File size: 706 Bytes
5c08af6
 
 
1b75e7c
f6a28b6
23759d9
f6a28b6
 
369ab27
 
298e285
1
2
3
4
5
6
7
8
9
10
11
---
license: other
---

FP16 model of airoboros 70b 1.4.1 (https://huggingface.co/jondurbin/airoboros-l2-70b-gpt4-1.4.1) from .bin to . safetensors, to be used to quant on exllama2.

It can also be used to load faster at FP16 using transformers.

There is a script inside bin2safetensors folder, that you can use to convert .bin files into .safetensor ones for other models.

Also, I included 2 measurements.json to be used to quant. First one (called old) was made with https://huggingface.co/datasets/EleutherAI/the_pile_deduplicated/blob/refs%2Fconvert%2Fparquet/default/train/0000.parquet and first exllamav2 version, and the second one is a cleaned pippa, with good formatting on 17/09/2023 exllamav2.