jpacifico/French-Alpaca-Llama3-8B-Instruct-v1.0

Hugging Face
TEXT GENERATIONConcurrency Cost:1Model Size:8BQuant:FP8Ctx Length:8kLicense:apache-2.0Architecture:Transformer0.0K Open Weights Warm

jpacifico/French-Alpaca-Llama3-8B-Instruct-v1.0 is an 8 billion parameter instruction-tuned causal language model developed by Jonathan Pacifico. Fine-tuned from Llama3-8B-Instruct using a French-Alpaca dataset, this model is specifically optimized for generating responses in French. It serves as a general-purpose French language model, suitable for various applications and as a base for further specialization.

Loading preview...

French-Alpaca-Llama3-8B-Instruct-v1.0 Overview

This model, developed by Jonathan Pacifico, is an 8 billion parameter instruction-tuned language model based on Llama3-8B-Instruct. Its primary distinction is its fine-tuning on a French-Alpaca dataset, which was entirely generated using OpenAI GPT-3.5-turbo. This process, inspired by the Stanford Alpaca method, aims to specialize the base Llama3 model for French language tasks.

Key Capabilities

  • French Language Proficiency: Optimized for understanding and generating text in French.
  • General-Purpose French LLM: Designed to handle a wide range of French language tasks.
  • Fine-tuning Base: Can be further fine-tuned for more specialized French use cases.
  • Instruction Following: Capable of responding appropriately to given instructions.

Good For

  • Developers requiring a robust, instruction-tuned French language model.
  • Applications focused on French content generation, translation, or conversational AI.
  • As a foundation for creating highly specialized French LLMs for specific domains.

Limitations

As a demonstration model, it currently lacks moderation mechanisms. A quantized Q4_K_M GGUF 4-bit version is also available for more efficient deployment.

Popular Sampler Settings

Top 3 parameter combinations used by Featherless users for this model. Click a tab to see each config.

temperature
top_p
top_k
frequency_penalty
presence_penalty
repetition_penalty
min_p