Uploaded model

This model was fine-tuned for 3 epochs on the RAG-12.000 dataset to perform better in RAG pipelines. Using RAGAs metrics on own testset, mean performance increased: Faithfulness 0.783 -> 0.822, Answer Correctness 0.591 -> 0.613, Answer Relevance 0.765 -> 0.874 and mean inference time decreased 9.95s -> 6.12s

  • Developed by: R0bfried
  • License: apache-2.0
  • Finetuned from model : unsloth/Meta-Llama-3.1-8B-Instruct-unsloth-bnb-4bit

This llama model was trained 2x faster with Unsloth and Huggingface's TRL library.

Downloads last month
15
GGUF
Model size
8.03B params
Architecture
llama
Hardware compatibility
Log In to view the estimation

4-bit

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

Model tree for R0bfried/RAG12000-LLaMA3.1-8B-gguf

Dataset used to train R0bfried/RAG12000-LLaMA3.1-8B-gguf