|
--- |
|
datasets: |
|
- PygmalionAI/PIPPA |
|
- lemonilia/LimaRP |
|
--- |
|
|
|
## Gen Settings & Prompting |
|
|
|
https://rentry.org/tsukasamodel |
|
|
|
## GPTQ |
|
|
|
2048 sequence length |
|
|
|
VMware/open-instruct dataset |
|
|
|
## Training |
|
|
|
[axolotl](https://github.com/OpenAccess-AI-Collective/axolotl) was used for training |
|
on a 8x nvidia a40 gpu cluster. |
|
|
|
the a40 GPU cluster has been graciously provided by [Arc Compute](https://www.arccompute.io/). |
|
|
|
rank 16 qlora (all modules) tune |
|
|
|
base model mistralai/Mistral-7B-v0.1 tuned on koishi commit 6e675d1 for one epoch |
|
|
|
then tuned on pippa 6412b0c for one epoch (metharme completion) |
|
|
|
then tuned on limarp Version 2023-10-19 for 2 epochs in metharme completion format |
|
|
