File size: 2,609 Bytes
53f7f4a
 
 
7e4e2b5
53f7f4a
 
 
 
7e4e2b5
 
53f7f4a
 
 
 
 
 
 
 
 
 
7e4e2b5
53f7f4a
7e4e2b5
 
 
 
 
 
 
 
 
53f7f4a
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
a9f1e80
 
 
53f7f4a
 
 
 
 
 
 
 
 
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
---
library_name: peft
tags:
- alignment-handbook
- trl
- dpo
- generated_from_trainer
base_model: norallm/normistral-7b-warm
datasets:
- hugodk-sch/aftonposten_title_prefs
model-index:
- name: ap-normistral-7b-align-scan
  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. -->

# ap-normistral-7b-align-scan

This model is a fine-tuned version of [data/ap-normistral-7b-sft-qlora](https://huggingface.co/data/ap-normistral-7b-sft-qlora) on the hugodk-sch/aftonposten_title_prefs dataset.
It achieves the following results on the evaluation set:
- Loss: 1.7667
- Rewards/chosen: 0.0466
- Rewards/rejected: 0.0526
- Rewards/accuracies: 0.4880
- Rewards/margins: -0.0060
- Logps/rejected: -35.9008
- Logps/chosen: -32.3849
- Logits/rejected: 98.9812
- Logits/chosen: 98.9874

## 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: 4
- eval_batch_size: 8
- seed: 42
- distributed_type: multi-GPU
- gradient_accumulation_steps: 2
- total_train_batch_size: 8
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: cosine
- lr_scheduler_warmup_ratio: 0.1
- 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 |
|:-------------:|:-----:|:----:|:---------------:|:--------------:|:----------------:|:------------------:|:---------------:|:--------------:|:------------:|:---------------:|:-------------:|
| 1.5488        | 0.26  | 100  | 1.7743          | -0.0738        | -0.1533          | 0.5378             | 0.0795          | -36.1581       | -32.5354     | 98.8023         | 98.8147       |
| 3.6133        | 0.52  | 200  | 1.8922          | -0.0939        | -0.1399          | 0.5166             | 0.0460          | -36.1414       | -32.5606     | 99.0488         | 99.0652       |
| 2.1193        | 0.78  | 300  | 1.5939          | 0.0537         | 0.0702           | 0.5191             | -0.0166         | -35.8787       | -32.3761     | 98.9855         | 98.9917       |


### Framework versions

- PEFT 0.10.0
- Transformers 4.39.0.dev0
- Pytorch 2.1.2+cu121
- Datasets 2.14.6
- Tokenizers 0.15.1