717
+ "learning_rate": 0.0002,
718
+ "loss": 0.0163,
719
+ "step": 113
720
+ },
721
+ {
722
+ "epoch": 4.22,
723
+ "learning_rate": 0.0002,
724
+ "loss": 0.0112,
725
+ "step": 114
726
+ },
727
+ {
728
+ "epoch": 4.26,
729
+ "learning_rate": 0.0002,
730
+ "loss": 0.0044,
731
+ "step": 115
732
+ },
733
+ {
734
+ "epoch": 4.3,
735
+ "learning_rate": 0.0002,
736
+ "loss": 0.002,
737
+ "step": 116
738
+ },
739
+ {
740
+ "epoch": 4.33,
741
+ "learning_rate": 0.0002,
742
+ "loss": 0.0059,
743
+ "step": 117
744
+ },
745
+ {
746
+ "epoch": 4.37,
747
+ "learning_rate": 0.0002,
748
+ "loss": 0.002,
749
+ "step": 118
750
+ },
751
+ {
752
+ "epoch": 4.41,
753
+ "learning_rate": 0.0002,
754
+ "loss": 0.0025,
755
+ "step": 119
756
+ },
757
+ {
758
+ "epoch": 4.44,
759
+ "learning_rate": 0.0002,
760
+ "loss": 0.005,
761
+ "step": 120
762
+ },
763
+ {
764
+ "epoch": 4.48,
765
+ "learning_rate": 0.0002,
766
+ "loss": 0.0048,
767
+ "step": 121
768
+ },
769
+ {
770
+ "epoch": 4.52,
771
+ "learning_rate": 0.0002,
772
+ "loss": 0.0097,
773
+ "step": 122
774
+ },
775
+ {
776
+ "epoch": 4.56,
777
+ "learning_rate": 0.0002,
778
+ "loss": 0.0034,
779
+ "step": 123
780
+ },
781
+ {
782
+ "epoch": 4.59,
783
+ "learning_rate": 0.0002,
784
+ "loss": 0.024,
785
+ "step": 124
786
+ },
787
+ {
788
+ "epoch": 4.63,
789
+ "learning_rate": 0.0002,
790
+ "loss": 0.0018,
791
+ "step": 125
792
+ },
793
+ {
794
+ "epoch": 4.67,
795
+ "learning_rate": 0.0002,
796
+ "loss": 0.0013,
797
+ "step": 126
798
+ },
799
+ {
800
+ "epoch": 4.7,
801
+ "learning_rate": 0.0002,
802
+ "loss": 0.0046,
803
+ "step": 127
804
+ },
805
+ {
806
+ "epoch": 4.74,
807
+ "learning_rate": 0.0002,
808
+ "loss": 0.0112,
809
+ "step": 128
810
+ },
811
+ {
812
+ "epoch": 4.78,
813
+ "learning_rate": 0.0002,
814
+ "loss": 0.004,
815
+ "step": 129
816
+ },
817
+ {
818
+ "epoch": 4.81,
819
+ "learning_rate": 0.0002,
820
+ "loss": 0.0088,
821
+ "step": 130
822
+ },
823
+ {
824
+ "epoch": 4.85,
825
+ "learning_rate": 0.0002,
826
+ "loss": 0.0048,
827
+ "step": 131
828
+ },
829
+ {
830
+ "epoch": 4.89,
831
+ "learning_rate": 0.0002,
832
+ "loss": 0.0048,
833
+ "step": 132
834
+ },
835
+ {
836
+ "epoch": 4.93,
837
+ "learning_rate": 0.0002,
838
+ "loss": 0.0073,
839
+ "step": 133
840
+ },
841
+ {
842
+ "epoch": 4.96,
843
+ "learning_rate": 0.0002,
844
+ "loss": 0.0016,
845
+ "step": 134
846
+ },
847
+ {
848
+ "epoch": 5.0,
849
+ "learning_rate": 0.0002,
850
+ "loss": 0.0034,
851
+ "step": 135
852
+ },
853
+ {
854
+ "epoch": 5.0,
855
+ "eval_loss": 0.0039200917817652225,
856
+ "eval_runtime": 48.0639,
857
+ "eval_samples_per_second": 8.842,
858
+ "eval_steps_per_second": 0.562,
859
+ "step": 135
860
+ },
861
+ {
862
+ "epoch": 5.04,
863
+ "learning_rate": 0.0002,
864
+ "loss": 0.0013,
865
+ "step": 136
866
+ },
867
+ {
868
+ "epoch": 5.07,
869
+ "learning_rate": 0.0002,
870
+ "loss": 0.0011,
871
+ "step": 137
872
+ },
873
+ {
874
+ "epoch": 5.11,
875
+ "learning_rate": 0.0002,
876
+ "loss": 0.002,
877
+ "step": 138
878
+ },
879
+ {
880
+ "epoch": 5.15,
881
+ "learning_rate": 0.0002,
882
+ "loss": 0.0013,
883
+ "step": 139
884
+ },
885
+ {
886
+ "epoch": 5.19,
887
+ "learning_rate": 0.0002,
888
+ "loss": 0.0046,
889
+ "step": 140
890
+ },
891
+ {
892
+ "epoch": 5.22,
893
+ "learning_rate": 0.0002,
894
+ "loss": 0.0075,
895
+ "step": 141
896
+ },
897
+ {
898
+ "epoch": 5.26,
899
+ "learning_rate": 0.0002,
900
+ "loss": 0.0032,
901
+ "step": 142
902
+ },
903
+ {
904
+ "epoch": 5.3,
905
+ "learning_rate": 0.0002,
906
+ "loss": 0.0005,
907
+ "step": 143
908
+ },
909
+ {
910
+ "epoch": 5.33,
911
+ "learning_rate": 0.0002,
912
+ "loss": 0.0014,
913
+ "step": 144
914
+ },
915
+ {
916
+ "epoch": 5.37,
917
+ "learning_rate": 0.0002,
918
+ "loss": 0.0021,
919
+ "step": 145
920
+ },
921
+ {
922
+ "epoch": 5.41,
923
+ "learning_rate": 0.0002,
924
+ "loss": 0.0038,
925
+ "step": 146
926
+ },
927
+ {
928
+ "epoch": 5.44,
929
+ "learning_rate": 0.0002,
930
+ "loss": 0.0032,
931
+ "step": 147
932
+ },
933
+ {
934
+ "epoch": 5.48,
935
+ "learning_rate": 0.0002,
936
+ "loss": 0.0007,
937
+ "step": 148
938
+ },
939
+ {
940
+ "epoch": 5.52,
941
+ "learning_rate": 0.0002,
942
+ "loss": 0.0031,
943
+ "step": 149
944
+ },
945
+ {
946
+ "epoch": 5.56,
947
+ "learning_rate": 0.0002,
948
+ "loss": 0.0035,
949
+ "step": 150
950
+ },
951
+ {
952
+ "epoch": 5.59,
953
+ "learning_rate": 0.0002,
954
+ "loss": 0.0044,
955
+ "step": 151
956
+ },
957
+ {
958
+ "epoch": 5.63,
959
+ "learning_rate": 0.0002,
960
+ "loss": 0.0018,
961
+ "step": 152
962
+ },
963
+ {
964
+ "epoch": 5.67,
965
+ "learning_rate": 0.0002,
966
+ "loss": 0.0031,
967
+ "step": 153
968
+ },
969
+ {
970
+ "epoch": 5.7,
971
+ "learning_rate": 0.0002,
972
+ "loss": 0.009,
973
+ "step": 154
974
+ },
975
+ {
976
+ "epoch": 5.74,
977
+ "learning_rate": 0.0002,
978
+ "loss": 0.0057,
979
+ "step": 155
980
+ },
981
+ {
982
+ "epoch": 5.78,
983
+ "learning_rate": 0.0002,
984
+ "loss": 0.0024,
985
+ "step": 156
986
+ },
987
+ {
988
+ "epoch": 5.81,
989
+ "learning_rate": 0.0002,
990
+ "loss": 0.0013,
991
+ "step": 157
992
+ },
993
+ {
994
+ "epoch": 5.85,
995
+ "learning_rate": 0.0002,
996
+ "loss": 0.0121,
997
+ "step": 158
998
+ },
999
+ {
1000
+ "epoch": 5.89,
1001
+ "learning_rate": 0.0002,
1002
+ "loss": 0.0064,
1003
+ "step": 159
1004
+ },
1005
+ {
1006
+ "epoch": 5.93,
1007
+ "learning_rate": 0.0002,
1008
+ "loss": 0.0074,
1009
+ "step": 160
1010
+ },
1011
+ {
1012
+ "epoch": 5.96,
1013
+ "learning_rate": 0.0002,
1014
+ "loss": 0.0055,
1015
+ "step": 161
1016
+ },
1017
+ {
1018
+ "epoch": 6.0,
1019
+ "learning_rate": 0.0002,
1020
+ "loss": 0.0017,
1021
+ "step": 162
1022
+ },
1023
+ {
1024
+ "epoch": 6.0,
1025
+ "eval_loss": 0.003504824358969927,
1026
+ "eval_runtime": 48.0525,
1027
+ "eval_samples_per_second": 8.844,
1028
+ "eval_steps_per_second": 0.562,
1029
+ "step": 162
1030
+ },
1031
+ {
1032
+ "epoch": 6.04,
1033
+ "learning_rate": 0.0002,
1034
+ "loss": 0.005,
1035
+ "step": 163
1036
+ },
1037
+ {
1038
+ "epoch": 6.07,
1039
+ "learning_rate": 0.0002,
1040
+ "loss": 0.0021,
1041
+ "step": 164
1042
+ },
1043
+ {
1044
+ "epoch": 6.11,
1045
+ "learning_rate": 0.0002,
1046
+ "loss": 0.0045,
1047
+ "step": 165
1048
+ },
1049
+ {
1050
+ "epoch": 6.15,
1051
+ "learning_rate": 0.0002,
1052
+ "loss": 0.0005,
1053
+ "step": 166
1054
+ },
1055
+ {
1056
+ "epoch": 6.19,
1057
+ "learning_rate": 0.0002,
1058
+ "loss": 0.001,
1059
+ "step": 167
1060
+ },
1061
+ {
1062
+ "epoch": 6.22,
1063
+ "learning_rate": 0.0002,
1064
+ "loss": 0.0013,
1065
+ "step": 168
1066
+ },
1067
+ {
1068
+ "epoch": 6.26,
1069
+ "learning_rate": 0.0002,
1070
+ "loss": 0.001,
1071
+ "step": 169
1072
+ },
1073
+ {
1074
+ "epoch": 6.3,
1075
+ "learning_rate": 0.0002,
1076
+ "loss": 0.0053,
1077
+ "step": 170
1078
+ },
1079
+ {
1080
+ "epoch": 6.33,
1081
+ "learning_rate": 0.0002,
1082
+ "loss": 0.0012,
1083
+ "step": 171
1084
+ },
1085
+ {
1086
+ "epoch": 6.37,
1087
+ "learning_rate": 0.0002,
1088
+ "loss": 0.0017,
1089
+ "step": 172
1090
+ },
1091
+ {
1092
+ "epoch": 6.41,
1093
+ "learning_rate": 0.0002,
1094
+ "loss": 0.0036,
1095
+ "step": 173
1096
+ },
1097
+ {
1098
+ "epoch": 6.44,
1099
+ "learning_rate": 0.0002,
1100
+ "loss": 0.0019,
1101
+ "step": 174
1102
+ },
1103
+ {
1104
+ "epoch": 6.48,
1105
+ "learning_rate": 0.0002,
1106
+ "loss": 0.0005,
1107
+ "step": 175
1108
+ },
1109
+ {
1110
+ "epoch": 6.52,
1111
+ "learning_rate": 0.0002,
1112
+ "loss": 0.0018,
1113
+ "step": 176
1114
+ },
1115
+ {
1116
+ "epoch": 6.56,
1117
+ "learning_rate": 0.0002,
1118
+ "loss": 0.0023,
1119
+ "step": 177
1120
+ },
1121
+ {
1122
+ "epoch": 6.59,
1123
+ "learning_rate": 0.0002,
1124
+ "loss": 0.0098,
1125
+ "step": 178
1126
+ },
1127
+ {
1128
+ "epoch": 6.63,
1129
+ "learning_rate": 0.0002,
1130
+ "loss": 0.0069,
1131
+ "step": 179
1132
+ },
1133
+ {
1134
+ "epoch": 6.67,
1135
+ "learning_rate": 0.0002,
1136
+ "loss": 0.002,
1137
+ "step": 180
1138
+ },
1139
+ {
1140
+ "epoch": 6.7,
1141
+ "learning_rate": 0.0002,
1142
+ "loss": 0.0022,
1143
+ "step": 181
1144
+ },
1145
+ {
1146
+ "epoch": 6.74,
1147
+ "learning_rate": 0.0002,
1148
+ "loss": 0.0023,
1149
+ "step": 182
1150
+ },
1151
+ {
1152
+ "epoch": 6.78,
1153
+ "learning_rate": 0.0002,
1154
+ "loss": 0.0054,
1155
+ "step": 183
1156
+ },
1157
+ {
1158
+ "epoch": 6.81,
1159
+ "learning_rate": 0.0002,
1160
+ "loss": 0.0029,
1161
+ "step": 184
1162
+ },
1163
+ {
1164
+ "epoch": 6.85,
1165
+ "learning_rate": 0.0002,
1166
+ "loss": 0.0031,
1167
+ "step": 185
1168
+ },
1169
+ {
1170
+ "epoch": 6.89,
1171
+ "learning_rate": 0.0002,
1172
+ "loss": 0.0022,
1173
+ "step": 186
1174
+ },
1175
+ {
1176
+ "epoch": 6.93,
1177
+ "learning_rate": 0.0002,
1178
+ "loss": 0.0045,
1179
+ "step": 187
1180
+ },
1181
+ {
1182
+ "epoch": 6.96,
1183
+ "learning_rate": 0.0002,
1184
+ "loss": 0.0022,
1185
+ "step": 188
1186
+ },
1187
+ {
1188
+ "epoch": 7.0,
1189
+ "learning_rate": 0.0002,
1190
+ "loss": 0.0018,
1191
+ "step": 189
1192
+ },
1193
+ {
1194
+ "epoch": 7.0,
1195
+ "eval_loss": 0.003912605345249176,
1196
+ "eval_runtime": 48.0378,
1197
+ "eval_samples_per_second": 8.847,
1198
+ "eval_steps_per_second": 0.562,
1199
+ "step": 189
1200
+ },
1201
+ {
1202
+ "epoch": 7.04,
1203
+ "learning_rate": 0.0002,
1204
+ "loss": 0.0028,
1205
+ "step": 190
1206
+ },
1207
+ {
1208
+ "epoch": 7.07,
1209
+ "learning_rate": 0.0002,
1210
+ "loss": 0.0042,
1211
+ "step": 191
1212
+ },
1213
+ {
1214
+ "epoch": 7.11,
1215
+ "learning_rate": 0.0002,
1216
+ "loss": 0.0007,
1217
+ "step": 192
1218
+ },
1219
+ {
1220
+ "epoch": 7.15,
1221
+ "learning_rate": 0.0002,
1222
+ "loss": 0.0047,
1223
+ "step": 193
1224
+ },
1225
+ {
1226
+ "epoch": 7.19,
1227
+ "learning_rate": 0.0002,
1228
+ "loss": 0.0047,
1229
+ "step": 194
1230
+ },
1231
+ {
1232
+ "epoch": 7.22,
1233
+ "learning_rate": 0.0002,
1234
+ "loss": 0.0034,
1235
+ "step": 195
1236
+ },
1237
+ {
1238
+ "epoch": 7.26,
1239
+ "learning_rate": 0.0002,
1240
+ "loss": 0.0032,
1241
+ "step": 196
1242
+ },
1243
+ {
1244
+ "epoch": 7.3,
1245
+ "learning_rate": 0.0002,
1246
+ "loss": 0.0031,
1247
+ "step": 197
1248
+ },
1249
+ {
1250
+ "epoch": 7.33,
1251
+ "learning_rate": 0.0002,
1252
+ "loss": 0.002,
1253
+ "step": 198
1254
+ },
1255
+ {
1256
+ "epoch": 7.37,
1257
+ "learning_rate": 0.0002,
1258
+ "loss": 0.0011,
1259
+ "step": 199
1260
+ },
1261
+ {
1262
+ "epoch": 7.41,
1263
+ "learning_rate": 0.0002,
1264
+ "loss": 0.0021,
1265
+ "step": 200
1266
+ },
1267
+ {
1268
+ "epoch": 7.44,
1269
+ "learning_rate": 0.0002,
1270
+ "loss": 0.0072,
1271
+ "step": 201
1272
+ },
1273
+ {
1274
+ "epoch": 7.48,
1275
+ "learning_rate": 0.0002,
1276
+ "loss": 0.0022,
1277
+ "step": 202
1278
+ },
1279
+ {
1280
+ "epoch": 7.52,
1281
+ "learning_rate": 0.0002,
1282
+ "loss": 0.0009,
1283
+ "step": 203
1284
+ },
1285
+ {
1286
+ "epoch": 7.56,
1287
+ "learning_rate": 0.0002,
1288
+ "loss": 0.0015,
1289
+ "step": 204
1290
+ },
1291
+ {
1292
+ "epoch": 7.59,
1293
+ "learning_rate": 0.0002,
1294
+ "loss": 0.0059,
1295
+ "step": 205
1296
+ },
1297
+ {
1298
+ "epoch": 7.63,
1299
+ "learning_rate": 0.0002,
1300
+ "loss": 0.001,
1301
+ "step": 206
1302
+ },
1303
+ {
1304
+ "epoch": 7.67,
1305
+ "learning_rate": 0.0002,
1306
+ "loss": 0.0015,
1307
+ "step": 207
1308
+ },
1309
+ {
1310
+ "epoch": 7.7,
1311
+ "learning_rate": 0.0002,
1312
+ "loss": 0.0015,
1313
+ "step": 208
1314
+ },
1315
+ {
1316
+ "epoch": 7.74,
1317
+ "learning_rate": 0.0002,
1318
+ "loss": 0.0027,
1319
+ "step": 209
1320
+ },
1321
+ {
1322
+ "epoch": 7.78,
1323
+ "learning_rate": 0.0002,
1324
+ "loss": 0.0047,
1325
+ "step": 210
1326
+ },
1327
+ {
1328
+ "epoch": 7.81,
1329
+ "learning_rate": 0.0002,
1330
+ "loss": 0.0006,
1331
+ "step": 211
1332
+ },
1333
+ {
1334
+ "epoch": 7.85,
1335
+ "learning_rate": 0.0002,
1336
+ "loss": 0.0012,
1337
+ "step": 212
1338
+ },
1339
+ {
1340
+ "epoch": 7.89,
1341
+ "learning_rate": 0.0002,
1342
+ "loss": 0.0044,
1343
+ "step": 213
1344
+ },
1345
+ {
1346
+ "epoch": 7.93,
1347
+ "learning_rate": 0.0002,
1348
+ "loss": 0.0063,
1349
+ "step": 214
1350
+ },
1351
+ {
1352
+ "epoch": 7.96,
1353
+ "learning_rate": 0.0002,
1354
+ "loss": 0.003,
1355
+ "step": 215
1356
+ },
1357
+ {
1358
+ "epoch": 8.0,
1359
+ "learning_rate": 0.0002,
1360
+ "loss": 0.0099,
1361
+ "step": 216
1362
+ },
1363
+ {
1364
+ "epoch": 8.0,
1365
+ "eval_loss": 0.0027730318251997232,
1366
+ "eval_runtime": 48.0783,
1367
+ "eval_samples_per_second": 8.84,
1368
+ "eval_steps_per_second": 0.562,
1369
+ "step": 216
1370
+ },
1371
+ {
1372
+ "epoch": 8.04,
1373
+ "learning_rate": 0.0002,
1374
+ "loss": 0.004,
1375
+ "step": 217
1376
+ },
1377
+ {
1378
+ "epoch": 8.07,
1379
+ "learning_rate": 0.0002,
1380
+ "loss": 0.0008,
1381
+ "step": 218
1382
+ },
1383
+ {
1384
+ "epoch": 8.11,
1385
+ "learning_rate": 0.0002,
1386
+ "loss": 0.0018,
1387
+ "step": 219
1388
+ },
1389
+ {
1390
+ "epoch": 8.15,
1391
+ "learning_rate": 0.0002,
1392
+ "loss": 0.0048,
1393
+ "step": 220
1394
+ },
1395
+ {
1396
+ "epoch": 8.19,
1397
+ "learning_rate": 0.0002,
1398
+ "loss": 0.011,
1399
+ "step": 221
1400
+ },
1401
+ {
1402
+ "epoch": 8.22,
1403
+ "learning_rate": 0.0002,
1404
+ "loss": 0.0051,
1405
+ "step": 222
1406
+ },
1407
+ {
1408
+ "epoch": 8.26,
1409
+ "learning_rate": 0.0002,
1410
+ "loss": 0.0046,
1411
+ "step": 223
1412
+ },
1413
+ {
1414
+ "epoch": 8.3,
1415
+ "learning_rate": 0.0002,
1416
+ "loss": 0.0041,
1417
+ "step": 224
1418
+ },
1419
+ {
1420
+ "epoch": 8.33,
1421
+ "learning_rate": 0.0002,
1422
+ "loss": 0.0055,
1423
+ "step": 225
1424
+ },
1425
+ {
1426
+ "epoch": 8.37,
1427
+ "learning_rate": 0.0002,
1428
+ "loss": 0.0016,
1429
+ "step": 226
1430
+ },
1431
+ {
1432
+ "epoch": 8.41,
1433
+ "learning_rate": 0.0002,
1434
+ "loss": 0.0047,
1435
+ "step": 227
1436
+ },
1437
+ {
1438
+ "epoch": 8.44,
1439
+ "learning_rate": 0.0002,
1440
+ "loss": 0.0024,
1441
+ "step": 228
1442
+ },
1443
+ {
1444
+ "epoch": 8.48,
1445
+ "learning_rate": 0.0002,
1446
+ "loss": 0.0057,
1447
+ "step": 229
1448
+ },
1449
+ {
1450
+ "epoch": 8.52,
1451
+ "learning_rate": 0.0002,
1452
+ "loss": 0.0033,
1453
+ "step": 230
1454
+ },
1455
+ {
1456
+ "epoch": 8.56,
1457
+ "learning_rate": 0.0002,
1458
+ "loss": 0.0124,
1459
+ "step": 231
1460
+ },
1461
+ {
1462
+ "epoch": 8.59,
1463
+ "learning_rate": 0.0002,
1464
+ "loss": 0.0107,
1465
+ "step": 232
1466
+ },
1467
+ {
1468
+ "epoch": 8.63,
1469
+ "learning_rate": 0.0002,
1470
+ "loss": 0.0041,
1471
+ "step": 233
1472
+ },
1473
+ {
1474
+ "epoch": 8.67,
1475
+ "learning_rate": 0.0002,
1476
+ "loss": 0.0035,
1477
+ "step": 234
1478
+ },
1479
+ {
1480
+ "epoch": 8.7,
1481
+ "learning_rate": 0.0002,
1482
+ "loss": 0.024,
1483
+ "step": 235
1484
+ },
1485
+ {
1486
+ "epoch": 8.74,
1487
+ "learning_rate": 0.0002,
1488
+ "loss": 0.0011,
1489
+ "step": 236
1490
+ },
1491
+ {
1492
+ "epoch": 8.78,
1493
+ "learning_rate": 0.0002,
1494
+ "loss": 0.0082,
1495
+ "step": 237
1496
+ },
1497
+ {
1498
+ "epoch": 8.81,
1499
+ "learning_rate": 0.0002,
1500
+ "loss": 0.0111,
1501
+ "step": 238
1502
+ },
1503
+ {
1504
+ "epoch": 8.85,
1505
+ "learning_rate": 0.0002,
1506
+ "loss": 0.0137,
1507
+ "step": 239
1508
+ },
1509
+ {
1510
+ "epoch": 8.89,
1511
+ "learning_rate": 0.0002,
1512
+ "loss": 0.0022,
1513
+ "step": 240
1514
+ },
1515
+ {
1516
+ "epoch": 8.93,
1517
+ "learning_rate": 0.0002,
1518
+ "loss": 0.0026,
1519
+ "step": 241
1520
+ },
1521
+ {
1522
+ "epoch": 8.96,
1523
+ "learning_rate": 0.0002,
1524
+ "loss": 0.0108,
1525
+ "step": 242
1526
+ },
1527
+ {
1528
+ "epoch": 9.0,
1529
+ "learning_rate": 0.0002,
1530
+ "loss": 0.0044,
1531
+ "step": 243
1532
+ },
1533
+ {
1534
+ "epoch": 9.0,
1535
+ "eval_loss": 0.0027062674053013325,
1536
+ "eval_runtime": 47.8752,
1537
+ "eval_samples_per_second": 8.877,
1538
+ "eval_steps_per_second": 0.564,
1539
+ "step": 243
1540
+ },
1541
+ {
1542
+ "epoch": 9.04,
1543
+ "learning_rate": 0.0002,
1544
+ "loss": 0.0016,
1545
+ "step": 244
1546
+ },
1547
+ {
1548
+ "epoch": 9.07,
1549
+ "learning_rate": 0.0002,
1550
+ "loss": 0.0017,
1551
+ "step": 245
1552
+ },
1553
+ {
1554
+ "epoch": 9.11,
1555
+ "learning_rate": 0.0002,
1556
+ "loss": 0.0036,
1557
+ "step": 246
1558
+ },
1559
+ {
1560
+ "epoch": 9.15,
1561
+ "learning_rate": 0.0002,
1562
+ "loss": 0.0045,
1563
+ "step": 247
1564
+ },
1565
+ {
1566
+ "epoch": 9.19,
1567
+ "learning_rate": 0.0002,
1568
+ "loss": 0.003,
1569
+ "step": 248
1570
+ },
1571
+ {
1572
+ "epoch": 9.22,
1573
+ "learning_rate": 0.0002,
1574
+ "loss": 0.0021,
1575
+ "step": 249
1576
+ },
1577
+ {
1578
+ "epoch": 9.26,
1579
+ "learning_rate": 0.0002,
1580
+ "loss": 0.0012,
1581
+ "step": 250
1582
+ },
1583
+ {
1584
+ "epoch": 9.3,
1585
+ "learning_rate": 0.0002,
1586
+ "loss": 0.0103,
1587
+ "step": 251
1588
+ },
1589
+ {
1590
+ "epoch": 9.33,
1591
+ "learning_rate": 0.0002,
1592
+ "loss": 0.0049,
1593
+ "step": 252
1594
+ },
1595
+ {
1596
+ "epoch": 9.37,
1597
+ "learning_rate": 0.0002,
1598
+ "loss": 0.0011,
1599
+ "step": 253
1600
+ },
1601
+ {
1602
+ "epoch": 9.41,
1603
+ "learning_rate": 0.0002,
1604
+ "loss": 0.0007,
1605
+ "step": 254
1606
+ },
1607
+ {
1608
+ "epoch": 9.44,
1609
+ "learning_rate": 0.0002,
1610
+ "loss": 0.0032,
1611
+ "step": 255
1612
+ },
1613
+ {
1614
+ "epoch": 9.48,
1615
+ "learning_rate": 0.0002,
1616
+ "loss": 0.0039,
1617
+ "step": 256
1618
+ },
1619
+ {
1620
+ "epoch": 9.52,
1621
+ "learning_rate": 0.0002,
1622
+ "loss": 0.0021,
1623
+ "step": 257
1624
+ },
1625
+ {
1626
+ "epoch": 9.56,
1627
+ "learning_rate": 0.0002,
1628
+ "loss": 0.0007,
1629
+ "step": 258
1630
+ },
1631
+ {
1632
+ "epoch": 9.59,
1633
+ "learning_rate": 0.0002,
1634
+ "loss": 0.0068,
1635
+ "step": 259
1636
+ },
1637
+ {
1638
+ "epoch": 9.63,
1639
+ "learning_rate": 0.0002,
1640
+ "loss": 0.0026,
1641
+ "step": 260
1642
+ },
1643
+ {
1644
+ "epoch": 9.67,
1645
+ "learning_rate": 0.0002,
1646
+ "loss": 0.0012,
1647
+ "step": 261
1648
+ },
1649
+ {
1650
+ "epoch": 9.7,
1651
+ "learning_rate": 0.0002,
1652
+ "loss": 0.001,
1653
+ "step": 262
1654
+ },
1655
+ {
1656
+ "epoch": 9.74,
1657
+ "learning_rate": 0.0002,
1658
+ "loss": 0.0033,
1659
+ "step": 263
1660
+ },
1661
+ {
1662
+ "epoch": 9.78,
1663
+ "learning_rate": 0.0002,
1664
+ "loss": 0.0024,
1665
+ "step": 264
1666
+ },
1667
+ {
1668
+ "epoch": 9.81,
1669
+ "learning_rate": 0.0002,
1670
+ "loss": 0.0051,
1671
+ "step": 265
1672
+ },
1673
+ {
1674
+ "epoch": 9.85,
1675
+ "learning_rate": 0.0002,
1676
+ "loss": 0.0071,
1677
+ "step": 266
1678
+ },
1679
+ {
1680
+ "epoch": 9.89,
1681
+ "learning_rate": 0.0002,
1682
+ "loss": 0.0034,
1683
+ "step": 267
1684
+ },
1685
+ {
1686
+ "epoch": 9.93,
1687
+ "learning_rate": 0.0002,
1688
+ "loss": 0.0008,
1689
+ "step": 268
1690
+ },
1691
+ {
1692
+ "epoch": 9.96,
1693
+ "learning_rate": 0.0002,
1694
+ "loss": 0.0056,
1695
+ "step": 269
1696
+ },
1697
+ {
1698
+ "epoch": 10.0,
1699
+ "learning_rate": 0.0002,
1700
+ "loss": 0.0012,
1701
+ "step": 270
1702
+ },
1703
+ {
1704
+ "epoch": 10.0,
1705
+ "eval_loss": 0.0049278344959020615,
1706
+ "eval_runtime": 47.9805,
1707
+ "eval_samples_per_second": 8.858,
1708
+ "eval_steps_per_second": 0.563,
1709
+ "step": 270
1710
+ },
1711
+ {
1712
+ "epoch": 10.04,
1713
+ "learning_rate": 0.0002,
1714
+ "loss": 0.0064,
1715
+ "step": 271
1716
+ },
1717
+ {
1718
+ "epoch": 10.07,
1719
+ "learning_rate": 0.0002,
1720
+ "loss": 0.0056,
1721
+ "step": 272
1722
+ },
1723
+ {
1724
+ "epoch": 10.11,
1725
+ "learning_rate": 0.0002,
1726
+ "loss": 0.0066,
1727
+ "step": 273
1728
+ },
1729
+ {
1730
+ "epoch": 10.15,
1731
+ "learning_rate": 0.0002,
1732
+ "loss": 0.0102,
1733
+ "step": 274
1734
+ },
1735
+ {
1736
+ "epoch": 10.19,
1737
+ "learning_rate": 0.0002,
1738
+ "loss": 0.008,
1739
+ "step": 275
1740
+ },
1741
+ {
1742
+ "epoch": 10.22,
1743
+ "learning_rate": 0.0002,
1744
+ "loss": 0.0004,
1745
+ "step": 276
1746
+ },
1747
+ {
1748
+ "epoch": 10.26,
1749
+ "learning_rate": 0.0002,
1750
+ "loss": 0.0046,
1751
+ "step": 277
1752
+ },
1753
+ {
1754
+ "epoch": 10.3,
1755
+ "learning_rate": 0.0002,
1756
+ "loss": 0.0038,
1757
+ "step": 278
1758
+ },
1759
+ {
1760
+ "epoch": 10.33,
1761
+ "learning_rate": 0.0002,
1762
+ "loss": 0.0053,
1763
+ "step": 279
1764
+ },
1765
+ {
1766
+ "epoch": 10.37,
1767
+ "learning_rate": 0.0002,
1768
+ "loss": 0.0035,
1769
+ "step": 280
1770
+ },
1771
+ {
1772
+ "epoch": 10.41,
1773
+ "learning_rate": 0.0002,
1774
+ "loss": 0.0044,
1775
+ "step": 281
1776
+ },
1777
+ {
1778
+ "epoch": 10.44,
1779
+ "learning_rate": 0.0002,
1780
+ "loss": 0.0043,
1781
+ "step": 282
1782
+ },
1783
+ {
1784
+ "epoch": 10.48,
1785
+ "learning_rate": 0.0002,
1786
+ "loss": 0.0083,
1787
+ "step": 283
1788
+ },
1789
+ {
1790
+ "epoch": 10.52,
1791
+ "learning_rate": 0.0002,
1792
+ "loss": 0.0063,
1793
+ "step": 284
1794
+ },
1795
+ {
1796
+ "epoch": 10.56,
1797
+ "learning_rate": 0.0002,
1798
+ "loss": 0.002,
1799
+ "step": 285
1800
+ },
1801
+ {
1802
+ "epoch": 10.59,
1803
+ "learning_rate": 0.0002,
1804
+ "loss": 0.006,
1805
+ "step": 286
1806
+ },
1807
+ {
1808
+ "epoch": 10.63,
1809
+ "learning_rate": 0.0002,
1810
+ "loss": 0.0026,
1811
+ "step": 287
1812
+ },
1813
+ {
1814
+ "epoch": 10.67,
1815
+ "learning_rate": 0.0002,
1816
+ "loss": 0.0022,
1817
+ "step": 288
1818
+ },
1819
+ {
1820
+ "epoch": 10.7,
1821
+ "learning_rate": 0.0002,
1822
+ "loss": 0.002,
1823
+ "step": 289
1824
+ },
1825
+ {
1826
+ "epoch": 10.74,
1827
+ "learning_rate": 0.0002,
1828
+ "loss": 0.0081,
1829
+ "step": 290
1830
+ },
1831
+ {
1832
+ "epoch": 10.78,
1833
+ "learning_rate": 0.0002,
1834
+ "loss": 0.0019,
1835
+ "step": 291
1836
+ },
1837
+ {
1838
+ "epoch": 10.81,
1839
+ "learning_rate": 0.0002,
1840
+ "loss": 0.0045,
1841
+ "step": 292
1842
+ },
1843
+ {
1844
+ "epoch": 10.85,
1845
+ "learning_rate": 0.0002,
1846
+ "loss": 0.0011,
1847
+ "step": 293
1848
+ },
1849
+ {
1850
+ "epoch": 10.89,
1851
+ "learning_rate": 0.0002,
1852
+ "loss": 0.0156,
1853
+ "step": 294
1854
+ },
1855
+ {
1856
+ "epoch": 10.93,
1857
+ "learning_rate": 0.0002,
1858
+ "loss": 0.0042,
1859
+ "step": 295
1860
+ },
1861
+ {
1862
+ "epoch": 10.96,
1863
+ "learning_rate": 0.0002,
1864
+ "loss": 0.0035,
1865
+ "step": 296
1866
+ },
1867
+ {
1868
+ "epoch": 11.0,
1869
+ "learning_rate": 0.0002,
1870
+ "loss": 0.0071,
1871
+ "step": 297
1872
+ },
1873
+ {
1874
+ "epoch": 11.0,
1875
+ "eval_loss": 0.0035311884712427855,
1876
+ "eval_runtime": 48.1845,
1877
+ "eval_samples_per_second": 8.82,
1878
+ "eval_steps_per_second": 0.56,
1879
+ "step": 297
1880
+ },
1881
+ {
1882
+ "epoch": 11.04,
1883
+ "learning_rate": 0.0002,
1884
+ "loss": 0.004,
1885
+ "step": 298
1886
+ },
1887
+ {
1888
+ "epoch": 11.07,
1889
+ "learning_rate": 0.0002,
1890
+ "loss": 0.0023,
1891
+ "step": 299
1892
+ },
1893
+ {
1894
+ "epoch": 11.11,
1895
+ "learning_rate": 0.0002,
1896
+ "loss": 0.0044,
1897
+ "step": 300
1898
+ },
1899
+ {
1900
+ "epoch": 11.15,
1901
+ "learning_rate": 0.0002,
1902
+ "loss": 0.0092,
1903
+ "step": 301
1904
+ },
1905
+ {
1906
+ "epoch": 11.19,
1907
+ "learning_rate": 0.0002,
1908
+ "loss": 0.0129,
1909
+ "step": 302
1910
+ },
1911
+ {
1912
+ "epoch": 11.22,
1913
+ "learning_rate": 0.0002,
1914
+ "loss": 0.0076,
1915
+ "step": 303
1916
+ },
1917
+ {
1918
+ "epoch": 11.26,
1919
+ "learning_rate": 0.0002,
1920
+ "loss": 0.0066,
1921
+ "step": 304
1922
+ },
1923
+ {
1924
+ "epoch": 11.3,
1925
+ "learning_rate": 0.0002,
1926
+ "loss": 0.0039,
1927
+ "step": 305
1928
+ },
1929
+ {
1930
+ "epoch": 11.33,
1931
+ "learning_rate": 0.0002,
1932
+ "loss": 0.005,
1933
+ "step": 306
1934
+ },
1935
+ {
1936
+ "epoch": 11.37,
1937
+ "learning_rate": 0.0002,
1938
+ "loss": 0.0026,
1939
+ "step": 307
1940
+ },
1941
+ {
1942
+ "epoch": 11.41,
1943
+ "learning_rate": 0.0002,
1944
+ "loss": 0.0033,
1945
+ "step": 308
1946
+ },
1947
+ {
1948
+ "epoch": 11.44,
1949
+ "learning_rate": 0.0002,
1950
+ "loss": 0.0051,
1951
+ "step": 309
1952
+ },
1953
+ {
1954
+ "epoch": 11.48,
1955
+ "learning_rate": 0.0002,
1956
+ "loss": 0.0049,
1957
+ "step": 310
1958
+ },
1959
+ {
1960
+ "epoch": 11.52,
1961
+ "learning_rate": 0.0002,
1962
+ "loss": 0.003,
1963
+ "step": 311
1964
+ },
1965
+ {
1966
+ "epoch": 11.56,
1967
+ "learning_rate": 0.0002,
1968
+ "loss": 0.0038,
1969
+ "step": 312
1970
+ },
1971
+ {
1972
+ "epoch": 11.59,
1973
+ "learning_rate": 0.0002,
1974
+ "loss": 0.0033,
1975
+ "step": 313
1976
+ },
1977
+ {
1978
+ "epoch": 11.63,
1979
+ "learning_rate": 0.0002,
1980
+ "loss": 0.0057,
1981
+ "step": 314
1982
+ },
1983
+ {
1984
+ "epoch": 11.67,
1985
+ "learning_rate": 0.0002,
1986
+ "loss": 0.0299,
1987
+ "step": 315
1988
+ },
1989
+ {
1990
+ "epoch": 11.7,
1991
+ "learning_rate": 0.0002,
1992
+ "loss": 0.0033,
1993
+ "step": 316
1994
+ },
1995
+ {
1996
+ "epoch": 11.74,
1997
+ "learning_rate": 0.0002,
1998
+ "loss": 0.0052,
1999
+ "step": 317
2000
+ },
2001
+ {
2002
+ "epoch": 11.78,
2003
+ "learning_rate": 0.0002,
2004
+ "loss": 0.0172,
2005
+ "step": 318
2006
+ },
2007
+ {
2008
+ "epoch": 11.81,
2009
+ "learning_rate": 0.0002,
2010
+ "loss": 0.0031,
2011
+ "step": 319
2012
+ },
2013
+ {
2014
+ "epoch": 11.85,
2015
+ "learning_rate": 0.0002,
2016
+ "loss": 0.0056,
2017
+ "step": 320
2018
+ },
2019
+ {
2020
+ "epoch": 11.89,
2021
+ "learning_rate": 0.0002,
2022
+ "loss": 0.014,
2023
+ "step": 321
2024
+ },
2025
+ {
2026
+ "epoch": 11.93,
2027
+ "learning_rate": 0.0002,
2028
+ "loss": 0.0137,
2029
+ "step": 322
2030
+ },
2031
+ {
2032
+ "epoch": 11.96,
2033
+ "learning_rate": 0.0002,
2034
+ "loss": 0.0036,
2035
+ "step": 323
2036
+ },
2037
+ {
2038
+ "epoch": 12.0,
2039
+ "learning_rate": 0.0002,
2040
+ "loss": 0.0066,
2041
+ "step": 324
2042
+ },
2043
+ {
2044
+ "epoch": 12.0,
2045
+ "eval_loss": 0.013012222945690155,
2046
+ "eval_runtime": 47.9144,
2047
+ "eval_samples_per_second": 8.87,
2048
+ "eval_steps_per_second": 0.564,
2049
+ "step": 324
2050
+ },
2051
+ {
2052
+ "epoch": 12.04,
2053
+ "learning_rate": 0.0002,
2054
+ "loss": 0.0128,
2055
+ "step": 325
2056
+ },
2057
+ {
2058
+ "epoch": 12.07,
2059
+ "learning_rate": 0.0002,
2060
+ "loss": 0.0122,
2061
+ "step": 326
2062
+ },
2063
+ {
2064
+ "epoch": 12.11,
2065
+ "learning_rate": 0.0002,
2066
+ "loss": 0.0031,
2067
+ "step": 327
2068
+ },
2069
+ {
2070
+ "epoch": 12.15,
2071
+ "learning_rate": 0.0002,
2072
+ "loss": 0.0041,
2073
+ "step": 328
2074
+ },
2075
+ {
2076
+ "epoch": 12.19,
2077
+ "learning_rate": 0.0002,
2078
+ "loss": 0.0123,
2079
+ "step": 329
2080
+ },
2081
+ {
2082
+ "epoch": 12.22,
2083
+ "learning_rate": 0.0002,
2084
+ "loss": 0.0062,
2085
+ "step": 330
2086
+ },
2087
+ {
2088
+ "epoch": 12.26,
2089
+ "learning_rate": 0.0002,
2090
+ "loss": 0.0218,
2091
+ "step": 331
2092
+ },
2093
+ {
2094
+ "epoch": 12.3,
2095
+ "learning_rate": 0.0002,
2096
+ "loss": 0.0045,
2097
+ "step": 332
2098
+ },
2099
+ {
2100
+ "epoch": 12.33,
2101
+ "learning_rate": 0.0002,
2102
+ "loss": 0.0039,
2103
+ "step": 333
2104
+ },
2105
+ {
2106
+ "epoch": 12.37,
2107
+ "learning_rate": 0.0002,
2108
+ "loss": 0.0103,
2109
+ "step": 334
2110
+ },
2111
+ {
2112
+ "epoch": 12.41,
2113
+ "learning_rate": 0.0002,
2114
+ "loss": 0.0121,
2115
+ "step": 335
2116
+ },
2117
+ {
2118
+ "epoch": 12.44,
2119
+ "learning_rate": 0.0002,
2120
+ "loss": 0.0071,
2121
+ "step": 336
2122
+ },
2123
+ {
2124
+ "epoch": 12.48,
2125
+ "learning_rate": 0.0002,
2126
+ "loss": 0.0068,
2127
+ "step": 337
2128
+ },
2129
+ {
2130
+ "epoch": 12.52,
2131
+ "learning_rate": 0.0002,
2132
+ "loss": 0.0067,
2133
+ "step": 338
2134
+ },
2135
+ {
2136
+ "epoch": 12.56,
2137
+ "learning_rate": 0.0002,
2138
+ "loss": 0.0118,
2139
+ "step": 339
2140
+ },
2141
+ {
2142
+ "epoch": 12.59,
2143
+ "learning_rate": 0.0002,
2144
+ "loss": 0.0178,
2145
+ "step": 340
2146
+ },
2147
+ {
2148
+ "epoch": 12.63,
2149
+ "learning_rate": 0.0002,
2150
+ "loss": 0.0357,
2151
+ "step": 341
2152
+ },
2153
+ {
2154
+ "epoch": 12.67,
2155
+ "learning_rate": 0.0002,
2156
+ "loss": 0.0171,
2157
+ "step": 342
2158
+ },
2159
+ {
2160
+ "epoch": 12.7,
2161
+ "learning_rate": 0.0002,
2162
+ "loss": 0.004,
2163
+ "step": 343
2164
+ },
2165
+ {
2166
+ "epoch": 12.74,
2167
+ "learning_rate": 0.0002,
2168
+ "loss": 0.0078,
2169
+ "step": 344
2170
+ },
2171
+ {
2172
+ "epoch": 12.78,
2173
+ "learning_rate": 0.0002,
2174
+ "loss": 0.0122,
2175
+ "step": 345
2176
+ },
2177
+ {
2178
+ "epoch": 12.81,
2179
+ "learning_rate": 0.0002,
2180
+ "loss": 0.0039,
2181
+ "step": 346
2182
+ },
2183
+ {
2184
+ "epoch": 12.85,
2185
+ "learning_rate": 0.0002,
2186
+ "loss": 0.0144,
2187
+ "step": 347
2188
+ },
2189
+ {
2190
+ "epoch": 12.89,
2191
+ "learning_rate": 0.0002,
2192
+ "loss": 0.013,
2193
+ "step": 348
2194
+ },
2195
+ {
2196
+ "epoch": 12.93,
2197
+ "learning_rate": 0.0002,
2198
+ "loss": 0.0137,
2199
+ "step": 349
2200
+ },
2201
+ {
2202
+ "epoch": 12.96,
2203
+ "learning_rate": 0.0002,
2204
+ "loss": 0.0057,
2205
+ "step": 350
2206
+ },
2207
+ {
2208
+ "epoch": 13.0,
2209
+ "learning_rate": 0.0002,
2210
+ "loss": 0.0107,
2211
+ "step": 351
2212
+ },
2213
+ {
2214
+ "epoch": 13.0,
2215
+ "eval_loss": 0.006607715506106615,
2216
+ "eval_runtime": 47.9517,
2217
+ "eval_samples_per_second": 8.863,
2218
+ "eval_steps_per_second": 0.563,
2219
+ "step": 351
2220
+ },
2221
+ {
2222
+ "epoch": 13.04,
2223
+ "learning_rate": 0.0002,
2224
+ "loss": 0.0047,
2225
+ "step": 352
2226
+ },
2227
+ {
2228
+ "epoch": 13.07,
2229
+ "learning_rate": 0.0002,
2230
+ "loss": 0.0094,
2231
+ "step": 353
2232
+ },
2233
+ {
2234
+ "epoch": 13.11,
2235
+ "learning_rate": 0.0002,
2236
+ "loss": 0.0125,
2237
+ "step": 354
2238
+ },
2239
+ {
2240
+ "epoch": 13.15,
2241
+ "learning_rate": 0.0002,
2242
+ "loss": 0.0021,
2243
+ "step": 355
2244
+ },
2245
+ {
2246
+ "epoch": 13.19,
2247
+ "learning_rate": 0.0002,
2248
+ "loss": 0.0065,
2249
+ "step": 356
2250
+ },
2251
+ {
2252
+ "epoch": 13.22,
2253
+ "learning_rate": 0.0002,
2254
+ "loss": 0.0108,
2255
+ "step": 357
2256
+ },
2257
+ {
2258
+ "epoch": 13.26,
2259
+ "learning_rate": 0.0002,
2260
+ "loss": 0.0021,
2261
+ "step": 358
2262
+ },
2263
+ {
2264
+ "epoch": 13.3,
2265
+ "learning_rate": 0.0002,
2266
+ "loss": 0.0064,
2267
+ "step": 359
2268
+ },
2269
+ {
2270
+ "epoch": 13.33,
2271
+ "learning_rate": 0.0002,
2272
+ "loss": 0.0047,
2273
+ "step": 360
2274
+ },
2275
+ {
2276
+ "epoch": 13.37,
2277
+ "learning_rate": 0.0002,
2278
+ "loss": 0.0079,
2279
+ "step": 361
2280
+ },
2281
+ {
2282
+ "epoch": 13.41,
2283
+ "learning_rate": 0.0002,
2284
+ "loss": 0.0056,
2285
+ "step": 362
2286
+ },
2287
+ {
2288
+ "epoch": 13.44,
2289
+ "learning_rate": 0.0002,
2290
+ "loss": 0.0096,
2291
+ "step": 363
2292
+ },
2293
+ {
2294
+ "epoch": 13.48,
2295
+ "learning_rate": 0.0002,
2296
+ "loss": 0.0011,
2297
+ "step": 364
2298
+ },
2299
+ {
2300
+ "epoch": 13.52,
2301
+ "learning_rate": 0.0002,
2302
+ "loss": 0.003,
2303
+ "step": 365
2304
+ },
2305
+ {
2306
+ "epoch": 13.56,
2307
+ "learning_rate": 0.0002,
2308
+ "loss": 0.0026,
2309
+ "step": 366
2310
+ },
2311
+ {
2312
+ "epoch": 13.59,
2313
+ "learning_rate": 0.0002,
2314
+ "loss": 0.0035,
2315
+ "step": 367
2316
+ },
2317
+ {
2318
+ "epoch": 13.63,
2319
+ "learning_rate": 0.0002,
2320
+ "loss": 0.0046,
2321
+ "step": 368
2322
+ },
2323
+ {
2324
+ "epoch": 13.67,
2325
+ "learning_rate": 0.0002,
2326
+ "loss": 0.0053,
2327
+ "step": 369
2328
+ },
2329
+ {
2330
+ "epoch": 13.7,
2331
+ "learning_rate": 0.0002,
2332
+ "loss": 0.008,
2333
+ "step": 370
2334
+ },
2335
+ {
2336
+ "epoch": 13.74,
2337
+ "learning_rate": 0.0002,
2338
+ "loss": 0.0065,
2339
+ "step": 371
2340
+ },
2341
+ {
2342
+ "epoch": 13.78,
2343
+ "learning_rate": 0.0002,
2344
+ "loss": 0.0028,
2345
+ "step": 372
2346
+ },
2347
+ {
2348
+ "epoch": 13.81,
2349
+ "learning_rate": 0.0002,
2350
+ "loss": 0.0045,
2351
+ "step": 373
2352
+ },
2353
+ {
2354
+ "epoch": 13.85,
2355
+ "learning_rate": 0.0002,
2356
+ "loss": 0.0039,
2357
+ "step": 374
2358
+ },
2359
+ {
2360
+ "epoch": 13.89,
2361
+ "learning_rate": 0.0002,
2362
+ "loss": 0.0099,
2363
+ "step": 375
2364
+ },
2365
+ {
2366
+ "epoch": 13.93,
2367
+ "learning_rate": 0.0002,
2368
+ "loss": 0.0048,
2369
+ "step": 376
2370
+ },
2371
+ {
2372
+ "epoch": 13.96,
2373
+ "learning_rate": 0.0002,
2374
+ "loss": 0.0024,
2375
+ "step": 377
2376
+ },
2377
+ {
2378
+ "epoch": 14.0,
2379
+ "learning_rate": 0.0002,
2380
+ "loss": 0.0095,
2381
+ "step": 378
2382
+ },
2383
+ {
2384
+ "epoch": 14.0,
2385
+ "eval_loss": 0.005205091089010239,
2386
+ "eval_runtime": 47.986,
2387
+ "eval_samples_per_second": 8.857,
2388
+ "eval_steps_per_second": 0.563,
2389
+ "step": 378
2390
+ },
2391
+ {
2392
+ "epoch": 14.04,
2393
+ "learning_rate": 0.0002,
2394
+ "loss": 0.0035,
2395
+ "step": 379
2396
+ },
2397
+ {
2398
+ "epoch": 14.07,
2399
+ "learning_rate": 0.0002,
2400
+ "loss": 0.0026,
2401
+ "step": 380
2402
+ },
2403
+ {
2404
+ "epoch": 14.11,
2405
+ "learning_rate": 0.0002,
2406
+ "loss": 0.0042,
2407
+ "step": 381
2408
+ },
2409
+ {
2410
+ "epoch": 14.15,
2411
+ "learning_rate": 0.0002,
2412
+ "loss": 0.0059,
2413
+ "step": 382
2414
+ },
2415
+ {
2416
+ "epoch": 14.19,
2417
+ "learning_rate": 0.0002,
2418
+ "loss": 0.0068,
2419
+ "step": 383
2420
+ },
2421
+ {
2422
+ "epoch": 14.22,
2423
+ "learning_rate": 0.0002,
2424
+ "loss": 0.0062,
2425
+ "step": 384
2426
+ },
2427
+ {
2428
+ "epoch": 14.26,
2429
+ "learning_rate": 0.0002,
2430
+ "loss": 0.0065,
2431
+ "step": 385
2432
+ },
2433
+ {
2434
+ "epoch": 14.3,
2435
+ "learning_rate": 0.0002,
2436
+ "loss": 0.0026,
2437
+ "step": 386
2438
+ },
2439
+ {
2440
+ "epoch": 14.33,
2441
+ "learning_rate": 0.0002,
2442
+ "loss": 0.0103,
2443
+ "step": 387
2444
+ },
2445
+ {
2446
+ "epoch": 14.37,
2447
+ "learning_rate": 0.0002,
2448
+ "loss": 0.005,
2449
+ "step": 388
2450
+ },
2451
+ {
2452
+ "epoch": 14.41,
2453
+ "learning_rate": 0.0002,
2454
+ "loss": 0.0079,
2455
+ "step": 389
2456
+ },
2457
+ {
2458
+ "epoch": 14.44,
2459
+ "learning_rate": 0.0002,
2460
+ "loss": 0.0114,
2461
+ "step": 390
2462
+ },
2463
+ {
2464
+ "epoch": 14.48,
2465
+ "learning_rate": 0.0002,
2466
+ "loss": 0.0094,
2467
+ "step": 391
2468
+ },
2469
+ {
2470
+ "epoch": 14.52,
2471
+ "learning_rate": 0.0002,
2472
+ "loss": 0.0036,
2473
+ "step": 392
2474
+ },
2475
+ {
2476
+ "epoch": 14.56,
2477
+ "learning_rate": 0.0002,
2478
+ "loss": 0.0137,
2479
+ "step": 393
2480
+ },
2481
+ {
2482
+ "epoch": 14.59,
2483
+ "learning_rate": 0.0002,
2484
+ "loss": 0.0054,
2485
+ "step": 394
2486
+ },
2487
+ {
2488
+ "epoch": 14.63,
2489
+ "learning_rate": 0.0002,
2490
+ "loss": 0.0036,
2491
+ "step": 395
2492
+ },
2493
+ {
2494
+ "epoch": 14.67,
2495
+ "learning_rate": 0.0002,
2496
+ "loss": 0.0043,
2497
+ "step": 396
2498
+ },
2499
+ {
2500
+ "epoch": 14.7,
2501
+ "learning_rate": 0.0002,
2502
+ "loss": 0.0018,
2503
+ "step": 397
2504
+ },
2505
+ {
2506
+ "epoch": 14.74,
2507
+ "learning_rate": 0.0002,
2508
+ "loss": 0.0027,
2509
+ "step": 398
2510
+ },
2511
+ {
2512
+ "epoch": 14.78,
2513
+ "learning_rate": 0.0002,
2514
+ "loss": 0.0043,
2515
+ "step": 399
2516
+ },
2517
+ {
2518
+ "epoch": 14.81,
2519
+ "learning_rate": 0.0002,
2520
+ "loss": 0.0081,
2521
+ "step": 400
2522
+ },
2523
+ {
2524
+ "epoch": 14.85,
2525
+ "learning_rate": 0.0002,
2526
+ "loss": 0.0032,
2527
+ "step": 401
2528
+ },
2529
+ {
2530
+ "epoch": 14.89,
2531
+ "learning_rate": 0.0002,
2532
+ "loss": 0.0068,
2533
+ "step": 402
2534
+ },
2535
+ {
2536
+ "epoch": 14.93,
2537
+ "learning_rate": 0.0002,
2538
+ "loss": 0.0018,
2539
+ "step": 403
2540
+ },
2541
+ {
2542
+ "epoch": 14.96,
2543
+ "learning_rate": 0.0002,
2544
+ "loss": 0.0056,
2545
+ "step": 404
2546
+ },
2547
+ {
2548
+ "epoch": 15.0,
2549
+ "learning_rate": 0.0002,
2550
+ "loss": 0.0044,
2551
+ "step": 405
2552
+ },
2553
+ {
2554
+ "epoch": 15.0,
2555
+ "eval_loss": 0.002976334420964122,
2556
+ "eval_runtime": 48.0244,
2557
+ "eval_samples_per_second": 8.85,
2558
+ "eval_steps_per_second": 0.562,
2559
+ "step": 405
2560
+ },
2561
+ {
2562
+ "epoch": 15.04,
2563
+ "learning_rate": 0.0002,
2564
+ "loss": 0.0025,
2565
+ "step": 406
2566
+ },
2567
+ {
2568
+ "epoch": 15.07,
2569
+ "learning_rate": 0.0002,
2570
+ "loss": 0.002,
2571
+ "step": 407
2572
+ },
2573
+ {
2574
+ "epoch": 15.11,
2575
+ "learning_rate": 0.0002,
2576
+ "loss": 0.0014,
2577
+ "step": 408
2578
+ },
2579
+ {
2580
+ "epoch": 15.15,
2581
+ "learning_rate": 0.0002,
2582
+ "loss": 0.0024,
2583
+ "step": 409
2584
+ },
2585
+ {
2586
+ "epoch": 15.19,
2587
+ "learning_rate": 0.0002,
2588
+ "loss": 0.0066,
2589
+ "step": 410
2590
+ },
2591
+ {
2592
+ "epoch": 15.22,
2593
+ "learning_rate": 0.0002,
2594
+ "loss": 0.002,
2595
+ "step": 411
2596
+ },
2597
+ {
2598
+ "epoch": 15.26,
2599
+ "learning_rate": 0.0002,
2600
+ "loss": 0.0022,
2601
+ "step": 412
2602
+ },
2603
+ {
2604
+ "epoch": 15.3,
2605
+ "learning_rate": 0.0002,
2606
+ "loss": 0.0098,
2607
+ "step": 413
2608
+ },
2609
+ {
2610
+ "epoch": 15.33,
2611
+ "learning_rate": 0.0002,
2612
+ "loss": 0.0094,
2613
+ "step": 414
2614
+ },
2615
+ {
2616
+ "epoch": 15.37,
2617
+ "learning_rate": 0.0002,
2618
+ "loss": 0.0048,
2619
+ "step": 415
2620
+ },
2621
+ {
2622
+ "epoch": 15.41,
2623
+ "learning_rate": 0.0002,
2624
+ "loss": 0.0042,
2625
+ "step": 416
2626
+ },
2627
+ {
2628
+ "epoch": 15.44,
2629
+ "learning_rate": 0.0002,
2630
+ "loss": 0.0055,
2631
+ "step": 417
2632
+ },
2633
+ {
2634
+ "epoch": 15.48,
2635
+ "learning_rate": 0.0002,
2636
+ "loss": 0.003,
2637
+ "step": 418
2638
+ },
2639
+ {
2640
+ "epoch": 15.52,
2641
+ "learning_rate": 0.0002,
2642
+ "loss": 0.0016,
2643
+ "step": 419
2644
+ },
2645
+ {
2646
+ "epoch": 15.56,
2647
+ "learning_rate": 0.0002,
2648
+ "loss": 0.0077,
2649
+ "step": 420
2650
+ },
2651
+ {
2652
+ "epoch": 15.59,
2653
+ "learning_rate": 0.0002,
2654
+ "loss": 0.0031,
2655
+ "step": 421
2656
+ },
2657
+ {
2658
+ "epoch": 15.63,
2659
+ "learning_rate": 0.0002,
2660
+ "loss": 0.0041,
2661
+ "step": 422
2662
+ },
2663
+ {
2664
+ "epoch": 15.67,
2665
+ "learning_rate": 0.0002,
2666
+ "loss": 0.031,
2667
+ "step": 423
2668
+ },
2669
+ {
2670
+ "epoch": 15.7,
2671
+ "learning_rate": 0.0002,
2672
+ "loss": 0.0038,
2673
+ "step": 424
2674
+ },
2675
+ {
2676
+ "epoch": 15.74,
2677
+ "learning_rate": 0.0002,
2678
+ "loss": 0.0051,
2679
+ "step": 425
2680
+ },
2681
+ {
2682
+ "epoch": 15.78,
2683
+ "learning_rate": 0.0002,
2684
+ "loss": 0.0057,
2685
+ "step": 426
2686
+ },
2687
+ {
2688
+ "epoch": 15.81,
2689
+ "learning_rate": 0.0002,
2690
+ "loss": 0.0033,
2691
+ "step": 427
2692
+ },
2693
+ {
2694
+ "epoch": 15.85,
2695
+ "learning_rate": 0.0002,
2696
+ "loss": 0.0032,
2697
+ "step": 428
2698
+ },
2699
+ {
2700
+ "epoch": 15.89,
2701
+ "learning_rate": 0.0002,
2702
+ "loss": 0.0105,
2703
+ "step": 429
2704
+ },
2705
+ {
2706
+ "epoch": 15.93,
2707
+ "learning_rate": 0.0002,
2708
+ "loss": 0.0074,
2709
+ "step": 430
2710
+ },
2711
+ {
2712
+ "epoch": 15.96,
2713
+ "learning_rate": 0.0002,
2714
+ "loss": 0.0031,
2715
+ "step": 431
2716
+ },
2717
+ {
2718
+ "epoch": 16.0,
2719
+ "learning_rate": 0.0002,
2720
+ "loss": 0.0051,
2721
+ "step": 432
2722
+ },
2723
+ {
2724
+ "epoch": 16.0,
2725
+ "eval_loss": 0.007675724104046822,
2726
+ "eval_runtime": 48.0403,
2727
+ "eval_samples_per_second": 8.847,
2728
+ "eval_steps_per_second": 0.562,
2729
+ "step": 432
2730
+ },
2731
+ {
2732
+ "epoch": 16.04,
2733
+ "learning_rate": 0.0002,
2734
+ "loss": 0.0071,
2735
+ "step": 433
2736
+ },
2737
+ {
2738
+ "epoch": 16.07,
2739
+ "learning_rate": 0.0002,
2740
+ "loss": 0.0037,
2741
+ "step": 434
2742
+ },
2743
+ {
2744
+ "epoch": 16.11,
2745
+ "learning_rate": 0.0002,
2746
+ "loss": 0.0092,
2747
+ "step": 435
2748
+ },
2749
+ {
2750
+ "epoch": 16.15,
2751
+ "learning_rate": 0.0002,
2752
+ "loss": 0.0031,
2753
+ "step": 436
2754
+ },
2755
+ {
2756
+ "epoch": 16.19,
2757
+ "learning_rate": 0.0002,
2758
+ "loss": 0.0072,
2759
+ "step": 437
2760
+ },
2761
+ {
2762
+ "epoch": 16.22,
2763
+ "learning_rate": 0.0002,
2764
+ "loss": 0.0007,
2765
+ "step": 438
2766
+ },
2767
+ {
2768
+ "epoch": 16.26,
2769
+ "learning_rate": 0.0002,
2770
+ "loss": 0.0046,
2771
+ "step": 439
2772
+ },
2773
+ {
2774
+ "epoch": 16.3,
2775
+ "learning_rate": 0.0002,
2776
+ "loss": 0.0144,
2777
+ "step": 440
2778
+ },
2779
+ {
2780
+ "epoch": 16.33,
2781
+ "learning_rate": 0.0002,
2782
+ "loss": 0.0023,
2783
+ "step": 441
2784
+ },
2785
+ {
2786
+ "epoch": 16.37,
2787
+ "learning_rate": 0.0002,
2788
+ "loss": 0.0009,
2789
+ "step": 442
2790
+ },
2791
+ {
2792
+ "epoch": 16.41,
2793
+ "learning_rate": 0.0002,
2794
+ "loss": 0.005,
2795
+ "step": 443
2796
+ },
2797
+ {
2798
+ "epoch": 16.44,
2799
+ "learning_rate": 0.0002,
2800
+ "loss": 0.014,
2801
+ "step": 444
2802
+ },
2803
+ {
2804
+ "epoch": 16.48,
2805
+ "learning_rate": 0.0002,
2806
+ "loss": 0.006,
2807
+ "step": 445
2808
+ },
2809
+ {
2810
+ "epoch": 16.52,
2811
+ "learning_rate": 0.0002,
2812
+ "loss": 0.0028,
2813
+ "step": 446
2814
+ },
2815
+ {
2816
+ "epoch": 16.56,
2817
+ "learning_rate": 0.0002,
2818
+ "loss": 0.0028,
2819
+ "step": 447
2820
+ },
2821
+ {
2822
+ "epoch": 16.59,
2823
+ "learning_rate": 0.0002,
2824
+ "loss": 0.0034,
2825
+ "step": 448
2826
+ },
2827
+ {
2828
+ "epoch": 16.63,
2829
+ "learning_rate": 0.0002,
2830
+ "loss": 0.0072,
2831
+ "step": 449
2832
+ },
2833
+ {
2834
+ "epoch": 16.67,
2835
+ "learning_rate": 0.0002,
2836
+ "loss": 0.001,
2837
+ "step": 450
2838
+ },
2839
+ {
2840
+ "epoch": 16.7,
2841
+ "learning_rate": 0.0002,
2842
+ "loss": 0.003,
2843
+ "step": 451
2844
+ },
2845
+ {
2846
+ "epoch": 16.74,
2847
+ "learning_rate": 0.0002,
2848
+ "loss": 0.0125,
2849
+ "step": 452
2850
+ },
2851
+ {
2852
+ "epoch": 16.78,
2853
+ "learning_rate": 0.0002,
2854
+ "loss": 0.0071,
2855
+ "step": 453
2856
+ },
2857
+ {
2858
+ "epoch": 16.81,
2859
+ "learning_rate": 0.0002,
2860
+ "loss": 0.0024,
2861
+ "step": 454
2862
+ },
2863
+ {
2864
+ "epoch": 16.85,
2865
+ "learning_rate": 0.0002,
2866
+ "loss": 0.0009,
2867
+ "step": 455
2868
+ },
2869
+ {
2870
+ "epoch": 16.89,
2871
+ "learning_rate": 0.0002,
2872
+ "loss": 0.0031,
2873
+ "step": 456
2874
+ },
2875
+ {
2876
+ "epoch": 16.93,
2877
+ "learning_rate": 0.0002,
2878
+ "loss": 0.0014,
2879
+ "step": 457
2880
+ },
2881
+ {
2882
+ "epoch": 16.96,
2883
+ "learning_rate": 0.0002,
2884
+ "loss": 0.0093,
2885
+ "step": 458
2886
+ },
2887
+ {
2888
+ "epoch": 17.0,
2889
+ "learning_rate": 0.0002,
2890
+ "loss": 0.0012,
2891
+ "step": 459
2892
+ },
2893
+ {
2894
+ "epoch": 17.0,
2895
+ "eval_loss": 0.0073828850872814655,
2896
+ "eval_runtime": 47.9949,
2897
+ "eval_samples_per_second": 8.855,
2898
+ "eval_steps_per_second": 0.563,
2899
+ "step": 459
2900
+ },
2901
+ {
2902
+ "epoch": 17.04,
2903
+ "learning_rate": 0.0002,
2904
+ "loss": 0.0117,
2905
+ "step": 460
2906
+ },
2907
+ {
2908
+ "epoch": 17.07,
2909
+ "learning_rate": 0.0002,
2910
+ "loss": 0.0019,
2911
+ "step": 461
2912
+ },
2913
+ {
2914
+ "epoch": 17.11,
2915
+ "learning_rate": 0.0002,
2916
+ "loss": 0.0051,
2917
+ "step": 462
2918
+ },
2919
+ {
2920
+ "epoch": 17.15,
2921
+ "learning_rate": 0.0002,
2922
+ "loss": 0.0194,
2923
+ "step": 463
2924
+ },
2925
+ {
2926
+ "epoch": 17.19,
2927
+ "learning_rate": 0.0002,
2928
+ "loss": 0.0076,
2929
+ "step": 464
2930
+ },
2931
+ {
2932
+ "epoch": 17.22,
2933
+ "learning_rate": 0.0002,
2934
+ "loss": 0.0027,
2935
+ "step": 465
2936
+ },
2937
+ {
2938
+ "epoch": 17.26,
2939
+ "learning_rate": 0.0002,
2940
+ "loss": 0.0027,
2941
+ "step": 466
2942
+ },
2943
+ {
2944
+ "epoch": 17.3,
2945
+ "learning_rate": 0.0002,
2946
+ "loss": 0.0058,
2947
+ "step": 467
2948
+ },
2949
+ {
2950
+ "epoch": 17.33,
2951
+ "learning_rate": 0.0002,
2952
+ "loss": 0.0014,
2953
+ "step": 468
2954
+ },
2955
+ {
2956
+ "epoch": 17.37,
2957
+ "learning_rate": 0.0002,
2958
+ "loss": 0.0065,
2959
+ "step": 469
2960
+ },
2961
+ {
2962
+ "epoch": 17.41,
2963
+ "learning_rate": 0.0002,
2964
+ "loss": 0.0028,
2965
+ "step": 470
2966
+ },
2967
+ {
2968
+ "epoch": 17.44,
2969
+ "learning_rate": 0.0002,
2970
+ "loss": 0.0026,
2971
+ "step": 471
2972
+ },
2973
+ {
2974
+ "epoch": 17.48,
2975
+ "learning_rate": 0.0002,
2976
+ "loss": 0.0149,
2977
+ "step": 472
2978
+ },
2979
+ {
2980
+ "epoch": 17.52,
2981
+ "learning_rate": 0.0002,
2982
+ "loss": 0.0217,
2983
+ "step": 473
2984
+ },
2985
+ {
2986
+ "epoch": 17.56,
2987
+ "learning_rate": 0.0002,
2988
+ "loss": 0.0039,
2989
+ "step": 474
2990
+ },
2991
+ {
2992
+ "epoch": 17.59,
2993
+ "learning_rate": 0.0002,
2994
+ "loss": 0.0013,
2995
+ "step": 475
2996
+ },
2997
+ {
2998
+ "epoch": 17.63,
2999
+ "learning_rate": 0.0002,
3000
+ "loss": 0.0052,
3001
+ "step": 476
3002
+ },
3003
+ {
3004
+ "epoch": 17.67,
3005
+ "learning_rate": 0.0002,
3006
+ "loss": 0.0012,
3007
+ "step": 477
3008
+ },
3009
+ {
3010
+ "epoch": 17.7,
3011
+ "learning_rate": 0.0002,
3012
+ "loss": 0.0021,
3013
+ "step": 478
3014
+ },
3015
+ {
3016
+ "epoch": 17.74,
3017
+ "learning_rate": 0.0002,
3018
+ "loss": 0.0014,
3019
+ "step": 479
3020
+ },
3021
+ {
3022
+ "epoch": 17.78,
3023
+ "learning_rate": 0.0002,
3024
+ "loss": 0.015,
3025
+ "step": 480
3026
+ },
3027
+ {
3028
+ "epoch": 17.81,
3029
+ "learning_rate": 0.0002,
3030
+ "loss": 0.0008,
3031
+ "step": 481
3032
+ },
3033
+ {
3034
+ "epoch": 17.85,
3035
+ "learning_rate": 0.0002,
3036
+ "loss": 0.009,
3037
+ "step": 482
3038
+ },
3039
+ {
3040
+ "epoch": 17.89,
3041
+ "learning_rate": 0.0002,
3042
+ "loss": 0.0023,
3043
+ "step": 483
3044
+ },
3045
+ {
3046
+ "epoch": 17.93,
3047
+ "learning_rate": 0.0002,
3048
+ "loss": 0.0095,
3049
+ "step": 484
3050
+ },
3051
+ {
3052
+ "epoch": 17.96,
3053
+ "learning_rate": 0.0002,
3054
+ "loss": 0.0344,
3055
+ "step": 485
3056
+ },
3057
+ {
3058
+ "epoch": 18.0,
3059
+ "learning_rate": 0.0002,
3060
+ "loss": 0.0012,
3061
+ "step": 486
3062
+ },
3063
+ {
3064
+ "epoch": 18.0,
3065
+ "eval_loss": 0.012661315500736237,
3066
+ "eval_runtime": 47.9578,
3067
+ "eval_samples_per_second": 8.862,
3068
+ "eval_steps_per_second": 0.563,
3069
+ "step": 486
3070
+ },
3071
+ {
3072
+ "epoch": 18.04,
3073
+ "learning_rate": 0.0002,
3074
+ "loss": 0.0142,
3075
+ "step": 487
3076
+ },
3077
+ {
3078
+ "epoch": 18.07,
3079
+ "learning_rate": 0.0002,
3080
+ "loss": 0.0036,
3081
+ "step": 488
3082
+ },
3083
+ {
3084
+ "epoch": 18.11,
3085
+ "learning_rate": 0.0002,
3086
+ "loss": 0.0046,
3087
+ "step": 489
3088
+ },
3089
+ {
3090
+ "epoch": 18.15,
3091
+ "learning_rate": 0.0002,
3092
+ "loss": 0.0222,
3093
+ "step": 490
3094
+ },
3095
+ {
3096
+ "epoch": 18.19,
3097
+ "learning_rate": 0.0002,
3098
+ "loss": 0.0162,
3099
+ "step": 491
3100
+ },
3101
+ {
3102
+ "epoch": 18.22,
3103
+ "learning_rate": 0.0002,
3104
+ "loss": 0.0188,
3105
+ "step": 492
3106
+ },
3107
+ {
3108
+ "epoch": 18.26,
3109
+ "learning_rate": 0.0002,
3110
+ "loss": 0.0059,
3111
+ "step": 493
3112
+ },
3113
+ {
3114
+ "epoch": 18.3,
3115
+ "learning_rate": 0.0002,
3116
+ "loss": 0.0086,
3117
+ "step": 494
3118
+ },
3119
+ {
3120
+ "epoch": 18.33,
3121
+ "learning_rate": 0.0002,
3122
+ "loss": 0.0155,
3123
+ "step": 495
3124
+ },
3125
+ {
3126
+ "epoch": 18.37,
3127
+ "learning_rate": 0.0002,
3128
+ "loss": 0.0035,
3129
+ "step": 496
3130
+ },
3131
+ {
3132
+ "epoch": 18.41,
3133
+ "learning_rate": 0.0002,
3134
+ "loss": 0.0122,
3135
+ "step": 497
3136
+ },
3137
+ {
3138
+ "epoch": 18.44,
3139
+ "learning_rate": 0.0002,
3140
+ "loss": 0.0126,
3141
+ "step": 498
3142
+ },
3143
+ {
3144
+ "epoch": 18.48,
3145
+ "learning_rate": 0.0002,
3146
+ "loss": 0.0211,
3147
+ "step": 499
3148
+ },
3149
+ {
3150
+ "epoch": 18.52,
3151
+ "learning_rate": 0.0002,
3152
+ "loss": 0.0186,
3153
+ "step": 500
3154
+ },
3155
+ {
3156
+ "epoch": 18.56,
3157
+ "learning_rate": 0.0002,
3158
+ "loss": 0.0331,
3159
+ "step": 501
3160
+ },
3161
+ {
3162
+ "epoch": 18.59,
3163
+ "learning_rate": 0.0002,
3164
+ "loss": 0.007,
3165
+ "step": 502
3166
+ },
3167
+ {
3168
+ "epoch": 18.63,
3169
+ "learning_rate": 0.0002,
3170
+ "loss": 0.0186,
3171
+ "step": 503
3172
+ },
3173
+ {
3174
+ "epoch": 18.67,
3175
+ "learning_rate": 0.0002,
3176
+ "loss": 0.0336,
3177
+ "step": 504
3178
+ },
3179
+ {
3180
+ "epoch": 18.7,
3181
+ "learning_rate": 0.0002,
3182
+ "loss": 0.0086,
3183
+ "step": 505
3184
+ },
3185
+ {
3186
+ "epoch": 18.74,
3187
+ "learning_rate": 0.0002,
3188
+ "loss": 0.0136,
3189
+ "step": 506
3190
+ },
3191
+ {
3192
+ "epoch": 18.78,
3193
+ "learning_rate": 0.0002,
3194
+ "loss": 0.0129,
3195
+ "step": 507
3196
+ },
3197
+ {
3198
+ "epoch": 18.81,
3199
+ "learning_rate": 0.0002,
3200
+ "loss": 0.0217,
3201
+ "step": 508
3202
+ },
3203
+ {
3204
+ "epoch": 18.85,
3205
+ "learning_rate": 0.0002,
3206
+ "loss": 0.0491,
3207
+ "step": 509
3208
+ },
3209
+ {
3210
+ "epoch": 18.89,
3211
+ "learning_rate": 0.0002,
3212
+ "loss": 0.0076,
3213
+ "step": 510
3214
+ },
3215
+ {
3216
+ "epoch": 18.93,
3217
+ "learning_rate": 0.0002,
3218
+ "loss": 0.0069,
3219
+ "step": 511
3220
+ },
3221
+ {
3222
+ "epoch": 18.96,
3223
+ "learning_rate": 0.0002,
3224
+ "loss": 0.0169,
3225
+ "step": 512
3226
+ },
3227
+ {
3228
+ "epoch": 19.0,
3229
+ "learning_rate": 0.0002,
3230
+ "loss": 0.0113,
3231
+ "step": 513
3232
+ },
3233
+ {
3234
+ "epoch": 19.0,
3235
+ "eval_loss": 0.014072887599468231,
3236
+ "eval_runtime": 48.0124,
3237
+ "eval_samples_per_second": 8.852,
3238
+ "eval_steps_per_second": 0.562,
3239
+ "step": 513
3240
+ },
3241
+ {
3242
+ "epoch": 19.04,
3243
+ "learning_rate": 0.0002,
3244
+ "loss": 0.0189,
3245
+ "step": 514
3246
+ },
3247
+ {
3248
+ "epoch": 19.07,
3249
+ "learning_rate": 0.0002,
3250
+ "loss": 0.0018,
3251
+ "step": 515
3252
+ },
3253
+ {
3254
+ "epoch": 19.11,
3255
+ "learning_rate": 0.0002,
3256
+ "loss": 0.0087,
3257
+ "step": 516
3258
+ },
3259
+ {
3260
+ "epoch": 19.15,
3261
+ "learning_rate": 0.0002,
3262
+ "loss": 0.0155,
3263
+ "step": 517
3264
+ },
3265
+ {
3266
+ "epoch": 19.19,
3267
+ "learning_rate": 0.0002,
3268
+ "loss": 0.0083,
3269
+ "step": 518
3270
+ },
3271
+ {
3272
+ "epoch": 19.22,
3273
+ "learning_rate": 0.0002,
3274
+ "loss": 0.0252,
3275
+ "step": 519
3276
+ },
3277
+ {
3278
+ "epoch": 19.26,
3279
+ "learning_rate": 0.0002,
3280
+ "loss": 0.0158,
3281
+ "step": 520
3282
+ },
3283
+ {
3284
+ "epoch": 19.3,
3285
+ "learning_rate": 0.0002,
3286
+ "loss": 0.0081,
3287
+ "step": 521
3288
+ },
3289
+ {
3290
+ "epoch": 19.33,
3291
+ "learning_rate": 0.0002,
3292
+ "loss": 0.0144,
3293
+ "step": 522
3294
+ },
3295
+ {
3296
+ "epoch": 19.37,
3297
+ "learning_rate": 0.0002,
3298
+ "loss": 0.0102,
3299
+ "step": 523
3300
+ },
3301
+ {
3302
+ "epoch": 19.41,
3303
+ "learning_rate": 0.0002,
3304
+ "loss": 0.0091,
3305
+ "step": 524
3306
+ },
3307
+ {
3308
+ "epoch": 19.44,
3309
+ "learning_rate": 0.0002,
3310
+ "loss": 0.009,
3311
+ "step": 525
3312
+ },
3313
+ {
3314
+ "epoch": 19.48,
3315
+ "learning_rate": 0.0002,
3316
+ "loss": 0.012,
3317
+ "step": 526
3318
+ },
3319
+ {
3320
+ "epoch": 19.52,
3321
+ "learning_rate": 0.0002,
3322
+ "loss": 0.0134,
3323
+ "step": 527
3324
+ },
3325
+ {
3326
+ "epoch": 19.56,
3327
+ "learning_rate": 0.0002,
3328
+ "loss": 0.0178,
3329
+ "step": 528
3330
+ },
3331
+ {
3332
+ "epoch": 19.59,
3333
+ "learning_rate": 0.0002,
3334
+ "loss": 0.0099,
3335
+ "step": 529
3336
+ },
3337
+ {
3338
+ "epoch": 19.63,
3339
+ "learning_rate": 0.0002,
3340
+ "loss": 0.0058,
3341
+ "step": 530
3342
+ },
3343
+ {
3344
+ "epoch": 19.67,
3345
+ "learning_rate": 0.0002,
3346
+ "loss": 0.0122,
3347
+ "step": 531
3348
+ },
3349
+ {
3350
+ "epoch": 19.7,
3351
+ "learning_rate": 0.0002,
3352
+ "loss": 0.0114,
3353
+ "step": 532
3354
+ },
3355
+ {
3356
+ "epoch": 19.74,
3357
+ "learning_rate": 0.0002,
3358
+ "loss": 0.0099,
3359
+ "step": 533
3360
+ },
3361
+ {
3362
+ "epoch": 19.78,
3363
+ "learning_rate": 0.0002,
3364
+ "loss": 0.0053,
3365
+ "step": 534
3366
+ },
3367
+ {
3368
+ "epoch": 19.81,
3369
+ "learning_rate": 0.0002,
3370
+ "loss": 0.0095,
3371
+ "step": 535
3372
+ },
3373
+ {
3374
+ "epoch": 19.85,
3375
+ "learning_rate": 0.0002,
3376
+ "loss": 0.0041,
3377
+ "step": 536
3378
+ },
3379
+ {
3380
+ "epoch": 19.89,
3381
+ "learning_rate": 0.0002,
3382
+ "loss": 0.0172,
3383
+ "step": 537
3384
+ },
3385
+ {
3386
+ "epoch": 19.93,
3387
+ "learning_rate": 0.0002,
3388
+ "loss": 0.0074,
3389
+ "step": 538
3390
+ },
3391
+ {
3392
+ "epoch": 19.96,
3393
+ "learning_rate": 0.0002,
3394
+ "loss": 0.018,
3395
+ "step": 539
3396
+ },
3397
+ {
3398
+ "epoch": 20.0,
3399
+ "learning_rate": 0.0002,
3400
+ "loss": 0.0232,
3401
+ "step": 540
3402
+ },
3403
+ {
3404
+ "epoch": 20.0,
3405
+ "eval_loss": 0.009742139838635921,
3406
+ "eval_runtime": 48.0721,
3407
+ "eval_samples_per_second": 8.841,
3408
+ "eval_steps_per_second": 0.562,
3409
+ "step": 540
3410
+ },
3411
+ {
3412
+ "epoch": 20.0,
3413
+ "step": 540,
3414
+ "total_flos": 8.143219945845555e+17,
3415
+ "train_loss": 0.04948140804567461,
3416
+ "train_runtime": 6016.1313,
3417
+ "train_samples_per_second": 1.413,
3418
+ "train_steps_per_second": 0.09
3419
+ }
3420
+ ],
3421
+ "logging_steps": 1.0,
3422
+ "max_steps": 540,
3423
+ "num_input_tokens_seen": 0,
3424
+ "num_train_epochs": 20,
3425
+ "save_steps": 50000,
3426
+ "total_flos": 8.143219945845555e+17,
3427
+ "train_batch_size": 4,
3428
+ "trial_name": null,
3429
+ "trial_params": null
3430
+ }
training_args.bin CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:a5f8e6f154ab077d6c0a1db7ffd281c4ace88c1e23119cf6949362b34391609b
3
  size 6840
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:044e9b2d025bc9ec7774e2eb37aa6fee7a0384b1e20dec78d3770c07373e38a2
3
  size 6840