starcoder2-7b/dp3/train.log

2026-03-21 12:35:00,834 [INFO] new_opacus_codex.train_steps: epoch=1 step=10 loss=1.7079
2026-03-21 12:37:56,393 [INFO] new_opacus_codex.train_steps: epoch=1 step=20 loss=1.7428
2026-03-21 12:40:52,701 [INFO] new_opacus_codex.train_steps: epoch=1 step=30 loss=1.6670
2026-03-21 12:43:49,022 [INFO] new_opacus_codex.train_steps: epoch=1 step=40 loss=1.8210
2026-03-21 12:46:44,353 [INFO] new_opacus_codex.train_steps: epoch=1 step=50 loss=1.7066
2026-03-21 12:47:11,390 [INFO] new_opacus_codex.train_steps: eval event=eval_step epoch=1 step=50 eval_loss=0.8423 duration_sec=27.03
2026-03-21 12:50:08,596 [INFO] new_opacus_codex.train_steps: epoch=1 step=60 loss=1.7660
2026-03-21 12:53:04,098 [INFO] new_opacus_codex.train_steps: epoch=1 step=70 loss=1.7122
2026-03-21 12:55:58,606 [INFO] new_opacus_codex.train_steps: epoch=1 step=80 loss=1.6304
2026-03-21 12:58:54,249 [INFO] new_opacus_codex.train_steps: epoch=1 step=90 loss=1.6471
2026-03-21 13:01:50,464 [INFO] new_opacus_codex.train_steps: epoch=1 step=100 loss=1.7788
2026-03-21 13:02:17,474 [INFO] new_opacus_codex.train_steps: eval event=eval_step epoch=1 step=100 eval_loss=0.8425 duration_sec=27.01
2026-03-21 13:05:44,630 [INFO] new_opacus_codex.train_steps: epoch=2 step=110 loss=1.6096
2026-03-21 13:08:40,748 [INFO] new_opacus_codex.train_steps: epoch=2 step=120 loss=1.6643
2026-03-21 13:11:35,815 [INFO] new_opacus_codex.train_steps: epoch=2 step=130 loss=1.6273
2026-03-21 13:14:31,647 [INFO] new_opacus_codex.train_steps: epoch=2 step=140 loss=1.5936
2026-03-21 13:17:27,329 [INFO] new_opacus_codex.train_steps: epoch=2 step=150 loss=1.6516
2026-03-21 13:17:54,372 [INFO] new_opacus_codex.train_steps: eval event=eval_step epoch=2 step=150 eval_loss=0.8426 duration_sec=27.04
2026-03-21 13:20:49,864 [INFO] new_opacus_codex.train_steps: epoch=2 step=160 loss=1.7516
2026-03-21 13:23:44,502 [INFO] new_opacus_codex.train_steps: epoch=2 step=170 loss=1.6953
2026-03-21 13:26:42,780 [INFO] new_opacus_codex.train_steps: epoch=2 step=180 loss=1.5992
2026-03-21 13:29:46,885 [INFO] new_opacus_codex.train_steps: epoch=2 step=190 loss=1.6480
2026-03-21 13:32:43,419 [INFO] new_opacus_codex.train_steps: epoch=2 step=200 loss=1.6365
2026-03-21 13:33:10,663 [INFO] new_opacus_codex.train_steps: eval event=eval_step epoch=2 step=200 eval_loss=0.8426 duration_sec=27.24