File size: 4,105 Bytes
ec5e7b3
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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
100
101
102
103
104
105
106
107
108
109
110
111
---
library_name: peft
license: apache-2.0
base_model: mistral-community/pixtral-12b
tags:
- generated_from_trainer
model-index:
- name: pixtral-12b-transformers-v0
  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. -->

# pixtral-12b-transformers-v0

This model is a fine-tuned version of [mistral-community/pixtral-12b](https://huggingface.co/mistral-community/pixtral-12b) on an unknown dataset.
It achieves the following results on the evaluation set:
- Loss: 0.0344

## 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: 1
- eval_batch_size: 1
- seed: 42
- gradient_accumulation_steps: 4
- total_train_batch_size: 4
- optimizer: Use OptimizerNames.PAGED_ADAMW_8BIT with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
- lr_scheduler_type: constant
- lr_scheduler_warmup_ratio: 0.03
- num_epochs: 10

### Training results

| Training Loss | Epoch  | Step | Validation Loss |
|:-------------:|:------:|:----:|:---------------:|
| 0.0752        | 0.1998 | 50   | 0.0506          |
| 0.035         | 0.3996 | 100  | 0.0443          |
| 0.0318        | 0.5994 | 150  | 0.0392          |
| 0.0296        | 0.7992 | 200  | 0.0329          |
| 0.0239        | 0.9990 | 250  | 0.0300          |
| 0.022         | 1.1958 | 300  | 0.0301          |
| 0.0201        | 1.3956 | 350  | 0.0343          |
| 0.0225        | 1.5954 | 400  | 0.0296          |
| 0.0197        | 1.7952 | 450  | 0.0272          |
| 0.0174        | 1.9950 | 500  | 0.0285          |
| 0.0167        | 2.1918 | 550  | 0.0266          |
| 0.0159        | 2.3916 | 600  | 0.0267          |
| 0.0141        | 2.5914 | 650  | 0.0262          |
| 0.017         | 2.7912 | 700  | 0.0270          |
| 0.0138        | 2.9910 | 750  | 0.0263          |
| 0.0126        | 3.1878 | 800  | 0.0272          |
| 0.0123        | 3.3876 | 850  | 0.0260          |
| 0.0111        | 3.5874 | 900  | 0.0249          |
| 0.011         | 3.7872 | 950  | 0.0257          |
| 0.015         | 3.9870 | 1000 | 0.0258          |
| 0.0094        | 4.1838 | 1050 | 0.0275          |
| 0.0093        | 4.3836 | 1100 | 0.0275          |
| 0.01          | 4.5834 | 1150 | 0.0272          |
| 0.0115        | 4.7832 | 1200 | 0.0269          |
| 0.012         | 4.9830 | 1250 | 0.0250          |
| 0.0079        | 5.1798 | 1300 | 0.0284          |
| 0.0081        | 5.3796 | 1350 | 0.0309          |
| 0.0096        | 5.5794 | 1400 | 0.0279          |
| 0.0094        | 5.7792 | 1450 | 0.0280          |
| 0.009         | 5.9790 | 1500 | 0.0276          |
| 0.0081        | 6.1758 | 1550 | 0.0329          |
| 0.008         | 6.3756 | 1600 | 0.0275          |
| 0.0076        | 6.5754 | 1650 | 0.0271          |
| 0.008         | 6.7752 | 1700 | 0.0292          |
| 0.0082        | 6.9750 | 1750 | 0.0270          |
| 0.005         | 7.1718 | 1800 | 0.0315          |
| 0.0069        | 7.3716 | 1850 | 0.0271          |
| 0.0063        | 7.5714 | 1900 | 0.0305          |
| 0.007         | 7.7712 | 1950 | 0.0282          |
| 0.006         | 7.9710 | 2000 | 0.0293          |
| 0.0059        | 8.1678 | 2050 | 0.0302          |
| 0.0042        | 8.3676 | 2100 | 0.0332          |
| 0.0053        | 8.5674 | 2150 | 0.0297          |
| 0.0053        | 8.7672 | 2200 | 0.0327          |
| 0.0057        | 8.9670 | 2250 | 0.0300          |
| 0.0066        | 9.1638 | 2300 | 0.0318          |
| 0.0042        | 9.3636 | 2350 | 0.0316          |
| 0.0037        | 9.5634 | 2400 | 0.0342          |
| 0.0066        | 9.7632 | 2450 | 0.0323          |
| 0.0045        | 9.9630 | 2500 | 0.0344          |


### Framework versions

- PEFT 0.14.0
- Transformers 4.48.3
- Pytorch 2.5.1+cu124
- Datasets 3.3.2
- Tokenizers 0.21.0