aimosprite/nemotron-super-120b-a12b-nvfp4-count-reap-keep95

Nemotron Super NVFP4 Count-REAP Keep95

This repository contains a structurally pruned variant of nvidia/NVIDIA-Nemotron-3-Super-120B-A12B-NVFP4.

Pruning details

• —Metric: count
• —Keep ratio: 0.95
• —Routed experts kept per MoE layer: 487 / 512
• —Experts removed per MoE layer: 25 / 512
• —Usage source: merged representative imo-answerbench final count dumps (pid61 + pid63)
• —Materialization script: compress/reap/scripts/materialize_pruned_nemotron_model.py

Notes

• —This is count-based REAP, not score-based REAP.
• —reap_prune_plan.json records the exact per-layer expert remapping.