README.md

metadata

title: README
emoji: 🏢
colorFrom: blue
colorTo: green
sdk: static
pinned: true
tags:
  - function calling
  - long context
  - mistral
  - deepseek
  - yi
  - code-llama
  - quantization
  - SFT
  - fine-tuning

last updated Nov 19th 2023

About

Trelis Research provides specialised fine-tuned models AND fine-tuning scripts.

Function-calling Models and Datasets

• Fine-tuned models for function calling - from 1.3B to 70B in size, incl. Yi, Deepseek, Llama and more
• Dataset to fine-tune models for function calling
• Scripts to fine-tune models for function calling and structured responses

ADVANCED Fine-tuning Scripts

Learn more here

1. Unsupervised + Supervised Fine-tuning Notebooks
2. Q&A Dataset Preparation + Cleaning Scripts
3. Scripts to create and use Embeddings
4. Quantization Scripts
5. Chat fine-tuning Scripts

LLM Server Setup

• Server setup and inference guide for open source models.
• Inference for function calling.
• Video: Run Llama 2 on AWS
• Video: Deploy a Llama API in 5 clicks
• Learn more here

One-click LLM API/Server Templates

• Runpod affiliate link
• Llama 2 70B

Summarise-Me.com

• Summarise PDFs and websites up to 60,000 words

Other Top Tutorials

• Run Code Llama on a Mac with an M1 Chip
• Run Llama 2 with 32k Context Length