File size: 3,281 Bytes
541807f
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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
87
88
89
90
91
92
93
94
95
96
97
98
99
---
license: llama3.1
base_model: meta-llama/Llama-3.1-8B-Instruct
tags:
- trl
- sft
- generated_from_trainer
model-index:
- name: Llama-3.1-8B-Instruct-EI1-2ep-sft
  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. -->

# Llama-3.1-8B-Instruct-EI1-2ep-sft

This model is a fine-tuned version of [meta-llama/Llama-3.1-8B-Instruct](https://huggingface.co./meta-llama/Llama-3.1-8B-Instruct) on an unknown dataset.
It achieves the following results on the evaluation set:
- Loss: 0.3970

## 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: 6e-06
- train_batch_size: 1
- eval_batch_size: 8
- seed: 42
- distributed_type: multi-GPU
- num_devices: 32
- gradient_accumulation_steps: 2
- total_train_batch_size: 64
- total_eval_batch_size: 256
- optimizer: Adam with betas=(0.9,0.95) and epsilon=1e-08
- lr_scheduler_type: cosine
- num_epochs: 2.0

### Training results

| Training Loss | Epoch  | Step | Validation Loss |
|:-------------:|:------:|:----:|:---------------:|
| No log        | 0.0562 | 100  | 0.5980          |
| No log        | 0.1124 | 200  | 0.5609          |
| No log        | 0.1685 | 300  | 0.5369          |
| No log        | 0.2247 | 400  | 0.5156          |
| 0.5582        | 0.2809 | 500  | 0.4955          |
| 0.5582        | 0.3371 | 600  | 0.4795          |
| 0.5582        | 0.3933 | 700  | 0.4655          |
| 0.5582        | 0.4494 | 800  | 0.4522          |
| 0.5582        | 0.5056 | 900  | 0.4433          |
| 0.448         | 0.5618 | 1000 | 0.4355          |
| 0.448         | 0.6180 | 1100 | 0.4295          |
| 0.448         | 0.6742 | 1200 | 0.4252          |
| 0.448         | 0.7303 | 1300 | 0.4200          |
| 0.448         | 0.7865 | 1400 | 0.4159          |
| 0.4123        | 0.8427 | 1500 | 0.4124          |
| 0.4123        | 0.8989 | 1600 | 0.4098          |
| 0.4123        | 0.9551 | 1700 | 0.4075          |
| 0.4123        | 1.0112 | 1800 | 0.4086          |
| 0.4123        | 1.0674 | 1900 | 0.4075          |
| 0.3815        | 1.1236 | 2000 | 0.4069          |
| 0.3815        | 1.1798 | 2100 | 0.4054          |
| 0.3815        | 1.2360 | 2200 | 0.4043          |
| 0.3815        | 1.2921 | 2300 | 0.4029          |
| 0.3815        | 1.3483 | 2400 | 0.4022          |
| 0.3532        | 1.4045 | 2500 | 0.4012          |
| 0.3532        | 1.4607 | 2600 | 0.4002          |
| 0.3532        | 1.5169 | 2700 | 0.3996          |
| 0.3532        | 1.5730 | 2800 | 0.3986          |
| 0.3532        | 1.6292 | 2900 | 0.3982          |
| 0.35          | 1.6854 | 3000 | 0.3978          |
| 0.35          | 1.7416 | 3100 | 0.3975          |
| 0.35          | 1.7978 | 3200 | 0.3971          |
| 0.35          | 1.8539 | 3300 | 0.3971          |
| 0.35          | 1.9101 | 3400 | 0.3970          |
| 0.3468        | 1.9663 | 3500 | 0.3970          |


### Framework versions

- Transformers 4.43.4
- Pytorch 2.4.0+cu121
- Datasets 3.0.1
- Tokenizers 0.19.1