File size: 2,329 Bytes
fe9c437
 
 
258753f
fe9c437
 
 
258753f
 
 
 
 
fe9c437
 
 
 
 
 
 
 
 
 
258753f
fe9c437
258753f
 
 
 
 
 
 
 
 
fe9c437
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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
---
base_model: Minbyul/llama2-7b-wo-medication_qa-sft
tags:
- alignment-handbook
- trl
- dpo
- generated_from_trainer
- trl
- dpo
- generated_from_trainer
datasets:
- HuggingFaceH4/ultrafeedback_binarized
model-index:
- name: llama2-7b-dpo-full-sft-wo-medication_qa
  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. -->

# llama2-7b-dpo-full-sft-wo-medication_qa

This model is a fine-tuned version of [Minbyul/llama2-7b-wo-medication_qa-sft](https://huggingface.co/Minbyul/llama2-7b-wo-medication_qa-sft) on the HuggingFaceH4/ultrafeedback_binarized dataset.
It achieves the following results on the evaluation set:
- Loss: 0.4396
- Rewards/chosen: -0.1779
- Rewards/rejected: -1.2468
- Rewards/accuracies: 0.9500
- Rewards/margins: 1.0689
- Logps/rejected: -650.3414
- Logps/chosen: -477.8221
- Logits/rejected: -0.4720
- Logits/chosen: -0.4277

## 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-07
- train_batch_size: 8
- eval_batch_size: 8
- seed: 42
- distributed_type: multi-GPU
- num_devices: 4
- gradient_accumulation_steps: 2
- total_train_batch_size: 64
- total_eval_batch_size: 32
- 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 | Logits/chosen | Logits/rejected | Logps/chosen | Logps/rejected | Validation Loss | Rewards/accuracies | Rewards/chosen | Rewards/margins | Rewards/rejected |
|:-------------:|:-----:|:----:|:-------------:|:---------------:|:------------:|:--------------:|:---------------:|:------------------:|:--------------:|:---------------:|:----------------:|
| 0.2708        | 0.76  | 100  | -0.4292       | -0.4708         | -476.1255    | -635.9033      | 0.4682          | 0.9250             | -0.1609        | 0.9415          | -1.1024          |


### Framework versions

- Transformers 4.39.0.dev0
- Pytorch 2.1.2
- Datasets 2.14.6
- Tokenizers 0.15.2