File size: 1,322 Bytes
e73eb09
3a591b5
d7256a2
 
e73eb09
 
 
 
 
 
51ff12b
e73eb09
 
 
 
 
ffed785
47c6828
ffed785
e73eb09
4f1bd87
ffed785
3c81959
ffed785
 
dab64fe
 
e73eb09
9a59065
1bb21cc
 
 
 
 
e73eb09
 
 
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
---
base_model: mistralai/Mistral-Small-24B-Instruct-2501
datasets:
  - ServiceNow-AI/R1-Distill-SFT
tags:
- text-generation-inference
- transformers
- unsloth
- mistral
- trl
- sft
license: apache-2.0
language:
- en
---

<picture>
<img alt="image" src="https://huggingface.co./tensopolis/assets/resolve/main/logo_512.png">
</picture>

## mistral-small-r1-tensopolis 

This model is a **reasoning** fine-tune of unsloth/**mistral-small-24b-instruct-2501**-unsloth-bnb-4bit. Trained in **1xA100** for about **100 hours**. Please refer to the base model and dataset for more information about license, prompt format, etc. 

Base model: [**mistralai/Mistral-Small-24B-Instruct-2501**](https://huggingface.co./mistralai/Mistral-Small-24B-Instruct-2501)

Dataset: [**ServiceNow-AI/R1-Distill-SFT**](https://huggingface.co./datasets/ServiceNow-AI/R1-Distill-SFT)

#### Basic Instruct Template (V7-Tekken)

```
<s>[SYSTEM_PROMPT]<system prompt>[/SYSTEM_PROMPT][INST]<user message>[/INST]<assistant response></s>[INST]<user message>[/INST]
```

This mistral model was trained 2x faster with [Unsloth](https://github.com/unslothai/unsloth) and Huggingface's TRL library.

[<img src="https://raw.githubusercontent.com/unslothai/unsloth/main/images/unsloth%20made%20with%20love.png" width="200"/>](https://github.com/unslothai/unsloth)