Official implementation of 'LLaMA-Adapter: Efficient Fine-tuning of Language Models with Zero-init Attention' and 'LLaMA-Adapter V2: Parameter-Efficient Visual Instruction Model'.
Utilizing LangKit, you can continuously evaluate the LLM-powered applications using essential metrics extracted from prompts, responses, and user interactions.
"Scale AI's LLM toolkit with three main functions: Demonstration Data: Collect human-generated text over various domains and use cases Comparison Data: Evaluate your model outputs with human feedback Trust & Safety: Align your model to produce factually correct and safe content "