TehVenom/Pygmalion_AlpacaLora-7b

TEXT GENERATIONConcurrency Cost:1Model Size:7BQuant:FP8Ctx Length:4kPublished:Apr 30, 2023Architecture:Transformer0.0K Cold

TehVenom/Pygmalion_AlpacaLora-7b is a 7 billion parameter LLaMA-based model, created by TehVenom, that merges the Pygmalion-7b model with Tloen's Alpaca LoRA. This model is designed to test the compatibility of LoRAs trained for other LLaMA fine-tunes with Pygmalion, and to provide a merged model for inference backends without LoRA support. It functions as a standard Hugging Face Transformers model, offering a combined instruction-following and conversational capability.

Loading preview...

Model Overview

TehVenom/Pygmalion_AlpacaLora-7b is a 7 billion parameter language model built upon the LLaMA architecture. This model represents a unique merge, combining the conversational capabilities of the Pygmalion-7b model with the instruction-following fine-tuning of Tloen's Alpaca LoRA.

Key Characteristics

  • Architecture: LLaMA-based, 7 billion parameters.
  • Composition: A direct merge of PygmalionAI/pygmalion-7b and tloen/alpaca-lora-7b.
  • Purpose: Primarily created to evaluate the interoperability of LoRAs across different LLaMA fine-tunes and to provide a consolidated model for inference environments that do not natively support LoRA adapters.
  • Compatibility: Functions as a standard Hugging Face Transformers model, allowing for straightforward integration and use.

Potential Use Cases

  • Conversational AI: Leveraging the Pygmalion base for engaging dialogue and character-based interactions.
  • Instruction Following: Utilizing the Alpaca LoRA component for tasks requiring adherence to specific instructions or prompts.
  • Research & Development: Ideal for researchers exploring model merging techniques and LoRA compatibility within the LLaMA ecosystem.
  • Simplified Deployment: Suitable for environments where deploying separate LoRA adapters is not feasible, offering a pre-merged solution.