Hugging Face's logo

C0uchP0tat0
/
my_rugpt3medium_finetune

Text Generation
Transformers
TensorBoard
Safetensors
gpt2
Generated from Trainer
text-generation-inference
Inference Endpoints
Model card Files Metrics Community
C0uchP0tat0 commited on
Commit
6827db8
•
1 Parent(s): 5edbf7d

End of training

Browse files
Files changed (3) hide show
  1. README.md +57 -11
  2. model.safetensors +1 -1
  3. runs/Dec28_07-10-19_782f0ab6a697/events.out.tfevents.1703747423.782f0ab6a697.5258.0 +2 -2
README.md CHANGED
@@ -14,7 +14,7 @@ should probably proofread and complete it, then remove this comment. -->
14
 
15
  This model is a fine-tuned version of [ai-forever/rugpt3medium_based_on_gpt2](https://huggingface.co/ai-forever/rugpt3medium_based_on_gpt2) on an unknown dataset.
16
  It achieves the following results on the evaluation set:
17
- - Loss: 9.1812
18
 
19
  ## Model description
20
 
@@ -41,22 +41,68 @@ The following hyperparameters were used during training:
41
  - total_train_batch_size: 24
42
  - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
43
  - lr_scheduler_type: cosine
44
- - lr_scheduler_warmup_steps: 500
45
- - num_epochs: 3
46
  - mixed_precision_training: Native AMP
47
 
48
  ### Training results
49
 
50
  | Training Loss | Epoch | Step | Validation Loss |
51
  |:-------------:|:-----:|:----:|:---------------:|
52
- | 10.8819 | 0.37 | 20 | 10.5196 |
53
- | 10.2462 | 0.74 | 40 | 9.9291 |
54
- | 9.8184 | 1.1 | 60 | 9.6638 |
55
- | 9.6334 | 1.47 | 80 | 9.5486 |
56
- | 9.5381 | 1.84 | 100 | 9.4843 |
57
- | 9.4918 | 2.21 | 120 | 9.4058 |
58
- | 9.3774 | 2.58 | 140 | 9.2999 |
59
- | 9.2801 | 2.94 | 160 | 9.1812 |
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
60
 
61
 
62
  ### Framework versions
 
14
 
15
  This model is a fine-tuned version of [ai-forever/rugpt3medium_based_on_gpt2](https://huggingface.co/ai-forever/rugpt3medium_based_on_gpt2) on an unknown dataset.
16
  It achieves the following results on the evaluation set:
17
+ - Loss: 4.3387
18
 
19
  ## Model description
20
 
 
41
  - total_train_batch_size: 24
42
  - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
43
  - lr_scheduler_type: cosine
44
+ - lr_scheduler_warmup_steps: 1000
45
+ - num_epochs: 25
46
  - mixed_precision_training: Native AMP
47
 
48
  ### Training results
49
 
50
  | Training Loss | Epoch | Step | Validation Loss |
51
  |:-------------:|:-----:|:----:|:---------------:|
52
+ | 10.916 | 0.46 | 25 | 10.6340 |
53
+ | 10.3795 | 0.92 | 50 | 9.9985 |
54
+ | 9.9003 | 1.38 | 75 | 9.7015 |
55
+ | 9.6822 | 1.84 | 100 | 9.5795 |
56
+ | 9.5804 | 2.3 | 125 | 9.5130 |
57
+ | 9.5294 | 2.76 | 150 | 9.4485 |
58
+ | 9.439 | 3.22 | 175 | 9.3772 |
59
+ | 9.3698 | 3.68 | 200 | 9.2804 |
60
+ | 9.2964 | 4.14 | 225 | 9.1746 |
61
+ | 9.1945 | 4.6 | 250 | 9.0623 |
62
+ | 9.0492 | 5.06 | 275 | 8.9352 |
63
+ | 8.9521 | 5.52 | 300 | 8.8157 |
64
+ | 8.8634 | 5.98 | 325 | 8.6838 |
65
+ | 8.7197 | 6.44 | 350 | 8.5445 |
66
+ | 8.6485 | 6.9 | 375 | 8.4181 |
67
+ | 8.522 | 7.36 | 400 | 8.2732 |
68
+ | 8.4227 | 7.82 | 425 | 8.1704 |
69
+ | 8.3083 | 8.28 | 450 | 8.0290 |
70
+ | 8.1897 | 8.74 | 475 | 7.8989 |
71
+ | 8.0876 | 9.2 | 500 | 7.7778 |
72
+ | 7.9824 | 9.66 | 525 | 7.6368 |
73
+ | 7.8762 | 10.12 | 550 | 7.4974 |
74
+ | 7.7408 | 10.58 | 575 | 7.3658 |
75
+ | 7.6855 | 11.04 | 600 | 7.2416 |
76
+ | 7.5163 | 11.5 | 625 | 7.1291 |
77
+ | 7.5079 | 11.96 | 650 | 7.0295 |
78
+ | 7.2873 | 12.42 | 675 | 6.8522 |
79
+ | 7.2856 | 12.88 | 700 | 6.7573 |
80
+ | 7.0868 | 13.34 | 725 | 6.6651 |
81
+ | 7.0886 | 13.8 | 750 | 6.5239 |
82
+ | 6.9283 | 14.26 | 775 | 6.3561 |
83
+ | 6.8257 | 14.72 | 800 | 6.2392 |
84
+ | 6.7328 | 15.18 | 825 | 6.1004 |
85
+ | 6.6153 | 15.64 | 850 | 5.9846 |
86
+ | 6.5824 | 16.1 | 875 | 5.8627 |
87
+ | 6.3905 | 16.56 | 900 | 5.7724 |
88
+ | 6.359 | 17.02 | 925 | 5.6321 |
89
+ | 6.1679 | 17.48 | 950 | 5.5329 |
90
+ | 6.1526 | 17.94 | 975 | 5.4058 |
91
+ | 5.9604 | 18.4 | 1000 | 5.3046 |
92
+ | 5.9669 | 18.87 | 1025 | 5.1939 |
93
+ | 5.6807 | 19.33 | 1050 | 5.0499 |
94
+ | 5.7445 | 19.79 | 1075 | 4.9479 |
95
+ | 5.6578 | 20.25 | 1100 | 4.8343 |
96
+ | 5.4919 | 20.71 | 1125 | 4.7547 |
97
+ | 5.4427 | 21.17 | 1150 | 4.6506 |
98
+ | 5.3212 | 21.63 | 1175 | 4.5628 |
99
+ | 5.2953 | 22.09 | 1200 | 4.4814 |
100
+ | 5.1872 | 22.55 | 1225 | 4.4373 |
101
+ | 5.1285 | 23.01 | 1250 | 4.3966 |
102
+ | 5.047 | 23.47 | 1275 | 4.3611 |
103
+ | 5.0698 | 23.93 | 1300 | 4.3520 |
104
+ | 5.1259 | 24.39 | 1325 | 4.3408 |
105
+ | 4.9851 | 24.85 | 1350 | 4.3387 |
106
 
107
 
108
  ### Framework versions
model.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:7862a5feed70f1659b46c0947c48c856d41f6f9b359277b0a756de394bff3002
3
  size 1423517184
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b93af633e0f28d0805556ff8d3a6ea5aded034a53df0470a8ee26bd14cb8fb6b
3
  size 1423517184
runs/Dec28_07-10-19_782f0ab6a697/events.out.tfevents.1703747423.782f0ab6a697.5258.0 CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:075b0ca4eeadeaeebfb220019573dd31bcbda2af9985f69026ba80ad764c6bd3
3
- size 21634
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:12130b423b70067cc1ca76a9c37f87b22299154dc525353445e060fa2efced75
3
+ size 27980