simonbutt
/

am_llama3_dpo

text-generation-inference

Model card Files Files and versions

Llama3 Amharic DPO

Amharic Llama3 8B Alpaca further DPO tuned on an amharic translated dolly-15k dataset to always respond in Amharic.

Very token inefficient.

Developed by: simonbutt
License: apache-2.0
Finetuned from model:
- unsloth/llama-3-8b-bnb-4bit
- simonbutt/am_llama3_alpaca

Downloads last month: -; Downloads are not tracked for this model. How to track

Inference Providers NEW

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

Model tree for simonbutt/am_llama3_dpo

Base model

meta-llama/Meta-Llama-3-8B

Quantized

unsloth/llama-3-8b-bnb-4bit

Finetuned

(2851)

this model

Datasets used to train simonbutt/am_llama3_dpo

Collection including simonbutt/am_llama3_dpo

amharic_llama

Goal: to create the best possible Amharic language models that can run offline on a budget smartphone • 4 items • Updated Apr 27, 2024