File size: 1,518 Bytes
0033a9a
 
 
 
 
 
a98ee5c
 
 
6d0405f
 
0033a9a
 
 
 
 
 
 
f94e606
a458698
 
 
0033a9a
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
a98ee5c
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
---
tags:
- generated_from_trainer
model-index:
- name: TinyIG
  results: []
language:
- en
pipeline_tag: text-generation
datasets:
- nampdn-ai/tiny-textbooks
---

<!-- 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. -->

# TinyIG

This model is a pre-trained version of LLama2  on the [nampdn-ai/tiny-textbooks](https://huggingface.co/datasets/nampdn-ai/tiny-textbooks) dataset.

Inspired by [TinyStories](https://arxiv.org/abs/2305.07759).

It achieves the following results on the evaluation set:
- Loss: 3.5677

## 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: 0.0005
- train_batch_size: 2
- eval_batch_size: 1
- seed: 42
- gradient_accumulation_steps: 2
- total_train_batch_size: 4
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: cosine
- lr_scheduler_warmup_steps: 500
- num_epochs: 3
- mixed_precision_training: Native AMP

### Training results

| Training Loss | Epoch | Step  | Validation Loss |
|:-------------:|:-----:|:-----:|:---------------:|
| 3.4647        | 2.02  | 30000 | 3.5677          |


### Framework versions

- Transformers 4.35.2
- Pytorch 2.1.0+cu121
- Datasets 2.16.1
- Tokenizers 0.15.0