Rosetta 0.2

Model Details

Model Description

Base model is the Unsloth Mistral 7b at full percision (trained on 4bit). A 4 bit model is avali on earlier commits, and exists on newer commits as model.safetensors.

Developed by: smashmaster
Funded by: myself
Model type: Mistral 7B Finetune
Language(s) (NLP): TBD
License: All rights reserved (for now, def subject to change with better revisions of the model)
Finetuned from model: Mistral 7B (Unsloth 4bit)

Framework versions

PEFT 0.7.1

Downloads last month: -

Safetensors

Model size

4B params

Tensor type

F32

BF16

Inference Providers NEW

This model isn't deployed by any Inference Provider. 🙋 Ask for provider support

Model tree for smashmaster/Rosetta-v0.2

Base model

unsloth/mistral-7b-bnb-4bit

Adapter

(92)

this model