
Loading weights:   0%|          | 0/311 [00:00<?, ?it/s]
Loading weights: 100%|██████████| 311/311 [00:00<00:00, 4092.59it/s]
mlp2 1e-05 step 0 loss 1.34 exact 5/32 train_s 0
mlp2 1e-05 step 4 loss 0.8602 exact 15/32 train_s 0.52
mlp2 1e-05 step 8 loss 0.6519 exact 20/32 train_s 0.99
mlp2 1e-05 step 16 loss 0.5314 exact 21/32 train_s 1.94
Saved artifacts/qwen3-06b/2026-09-08-efficiency-screen/mlp2-lr1e-5
