
Loading weights:   0%|          | 0/311 [00:00<?, ?it/s]
Loading weights: 100%|██████████| 311/311 [00:00<00:00, 4347.89it/s]
full 3e-06 step 0 loss 1.2652 exact 32/128 train_s 0
full 3e-06 step 16 loss 0.2928 exact 104/128 train_s 23.59
Saved artifacts/qwen3-06b/2026-09-08-efficiency-replication/full-seed101
