File size: 1,960 Bytes
c403c7c
8d61982
 
 
 
 
 
 
 
c403c7c
 
8d61982
 
c403c7c
8d61982
c403c7c
8d61982
 
 
c403c7c
8d61982
c403c7c
8d61982
c403c7c
8d61982
c403c7c
8d61982
c403c7c
8d61982
c403c7c
8d61982
c403c7c
8d61982
c403c7c
8d61982
c403c7c
8d61982
 
 
 
 
 
 
 
 
 
 
 
c403c7c
8d61982
c403c7c
8d61982
 
 
 
 
 
 
 
 
 
 
 
c403c7c
 
8d61982
c403c7c
8d61982
 
 
 
 
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
---
license: apache-2.0
library_name: peft
tags:
- generated_from_trainer
base_model: TheBloke/Mistral-7B-Instruct-v0.2-GPTQ
model-index:
- name: Mygpt
  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. -->

# Mygpt

This model is a fine-tuned version of [TheBloke/Mistral-7B-Instruct-v0.2-GPTQ](https://huggingface.co/TheBloke/Mistral-7B-Instruct-v0.2-GPTQ) on an unknown dataset.
It achieves the following results on the evaluation set:
- Loss: 1.9196

## 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: 0.0002
- train_batch_size: 4
- eval_batch_size: 4
- seed: 42
- gradient_accumulation_steps: 4
- total_train_batch_size: 16
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_steps: 2
- num_epochs: 10
- mixed_precision_training: Native AMP

### Training results

| Training Loss | Epoch  | Step | Validation Loss |
|:-------------:|:------:|:----:|:---------------:|
| 4.5937        | 0.9231 | 3    | 3.9657          |
| 4.0542        | 1.8462 | 6    | 3.4486          |
| 3.4784        | 2.7692 | 9    | 2.9966          |
| 2.2705        | 4.0    | 13   | 2.5821          |
| 2.7175        | 4.9231 | 16   | 2.3590          |
| 2.4291        | 5.8462 | 19   | 2.1768          |
| 2.2157        | 6.7692 | 22   | 2.0523          |
| 1.5752        | 8.0    | 26   | 1.9837          |
| 2.0503        | 8.9231 | 29   | 1.9356          |
| 1.4184        | 9.2308 | 30   | 1.9196          |


### Framework versions

- PEFT 0.10.0
- Transformers 4.40.0
- Pytorch 2.1.0+cu121
- Datasets 2.19.0
- Tokenizers 0.19.1