Hugging Face's logo

augustocsc
/
Se124M10KInfPrompt

PEFT
Safetensors
Generated from Trainer
Model card Files
xet
Community
augustocsc commited on
Commit
bb1886f
·
verified ·
1 Parent(s): 783e782

Model save

Browse files
Files changed (4) hide show
  1. README.md +54 -56
  2. adapter_model.safetensors +1 -1
  3. all_results.json +6 -6
  4. train_results.json +6 -6
README.md CHANGED
@@ -5,18 +5,18 @@ base_model: gpt2
5
  tags:
6
  - generated_from_trainer
7
  model-index:
8
- - name: Se124M10KInfPrompt_endtoken_ls
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
- # Se124M10KInfPrompt_endtoken_ls
16
 
17
  This model is a fine-tuned version of [gpt2](https://huggingface.co/gpt2) on an unknown dataset.
18
  It achieves the following results on the evaluation set:
19
- - Loss: 2.0494
20
 
21
  ## Model description
22
 
@@ -35,68 +35,66 @@ More information needed
35
  ### Training hyperparameters
36
 
37
  The following hyperparameters were used during training:
38
- - learning_rate: 0.0005
39
- - train_batch_size: 4
40
- - eval_batch_size: 8
41
  - seed: 42
42
- - gradient_accumulation_steps: 8
43
- - total_train_batch_size: 32
44
  - optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
45
- - lr_scheduler_type: cosine
46
- - lr_scheduler_warmup_steps: 200
47
  - num_epochs: 50
48
  - mixed_precision_training: Native AMP
49
- - label_smoothing_factor: 0.1
50
 
51
  ### Training results
52
 
53
  | Training Loss | Epoch | Step | Validation Loss |
54
  |:-------------:|:-----:|:-----:|:---------------:|
55
- | 19.0863 | 1.0 | 267 | 2.1942 |
56
- | 17.6413 | 2.0 | 534 | 2.1318 |
57
- | 17.3454 | 3.0 | 801 | 2.1143 |
58
- | 17.2455 | 4.0 | 1068 | 2.0979 |
59
- | 17.112 | 5.0 | 1335 | 2.0918 |
60
- | 17.0311 | 6.0 | 1602 | 2.0852 |
61
- | 16.9714 | 7.0 | 1869 | 2.0805 |
62
- | 16.8883 | 8.0 | 2136 | 2.0760 |
63
- | 16.8675 | 9.0 | 2403 | 2.0727 |
64
- | 16.8491 | 10.0 | 2670 | 2.0699 |
65
- | 16.8653 | 11.0 | 2937 | 2.0698 |
66
- | 16.7795 | 12.0 | 3204 | 2.0718 |
67
- | 16.8033 | 13.0 | 3471 | 2.0635 |
68
- | 16.7715 | 14.0 | 3738 | 2.0644 |
69
- | 16.7677 | 15.0 | 4005 | 2.0632 |
70
- | 16.7682 | 16.0 | 4272 | 2.0615 |
71
- | 16.7473 | 17.0 | 4539 | 2.0598 |
72
- | 16.7306 | 18.0 | 4806 | 2.0615 |
73
- | 16.6896 | 19.0 | 5073 | 2.0586 |
74
- | 16.7027 | 20.0 | 5340 | 2.0589 |
75
- | 16.6991 | 21.0 | 5607 | 2.0581 |
76
- | 16.6864 | 22.0 | 5874 | 2.0573 |
77
- | 16.6749 | 23.0 | 6141 | 2.0562 |
78
- | 16.6714 | 24.0 | 6408 | 2.0551 |
79
- | 16.6603 | 25.0 | 6675 | 2.0546 |
80
- | 16.6801 | 26.0 | 6942 | 2.0542 |
81
- | 16.6263 | 27.0 | 7209 | 2.0541 |
82
- | 16.6436 | 28.0 | 7476 | 2.0531 |
83
- | 16.6471 | 29.0 | 7743 | 2.0523 |
84
- | 16.6412 | 30.0 | 8010 | 2.0549 |
85
- | 16.6017 | 31.0 | 8277 | 2.0529 |
86
- | 16.6352 | 32.0 | 8544 | 2.0510 |
87
- | 16.5937 | 33.0 | 8811 | 2.0522 |
88
- | 16.6165 | 34.0 | 9078 | 2.0511 |
89
- | 16.5961 | 35.0 | 9345 | 2.0518 |
90
- | 16.5675 | 36.0 | 9612 | 2.0514 |
91
- | 16.5565 | 37.0 | 9879 | 2.0499 |
92
- | 16.6215 | 38.0 | 10146 | 2.0504 |
93
- | 16.6133 | 39.0 | 10413 | 2.0505 |
94
- | 16.5901 | 40.0 | 10680 | 2.0492 |
95
- | 16.5841 | 41.0 | 10947 | 2.0500 |
96
- | 16.5856 | 42.0 | 11214 | 2.0493 |
97
- | 16.5775 | 43.0 | 11481 | 2.0494 |
98
- | 16.5873 | 44.0 | 11748 | 2.0497 |
99
- | 16.5285 | 45.0 | 12015 | 2.0494 |
 
 
100
 
101
 
102
  ### Framework versions
 
5
  tags:
6
  - generated_from_trainer
7
  model-index:
8
+ - name: Se124M10KInfPrompt
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
+ # Se124M10KInfPrompt
16
 
17
  This model is a fine-tuned version of [gpt2](https://huggingface.co/gpt2) on an unknown dataset.
18
  It achieves the following results on the evaluation set:
19
+ - Loss: 0.7133
20
 
21
  ## Model description
22
 
 
35
  ### Training hyperparameters
36
 
37
  The following hyperparameters were used during training:
38
+ - learning_rate: 5e-05
39
+ - train_batch_size: 32
40
+ - eval_batch_size: 32
41
  - seed: 42
 
 
42
  - optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
43
+ - lr_scheduler_type: linear
 
44
  - num_epochs: 50
45
  - mixed_precision_training: Native AMP
 
46
 
47
  ### Training results
48
 
49
  | Training Loss | Epoch | Step | Validation Loss |
50
  |:-------------:|:-----:|:-----:|:---------------:|
51
+ | 0.4014 | 1.0 | 267 | 1.0141 |
52
+ | 0.2422 | 2.0 | 534 | 0.8523 |
53
+ | 0.2202 | 3.0 | 801 | 0.8168 |
54
+ | 0.2129 | 4.0 | 1068 | 0.7993 |
55
+ | 0.2059 | 5.0 | 1335 | 0.7837 |
56
+ | 0.2041 | 6.0 | 1602 | 0.7695 |
57
+ | 0.2031 | 7.0 | 1869 | 0.7635 |
58
+ | 0.1982 | 8.0 | 2136 | 0.7586 |
59
+ | 0.1975 | 9.0 | 2403 | 0.7532 |
60
+ | 0.1974 | 10.0 | 2670 | 0.7483 |
61
+ | 0.1978 | 11.0 | 2937 | 0.7467 |
62
+ | 0.1939 | 12.0 | 3204 | 0.7445 |
63
+ | 0.1953 | 13.0 | 3471 | 0.7439 |
64
+ | 0.1929 | 14.0 | 3738 | 0.7362 |
65
+ | 0.1937 | 15.0 | 4005 | 0.7328 |
66
+ | 0.1934 | 16.0 | 4272 | 0.7329 |
67
+ | 0.1927 | 17.0 | 4539 | 0.7323 |
68
+ | 0.1927 | 18.0 | 4806 | 0.7257 |
69
+ | 0.1909 | 19.0 | 5073 | 0.7276 |
70
+ | 0.1919 | 20.0 | 5340 | 0.7251 |
71
+ | 0.1919 | 21.0 | 5607 | 0.7239 |
72
+ | 0.1912 | 22.0 | 5874 | 0.7260 |
73
+ | 0.1897 | 23.0 | 6141 | 0.7241 |
74
+ | 0.1916 | 24.0 | 6408 | 0.7235 |
75
+ | 0.1905 | 25.0 | 6675 | 0.7225 |
76
+ | 0.1919 | 26.0 | 6942 | 0.7188 |
77
+ | 0.1883 | 27.0 | 7209 | 0.7207 |
78
+ | 0.1898 | 28.0 | 7476 | 0.7198 |
79
+ | 0.1874 | 29.0 | 7743 | 0.7195 |
80
+ | 0.188 | 30.0 | 8010 | 0.7194 |
81
+ | 0.1873 | 31.0 | 8277 | 0.7182 |
82
+ | 0.1878 | 32.0 | 8544 | 0.7212 |
83
+ | 0.1866 | 33.0 | 8811 | 0.7171 |
84
+ | 0.1883 | 34.0 | 9078 | 0.7151 |
85
+ | 0.1881 | 35.0 | 9345 | 0.7176 |
86
+ | 0.1868 | 36.0 | 9612 | 0.7149 |
87
+ | 0.1871 | 37.0 | 9879 | 0.7157 |
88
+ | 0.1876 | 38.0 | 10146 | 0.7162 |
89
+ | 0.188 | 39.0 | 10413 | 0.7142 |
90
+ | 0.1861 | 40.0 | 10680 | 0.7149 |
91
+ | 0.1862 | 41.0 | 10947 | 0.7144 |
92
+ | 0.1862 | 42.0 | 11214 | 0.7128 |
93
+ | 0.186 | 43.0 | 11481 | 0.7136 |
94
+ | 0.1868 | 44.0 | 11748 | 0.7137 |
95
+ | 0.1837 | 45.0 | 12015 | 0.7138 |
96
+ | 0.1868 | 46.0 | 12282 | 0.7141 |
97
+ | 0.187 | 47.0 | 12549 | 0.7133 |
98
 
99
 
100
  ### Framework versions
adapter_model.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:98f4ac492147ad276077dbf256d43da87d4144b1212a22da1e960ce2c439e49e
3
  size 309980480
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:84669f99a89075044443690a296320a3a4634b7bf5db8b82193fbdbcab618fce
3
  size 309980480
all_results.json CHANGED
@@ -1,13 +1,13 @@
1
  {
2
- "epoch": 45.0,
3
  "eval_loss": 2.0493669509887695,
4
  "eval_runtime": 3.2905,
5
  "eval_samples_per_second": 555.532,
6
  "eval_steps_per_second": 69.594,
7
  "perplexity": 7.7629852003609585,
8
- "total_flos": 1.975920863064883e+16,
9
- "train_loss": 16.87015275197182,
10
- "train_runtime": 3210.0048,
11
- "train_samples_per_second": 132.835,
12
- "train_steps_per_second": 4.143
13
  }
 
1
  {
2
+ "epoch": 47.0,
3
  "eval_loss": 2.0493669509887695,
4
  "eval_runtime": 3.2905,
5
  "eval_samples_per_second": 555.532,
6
  "eval_steps_per_second": 69.594,
7
  "perplexity": 7.7629852003609585,
8
+ "total_flos": 2.5044104326938624e+16,
9
+ "train_loss": 0.1997737506061201,
10
+ "train_runtime": 1307.9406,
11
+ "train_samples_per_second": 326.009,
12
+ "train_steps_per_second": 10.207
13
  }
train_results.json CHANGED
@@ -1,8 +1,8 @@
1
  {
2
- "epoch": 45.0,
3
- "total_flos": 1.975920863064883e+16,
4
- "train_loss": 16.87015275197182,
5
- "train_runtime": 3210.0048,
6
- "train_samples_per_second": 132.835,
7
- "train_steps_per_second": 4.143
8
  }
 
1
  {
2
+ "epoch": 47.0,
3
+ "total_flos": 2.5044104326938624e+16,
4
+ "train_loss": 0.1997737506061201,
5
+ "train_runtime": 1307.9406,
6
+ "train_samples_per_second": 326.009,
7
+ "train_steps_per_second": 10.207
8
  }