Hugging Face's logo

Jellywibble
/
CHAI_alignment_reward_model

Text Classification
Transformers
Safetensors
gpt2
Generated from Trainer
text-generation-inference
Inference Endpoints
Model card Files Community
Jellywibble commited on
Commit
5aef637
•
1 Parent(s): 81a3339

Jellywibble/CHAI_alignment_reward_model

Browse files
Files changed (4) hide show
  1. README.md +171 -0
  2. config.json +41 -0
  3. model.safetensors +3 -0
  4. training_args.bin +3 -0
README.md ADDED
@@ -0,0 +1,171 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ license: mit
3
+ base_model: gpt2
4
+ tags:
5
+ - generated_from_trainer
6
+ metrics:
7
+ - accuracy
8
+ model-index:
9
+ - name: output
10
+ results: []
11
+ ---
12
+
13
+ <!-- This model card has been generated automatically according to the information the Trainer had access to. You
14
+ should probably proofread and complete it, then remove this comment. -->
15
+
16
+ # output
17
+
18
+ This model is a fine-tuned version of [gpt2](https://huggingface.co/gpt2) on the None dataset.
19
+ It achieves the following results on the evaluation set:
20
+ - Loss: 0.5590
21
+ - Accuracy: 0.7005
22
+
23
+ ## Model description
24
+
25
+ More information needed
26
+
27
+ ## Intended uses & limitations
28
+
29
+ More information needed
30
+
31
+ ## Training and evaluation data
32
+
33
+ More information needed
34
+
35
+ ## Training procedure
36
+
37
+ ### Training hyperparameters
38
+
39
+ The following hyperparameters were used during training:
40
+ - learning_rate: 1e-05
41
+ - train_batch_size: 16
42
+ - eval_batch_size: 64
43
+ - seed: 42
44
+ - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
45
+ - lr_scheduler_type: linear
46
+ - num_epochs: 3
47
+ - mixed_precision_training: Native AMP
48
+
49
+ ### Training results
50
+
51
+ | Training Loss | Epoch | Step | Validation Loss | Accuracy |
52
+ |:-------------:|:------:|:-----:|:---------------:|:--------:|
53
+ | 0.6976 | 0.0268 | 250 | 0.6614 | 0.6728 |
54
+ | 0.6155 | 0.0537 | 500 | 0.5858 | 0.6811 |
55
+ | 0.5869 | 0.0805 | 750 | 0.5820 | 0.6856 |
56
+ | 0.5911 | 0.1073 | 1000 | 0.5843 | 0.6811 |
57
+ | 0.5788 | 0.1341 | 1250 | 0.5750 | 0.6790 |
58
+ | 0.5913 | 0.1610 | 1500 | 0.5810 | 0.6864 |
59
+ | 0.5712 | 0.1878 | 1750 | 0.5731 | 0.6892 |
60
+ | 0.5793 | 0.2146 | 2000 | 0.5717 | 0.6882 |
61
+ | 0.5788 | 0.2415 | 2250 | 0.5868 | 0.6838 |
62
+ | 0.5802 | 0.2683 | 2500 | 0.5653 | 0.6942 |
63
+ | 0.583 | 0.2951 | 2750 | 0.5631 | 0.6984 |
64
+ | 0.5762 | 0.3220 | 3000 | 0.5654 | 0.6916 |
65
+ | 0.5678 | 0.3488 | 3250 | 0.5635 | 0.6906 |
66
+ | 0.5679 | 0.3756 | 3500 | 0.5706 | 0.6838 |
67
+ | 0.56 | 0.4024 | 3750 | 0.5661 | 0.6932 |
68
+ | 0.562 | 0.4293 | 4000 | 0.5994 | 0.6885 |
69
+ | 0.5861 | 0.4561 | 4250 | 0.5659 | 0.6979 |
70
+ | 0.5845 | 0.4829 | 4500 | 0.5631 | 0.6992 |
71
+ | 0.5665 | 0.5098 | 4750 | 0.5621 | 0.6987 |
72
+ | 0.5795 | 0.5366 | 5000 | 0.5698 | 0.6934 |
73
+ | 0.5722 | 0.5634 | 5250 | 0.5615 | 0.6895 |
74
+ | 0.5765 | 0.5903 | 5500 | 0.5610 | 0.7010 |
75
+ | 0.5627 | 0.6171 | 5750 | 0.5594 | 0.6932 |
76
+ | 0.5761 | 0.6439 | 6000 | 0.5581 | 0.6997 |
77
+ | 0.5682 | 0.6707 | 6250 | 0.5693 | 0.6856 |
78
+ | 0.566 | 0.6976 | 6500 | 0.5634 | 0.6895 |
79
+ | 0.5628 | 0.7244 | 6750 | 0.5594 | 0.7026 |
80
+ | 0.5739 | 0.7512 | 7000 | 0.5634 | 0.6926 |
81
+ | 0.5762 | 0.7781 | 7250 | 0.5593 | 0.7015 |
82
+ | 0.572 | 0.8049 | 7500 | 0.5612 | 0.6853 |
83
+ | 0.5657 | 0.8317 | 7750 | 0.5593 | 0.6974 |
84
+ | 0.5665 | 0.8586 | 8000 | 0.5614 | 0.6916 |
85
+ | 0.578 | 0.8854 | 8250 | 0.5600 | 0.6995 |
86
+ | 0.571 | 0.9122 | 8500 | 0.5635 | 0.6934 |
87
+ | 0.5703 | 0.9390 | 8750 | 0.5628 | 0.7052 |
88
+ | 0.5801 | 0.9659 | 9000 | 0.5582 | 0.7010 |
89
+ | 0.5691 | 0.9927 | 9250 | 0.5673 | 0.6958 |
90
+ | 0.551 | 1.0195 | 9500 | 0.5631 | 0.6913 |
91
+ | 0.5625 | 1.0464 | 9750 | 0.5583 | 0.6987 |
92
+ | 0.5679 | 1.0732 | 10000 | 0.5633 | 0.7015 |
93
+ | 0.5693 | 1.1000 | 10250 | 0.5590 | 0.6934 |
94
+ | 0.5649 | 1.1269 | 10500 | 0.5580 | 0.6966 |
95
+ | 0.5558 | 1.1537 | 10750 | 0.5661 | 0.6879 |
96
+ | 0.5674 | 1.1805 | 11000 | 0.5595 | 0.7026 |
97
+ | 0.5507 | 1.2073 | 11250 | 0.5594 | 0.7015 |
98
+ | 0.5656 | 1.2342 | 11500 | 0.5592 | 0.6976 |
99
+ | 0.5696 | 1.2610 | 11750 | 0.5604 | 0.6926 |
100
+ | 0.5605 | 1.2878 | 12000 | 0.5618 | 0.7026 |
101
+ | 0.5572 | 1.3147 | 12250 | 0.5649 | 0.7000 |
102
+ | 0.5553 | 1.3415 | 12500 | 0.5621 | 0.6984 |
103
+ | 0.546 | 1.3683 | 12750 | 0.5630 | 0.6966 |
104
+ | 0.5614 | 1.3951 | 13000 | 0.5605 | 0.6955 |
105
+ | 0.5635 | 1.4220 | 13250 | 0.5587 | 0.6971 |
106
+ | 0.5561 | 1.4488 | 13500 | 0.5647 | 0.6947 |
107
+ | 0.5634 | 1.4756 | 13750 | 0.5607 | 0.6995 |
108
+ | 0.5585 | 1.5025 | 14000 | 0.5577 | 0.7023 |
109
+ | 0.5599 | 1.5293 | 14250 | 0.5740 | 0.6788 |
110
+ | 0.5697 | 1.5561 | 14500 | 0.5570 | 0.7023 |
111
+ | 0.5453 | 1.5830 | 14750 | 0.5624 | 0.6921 |
112
+ | 0.5642 | 1.6098 | 15000 | 0.5687 | 0.6864 |
113
+ | 0.5692 | 1.6366 | 15250 | 0.5643 | 0.6924 |
114
+ | 0.558 | 1.6634 | 15500 | 0.5625 | 0.6961 |
115
+ | 0.5465 | 1.6903 | 15750 | 0.5627 | 0.6997 |
116
+ | 0.5744 | 1.7171 | 16000 | 0.5594 | 0.6992 |
117
+ | 0.5683 | 1.7439 | 16250 | 0.5577 | 0.6961 |
118
+ | 0.5638 | 1.7708 | 16500 | 0.5579 | 0.6961 |
119
+ | 0.5512 | 1.7976 | 16750 | 0.5613 | 0.6945 |
120
+ | 0.5652 | 1.8244 | 17000 | 0.5596 | 0.6987 |
121
+ | 0.5771 | 1.8513 | 17250 | 0.5575 | 0.6997 |
122
+ | 0.5624 | 1.8781 | 17500 | 0.5628 | 0.6971 |
123
+ | 0.5719 | 1.9049 | 17750 | 0.5575 | 0.6937 |
124
+ | 0.5577 | 1.9317 | 18000 | 0.5686 | 0.6895 |
125
+ | 0.5599 | 1.9586 | 18250 | 0.5632 | 0.6981 |
126
+ | 0.5622 | 1.9854 | 18500 | 0.5574 | 0.7008 |
127
+ | 0.56 | 2.0122 | 18750 | 0.5577 | 0.7008 |
128
+ | 0.5447 | 2.0391 | 19000 | 0.5590 | 0.7036 |
129
+ | 0.5599 | 2.0659 | 19250 | 0.5604 | 0.7005 |
130
+ | 0.5512 | 2.0927 | 19500 | 0.5584 | 0.7000 |
131
+ | 0.549 | 2.1196 | 19750 | 0.5593 | 0.6987 |
132
+ | 0.5485 | 2.1464 | 20000 | 0.5680 | 0.6947 |
133
+ | 0.5528 | 2.1732 | 20250 | 0.5619 | 0.6955 |
134
+ | 0.5549 | 2.2000 | 20500 | 0.5593 | 0.7021 |
135
+ | 0.5505 | 2.2269 | 20750 | 0.5608 | 0.7029 |
136
+ | 0.5424 | 2.2537 | 21000 | 0.5644 | 0.7021 |
137
+ | 0.5405 | 2.2805 | 21250 | 0.5607 | 0.7013 |
138
+ | 0.5492 | 2.3074 | 21500 | 0.5611 | 0.6984 |
139
+ | 0.5589 | 2.3342 | 21750 | 0.5621 | 0.6961 |
140
+ | 0.5531 | 2.3610 | 22000 | 0.5615 | 0.6995 |
141
+ | 0.5539 | 2.3879 | 22250 | 0.5623 | 0.6950 |
142
+ | 0.5479 | 2.4147 | 22500 | 0.5615 | 0.7021 |
143
+ | 0.5476 | 2.4415 | 22750 | 0.5600 | 0.7015 |
144
+ | 0.5589 | 2.4683 | 23000 | 0.5596 | 0.6981 |
145
+ | 0.5511 | 2.4952 | 23250 | 0.5603 | 0.6997 |
146
+ | 0.5517 | 2.5220 | 23500 | 0.5594 | 0.7015 |
147
+ | 0.5439 | 2.5488 | 23750 | 0.5623 | 0.6947 |
148
+ | 0.5442 | 2.5757 | 24000 | 0.5612 | 0.7044 |
149
+ | 0.5455 | 2.6025 | 24250 | 0.5596 | 0.6966 |
150
+ | 0.5525 | 2.6293 | 24500 | 0.5613 | 0.6981 |
151
+ | 0.5384 | 2.6561 | 24750 | 0.5622 | 0.7010 |
152
+ | 0.552 | 2.6830 | 25000 | 0.5611 | 0.6981 |
153
+ | 0.5551 | 2.7098 | 25250 | 0.5642 | 0.6940 |
154
+ | 0.5411 | 2.7366 | 25500 | 0.5615 | 0.7005 |
155
+ | 0.5661 | 2.7635 | 25750 | 0.5614 | 0.6979 |
156
+ | 0.5528 | 2.7903 | 26000 | 0.5593 | 0.7002 |
157
+ | 0.5603 | 2.8171 | 26250 | 0.5588 | 0.7002 |
158
+ | 0.5514 | 2.8440 | 26500 | 0.5590 | 0.7000 |
159
+ | 0.5559 | 2.8708 | 26750 | 0.5591 | 0.7010 |
160
+ | 0.5587 | 2.8976 | 27000 | 0.5597 | 0.6997 |
161
+ | 0.5368 | 2.9244 | 27250 | 0.5597 | 0.7008 |
162
+ | 0.5624 | 2.9513 | 27500 | 0.5592 | 0.7008 |
163
+ | 0.571 | 2.9781 | 27750 | 0.5590 | 0.7005 |
164
+
165
+
166
+ ### Framework versions
167
+
168
+ - Transformers 4.41.2
169
+ - Pytorch 2.0.1+cu118
170
+ - Datasets 2.19.2
171
+ - Tokenizers 0.19.1
config.json ADDED
@@ -0,0 +1,41 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "gpt2",
3
+ "activation_function": "gelu_new",
4
+ "architectures": [
5
+ "GPT2ForSequenceClassification"
6
+ ],
7
+ "attn_pdrop": 0.1,
8
+ "bos_token_id": 50256,
9
+ "embd_pdrop": 0.1,
10
+ "eos_token_id": 50256,
11
+ "initializer_range": 0.02,
12
+ "layer_norm_epsilon": 1e-05,
13
+ "model_type": "gpt2",
14
+ "n_ctx": 1024,
15
+ "n_embd": 768,
16
+ "n_head": 12,
17
+ "n_inner": null,
18
+ "n_layer": 12,
19
+ "n_positions": 1024,
20
+ "pad_token_id": 50256,
21
+ "problem_type": "single_label_classification",
22
+ "reorder_and_upcast_attn": false,
23
+ "resid_pdrop": 0.1,
24
+ "scale_attn_by_inverse_layer_idx": false,
25
+ "scale_attn_weights": true,
26
+ "summary_activation": null,
27
+ "summary_first_dropout": 0.1,
28
+ "summary_proj_to_labels": true,
29
+ "summary_type": "cls_index",
30
+ "summary_use_proj": true,
31
+ "task_specific_params": {
32
+ "text-generation": {
33
+ "do_sample": true,
34
+ "max_length": 50
35
+ }
36
+ },
37
+ "torch_dtype": "float32",
38
+ "transformers_version": "4.41.2",
39
+ "use_cache": true,
40
+ "vocab_size": 50257
41
+ }
model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:0d6e3c523722ef07337cb168a16c58564cc3e070e65e1c6b99302a9fb13bcd12
3
+ size 497780432
training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c9b14bcee9cf13cae4f4900dfe990046d5286bd8a7712b1e4761cd080936a7b4
3
+ size 4603