File size: 2,449 Bytes
9160844
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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
---
license: llama2
library_name: peft
tags:
- generated_from_trainer
base_model: codellama/CodeLlama-7b-Instruct-hf
model-index:
- name: codellama-hugcoder-v2
  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. -->

# codellama-hugcoder-v2

This model is a fine-tuned version of [codellama/CodeLlama-7b-Instruct-hf](https://huggingface.co/codellama/CodeLlama-7b-Instruct-hf) on an unknown dataset.
It achieves the following results on the evaluation set:
- Loss: 0.4602

## 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: 16
- eval_batch_size: 16
- seed: 11
- gradient_accumulation_steps: 4
- total_train_batch_size: 64
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: cosine
- lr_scheduler_warmup_ratio: 0.1
- training_steps: 2000

### Training results

| Training Loss | Epoch | Step | Validation Loss |
|:-------------:|:-----:|:----:|:---------------:|
| 0.5827        | 0.05  | 100  | 0.6188          |
| 0.5648        | 0.1   | 200  | 0.5643          |
| 0.5316        | 0.15  | 300  | 0.5359          |
| 0.5008        | 0.2   | 400  | 0.5202          |
| 0.4919        | 0.25  | 500  | 0.5042          |
| 0.4665        | 0.3   | 600  | 0.4962          |
| 0.4324        | 0.35  | 700  | 0.4856          |
| 0.4179        | 0.4   | 800  | 0.4804          |
| 0.3614        | 0.45  | 900  | 0.4738          |
| 0.4344        | 0.5   | 1000 | 0.4703          |
| 0.3473        | 0.55  | 1100 | 0.4672          |
| 0.3777        | 0.6   | 1200 | 0.4648          |
| 0.3378        | 0.65  | 1300 | 0.4620          |
| 0.3744        | 0.7   | 1400 | 0.4614          |
| 0.3834        | 0.75  | 1500 | 0.4610          |
| 0.2859        | 0.8   | 1600 | 0.4603          |
| 0.3787        | 0.85  | 1700 | 0.4598          |
| 0.3132        | 0.9   | 1800 | 0.4597          |
| 0.3607        | 0.95  | 1900 | 0.4595          |
| 0.3684        | 1.0   | 2000 | 0.4602          |


### Framework versions

- PEFT 0.8.2
- Transformers 4.38.0.dev0
- Pytorch 2.1.0+cu121
- Datasets 2.16.1
- Tokenizers 0.15.1