File size: 4,520 Bytes
61d0bcd
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
---
license: mit
library_name: peft
tags:
- trl
- reward-trainer
- generated_from_trainer
base_model: openai-community/gpt2-large
metrics:
- accuracy
model-index:
- name: RM-HH-AllMix_helpful_gpt3_loraR64_20000_gpt2-large_shuffleTrue_extractchosenFalse
  results: []
---

<!-- This model card has been generated automatically according to the information the Trainer had access to. You
should probably proofread and complete it, then remove this comment. -->

# RM-HH-AllMix_helpful_gpt3_loraR64_20000_gpt2-large_shuffleTrue_extractchosenFalse

This model is a fine-tuned version of [openai-community/gpt2-large](https://huggingface.co./openai-community/gpt2-large) on the None dataset.
It achieves the following results on the evaluation set:
- Loss: 0.4894
- Accuracy: 0.7351

## Model description

More information needed

## Intended uses & limitations

More information needed

## Training and evaluation data

More information needed

## Training procedure

### Training hyperparameters

The following hyperparameters were used during training:
- learning_rate: 1.41e-05
- train_batch_size: 1
- eval_batch_size: 8
- seed: 42
- gradient_accumulation_steps: 4
- total_train_batch_size: 4
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 2.0

### Training results

| Training Loss | Epoch | Step  | Validation Loss | Accuracy |
|:-------------:|:-----:|:-----:|:---------------:|:--------:|
| 0.6628        | 0.04  | 250   | 0.6390          | 0.6277   |
| 0.598         | 0.08  | 500   | 0.5673          | 0.6933   |
| 0.5479        | 0.13  | 750   | 0.5415          | 0.7076   |
| 0.5397        | 0.17  | 1000  | 0.5308          | 0.7110   |
| 0.5094        | 0.21  | 1250  | 0.5261          | 0.7159   |
| 0.5142        | 0.25  | 1500  | 0.5203          | 0.7193   |
| 0.5414        | 0.29  | 1750  | 0.5161          | 0.7197   |
| 0.5189        | 0.33  | 2000  | 0.5131          | 0.7189   |
| 0.5151        | 0.38  | 2250  | 0.5100          | 0.7216   |
| 0.4942        | 0.42  | 2500  | 0.5089          | 0.7208   |
| 0.5067        | 0.46  | 2750  | 0.5057          | 0.7216   |
| 0.5026        | 0.5   | 3000  | 0.5041          | 0.7238   |
| 0.4926        | 0.54  | 3250  | 0.5038          | 0.7265   |
| 0.4931        | 0.59  | 3500  | 0.5022          | 0.7310   |
| 0.4946        | 0.63  | 3750  | 0.4993          | 0.7329   |
| 0.5058        | 0.67  | 4000  | 0.4968          | 0.7313   |
| 0.4822        | 0.71  | 4250  | 0.4963          | 0.7306   |
| 0.4924        | 0.75  | 4500  | 0.4961          | 0.7329   |
| 0.4654        | 0.8   | 4750  | 0.4959          | 0.7302   |
| 0.4924        | 0.84  | 5000  | 0.4971          | 0.7310   |
| 0.4674        | 0.88  | 5250  | 0.4948          | 0.7310   |
| 0.4704        | 0.92  | 5500  | 0.4950          | 0.7336   |
| 0.5089        | 0.96  | 5750  | 0.4905          | 0.7306   |
| 0.4673        | 1.0   | 6000  | 0.4929          | 0.7313   |
| 0.4594        | 1.05  | 6250  | 0.4932          | 0.7291   |
| 0.479         | 1.09  | 6500  | 0.4919          | 0.7332   |
| 0.5112        | 1.13  | 6750  | 0.4895          | 0.7355   |
| 0.4794        | 1.17  | 7000  | 0.4888          | 0.7332   |
| 0.5188        | 1.21  | 7250  | 0.4881          | 0.7340   |
| 0.4541        | 1.26  | 7500  | 0.4892          | 0.7359   |
| 0.4617        | 1.3   | 7750  | 0.4898          | 0.7366   |
| 0.4747        | 1.34  | 8000  | 0.4898          | 0.7362   |
| 0.4834        | 1.38  | 8250  | 0.4893          | 0.7389   |
| 0.4954        | 1.42  | 8500  | 0.4875          | 0.7385   |
| 0.5029        | 1.47  | 8750  | 0.4875          | 0.7385   |
| 0.4742        | 1.51  | 9000  | 0.4872          | 0.7400   |
| 0.4802        | 1.55  | 9250  | 0.4884          | 0.7393   |
| 0.5009        | 1.59  | 9500  | 0.4877          | 0.7400   |
| 0.4619        | 1.63  | 9750  | 0.4875          | 0.7396   |
| 0.4433        | 1.67  | 10000 | 0.4902          | 0.7404   |
| 0.4844        | 1.72  | 10250 | 0.4903          | 0.7400   |
| 0.4337        | 1.76  | 10500 | 0.4917          | 0.7400   |
| 0.4897        | 1.8   | 10750 | 0.4901          | 0.7396   |
| 0.4783        | 1.84  | 11000 | 0.4894          | 0.7366   |
| 0.4929        | 1.88  | 11250 | 0.4892          | 0.7359   |
| 0.4776        | 1.93  | 11500 | 0.4891          | 0.7362   |
| 0.4574        | 1.97  | 11750 | 0.4894          | 0.7351   |


### Framework versions

- PEFT 0.9.0
- Transformers 4.38.2
- Pytorch 2.1.2
- Datasets 2.18.0
- Tokenizers 0.15.2