File size: 2,957 Bytes
05a3dde
 
 
 
 
b669047
05a3dde
 
 
b669047
 
 
 
 
05a3dde
 
 
 
 
 
 
 
 
 
b669047
05a3dde
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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
---
library_name: transformers
license: gemma
base_model: google/gemma-7b
tags:
- alignment-handbook
- trl
- orpo
- generated_from_trainer
- trl
- orpo
- generated_from_trainer
datasets:
- silviasapora/low_quality_dpo7k
model-index:
- name: gemma-7b-borpo-low-quality
  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. -->

# gemma-7b-borpo-low-quality

This model is a fine-tuned version of [google/gemma-7b](https://huggingface.co/google/gemma-7b) on the silviasapora/low_quality_dpo7k dataset.
It achieves the following results on the evaluation set:
- Loss: 1.5380
- Rewards/chosen: -0.0547
- Rewards/rejected: -0.0625
- Rewards/accuracies: 0.5468
- Rewards/margins: 0.0079
- Logps/rejected: -1.2508
- Logps/chosen: -1.0933
- Logits/rejected: 267.2346
- Logits/chosen: 296.6808
- Nll Loss: 1.4703
- Log Odds Ratio: -0.7039
- Log Odds Chosen: 0.2721

## 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: 5e-06
- train_batch_size: 2
- eval_batch_size: 1
- seed: 42
- distributed_type: multi-GPU
- num_devices: 4
- gradient_accumulation_steps: 4
- total_train_batch_size: 32
- total_eval_batch_size: 4
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: inverse_sqrt
- lr_scheduler_warmup_steps: 100
- num_epochs: 3

### 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 |
|:-------------:|:------:|:----:|:---------------:|:--------------:|:----------------:|:------------------:|:---------------:|:--------------:|:------------:|:---------------:|:-------------:|:--------:|:--------------:|:---------------:|
| 1.436         | 0.9955 | 167  | 1.4639          | -0.0502        | -0.0571          | 0.5540             | 0.0068          | -1.1413        | -1.0048      | 294.2689        | 322.9157      | 1.4152   | -0.6882        | 0.2192          |
| 1.0918        | 1.9970 | 335  | 1.4233          | -0.0501        | -0.0574          | 0.4964             | 0.0073          | -1.1475        | -1.0012      | 284.8744        | 313.3100      | 1.3661   | -0.7028        | 0.2209          |
| 0.576         | 2.9866 | 501  | 1.5380          | -0.0547        | -0.0625          | 0.5468             | 0.0079          | -1.2508        | -1.0933      | 267.2346        | 296.6808      | 1.4703   | -0.7039        | 0.2721          |


### Framework versions

- Transformers 4.44.2
- Pytorch 2.4.0+cu121
- Datasets 3.0.0
- Tokenizers 0.19.1