artificialguybr/LLAMA3.2-1B-Synthia-I-Redmond

Hugging Face
TEXT GENERATIONConcurrent Unit Cost:1Model Size:1BQuant:BF16Context Size:32kTool Calling:SupportedPublished:Nov 24, 2024License:llama3.2Architecture:Transformer0.0K Featherless Exclusive Warm

artificialguybr/LLAMA3.2-1B-Synthia-I-Redmond is a 1 billion parameter Llama 3.2 base model, fine-tuned by artificialguybr on the Synthia-v1.5-I instruction dataset. This model is optimized for instruction-following tasks and conversational AI applications, leveraging its multilingual capabilities. It is particularly suited for research and development in natural language processing where a smaller, instruction-tuned model is beneficial.

Loading preview...

Model Overview

This model, artificialguybr/LLAMA3.2-1B-Synthia-I-Redmond, is a fine-tuned version of the 1 billion parameter NousResearch/Llama-3.2-1B base model. It has been specifically adapted using the Synthia-v1.5-I instruction dataset to enhance its ability to follow instructions.

Key Capabilities

  • Instruction Following: Improved performance on tasks requiring precise adherence to given instructions.
  • Conversational AI: Suitable for developing chatbots and other interactive AI applications.
  • Multilingual Support: Inherits the multilingual capabilities of the Llama 3.2 base model.
  • Research & Development: A practical choice for NLP research, especially where a smaller, efficient model is preferred.

Training Details

The model was fine-tuned on 20.7k training examples from the Synthia-v1.5-I dataset over 3 epochs, utilizing a Paged AdamW 8bit optimizer and a Cosine learning rate scheduler with 100 warmup steps. The training was supported by RedmondAI and conducted using the Axolotl framework version 0.5.0.

Intended Use Cases

  • Developing instruction-based agents.
  • Building conversational interfaces.
  • Experimenting with fine-tuned LLMs in NLP research.