Safetensors
qwen2
File size: 757 Bytes
6748fb6
 
 
 
 
 
 
 
 
cd67c2b
9b3492c
cd67c2b
 
9b3492c
 
7cd612f
 
 
 
 
 
 
 
9b3492c
 
6748fb6
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
---
license: agpl-3.0
datasets:
- kalomaze/Opus_Instruct_25k
- MangoHQ/Claude-Data-Anon-Killed
---

Experimental Finetune of Qwen 1.5B with Kalomaze Opus Instruct first and then Claude C2 logs. Model may produce NSFW results. 


First run on Opus Instruct -> https://raw.githubusercontent.com/MangyMango/AxolotlConfigs/main/step1two.yaml


Second run on C2 logs -> https://raw.githubusercontent.com/MangyMango/AxolotlConfigs/main/step1two.yaml

Model uses Alpaca formatting
```
### Instruction:
<Prompt>
### Input:
<Insert Context Here>
### Response:
```


[<img src="https://raw.githubusercontent.com/axolotl-ai-cloud/axolotl/main/image/axolotl-badge-web.png" alt="Built with Axolotl" width="200" height="32"/>](https://github.com/axolotl-ai-cloud/axolotl)