Hugging Face's logo

ke-lly
/
47163343_0

Text Generation
Transformers
Safetensors
gpt2
trl
sft
Generated from Trainer
text-generation-inference
Inference Endpoints
Model card Files Community
ke-lly commited on
Commit
6ee0f51
•
1 Parent(s): c635202

Model save

Browse files
Files changed (2) hide show
  1. README.md +146 -0
  2. generation_config.json +6 -0
README.md ADDED
@@ -0,0 +1,146 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ license: mit
3
+ base_model: openai-community/gpt2
4
+ tags:
5
+ - trl
6
+ - sft
7
+ - generated_from_trainer
8
+ metrics:
9
+ - accuracy
10
+ model-index:
11
+ - name: '47163343_0'
12
+ results: []
13
+ ---
14
+
15
+ <!-- This model card has been generated automatically according to the information the Trainer had access to. You
16
+ should probably proofread and complete it, then remove this comment. -->
17
+
18
+ # 47163343_0
19
+
20
+ This model is a fine-tuned version of [openai-community/gpt2](https://huggingface.co/openai-community/gpt2) on an unknown dataset.
21
+ It achieves the following results on the evaluation set:
22
+ - Loss: 0.6417
23
+ - Accuracy: 0.0001
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: 1.41e-05
43
+ - train_batch_size: 32
44
+ - eval_batch_size: 4
45
+ - seed: 42
46
+ - distributed_type: multi-GPU
47
+ - num_devices: 2
48
+ - gradient_accumulation_steps: 4
49
+ - total_train_batch_size: 256
50
+ - total_eval_batch_size: 8
51
+ - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
52
+ - lr_scheduler_type: linear
53
+ - training_steps: 2000
54
+
55
+ ### Training results
56
+
57
+ | Training Loss | Epoch | Step | Validation Loss | Accuracy |
58
+ |:-------------:|:-----:|:----:|:---------------:|:--------:|
59
+ | 1.0667 | 0.48 | 25 | 0.9624 | 0.0001 |
60
+ | 0.854 | 0.97 | 50 | 0.8076 | 0.0001 |
61
+ | 0.7965 | 1.45 | 75 | 0.7675 | 0.0002 |
62
+ | 0.7817 | 1.93 | 100 | 0.7461 | 0.0001 |
63
+ | 0.7668 | 2.42 | 125 | 0.7326 | 0.0001 |
64
+ | 0.7533 | 2.9 | 150 | 0.7225 | 0.0001 |
65
+ | 0.7479 | 3.38 | 175 | 0.7150 | 0.0001 |
66
+ | 0.7325 | 3.86 | 200 | 0.7104 | 0.0001 |
67
+ | 0.7397 | 4.35 | 225 | 0.7055 | 0.0001 |
68
+ | 0.7255 | 4.83 | 250 | 0.7020 | 0.0001 |
69
+ | 0.7182 | 5.31 | 275 | 0.6983 | 0.0001 |
70
+ | 0.7161 | 5.8 | 300 | 0.6955 | 0.0001 |
71
+ | 0.7162 | 6.28 | 325 | 0.6925 | 0.0001 |
72
+ | 0.7043 | 6.76 | 350 | 0.6901 | 0.0001 |
73
+ | 0.7076 | 7.25 | 375 | 0.6868 | 0.0001 |
74
+ | 0.7042 | 7.73 | 400 | 0.6844 | 0.0001 |
75
+ | 0.6975 | 8.21 | 425 | 0.6820 | 0.0001 |
76
+ | 0.7021 | 8.7 | 450 | 0.6799 | 0.0001 |
77
+ | 0.6955 | 9.18 | 475 | 0.6778 | 0.0001 |
78
+ | 0.6868 | 9.66 | 500 | 0.6763 | 0.0001 |
79
+ | 0.6866 | 10.14 | 525 | 0.6744 | 0.0001 |
80
+ | 0.6903 | 10.63 | 550 | 0.6723 | 0.0001 |
81
+ | 0.6786 | 11.11 | 575 | 0.6709 | 0.0001 |
82
+ | 0.6843 | 11.59 | 600 | 0.6695 | 0.0001 |
83
+ | 0.6835 | 12.08 | 625 | 0.6680 | 0.0001 |
84
+ | 0.6819 | 12.56 | 650 | 0.6669 | 0.0001 |
85
+ | 0.6804 | 13.04 | 675 | 0.6656 | 0.0001 |
86
+ | 0.6748 | 13.53 | 700 | 0.6642 | 0.0001 |
87
+ | 0.674 | 14.01 | 725 | 0.6637 | 0.0001 |
88
+ | 0.6731 | 14.49 | 750 | 0.6624 | 0.0001 |
89
+ | 0.681 | 14.98 | 775 | 0.6611 | 0.0001 |
90
+ | 0.6763 | 15.46 | 800 | 0.6602 | 0.0001 |
91
+ | 0.677 | 15.94 | 825 | 0.6597 | 0.0001 |
92
+ | 0.6725 | 16.43 | 850 | 0.6583 | 0.0001 |
93
+ | 0.6669 | 16.91 | 875 | 0.6574 | 0.0001 |
94
+ | 0.6682 | 17.39 | 900 | 0.6567 | 0.0001 |
95
+ | 0.669 | 17.87 | 925 | 0.6559 | 0.0001 |
96
+ | 0.6647 | 18.36 | 950 | 0.6554 | 0.0001 |
97
+ | 0.664 | 18.84 | 975 | 0.6549 | 0.0001 |
98
+ | 0.6563 | 19.32 | 1000 | 0.6542 | 0.0001 |
99
+ | 0.6656 | 19.81 | 1025 | 0.6533 | 0.0001 |
100
+ | 0.6634 | 20.29 | 1050 | 0.6530 | 0.0001 |
101
+ | 0.6592 | 20.77 | 1075 | 0.6521 | 0.0001 |
102
+ | 0.6558 | 21.26 | 1100 | 0.6514 | 0.0001 |
103
+ | 0.6664 | 21.74 | 1125 | 0.6511 | 0.0001 |
104
+ | 0.6561 | 22.22 | 1150 | 0.6504 | 0.0001 |
105
+ | 0.6634 | 22.71 | 1175 | 0.6499 | 0.0001 |
106
+ | 0.6679 | 23.19 | 1200 | 0.6491 | 0.0001 |
107
+ | 0.6625 | 23.67 | 1225 | 0.6489 | 0.0001 |
108
+ | 0.6619 | 24.15 | 1250 | 0.6483 | 0.0001 |
109
+ | 0.6495 | 24.64 | 1275 | 0.6479 | 0.0001 |
110
+ | 0.6547 | 25.12 | 1300 | 0.6474 | 0.0001 |
111
+ | 0.6649 | 25.6 | 1325 | 0.6469 | 0.0001 |
112
+ | 0.6551 | 26.09 | 1350 | 0.6466 | 0.0001 |
113
+ | 0.6547 | 26.57 | 1375 | 0.6463 | 0.0001 |
114
+ | 0.6546 | 27.05 | 1400 | 0.6458 | 0.0001 |
115
+ | 0.6576 | 27.54 | 1425 | 0.6456 | 0.0001 |
116
+ | 0.6576 | 28.02 | 1450 | 0.6452 | 0.0001 |
117
+ | 0.6568 | 28.5 | 1475 | 0.6448 | 0.0001 |
118
+ | 0.6596 | 28.99 | 1500 | 0.6446 | 0.0001 |
119
+ | 0.6538 | 29.47 | 1525 | 0.6443 | 0.0001 |
120
+ | 0.6488 | 29.95 | 1550 | 0.6440 | 0.0001 |
121
+ | 0.6433 | 30.43 | 1575 | 0.6437 | 0.0001 |
122
+ | 0.6583 | 30.92 | 1600 | 0.6435 | 0.0001 |
123
+ | 0.6575 | 31.4 | 1625 | 0.6432 | 0.0001 |
124
+ | 0.6465 | 31.88 | 1650 | 0.6430 | 0.0001 |
125
+ | 0.6495 | 32.37 | 1675 | 0.6429 | 0.0001 |
126
+ | 0.6487 | 32.85 | 1700 | 0.6427 | 0.0001 |
127
+ | 0.6571 | 33.33 | 1725 | 0.6426 | 0.0001 |
128
+ | 0.6463 | 33.82 | 1750 | 0.6425 | 0.0001 |
129
+ | 0.648 | 34.3 | 1775 | 0.6423 | 0.0001 |
130
+ | 0.6537 | 34.78 | 1800 | 0.6422 | 0.0001 |
131
+ | 0.6564 | 35.27 | 1825 | 0.6420 | 0.0001 |
132
+ | 0.6491 | 35.75 | 1850 | 0.6420 | 0.0001 |
133
+ | 0.6549 | 36.23 | 1875 | 0.6419 | 0.0001 |
134
+ | 0.6524 | 36.71 | 1900 | 0.6418 | 0.0001 |
135
+ | 0.6522 | 37.2 | 1925 | 0.6418 | 0.0001 |
136
+ | 0.655 | 37.68 | 1950 | 0.6417 | 0.0001 |
137
+ | 0.6614 | 38.16 | 1975 | 0.6417 | 0.0001 |
138
+ | 0.6451 | 38.65 | 2000 | 0.6417 | 0.0001 |
139
+
140
+
141
+ ### Framework versions
142
+
143
+ - Transformers 4.37.0
144
+ - Pytorch 2.0.0+cu118
145
+ - Datasets 2.16.1
146
+ - Tokenizers 0.15.1
generation_config.json ADDED
@@ -0,0 +1,6 @@
 
 
 
 
 
 
 
1
+ {
2
+ "_from_model_config": true,
3
+ "bos_token_id": 50256,
4
+ "eos_token_id": 50256,
5
+ "transformers_version": "4.37.0"
6
+ }