Skip to content

fzzhang

pearl_gsm8k_quantized_merged

fzzhang/pearl_gsm8k_quantized_merged

pearl_gsm8k_quantized_merged is a text generation model from fzzhang. Use it when you need the model to write or continue text. It is set up for transformers.

Downloads · 30 days

9

16% of all-time downloads

All-time downloads

56

Public

Parameters

7.2B

29 GB on disk

Likes

0

Public

Hugging Face

Repo makeup

Click a slice to open those files.

.safetensors29 GB · 100%

At a glance

Task
Text Generation
Library
transformers
Model type
mistral
Access
Public
Created
Feb 19, 2024
Updated
Feb 19, 2024
SHA
447602e9

Try a prompt

Task
Text Generation
Library
transformers
Type
mistral
Created
Feb 19, 2024
Updated
Feb 19, 2024