File size: 1,336 Bytes
310cde1
 
 
 
 
 
d91e7be
310cde1
 
 
 
 
ed26bc7
 
310cde1
 
 
 
 
 
 
9576c36
310cde1
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
ed26bc7
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
---
license: mit
tags:
- trl
- reward-trainer
- generated_from_trainer
base_model: OpenAssistant/reward-model-deberta-v3-base
metrics:
- accuracy
model-index:
- name: Reward_model
  results: []
datasets:
- Anthropic/hh-rlhf
---

<!-- 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. -->

# Reward_model

This model is a fine-tuned version of [OpenAssistant/reward-model-deberta-v3-base](https://huggingface.co/OpenAssistant/reward-model-deberta-v3-base) on [Anthropic/hh-rlhf](https://huggingface.co/datasets/Anthropic/hh-rlhf) dataset.
It achieves the following results on the evaluation set:
- Loss: 0.6931
- Accuracy: 1.0

### Training hyperparameters

The following hyperparameters were used during training:
- learning_rate: 5e-05
- train_batch_size: 12
- eval_batch_size: 2
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 1

### Training results

| Training Loss | Epoch | Step  | Validation Loss | Accuracy |
|:-------------:|:-----:|:-----:|:---------------:|:--------:|
| 0.6936        | 1.0   | 13400 | 0.6931          | 1.0      |


### Framework versions

- Transformers 4.39.3
- Pytorch 2.1.2
- Datasets 2.18.0
- Tokenizers 0.15.2