andrewprayle/llama-2-7b-miniguanaco

TEXT GENERATIONConcurrency Cost:1Model Size:7BQuant:FP8Ctx Length:4kArchitecture:Transformer0.0K Cold

The andrewprayle/llama-2-7b-miniguanaco model is a 7 billion parameter fine-tuned version of the Llama 2 architecture, specifically based on NousResearch/llama-2-7b-chat-hf. This model was fine-tuned using the mlabonne/guanaco-llama2-1k dataset, primarily as a learning exercise by its creator, Andrew Prayle. It is intended for internal use to explore the automation of medical literature searching within CochraneCF.

Loading preview...

Model Overview

The andrewprayle/llama-2-7b-miniguanaco is a 7 billion parameter language model, representing a fine-tuned iteration of the Llama 2 architecture. Specifically, it is built upon the NousResearch/llama-2-7b-chat-hf base model.

Key Characteristics

  • Base Model: Fine-tuned from NousResearch/llama-2-7b-chat-hf.
  • Training Data: Utilizes the mlabonne/guanaco-llama2-1k dataset for fine-tuning.
  • Context Length: Inherits a context length of 4096 tokens from its Llama 2 base.
  • Development Purpose: Primarily developed as a personal learning project by Andrew Prayle to understand model fine-tuning.

Intended Use Cases

This model is currently intended for internal exploration within CochraneCF to investigate:

  • Automating aspects of medical literature searching.
  • Understanding the practical application of large language models in a research context.

Limitations

As a fine-tuned version of Llama 2 7B, it is expected to share similar risks, biases, and limitations as its foundational model. Users are advised to be aware of these inherent characteristics, which are still under exploration by the developer.