File size: 3,119 Bytes
f1de845
 
 
 
 
 
 
 
 
860a14a
 
f1de845
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
860a14a
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
---
library_name: transformers
license: apache-2.0
base_model: EleutherAI/pythia-70m
tags:
- generated_from_trainer
model-index:
- name: pythia_70m_sft
  results: []
datasets:
- tatsu-lab/alpaca_farm
---

<!-- 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. -->

# pythia_70m_sft

This model is a fine-tuned version of [EleutherAI/pythia-70m](https://huggingface.co/EleutherAI/pythia-70m) on an unknown dataset.
It achieves the following results on the evaluation set:
- Loss: 2.4023

## 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: 2e-05
- train_batch_size: 8
- eval_batch_size: 8
- seed: 42
- optimizer: Use adamw_torch with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
- lr_scheduler_type: linear
- num_epochs: 3
- mixed_precision_training: Native AMP

### Training results

| Training Loss | Epoch  | Step | Validation Loss |
|:-------------:|:------:|:----:|:---------------:|
| 2.8612        | 0.0889 | 100  | 2.7705          |
| 2.7799        | 0.1778 | 200  | 2.7116          |
| 2.754         | 0.2667 | 300  | 2.6827          |
| 2.7035        | 0.3556 | 400  | 2.6482          |
| 2.6667        | 0.4444 | 500  | 2.6296          |
| 2.6568        | 0.5333 | 600  | 2.6048          |
| 2.6233        | 0.6222 | 700  | 2.5862          |
| 2.5956        | 0.7111 | 800  | 2.5790          |
| 2.5635        | 0.8    | 900  | 2.5436          |
| 2.5469        | 0.8889 | 1000 | 2.5445          |
| 2.499         | 0.9778 | 1100 | 2.5155          |
| 2.3677        | 1.0667 | 1200 | 2.5158          |
| 2.326         | 1.1556 | 1300 | 2.5018          |
| 2.3132        | 1.2444 | 1400 | 2.5023          |
| 2.3401        | 1.3333 | 1500 | 2.5026          |
| 2.3027        | 1.4222 | 1600 | 2.4953          |
| 2.3223        | 1.5111 | 1700 | 2.4804          |
| 2.3186        | 1.6    | 1800 | 2.4776          |
| 2.3187        | 1.6889 | 1900 | 2.4709          |
| 2.3102        | 1.7778 | 2000 | 2.4618          |
| 2.3129        | 1.8667 | 2100 | 2.4526          |
| 2.287         | 1.9556 | 2200 | 2.4456          |
| 2.1946        | 2.0444 | 2300 | 2.4423          |
| 2.1757        | 2.1333 | 2400 | 2.4408          |
| 2.1308        | 2.2222 | 2500 | 2.4383          |
| 2.1475        | 2.3111 | 2600 | 2.4283          |
| 2.155         | 2.4    | 2700 | 2.4231          |
| 2.1197        | 2.4889 | 2800 | 2.4222          |
| 2.1225        | 2.5778 | 2900 | 2.4173          |
| 2.1196        | 2.6667 | 3000 | 2.4118          |
| 2.146         | 2.7556 | 3100 | 2.4075          |
| 2.1361        | 2.8444 | 3200 | 2.4046          |
| 2.0965        | 2.9333 | 3300 | 2.4023          |


### Framework versions

- Transformers 4.46.2
- Pytorch 2.5.1+cu121
- Datasets 3.1.0
- Tokenizers 0.20.3