README.md

metadata

datasets:
  - cerebras/SlimPajama-627B
tags:
  - ggml

BTLM-3B GGML

Bittensor Language Model (BTLM-3B-8k-base) is a 3 billion parameter language model with an 8k context length trained on 627B tokens of SlimPajama.

This is just model conversion into ggml format, model is not actaully implemented so you cannot use it. Stay tuned!

Ref: https://huggingface.co/cerebras/btlm-3b-8k-base GGML ISSUE https://github.com/ggerganov/ggml/issues/427