HyperPrune-Llama-3.1-8B-2to4

Description

HyperPrune-Llama-3.1-8B-2to4 is an open-weight text-generation model with roughly 8B parameters published by elastix-ai on the Hugging Face Hub. It has 410 downloads. Runs with transformers.

Categories

• Writing

Tags

• llama
• text-generation
• hyperprune
• pruning
• sparsity
• 2-4-sparsity
• semi-structured

Capabilities

• Text generation
• Library: transformers
• 8B parameters