Model save
Browse files- README.md +69 -0
- adapter_model.safetensors +1 -1
- all_results.json +13 -0
- eval_results.json +8 -0
- runs/Mar04_05-15-04_SYS-4029GP-TRT/events.out.tfevents.1709501213.SYS-4029GP-TRT.1795489.0 +2 -2
- runs/Mar04_05-15-04_SYS-4029GP-TRT/events.out.tfevents.1709555650.SYS-4029GP-TRT.1795489.1 +3 -0
- train_results.json +8 -0
- trainer_state.json +728 -0
README.md
ADDED
@@ -0,0 +1,69 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
---
|
2 |
+
license: llama2
|
3 |
+
library_name: peft
|
4 |
+
tags:
|
5 |
+
- trl
|
6 |
+
- sft
|
7 |
+
- generated_from_trainer
|
8 |
+
datasets:
|
9 |
+
- generator
|
10 |
+
base_model: lmsys/vicuna-7b-v1.5
|
11 |
+
model-index:
|
12 |
+
- name: vicuna-qlora
|
13 |
+
results: []
|
14 |
+
---
|
15 |
+
|
16 |
+
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
17 |
+
should probably proofread and complete it, then remove this comment. -->
|
18 |
+
|
19 |
+
# vicuna-qlora
|
20 |
+
|
21 |
+
This model is a fine-tuned version of [lmsys/vicuna-7b-v1.5](https://huggingface.co/lmsys/vicuna-7b-v1.5) on the generator dataset.
|
22 |
+
It achieves the following results on the evaluation set:
|
23 |
+
- Loss: 0.9819
|
24 |
+
|
25 |
+
## Model description
|
26 |
+
|
27 |
+
More information needed
|
28 |
+
|
29 |
+
## Intended uses & limitations
|
30 |
+
|
31 |
+
More information needed
|
32 |
+
|
33 |
+
## Training and evaluation data
|
34 |
+
|
35 |
+
More information needed
|
36 |
+
|
37 |
+
## Training procedure
|
38 |
+
|
39 |
+
### Training hyperparameters
|
40 |
+
|
41 |
+
The following hyperparameters were used during training:
|
42 |
+
- learning_rate: 0.0002
|
43 |
+
- train_batch_size: 2
|
44 |
+
- eval_batch_size: 8
|
45 |
+
- seed: 42
|
46 |
+
- distributed_type: multi-GPU
|
47 |
+
- num_devices: 4
|
48 |
+
- gradient_accumulation_steps: 32
|
49 |
+
- total_train_batch_size: 256
|
50 |
+
- total_eval_batch_size: 32
|
51 |
+
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
52 |
+
- lr_scheduler_type: cosine
|
53 |
+
- lr_scheduler_warmup_ratio: 0.1
|
54 |
+
- num_epochs: 1
|
55 |
+
|
56 |
+
### Training results
|
57 |
+
|
58 |
+
| Training Loss | Epoch | Step | Validation Loss |
|
59 |
+
|:-------------:|:-----:|:----:|:---------------:|
|
60 |
+
| 0.9771 | 1.0 | 570 | 0.9819 |
|
61 |
+
|
62 |
+
|
63 |
+
### Framework versions
|
64 |
+
|
65 |
+
- PEFT 0.7.1
|
66 |
+
- Transformers 4.36.2
|
67 |
+
- Pytorch 2.2.1+cu121
|
68 |
+
- Datasets 2.14.6
|
69 |
+
- Tokenizers 0.15.2
|
adapter_model.safetensors
CHANGED
@@ -1,3 +1,3 @@
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:
|
3 |
size 80014016
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:2521e573b10c0387a3694f2d465b5ae9b4ea4c2f0abfff575ab96540450da928
|
3 |
size 80014016
|
all_results.json
ADDED
@@ -0,0 +1,13 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"epoch": 1.0,
|
3 |
+
"eval_loss": 0.981906533241272,
|
4 |
+
"eval_runtime": 2117.4688,
|
5 |
+
"eval_samples": 23110,
|
6 |
+
"eval_samples_per_second": 7.635,
|
7 |
+
"eval_steps_per_second": 0.239,
|
8 |
+
"train_loss": 0.9983171429550438,
|
9 |
+
"train_runtime": 52319.2765,
|
10 |
+
"train_samples": 207865,
|
11 |
+
"train_samples_per_second": 2.792,
|
12 |
+
"train_steps_per_second": 0.011
|
13 |
+
}
|
eval_results.json
ADDED
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"epoch": 1.0,
|
3 |
+
"eval_loss": 0.981906533241272,
|
4 |
+
"eval_runtime": 2117.4688,
|
5 |
+
"eval_samples": 23110,
|
6 |
+
"eval_samples_per_second": 7.635,
|
7 |
+
"eval_steps_per_second": 0.239
|
8 |
+
}
|
runs/Mar04_05-15-04_SYS-4029GP-TRT/events.out.tfevents.1709501213.SYS-4029GP-TRT.1795489.0
CHANGED
@@ -1,3 +1,3 @@
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:
|
3 |
-
size
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:b4eadbfe98e66b97a5e87676b07b4deeb993d5dcf311a8550c30fee1d60cd940
|
3 |
+
size 23438
|
runs/Mar04_05-15-04_SYS-4029GP-TRT/events.out.tfevents.1709555650.SYS-4029GP-TRT.1795489.1
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:1dcf99d0c9ae44767ac870a30434913209063572cc96ae4549fc8481241ce9a0
|
3 |
+
size 359
|
train_results.json
ADDED
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"epoch": 1.0,
|
3 |
+
"train_loss": 0.9983171429550438,
|
4 |
+
"train_runtime": 52319.2765,
|
5 |
+
"train_samples": 207865,
|
6 |
+
"train_samples_per_second": 2.792,
|
7 |
+
"train_steps_per_second": 0.011
|
8 |
+
}
|
trainer_state.json
ADDED
@@ -0,0 +1,728 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"best_metric": null,
|
3 |
+
"best_model_checkpoint": null,
|
4 |
+
"epoch": 0.999068850303993,
|
5 |
+
"eval_steps": 500,
|
6 |
+
"global_step": 570,
|
7 |
+
"is_hyper_param_search": false,
|
8 |
+
"is_local_process_zero": true,
|
9 |
+
"is_world_process_zero": true,
|
10 |
+
"log_history": [
|
11 |
+
{
|
12 |
+
"epoch": 0.0,
|
13 |
+
"learning_rate": 3.5087719298245615e-06,
|
14 |
+
"loss": 1.2582,
|
15 |
+
"step": 1
|
16 |
+
},
|
17 |
+
{
|
18 |
+
"epoch": 0.01,
|
19 |
+
"learning_rate": 1.7543859649122806e-05,
|
20 |
+
"loss": 1.2488,
|
21 |
+
"step": 5
|
22 |
+
},
|
23 |
+
{
|
24 |
+
"epoch": 0.02,
|
25 |
+
"learning_rate": 3.508771929824561e-05,
|
26 |
+
"loss": 1.2359,
|
27 |
+
"step": 10
|
28 |
+
},
|
29 |
+
{
|
30 |
+
"epoch": 0.03,
|
31 |
+
"learning_rate": 5.2631578947368424e-05,
|
32 |
+
"loss": 1.2016,
|
33 |
+
"step": 15
|
34 |
+
},
|
35 |
+
{
|
36 |
+
"epoch": 0.04,
|
37 |
+
"learning_rate": 7.017543859649122e-05,
|
38 |
+
"loss": 1.155,
|
39 |
+
"step": 20
|
40 |
+
},
|
41 |
+
{
|
42 |
+
"epoch": 0.04,
|
43 |
+
"learning_rate": 8.771929824561403e-05,
|
44 |
+
"loss": 1.1231,
|
45 |
+
"step": 25
|
46 |
+
},
|
47 |
+
{
|
48 |
+
"epoch": 0.05,
|
49 |
+
"learning_rate": 0.00010526315789473685,
|
50 |
+
"loss": 1.1019,
|
51 |
+
"step": 30
|
52 |
+
},
|
53 |
+
{
|
54 |
+
"epoch": 0.06,
|
55 |
+
"learning_rate": 0.00012280701754385965,
|
56 |
+
"loss": 1.0873,
|
57 |
+
"step": 35
|
58 |
+
},
|
59 |
+
{
|
60 |
+
"epoch": 0.07,
|
61 |
+
"learning_rate": 0.00014035087719298245,
|
62 |
+
"loss": 1.0644,
|
63 |
+
"step": 40
|
64 |
+
},
|
65 |
+
{
|
66 |
+
"epoch": 0.08,
|
67 |
+
"learning_rate": 0.00015789473684210527,
|
68 |
+
"loss": 1.0655,
|
69 |
+
"step": 45
|
70 |
+
},
|
71 |
+
{
|
72 |
+
"epoch": 0.09,
|
73 |
+
"learning_rate": 0.00017543859649122806,
|
74 |
+
"loss": 1.0574,
|
75 |
+
"step": 50
|
76 |
+
},
|
77 |
+
{
|
78 |
+
"epoch": 0.1,
|
79 |
+
"learning_rate": 0.00019298245614035088,
|
80 |
+
"loss": 1.0447,
|
81 |
+
"step": 55
|
82 |
+
},
|
83 |
+
{
|
84 |
+
"epoch": 0.11,
|
85 |
+
"learning_rate": 0.00019998312416333227,
|
86 |
+
"loss": 1.0356,
|
87 |
+
"step": 60
|
88 |
+
},
|
89 |
+
{
|
90 |
+
"epoch": 0.11,
|
91 |
+
"learning_rate": 0.0001998800146766861,
|
92 |
+
"loss": 1.0223,
|
93 |
+
"step": 65
|
94 |
+
},
|
95 |
+
{
|
96 |
+
"epoch": 0.12,
|
97 |
+
"learning_rate": 0.00019968326771610797,
|
98 |
+
"loss": 1.031,
|
99 |
+
"step": 70
|
100 |
+
},
|
101 |
+
{
|
102 |
+
"epoch": 0.13,
|
103 |
+
"learning_rate": 0.00019939306773179497,
|
104 |
+
"loss": 1.0194,
|
105 |
+
"step": 75
|
106 |
+
},
|
107 |
+
{
|
108 |
+
"epoch": 0.14,
|
109 |
+
"learning_rate": 0.00019900968678611666,
|
110 |
+
"loss": 1.0126,
|
111 |
+
"step": 80
|
112 |
+
},
|
113 |
+
{
|
114 |
+
"epoch": 0.15,
|
115 |
+
"learning_rate": 0.00019853348429855672,
|
116 |
+
"loss": 1.0082,
|
117 |
+
"step": 85
|
118 |
+
},
|
119 |
+
{
|
120 |
+
"epoch": 0.16,
|
121 |
+
"learning_rate": 0.0001979649067087574,
|
122 |
+
"loss": 1.0099,
|
123 |
+
"step": 90
|
124 |
+
},
|
125 |
+
{
|
126 |
+
"epoch": 0.17,
|
127 |
+
"learning_rate": 0.00019730448705798239,
|
128 |
+
"loss": 1.0018,
|
129 |
+
"step": 95
|
130 |
+
},
|
131 |
+
{
|
132 |
+
"epoch": 0.18,
|
133 |
+
"learning_rate": 0.00019655284448939094,
|
134 |
+
"loss": 1.0068,
|
135 |
+
"step": 100
|
136 |
+
},
|
137 |
+
{
|
138 |
+
"epoch": 0.18,
|
139 |
+
"learning_rate": 0.00019571068366759143,
|
140 |
+
"loss": 1.0011,
|
141 |
+
"step": 105
|
142 |
+
},
|
143 |
+
{
|
144 |
+
"epoch": 0.19,
|
145 |
+
"learning_rate": 0.00019477879411801844,
|
146 |
+
"loss": 1.0074,
|
147 |
+
"step": 110
|
148 |
+
},
|
149 |
+
{
|
150 |
+
"epoch": 0.2,
|
151 |
+
"learning_rate": 0.00019375804948675306,
|
152 |
+
"loss": 0.9902,
|
153 |
+
"step": 115
|
154 |
+
},
|
155 |
+
{
|
156 |
+
"epoch": 0.21,
|
157 |
+
"learning_rate": 0.00019264940672148018,
|
158 |
+
"loss": 0.996,
|
159 |
+
"step": 120
|
160 |
+
},
|
161 |
+
{
|
162 |
+
"epoch": 0.22,
|
163 |
+
"learning_rate": 0.00019145390517435012,
|
164 |
+
"loss": 1.0023,
|
165 |
+
"step": 125
|
166 |
+
},
|
167 |
+
{
|
168 |
+
"epoch": 0.23,
|
169 |
+
"learning_rate": 0.00019017266562758659,
|
170 |
+
"loss": 1.0005,
|
171 |
+
"step": 130
|
172 |
+
},
|
173 |
+
{
|
174 |
+
"epoch": 0.24,
|
175 |
+
"learning_rate": 0.00018880688924275378,
|
176 |
+
"loss": 0.9882,
|
177 |
+
"step": 135
|
178 |
+
},
|
179 |
+
{
|
180 |
+
"epoch": 0.25,
|
181 |
+
"learning_rate": 0.00018735785643466784,
|
182 |
+
"loss": 0.9905,
|
183 |
+
"step": 140
|
184 |
+
},
|
185 |
+
{
|
186 |
+
"epoch": 0.25,
|
187 |
+
"learning_rate": 0.00018582692567100867,
|
188 |
+
"loss": 0.9949,
|
189 |
+
"step": 145
|
190 |
+
},
|
191 |
+
{
|
192 |
+
"epoch": 0.26,
|
193 |
+
"learning_rate": 0.00018421553219875658,
|
194 |
+
"loss": 0.9976,
|
195 |
+
"step": 150
|
196 |
+
},
|
197 |
+
{
|
198 |
+
"epoch": 0.27,
|
199 |
+
"learning_rate": 0.00018252518669864936,
|
200 |
+
"loss": 0.9847,
|
201 |
+
"step": 155
|
202 |
+
},
|
203 |
+
{
|
204 |
+
"epoch": 0.28,
|
205 |
+
"learning_rate": 0.0001807574738689193,
|
206 |
+
"loss": 0.9881,
|
207 |
+
"step": 160
|
208 |
+
},
|
209 |
+
{
|
210 |
+
"epoch": 0.29,
|
211 |
+
"learning_rate": 0.00017891405093963938,
|
212 |
+
"loss": 0.9812,
|
213 |
+
"step": 165
|
214 |
+
},
|
215 |
+
{
|
216 |
+
"epoch": 0.3,
|
217 |
+
"learning_rate": 0.00017699664611907072,
|
218 |
+
"loss": 1.0059,
|
219 |
+
"step": 170
|
220 |
+
},
|
221 |
+
{
|
222 |
+
"epoch": 0.31,
|
223 |
+
"learning_rate": 0.0001750070569734681,
|
224 |
+
"loss": 0.9895,
|
225 |
+
"step": 175
|
226 |
+
},
|
227 |
+
{
|
228 |
+
"epoch": 0.32,
|
229 |
+
"learning_rate": 0.0001729471487418621,
|
230 |
+
"loss": 0.9902,
|
231 |
+
"step": 180
|
232 |
+
},
|
233 |
+
{
|
234 |
+
"epoch": 0.32,
|
235 |
+
"learning_rate": 0.00017081885258739846,
|
236 |
+
"loss": 0.9846,
|
237 |
+
"step": 185
|
238 |
+
},
|
239 |
+
{
|
240 |
+
"epoch": 0.33,
|
241 |
+
"learning_rate": 0.0001686241637868734,
|
242 |
+
"loss": 0.9848,
|
243 |
+
"step": 190
|
244 |
+
},
|
245 |
+
{
|
246 |
+
"epoch": 0.34,
|
247 |
+
"learning_rate": 0.00016636513986016213,
|
248 |
+
"loss": 0.9778,
|
249 |
+
"step": 195
|
250 |
+
},
|
251 |
+
{
|
252 |
+
"epoch": 0.35,
|
253 |
+
"learning_rate": 0.00016404389864129533,
|
254 |
+
"loss": 0.9812,
|
255 |
+
"step": 200
|
256 |
+
},
|
257 |
+
{
|
258 |
+
"epoch": 0.36,
|
259 |
+
"learning_rate": 0.00016166261629298995,
|
260 |
+
"loss": 1.0017,
|
261 |
+
"step": 205
|
262 |
+
},
|
263 |
+
{
|
264 |
+
"epoch": 0.37,
|
265 |
+
"learning_rate": 0.00015922352526649803,
|
266 |
+
"loss": 0.9853,
|
267 |
+
"step": 210
|
268 |
+
},
|
269 |
+
{
|
270 |
+
"epoch": 0.38,
|
271 |
+
"learning_rate": 0.00015672891220868432,
|
272 |
+
"loss": 0.9849,
|
273 |
+
"step": 215
|
274 |
+
},
|
275 |
+
{
|
276 |
+
"epoch": 0.39,
|
277 |
+
"learning_rate": 0.00015418111581829574,
|
278 |
+
"loss": 0.9827,
|
279 |
+
"step": 220
|
280 |
+
},
|
281 |
+
{
|
282 |
+
"epoch": 0.39,
|
283 |
+
"learning_rate": 0.00015158252465343242,
|
284 |
+
"loss": 0.9908,
|
285 |
+
"step": 225
|
286 |
+
},
|
287 |
+
{
|
288 |
+
"epoch": 0.4,
|
289 |
+
"learning_rate": 0.00014893557489227517,
|
290 |
+
"loss": 0.9773,
|
291 |
+
"step": 230
|
292 |
+
},
|
293 |
+
{
|
294 |
+
"epoch": 0.41,
|
295 |
+
"learning_rate": 0.00014624274804916958,
|
296 |
+
"loss": 0.988,
|
297 |
+
"step": 235
|
298 |
+
},
|
299 |
+
{
|
300 |
+
"epoch": 0.42,
|
301 |
+
"learning_rate": 0.00014350656864820733,
|
302 |
+
"loss": 0.9897,
|
303 |
+
"step": 240
|
304 |
+
},
|
305 |
+
{
|
306 |
+
"epoch": 0.43,
|
307 |
+
"learning_rate": 0.00014072960185648577,
|
308 |
+
"loss": 0.9757,
|
309 |
+
"step": 245
|
310 |
+
},
|
311 |
+
{
|
312 |
+
"epoch": 0.44,
|
313 |
+
"learning_rate": 0.00013791445107926478,
|
314 |
+
"loss": 0.9731,
|
315 |
+
"step": 250
|
316 |
+
},
|
317 |
+
{
|
318 |
+
"epoch": 0.45,
|
319 |
+
"learning_rate": 0.00013506375551927547,
|
320 |
+
"loss": 0.9815,
|
321 |
+
"step": 255
|
322 |
+
},
|
323 |
+
{
|
324 |
+
"epoch": 0.46,
|
325 |
+
"learning_rate": 0.00013218018770246858,
|
326 |
+
"loss": 0.9755,
|
327 |
+
"step": 260
|
328 |
+
},
|
329 |
+
{
|
330 |
+
"epoch": 0.46,
|
331 |
+
"learning_rate": 0.0001292664509725226,
|
332 |
+
"loss": 0.9746,
|
333 |
+
"step": 265
|
334 |
+
},
|
335 |
+
{
|
336 |
+
"epoch": 0.47,
|
337 |
+
"learning_rate": 0.00012632527695645993,
|
338 |
+
"loss": 0.9744,
|
339 |
+
"step": 270
|
340 |
+
},
|
341 |
+
{
|
342 |
+
"epoch": 0.48,
|
343 |
+
"learning_rate": 0.00012335942300374788,
|
344 |
+
"loss": 0.9807,
|
345 |
+
"step": 275
|
346 |
+
},
|
347 |
+
{
|
348 |
+
"epoch": 0.49,
|
349 |
+
"learning_rate": 0.00012037166960128443,
|
350 |
+
"loss": 0.9888,
|
351 |
+
"step": 280
|
352 |
+
},
|
353 |
+
{
|
354 |
+
"epoch": 0.5,
|
355 |
+
"learning_rate": 0.00011736481776669306,
|
356 |
+
"loss": 0.9848,
|
357 |
+
"step": 285
|
358 |
+
},
|
359 |
+
{
|
360 |
+
"epoch": 0.51,
|
361 |
+
"learning_rate": 0.00011434168642236964,
|
362 |
+
"loss": 0.9797,
|
363 |
+
"step": 290
|
364 |
+
},
|
365 |
+
{
|
366 |
+
"epoch": 0.52,
|
367 |
+
"learning_rate": 0.00011130510975274409,
|
368 |
+
"loss": 0.9799,
|
369 |
+
"step": 295
|
370 |
+
},
|
371 |
+
{
|
372 |
+
"epoch": 0.53,
|
373 |
+
"learning_rate": 0.00010825793454723325,
|
374 |
+
"loss": 0.9939,
|
375 |
+
"step": 300
|
376 |
+
},
|
377 |
+
{
|
378 |
+
"epoch": 0.53,
|
379 |
+
"learning_rate": 0.00010520301753137724,
|
380 |
+
"loss": 0.9693,
|
381 |
+
"step": 305
|
382 |
+
},
|
383 |
+
{
|
384 |
+
"epoch": 0.54,
|
385 |
+
"learning_rate": 0.00010214322268866032,
|
386 |
+
"loss": 0.9846,
|
387 |
+
"step": 310
|
388 |
+
},
|
389 |
+
{
|
390 |
+
"epoch": 0.55,
|
391 |
+
"learning_rate": 9.908141857552737e-05,
|
392 |
+
"loss": 0.9754,
|
393 |
+
"step": 315
|
394 |
+
},
|
395 |
+
{
|
396 |
+
"epoch": 0.56,
|
397 |
+
"learning_rate": 9.602047563211359e-05,
|
398 |
+
"loss": 0.9912,
|
399 |
+
"step": 320
|
400 |
+
},
|
401 |
+
{
|
402 |
+
"epoch": 0.57,
|
403 |
+
"learning_rate": 9.296326349120785e-05,
|
404 |
+
"loss": 0.9778,
|
405 |
+
"step": 325
|
406 |
+
},
|
407 |
+
{
|
408 |
+
"epoch": 0.58,
|
409 |
+
"learning_rate": 8.991264828797319e-05,
|
410 |
+
"loss": 0.9828,
|
411 |
+
"step": 330
|
412 |
+
},
|
413 |
+
{
|
414 |
+
"epoch": 0.59,
|
415 |
+
"learning_rate": 8.687148997294621e-05,
|
416 |
+
"loss": 0.9717,
|
417 |
+
"step": 335
|
418 |
+
},
|
419 |
+
{
|
420 |
+
"epoch": 0.6,
|
421 |
+
"learning_rate": 8.384263963083453e-05,
|
422 |
+
"loss": 0.9907,
|
423 |
+
"step": 340
|
424 |
+
},
|
425 |
+
{
|
426 |
+
"epoch": 0.6,
|
427 |
+
"learning_rate": 8.082893680762619e-05,
|
428 |
+
"loss": 0.9787,
|
429 |
+
"step": 345
|
430 |
+
},
|
431 |
+
{
|
432 |
+
"epoch": 0.61,
|
433 |
+
"learning_rate": 7.783320684851614e-05,
|
434 |
+
"loss": 0.9793,
|
435 |
+
"step": 350
|
436 |
+
},
|
437 |
+
{
|
438 |
+
"epoch": 0.62,
|
439 |
+
"learning_rate": 7.485825824914659e-05,
|
440 |
+
"loss": 0.9698,
|
441 |
+
"step": 355
|
442 |
+
},
|
443 |
+
{
|
444 |
+
"epoch": 0.63,
|
445 |
+
"learning_rate": 7.190688002264308e-05,
|
446 |
+
"loss": 0.9792,
|
447 |
+
"step": 360
|
448 |
+
},
|
449 |
+
{
|
450 |
+
"epoch": 0.64,
|
451 |
+
"learning_rate": 6.898183908491617e-05,
|
452 |
+
"loss": 0.9821,
|
453 |
+
"step": 365
|
454 |
+
},
|
455 |
+
{
|
456 |
+
"epoch": 0.65,
|
457 |
+
"learning_rate": 6.608587766067852e-05,
|
458 |
+
"loss": 0.9796,
|
459 |
+
"step": 370
|
460 |
+
},
|
461 |
+
{
|
462 |
+
"epoch": 0.66,
|
463 |
+
"learning_rate": 6.322171071261071e-05,
|
464 |
+
"loss": 0.9832,
|
465 |
+
"step": 375
|
466 |
+
},
|
467 |
+
{
|
468 |
+
"epoch": 0.67,
|
469 |
+
"learning_rate": 6.039202339608432e-05,
|
470 |
+
"loss": 0.9853,
|
471 |
+
"step": 380
|
472 |
+
},
|
473 |
+
{
|
474 |
+
"epoch": 0.67,
|
475 |
+
"learning_rate": 5.7599468541830356e-05,
|
476 |
+
"loss": 0.9659,
|
477 |
+
"step": 385
|
478 |
+
},
|
479 |
+
{
|
480 |
+
"epoch": 0.68,
|
481 |
+
"learning_rate": 5.484666416891109e-05,
|
482 |
+
"loss": 0.971,
|
483 |
+
"step": 390
|
484 |
+
},
|
485 |
+
{
|
486 |
+
"epoch": 0.69,
|
487 |
+
"learning_rate": 5.2136191030328455e-05,
|
488 |
+
"loss": 0.9705,
|
489 |
+
"step": 395
|
490 |
+
},
|
491 |
+
{
|
492 |
+
"epoch": 0.7,
|
493 |
+
"learning_rate": 4.9470590193569044e-05,
|
494 |
+
"loss": 0.9829,
|
495 |
+
"step": 400
|
496 |
+
},
|
497 |
+
{
|
498 |
+
"epoch": 0.71,
|
499 |
+
"learning_rate": 4.685236065835443e-05,
|
500 |
+
"loss": 0.9638,
|
501 |
+
"step": 405
|
502 |
+
},
|
503 |
+
{
|
504 |
+
"epoch": 0.72,
|
505 |
+
"learning_rate": 4.4283957013829846e-05,
|
506 |
+
"loss": 0.9592,
|
507 |
+
"step": 410
|
508 |
+
},
|
509 |
+
{
|
510 |
+
"epoch": 0.73,
|
511 |
+
"learning_rate": 4.176778713738787e-05,
|
512 |
+
"loss": 0.9758,
|
513 |
+
"step": 415
|
514 |
+
},
|
515 |
+
{
|
516 |
+
"epoch": 0.74,
|
517 |
+
"learning_rate": 3.9306209937284346e-05,
|
518 |
+
"loss": 0.9843,
|
519 |
+
"step": 420
|
520 |
+
},
|
521 |
+
{
|
522 |
+
"epoch": 0.74,
|
523 |
+
"learning_rate": 3.69015331411628e-05,
|
524 |
+
"loss": 0.979,
|
525 |
+
"step": 425
|
526 |
+
},
|
527 |
+
{
|
528 |
+
"epoch": 0.75,
|
529 |
+
"learning_rate": 3.455601113256073e-05,
|
530 |
+
"loss": 0.9816,
|
531 |
+
"step": 430
|
532 |
+
},
|
533 |
+
{
|
534 |
+
"epoch": 0.76,
|
535 |
+
"learning_rate": 3.227184283742591e-05,
|
536 |
+
"loss": 0.9885,
|
537 |
+
"step": 435
|
538 |
+
},
|
539 |
+
{
|
540 |
+
"epoch": 0.77,
|
541 |
+
"learning_rate": 3.0051169662624225e-05,
|
542 |
+
"loss": 0.9757,
|
543 |
+
"step": 440
|
544 |
+
},
|
545 |
+
{
|
546 |
+
"epoch": 0.78,
|
547 |
+
"learning_rate": 2.789607348837153e-05,
|
548 |
+
"loss": 0.9678,
|
549 |
+
"step": 445
|
550 |
+
},
|
551 |
+
{
|
552 |
+
"epoch": 0.79,
|
553 |
+
"learning_rate": 2.5808574716471856e-05,
|
554 |
+
"loss": 0.9757,
|
555 |
+
"step": 450
|
556 |
+
},
|
557 |
+
{
|
558 |
+
"epoch": 0.8,
|
559 |
+
"learning_rate": 2.379063037619146e-05,
|
560 |
+
"loss": 0.979,
|
561 |
+
"step": 455
|
562 |
+
},
|
563 |
+
{
|
564 |
+
"epoch": 0.81,
|
565 |
+
"learning_rate": 2.184413228954468e-05,
|
566 |
+
"loss": 0.9789,
|
567 |
+
"step": 460
|
568 |
+
},
|
569 |
+
{
|
570 |
+
"epoch": 0.82,
|
571 |
+
"learning_rate": 1.9970905297711606e-05,
|
572 |
+
"loss": 0.9728,
|
573 |
+
"step": 465
|
574 |
+
},
|
575 |
+
{
|
576 |
+
"epoch": 0.82,
|
577 |
+
"learning_rate": 1.8172705550250092e-05,
|
578 |
+
"loss": 0.9863,
|
579 |
+
"step": 470
|
580 |
+
},
|
581 |
+
{
|
582 |
+
"epoch": 0.83,
|
583 |
+
"learning_rate": 1.6451218858706374e-05,
|
584 |
+
"loss": 0.9887,
|
585 |
+
"step": 475
|
586 |
+
},
|
587 |
+
{
|
588 |
+
"epoch": 0.84,
|
589 |
+
"learning_rate": 1.4808059116167305e-05,
|
590 |
+
"loss": 0.9824,
|
591 |
+
"step": 480
|
592 |
+
},
|
593 |
+
{
|
594 |
+
"epoch": 0.85,
|
595 |
+
"learning_rate": 1.3244766784236307e-05,
|
596 |
+
"loss": 0.9743,
|
597 |
+
"step": 485
|
598 |
+
},
|
599 |
+
{
|
600 |
+
"epoch": 0.86,
|
601 |
+
"learning_rate": 1.176280744885121e-05,
|
602 |
+
"loss": 0.9859,
|
603 |
+
"step": 490
|
604 |
+
},
|
605 |
+
{
|
606 |
+
"epoch": 0.87,
|
607 |
+
"learning_rate": 1.0363570446297999e-05,
|
608 |
+
"loss": 0.9732,
|
609 |
+
"step": 495
|
610 |
+
},
|
611 |
+
{
|
612 |
+
"epoch": 0.88,
|
613 |
+
"learning_rate": 9.048367560708604e-06,
|
614 |
+
"loss": 0.9768,
|
615 |
+
"step": 500
|
616 |
+
},
|
617 |
+
{
|
618 |
+
"epoch": 0.89,
|
619 |
+
"learning_rate": 7.818431794263836e-06,
|
620 |
+
"loss": 0.9708,
|
621 |
+
"step": 505
|
622 |
+
},
|
623 |
+
{
|
624 |
+
"epoch": 0.89,
|
625 |
+
"learning_rate": 6.674916211254289e-06,
|
626 |
+
"loss": 0.9679,
|
627 |
+
"step": 510
|
628 |
+
},
|
629 |
+
{
|
630 |
+
"epoch": 0.9,
|
631 |
+
"learning_rate": 5.618892857083069e-06,
|
632 |
+
"loss": 0.9739,
|
633 |
+
"step": 515
|
634 |
+
},
|
635 |
+
{
|
636 |
+
"epoch": 0.91,
|
637 |
+
"learning_rate": 4.65135175322361e-06,
|
638 |
+
"loss": 0.9795,
|
639 |
+
"step": 520
|
640 |
+
},
|
641 |
+
{
|
642 |
+
"epoch": 0.92,
|
643 |
+
"learning_rate": 3.7731999690749585e-06,
|
644 |
+
"loss": 0.9785,
|
645 |
+
"step": 525
|
646 |
+
},
|
647 |
+
{
|
648 |
+
"epoch": 0.93,
|
649 |
+
"learning_rate": 2.9852607715846193e-06,
|
650 |
+
"loss": 0.9904,
|
651 |
+
"step": 530
|
652 |
+
},
|
653 |
+
{
|
654 |
+
"epoch": 0.94,
|
655 |
+
"learning_rate": 2.288272853436013e-06,
|
656 |
+
"loss": 0.9778,
|
657 |
+
"step": 535
|
658 |
+
},
|
659 |
+
{
|
660 |
+
"epoch": 0.95,
|
661 |
+
"learning_rate": 1.6828896405244988e-06,
|
662 |
+
"loss": 0.972,
|
663 |
+
"step": 540
|
664 |
+
},
|
665 |
+
{
|
666 |
+
"epoch": 0.96,
|
667 |
+
"learning_rate": 1.1696786793707781e-06,
|
668 |
+
"loss": 0.9647,
|
669 |
+
"step": 545
|
670 |
+
},
|
671 |
+
{
|
672 |
+
"epoch": 0.96,
|
673 |
+
"learning_rate": 7.491211050462798e-07,
|
674 |
+
"loss": 0.9865,
|
675 |
+
"step": 550
|
676 |
+
},
|
677 |
+
{
|
678 |
+
"epoch": 0.97,
|
679 |
+
"learning_rate": 4.216111901092501e-07,
|
680 |
+
"loss": 0.9742,
|
681 |
+
"step": 555
|
682 |
+
},
|
683 |
+
{
|
684 |
+
"epoch": 0.98,
|
685 |
+
"learning_rate": 1.8745597497433765e-07,
|
686 |
+
"loss": 0.9649,
|
687 |
+
"step": 560
|
688 |
+
},
|
689 |
+
{
|
690 |
+
"epoch": 0.99,
|
691 |
+
"learning_rate": 4.687498006236135e-08,
|
692 |
+
"loss": 0.9745,
|
693 |
+
"step": 565
|
694 |
+
},
|
695 |
+
{
|
696 |
+
"epoch": 1.0,
|
697 |
+
"learning_rate": 0.0,
|
698 |
+
"loss": 0.9771,
|
699 |
+
"step": 570
|
700 |
+
},
|
701 |
+
{
|
702 |
+
"epoch": 1.0,
|
703 |
+
"eval_loss": 0.981906533241272,
|
704 |
+
"eval_runtime": 2107.2433,
|
705 |
+
"eval_samples_per_second": 7.672,
|
706 |
+
"eval_steps_per_second": 0.24,
|
707 |
+
"step": 570
|
708 |
+
},
|
709 |
+
{
|
710 |
+
"epoch": 1.0,
|
711 |
+
"step": 570,
|
712 |
+
"total_flos": 1.1919078088043921e+19,
|
713 |
+
"train_loss": 0.9983171429550438,
|
714 |
+
"train_runtime": 52319.2765,
|
715 |
+
"train_samples_per_second": 2.792,
|
716 |
+
"train_steps_per_second": 0.011
|
717 |
+
}
|
718 |
+
],
|
719 |
+
"logging_steps": 5,
|
720 |
+
"max_steps": 570,
|
721 |
+
"num_input_tokens_seen": 0,
|
722 |
+
"num_train_epochs": 1,
|
723 |
+
"save_steps": 100,
|
724 |
+
"total_flos": 1.1919078088043921e+19,
|
725 |
+
"train_batch_size": 2,
|
726 |
+
"trial_name": null,
|
727 |
+
"trial_params": null
|
728 |
+
}
|