pa_warm_start_sft_120b_light1bmix_32k

Description

pa_warm_start_sft_120b_light1bmix_32k is an open-weight text-generation model with roughly 120B parameters published by geodesic-research on the Hugging Face Hub. It has 434 downloads. Runs with transformers.

Categories

• Research
• Productivity

Tags

• nemotron_h
• text-generation
• nemotron
• nemotron-h
• mamba2
• moe
• reasoning
• sft

Capabilities

• Text generation
• Conversational
• Library: transformers
• 120B parameters