MobileLLM-60M / README.md
Yangyang1205's picture
Update README.md
8c1c2d8 verified
metadata
license: mit
language:
  - en
pipeline_tag: text-generation
tags:
  - mobilellm
  - pytorch

MobileLLM 60M (Replication)

This is a replicated version of the MobileLLM 60M model, based on the paper MobileLLM: Optimizing Sub-billion Parameter Language Models for On-Device Use Cases.

  • Model Size: 60M parameters
  • Architecture: Llama-based (Deep & Thin)
  • Status: Research / Testing