Hugging Face's logo

augustocsc
/
Se124M10KInfPrompt_endtoken

PEFT
Safetensors
Generated from Trainer
Model card Files
xet
 Community
augustocsc commited on
Commit
bd38aa3
·
verified ·
1 Parent(s): 494bb23

Model save

Browse files
Files changed (4) hide show
  1. README.md +28 -25
  2. adapter_model.safetensors +1 -1
  3. all_results.json +6 -6
  4. train_results.json +6 -6
README.md CHANGED
@@ -16,7 +16,7 @@ should probably proofread and complete it, then remove this comment. -->
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: 0.7284
20
 
21
  ## Model description
22
 
@@ -35,12 +35,13 @@ More information needed
35
  ### Training hyperparameters
36
 
37
  The following hyperparameters were used during training:
38
- - learning_rate: 5e-05
39
  - train_batch_size: 8
40
  - eval_batch_size: 8
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
 
@@ -48,28 +49,30 @@ The following hyperparameters were used during training:
48
 
49
  | Training Loss | Epoch | Step | Validation Loss |
50
  |:-------------:|:-----:|:-----:|:---------------:|
51
- | 0.9762 | 1.0 | 610 | 0.8681 |
52
- | 0.8832 | 2.0 | 1220 | 0.8117 |
53
- | 0.8432 | 3.0 | 1830 | 0.7922 |
54
- | 0.8183 | 4.0 | 2440 | 0.7809 |
55
- | 0.8131 | 5.0 | 3050 | 0.7687 |
56
- | 0.8004 | 6.0 | 3660 | 0.7640 |
57
- | 0.7972 | 7.0 | 4270 | 0.7584 |
58
- | 0.7941 | 8.0 | 4880 | 0.7528 |
59
- | 0.789 | 9.0 | 5490 | 0.7480 |
60
- | 0.7806 | 10.0 | 6100 | 0.7454 |
61
- | 0.7736 | 11.0 | 6710 | 0.7442 |
62
- | 0.769 | 12.0 | 7320 | 0.7444 |
63
- | 0.7734 | 13.0 | 7930 | 0.7402 |
64
- | 0.7671 | 14.0 | 8540 | 0.7385 |
65
- | 0.7605 | 15.0 | 9150 | 0.7365 |
66
- | 0.7651 | 16.0 | 9760 | 0.7357 |
67
- | 0.7657 | 17.0 | 10370 | 0.7340 |
68
- | 0.763 | 18.0 | 10980 | 0.7318 |
69
- | 0.7552 | 19.0 | 11590 | 0.7305 |
70
- | 0.7563 | 20.0 | 12200 | 0.7284 |
71
- | 0.7558 | 21.0 | 12810 | 0.7285 |
72
- | 0.7465 | 22.0 | 13420 | 0.7290 |
 
 
73
 
74
 
75
  ### Framework versions
 
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: 0.6875
20
 
21
  ## Model description
22
 
 
35
  ### Training hyperparameters
36
 
37
  The following hyperparameters were used during training:
38
+ - learning_rate: 0.0005
39
  - train_batch_size: 8
40
  - eval_batch_size: 8
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: cosine
44
+ - lr_scheduler_warmup_steps: 200
45
  - num_epochs: 50
46
  - mixed_precision_training: Native AMP
47
 
 
49
 
50
  | Training Loss | Epoch | Step | Validation Loss |
51
  |:-------------:|:-----:|:-----:|:---------------:|
52
+ | 0.8085 | 1.0 | 610 | 0.7760 |
53
+ | 0.7801 | 2.0 | 1220 | 0.7436 |
54
+ | 0.7608 | 3.0 | 1830 | 0.7269 |
55
+ | 0.7438 | 4.0 | 2440 | 0.7199 |
56
+ | 0.7413 | 5.0 | 3050 | 0.7118 |
57
+ | 0.7343 | 6.0 | 3660 | 0.7121 |
58
+ | 0.7332 | 7.0 | 4270 | 0.7089 |
59
+ | 0.7319 | 8.0 | 4880 | 0.7025 |
60
+ | 0.7289 | 9.0 | 5490 | 0.7001 |
61
+ | 0.7236 | 10.0 | 6100 | 0.6965 |
62
+ | 0.7147 | 11.0 | 6710 | 0.6970 |
63
+ | 0.7126 | 12.0 | 7320 | 0.6973 |
64
+ | 0.7167 | 13.0 | 7930 | 0.6935 |
65
+ | 0.711 | 14.0 | 8540 | 0.6927 |
66
+ | 0.7057 | 15.0 | 9150 | 0.6940 |
67
+ | 0.7109 | 16.0 | 9760 | 0.6924 |
68
+ | 0.7117 | 17.0 | 10370 | 0.6928 |
69
+ | 0.7086 | 18.0 | 10980 | 0.6882 |
70
+ | 0.7004 | 19.0 | 11590 | 0.6872 |
71
+ | 0.7016 | 20.0 | 12200 | 0.6895 |
72
+ | 0.7027 | 21.0 | 12810 | 0.6884 |
73
+ | 0.6928 | 22.0 | 13420 | 0.6885 |
74
+ | 0.7059 | 23.0 | 14030 | 0.6894 |
75
+ | 0.6916 | 24.0 | 14640 | 0.6875 |
76
 
77
 
78
  ### Framework versions
adapter_model.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:857eb4f5c1eec783f6b4a13739f0ca15d4d1d364d1ef3b8f19315a7e182fb29f
3
  size 309980480
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:9fa14457372498550554059b25cf425279046c0b5d83f5f67fdc09a283753333
3
  size 309980480
all_results.json CHANGED
@@ -1,13 +1,13 @@
1
  {
2
- "epoch": 22.0,
3
  "eval_loss": 0.7283757925033569,
4
  "eval_runtime": 1.828,
5
  "eval_samples_per_second": 578.237,
6
  "eval_steps_per_second": 72.758,
7
  "perplexity": 2.071712981796928,
8
- "total_flos": 7033068097634304.0,
9
- "train_loss": 0.8158138496865104,
10
- "train_runtime": 912.3512,
11
- "train_samples_per_second": 267.276,
12
- "train_steps_per_second": 33.43
13
  }
 
1
  {
2
+ "epoch": 24.0,
3
  "eval_loss": 0.7283757925033569,
4
  "eval_runtime": 1.828,
5
  "eval_samples_per_second": 578.237,
6
  "eval_steps_per_second": 72.758,
7
  "perplexity": 2.071712981796928,
8
+ "total_flos": 7672437924691968.0,
9
+ "train_loss": 0.7372308338926138,
10
+ "train_runtime": 572.2889,
11
+ "train_samples_per_second": 426.096,
12
+ "train_steps_per_second": 53.295
13
  }
train_results.json CHANGED
@@ -1,8 +1,8 @@
1
  {
2
- "epoch": 22.0,
3
- "total_flos": 7033068097634304.0,
4
- "train_loss": 0.8158138496865104,
5
- "train_runtime": 912.3512,
6
- "train_samples_per_second": 267.276,
7
- "train_steps_per_second": 33.43
8
  }
 
1
  {
2
+ "epoch": 24.0,
3
+ "total_flos": 7672437924691968.0,
4
+ "train_loss": 0.7372308338926138,
5
+ "train_runtime": 572.2889,
6
+ "train_samples_per_second": 426.096,
7
+ "train_steps_per_second": 53.295
8
  }