List your company

LLM Pre-training Resource Utilization on C4

12.55Model Weights ShareFull params
11.922512.2362512.5512.86375Mar 6, 2026

Evaluation results

MethodLinks
Full paramsBackbone=LLaMA-7B2026.0312.5512.5525.114.6664.86
GaLoreBackbone=LLaMA-7B, rank=1282026.0312.5512.551.734.431.23
GoLoreBackbone=LLaMA-7B, rank=1282026.0312.5512.551.734.431.23
LISABackbone=LLaMA-7B, sampling layers=22026.0312.551.242.483.2919.56
LISA-worBackbone=LLaMA-7B, sampling layers=22026.0312.551.242.483.2919.56