starcoder2-7b/dp8/train.log

2026-03-21 14:43:51,342 [INFO] new_opacus_codex.train_steps: epoch=1 step=10 loss=1.7169
2026-03-21 14:46:31,374 [INFO] new_opacus_codex.train_steps: epoch=1 step=20 loss=1.7233
2026-03-21 14:49:11,893 [INFO] new_opacus_codex.train_steps: epoch=1 step=30 loss=1.6529
2026-03-21 14:51:54,577 [INFO] new_opacus_codex.train_steps: epoch=1 step=40 loss=1.7956
2026-03-21 14:54:35,859 [INFO] new_opacus_codex.train_steps: epoch=1 step=50 loss=1.6767
2026-03-21 14:55:02,994 [INFO] new_opacus_codex.train_steps: eval event=eval_step epoch=1 step=50 eval_loss=0.8407 duration_sec=27.13
2026-03-21 14:57:43,263 [INFO] new_opacus_codex.train_steps: epoch=1 step=60 loss=1.7326
2026-03-21 15:00:22,484 [INFO] new_opacus_codex.train_steps: epoch=1 step=70 loss=1.6678
2026-03-21 15:03:03,164 [INFO] new_opacus_codex.train_steps: epoch=1 step=80 loss=1.5879
2026-03-21 15:05:44,501 [INFO] new_opacus_codex.train_steps: epoch=1 step=90 loss=1.6269
2026-03-21 15:08:26,236 [INFO] new_opacus_codex.train_steps: epoch=1 step=100 loss=1.7528
2026-03-21 15:08:53,353 [INFO] new_opacus_codex.train_steps: eval event=eval_step epoch=1 step=100 eval_loss=0.8408 duration_sec=27.11
2026-03-21 15:12:05,824 [INFO] new_opacus_codex.train_steps: epoch=2 step=110 loss=1.5950
2026-03-21 15:14:46,773 [INFO] new_opacus_codex.train_steps: epoch=2 step=120 loss=1.6361
2026-03-21 15:17:26,290 [INFO] new_opacus_codex.train_steps: epoch=2 step=130 loss=1.5945
2026-03-21 15:20:05,432 [INFO] new_opacus_codex.train_steps: epoch=2 step=140 loss=1.5773
2026-03-21 15:22:44,899 [INFO] new_opacus_codex.train_steps: epoch=2 step=150 loss=1.6216
2026-03-21 15:23:11,983 [INFO] new_opacus_codex.train_steps: eval event=eval_step epoch=2 step=150 eval_loss=0.8408 duration_sec=27.08
2026-03-21 15:25:53,056 [INFO] new_opacus_codex.train_steps: epoch=2 step=160 loss=1.7143
2026-03-21 15:28:34,111 [INFO] new_opacus_codex.train_steps: epoch=2 step=170 loss=1.6569
2026-03-21 15:31:15,007 [INFO] new_opacus_codex.train_steps: epoch=2 step=180 loss=1.5789
2026-03-21 15:33:54,650 [INFO] new_opacus_codex.train_steps: epoch=2 step=190 loss=1.5876
2026-03-21 15:36:34,437 [INFO] new_opacus_codex.train_steps: epoch=2 step=200 loss=1.6100
2026-03-21 15:37:01,524 [INFO] new_opacus_codex.train_steps: eval event=eval_step epoch=2 step=200 eval_loss=0.8409 duration_sec=27.08