PEFT
Not-For-All-Audiences
File size: 2,176 Bytes
cc5aa64
7aa2af3
6527e25
7aa2af3
 
cc5aa64
7aa2af3
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
f30968e
 
 
 
 
 
 
 
 
 
7aa2af3
 
 
 
 
 
 
 
 
 
f30968e
 
 
 
 
7aa2af3
 
6527e25
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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
---
license: agpl-3.0
library_name: peft
tags:
- not-for-all-audiences
---

# Pippa-13b-qlora

This is a repository of my Llama-2-13b Qlora checkpoints of the [PIPPA-13b-ShareGPT](https://huggingface.co/datasets/kingbri/PIPPA-ShareGPT) dataset.

You can read more about the dataset on its relevant page. It's a ShareGPT reformat of the [PIPPA dataset](https://huggingface.co/datasets/PygmalionAI/PIPPA) by PygmalionAI. The reformat was done to allow for axolotl compatability.

### Architecture

- **Model Architecture**: Llama-2-13b
- **Training Algorithm**: QLora
- **Dataset Used**: PIPPA-ShareGPT (pippa_sharegpt_trimmed.jsonl)

### Training Details

- **Dataset**: [PIPPA-ShareGPT](https://huggingface.co/datasets/kingbri/PIPPA-ShareGPT)
- **Datset type**: ShareGPT
- **Training Parameters**: [See Here](https://gist.github.com/bdashore3/55ae04892f31609f2c3779c4a8a55408)
- **Training Environment**: Axolotl
- **sequence_len**: 4096

## Instruct Format

ShareGPT gets converted to vicuna format. The dataset uses modified roles of `USER` and `CHARACTER` instead of `USER` and `ASSISTANT`.

```
SYSTEM: Enter roleplay mode...
USER: {prompt}
CHARACTER:
```

## Notes

This Qlora was produced as an experiment to see how the public version of PIPPA can affect a model. As a result, I have no idea if this lora is of great quality or absolute garbage.

## Acknowledgments

Thanks to:
- PygmalionAI: The creators of the PIPPA dataset
- Axolotl: Finetuning suite

## Donate?
All my infrastructure and cloud expenses are paid out of pocket. If you'd like to donate, you can do so here: [https://ko-fi.com/kingbri](https://ko-fi.com/kingbri)

You should not feel obligated to donate, but if you do, I'd appreciate it.

## Axolotl stuff

## Training procedure


The following `bitsandbytes` quantization config was used during training:
- quant_method: bitsandbytes
- load_in_8bit: False
- load_in_4bit: True
- llm_int8_threshold: 6.0
- llm_int8_skip_modules: None
- llm_int8_enable_fp32_cpu_offload: False
- llm_int8_has_fp16_weight: False
- bnb_4bit_quant_type: nf4
- bnb_4bit_use_double_quant: True
- bnb_4bit_compute_dtype: bfloat16
### Framework versions


- PEFT 0.6.0.dev0