Hugging Face's logo

NasimB
/
gpt2_left_out_wikipedia

Text Generation
Transformers
PyTorch
gpt2
Generated from Trainer
text-generation-inference
Model card Files
xet
 Community
1
NasimB commited on
Commit
e4eb547
·
1 Parent(s): 5d2c3e4

update model card README.md

Browse files
Files changed (1) hide show
  1. README.md +96 -0
README.md ADDED
@@ -0,0 +1,96 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ license: mit
3
+ tags:
4
+ - generated_from_trainer
5
+ datasets:
6
+ - generator
7
+ model-index:
8
+ - name: gpt2_left_out_wikipedia
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
+ # gpt2_left_out_wikipedia
16
+
17
+ This model is a fine-tuned version of [gpt2](https://huggingface.co/gpt2) on the generator dataset.
18
+ It achieves the following results on the evaluation set:
19
+ - Loss: 3.8366
20
+
21
+ ## Model description
22
+
23
+ More information needed
24
+
25
+ ## Intended uses & limitations
26
+
27
+ More information needed
28
+
29
+ ## Training and evaluation data
30
+
31
+ More information needed
32
+
33
+ ## Training procedure
34
+
35
+ ### Training hyperparameters
36
+
37
+ The following hyperparameters were used during training:
38
+ - learning_rate: 0.0005
39
+ - train_batch_size: 64
40
+ - eval_batch_size: 64
41
+ - seed: 42
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: 10
46
+ - mixed_precision_training: Native AMP
47
+
48
+ ### Training results
49
+
50
+ | Training Loss | Epoch | Step | Validation Loss |
51
+ |:-------------:|:-----:|:-----:|:---------------:|
52
+ | 5.8141 | 0.27 | 500 | 4.8520 |
53
+ | 4.5861 | 0.53 | 1000 | 4.4909 |
54
+ | 4.3045 | 0.8 | 1500 | 4.2742 |
55
+ | 4.0861 | 1.07 | 2000 | 4.1490 |
56
+ | 3.9278 | 1.33 | 2500 | 4.0562 |
57
+ | 3.8591 | 1.6 | 3000 | 3.9800 |
58
+ | 3.7835 | 1.87 | 3500 | 3.9083 |
59
+ | 3.6499 | 2.13 | 4000 | 3.8799 |
60
+ | 3.567 | 2.4 | 4500 | 3.8381 |
61
+ | 3.5361 | 2.67 | 5000 | 3.7975 |
62
+ | 3.5278 | 2.93 | 5500 | 3.7552 |
63
+ | 3.3555 | 3.2 | 6000 | 3.7622 |
64
+ | 3.3265 | 3.47 | 6500 | 3.7426 |
65
+ | 3.3305 | 3.73 | 7000 | 3.7122 |
66
+ | 3.3246 | 4.0 | 7500 | 3.6889 |
67
+ | 3.0968 | 4.27 | 8000 | 3.7216 |
68
+ | 3.1248 | 4.53 | 8500 | 3.7057 |
69
+ | 3.1354 | 4.8 | 9000 | 3.6846 |
70
+ | 3.0701 | 5.07 | 9500 | 3.7066 |
71
+ | 2.8974 | 5.33 | 10000 | 3.7183 |
72
+ | 2.9258 | 5.6 | 10500 | 3.7096 |
73
+ | 2.9387 | 5.87 | 11000 | 3.6943 |
74
+ | 2.7975 | 6.13 | 11500 | 3.7369 |
75
+ | 2.6972 | 6.4 | 12000 | 3.7468 |
76
+ | 2.7193 | 6.67 | 12500 | 3.7422 |
77
+ | 2.7233 | 6.93 | 13000 | 3.7337 |
78
+ | 2.5434 | 7.2 | 13500 | 3.7783 |
79
+ | 2.5072 | 7.47 | 14000 | 3.7864 |
80
+ | 2.5183 | 7.73 | 14500 | 3.7869 |
81
+ | 2.5263 | 8.0 | 15000 | 3.7838 |
82
+ | 2.3533 | 8.27 | 15500 | 3.8174 |
83
+ | 2.3661 | 8.53 | 16000 | 3.8220 |
84
+ | 2.3659 | 8.8 | 16500 | 3.8246 |
85
+ | 2.3462 | 9.07 | 17000 | 3.8313 |
86
+ | 2.286 | 9.33 | 17500 | 3.8359 |
87
+ | 2.2867 | 9.6 | 18000 | 3.8367 |
88
+ | 2.2885 | 9.87 | 18500 | 3.8366 |
89
+
90
+
91
+ ### Framework versions
92
+
93
+ - Transformers 4.26.1
94
+ - Pytorch 1.11.0+cu113
95
+ - Datasets 2.13.0
96
+ - Tokenizers 0.13.3