zgce's picture
Update README.md
9289640
|
raw
history blame
649 Bytes
metadata
license: apache-2.0

Mistral-RP-0.1 7B EXL2-3.5bpw

Description

  • 3.5bpw per weight
  • 应 Surdo 要求为小显存做的试验模型

I converted the model using the convert.py script from the exllamav2 repo: https://github.com/turboderp/exllamav2

Its documentation: https://github.com/turboderp/exllamav2/blob/master/doc/convert.md

I used the WikiText-2-v1 dataset for calibration: https://huggingface.co/datasets/wikitext/blob/refs%2Fconvert%2Fparquet/wikitext-2-v1/test/0000.parquet