Sign up to save tools and stay up to date with the latest in AI
bg
bg
1

knn-router/deploy/pulze-intent-v0.1 at main · pulzeai-oss/knn-router

May 07, 2024 - github.com
The article describes the usage and implementation of the pulze-intent-v0.1 model and dataset. This model is an Intent-tuned LLM router that selects the best LLM for a user query. It can be used locally by fetching artifacts from Huggingface and starting the services using Docker. The model can also be used in Kubernetes. The model uses several other models including claude-3-haiku-20240307, claude-3-opus-20240229, and gpt-4-turbo-2024-04-09 among others.

The data used in the model includes prompts and intent categories derived from the GAIR-NLP/Auto-J scenario classification dataset. The responses from the model were evaluated pairwise using `openai/gpt-4-turbo-2024-04-09`. The embedding model was generated by first fine-tuning BAAI/bge-base-en-v1.5 with the intent categories from the dataset above, using contrastive learning with cosine similarity loss, and subsequently merging the resultant model with the base model at a 3:2 ratio.

Key takeaways:

  • The pulze-intent-v0.1 is an Intent-tuned LLM router that selects the best LLM for a user query.
  • The model uses a variety of models including claude-3-haiku-20240307, claude-3-opus-20240229, and gpt-4-turbo-2024-04-09 among others.
  • Prompt and intent categories are derived from the GAIR-NLP/Auto-J scenario classification dataset.
  • The embedding model was generated by first fine-tuning BAAI/bge-base-en-v1.5 with the intent categories from the dataset above, using contrastive learning with cosine similarity loss.
View Full Article

Comments (0)

Be the first to comment!