yitong241/llama-recipe-7B-3epoch-12batch
TEXT GENERATIONConcurrency Cost:1Model Size:7BQuant:FP8Ctx Length:4kArchitecture:Transformer Cold
The yitong241/llama-recipe-7B-3epoch-12batch model is a fine-tuned variant of the Meta Llama 2 7B Chat architecture, specifically optimized using LoRA PEFT methods over 3 epochs with a batch size of 12. This model incorporates quantization during training, making it suitable for efficient deployment. It is primarily designed for tasks similar to those in the Alpaca dataset, leveraging its base Llama 2 capabilities for conversational AI and instruction following.
Loading preview...