File size: 2,238 Bytes
d29e17e
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
36ecaff
 
 
0bbeacf
36ecaff
 
 
 
 
 
 
 
d29e17e
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
18f3a1d
 
36ecaff
d29e17e
 
 
 
 
 
b21560b
d29e17e
 
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
---
base_model: NousResearch/Meta-Llama-3-8B-Instruct
library_name: peft
license: other
tags:
- trl
- orpo
- generated_from_trainer
model-index:
- name: results
  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. -->

# results

This model is a fine-tuned version of [NousResearch/Meta-Llama-3-8B-Instruct](https://huggingface.co/NousResearch/Meta-Llama-3-8B-Instruct) on the None dataset.
It achieves the following results on the evaluation set:
- Loss: 0.1433
- Rewards/chosen: -0.0156
- Rewards/rejected: -1.1268
- Rewards/accuracies: 1.0
- Rewards/margins: 1.1112
- Logps/rejected: -11.2681
- Logps/chosen: -0.1556
- Logits/rejected: -0.2138
- Logits/chosen: 0.6631
- Nll Loss: 0.1433
- Log Odds Ratio: -0.0000
- Log Odds Chosen: 13.2065

## 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: 8e-05
- train_batch_size: 1
- eval_batch_size: 1
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_steps: 10
- num_epochs: 1

### Training results

| Training Loss | Epoch  | Step | Validation Loss | Rewards/chosen | Rewards/rejected | Rewards/accuracies | Rewards/margins | Logps/rejected | Logps/chosen | Logits/rejected | Logits/chosen | Nll Loss | Log Odds Ratio | Log Odds Chosen |
|:-------------:|:------:|:----:|:---------------:|:--------------:|:----------------:|:------------------:|:---------------:|:--------------:|:------------:|:---------------:|:-------------:|:--------:|:--------------:|:---------------:|
| 0.1869        | 0.5001 | 2175 | 0.1433          | -0.0156        | -1.1268          | 1.0                | 1.1112          | -11.2681       | -0.1556      | -0.2138         | 0.6631        | 0.1433   | -0.0000        | 13.2065         |


### Framework versions

- PEFT 0.11.1
- Transformers 4.42.3
- Pytorch 2.1.2+cu121
- Datasets 2.20.0
- Tokenizers 0.19.1