Hugging Face's logo

ke-lly
/
47015772_2

Text Generation
Transformers
Safetensors
gpt2
trl
sft
Generated from Trainer
text-generation-inference
Inference Endpoints
Model card Files Community
ke-lly commited on
Commit
945141e
·
verified ·
1 Parent(s): 01fac33

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: '47015772_2'
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
+ # 47015772_2
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: 0.6622
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
+ | 1.0551 | 0.01 | 25 | 0.9816 | 0.0001 |
60
+ | 0.8806 | 0.01 | 50 | 0.8239 | 0.0001 |
61
+ | 0.8218 | 0.02 | 75 | 0.7796 | 0.0001 |
62
+ | 0.7854 | 0.02 | 100 | 0.7578 | 0.0001 |
63
+ | 0.7887 | 0.03 | 125 | 0.7440 | 0.0000 |
64
+ | 0.7675 | 0.03 | 150 | 0.7347 | 0.0001 |
65
+ | 0.7552 | 0.04 | 175 | 0.7275 | 0.0001 |
66
+ | 0.7493 | 0.04 | 200 | 0.7230 | 0.0001 |
67
+ | 0.7471 | 0.05 | 225 | 0.7187 | 0.0000 |
68
+ | 0.7376 | 0.06 | 250 | 0.7147 | 0.0001 |
69
+ | 0.7341 | 0.06 | 275 | 0.7119 | 0.0001 |
70
+ | 0.7387 | 0.07 | 300 | 0.7091 | 0.0001 |
71
+ | 0.7364 | 0.07 | 325 | 0.7069 | 0.0001 |
72
+ | 0.7179 | 0.08 | 350 | 0.7039 | 0.0001 |
73
+ | 0.7145 | 0.08 | 375 | 0.7016 | 0.0000 |
74
+ | 0.7204 | 0.09 | 400 | 0.6989 | 0.0000 |
75
+ | 0.7236 | 0.1 | 425 | 0.6963 | 0.0000 |
76
+ | 0.7099 | 0.1 | 450 | 0.6944 | 0.0000 |
77
+ | 0.7137 | 0.11 | 475 | 0.6927 | 0.0001 |
78
+ | 0.7106 | 0.11 | 500 | 0.6900 | 0.0000 |
79
+ | 0.7074 | 0.12 | 525 | 0.6884 | 0.0001 |
80
+ | 0.7141 | 0.12 | 550 | 0.6871 | 0.0000 |
81
+ | 0.7053 | 0.13 | 575 | 0.6856 | 0.0001 |
82
+ | 0.7082 | 0.13 | 600 | 0.6849 | 0.0000 |
83
+ | 0.704 | 0.14 | 625 | 0.6841 | 0.0001 |
84
+ | 0.7057 | 0.15 | 650 | 0.6828 | 0.0001 |
85
+ | 0.7027 | 0.15 | 675 | 0.6813 | 0.0001 |
86
+ | 0.6951 | 0.16 | 700 | 0.6807 | 0.0001 |
87
+ | 0.6943 | 0.16 | 725 | 0.6795 | 0.0001 |
88
+ | 0.7063 | 0.17 | 750 | 0.6781 | 0.0001 |
89
+ | 0.6888 | 0.17 | 775 | 0.6777 | 0.0001 |
90
+ | 0.7 | 0.18 | 800 | 0.6769 | 0.0001 |
91
+ | 0.6905 | 0.19 | 825 | 0.6762 | 0.0001 |
92
+ | 0.7005 | 0.19 | 850 | 0.6756 | 0.0001 |
93
+ | 0.6968 | 0.2 | 875 | 0.6747 | 0.0000 |
94
+ | 0.6895 | 0.2 | 900 | 0.6743 | 0.0001 |
95
+ | 0.6935 | 0.21 | 925 | 0.6732 | 0.0001 |
96
+ | 0.7006 | 0.21 | 950 | 0.6727 | 0.0001 |
97
+ | 0.6862 | 0.22 | 975 | 0.6722 | 0.0001 |
98
+ | 0.6921 | 0.22 | 1000 | 0.6716 | 0.0001 |
99
+ | 0.6875 | 0.23 | 1025 | 0.6710 | 0.0001 |
100
+ | 0.6971 | 0.24 | 1050 | 0.6705 | 0.0001 |
101
+ | 0.692 | 0.24 | 1075 | 0.6702 | 0.0001 |
102
+ | 0.686 | 0.25 | 1100 | 0.6693 | 0.0001 |
103
+ | 0.6876 | 0.25 | 1125 | 0.6692 | 0.0001 |
104
+ | 0.6834 | 0.26 | 1150 | 0.6686 | 0.0001 |
105
+ | 0.7024 | 0.26 | 1175 | 0.6683 | 0.0001 |
106
+ | 0.6931 | 0.27 | 1200 | 0.6678 | 0.0001 |
107
+ | 0.6764 | 0.28 | 1225 | 0.6677 | 0.0001 |
108
+ | 0.6914 | 0.28 | 1250 | 0.6672 | 0.0001 |
109
+ | 0.6916 | 0.29 | 1275 | 0.6667 | 0.0001 |
110
+ | 0.6808 | 0.29 | 1300 | 0.6665 | 0.0001 |
111
+ | 0.6789 | 0.3 | 1325 | 0.6661 | 0.0001 |
112
+ | 0.6803 | 0.3 | 1350 | 0.6659 | 0.0001 |
113
+ | 0.6892 | 0.31 | 1375 | 0.6656 | 0.0001 |
114
+ | 0.6749 | 0.31 | 1400 | 0.6654 | 0.0001 |
115
+ | 0.6823 | 0.32 | 1425 | 0.6648 | 0.0001 |
116
+ | 0.6827 | 0.33 | 1450 | 0.6648 | 0.0001 |
117
+ | 0.6826 | 0.33 | 1475 | 0.6647 | 0.0001 |
118
+ | 0.6813 | 0.34 | 1500 | 0.6645 | 0.0001 |
119
+ | 0.6864 | 0.34 | 1525 | 0.6641 | 0.0001 |
120
+ | 0.6809 | 0.35 | 1550 | 0.6639 | 0.0001 |
121
+ | 0.6779 | 0.35 | 1575 | 0.6639 | 0.0001 |
122
+ | 0.687 | 0.36 | 1600 | 0.6635 | 0.0001 |
123
+ | 0.6824 | 0.37 | 1625 | 0.6635 | 0.0001 |
124
+ | 0.6816 | 0.37 | 1650 | 0.6631 | 0.0001 |
125
+ | 0.6773 | 0.38 | 1675 | 0.6632 | 0.0001 |
126
+ | 0.6788 | 0.38 | 1700 | 0.6630 | 0.0001 |
127
+ | 0.6739 | 0.39 | 1725 | 0.6630 | 0.0001 |
128
+ | 0.6773 | 0.39 | 1750 | 0.6629 | 0.0001 |
129
+ | 0.6742 | 0.4 | 1775 | 0.6628 | 0.0001 |
130
+ | 0.6797 | 0.4 | 1800 | 0.6626 | 0.0001 |
131
+ | 0.6861 | 0.41 | 1825 | 0.6627 | 0.0001 |
132
+ | 0.686 | 0.42 | 1850 | 0.6625 | 0.0001 |
133
+ | 0.6711 | 0.42 | 1875 | 0.6624 | 0.0001 |
134
+ | 0.6773 | 0.43 | 1900 | 0.6624 | 0.0001 |
135
+ | 0.6844 | 0.43 | 1925 | 0.6623 | 0.0001 |
136
+ | 0.6809 | 0.44 | 1950 | 0.6622 | 0.0001 |
137
+ | 0.6766 | 0.44 | 1975 | 0.6622 | 0.0001 |
138
+ | 0.6733 | 0.45 | 2000 | 0.6622 | 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
+ }