Downloads · 30 days
0
0% of all-time downloads
mistral-7b-math
akashvshroff/mistral-7b-math
mistral-7b-math is a machine learning model from akashvshroff. Use it for the machine learning task on the model card, and read the license before you ship it in a product.
The model is a finetuned version of Mistral 7B, tuned on a small subset of the MathInstruct database by TIGER-Lab.
All-time downloads
6
Public
Repo size
650 MB
Likes
0
Public
Repo makeup
Click a slice to open those files.
.safetensors609 MB · 100%
At a glance
- Access
- Public
- Created
- Jan 8, 2024
- Updated
- Jan 10, 2024
- SHA
- bacc6c0c
Model card
From the Hugging Face model README
Model Details
The model is a finetuned version of Mistral 7B, tuned on a small subset of the MathInstruct database by TIGER-Lab.
This finetuning was done to see whether an incredibly small subset, roughly 5000 data points, could cause a noticeable increase in the mathematical performance of the model as well as allow me to experiment with hosting and running LLMs locally.
The framework employed for finetuning was PEFT LoRA or Low-Rank Adaptation.
More about the training process and some example results can be seen on my GitHub repo.
Framework versions PEFT 0.7.1
- Created
- Jan 8, 2024
- Updated
- Jan 10, 2024