[2026-04-17 08:57:56] CUDA_VISIBLE_DEVICES: 0,1 [2026-04-17 08:57:56] Number of processes: 2 [2026-04-17 08:57:56] Process index: 0 [2026-04-17 08:57:56] Mixed precision: bf16 [2026-04-17 08:57:56] ============================================================ [2026-04-17 08:57:56] HNet Training Pipeline (Hydra + Trackio + Accelerate) [2026-04-17 08:57:56] ============================================================ [2026-04-17 08:57:56] Config: model: config_path: /workspace/byte-llms-code/hnet_project/configs/hnet_2stage_XL_code.json checkpoint_path: /workspace/byte-llms-code/hnet_project/checkpoints/hnet_2stage_XL_code.pt training: epochs: 3 batch_size: 4 eval_batch_size: 24 gradient_accumulation_steps: 4 lr: 0.0001 weight_decay: 0.1 betas: - 0.9 - 0.95 eps: 1.0e-08 lr_scheduler: wsd warmup_ratio: 0.1 decay_ratio: 0.2 warmup_steps: 100 min_lr_ratio: 0.1 lr_multiplier: - 2.0 - 1.5 - 1.0 load_balancing_weight: 0.01 load_balancing_N: 4.0 max_grad_norm: 1.0 use_amp: true resume: false resume_checkpoint: null warmup_model: true data: path: /workspace/byte-llms-code/code_completion_exp/datasets/data_V5_full max_context_len: 4096 max_target_len: 256 num_workers: 0 pin_memory: true max_train_samples: null max_val_samples: null logging: log_interval: 10 save_interval: 3000 eval_interval: 1000 save_every_epoch: true tracking: enabled: true backend: wandb project: code-completion-full-docstring run_name: hnet_train entity: null base_url: https://wandb.platun0v.ru paths: output_dir: outputs/2026-04-17/08-57-56 seed: 42 device: cuda [2026-04-17 08:57:58] Initializing tokenizer... [2026-04-17 08:57:58] Loading model... [2026-04-17 08:58:02] Loaded pretrained: /workspace/byte-llms-code/hnet_project/checkpoints/hnet_2stage_XL_code.pt [2026-04-17 08:58:02] Applied LR multipliers: [2.0, 1.5, 1.0] [2026-04-17 08:58:02] Warming up model... [2026-04-17 09:00:54] Total params: 1,654,090,112 [2026-04-17 09:00:54] Trainable params: 1,654,090,112 [2026-04-17 09:00:54] Creating dataloaders... [2026-04-17 09:00:54] Train dataset size: 338932 [2026-04-17 09:00:54] Train batches per epoch (before DDP split): 84733 [2026-04-17 09:00:54] Validation dataset size: 37592 [2026-04-17 09:00:54] Validation batches: 1567 [2026-04-17 09:00:54] Creating optimizer... [2026-04-17 09:00:54] Total steps: 31775, Steps per epoch: 42367 [2026-04-17 09:00:54] Preparing model, optimizer, and dataloaders with Accelerate... [2026-04-17 09:00:55] Train batches per epoch (after DDP split): 42367 [2026-04-17 09:00:55] Starting training... [2026-04-17 09:00:55] ============================================================ [2026-04-17 09:00:55] EPOCH 1/3 [2026-04-17 09:00:55] ============================================================ [2026-04-17 09:03:20] Epoch 1 | Step 10 | Loss: 0.7009 | LM: 0.6808 | LB: 1.1460 | CL0: 2.8 | CL1: 2.2 | HR0: 0.357/SR0: 0.357 | HR1: 0.463/SR1: 0.448 | LR: 1.06e-05 [2026-04-17 09:03:27] Epoch 1 | Step 20 | Loss: 0.6857 | LM: 0.6361 | LB: 1.1458 | CL0: 2.9 | CL1: 2.1 | HR0: 0.351/SR0: 0.352 | HR1: 0.467/SR1: 0.451 | LR: 1.11e-05 [2026-04-17 09:03:34] Epoch 1 | Step 30 | Loss: 0.6687 | LM: 0.6387 | LB: 1.1470 | CL0: 2.9 | CL1: 2.1 | HR0: 0.347/SR0: 0.348 | HR1: 0.469/SR1: 0.454 | LR: 1.17e-05 [2026-04-17 09:03:43] Epoch 1 | Step 40 | Loss: 0.6616 | LM: 0.6364 | LB: 1.1476 | CL0: 2.9 | CL1: 2.1 | HR0: 0.347/SR0: 0.348 | HR1: 0.469/SR1: 0.454 | LR: 1.23e-05 [2026-04-17 09:03:50] Epoch 1 | Step 50 | Loss: 0.6459 | LM: 0.6315 | LB: 1.1499 | CL0: 2.9 | CL1: 2.1 | HR0: 0.351/SR0: 0.351 | HR1: 0.470/SR1: 0.453 | LR: 1.28e-05 [2026-04-17 09:03:56] Epoch 1 | Step 60 | Loss: 0.6368 | LM: 0.5975 | LB: 1.1501 | CL0: 2.9 | CL1: 2.1 | HR0: 0.353/SR0: 0.353 | HR1: 0.470/SR1: 0.452 | LR: 1.34e-05 [2026-04-17 09:04:02] Epoch 1 | Step 70 | Loss: 0.6154 | LM: 0.5864 | LB: 1.1514 | CL0: 2.9 | CL1: 2.1 | HR0: 0.350/SR0: 0.351 | HR1: 0.472/SR1: 0.455 | LR: 1.40e-05 [2026-04-17 09:04:09] Epoch 1 | Step 80 | Loss: 0.5992 | LM: 0.5676 | LB: 1.1508 | CL0: 2.9 | CL1: 2.1 | HR0: 0.346/SR0: 0.348 | HR1: 0.473/SR1: 0.455 | LR: 1.45e-05 [2026-04-17 09:04:15] Epoch 1 | Step 90 | Loss: 0.5928 | LM: 0.5707 | LB: 1.1523 | CL0: 2.9 | CL1: 2.1 | HR0: 0.347/SR0: 0.349 | HR1: 0.474/SR1: 0.456 | LR: 1.51e-05 [2026-04-17 09:04:22] Epoch 1 | Step 100 | Loss: 0.5827 | LM: 0.5594 | LB: 1.1537 | CL0: 2.9 | CL1: 2.1 | HR0: 0.349/SR0: 0.350 | HR1: 0.475/SR1: 0.457 | LR: 1.57e-05 [2026-04-17 09:04:28] Epoch 1 | Step 110 | Loss: 0.5730 | LM: 0.5515 | LB: 1.1542 | CL0: 2.9 | CL1: 2.1 | HR0: 0.349/SR0: 0.350 | HR1: 0.475/SR1: 0.457 | LR: 1.62e-05 [2026-04-17 09:04:34] Epoch 1 | Step 120 | Loss: 0.5577 | LM: 0.5408 | LB: 1.1549 | CL0: 2.9 | CL1: 2.1 | HR0: 0.348/SR0: 0.350 | HR1: 0.477/SR1: 0.458 | LR: 1.68e-05 [2026-04-17 09:04:41] Epoch 1 | Step 130 | Loss: 0.5467 | LM: 0.5349 | LB: 1.1560 | CL0: 2.9 | CL1: 2.1 | HR0: 0.348/SR0: 0.350 | HR1: 0.478/SR1: 0.458 | LR: 1.74e-05 [2026-04-17 09:04:47] Epoch 1 | Step 140 | Loss: 0.5330 | LM: 0.5263 | LB: 1.1561 | CL0: 2.9 | CL1: 2.1 | HR0: 0.348/SR0: 0.349 | HR1: 0.478/SR1: 0.459 | LR: 1.79e-05 [2026-04-17 09:04:54] Epoch 1 | Step 150 | Loss: 0.5231 | LM: 0.5171 | LB: 1.1551 | CL0: 2.9 | CL1: 2.1 | HR0: 0.348/SR0: 0.350 | HR1: 0.477/SR1: 0.458 | LR: 1.85e-05 [2026-04-17 09:05:00] Epoch 1 | Step 160 | Loss: 0.5127 | LM: 0.5005 | LB: 1.1552 | CL0: 2.9 | CL1: 2.1 | HR0: 0.348/SR0: 0.350 | HR1: 0.477/SR1: 0.458 | LR: 1.91e-05 [2026-04-17 09:05:06] Epoch 1 | Step 170 | Loss: 0.5046 | LM: 0.4946 | LB: 1.1559 | CL0: 2.9 | CL1: 2.1 | HR0: 0.348/SR0: 0.350 | HR1: 0.477/SR1: 0.458 | LR: 1.96e-05 [2026-04-17 09:05:13] Epoch 1 | Step 180 | Loss: 0.4949 | LM: 0.4861 | LB: 1.1563 | CL0: 2.9 | CL1: 2.1 | HR0: 0.349/SR0: 0.350 | HR1: 0.478/SR1: 0.459 | LR: 2.02e-05 [2026-04-17 09:05:19] Epoch 1 | Step 190 | Loss: 0.4860 | LM: 0.4742 | LB: 1.1563 | CL0: 2.9 | CL1: 2.1 | HR0: 0.349/SR0: 0.350 | HR1: 0.478/SR1: 0.459 | LR: 2.08e-05 [2026-04-17 09:05:26] Epoch 1 | Step 200 | Loss: 0.4808 | LM: 0.4693 | LB: 1.1558 | CL0: 2.9 | CL1: 2.1 | HR0: 0.349/SR0: 0.350 | HR1: 0.477/SR1: 0.459 | LR: 2.13e-05 [2026-04-17 09:05:32] Epoch 1 | Step 210 | Loss: 0.4733 | LM: 0.4604 | LB: 1.1560 | CL0: 2.9 | CL1: 2.1 | HR0: 0.349/SR0: 0.350 | HR1: 0.477/SR1: 0.459 | LR: 2.19e-05 [2026-04-17 09:05:39] Epoch 1 | Step 220 | Loss: 0.4690 | LM: 0.4574 | LB: 1.1565 | CL0: 2.9 | CL1: 2.1 | HR0: 0.349/SR0: 0.351 | HR1: 0.477/SR1: 0.459 | LR: 2.25e-05 [2026-04-17 09:05:45] Epoch 1 | Step 230 | Loss: 0.4639 | LM: 0.4545 | LB: 1.1556 | CL0: 2.9 | CL1: 2.1 | HR0: 0.349/SR0: 0.351 | HR1: 0.477/SR1: 0.458 | LR: 2.30e-05 [2026-04-17 09:05:52] Epoch 1 | Step 240 | Loss: 0.4595 | LM: 0.4496 | LB: 1.1557 | CL0: 2.9 | CL1: 2.1 | HR0: 0.349/SR0: 0.350 | HR1: 0.477/SR1: 0.458 | LR: 2.36e-05 [2026-04-17 09:05:58] Epoch 1 | Step 250 | Loss: 0.4563 | LM: 0.4480 | LB: 1.1560 | CL0: 2.9 | CL1: 2.1 | HR0: 0.349/SR0: 0.350 | HR1: 0.477/SR1: 0.459 | LR: 2.42e-05 [2026-04-17 09:06:04] Epoch 1 | Step 260 | Loss: 0.4537 | LM: 0.4455 | LB: 1.1570 | CL0: 2.9 | CL1: 2.1 | HR0: 0.349/SR0: 0.351 | HR1: 0.478/SR1: 0.459 | LR: 2.47e-05 [2026-04-17 09:06:11] Epoch 1 | Step 270 | Loss: 0.4513 | LM: 0.4414 | LB: 1.1569 | CL0: 2.9 | CL1: 2.1 | HR0: 0.349/SR0: 0.350 | HR1: 0.478/SR1: 0.459 | LR: 2.53e-05 [2026-04-17 09:06:17] Epoch 1 | Step 280 | Loss: 0.4489 | LM: 0.4363 | LB: 1.1568 | CL0: 2.9 | CL1: 2.1 | HR0: 0.349/SR0: 0.351 | HR1: 0.478/SR1: 0.459 | LR: 2.59e-05 [2026-04-17 09:06:23] Epoch 1 | Step 290 | Loss: 0.4468 | LM: 0.4355 | LB: 1.1562 | CL0: 2.9 | CL1: 2.1 | HR0: 0.349/SR0: 0.351 | HR1: 0.477/SR1: 0.459 | LR: 2.64e-05 [2026-04-17 09:06:30] Epoch 1 | Step 300 | Loss: 0.4436 | LM: 0.4331 | LB: 1.1562 | CL0: 2.9 | CL1: 2.1 | HR0: 0.349/SR0: 0.350 | HR1: 0.478/SR1: 0.459 | LR: 2.70e-05 [2026-04-17 09:06:36] Epoch 1 | Step 310 | Loss: 0.4422 | LM: 0.4293 | LB: 1.1564 | CL0: 2.9 | CL1: 2.1 | HR0: 0.349/SR0: 0.351 | HR1: 0.478/SR1: 0.459 | LR: 2.76e-05 [2026-04-17 09:06:43] Epoch 1 | Step 320 | Loss: 0.4403 | LM: 0.4262 | LB: 1.1563 | CL0: 2.9 | CL1: 2.1 | HR0: 0.349/SR0: 0.350 | HR1: 0.478/SR1: 0.459 | LR: 2.81e-05 [2026-04-17 09:06:49] Epoch 1 | Step 330 | Loss: 0.4384 | LM: 0.4257 | LB: 1.1559 | CL0: 2.9 | CL1: 2.1 | HR0: 0.349/SR0: 0.351 | HR1: 0.477/SR1: 0.458 | LR: 2.87e-05 [2026-04-17 09:06:55] Epoch 1 | Step 340 | Loss: 0.4357 | LM: 0.4231 | LB: 1.1559 | CL0: 2.9 | CL1: 2.1 | HR0: 0.349/SR0: 0.351 | HR1: 0.477/SR1: 0.458 | LR: 2.93e-05 [2026-04-17 09:07:02] Epoch 1 | Step 350 | Loss: 0.4331 | LM: 0.4236 | LB: 1.1556 | CL0: 2.9 | CL1: 2.1 | HR0: 0.349/SR0: 0.351 | HR1: 0.477/SR1: 0.458 | LR: 2.98e-05 [2026-04-17 09:07:08] Epoch 1 | Step 360 | Loss: 0.4309 | LM: 0.4217 | LB: 1.1556 | CL0: 2.9 | CL1: 2.1 | HR0: 0.349/SR0: 0.351 | HR1: 0.477/SR1: 0.458 | LR: 3.04e-05 [2026-04-17 09:07:14] Epoch 1 | Step 370 | Loss: 0.4277 | LM: 0.4194 | LB: 1.1553 | CL0: 2.9 | CL1: 2.1 | HR0: 0.349/SR0: 0.350 | HR1: 0.477/SR1: 0.458 | LR: 3.10e-05 [2026-04-17 09:07:21] Epoch 1 | Step 380 | Loss: 0.4240 | LM: 0.4169 | LB: 1.1551 | CL0: 2.9 | CL1: 2.1 | HR0: 0.348/SR0: 0.350 | HR1: 0.477/SR1: 0.458 | LR: 3.15e-05 [2026-04-17 09:07:28] Epoch 1 | Step 390 | Loss: 0.4230 | LM: 0.4150 | LB: 1.1549 | CL0: 2.9 | CL1: 2.1 | HR0: 0.348/SR0: 0.350 | HR1: 0.477/SR1: 0.458 | LR: 3.21e-05 [2026-04-17 09:07:34] Epoch 1 | Step 400 | Loss: 0.4211 | LM: 0.4141 | LB: 1.1549 | CL0: 2.9 | CL1: 2.1 | HR0: 0.348/SR0: 0.350 | HR1: 0.477/SR1: 0.458 | LR: 3.27e-05 [2026-04-17 09:07:40] Epoch 1 | Step 410 | Loss: 0.4180 | LM: 0.4096 | LB: 1.1552 | CL0: 2.9 | CL1: 2.1 | HR0: 0.348/SR0: 0.350 | HR1: 0.477/SR1: 0.458 | LR: 3.32e-05 [2026-04-17 09:07:47] Epoch 1 | Step 420 | Loss: 0.4162 | LM: 0.4064 | LB: 1.1554 | CL0: 2.9 | CL1: 2.1 | HR0: 0.348/SR0: 0.350 | HR1: 0.477/SR1: 0.458 | LR: 3.38e-05 [2026-04-17 09:07:53] Epoch 1 | Step 430 | Loss: 0.4140 | LM: 0.4049 | LB: 1.1551 | CL0: 2.9 | CL1: 2.1 | HR0: 0.349/SR0: 0.350 | HR1: 0.477/SR1: 0.458 | LR: 3.44e-05 [2026-04-17 09:07:59] Epoch 1 | Step 440 | Loss: 0.4119 | LM: 0.4017 | LB: 1.1553 | CL0: 2.9 | CL1: 2.1 | HR0: 0.349/SR0: 0.350 | HR1: 0.477/SR1: 0.458 | LR: 3.49e-05 [2026-04-17 09:08:05] Epoch 1 | Step 450 | Loss: 0.4103 | LM: 0.3998 | LB: 1.1551 | CL0: 2.9 | CL1: 2.1 | HR0: 0.349/SR0: 0.350 | HR1: 0.476/SR1: 0.458 | LR: 3.55e-05 [2026-04-17 09:08:12] Epoch 1 | Step 460 | Loss: 0.4091 | LM: 0.3984 | LB: 1.1549 | CL0: 2.9 | CL1: 2.1 | HR0: 0.349/SR0: 0.350 | HR1: 0.476/SR1: 0.458 | LR: 3.61e-05 [2026-04-17 09:08:18] Epoch 1 | Step 470 | Loss: 0.4071 | LM: 0.3944 | LB: 1.1546 | CL0: 2.9 | CL1: 2.1 | HR0: 0.349/SR0: 0.350 | HR1: 0.476/SR1: 0.458 | LR: 3.66e-05 [2026-04-17 09:08:25] Epoch 1 | Step 480 | Loss: 0.4052 | LM: 0.3925 | LB: 1.1546 | CL0: 2.9 | CL1: 2.1 | HR0: 0.349/SR0: 0.350 | HR1: 0.476/SR1: 0.458 | LR: 3.72e-05 [2026-04-17 09:08:31] Epoch 1 | Step 490 | Loss: 0.4043 | LM: 0.3905 | LB: 1.1543 | CL0: 2.9 | CL1: 2.1 | HR0: 0.348/SR0: 0.350 | HR1: 0.476/SR1: 0.457 | LR: 3.78e-05 [2026-04-17 09:08:37] Epoch 1 | Step 500 | Loss: 0.4036 | LM: 0.3882 | LB: 1.1543 | CL0: 2.9 | CL1: 2.1 | HR0: 0.348/SR0: 0.350 | HR1: 0.476/SR1: 0.457 | LR: 3.83e-05 [2026-04-17 09:08:44] Epoch 1 | Step 510 | Loss: 0.4026 | LM: 0.3876 | LB: 1.1543 | CL0: 2.9 | CL1: 2.1 | HR0: 0.348/SR0: 0.350 | HR1: 0.476/SR1: 0.457 | LR: 3.89e-05 [2026-04-17 09:08:50] Epoch 1 | Step 520 | Loss: 0.4016 | LM: 0.3861 | LB: 1.1541 | CL0: 2.9 | CL1: 2.1 | HR0: 0.348/SR0: 0.350 | HR1: 0.476/SR1: 0.457 | LR: 3.95e-05 [2026-04-17 09:08:57] Epoch 1 | Step 530 | Loss: 0.4003 | LM: 0.3838 | LB: 1.1540 | CL0: 2.9 | CL1: 2.1 | HR0: 0.348/SR0: 0.349 | HR1: 0.476/SR1: 0.457 | LR: 4.00e-05 [2026-04-17 09:09:03] Epoch 1 | Step 540 | Loss: 0.3987 | LM: 0.3818 | LB: 1.1541 | CL0: 2.9 | CL1: 2.1 | HR0: 0.348/SR0: 0.349 | HR1: 0.476/SR1: 0.458 | LR: 4.06e-05 [2026-04-17 09:09:09] Epoch 1 | Step 550 | Loss: 0.3965 | LM: 0.3791 | LB: 1.1540 | CL0: 2.9 | CL1: 2.1 | HR0: 0.348/SR0: 0.349 | HR1: 0.476/SR1: 0.457 | LR: 4.12e-05 [2026-04-17 09:09:16] Epoch 1 | Step 560 | Loss: 0.3954 | LM: 0.3780 | LB: 1.1542 | CL0: 2.9 | CL1: 2.1 | HR0: 0.348/SR0: 0.349 | HR1: 0.476/SR1: 0.458 | LR: 4.17e-05 [2026-04-17 09:09:22] Epoch 1 | Step 570 | Loss: 0.3937 | LM: 0.3765 | LB: 1.1539 | CL0: 2.9 | CL1: 2.1 | HR0: 0.348/SR0: 0.349 | HR1: 0.476/SR1: 0.457 | LR: 4.23e-05 [2026-04-17 09:09:28] Epoch 1 | Step 580 | Loss: 0.3922 | LM: 0.3752 | LB: 1.1540 | CL0: 2.9 | CL1: 2.1 | HR0: 0.348/SR0: 0.349 | HR1: 0.476/SR1: 0.457 | LR: 4.29e-05 [2026-04-17 09:09:35] Epoch 1 | Step 590 | Loss: 0.3917 | LM: 0.3744 | LB: 1.1541 | CL0: 2.9 | CL1: 2.1 | HR0: 0.348/SR0: 0.349 | HR1: 0.476/SR1: 0.457 | LR: 4.34e-05 [2026-04-17 09:09:41] Epoch 1 | Step 600 | Loss: 0.3907 | LM: 0.3741 | LB: 1.1543 | CL0: 2.9 | CL1: 2.1 | HR0: 0.348/SR0: 0.349 | HR1: 0.476/SR1: 0.457 | LR: 4.40e-05 [2026-04-17 09:09:48] Epoch 1 | Step 610 | Loss: 0.3897 | LM: 0.3734 | LB: 1.1542 | CL0: 2.9 | CL1: 2.1 | HR0: 0.348/SR0: 0.350 | HR1: 0.476/SR1: 0.457 | LR: 4.46e-05 [2026-04-17 09:09:54] Epoch 1 | Step 620 | Loss: 0.3882 | LM: 0.3717 | LB: 1.1541 | CL0: 2.9 | CL1: 2.1 | HR0: 0.348/SR0: 0.350 | HR1: 0.476/SR1: 0.457 | LR: 4.51e-05 [2026-04-17 09:10:00] Epoch 1 | Step 630 | Loss: 0.3870 | LM: 0.3708 | LB: 1.1541 | CL0: 2.9 | CL1: 2.1 | HR0: 0.348/SR0: 0.349 | HR1: 0.476/SR1: 0.457 | LR: 4.57e-05 [2026-04-17 09:10:09] Epoch 1 | Step 640 | Loss: 0.3863 | LM: 0.3711 | LB: 1.1541 | CL0: 2.9 | CL1: 2.1 | HR0: 0.348/SR0: 0.349 | HR1: 0.476/SR1: 0.457 | LR: 4.63e-05 [2026-04-17 09:10:15] Epoch 1 | Step 650 | Loss: 0.3854 | LM: 0.3707 | LB: 1.1539 | CL0: 2.9 | CL1: 2.1 | HR0: 0.348/SR0: 0.349 | HR1: 0.476/SR1: 0.457 | LR: 4.68e-05 [2026-04-17 09:10:22] Epoch 1 | Step 660 | Loss: 0.3849 | LM: 0.3702 | LB: 1.1539 | CL0: 2.9 | CL1: 2.1 | HR0: 0.348/SR0: 0.349 | HR1: 0.476/SR1: 0.457 | LR: 4.74e-05 [2026-04-17 09:10:28] Epoch 1 | Step 670 | Loss: 0.3841 | LM: 0.3696 | LB: 1.1539 | CL0: 2.9 | CL1: 2.1 | HR0: 0.348/SR0: 0.349 | HR1: 0.476/SR1: 0.457 | LR: 4.80e-05 [2026-04-17 09:10:35] Epoch 1 | Step 680 | Loss: 0.3830 | LM: 0.3694 | LB: 1.1539 | CL0: 2.9 | CL1: 2.1 | HR0: 0.348/SR0: 0.349 | HR1: 0.476/SR1: 0.457 | LR: 4.85e-05 [2026-04-17 09:10:41] Epoch 1 | Step 690 | Loss: 0.3820 | LM: 0.3686 | LB: 1.1537 | CL0: 2.9 | CL1: 2.1 | HR0: 0.348/SR0: 0.349 | HR1: 0.476/SR1: 0.457 | LR: 4.91e-05 [2026-04-17 09:10:47] Epoch 1 | Step 700 | Loss: 0.3818 | LM: 0.3689 | LB: 1.1534 | CL0: 2.9 | CL1: 2.1 | HR0: 0.348/SR0: 0.349 | HR1: 0.476/SR1: 0.457 | LR: 4.97e-05 [2026-04-17 09:10:54] Epoch 1 | Step 710 | Loss: 0.3817 | LM: 0.3690 | LB: 1.1533 | CL0: 2.9 | CL1: 2.1 | HR0: 0.348/SR0: 0.349 | HR1: 0.475/SR1: 0.457 | LR: 5.02e-05 [2026-04-17 09:11:00] Epoch 1 | Step 720 | Loss: 0.3810 | LM: 0.3688 | LB: 1.1532 | CL0: 2.9 | CL1: 2.1 | HR0: 0.348/SR0: 0.349 | HR1: 0.475/SR1: 0.456 | LR: 5.08e-05 [2026-04-17 09:11:07] Epoch 1 | Step 730 | Loss: 0.3803 | LM: 0.3680 | LB: 1.1534 | CL0: 2.9 | CL1: 2.1 | HR0: 0.348/SR0: 0.350 | HR1: 0.475/SR1: 0.456 | LR: 5.14e-05 [2026-04-17 09:11:13] Epoch 1 | Step 740 | Loss: 0.3799 | LM: 0.3670 | LB: 1.1532 | CL0: 2.9 | CL1: 2.1 | HR0: 0.348/SR0: 0.349 | HR1: 0.475/SR1: 0.456 | LR: 5.19e-05 [2026-04-17 09:11:19] Epoch 1 | Step 750 | Loss: 0.3794 | LM: 0.3663 | LB: 1.1529 | CL0: 2.9 | CL1: 2.1 | HR0: 0.348/SR0: 0.349 | HR1: 0.475/SR1: 0.456 | LR: 5.25e-05 [2026-04-17 09:11:26] Epoch 1 | Step 760 | Loss: 0.3792 | LM: 0.3662 | LB: 1.1527 | CL0: 2.9 | CL1: 2.1 | HR0: 0.348/SR0: 0.349 | HR1: 0.475/SR1: 0.456 | LR: 5.31e-05 [2026-04-17 09:11:32] Epoch 1 | Step 770 | Loss: 0.3786 | LM: 0.3662 | LB: 1.1526 | CL0: 2.9 | CL1: 2.1 | HR0: 0.348/SR0: 0.349 | HR1: 0.475/SR1: 0.456 | LR: 5.36e-05 [2026-04-17 09:11:38] Epoch 1 | Step 780 | Loss: 0.3779 | LM: 0.3650 | LB: 1.1528 | CL0: 2.9 | CL1: 2.1 | HR0: 0.348/SR0: 0.349 | HR1: 0.475/SR1: 0.456 | LR: 5.42e-05 [2026-04-17 09:11:44] Epoch 1 | Step 790 | Loss: 0.3775 | LM: 0.3664 | LB: 1.1528 | CL0: 2.9 | CL1: 2.1 | HR0: 0.348/SR0: 0.349 | HR1: 0.475/SR1: 0.456 | LR: 5.48e-05 [2026-04-17 09:11:51] Epoch 1 | Step 800 | Loss: 0.3769 | LM: 0.3647 | LB: 1.1526 | CL0: 2.9 | CL1: 2.1 | HR0: 0.348/SR0: 0.349 | HR1: 0.475/SR1: 0.456 | LR: 5.53e-05 [2026-04-17 09:11:57] Epoch 1 | Step 810 | Loss: 0.3772 | LM: 0.3644 | LB: 1.1525 | CL0: 2.9 | CL1: 2.1 | HR0: 0.348/SR0: 0.349 | HR1: 0.475/SR1: 0.456 | LR: 5.59e-05 [2026-04-17 09:12:03] Epoch 1 | Step 820 | Loss: 0.3766 | LM: 0.3640 | LB: 1.1524 | CL0: 2.9 | CL1: 2.1 | HR0: 0.348/SR0: 0.349 | HR1: 0.475/SR1: 0.456 | LR: 5.65e-05 [2026-04-17 09:12:09] Epoch 1 | Step 830 | Loss: 0.3764 | LM: 0.3638 | LB: 1.1523 | CL0: 2.9 | CL1: 2.1 | HR0: 0.348/SR0: 0.349 | HR1: 0.475/SR1: 0.456 | LR: 5.70e-05 [2026-04-17 09:12:16] Epoch 1 | Step 840 | Loss: 0.3755 | LM: 0.3628 | LB: 1.1523 | CL0: 2.9 | CL1: 2.1 | HR0: 0.348/SR0: 0.349 | HR1: 0.475/SR1: 0.456 | LR: 5.76e-05 [2026-04-17 09:12:22] Epoch 1 | Step 850 | Loss: 0.3749 | LM: 0.3622 | LB: 1.1522 | CL0: 2.9 | CL1: 2.1 | HR0: 0.347/SR0: 0.349 | HR1: 0.475/SR1: 0.456 | LR: 5.82e-05 [2026-04-17 09:12:28] Epoch 1 | Step 860 | Loss: 0.3749 | LM: 0.3612 | LB: 1.1522 | CL0: 2.9 | CL1: 2.1 | HR0: 0.347/SR0: 0.349 | HR1: 0.475/SR1: 0.456 | LR: 5.87e-05 [2026-04-17 09:12:35] Epoch 1 | Step 870 | Loss: 0.3745 | LM: 0.3609 | LB: 1.1520 | CL0: 2.9 | CL1: 2.1 | HR0: 0.347/SR0: 0.349 | HR1: 0.475/SR1: 0.456 | LR: 5.93e-05 [2026-04-17 09:12:41] Epoch 1 | Step 880 | Loss: 0.3741 | LM: 0.3609 | LB: 1.1522 | CL0: 2.9 | CL1: 2.1 | HR0: 0.348/SR0: 0.349 | HR1: 0.475/SR1: 0.456 | LR: 5.99e-05 [2026-04-17 09:12:48] Epoch 1 | Step 890 | Loss: 0.3735 | LM: 0.3604 | LB: 1.1520 | CL0: 2.9 | CL1: 2.1 | HR0: 0.348/SR0: 0.349 | HR1: 0.475/SR1: 0.455 | LR: 6.04e-05 [2026-04-17 09:12:54] Epoch 1 | Step 900 | Loss: 0.3732 | LM: 0.3594 | LB: 1.1521 | CL0: 2.9 | CL1: 2.1 | HR0: 0.348/SR0: 0.349 | HR1: 0.475/SR1: 0.456 | LR: 6.10e-05 [2026-04-17 09:13:00] Epoch 1 | Step 910 | Loss: 0.3726 | LM: 0.3584 | LB: 1.1519 | CL0: 2.9 | CL1: 2.1 | HR0: 0.347/SR0: 0.349 | HR1: 0.474/SR1: 0.455 | LR: 6.16e-05 [2026-04-17 09:13:07] Epoch 1 | Step 920 | Loss: 0.3721 | LM: 0.3574 | LB: 1.1518 | CL0: 2.9 | CL1: 2.1 | HR0: 0.347/SR0: 0.349 | HR1: 0.474/SR1: 0.455 | LR: 6.21e-05 [2026-04-17 09:13:13] Epoch 1 | Step 930 | Loss: 0.3710 | LM: 0.3565 | LB: 1.1515 | CL0: 2.9 | CL1: 2.1 | HR0: 0.347/SR0: 0.349 | HR1: 0.474/SR1: 0.455 | LR: 6.27e-05 [2026-04-17 09:13:20] Epoch 1 | Step 940 | Loss: 0.3700 | LM: 0.3558 | LB: 1.1514 | CL0: 2.9 | CL1: 2.1 | HR0: 0.347/SR0: 0.349 | HR1: 0.474/SR1: 0.455 | LR: 6.33e-05 [2026-04-17 09:13:26] Epoch 1 | Step 950 | Loss: 0.3694 | LM: 0.3550 | LB: 1.1512 | CL0: 2.9 | CL1: 2.1 | HR0: 0.347/SR0: 0.349 | HR1: 0.474/SR1: 0.455 | LR: 6.38e-05 [2026-04-17 09:13:32] Epoch 1 | Step 960 | Loss: 0.3695 | LM: 0.3546 | LB: 1.1513 | CL0: 2.9 | CL1: 2.1 | HR0: 0.347/SR0: 0.349 | HR1: 0.474/SR1: 0.455 | LR: 6.44e-05 [2026-04-17 09:13:39] Epoch 1 | Step 970 | Loss: 0.3691 | LM: 0.3544 | LB: 1.1512 | CL0: 2.9 | CL1: 2.1 | HR0: 0.347/SR0: 0.349 | HR1: 0.474/SR1: 0.455 | LR: 6.50e-05 [2026-04-17 09:13:45] Epoch 1 | Step 980 | Loss: 0.3688 | LM: 0.3532 | LB: 1.1511 | CL0: 2.9 | CL1: 2.1 | HR0: 0.347/SR0: 0.349 | HR1: 0.474/SR1: 0.455 | LR: 6.55e-05 [2026-04-17 09:13:51] Epoch 1 | Step 990 | Loss: 0.3684 | LM: 0.3533 | LB: 1.1510 | CL0: 2.9 | CL1: 2.1 | HR0: 0.347/SR0: 0.349 | HR1: 0.474/SR1: 0.455 | LR: 6.61e-05 [2026-04-17 09:13:58] Epoch 1 | Step 1000 | Loss: 0.3678 | LM: 0.3532 | LB: 1.1509 | CL0: 2.9 | CL1: 2.1 | HR0: 0.347/SR0: 0.349 | HR1: 0.474/SR1: 0.454 | LR: 6.67e-05 [2026-04-17 09:13:59] Validation | Batch 10/784 | Loss: 0.3496 | LM_LOSS: 0.3382 | LB_LOSS: 1.1399 [2026-04-17 09:14:00] Validation | Batch 20/784 | Loss: 0.3707 | LM_LOSS: 0.3593 | LB_LOSS: 1.1388 [2026-04-17 09:14:02] Validation | Batch 30/784 | Loss: 0.3501 | LM_LOSS: 0.3388 | LB_LOSS: 1.1378 [2026-04-17 09:14:03] Validation | Batch 40/784 | Loss: 0.3493 | LM_LOSS: 0.3379 | LB_LOSS: 1.1376 [2026-04-17 09:14:05] Validation | Batch 50/784 | Loss: 0.3438 | LM_LOSS: 0.3325 | LB_LOSS: 1.1367 [2026-04-17 09:14:06] Validation | Batch 60/784 | Loss: 0.3444 | LM_LOSS: 0.3330 | LB_LOSS: 1.1364 [2026-04-17 09:14:07] Validation | Batch 70/784 | Loss: 0.3415 | LM_LOSS: 0.3301 | LB_LOSS: 1.1355 [2026-04-17 09:14:09] Validation | Batch 80/784 | Loss: 0.3361 | LM_LOSS: 0.3248 | LB_LOSS: 1.1349 [2026-04-17 09:14:10] Validation | Batch 90/784 | Loss: 0.3345 | LM_LOSS: 0.3231 | LB_LOSS: 1.1355 [2026-04-17 09:14:12] Validation | Batch 100/784 | Loss: 0.3363 | LM_LOSS: 0.3249 | LB_LOSS: 1.1364 [2026-04-17 09:14:13] Validation | Batch 110/784 | Loss: 0.3327 | LM_LOSS: 0.3213 | LB_LOSS: 1.1364 [2026-04-17 09:14:14] Validation | Batch 120/784 | Loss: 0.3353 | LM_LOSS: 0.3239 | LB_LOSS: 1.1361 [2026-04-17 09:14:16] Validation | Batch 130/784 | Loss: 0.3386 | LM_LOSS: 0.3273 | LB_LOSS: 1.1362 [2026-04-17 09:14:17] Validation | Batch 140/784 | Loss: 0.3379 | LM_LOSS: 0.3266 | LB_LOSS: 1.1358 [2026-04-17 09:14:19] Validation | Batch 150/784 | Loss: 0.3334 | LM_LOSS: 0.3221 | LB_LOSS: 1.1362 [2026-04-17 09:14:20] Validation | Batch 160/784 | Loss: 0.3335 | LM_LOSS: 0.3222 | LB_LOSS: 1.1359 [2026-04-17 09:14:22] Validation | Batch 170/784 | Loss: 0.3339 | LM_LOSS: 0.3226 | LB_LOSS: 1.1354 [2026-04-17 09:14:23] Validation | Batch 180/784 | Loss: 0.3310 | LM_LOSS: 0.3196 | LB_LOSS: 1.1356 [2026-04-17 09:14:25] Validation | Batch 190/784 | Loss: 0.3321 | LM_LOSS: 0.3207 | LB_LOSS: 1.1361 [2026-04-17 09:14:26] Validation | Batch 200/784 | Loss: 0.3320 | LM_LOSS: 0.3206 | LB_LOSS: 1.1362 [2026-04-17 09:14:27] Validation | Batch 210/784 | Loss: 0.3310 | LM_LOSS: 0.3197 | LB_LOSS: 1.1361 [2026-04-17 09:14:29] Validation | Batch 220/784 | Loss: 0.3315 | LM_LOSS: 0.3201 | LB_LOSS: 1.1361 [2026-04-17 09:14:30] Validation | Batch 230/784 | Loss: 0.3322 | LM_LOSS: 0.3208 | LB_LOSS: 1.1360 [2026-04-17 09:14:32] Validation | Batch 240/784 | Loss: 0.3326 | LM_LOSS: 0.3213 | LB_LOSS: 1.1366 [2026-04-17 09:14:33] Validation | Batch 250/784 | Loss: 0.3323 | LM_LOSS: 0.3210 | LB_LOSS: 1.1363 [2026-04-17 09:14:35] Validation | Batch 260/784 | Loss: 0.3322 | LM_LOSS: 0.3208 | LB_LOSS: 1.1366 [2026-04-17 09:14:37] Validation | Batch 270/784 | Loss: 0.3314 | LM_LOSS: 0.3201 | LB_LOSS: 1.1366 [2026-04-17 09:14:38] Validation | Batch 280/784 | Loss: 0.3318 | LM_LOSS: 0.3204 | LB_LOSS: 1.1369 [2026-04-17 09:14:39] Validation | Batch 290/784 | Loss: 0.3327 | LM_LOSS: 0.3213 | LB_LOSS: 1.1371 [2026-04-17 09:14:41] Validation | Batch 300/784 | Loss: 0.3332 | LM_LOSS: 0.3218 | LB_LOSS: 1.1371 [2026-04-17 09:14:42] Validation | Batch 310/784 | Loss: 0.3325 | LM_LOSS: 0.3211 | LB_LOSS: 1.1371 [2026-04-17 09:14:44] Validation | Batch 320/784 | Loss: 0.3338 | LM_LOSS: 0.3225 | LB_LOSS: 1.1371 [2026-04-17 09:14:45] Validation | Batch 330/784 | Loss: 0.3334 | LM_LOSS: 0.3221 | LB_LOSS: 1.1370 [2026-04-17 09:14:46] Validation | Batch 340/784 | Loss: 0.3323 | LM_LOSS: 0.3209 | LB_LOSS: 1.1373 [2026-04-17 09:14:48] Validation | Batch 350/784 | Loss: 0.3326 | LM_LOSS: 0.3212 | LB_LOSS: 1.1375 [2026-04-17 09:14:49] Validation | Batch 360/784 | Loss: 0.3323 | LM_LOSS: 0.3209 | LB_LOSS: 1.1376 [2026-04-17 09:14:50] Validation | Batch 370/784 | Loss: 0.3326 | LM_LOSS: 0.3212 | LB_LOSS: 1.1375 [2026-04-17 09:14:51] Validation | Batch 380/784 | Loss: 0.3326 | LM_LOSS: 0.3213 | LB_LOSS: 1.1376 [2026-04-17 09:14:53] Validation | Batch 390/784 | Loss: 0.3327 | LM_LOSS: 0.3213 | LB_LOSS: 1.1376 [2026-04-17 09:14:54] Validation | Batch 400/784 | Loss: 0.3328 | LM_LOSS: 0.3214 | LB_LOSS: 1.1376 [2026-04-17 09:14:55] Validation | Batch 410/784 | Loss: 0.3327 | LM_LOSS: 0.3213 | LB_LOSS: 1.1377 [2026-04-17 09:14:57] Validation | Batch 420/784 | Loss: 0.3332 | LM_LOSS: 0.3218 | LB_LOSS: 1.1378 [2026-04-17 09:14:58] Validation | Batch 430/784 | Loss: 0.3331 | LM_LOSS: 0.3218 | LB_LOSS: 1.1377 [2026-04-17 09:14:59] Validation | Batch 440/784 | Loss: 0.3327 | LM_LOSS: 0.3213 | LB_LOSS: 1.1378 [2026-04-17 09:15:01] Validation | Batch 450/784 | Loss: 0.3322 | LM_LOSS: 0.3208 | LB_LOSS: 1.1378 [2026-04-17 09:15:02] Validation | Batch 460/784 | Loss: 0.3325 | LM_LOSS: 0.3211 | LB_LOSS: 1.1379 [2026-04-17 09:15:04] Validation | Batch 470/784 | Loss: 0.3317 | LM_LOSS: 0.3203 | LB_LOSS: 1.1378 [2026-04-17 09:15:05] Validation | Batch 480/784 | Loss: 0.3318 | LM_LOSS: 0.3204 | LB_LOSS: 1.1378 [2026-04-17 09:15:06] Validation | Batch 490/784 | Loss: 0.3313 | LM_LOSS: 0.3199 | LB_LOSS: 1.1377 [2026-04-17 09:15:08] Validation | Batch 500/784 | Loss: 0.3319 | LM_LOSS: 0.3205 | LB_LOSS: 1.1377 [2026-04-17 09:15:09] Validation | Batch 510/784 | Loss: 0.3315 | LM_LOSS: 0.3201 | LB_LOSS: 1.1376 [2026-04-17 09:15:11] Validation | Batch 520/784 | Loss: 0.3314 | LM_LOSS: 0.3200 | LB_LOSS: 1.1375 [2026-04-17 09:15:12] Validation | Batch 530/784 | Loss: 0.3321 | LM_LOSS: 0.3208 | LB_LOSS: 1.1374 [2026-04-17 09:15:13] Validation | Batch 540/784 | Loss: 0.3324 | LM_LOSS: 0.3210 | LB_LOSS: 1.1374 [2026-04-17 09:15:15] Validation | Batch 550/784 | Loss: 0.3339 | LM_LOSS: 0.3225 | LB_LOSS: 1.1374 [2026-04-17 09:15:16] Validation | Batch 560/784 | Loss: 0.3338 | LM_LOSS: 0.3224 | LB_LOSS: 1.1374 [2026-04-17 09:15:18] Validation | Batch 570/784 | Loss: 0.3335 | LM_LOSS: 0.3221 | LB_LOSS: 1.1374 [2026-04-17 09:15:19] Validation | Batch 580/784 | Loss: 0.3328 | LM_LOSS: 0.3215 | LB_LOSS: 1.1374 [2026-04-17 09:15:21] Validation | Batch 590/784 | Loss: 0.3332 | LM_LOSS: 0.3218 | LB_LOSS: 1.1373 [2026-04-17 09:15:22] Validation | Batch 600/784 | Loss: 0.3331 | LM_LOSS: 0.3217 | LB_LOSS: 1.1372 [2026-04-17 09:15:24] Validation | Batch 610/784 | Loss: 0.3332 | LM_LOSS: 0.3218 | LB_LOSS: 1.1372 [2026-04-17 09:15:25] Validation | Batch 620/784 | Loss: 0.3332 | LM_LOSS: 0.3218 | LB_LOSS: 1.1372 [2026-04-17 09:15:27] Validation | Batch 630/784 | Loss: 0.3337 | LM_LOSS: 0.3223 | LB_LOSS: 1.1372 [2026-04-17 09:15:28] Validation | Batch 640/784 | Loss: 0.3340 | LM_LOSS: 0.3227 | LB_LOSS: 1.1371 [2026-04-17 09:15:30] Validation | Batch 650/784 | Loss: 0.3341 | LM_LOSS: 0.3228 | LB_LOSS: 1.1372 [2026-04-17 09:15:31] Validation | Batch 660/784 | Loss: 0.3344 | LM_LOSS: 0.3230 | LB_LOSS: 1.1372 [2026-04-17 09:15:33] Validation | Batch 670/784 | Loss: 0.3348 | LM_LOSS: 0.3234 | LB_LOSS: 1.1372 [2026-04-17 09:15:34] Validation | Batch 680/784 | Loss: 0.3345 | LM_LOSS: 0.3231 | LB_LOSS: 1.1372 [2026-04-17 09:15:36] Validation | Batch 690/784 | Loss: 0.3350 | LM_LOSS: 0.3237 | LB_LOSS: 1.1372 [2026-04-17 09:15:37] Validation | Batch 700/784 | Loss: 0.3351 | LM_LOSS: 0.3237 | LB_LOSS: 1.1371 [2026-04-17 09:15:38] Validation | Batch 710/784 | Loss: 0.3347 | LM_LOSS: 0.3234 | LB_LOSS: 1.1370 [2026-04-17 09:15:40] Validation | Batch 720/784 | Loss: 0.3345 | LM_LOSS: 0.3231 | LB_LOSS: 1.1369 [2026-04-17 09:15:41] Validation | Batch 730/784 | Loss: 0.3339 | LM_LOSS: 0.3226 | LB_LOSS: 1.1369 [2026-04-17 09:15:43] Validation | Batch 740/784 | Loss: 0.3340 | LM_LOSS: 0.3226 | LB_LOSS: 1.1370 [2026-04-17 09:15:44] Validation | Batch 750/784 | Loss: 0.3332 | LM_LOSS: 0.3218 | LB_LOSS: 1.1371 [2026-04-17 09:15:45] Validation | Batch 760/784 | Loss: 0.3333 | LM_LOSS: 0.3219 | LB_LOSS: 1.1371 [2026-04-17 09:15:46] Validation | Batch 770/784 | Loss: 0.3334 | LM_LOSS: 0.3220 | LB_LOSS: 1.1371 [2026-04-17 09:15:48] Validation | Batch 780/784 | Loss: 0.3336 | LM_LOSS: 0.3222 | LB_LOSS: 1.1371 [2026-04-17 09:15:49] Validation | Batch 784/784 | Loss: 0.3338 | LM_LOSS: 0.3224 | LB_LOSS: 1.1371 [2026-04-17 09:15:52] Validation | Loss: 0.3338 | LM_LOSS: 0.3224 | LB_LOSS: 1.1371 | PPL: 1.38 | Time: 111.01s [2026-04-17 09:15:55] New best model saved! Val loss: 0.3338 [2026-04-17 09:16:01] Epoch 1 | Step 1010 | Loss: 0.3675 | LM: 0.3524 | LB: 1.1509 | CL0: 2.9 | CL1: 2.1 | HR0: 0.347/SR0: 0.349 | HR1: 0.474/SR1: 0.454 | LR: 6.72e-05 [2026-04-17 09:16:08] Epoch 1 | Step 1020 | Loss: 0.3668 | LM: 0.3513 | LB: 1.1510 | CL0: 2.9 | CL1: 2.1 | HR0: 0.348/SR0: 0.349 | HR1: 0.474/SR1: 0.454 | LR: 6.78e-05 [2026-04-17 09:16:14] Epoch 1 | Step 1030 | Loss: 0.3663 | LM: 0.3508 | LB: 1.1509 | CL0: 2.9 | CL1: 2.1 | HR0: 0.348/SR0: 0.349 | HR1: 0.473/SR1: 0.454 | LR: 6.84e-05 [2026-04-17 09:16:21] Epoch 1 | Step 1040 | Loss: 0.3655 | LM: 0.3490 | LB: 1.1506 | CL0: 2.9 | CL1: 2.1 | HR0: 0.348/SR0: 0.349 | HR1: 0.473/SR1: 0.454 | LR: 6.89e-05 [2026-04-17 09:16:27] Epoch 1 | Step 1050 | Loss: 0.3650 | LM: 0.3484 | LB: 1.1503 | CL0: 2.9 | CL1: 2.1 | HR0: 0.348/SR0: 0.349 | HR1: 0.473/SR1: 0.454 | LR: 6.95e-05 [2026-04-17 09:16:34] Epoch 1 | Step 1060 | Loss: 0.3647 | LM: 0.3482 | LB: 1.1503 | CL0: 2.9 | CL1: 2.1 | HR0: 0.348/SR0: 0.349 | HR1: 0.473/SR1: 0.454 | LR: 7.01e-05 [2026-04-17 09:16:40] Epoch 1 | Step 1070 | Loss: 0.3646 | LM: 0.3495 | LB: 1.1502 | CL0: 2.9 | CL1: 2.1 | HR0: 0.348/SR0: 0.349 | HR1: 0.473/SR1: 0.454 | LR: 7.06e-05 [2026-04-17 09:16:46] Epoch 1 | Step 1080 | Loss: 0.3644 | LM: 0.3487 | LB: 1.1501 | CL0: 2.9 | CL1: 2.1 | HR0: 0.348/SR0: 0.349 | HR1: 0.473/SR1: 0.454 | LR: 7.12e-05 [2026-04-17 09:16:53] Epoch 1 | Step 1090 | Loss: 0.3642 | LM: 0.3481 | LB: 1.1500 | CL0: 2.9 | CL1: 2.1 | HR0: 0.348/SR0: 0.349 | HR1: 0.473/SR1: 0.453 | LR: 7.18e-05 [2026-04-17 09:16:59] Epoch 1 | Step 1100 | Loss: 0.3636 | LM: 0.3471 | LB: 1.1498 | CL0: 2.9 | CL1: 2.1 | HR0: 0.347/SR0: 0.349 | HR1: 0.473/SR1: 0.453 | LR: 7.23e-05 [2026-04-17 09:17:06] Epoch 1 | Step 1110 | Loss: 0.3631 | LM: 0.3466 | LB: 1.1498 | CL0: 2.9 | CL1: 2.1 | HR0: 0.348/SR0: 0.349 | HR1: 0.473/SR1: 0.453 | LR: 7.29e-05 [2026-04-17 09:17:12] Epoch 1 | Step 1120 | Loss: 0.3624 | LM: 0.3459 | LB: 1.1497 | CL0: 2.9 | CL1: 2.1 | HR0: 0.347/SR0: 0.349 | HR1: 0.473/SR1: 0.453 | LR: 7.35e-05 [2026-04-17 09:17:18] Epoch 1 | Step 1130 | Loss: 0.3622 | LM: 0.3463 | LB: 1.1496 | CL0: 2.9 | CL1: 2.1 | HR0: 0.347/SR0: 0.349 | HR1: 0.473/SR1: 0.453 | LR: 7.40e-05 [2026-04-17 09:17:25] Epoch 1 | Step 1140 | Loss: 0.3618 | LM: 0.3463 | LB: 1.1495 | CL0: 2.9 | CL1: 2.1 | HR0: 0.348/SR0: 0.349 | HR1: 0.472/SR1: 0.453 | LR: 7.46e-05 [2026-04-17 09:17:31] Epoch 1 | Step 1150 | Loss: 0.3615 | LM: 0.3463 | LB: 1.1495 | CL0: 2.9 | CL1: 2.1 | HR0: 0.347/SR0: 0.349 | HR1: 0.472/SR1: 0.453 | LR: 7.52e-05 [2026-04-17 09:17:37] Epoch 1 | Step 1160 | Loss: 0.3613 | LM: 0.3463 | LB: 1.1497 | CL0: 2.9 | CL1: 2.1 | HR0: 0.348/SR0: 0.349 | HR1: 0.473/SR1: 0.453 | LR: 7.57e-05 [2026-04-17 09:17:44] Epoch 1 | Step 1170 | Loss: 0.3607 | LM: 0.3458 | LB: 1.1495 | CL0: 2.9 | CL1: 2.1 | HR0: 0.347/SR0: 0.349 | HR1: 0.472/SR1: 0.453 | LR: 7.63e-05 [2026-04-17 09:17:50] Epoch 1 | Step 1180 | Loss: 0.3604 | LM: 0.3452 | LB: 1.1493 | CL0: 2.9 | CL1: 2.1 | HR0: 0.347/SR0: 0.349 | HR1: 0.472/SR1: 0.453 | LR: 7.69e-05 [2026-04-17 09:17:56] Epoch 1 | Step 1190 | Loss: 0.3600 | LM: 0.3451 | LB: 1.1493 | CL0: 2.9 | CL1: 2.1 | HR0: 0.347/SR0: 0.349 | HR1: 0.472/SR1: 0.453 | LR: 7.74e-05 [2026-04-17 09:18:03] Epoch 1 | Step 1200 | Loss: 0.3595 | LM: 0.3451 | LB: 1.1492 | CL0: 2.9 | CL1: 2.1 | HR0: 0.347/SR0: 0.349 | HR1: 0.472/SR1: 0.453 | LR: 7.80e-05 [2026-04-17 09:18:09] Epoch 1 | Step 1210 | Loss: 0.3590 | LM: 0.3444 | LB: 1.1492 | CL0: 2.9 | CL1: 2.1 | HR0: 0.348/SR0: 0.349 | HR1: 0.472/SR1: 0.453 | LR: 7.86e-05 [2026-04-17 09:18:15] Epoch 1 | Step 1220 | Loss: 0.3585 | LM: 0.3440 | LB: 1.1491 | CL0: 2.9 | CL1: 2.1 | HR0: 0.348/SR0: 0.349 | HR1: 0.472/SR1: 0.453 | LR: 7.91e-05 [2026-04-17 09:18:22] Epoch 1 | Step 1230 | Loss: 0.3581 | LM: 0.3429 | LB: 1.1490 | CL0: 2.9 | CL1: 2.1 | HR0: 0.348/SR0: 0.349 | HR1: 0.472/SR1: 0.453 | LR: 7.97e-05 [2026-04-17 09:18:28] Epoch 1 | Step 1240 | Loss: 0.3578 | LM: 0.3429 | LB: 1.1490 | CL0: 2.9 | CL1: 2.1 | HR0: 0.348/SR0: 0.349 | HR1: 0.472/SR1: 0.453 | LR: 8.03e-05 [2026-04-17 09:18:35] Epoch 1 | Step 1250 | Loss: 0.3582 | LM: 0.3429 | LB: 1.1489 | CL0: 2.9 | CL1: 2.1 | HR0: 0.348/SR0: 0.349 | HR1: 0.472/SR1: 0.453 | LR: 8.08e-05 [2026-04-17 09:18:41] Epoch 1 | Step 1260 | Loss: 0.3575 | LM: 0.3423 | LB: 1.1488 | CL0: 2.9 | CL1: 2.1 | HR0: 0.348/SR0: 0.349 | HR1: 0.472/SR1: 0.452 | LR: 8.14e-05 [2026-04-17 09:18:47] Epoch 1 | Step 1270 | Loss: 0.3570 | LM: 0.3417 | LB: 1.1487 | CL0: 2.9 | CL1: 2.1 | HR0: 0.348/SR0: 0.349 | HR1: 0.472/SR1: 0.452 | LR: 8.20e-05 [2026-04-17 09:18:54] Epoch 1 | Step 1280 | Loss: 0.3563 | LM: 0.3411 | LB: 1.1485 | CL0: 2.9 | CL1: 2.1 | HR0: 0.347/SR0: 0.349 | HR1: 0.472/SR1: 0.452 | LR: 8.25e-05 [2026-04-17 09:19:00] Epoch 1 | Step 1290 | Loss: 0.3559 | LM: 0.3411 | LB: 1.1484 | CL0: 2.9 | CL1: 2.1 | HR0: 0.348/SR0: 0.349 | HR1: 0.472/SR1: 0.452 | LR: 8.31e-05 [2026-04-17 09:19:06] Epoch 1 | Step 1300 | Loss: 0.3556 | LM: 0.3412 | LB: 1.1482 | CL0: 2.9 | CL1: 2.1 | HR0: 0.348/SR0: 0.349 | HR1: 0.471/SR1: 0.452 | LR: 8.37e-05 [2026-04-17 09:19:13] Epoch 1 | Step 1310 | Loss: 0.3550 | LM: 0.3405 | LB: 1.1481 | CL0: 2.9 | CL1: 2.1 | HR0: 0.348/SR0: 0.349 | HR1: 0.471/SR1: 0.452 | LR: 8.42e-05 [2026-04-17 09:19:19] Epoch 1 | Step 1320 | Loss: 0.3547 | LM: 0.3406 | LB: 1.1481 | CL0: 2.9 | CL1: 2.1 | HR0: 0.348/SR0: 0.349 | HR1: 0.471/SR1: 0.452 | LR: 8.48e-05 [2026-04-17 09:19:25] Epoch 1 | Step 1330 | Loss: 0.3542 | LM: 0.3404 | LB: 1.1481 | CL0: 2.9 | CL1: 2.1 | HR0: 0.348/SR0: 0.349 | HR1: 0.471/SR1: 0.452 | LR: 8.54e-05 [2026-04-17 09:19:32] Epoch 1 | Step 1340 | Loss: 0.3542 | LM: 0.3400 | LB: 1.1480 | CL0: 2.9 | CL1: 2.1 | HR0: 0.348/SR0: 0.349 | HR1: 0.471/SR1: 0.452 | LR: 8.59e-05 [2026-04-17 09:19:39] Epoch 1 | Step 1350 | Loss: 0.3543 | LM: 0.3405 | LB: 1.1479 | CL0: 2.9 | CL1: 2.1 | HR0: 0.348/SR0: 0.349 | HR1: 0.471/SR1: 0.451 | LR: 8.65e-05 [2026-04-17 09:19:45] Epoch 1 | Step 1360 | Loss: 0.3538 | LM: 0.3399 | LB: 1.1477 | CL0: 2.9 | CL1: 2.1 | HR0: 0.348/SR0: 0.349 | HR1: 0.471/SR1: 0.451 | LR: 8.71e-05 [2026-04-17 09:19:51] Epoch 1 | Step 1370 | Loss: 0.3532 | LM: 0.3394 | LB: 1.1475 | CL0: 2.9 | CL1: 2.1 | HR0: 0.348/SR0: 0.349 | HR1: 0.471/SR1: 0.451 | LR: 8.76e-05 [2026-04-17 09:19:58] Epoch 1 | Step 1380 | Loss: 0.3529 | LM: 0.3384 | LB: 1.1475 | CL0: 2.9 | CL1: 2.1 | HR0: 0.348/SR0: 0.349 | HR1: 0.471/SR1: 0.451 | LR: 8.82e-05 [2026-04-17 09:20:04] Epoch 1 | Step 1390 | Loss: 0.3529 | LM: 0.3387 | LB: 1.1475 | CL0: 2.9 | CL1: 2.1 | HR0: 0.348/SR0: 0.349 | HR1: 0.471/SR1: 0.451 | LR: 8.88e-05 [2026-04-17 09:20:11] Epoch 1 | Step 1400 | Loss: 0.3525 | LM: 0.3387 | LB: 1.1474 | CL0: 2.9 | CL1: 2.1 | HR0: 0.348/SR0: 0.349 | HR1: 0.471/SR1: 0.451 | LR: 8.93e-05 [2026-04-17 09:20:17] Epoch 1 | Step 1410 | Loss: 0.3524 | LM: 0.3385 | LB: 1.1472 | CL0: 2.9 | CL1: 2.1 | HR0: 0.348/SR0: 0.349 | HR1: 0.470/SR1: 0.451 | LR: 8.99e-05 [2026-04-17 09:20:23] Epoch 1 | Step 1420 | Loss: 0.3519 | LM: 0.3377 | LB: 1.1470 | CL0: 2.9 | CL1: 2.1 | HR0: 0.348/SR0: 0.349 | HR1: 0.470/SR1: 0.451 | LR: 9.05e-05 [2026-04-17 09:20:30] Epoch 1 | Step 1430 | Loss: 0.3518 | LM: 0.3382 | LB: 1.1469 | CL0: 2.9 | CL1: 2.1 | HR0: 0.348/SR0: 0.349 | HR1: 0.470/SR1: 0.451 | LR: 9.10e-05 [2026-04-17 09:20:36] Epoch 1 | Step 1440 | Loss: 0.3515 | LM: 0.3375 | LB: 1.1467 | CL0: 2.9 | CL1: 2.1 | HR0: 0.348/SR0: 0.349 | HR1: 0.470/SR1: 0.450 | LR: 9.16e-05 [2026-04-17 09:20:42] Epoch 1 | Step 1450 | Loss: 0.3509 | LM: 0.3371 | LB: 1.1466 | CL0: 2.9 | CL1: 2.1 | HR0: 0.348/SR0: 0.349 | HR1: 0.470/SR1: 0.450 | LR: 9.22e-05 [2026-04-17 09:20:49] Epoch 1 | Step 1460 | Loss: 0.3505 | LM: 0.3364 | LB: 1.1465 | CL0: 2.9 | CL1: 2.1 | HR0: 0.348/SR0: 0.349 | HR1: 0.470/SR1: 0.450 | LR: 9.27e-05 [2026-04-17 09:20:55] Epoch 1 | Step 1470 | Loss: 0.3502 | LM: 0.3364 | LB: 1.1465 | CL0: 2.9 | CL1: 2.1 | HR0: 0.347/SR0: 0.349 | HR1: 0.470/SR1: 0.450 | LR: 9.33e-05 [2026-04-17 09:21:02] Epoch 1 | Step 1480 | Loss: 0.3499 | LM: 0.3362 | LB: 1.1464 | CL0: 2.9 | CL1: 2.1 | HR0: 0.347/SR0: 0.349 | HR1: 0.470/SR1: 0.450 | LR: 9.39e-05 [2026-04-17 09:21:08] Epoch 1 | Step 1490 | Loss: 0.3501 | LM: 0.3369 | LB: 1.1464 | CL0: 2.9 | CL1: 2.1 | HR0: 0.348/SR0: 0.349 | HR1: 0.470/SR1: 0.450 | LR: 9.44e-05 [2026-04-17 09:21:14] Epoch 1 | Step 1500 | Loss: 0.3501 | LM: 0.3364 | LB: 1.1462 | CL0: 2.9 | CL1: 2.1 | HR0: 0.348/SR0: 0.349 | HR1: 0.470/SR1: 0.450 | LR: 9.50e-05 [2026-04-17 09:21:20] Epoch 1 | Step 1510 | Loss: 0.3498 | LM: 0.3362 | LB: 1.1462 | CL0: 2.9 | CL1: 2.1 | HR0: 0.348/SR0: 0.349 | HR1: 0.470/SR1: 0.450 | LR: 9.56e-05 [2026-04-17 09:21:27] Epoch 1 | Step 1520 | Loss: 0.3497 | LM: 0.3364 | LB: 1.1461 | CL0: 2.9 | CL1: 2.1 | HR0: 0.348/SR0: 0.349 | HR1: 0.469/SR1: 0.450 | LR: 9.61e-05 [2026-04-17 09:21:33] Epoch 1 | Step 1530 | Loss: 0.3495 | LM: 0.3362 | LB: 1.1461 | CL0: 2.9 | CL1: 2.1 | HR0: 0.348/SR0: 0.349 | HR1: 0.469/SR1: 0.450 | LR: 9.67e-05 [2026-04-17 09:21:39] Epoch 1 | Step 1540 | Loss: 0.3494 | LM: 0.3364 | LB: 1.1461 | CL0: 2.9 | CL1: 2.1 | HR0: 0.348/SR0: 0.349 | HR1: 0.470/SR1: 0.450 | LR: 9.73e-05 [2026-04-17 09:21:46] Epoch 1 | Step 1550 | Loss: 0.3490 | LM: 0.3358 | LB: 1.1461 | CL0: 2.9 | CL1: 2.1 | HR0: 0.348/SR0: 0.349 | HR1: 0.470/SR1: 0.450 | LR: 9.78e-05 [2026-04-17 09:21:52] Epoch 1 | Step 1560 | Loss: 0.3490 | LM: 0.3359 | LB: 1.1460 | CL0: 2.9 | CL1: 2.1 | HR0: 0.348/SR0: 0.349 | HR1: 0.469/SR1: 0.450 | LR: 9.84e-05 [2026-04-17 09:21:58] Epoch 1 | Step 1570 | Loss: 0.3486 | LM: 0.3355 | LB: 1.1459 | CL0: 2.9 | CL1: 2.1 | HR0: 0.348/SR0: 0.349 | HR1: 0.469/SR1: 0.450 | LR: 9.90e-05 [2026-04-17 09:22:04] Epoch 1 | Step 1580 | Loss: 0.3482 | LM: 0.3347 | LB: 1.1458 | CL0: 2.9 | CL1: 2.1 | HR0: 0.348/SR0: 0.349 | HR1: 0.469/SR1: 0.450 | LR: 9.95e-05 [2026-04-17 09:22:10] Epoch 1 | Step 1590 | Loss: 0.3479 | LM: 0.3345 | LB: 1.1457 | CL0: 2.9 | CL1: 2.1 | HR0: 0.347/SR0: 0.349 | HR1: 0.469/SR1: 0.450 | LR: 1.00e-04 [2026-04-17 09:22:17] Epoch 1 | Step 1600 | Loss: 0.3477 | LM: 0.3345 | LB: 1.1456 | CL0: 2.9 | CL1: 2.1 | HR0: 0.347/SR0: 0.349 | HR1: 0.469/SR1: 0.449 | LR: 1.00e-04 [2026-04-17 09:22:23] Epoch 1 | Step 1610 | Loss: 0.3473 | LM: 0.3341 | LB: 1.1454 | CL0: 2.9 | CL1: 2.1 | HR0: 0.347/SR0: 0.349 | HR1: 0.469/SR1: 0.449 | LR: 1.00e-04 [2026-04-17 09:22:30] Epoch 1 | Step 1620 | Loss: 0.3470 | LM: 0.3335 | LB: 1.1454 | CL0: 2.9 | CL1: 2.1 | HR0: 0.348/SR0: 0.349 | HR1: 0.469/SR1: 0.449 | LR: 1.00e-04 [2026-04-17 09:22:36] Epoch 1 | Step 1630 | Loss: 0.3467 | LM: 0.3332 | LB: 1.1453 | CL0: 2.9 | CL1: 2.1 | HR0: 0.348/SR0: 0.349 | HR1: 0.469/SR1: 0.449 | LR: 1.00e-04 [2026-04-17 09:22:42] Epoch 1 | Step 1640 | Loss: 0.3465 | LM: 0.3327 | LB: 1.1452 | CL0: 2.9 | CL1: 2.1 | HR0: 0.348/SR0: 0.349 | HR1: 0.469/SR1: 0.449 | LR: 1.00e-04 [2026-04-17 09:22:48] Epoch 1 | Step 1650 | Loss: 0.3464 | LM: 0.3325 | LB: 1.1451 | CL0: 2.9 | CL1: 2.1 | HR0: 0.348/SR0: 0.349 | HR1: 0.469/SR1: 0.449 | LR: 1.00e-04 [2026-04-17 09:22:54] Epoch 1 | Step 1660 | Loss: 0.3463 | LM: 0.3321 | LB: 1.1451 | CL0: 2.9 | CL1: 2.1 | HR0: 0.348/SR0: 0.349 | HR1: 0.469/SR1: 0.449 | LR: 1.00e-04 [2026-04-17 09:23:00] Epoch 1 | Step 1670 | Loss: 0.3458 | LM: 0.3316 | LB: 1.1449 | CL0: 2.9 | CL1: 2.1 | HR0: 0.348/SR0: 0.349 | HR1: 0.469/SR1: 0.449 | LR: 1.00e-04 [2026-04-17 09:23:07] Epoch 1 | Step 1680 | Loss: 0.3459 | LM: 0.3317 | LB: 1.1448 | CL0: 2.9 | CL1: 2.1 | HR0: 0.348/SR0: 0.349 | HR1: 0.469/SR1: 0.449 | LR: 1.00e-04 [2026-04-17 09:23:13] Epoch 1 | Step 1690 | Loss: 0.3459 | LM: 0.3319 | LB: 1.1448 | CL0: 2.9 | CL1: 2.1 | HR0: 0.348/SR0: 0.349 | HR1: 0.469/SR1: 0.449 | LR: 1.00e-04 [2026-04-17 09:23:19] Epoch 1 | Step 1700 | Loss: 0.3457 | LM: 0.3313 | LB: 1.1447 | CL0: 2.9 | CL1: 2.1 | HR0: 0.347/SR0: 0.349 | HR1: 0.468/SR1: 0.449 | LR: 1.00e-04 [2026-04-17 09:23:26] Epoch 1 | Step 1710 | Loss: 0.3455 | LM: 0.3311 | LB: 1.1446 | CL0: 2.9 | CL1: 2.1 | HR0: 0.348/SR0: 0.349 | HR1: 0.468/SR1: 0.449 | LR: 1.00e-04 [2026-04-17 09:23:32] Epoch 1 | Step 1720 | Loss: 0.3453 | LM: 0.3310 | LB: 1.1445 | CL0: 2.9 | CL1: 2.1 | HR0: 0.347/SR0: 0.349 | HR1: 0.468/SR1: 0.449 | LR: 1.00e-04 [2026-04-17 09:23:38] Epoch 1 | Step 1730 | Loss: 0.3451 | LM: 0.3306 | LB: 1.1444 | CL0: 2.9 | CL1: 2.1 | HR0: 0.347/SR0: 0.349 | HR1: 0.468/SR1: 0.448 | LR: 1.00e-04 [2026-04-17 09:23:45] Epoch 1 | Step 1740 | Loss: 0.3451 | LM: 0.3307 | LB: 1.1443 | CL0: 2.9 | CL1: 2.1 | HR0: 0.347/SR0: 0.349 | HR1: 0.468/SR1: 0.448 | LR: 1.00e-04 [2026-04-17 09:23:51] Epoch 1 | Step 1750 | Loss: 0.3451 | LM: 0.3306 | LB: 1.1442 | CL0: 2.9 | CL1: 2.1 | HR0: 0.347/SR0: 0.349 | HR1: 0.468/SR1: 0.448 | LR: 1.00e-04 [2026-04-17 09:23:57] Epoch 1 | Step 1760 | Loss: 0.3449 | LM: 0.3307 | LB: 1.1440 | CL0: 2.9 | CL1: 2.1 | HR0: 0.347/SR0: 0.349 | HR1: 0.468/SR1: 0.448 | LR: 1.00e-04 [2026-04-17 09:24:04] Epoch 1 | Step 1770 | Loss: 0.3446 | LM: 0.3303 | LB: 1.1439 | CL0: 2.9 | CL1: 2.1 | HR0: 0.347/SR0: 0.349 | HR1: 0.468/SR1: 0.448 | LR: 1.00e-04 [2026-04-17 09:24:10] Epoch 1 | Step 1780 | Loss: 0.3447 | LM: 0.3300 | LB: 1.1437 | CL0: 2.9 | CL1: 2.1 | HR0: 0.347/SR0: 0.348 | HR1: 0.468/SR1: 0.448 | LR: 1.00e-04 [2026-04-17 09:24:16] Epoch 1 | Step 1790 | Loss: 0.3448 | LM: 0.3297 | LB: 1.1436 | CL0: 2.9 | CL1: 2.1 | HR0: 0.347/SR0: 0.349 | HR1: 0.468/SR1: 0.448 | LR: 1.00e-04 [2026-04-17 09:24:22] Epoch 1 | Step 1800 | Loss: 0.3448 | LM: 0.3300 | LB: 1.1435 | CL0: 2.9 | CL1: 2.1 | HR0: 0.347/SR0: 0.349 | HR1: 0.467/SR1: 0.448 | LR: 1.00e-04 [2026-04-17 09:24:29] Epoch 1 | Step 1810 | Loss: 0.3448 | LM: 0.3302 | LB: 1.1435 | CL0: 2.9 | CL1: 2.1 | HR0: 0.347/SR0: 0.349 | HR1: 0.467/SR1: 0.448 | LR: 1.00e-04 [2026-04-17 09:24:35] Epoch 1 | Step 1820 | Loss: 0.3447 | LM: 0.3300 | LB: 1.1434 | CL0: 2.9 | CL1: 2.1 | HR0: 0.347/SR0: 0.349 | HR1: 0.467/SR1: 0.448 | LR: 1.00e-04 [2026-04-17 09:24:41] Epoch 1 | Step 1830 | Loss: 0.3445 | LM: 0.3298 | LB: 1.1433 | CL0: 2.9 | CL1: 2.1 | HR0: 0.347/SR0: 0.349 | HR1: 0.467/SR1: 0.447 | LR: 1.00e-04 [2026-04-17 09:24:47] Epoch 1 | Step 1840 | Loss: 0.3443 | LM: 0.3301 | LB: 1.1432 | CL0: 2.9 | CL1: 2.1 | HR0: 0.347/SR0: 0.349 | HR1: 0.467/SR1: 0.447 | LR: 1.00e-04 [2026-04-17 09:24:54] Epoch 1 | Step 1850 | Loss: 0.3442 | LM: 0.3302 | LB: 1.1432 | CL0: 2.9 | CL1: 2.1 | HR0: 0.347/SR0: 0.349 | HR1: 0.467/SR1: 0.447 | LR: 1.00e-04 [2026-04-17 09:25:00] Epoch 1 | Step 1860 | Loss: 0.3442 | LM: 0.3299 | LB: 1.1430 | CL0: 2.9 | CL1: 2.1 | HR0: 0.347/SR0: 0.349 | HR1: 0.467/SR1: 0.447 | LR: 1.00e-04 [2026-04-17 09:25:06] Epoch 1 | Step 1870 | Loss: 0.3441 | LM: 0.3297 | LB: 1.1429 | CL0: 2.9 | CL1: 2.1 | HR0: 0.347/SR0: 0.349 | HR1: 0.467/SR1: 0.447 | LR: 1.00e-04 [2026-04-17 09:25:12] Epoch 1 | Step 1880 | Loss: 0.3444 | LM: 0.3301 | LB: 1.1428 | CL0: 2.9 | CL1: 2.2 | HR0: 0.347/SR0: 0.349 | HR1: 0.467/SR1: 0.447 | LR: 1.00e-04 [2026-04-17 09:25:19] Epoch 1 | Step 1890 | Loss: 0.3444 | LM: 0.3304 | LB: 1.1427 | CL0: 2.9 | CL1: 2.2 | HR0: 0.347/SR0: 0.349 | HR1: 0.467/SR1: 0.447 | LR: 1.00e-04 [2026-04-17 09:25:25] Epoch 1 | Step 1900 | Loss: 0.3444 | LM: 0.3300 | LB: 1.1426 | CL0: 2.9 | CL1: 2.2 | HR0: 0.347/SR0: 0.349 | HR1: 0.467/SR1: 0.447 | LR: 1.00e-04 [2026-04-17 09:25:31] Epoch 1 | Step 1910 | Loss: 0.3444 | LM: 0.3295 | LB: 1.1425 | CL0: 2.9 | CL1: 2.2 | HR0: 0.347/SR0: 0.349 | HR1: 0.466/SR1: 0.447 | LR: 1.00e-04 [2026-04-17 09:25:40] Epoch 1 | Step 1920 | Loss: 0.3439 | LM: 0.3290 | LB: 1.1424 | CL0: 2.9 | CL1: 2.2 | HR0: 0.347/SR0: 0.349 | HR1: 0.466/SR1: 0.446 | LR: 1.00e-04 [2026-04-17 09:25:46] Epoch 1 | Step 1930 | Loss: 0.3440 | LM: 0.3292 | LB: 1.1422 | CL0: 2.9 | CL1: 2.2 | HR0: 0.347/SR0: 0.349 | HR1: 0.466/SR1: 0.446 | LR: 1.00e-04 [2026-04-17 09:25:52] Epoch 1 | Step 1940 | Loss: 0.3437 | LM: 0.3288 | LB: 1.1420 | CL0: 2.9 | CL1: 2.2 | HR0: 0.347/SR0: 0.349 | HR1: 0.466/SR1: 0.446 | LR: 1.00e-04 [2026-04-17 09:25:59] Epoch 1 | Step 1950 | Loss: 0.3437 | LM: 0.3291 | LB: 1.1419 | CL0: 2.9 | CL1: 2.2 | HR0: 0.347/SR0: 0.349 | HR1: 0.466/SR1: 0.446 | LR: 1.00e-04 [2026-04-17 09:26:05] Epoch 1 | Step 1960 | Loss: 0.3438 | LM: 0.3295 | LB: 1.1418 | CL0: 2.9 | CL1: 2.2 | HR0: 0.347/SR0: 0.349 | HR1: 0.466/SR1: 0.446 | LR: 1.00e-04 [2026-04-17 09:26:12] Epoch 1 | Step 1970 | Loss: 0.3434 | LM: 0.3292 | LB: 1.1417 | CL0: 2.9 | CL1: 2.2 | HR0: 0.348/SR0: 0.349 | HR1: 0.466/SR1: 0.446 | LR: 1.00e-04 [2026-04-17 09:26:18] Epoch 1 | Step 1980 | Loss: 0.3431 | LM: 0.3290 | LB: 1.1416 | CL0: 2.9 | CL1: 2.2 | HR0: 0.347/SR0: 0.349 | HR1: 0.466/SR1: 0.446 | LR: 1.00e-04 [2026-04-17 09:26:25] Epoch 1 | Step 1990 | Loss: 0.3431 | LM: 0.3287 | LB: 1.1414 | CL0: 2.9 | CL1: 2.2 | HR0: 0.347/SR0: 0.349 | HR1: 0.466/SR1: 0.446 | LR: 1.00e-04 [2026-04-17 09:26:31] Epoch 1 | Step 2000 | Loss: 0.3429 | LM: 0.3290 | LB: 1.1414 | CL0: 2.9 | CL1: 2.2 | HR0: 0.347/SR0: 0.349 | HR1: 0.465/SR1: 0.445 | LR: 1.00e-04 [2026-04-17 09:26:32] Validation | Batch 10/784 | Loss: 0.3425 | LM_LOSS: 0.3313 | LB_LOSS: 1.1139 [2026-04-17 09:26:34] Validation | Batch 20/784 | Loss: 0.3556 | LM_LOSS: 0.3444 | LB_LOSS: 1.1135 [2026-04-17 09:26:35] Validation | Batch 30/784 | Loss: 0.3379 | LM_LOSS: 0.3268 | LB_LOSS: 1.1123 [2026-04-17 09:26:37] Validation | Batch 40/784 | Loss: 0.3373 | LM_LOSS: 0.3262 | LB_LOSS: 1.1120 [2026-04-17 09:26:38] Validation | Batch 50/784 | Loss: 0.3329 | LM_LOSS: 0.3218 | LB_LOSS: 1.1111 [2026-04-17 09:26:40] Validation | Batch 60/784 | Loss: 0.3338 | LM_LOSS: 0.3227 | LB_LOSS: 1.1108 [2026-04-17 09:26:41] Validation | Batch 70/784 | Loss: 0.3314 | LM_LOSS: 0.3203 | LB_LOSS: 1.1099 [2026-04-17 09:26:42] Validation | Batch 80/784 | Loss: 0.3265 | LM_LOSS: 0.3154 | LB_LOSS: 1.1094 [2026-04-17 09:26:44] Validation | Batch 90/784 | Loss: 0.3251 | LM_LOSS: 0.3140 | LB_LOSS: 1.1099 [2026-04-17 09:26:45] Validation | Batch 100/784 | Loss: 0.3269 | LM_LOSS: 0.3157 | LB_LOSS: 1.1106 [2026-04-17 09:26:46] Validation | Batch 110/784 | Loss: 0.3228 | LM_LOSS: 0.3117 | LB_LOSS: 1.1106 [2026-04-17 09:26:48] Validation | Batch 120/784 | Loss: 0.3254 | LM_LOSS: 0.3143 | LB_LOSS: 1.1104 [2026-04-17 09:26:49] Validation | Batch 130/784 | Loss: 0.3282 | LM_LOSS: 0.3171 | LB_LOSS: 1.1104 [2026-04-17 09:26:51] Validation | Batch 140/784 | Loss: 0.3272 | LM_LOSS: 0.3161 | LB_LOSS: 1.1101 [2026-04-17 09:26:52] Validation | Batch 150/784 | Loss: 0.3230 | LM_LOSS: 0.3119 | LB_LOSS: 1.1105 [2026-04-17 09:26:53] Validation | Batch 160/784 | Loss: 0.3235 | LM_LOSS: 0.3124 | LB_LOSS: 1.1102 [2026-04-17 09:26:55] Validation | Batch 170/784 | Loss: 0.3240 | LM_LOSS: 0.3129 | LB_LOSS: 1.1097 [2026-04-17 09:26:56] Validation | Batch 180/784 | Loss: 0.3213 | LM_LOSS: 0.3102 | LB_LOSS: 1.1098 [2026-04-17 09:26:58] Validation | Batch 190/784 | Loss: 0.3227 | LM_LOSS: 0.3116 | LB_LOSS: 1.1103 [2026-04-17 09:26:59] Validation | Batch 200/784 | Loss: 0.3227 | LM_LOSS: 0.3116 | LB_LOSS: 1.1104 [2026-04-17 09:27:00] Validation | Batch 210/784 | Loss: 0.3216 | LM_LOSS: 0.3105 | LB_LOSS: 1.1103 [2026-04-17 09:27:02] Validation | Batch 220/784 | Loss: 0.3224 | LM_LOSS: 0.3113 | LB_LOSS: 1.1103 [2026-04-17 09:27:03] Validation | Batch 230/784 | Loss: 0.3228 | LM_LOSS: 0.3117 | LB_LOSS: 1.1102 [2026-04-17 09:27:05] Validation | Batch 240/784 | Loss: 0.3234 | LM_LOSS: 0.3123 | LB_LOSS: 1.1107 [2026-04-17 09:27:06] Validation | Batch 250/784 | Loss: 0.3232 | LM_LOSS: 0.3121 | LB_LOSS: 1.1105 [2026-04-17 09:27:08] Validation | Batch 260/784 | Loss: 0.3232 | LM_LOSS: 0.3121 | LB_LOSS: 1.1107 [2026-04-17 09:27:09] Validation | Batch 270/784 | Loss: 0.3225 | LM_LOSS: 0.3114 | LB_LOSS: 1.1107 [2026-04-17 09:27:11] Validation | Batch 280/784 | Loss: 0.3229 | LM_LOSS: 0.3118 | LB_LOSS: 1.1109 [2026-04-17 09:27:12] Validation | Batch 290/784 | Loss: 0.3239 | LM_LOSS: 0.3128 | LB_LOSS: 1.1111 [2026-04-17 09:27:13] Validation | Batch 300/784 | Loss: 0.3244 | LM_LOSS: 0.3133 | LB_LOSS: 1.1111 [2026-04-17 09:27:15] Validation | Batch 310/784 | Loss: 0.3238 | LM_LOSS: 0.3126 | LB_LOSS: 1.1111 [2026-04-17 09:27:16] Validation | Batch 320/784 | Loss: 0.3251 | LM_LOSS: 0.3139 | LB_LOSS: 1.1111 [2026-04-17 09:27:18] Validation | Batch 330/784 | Loss: 0.3247 | LM_LOSS: 0.3136 | LB_LOSS: 1.1111 [2026-04-17 09:27:19] Validation | Batch 340/784 | Loss: 0.3238 | LM_LOSS: 0.3127 | LB_LOSS: 1.1113 [2026-04-17 09:27:20] Validation | Batch 350/784 | Loss: 0.3238 | LM_LOSS: 0.3127 | LB_LOSS: 1.1115 [2026-04-17 09:27:21] Validation | Batch 360/784 | Loss: 0.3234 | LM_LOSS: 0.3122 | LB_LOSS: 1.1115 [2026-04-17 09:27:23] Validation | Batch 370/784 | Loss: 0.3237 | LM_LOSS: 0.3125 | LB_LOSS: 1.1114 [2026-04-17 09:27:24] Validation | Batch 380/784 | Loss: 0.3236 | LM_LOSS: 0.3125 | LB_LOSS: 1.1115 [2026-04-17 09:27:26] Validation | Batch 390/784 | Loss: 0.3235 | LM_LOSS: 0.3124 | LB_LOSS: 1.1115 [2026-04-17 09:27:27] Validation | Batch 400/784 | Loss: 0.3235 | LM_LOSS: 0.3124 | LB_LOSS: 1.1115 [2026-04-17 09:27:28] Validation | Batch 410/784 | Loss: 0.3236 | LM_LOSS: 0.3125 | LB_LOSS: 1.1116 [2026-04-17 09:27:29] Validation | Batch 420/784 | Loss: 0.3238 | LM_LOSS: 0.3127 | LB_LOSS: 1.1117 [2026-04-17 09:27:31] Validation | Batch 430/784 | Loss: 0.3238 | LM_LOSS: 0.3127 | LB_LOSS: 1.1116 [2026-04-17 09:27:32] Validation | Batch 440/784 | Loss: 0.3235 | LM_LOSS: 0.3124 | LB_LOSS: 1.1117 [2026-04-17 09:27:33] Validation | Batch 450/784 | Loss: 0.3231 | LM_LOSS: 0.3120 | LB_LOSS: 1.1116 [2026-04-17 09:27:35] Validation | Batch 460/784 | Loss: 0.3235 | LM_LOSS: 0.3124 | LB_LOSS: 1.1117 [2026-04-17 09:27:36] Validation | Batch 470/784 | Loss: 0.3226 | LM_LOSS: 0.3115 | LB_LOSS: 1.1116 [2026-04-17 09:27:37] Validation | Batch 480/784 | Loss: 0.3229 | LM_LOSS: 0.3118 | LB_LOSS: 1.1116 [2026-04-17 09:27:39] Validation | Batch 490/784 | Loss: 0.3223 | LM_LOSS: 0.3112 | LB_LOSS: 1.1115 [2026-04-17 09:27:40] Validation | Batch 500/784 | Loss: 0.3229 | LM_LOSS: 0.3118 | LB_LOSS: 1.1115 [2026-04-17 09:27:42] Validation | Batch 510/784 | Loss: 0.3226 | LM_LOSS: 0.3114 | LB_LOSS: 1.1114 [2026-04-17 09:27:43] Validation | Batch 520/784 | Loss: 0.3226 | LM_LOSS: 0.3115 | LB_LOSS: 1.1113 [2026-04-17 09:27:44] Validation | Batch 530/784 | Loss: 0.3234 | LM_LOSS: 0.3123 | LB_LOSS: 1.1113 [2026-04-17 09:27:46] Validation | Batch 540/784 | Loss: 0.3237 | LM_LOSS: 0.3126 | LB_LOSS: 1.1113 [2026-04-17 09:27:47] Validation | Batch 550/784 | Loss: 0.3251 | LM_LOSS: 0.3140 | LB_LOSS: 1.1112 [2026-04-17 09:27:49] Validation | Batch 560/784 | Loss: 0.3251 | LM_LOSS: 0.3140 | LB_LOSS: 1.1113 [2026-04-17 09:27:50] Validation | Batch 570/784 | Loss: 0.3247 | LM_LOSS: 0.3136 | LB_LOSS: 1.1112 [2026-04-17 09:27:51] Validation | Batch 580/784 | Loss: 0.3242 | LM_LOSS: 0.3131 | LB_LOSS: 1.1113 [2026-04-17 09:27:53] Validation | Batch 590/784 | Loss: 0.3245 | LM_LOSS: 0.3134 | LB_LOSS: 1.1112 [2026-04-17 09:27:54] Validation | Batch 600/784 | Loss: 0.3244 | LM_LOSS: 0.3133 | LB_LOSS: 1.1111 [2026-04-17 09:27:56] Validation | Batch 610/784 | Loss: 0.3245 | LM_LOSS: 0.3134 | LB_LOSS: 1.1111 [2026-04-17 09:27:57] Validation | Batch 620/784 | Loss: 0.3246 | LM_LOSS: 0.3135 | LB_LOSS: 1.1111 [2026-04-17 09:27:59] Validation | Batch 630/784 | Loss: 0.3252 | LM_LOSS: 0.3141 | LB_LOSS: 1.1111 [2026-04-17 09:28:00] Validation | Batch 640/784 | Loss: 0.3254 | LM_LOSS: 0.3143 | LB_LOSS: 1.1110 [2026-04-17 09:28:02] Validation | Batch 650/784 | Loss: 0.3254 | LM_LOSS: 0.3143 | LB_LOSS: 1.1111 [2026-04-17 09:28:03] Validation | Batch 660/784 | Loss: 0.3257 | LM_LOSS: 0.3146 | LB_LOSS: 1.1111 [2026-04-17 09:28:05] Validation | Batch 670/784 | Loss: 0.3261 | LM_LOSS: 0.3150 | LB_LOSS: 1.1112 [2026-04-17 09:28:06] Validation | Batch 680/784 | Loss: 0.3258 | LM_LOSS: 0.3147 | LB_LOSS: 1.1111 [2026-04-17 09:28:07] Validation | Batch 690/784 | Loss: 0.3261 | LM_LOSS: 0.3150 | LB_LOSS: 1.1111 [2026-04-17 09:28:09] Validation | Batch 700/784 | Loss: 0.3261 | LM_LOSS: 0.3150 | LB_LOSS: 1.1110 [2026-04-17 09:28:10] Validation | Batch 710/784 | Loss: 0.3259 | LM_LOSS: 0.3147 | LB_LOSS: 1.1110 [2026-04-17 09:28:12] Validation | Batch 720/784 | Loss: 0.3256 | LM_LOSS: 0.3145 | LB_LOSS: 1.1109 [2026-04-17 09:28:13] Validation | Batch 730/784 | Loss: 0.3250 | LM_LOSS: 0.3139 | LB_LOSS: 1.1108 [2026-04-17 09:28:14] Validation | Batch 740/784 | Loss: 0.3251 | LM_LOSS: 0.3140 | LB_LOSS: 1.1109 [2026-04-17 09:28:15] Validation | Batch 750/784 | Loss: 0.3244 | LM_LOSS: 0.3133 | LB_LOSS: 1.1110 [2026-04-17 09:28:17] Validation | Batch 760/784 | Loss: 0.3245 | LM_LOSS: 0.3134 | LB_LOSS: 1.1110 [2026-04-17 09:28:18] Validation | Batch 770/784 | Loss: 0.3246 | LM_LOSS: 0.3135 | LB_LOSS: 1.1110 [2026-04-17 09:28:19] Validation | Batch 780/784 | Loss: 0.3248 | LM_LOSS: 0.3137 | LB_LOSS: 1.1110 [2026-04-17 09:28:20] Validation | Batch 784/784 | Loss: 0.3250 | LM_LOSS: 0.3139 | LB_LOSS: 1.1110 [2026-04-17 09:28:23] Validation | Loss: 0.3250 | LM_LOSS: 0.3139 | LB_LOSS: 1.1110 | PPL: 1.37 | Time: 109.01s [2026-04-17 09:28:28] New best model saved! Val loss: 0.3250 [2026-04-17 09:28:34] Epoch 1 | Step 2010 | Loss: 0.3426 | LM: 0.3291 | LB: 1.1412 | CL0: 2.9 | CL1: 2.2 | HR0: 0.347/SR0: 0.349 | HR1: 0.465/SR1: 0.445 | LR: 1.00e-04 [2026-04-17 09:28:41] Epoch 1 | Step 2020 | Loss: 0.3426 | LM: 0.3290 | LB: 1.1411 | CL0: 2.9 | CL1: 2.2 | HR0: 0.347/SR0: 0.349 | HR1: 0.465/SR1: 0.445 | LR: 1.00e-04 [2026-04-17 09:28:47] Epoch 1 | Step 2030 | Loss: 0.3428 | LM: 0.3291 | LB: 1.1410 | CL0: 2.9 | CL1: 2.2 | HR0: 0.347/SR0: 0.349 | HR1: 0.465/SR1: 0.445 | LR: 1.00e-04 [2026-04-17 09:28:53] Epoch 1 | Step 2040 | Loss: 0.3424 | LM: 0.3287 | LB: 1.1409 | CL0: 2.9 | CL1: 2.2 | HR0: 0.347/SR0: 0.349 | HR1: 0.465/SR1: 0.445 | LR: 1.00e-04 [2026-04-17 09:28:59] Epoch 1 | Step 2050 | Loss: 0.3422 | LM: 0.3286 | LB: 1.1408 | CL0: 2.9 | CL1: 2.2 | HR0: 0.347/SR0: 0.349 | HR1: 0.465/SR1: 0.445 | LR: 1.00e-04 [2026-04-17 09:29:06] Epoch 1 | Step 2060 | Loss: 0.3420 | LM: 0.3282 | LB: 1.1407 | CL0: 2.9 | CL1: 2.2 | HR0: 0.347/SR0: 0.349 | HR1: 0.465/SR1: 0.445 | LR: 1.00e-04 [2026-04-17 09:29:12] Epoch 1 | Step 2070 | Loss: 0.3419 | LM: 0.3280 | LB: 1.1406 | CL0: 2.9 | CL1: 2.2 | HR0: 0.347/SR0: 0.349 | HR1: 0.465/SR1: 0.445 | LR: 1.00e-04 [2026-04-17 09:29:19] Epoch 1 | Step 2080 | Loss: 0.3417 | LM: 0.3275 | LB: 1.1405 | CL0: 2.9 | CL1: 2.2 | HR0: 0.347/SR0: 0.349 | HR1: 0.465/SR1: 0.445 | LR: 1.00e-04 [2026-04-17 09:29:25] Epoch 1 | Step 2090 | Loss: 0.3413 | LM: 0.3271 | LB: 1.1404 | CL0: 2.9 | CL1: 2.2 | HR0: 0.347/SR0: 0.349 | HR1: 0.465/SR1: 0.444 | LR: 1.00e-04 [2026-04-17 09:29:32] Epoch 1 | Step 2100 | Loss: 0.3412 | LM: 0.3271 | LB: 1.1403 | CL0: 2.9 | CL1: 2.2 | HR0: 0.347/SR0: 0.349 | HR1: 0.465/SR1: 0.444 | LR: 1.00e-04 [2026-04-17 09:29:38] Epoch 1 | Step 2110 | Loss: 0.3410 | LM: 0.3270 | LB: 1.1402 | CL0: 2.9 | CL1: 2.2 | HR0: 0.347/SR0: 0.349 | HR1: 0.464/SR1: 0.444 | LR: 1.00e-04 [2026-04-17 09:29:44] Epoch 1 | Step 2120 | Loss: 0.3409 | LM: 0.3266 | LB: 1.1400 | CL0: 2.9 | CL1: 2.2 | HR0: 0.347/SR0: 0.349 | HR1: 0.464/SR1: 0.444 | LR: 1.00e-04 [2026-04-17 09:29:51] Epoch 1 | Step 2130 | Loss: 0.3408 | LM: 0.3270 | LB: 1.1400 | CL0: 2.9 | CL1: 2.2 | HR0: 0.347/SR0: 0.349 | HR1: 0.464/SR1: 0.444 | LR: 1.00e-04 [2026-04-17 09:29:57] Epoch 1 | Step 2140 | Loss: 0.3407 | LM: 0.3267 | LB: 1.1399 | CL0: 2.9 | CL1: 2.2 | HR0: 0.347/SR0: 0.349 | HR1: 0.464/SR1: 0.444 | LR: 1.00e-04 [2026-04-17 09:30:04] Epoch 1 | Step 2150 | Loss: 0.3407 | LM: 0.3267 | LB: 1.1398 | CL0: 2.9 | CL1: 2.2 | HR0: 0.347/SR0: 0.349 | HR1: 0.464/SR1: 0.444 | LR: 1.00e-04 [2026-04-17 09:30:10] Epoch 1 | Step 2160 | Loss: 0.3406 | LM: 0.3264 | LB: 1.1397 | CL0: 2.9 | CL1: 2.2 | HR0: 0.347/SR0: 0.349 | HR1: 0.464/SR1: 0.444 | LR: 1.00e-04 [2026-04-17 09:30:17] Epoch 1 | Step 2170 | Loss: 0.3408 | LM: 0.3263 | LB: 1.1396 | CL0: 2.9 | CL1: 2.2 | HR0: 0.347/SR0: 0.349 | HR1: 0.464/SR1: 0.444 | LR: 1.00e-04 [2026-04-17 09:30:23] Epoch 1 | Step 2180 | Loss: 0.3407 | LM: 0.3259 | LB: 1.1395 | CL0: 2.9 | CL1: 2.2 | HR0: 0.347/SR0: 0.349 | HR1: 0.464/SR1: 0.444 | LR: 1.00e-04 [2026-04-17 09:30:29] Epoch 1 | Step 2190 | Loss: 0.3406 | LM: 0.3258 | LB: 1.1394 | CL0: 2.9 | CL1: 2.2 | HR0: 0.347/SR0: 0.349 | HR1: 0.464/SR1: 0.444 | LR: 1.00e-04 [2026-04-17 09:30:36] Epoch 1 | Step 2200 | Loss: 0.3406 | LM: 0.3257 | LB: 1.1392 | CL0: 2.9 | CL1: 2.2 | HR0: 0.347/SR0: 0.349 | HR1: 0.464/SR1: 0.443 | LR: 1.00e-04 [2026-04-17 09:30:42] Epoch 1 | Step 2210 | Loss: 0.3405 | LM: 0.3257 | LB: 1.1392 | CL0: 2.9 | CL1: 2.2 | HR0: 0.347/SR0: 0.349 | HR1: 0.464/SR1: 0.443 | LR: 1.00e-04 [2026-04-17 09:30:49] Epoch 1 | Step 2220 | Loss: 0.3403 | LM: 0.3254 | LB: 1.1391 | CL0: 2.9 | CL1: 2.2 | HR0: 0.347/SR0: 0.349 | HR1: 0.464/SR1: 0.443 | LR: 1.00e-04 [2026-04-17 09:30:55] Epoch 1 | Step 2230 | Loss: 0.3404 | LM: 0.3255 | LB: 1.1390 | CL0: 2.9 | CL1: 2.2 | HR0: 0.347/SR0: 0.349 | HR1: 0.464/SR1: 0.443 | LR: 1.00e-04 [2026-04-17 09:31:01] Epoch 1 | Step 2240 | Loss: 0.3401 | LM: 0.3251 | LB: 1.1389 | CL0: 2.9 | CL1: 2.2 | HR0: 0.347/SR0: 0.349 | HR1: 0.463/SR1: 0.443 | LR: 1.00e-04 [2026-04-17 09:31:08] Epoch 1 | Step 2250 | Loss: 0.3400 | LM: 0.3248 | LB: 1.1388 | CL0: 2.9 | CL1: 2.2 | HR0: 0.347/SR0: 0.349 | HR1: 0.463/SR1: 0.443 | LR: 1.00e-04 [2026-04-17 09:31:14] Epoch 1 | Step 2260 | Loss: 0.3400 | LM: 0.3248 | LB: 1.1386 | CL0: 2.9 | CL1: 2.2 | HR0: 0.347/SR0: 0.349 | HR1: 0.463/SR1: 0.443 | LR: 1.00e-04 [2026-04-17 09:31:21] Epoch 1 | Step 2270 | Loss: 0.3401 | LM: 0.3248 | LB: 1.1385 | CL0: 2.9 | CL1: 2.2 | HR0: 0.347/SR0: 0.349 | HR1: 0.463/SR1: 0.443 | LR: 1.00e-04 [2026-04-17 09:31:27] Epoch 1 | Step 2280 | Loss: 0.3399 | LM: 0.3248 | LB: 1.1385 | CL0: 2.9 | CL1: 2.2 | HR0: 0.347/SR0: 0.349 | HR1: 0.463/SR1: 0.443 | LR: 1.00e-04 [2026-04-17 09:31:34] Epoch 1 | Step 2290 | Loss: 0.3397 | LM: 0.3246 | LB: 1.1384 | CL0: 2.9 | CL1: 2.2 | HR0: 0.347/SR0: 0.349 | HR1: 0.463/SR1: 0.443 | LR: 1.00e-04 [2026-04-17 09:31:40] Epoch 1 | Step 2300 | Loss: 0.3395 | LM: 0.3242 | LB: 1.1383 | CL0: 2.9 | CL1: 2.2 | HR0: 0.347/SR0: 0.349 | HR1: 0.463/SR1: 0.442 | LR: 1.00e-04 [2026-04-17 09:31:46] Epoch 1 | Step 2310 | Loss: 0.3392 | LM: 0.3239 | LB: 1.1382 | CL0: 2.9 | CL1: 2.2 | HR0: 0.347/SR0: 0.349 | HR1: 0.463/SR1: 0.442 | LR: 1.00e-04 [2026-04-17 09:31:53] Epoch 1 | Step 2320 | Loss: 0.3393 | LM: 0.3242 | LB: 1.1382 | CL0: 2.9 | CL1: 2.2 | HR0: 0.347/SR0: 0.349 | HR1: 0.463/SR1: 0.442 | LR: 1.00e-04 [2026-04-17 09:31:59] Epoch 1 | Step 2330 | Loss: 0.3391 | LM: 0.3242 | LB: 1.1381 | CL0: 2.9 | CL1: 2.2 | HR0: 0.347/SR0: 0.349 | HR1: 0.463/SR1: 0.442 | LR: 1.00e-04 [2026-04-17 09:32:05] Epoch 1 | Step 2340 | Loss: 0.3391 | LM: 0.3244 | LB: 1.1381 | CL0: 2.9 | CL1: 2.2 | HR0: 0.347/SR0: 0.349 | HR1: 0.463/SR1: 0.442 | LR: 1.00e-04 [2026-04-17 09:32:12] Epoch 1 | Step 2350 | Loss: 0.3389 | LM: 0.3243 | LB: 1.1379 | CL0: 2.9 | CL1: 2.2 | HR0: 0.347/SR0: 0.349 | HR1: 0.463/SR1: 0.442 | LR: 1.00e-04 [2026-04-17 09:32:18] Epoch 1 | Step 2360 | Loss: 0.3389 | LM: 0.3245 | LB: 1.1379 | CL0: 2.9 | CL1: 2.2 | HR0: 0.347/SR0: 0.349 | HR1: 0.463/SR1: 0.442 | LR: 1.00e-04 [2026-04-17 09:32:25] Epoch 1 | Step 2370 | Loss: 0.3389 | LM: 0.3242 | LB: 1.1378 | CL0: 2.9 | CL1: 2.2 | HR0: 0.347/SR0: 0.349 | HR1: 0.462/SR1: 0.442 | LR: 1.00e-04 [2026-04-17 09:32:31] Epoch 1 | Step 2380 | Loss: 0.3388 | LM: 0.3240 | LB: 1.1376 | CL0: 2.9 | CL1: 2.2 | HR0: 0.347/SR0: 0.349 | HR1: 0.462/SR1: 0.442 | LR: 1.00e-04 [2026-04-17 09:32:37] Epoch 1 | Step 2390 | Loss: 0.3386 | LM: 0.3238 | LB: 1.1376 | CL0: 2.9 | CL1: 2.2 | HR0: 0.347/SR0: 0.349 | HR1: 0.462/SR1: 0.442 | LR: 1.00e-04 [2026-04-17 09:32:44] Epoch 1 | Step 2400 | Loss: 0.3384 | LM: 0.3237 | LB: 1.1375 | CL0: 2.9 | CL1: 2.2 | HR0: 0.347/SR0: 0.349 | HR1: 0.462/SR1: 0.442 | LR: 1.00e-04 [2026-04-17 09:32:50] Epoch 1 | Step 2410 | Loss: 0.3383 | LM: 0.3237 | LB: 1.1373 | CL0: 2.9 | CL1: 2.2 | HR0: 0.347/SR0: 0.349 | HR1: 0.462/SR1: 0.441 | LR: 1.00e-04 [2026-04-17 09:32:57] Epoch 1 | Step 2420 | Loss: 0.3381 | LM: 0.3236 | LB: 1.1371 | CL0: 2.9 | CL1: 2.2 | HR0: 0.347/SR0: 0.348 | HR1: 0.462/SR1: 0.441 | LR: 1.00e-04 [2026-04-17 09:33:03] Epoch 1 | Step 2430 | Loss: 0.3382 | LM: 0.3238 | LB: 1.1370 | CL0: 2.9 | CL1: 2.2 | HR0: 0.347/SR0: 0.348 | HR1: 0.462/SR1: 0.441 | LR: 1.00e-04 [2026-04-17 09:33:09] Epoch 1 | Step 2440 | Loss: 0.3380 | LM: 0.3235 | LB: 1.1369 | CL0: 2.9 | CL1: 2.2 | HR0: 0.347/SR0: 0.348 | HR1: 0.462/SR1: 0.441 | LR: 1.00e-04 [2026-04-17 09:33:16] Epoch 1 | Step 2450 | Loss: 0.3377 | LM: 0.3232 | LB: 1.1368 | CL0: 2.9 | CL1: 2.2 | HR0: 0.347/SR0: 0.348 | HR1: 0.462/SR1: 0.441 | LR: 1.00e-04 [2026-04-17 09:33:22] Epoch 1 | Step 2460 | Loss: 0.3377 | LM: 0.3230 | LB: 1.1366 | CL0: 2.9 | CL1: 2.2 | HR0: 0.347/SR0: 0.348 | HR1: 0.461/SR1: 0.441 | LR: 1.00e-04 [2026-04-17 09:33:28] Epoch 1 | Step 2470 | Loss: 0.3376 | LM: 0.3230 | LB: 1.1366 | CL0: 2.9 | CL1: 2.2 | HR0: 0.347/SR0: 0.348 | HR1: 0.461/SR1: 0.441 | LR: 1.00e-04 [2026-04-17 09:33:35] Epoch 1 | Step 2480 | Loss: 0.3376 | LM: 0.3231 | LB: 1.1364 | CL0: 2.9 | CL1: 2.2 | HR0: 0.347/SR0: 0.348 | HR1: 0.461/SR1: 0.441 | LR: 1.00e-04 [2026-04-17 09:33:41] Epoch 1 | Step 2490 | Loss: 0.3376 | LM: 0.3231 | LB: 1.1363 | CL0: 2.9 | CL1: 2.2 | HR0: 0.347/SR0: 0.348 | HR1: 0.461/SR1: 0.441 | LR: 1.00e-04 [2026-04-17 09:33:48] Epoch 1 | Step 2500 | Loss: 0.3374 | LM: 0.3229 | LB: 1.1362 | CL0: 2.9 | CL1: 2.2 | HR0: 0.347/SR0: 0.348 | HR1: 0.461/SR1: 0.441 | LR: 1.00e-04 [2026-04-17 09:33:54] Epoch 1 | Step 2510 | Loss: 0.3375 | LM: 0.3230 | LB: 1.1361 | CL0: 2.9 | CL1: 2.2 | HR0: 0.347/SR0: 0.348 | HR1: 0.461/SR1: 0.440 | LR: 1.00e-04 [2026-04-17 09:34:01] Epoch 1 | Step 2520 | Loss: 0.3373 | LM: 0.3227 | LB: 1.1360 | CL0: 2.9 | CL1: 2.2 | HR0: 0.347/SR0: 0.348 | HR1: 0.461/SR1: 0.440 | LR: 1.00e-04 [2026-04-17 09:34:07] Epoch 1 | Step 2530 | Loss: 0.3374 | LM: 0.3227 | LB: 1.1360 | CL0: 2.9 | CL1: 2.2 | HR0: 0.347/SR0: 0.348 | HR1: 0.461/SR1: 0.440 | LR: 1.00e-04 [2026-04-17 09:34:13] Epoch 1 | Step 2540 | Loss: 0.3373 | LM: 0.3224 | LB: 1.1358 | CL0: 2.9 | CL1: 2.2 | HR0: 0.347/SR0: 0.348 | HR1: 0.461/SR1: 0.440 | LR: 1.00e-04 [2026-04-17 09:34:20] Epoch 1 | Step 2550 | Loss: 0.3372 | LM: 0.3224 | LB: 1.1358 | CL0: 2.9 | CL1: 2.2 | HR0: 0.347/SR0: 0.348 | HR1: 0.461/SR1: 0.440 | LR: 1.00e-04 [2026-04-17 09:34:26] Epoch 1 | Step 2560 | Loss: 0.3371 | LM: 0.3225 | LB: 1.1357 | CL0: 2.9 | CL1: 2.2 | HR0: 0.347/SR0: 0.348 | HR1: 0.460/SR1: 0.440 | LR: 1.00e-04 [2026-04-17 09:34:33] Epoch 1 | Step 2570 | Loss: 0.3371 | LM: 0.3225 | LB: 1.1357 | CL0: 2.9 | CL1: 2.2 | HR0: 0.347/SR0: 0.348 | HR1: 0.460/SR1: 0.440 | LR: 1.00e-04 [2026-04-17 09:34:39] Epoch 1 | Step 2580 | Loss: 0.3369 | LM: 0.3222 | LB: 1.1356 | CL0: 2.9 | CL1: 2.2 | HR0: 0.347/SR0: 0.348 | HR1: 0.460/SR1: 0.440 | LR: 1.00e-04 [2026-04-17 09:34:45] Epoch 1 | Step 2590 | Loss: 0.3369 | LM: 0.3220 | LB: 1.1355 | CL0: 2.9 | CL1: 2.2 | HR0: 0.347/SR0: 0.348 | HR1: 0.460/SR1: 0.440 | LR: 1.00e-04 [2026-04-17 09:34:52] Epoch 1 | Step 2600 | Loss: 0.3366 | LM: 0.3221 | LB: 1.1354 | CL0: 2.9 | CL1: 2.2 | HR0: 0.347/SR0: 0.348 | HR1: 0.460/SR1: 0.440 | LR: 1.00e-04 [2026-04-17 09:34:58] Epoch 1 | Step 2610 | Loss: 0.3365 | LM: 0.3218 | LB: 1.1354 | CL0: 2.9 | CL1: 2.2 | HR0: 0.347/SR0: 0.348 | HR1: 0.460/SR1: 0.440 | LR: 1.00e-04 [2026-04-17 09:35:04] Epoch 1 | Step 2620 | Loss: 0.3366 | LM: 0.3217 | LB: 1.1353 | CL0: 2.9 | CL1: 2.2 | HR0: 0.347/SR0: 0.348 | HR1: 0.460/SR1: 0.440 | LR: 1.00e-04 [2026-04-17 09:35:11] Epoch 1 | Step 2630 | Loss: 0.3364 | LM: 0.3214 | LB: 1.1352 | CL0: 2.9 | CL1: 2.2 | HR0: 0.347/SR0: 0.348 | HR1: 0.460/SR1: 0.439 | LR: 1.00e-04 [2026-04-17 09:35:17] Epoch 1 | Step 2640 | Loss: 0.3364 | LM: 0.3213 | LB: 1.1351 | CL0: 2.9 | CL1: 2.2 | HR0: 0.347/SR0: 0.348 | HR1: 0.460/SR1: 0.439 | LR: 1.00e-04 [2026-04-17 09:35:24] Epoch 1 | Step 2650 | Loss: 0.3362 | LM: 0.3214 | LB: 1.1351 | CL0: 2.9 | CL1: 2.2 | HR0: 0.347/SR0: 0.348 | HR1: 0.460/SR1: 0.439 | LR: 1.00e-04 [2026-04-17 09:35:30] Epoch 1 | Step 2660 | Loss: 0.3360 | LM: 0.3211 | LB: 1.1350 | CL0: 2.9 | CL1: 2.2 | HR0: 0.347/SR0: 0.348 | HR1: 0.460/SR1: 0.439 | LR: 1.00e-04 [2026-04-17 09:35:36] Epoch 1 | Step 2670 | Loss: 0.3357 | LM: 0.3210 | LB: 1.1349 | CL0: 2.9 | CL1: 2.2 | HR0: 0.347/SR0: 0.348 | HR1: 0.460/SR1: 0.439 | LR: 1.00e-04 [2026-04-17 09:35:43] Epoch 1 | Step 2680 | Loss: 0.3356 | LM: 0.3209 | LB: 1.1348 | CL0: 2.9 | CL1: 2.2 | HR0: 0.347/SR0: 0.348 | HR1: 0.460/SR1: 0.439 | LR: 1.00e-04 [2026-04-17 09:35:49] Epoch 1 | Step 2690 | Loss: 0.3355 | LM: 0.3206 | LB: 1.1347 | CL0: 2.9 | CL1: 2.2 | HR0: 0.347/SR0: 0.348 | HR1: 0.460/SR1: 0.439 | LR: 1.00e-04 [2026-04-17 09:35:56] Epoch 1 | Step 2700 | Loss: 0.3354 | LM: 0.3207 | LB: 1.1347 | CL0: 2.9 | CL1: 2.2 | HR0: 0.347/SR0: 0.348 | HR1: 0.460/SR1: 0.439 | LR: 1.00e-04 [2026-04-17 09:36:02] Epoch 1 | Step 2710 | Loss: 0.3353 | LM: 0.3204 | LB: 1.1346 | CL0: 2.9 | CL1: 2.2 | HR0: 0.347/SR0: 0.348 | HR1: 0.459/SR1: 0.439 | LR: 1.00e-04 [2026-04-17 09:36:09] Epoch 1 | Step 2720 | Loss: 0.3351 | LM: 0.3202 | LB: 1.1345 | CL0: 2.9 | CL1: 2.2 | HR0: 0.347/SR0: 0.348 | HR1: 0.459/SR1: 0.439 | LR: 1.00e-04 [2026-04-17 09:36:15] Epoch 1 | Step 2730 | Loss: 0.3350 | LM: 0.3202 | LB: 1.1345 | CL0: 2.9 | CL1: 2.2 | HR0: 0.347/SR0: 0.348 | HR1: 0.459/SR1: 0.439 | LR: 1.00e-04 [2026-04-17 09:36:21] Epoch 1 | Step 2740 | Loss: 0.3349 | LM: 0.3204 | LB: 1.1344 | CL0: 2.9 | CL1: 2.2 | HR0: 0.347/SR0: 0.348 | HR1: 0.459/SR1: 0.439 | LR: 1.00e-04 [2026-04-17 09:36:28] Epoch 1 | Step 2750 | Loss: 0.3349 | LM: 0.3205 | LB: 1.1344 | CL0: 2.9 | CL1: 2.2 | HR0: 0.347/SR0: 0.348 | HR1: 0.459/SR1: 0.439 | LR: 1.00e-04 [2026-04-17 09:36:34] Epoch 1 | Step 2760 | Loss: 0.3350 | LM: 0.3204 | LB: 1.1343 | CL0: 2.9 | CL1: 2.2 | HR0: 0.347/SR0: 0.348 | HR1: 0.459/SR1: 0.439 | LR: 1.00e-04 [2026-04-17 09:36:41] Epoch 1 | Step 2770 | Loss: 0.3349 | LM: 0.3207 | LB: 1.1342 | CL0: 2.9 | CL1: 2.2 | HR0: 0.347/SR0: 0.348 | HR1: 0.459/SR1: 0.439 | LR: 1.00e-04 [2026-04-17 09:36:47] Epoch 1 | Step 2780 | Loss: 0.3348 | LM: 0.3208 | LB: 1.1342 | CL0: 2.9 | CL1: 2.2 | HR0: 0.347/SR0: 0.348 | HR1: 0.459/SR1: 0.438 | LR: 1.00e-04 [2026-04-17 09:36:53] Epoch 1 | Step 2790 | Loss: 0.3348 | LM: 0.3211 | LB: 1.1341 | CL0: 2.9 | CL1: 2.2 | HR0: 0.347/SR0: 0.348 | HR1: 0.459/SR1: 0.438 | LR: 1.00e-04 [2026-04-17 09:37:00] Epoch 1 | Step 2800 | Loss: 0.3347 | LM: 0.3210 | LB: 1.1340 | CL0: 2.9 | CL1: 2.2 | HR0: 0.347/SR0: 0.348 | HR1: 0.459/SR1: 0.438 | LR: 1.00e-04 [2026-04-17 09:37:06] Epoch 1 | Step 2810 | Loss: 0.3346 | LM: 0.3211 | LB: 1.1339 | CL0: 2.9 | CL1: 2.2 | HR0: 0.347/SR0: 0.348 | HR1: 0.459/SR1: 0.438 | LR: 1.00e-04 [2026-04-17 09:37:13] Epoch 1 | Step 2820 | Loss: 0.3345 | LM: 0.3210 | LB: 1.1338 | CL0: 2.9 | CL1: 2.2 | HR0: 0.347/SR0: 0.348 | HR1: 0.459/SR1: 0.438 | LR: 1.00e-04 [2026-04-17 09:37:19] Epoch 1 | Step 2830 | Loss: 0.3344 | LM: 0.3210 | LB: 1.1337 | CL0: 2.9 | CL1: 2.2 | HR0: 0.347/SR0: 0.348 | HR1: 0.459/SR1: 0.438 | LR: 1.00e-04 [2026-04-17 09:37:26] Epoch 1 | Step 2840 | Loss: 0.3343 | LM: 0.3211 | LB: 1.1336 | CL0: 2.9 | CL1: 2.2 | HR0: 0.347/SR0: 0.348 | HR1: 0.459/SR1: 0.438 | LR: 1.00e-04 [2026-04-17 09:37:32] Epoch 1 | Step 2850 | Loss: 0.3344 | LM: 0.3209 | LB: 1.1335 | CL0: 2.9 | CL1: 2.2 | HR0: 0.347/SR0: 0.348 | HR1: 0.459/SR1: 0.438 | LR: 1.00e-04 [2026-04-17 09:37:38] Epoch 1 | Step 2860 | Loss: 0.3344 | LM: 0.3207 | LB: 1.1335 | CL0: 2.9 | CL1: 2.2 | HR0: 0.347/SR0: 0.348 | HR1: 0.458/SR1: 0.438 | LR: 1.00e-04 [2026-04-17 09:37:45] Epoch 1 | Step 2870 | Loss: 0.3344 | LM: 0.3210 | LB: 1.1334 | CL0: 2.9 | CL1: 2.2 | HR0: 0.347/SR0: 0.348 | HR1: 0.458/SR1: 0.438 | LR: 1.00e-04 [2026-04-17 09:37:51] Epoch 1 | Step 2880 | Loss: 0.3343 | LM: 0.3211 | LB: 1.1332 | CL0: 2.9 | CL1: 2.2 | HR0: 0.347/SR0: 0.348 | HR1: 0.458/SR1: 0.438 | LR: 1.00e-04 [2026-04-17 09:37:57] Epoch 1 | Step 2890 | Loss: 0.3341 | LM: 0.3208 | LB: 1.1331 | CL0: 2.9 | CL1: 2.2 | HR0: 0.347/SR0: 0.348 | HR1: 0.458/SR1: 0.437 | LR: 1.00e-04 [2026-04-17 09:38:04] Epoch 1 | Step 2900 | Loss: 0.3340 | LM: 0.3208 | LB: 1.1330 | CL0: 2.9 | CL1: 2.2 | HR0: 0.347/SR0: 0.348 | HR1: 0.458/SR1: 0.437 | LR: 1.00e-04 [2026-04-17 09:38:10] Epoch 1 | Step 2910 | Loss: 0.3338 | LM: 0.3207 | LB: 1.1329 | CL0: 2.9 | CL1: 2.2 | HR0: 0.347/SR0: 0.348 | HR1: 0.458/SR1: 0.437 | LR: 1.00e-04 [2026-04-17 09:38:17] Epoch 1 | Step 2920 | Loss: 0.3337 | LM: 0.3207 | LB: 1.1328 | CL0: 2.9 | CL1: 2.2 | HR0: 0.347/SR0: 0.348 | HR1: 0.458/SR1: 0.437 | LR: 1.00e-04 [2026-04-17 09:38:23] Epoch 1 | Step 2930 | Loss: 0.3335 | LM: 0.3204 | LB: 1.1327 | CL0: 2.9 | CL1: 2.2 | HR0: 0.347/SR0: 0.348 | HR1: 0.458/SR1: 0.437 | LR: 1.00e-04 [2026-04-17 09:38:29] Epoch 1 | Step 2940 | Loss: 0.3335 | LM: 0.3204 | LB: 1.1326 | CL0: 2.9 | CL1: 2.2 | HR0: 0.347/SR0: 0.348 | HR1: 0.457/SR1: 0.437 | LR: 1.00e-04 [2026-04-17 09:38:36] Epoch 1 | Step 2950 | Loss: 0.3334 | LM: 0.3205 | LB: 1.1325 | CL0: 2.9 | CL1: 2.2 | HR0: 0.347/SR0: 0.348 | HR1: 0.457/SR1: 0.437 | LR: 1.00e-04 [2026-04-17 09:38:42] Epoch 1 | Step 2960 | Loss: 0.3332 | LM: 0.3204 | LB: 1.1324 | CL0: 2.9 | CL1: 2.2 | HR0: 0.347/SR0: 0.348 | HR1: 0.457/SR1: 0.437 | LR: 1.00e-04 [2026-04-17 09:38:49] Epoch 1 | Step 2970 | Loss: 0.3332 | LM: 0.3204 | LB: 1.1323 | CL0: 2.9 | CL1: 2.2 | HR0: 0.347/SR0: 0.348 | HR1: 0.457/SR1: 0.437 | LR: 1.00e-04 [2026-04-17 09:38:55] Epoch 1 | Step 2980 | Loss: 0.3332 | LM: 0.3206 | LB: 1.1322 | CL0: 2.9 | CL1: 2.2 | HR0: 0.347/SR0: 0.348 | HR1: 0.457/SR1: 0.437 | LR: 1.00e-04 [2026-04-17 09:39:02] Epoch 1 | Step 2990 | Loss: 0.3330 | LM: 0.3206 | LB: 1.1321 | CL0: 2.9 | CL1: 2.2 | HR0: 0.347/SR0: 0.348 | HR1: 0.457/SR1: 0.437 | LR: 1.00e-04 [2026-04-17 09:39:08] Epoch 1 | Step 3000 | Loss: 0.3331 | LM: 0.3207 | LB: 1.1320 | CL0: 2.9 | CL1: 2.2 | HR0: 0.347/SR0: 0.348 | HR1: 0.457/SR1: 0.436 | LR: 1.00e-04 [2026-04-17 09:39:17] Checkpoint saved: outputs/2026-04-17/08-57-56/checkpoints/checkpoint_step_3000.pt [2026-04-17 09:39:27] Validation | Batch 10/784 | Loss: 0.3377 | LM_LOSS: 0.3266 | LB_LOSS: 1.1065 [2026-04-17 09:39:29] Validation | Batch 20/784 | Loss: 0.3499 | LM_LOSS: 0.3389 | LB_LOSS: 1.1064 [2026-04-17 09:39:30] Validation | Batch 30/784 | Loss: 0.3327 | LM_LOSS: 0.3216 | LB_LOSS: 1.1053 [2026-04-17 09:39:32] Validation | Batch 40/784 | Loss: 0.3333 | LM_LOSS: 0.3222 | LB_LOSS: 1.1051 [2026-04-17 09:39:33] Validation | Batch 50/784 | Loss: 0.3288 | LM_LOSS: 0.3177 | LB_LOSS: 1.1044 [2026-04-17 09:39:35] Validation | Batch 60/784 | Loss: 0.3293 | LM_LOSS: 0.3183 | LB_LOSS: 1.1040 [2026-04-17 09:39:36] Validation | Batch 70/784 | Loss: 0.3262 | LM_LOSS: 0.3152 | LB_LOSS: 1.1033 [2026-04-17 09:39:37] Validation | Batch 80/784 | Loss: 0.3214 | LM_LOSS: 0.3103 | LB_LOSS: 1.1027 [2026-04-17 09:39:38] Validation | Batch 90/784 | Loss: 0.3199 | LM_LOSS: 0.3089 | LB_LOSS: 1.1034 [2026-04-17 09:39:40] Validation | Batch 100/784 | Loss: 0.3217 | LM_LOSS: 0.3107 | LB_LOSS: 1.1040 [2026-04-17 09:39:41] Validation | Batch 110/784 | Loss: 0.3174 | LM_LOSS: 0.3063 | LB_LOSS: 1.1041 [2026-04-17 09:39:43] Validation | Batch 120/784 | Loss: 0.3200 | LM_LOSS: 0.3090 | LB_LOSS: 1.1039 [2026-04-17 09:39:44] Validation | Batch 130/784 | Loss: 0.3230 | LM_LOSS: 0.3119 | LB_LOSS: 1.1039 [2026-04-17 09:39:45] Validation | Batch 140/784 | Loss: 0.3223 | LM_LOSS: 0.3112 | LB_LOSS: 1.1036 [2026-04-17 09:39:47] Validation | Batch 150/784 | Loss: 0.3182 | LM_LOSS: 0.3072 | LB_LOSS: 1.1040 [2026-04-17 09:39:48] Validation | Batch 160/784 | Loss: 0.3187 | LM_LOSS: 0.3077 | LB_LOSS: 1.1037 [2026-04-17 09:39:50] Validation | Batch 170/784 | Loss: 0.3195 | LM_LOSS: 0.3085 | LB_LOSS: 1.1033 [2026-04-17 09:39:51] Validation | Batch 180/784 | Loss: 0.3169 | LM_LOSS: 0.3059 | LB_LOSS: 1.1034 [2026-04-17 09:39:53] Validation | Batch 190/784 | Loss: 0.3184 | LM_LOSS: 0.3074 | LB_LOSS: 1.1038 [2026-04-17 09:39:54] Validation | Batch 200/784 | Loss: 0.3185 | LM_LOSS: 0.3074 | LB_LOSS: 1.1039 [2026-04-17 09:39:55] Validation | Batch 210/784 | Loss: 0.3175 | LM_LOSS: 0.3065 | LB_LOSS: 1.1038 [2026-04-17 09:39:57] Validation | Batch 220/784 | Loss: 0.3184 | LM_LOSS: 0.3074 | LB_LOSS: 1.1038 [2026-04-17 09:39:58] Validation | Batch 230/784 | Loss: 0.3190 | LM_LOSS: 0.3080 | LB_LOSS: 1.1037 [2026-04-17 09:39:59] Validation | Batch 240/784 | Loss: 0.3195 | LM_LOSS: 0.3085 | LB_LOSS: 1.1042 [2026-04-17 09:40:01] Validation | Batch 250/784 | Loss: 0.3194 | LM_LOSS: 0.3084 | LB_LOSS: 1.1040 [2026-04-17 09:40:02] Validation | Batch 260/784 | Loss: 0.3195 | LM_LOSS: 0.3084 | LB_LOSS: 1.1042 [2026-04-17 09:40:04] Validation | Batch 270/784 | Loss: 0.3188 | LM_LOSS: 0.3078 | LB_LOSS: 1.1043 [2026-04-17 09:40:05] Validation | Batch 280/784 | Loss: 0.3194 | LM_LOSS: 0.3084 | LB_LOSS: 1.1045 [2026-04-17 09:40:07] Validation | Batch 290/784 | Loss: 0.3203 | LM_LOSS: 0.3093 | LB_LOSS: 1.1046 [2026-04-17 09:40:08] Validation | Batch 300/784 | Loss: 0.3208 | LM_LOSS: 0.3098 | LB_LOSS: 1.1046 [2026-04-17 09:40:09] Validation | Batch 310/784 | Loss: 0.3203 | LM_LOSS: 0.3093 | LB_LOSS: 1.1046 [2026-04-17 09:40:11] Validation | Batch 320/784 | Loss: 0.3217 | LM_LOSS: 0.3107 | LB_LOSS: 1.1046 [2026-04-17 09:40:12] Validation | Batch 330/784 | Loss: 0.3215 | LM_LOSS: 0.3104 | LB_LOSS: 1.1046 [2026-04-17 09:40:13] Validation | Batch 340/784 | Loss: 0.3205 | LM_LOSS: 0.3095 | LB_LOSS: 1.1047 [2026-04-17 09:40:15] Validation | Batch 350/784 | Loss: 0.3206 | LM_LOSS: 0.3096 | LB_LOSS: 1.1049 [2026-04-17 09:40:16] Validation | Batch 360/784 | Loss: 0.3203 | LM_LOSS: 0.3092 | LB_LOSS: 1.1050 [2026-04-17 09:40:17] Validation | Batch 370/784 | Loss: 0.3206 | LM_LOSS: 0.3095 | LB_LOSS: 1.1049 [2026-04-17 09:40:18] Validation | Batch 380/784 | Loss: 0.3206 | LM_LOSS: 0.3096 | LB_LOSS: 1.1049 [2026-04-17 09:40:20] Validation | Batch 390/784 | Loss: 0.3206 | LM_LOSS: 0.3095 | LB_LOSS: 1.1049 [2026-04-17 09:40:21] Validation | Batch 400/784 | Loss: 0.3207 | LM_LOSS: 0.3096 | LB_LOSS: 1.1050 [2026-04-17 09:40:22] Validation | Batch 410/784 | Loss: 0.3208 | LM_LOSS: 0.3097 | LB_LOSS: 1.1050 [2026-04-17 09:40:24] Validation | Batch 420/784 | Loss: 0.3210 | LM_LOSS: 0.3099 | LB_LOSS: 1.1051 [2026-04-17 09:40:25] Validation | Batch 430/784 | Loss: 0.3209 | LM_LOSS: 0.3099 | LB_LOSS: 1.1050 [2026-04-17 09:40:26] Validation | Batch 440/784 | Loss: 0.3206 | LM_LOSS: 0.3095 | LB_LOSS: 1.1051 [2026-04-17 09:40:28] Validation | Batch 450/784 | Loss: 0.3200 | LM_LOSS: 0.3090 | LB_LOSS: 1.1050 [2026-04-17 09:40:29] Validation | Batch 460/784 | Loss: 0.3204 | LM_LOSS: 0.3094 | LB_LOSS: 1.1051 [2026-04-17 09:40:31] Validation | Batch 470/784 | Loss: 0.3195 | LM_LOSS: 0.3085 | LB_LOSS: 1.1051 [2026-04-17 09:40:32] Validation | Batch 480/784 | Loss: 0.3198 | LM_LOSS: 0.3088 | LB_LOSS: 1.1050 [2026-04-17 09:40:33] Validation | Batch 490/784 | Loss: 0.3193 | LM_LOSS: 0.3082 | LB_LOSS: 1.1049 [2026-04-17 09:40:35] Validation | Batch 500/784 | Loss: 0.3198 | LM_LOSS: 0.3088 | LB_LOSS: 1.1049 [2026-04-17 09:40:36] Validation | Batch 510/784 | Loss: 0.3195 | LM_LOSS: 0.3085 | LB_LOSS: 1.1048 [2026-04-17 09:40:37] Validation | Batch 520/784 | Loss: 0.3196 | LM_LOSS: 0.3085 | LB_LOSS: 1.1048 [2026-04-17 09:40:39] Validation | Batch 530/784 | Loss: 0.3204 | LM_LOSS: 0.3093 | LB_LOSS: 1.1047 [2026-04-17 09:40:40] Validation | Batch 540/784 | Loss: 0.3208 | LM_LOSS: 0.3097 | LB_LOSS: 1.1047 [2026-04-17 09:40:42] Validation | Batch 550/784 | Loss: 0.3222 | LM_LOSS: 0.3112 | LB_LOSS: 1.1047 [2026-04-17 09:40:43] Validation | Batch 560/784 | Loss: 0.3222 | LM_LOSS: 0.3112 | LB_LOSS: 1.1047 [2026-04-17 09:40:45] Validation | Batch 570/784 | Loss: 0.3219 | LM_LOSS: 0.3108 | LB_LOSS: 1.1046 [2026-04-17 09:40:46] Validation | Batch 580/784 | Loss: 0.3213 | LM_LOSS: 0.3102 | LB_LOSS: 1.1047 [2026-04-17 09:40:47] Validation | Batch 590/784 | Loss: 0.3216 | LM_LOSS: 0.3106 | LB_LOSS: 1.1046 [2026-04-17 09:40:49] Validation | Batch 600/784 | Loss: 0.3215 | LM_LOSS: 0.3105 | LB_LOSS: 1.1045 [2026-04-17 09:40:50] Validation | Batch 610/784 | Loss: 0.3216 | LM_LOSS: 0.3105 | LB_LOSS: 1.1045 [2026-04-17 09:40:51] Validation | Batch 620/784 | Loss: 0.3216 | LM_LOSS: 0.3105 | LB_LOSS: 1.1045 [2026-04-17 09:40:53] Validation | Batch 630/784 | Loss: 0.3222 | LM_LOSS: 0.3111 | LB_LOSS: 1.1045 [2026-04-17 09:40:55] Validation | Batch 640/784 | Loss: 0.3223 | LM_LOSS: 0.3113 | LB_LOSS: 1.1045 [2026-04-17 09:40:56] Validation | Batch 650/784 | Loss: 0.3223 | LM_LOSS: 0.3113 | LB_LOSS: 1.1045 [2026-04-17 09:40:57] Validation | Batch 660/784 | Loss: 0.3226 | LM_LOSS: 0.3115 | LB_LOSS: 1.1045 [2026-04-17 09:40:59] Validation | Batch 670/784 | Loss: 0.3230 | LM_LOSS: 0.3119 | LB_LOSS: 1.1046 [2026-04-17 09:41:00] Validation | Batch 680/784 | Loss: 0.3227 | LM_LOSS: 0.3116 | LB_LOSS: 1.1046 [2026-04-17 09:41:02] Validation | Batch 690/784 | Loss: 0.3229 | LM_LOSS: 0.3119 | LB_LOSS: 1.1045 [2026-04-17 09:41:03] Validation | Batch 700/784 | Loss: 0.3229 | LM_LOSS: 0.3119 | LB_LOSS: 1.1045 [2026-04-17 09:41:04] Validation | Batch 710/784 | Loss: 0.3226 | LM_LOSS: 0.3116 | LB_LOSS: 1.1044 [2026-04-17 09:41:06] Validation | Batch 720/784 | Loss: 0.3223 | LM_LOSS: 0.3113 | LB_LOSS: 1.1043 [2026-04-17 09:41:07] Validation | Batch 730/784 | Loss: 0.3218 | LM_LOSS: 0.3107 | LB_LOSS: 1.1043 [2026-04-17 09:41:08] Validation | Batch 740/784 | Loss: 0.3218 | LM_LOSS: 0.3108 | LB_LOSS: 1.1044 [2026-04-17 09:41:09] Validation | Batch 750/784 | Loss: 0.3212 | LM_LOSS: 0.3101 | LB_LOSS: 1.1044 [2026-04-17 09:41:11] Validation | Batch 760/784 | Loss: 0.3213 | LM_LOSS: 0.3102 | LB_LOSS: 1.1044 [2026-04-17 09:41:12] Validation | Batch 770/784 | Loss: 0.3214 | LM_LOSS: 0.3104 | LB_LOSS: 1.1044 [2026-04-17 09:41:14] Validation | Batch 780/784 | Loss: 0.3216 | LM_LOSS: 0.3106 | LB_LOSS: 1.1044 [2026-04-17 09:41:14] Validation | Batch 784/784 | Loss: 0.3219 | LM_LOSS: 0.3108 | LB_LOSS: 1.1044 [2026-04-17 09:41:17] Validation | Loss: 0.3219 | LM_LOSS: 0.3108 | LB_LOSS: 1.1044 | PPL: 1.36 | Time: 107.93s [2026-04-17 09:41:21] New best model saved! Val loss: 0.3219 [2026-04-17 09:41:27] Epoch 1 | Step 3010 | Loss: 0.3330 | LM: 0.3206 | LB: 1.1320 | CL0: 2.9 | CL1: 2.2 | HR0: 0.347/SR0: 0.348 | HR1: 0.457/SR1: 0.436 | LR: 1.00e-04 [2026-04-17 09:41:34] Epoch 1 | Step 3020 | Loss: 0.3329 | LM: 0.3203 | LB: 1.1319 | CL0: 2.9 | CL1: 2.2 | HR0: 0.347/SR0: 0.348 | HR1: 0.457/SR1: 0.436 | LR: 1.00e-04 [2026-04-17 09:41:40] Epoch 1 | Step 3030 | Loss: 0.3328 | LM: 0.3202 | LB: 1.1318 | CL0: 2.9 | CL1: 2.2 | HR0: 0.347/SR0: 0.348 | HR1: 0.457/SR1: 0.436 | LR: 1.00e-04 [2026-04-17 09:41:46] Epoch 1 | Step 3040 | Loss: 0.3327 | LM: 0.3200 | LB: 1.1318 | CL0: 2.9 | CL1: 2.2 | HR0: 0.347/SR0: 0.348 | HR1: 0.457/SR1: 0.436 | LR: 1.00e-04 [2026-04-17 09:41:53] Epoch 1 | Step 3050 | Loss: 0.3325 | LM: 0.3197 | LB: 1.1317 | CL0: 2.9 | CL1: 2.2 | HR0: 0.347/SR0: 0.348 | HR1: 0.457/SR1: 0.436 | LR: 1.00e-04 [2026-04-17 09:41:59] Epoch 1 | Step 3060 | Loss: 0.3324 | LM: 0.3197 | LB: 1.1316 | CL0: 2.9 | CL1: 2.2 | HR0: 0.347/SR0: 0.348 | HR1: 0.457/SR1: 0.436 | LR: 1.00e-04 [2026-04-17 09:42:06] Epoch 1 | Step 3070 | Loss: 0.3322 | LM: 0.3198 | LB: 1.1315 | CL0: 2.9 | CL1: 2.2 | HR0: 0.347/SR0: 0.348 | HR1: 0.456/SR1: 0.436 | LR: 1.00e-04 [2026-04-17 09:42:12] Epoch 1 | Step 3080 | Loss: 0.3322 | LM: 0.3199 | LB: 1.1314 | CL0: 2.9 | CL1: 2.2 | HR0: 0.347/SR0: 0.348 | HR1: 0.456/SR1: 0.436 | LR: 1.00e-04 [2026-04-17 09:42:18] Epoch 1 | Step 3090 | Loss: 0.3321 | LM: 0.3198 | LB: 1.1313 | CL0: 2.9 | CL1: 2.2 | HR0: 0.347/SR0: 0.348 | HR1: 0.456/SR1: 0.436 | LR: 1.00e-04 [2026-04-17 09:42:25] Epoch 1 | Step 3100 | Loss: 0.3321 | LM: 0.3197 | LB: 1.1313 | CL0: 2.9 | CL1: 2.2 | HR0: 0.347/SR0: 0.348 | HR1: 0.456/SR1: 0.436 | LR: 1.00e-04 [2026-04-17 09:42:31] Epoch 1 | Step 3110 | Loss: 0.3320 | LM: 0.3196 | LB: 1.1312 | CL0: 2.9 | CL1: 2.2 | HR0: 0.347/SR0: 0.348 | HR1: 0.456/SR1: 0.436 | LR: 1.00e-04 [2026-04-17 09:42:38] Epoch 1 | Step 3120 | Loss: 0.3318 | LM: 0.3195 | LB: 1.1311 | CL0: 2.9 | CL1: 2.2 | HR0: 0.347/SR0: 0.348 | HR1: 0.456/SR1: 0.436 | LR: 1.00e-04 [2026-04-17 09:42:44] Epoch 1 | Step 3130 | Loss: 0.3318 | LM: 0.3192 | LB: 1.1310 | CL0: 2.9 | CL1: 2.2 | HR0: 0.347/SR0: 0.348 | HR1: 0.456/SR1: 0.435 | LR: 1.00e-04 [2026-04-17 09:42:50] Epoch 1 | Step 3140 | Loss: 0.3317 | LM: 0.3193 | LB: 1.1309 | CL0: 2.9 | CL1: 2.2 | HR0: 0.347/SR0: 0.348 | HR1: 0.456/SR1: 0.435 | LR: 1.00e-04 [2026-04-17 09:42:57] Epoch 1 | Step 3150 | Loss: 0.3316 | LM: 0.3192 | LB: 1.1308 | CL0: 2.9 | CL1: 2.2 | HR0: 0.347/SR0: 0.348 | HR1: 0.456/SR1: 0.435 | LR: 1.00e-04 [2026-04-17 09:43:03] Epoch 1 | Step 3160 | Loss: 0.3316 | LM: 0.3192 | LB: 1.1308 | CL0: 2.9 | CL1: 2.2 | HR0: 0.347/SR0: 0.348 | HR1: 0.456/SR1: 0.435 | LR: 1.00e-04 [2026-04-17 09:43:10] Epoch 1 | Step 3170 | Loss: 0.3315 | LM: 0.3191 | LB: 1.1307 | CL0: 2.9 | CL1: 2.2 | HR0: 0.347/SR0: 0.348 | HR1: 0.456/SR1: 0.435 | LR: 1.00e-04 [2026-04-17 09:43:16] Epoch 1 | Step 3180 | Loss: 0.3313 | LM: 0.3189 | LB: 1.1307 | CL0: 2.9 | CL1: 2.2 | HR0: 0.347/SR0: 0.348 | HR1: 0.456/SR1: 0.435 | LR: 1.00e-04 [2026-04-17 09:43:23] Epoch 1 | Step 3190 | Loss: 0.3311 | LM: 0.3188 | LB: 1.1306 | CL0: 2.9 | CL1: 2.2 | HR0: 0.347/SR0: 0.348 | HR1: 0.456/SR1: 0.435 | LR: 1.00e-04 [2026-04-17 09:43:29] Epoch 1 | Step 3200 | Loss: 0.3311 | LM: 0.3188 | LB: 1.1306 | CL0: 2.9 | CL1: 2.2 | HR0: 0.347/SR0: 0.348 | HR1: 0.456/SR1: 0.435 | LR: 1.00e-04 [2026-04-17 09:43:35] Epoch 1 | Step 3210 | Loss: 0.3311 | LM: 0.3188 | LB: 1.1305 | CL0: 2.9 | CL1: 2.2 | HR0: 0.347/SR0: 0.348 | HR1: 0.456/SR1: 0.435 | LR: 1.00e-04 [2026-04-17 09:43:42] Epoch 1 | Step 3220 | Loss: 0.3310 | LM: 0.3185 | LB: 1.1305 | CL0: 2.9 | CL1: 2.2 | HR0: 0.347/SR0: 0.348 | HR1: 0.456/SR1: 0.435 | LR: 1.00e-04 [2026-04-17 09:43:48] Epoch 1 | Step 3230 | Loss: 0.3309 | LM: 0.3186 | LB: 1.1304 | CL0: 2.9 | CL1: 2.2 | HR0: 0.347/SR0: 0.348 | HR1: 0.455/SR1: 0.435 | LR: 1.00e-04 [2026-04-17 09:43:55] Epoch 1 | Step 3240 | Loss: 0.3308 | LM: 0.3186 | LB: 1.1303 | CL0: 2.9 | CL1: 2.2 | HR0: 0.347/SR0: 0.348 | HR1: 0.455/SR1: 0.435 | LR: 1.00e-04 [2026-04-17 09:44:01] Epoch 1 | Step 3250 | Loss: 0.3307 | LM: 0.3184 | LB: 1.1302 | CL0: 2.9 | CL1: 2.2 | HR0: 0.347/SR0: 0.348 | HR1: 0.455/SR1: 0.435 | LR: 1.00e-04 [2026-04-17 09:44:07] Epoch 1 | Step 3260 | Loss: 0.3306 | LM: 0.3181 | LB: 1.1301 | CL0: 2.9 | CL1: 2.2 | HR0: 0.347/SR0: 0.348 | HR1: 0.455/SR1: 0.435 | LR: 1.00e-04 [2026-04-17 09:44:14] Epoch 1 | Step 3270 | Loss: 0.3306 | LM: 0.3180 | LB: 1.1301 | CL0: 2.9 | CL1: 2.2 | HR0: 0.347/SR0: 0.348 | HR1: 0.455/SR1: 0.435 | LR: 1.00e-04 [2026-04-17 09:44:20] Epoch 1 | Step 3280 | Loss: 0.3304 | LM: 0.3178 | LB: 1.1300 | CL0: 2.9 | CL1: 2.2 | HR0: 0.347/SR0: 0.348 | HR1: 0.455/SR1: 0.435 | LR: 1.00e-04 [2026-04-17 09:44:27] Epoch 1 | Step 3290 | Loss: 0.3303 | LM: 0.3176 | LB: 1.1299 | CL0: 2.9 | CL1: 2.2 | HR0: 0.347/SR0: 0.348 | HR1: 0.455/SR1: 0.434 | LR: 1.00e-04 [2026-04-17 09:44:33] Epoch 1 | Step 3300 | Loss: 0.3303 | LM: 0.3176 | LB: 1.1299 | CL0: 2.9 | CL1: 2.2 | HR0: 0.347/SR0: 0.348 | HR1: 0.455/SR1: 0.434 | LR: 1.00e-04 [2026-04-17 09:44:39] Epoch 1 | Step 3310 | Loss: 0.3303 | LM: 0.3174 | LB: 1.1299 | CL0: 2.9 | CL1: 2.2 | HR0: 0.347/SR0: 0.348 | HR1: 0.455/SR1: 0.434 | LR: 1.00e-04 [2026-04-17 09:44:46] Epoch 1 | Step 3320 | Loss: 0.3301 | LM: 0.3173 | LB: 1.1298 | CL0: 2.9 | CL1: 2.2 | HR0: 0.347/SR0: 0.348 | HR1: 0.455/SR1: 0.434 | LR: 1.00e-04 [2026-04-17 09:44:52] Epoch 1 | Step 3330 | Loss: 0.3301 | LM: 0.3175 | LB: 1.1297 | CL0: 2.9 | CL1: 2.2 | HR0: 0.347/SR0: 0.348 | HR1: 0.455/SR1: 0.434 | LR: 1.00e-04 [2026-04-17 09:44:58] Epoch 1 | Step 3340 | Loss: 0.3301 | LM: 0.3174 | LB: 1.1297 | CL0: 2.9 | CL1: 2.2 | HR0: 0.347/SR0: 0.348 | HR1: 0.455/SR1: 0.434 | LR: 1.00e-04 [2026-04-17 09:45:05] Epoch 1 | Step 3350 | Loss: 0.3300 | LM: 0.3177 | LB: 1.1296 | CL0: 2.9 | CL1: 2.2 | HR0: 0.347/SR0: 0.348 | HR1: 0.455/SR1: 0.434 | LR: 1.00e-04 [2026-04-17 09:45:11] Epoch 1 | Step 3360 | Loss: 0.3299 | LM: 0.3175 | LB: 1.1295 | CL0: 2.9 | CL1: 2.2 | HR0: 0.347/SR0: 0.348 | HR1: 0.455/SR1: 0.434 | LR: 1.00e-04 [2026-04-17 09:45:18] Epoch 1 | Step 3370 | Loss: 0.3299 | LM: 0.3175 | LB: 1.1295 | CL0: 2.9 | CL1: 2.2 | HR0: 0.347/SR0: 0.348 | HR1: 0.455/SR1: 0.434 | LR: 1.00e-04 [2026-04-17 09:45:24] Epoch 1 | Step 3380 | Loss: 0.3298 | LM: 0.3172 | LB: 1.1294 | CL0: 2.9 | CL1: 2.2 | HR0: 0.347/SR0: 0.348 | HR1: 0.454/SR1: 0.434 | LR: 1.00e-04 [2026-04-17 09:45:30] Epoch 1 | Step 3390 | Loss: 0.3298 | LM: 0.3175 | LB: 1.1294 | CL0: 2.9 | CL1: 2.2 | HR0: 0.347/SR0: 0.348 | HR1: 0.454/SR1: 0.434 | LR: 1.00e-04 [2026-04-17 09:45:37] Epoch 1 | Step 3400 | Loss: 0.3297 | LM: 0.3172 | LB: 1.1293 | CL0: 2.9 | CL1: 2.2 | HR0: 0.347/SR0: 0.348 | HR1: 0.454/SR1: 0.434 | LR: 1.00e-04 [2026-04-17 09:45:43] Epoch 1 | Step 3410 | Loss: 0.3294 | LM: 0.3168 | LB: 1.1293 | CL0: 2.9 | CL1: 2.2 | HR0: 0.347/SR0: 0.348 | HR1: 0.454/SR1: 0.434 | LR: 1.00e-04 [2026-04-17 09:45:49] Epoch 1 | Step 3420 | Loss: 0.3295 | LM: 0.3168 | LB: 1.1292 | CL0: 2.9 | CL1: 2.2 | HR0: 0.347/SR0: 0.348 | HR1: 0.454/SR1: 0.434 | LR: 1.00e-04 [2026-04-17 09:45:56] Epoch 1 | Step 3430 | Loss: 0.3294 | LM: 0.3166 | LB: 1.1291 | CL0: 2.9 | CL1: 2.2 | HR0: 0.347/SR0: 0.348 | HR1: 0.454/SR1: 0.434 | LR: 1.00e-04 [2026-04-17 09:46:03] Epoch 1 | Step 3440 | Loss: 0.3294 | LM: 0.3167 | LB: 1.1291 | CL0: 2.9 | CL1: 2.2 | HR0: 0.347/SR0: 0.348 | HR1: 0.454/SR1: 0.433 | LR: 1.00e-04 [2026-04-17 09:46:09] Epoch 1 | Step 3450 | Loss: 0.3294 | LM: 0.3167 | LB: 1.1290 | CL0: 2.9 | CL1: 2.2 | HR0: 0.347/SR0: 0.348 | HR1: 0.454/SR1: 0.433 | LR: 1.00e-04 [2026-04-17 09:46:15] Epoch 1 | Step 3460 | Loss: 0.3295 | LM: 0.3168 | LB: 1.1290 | CL0: 2.9 | CL1: 2.2 | HR0: 0.347/SR0: 0.348 | HR1: 0.454/SR1: 0.433 | LR: 1.00e-04 [2026-04-17 09:46:22] Epoch 1 | Step 3470 | Loss: 0.3294 | LM: 0.3166 | LB: 1.1289 | CL0: 2.9 | CL1: 2.2 | HR0: 0.347/SR0: 0.348 | HR1: 0.454/SR1: 0.433 | LR: 1.00e-04 [2026-04-17 09:46:28] Epoch 1 | Step 3480 | Loss: 0.3294 | LM: 0.3164 | LB: 1.1288 | CL0: 2.9 | CL1: 2.2 | HR0: 0.347/SR0: 0.348 | HR1: 0.454/SR1: 0.433 | LR: 1.00e-04 [2026-04-17 09:46:34] Epoch 1 | Step 3490 | Loss: 0.3293 | LM: 0.3163 | LB: 1.1287 | CL0: 2.9 | CL1: 2.2 | HR0: 0.347/SR0: 0.348 | HR1: 0.454/SR1: 0.433 | LR: 1.00e-04 [2026-04-17 09:46:41] Epoch 1 | Step 3500 | Loss: 0.3294 | LM: 0.3165 | LB: 1.1287 | CL0: 2.9 | CL1: 2.2 | HR0: 0.347/SR0: 0.348 | HR1: 0.454/SR1: 0.433 | LR: 1.00e-04 [2026-04-17 09:46:47] Epoch 1 | Step 3510 | Loss: 0.3293 | LM: 0.3165 | LB: 1.1287 | CL0: 2.9 | CL1: 2.2 | HR0: 0.347/SR0: 0.348 | HR1: 0.454/SR1: 0.433 | LR: 1.00e-04 [2026-04-17 09:46:54] Epoch 1 | Step 3520 | Loss: 0.3294 | LM: 0.3165 | LB: 1.1286 | CL0: 2.9 | CL1: 2.2 | HR0: 0.347/SR0: 0.348 | HR1: 0.454/SR1: 0.433 | LR: 1.00e-04 [2026-04-17 09:47:00] Epoch 1 | Step 3530 | Loss: 0.3292 | LM: 0.3164 | LB: 1.1285 | CL0: 2.9 | CL1: 2.2 | HR0: 0.347/SR0: 0.348 | HR1: 0.454/SR1: 0.433 | LR: 1.00e-04 [2026-04-17 09:47:07] Epoch 1 | Step 3540 | Loss: 0.3292 | LM: 0.3163 | LB: 1.1285 | CL0: 2.9 | CL1: 2.2 | HR0: 0.347/SR0: 0.348 | HR1: 0.453/SR1: 0.433 | LR: 1.00e-04 [2026-04-17 09:47:13] Epoch 1 | Step 3550 | Loss: 0.3292 | LM: 0.3163 | LB: 1.1284 | CL0: 2.9 | CL1: 2.2 | HR0: 0.347/SR0: 0.348 | HR1: 0.453/SR1: 0.433 | LR: 1.00e-04 [2026-04-17 09:47:19] Epoch 1 | Step 3560 | Loss: 0.3292 | LM: 0.3164 | LB: 1.1283 | CL0: 2.9 | CL1: 2.2 | HR0: 0.347/SR0: 0.348 | HR1: 0.453/SR1: 0.433 | LR: 1.00e-04 [2026-04-17 09:47:26] Epoch 1 | Step 3570 | Loss: 0.3292 | LM: 0.3163 | LB: 1.1283 | CL0: 2.9 | CL1: 2.2 | HR0: 0.347/SR0: 0.348 | HR1: 0.453/SR1: 0.433 | LR: 1.00e-04 [2026-04-17 09:47:32] Epoch 1 | Step 3580 | Loss: 0.3292 | LM: 0.3162 | LB: 1.1282 | CL0: 2.9 | CL1: 2.2 | HR0: 0.347/SR0: 0.348 | HR1: 0.453/SR1: 0.433 | LR: 1.00e-04 [2026-04-17 09:47:39] Epoch 1 | Step 3590 | Loss: 0.3291 | LM: 0.3165 | LB: 1.1281 | CL0: 2.9 | CL1: 2.2 | HR0: 0.347/SR0: 0.348 | HR1: 0.453/SR1: 0.432 | LR: 1.00e-04 [2026-04-17 09:47:45] Epoch 1 | Step 3600 | Loss: 0.3291 | LM: 0.3166 | LB: 1.1280 | CL0: 2.9 | CL1: 2.2 | HR0: 0.347/SR0: 0.348 | HR1: 0.453/SR1: 0.432 | LR: 1.00e-04 [2026-04-17 09:47:51] Epoch 1 | Step 3610 | Loss: 0.3290 | LM: 0.3166 | LB: 1.1279 | CL0: 2.9 | CL1: 2.2 | HR0: 0.347/SR0: 0.348 | HR1: 0.453/SR1: 0.432 | LR: 1.00e-04 [2026-04-17 09:47:58] Epoch 1 | Step 3620 | Loss: 0.3289 | LM: 0.3166 | LB: 1.1279 | CL0: 2.9 | CL1: 2.2 | HR0: 0.347/SR0: 0.348 | HR1: 0.453/SR1: 0.432 | LR: 1.00e-04 [2026-04-17 09:48:04] Epoch 1 | Step 3630 | Loss: 0.3289 | LM: 0.3166 | LB: 1.1278 | CL0: 2.9 | CL1: 2.2 | HR0: 0.347/SR0: 0.348 | HR1: 0.453/SR1: 0.432 | LR: 1.00e-04 [2026-04-17 09:48:10] Epoch 1 | Step 3640 | Loss: 0.3288 | LM: 0.3165 | LB: 1.1277 | CL0: 2.9 | CL1: 2.2 | HR0: 0.347/SR0: 0.348 | HR1: 0.453/SR1: 0.432 | LR: 1.00e-04 [2026-04-17 09:48:17] Epoch 1 | Step 3650 | Loss: 0.3289 | LM: 0.3165 | LB: 1.1277 | CL0: 2.9 | CL1: 2.2 | HR0: 0.347/SR0: 0.348 | HR1: 0.453/SR1: 0.432 | LR: 1.00e-04 [2026-04-17 09:48:23] Epoch 1 | Step 3660 | Loss: 0.3288 | LM: 0.3165 | LB: 1.1276 | CL0: 2.9 | CL1: 2.2 | HR0: 0.347/SR0: 0.348 | HR1: 0.453/SR1: 0.432 | LR: 1.00e-04 [2026-04-17 09:48:30] Epoch 1 | Step 3670 | Loss: 0.3288 | LM: 0.3165 | LB: 1.1276 | CL0: 2.9 | CL1: 2.2 | HR0: 0.347/SR0: 0.348 | HR1: 0.453/SR1: 0.432 | LR: 1.00e-04 [2026-04-17 09:48:36] Epoch 1 | Step 3680 | Loss: 0.3287 | LM: 0.3164 | LB: 1.1275 | CL0: 2.9 | CL1: 2.2 | HR0: 0.347/SR0: 0.348 | HR1: 0.453/SR1: 0.432 | LR: 1.00e-04 [2026-04-17 09:48:42] Epoch 1 | Step 3690 | Loss: 0.3288 | LM: 0.3163 | LB: 1.1274 | CL0: 2.9 | CL1: 2.2 | HR0: 0.347/SR0: 0.348 | HR1: 0.453/SR1: 0.432 | LR: 1.00e-04 [2026-04-17 09:48:49] Epoch 1 | Step 3700 | Loss: 0.3288 | LM: 0.3164 | LB: 1.1274 | CL0: 2.9 | CL1: 2.2 | HR0: 0.347/SR0: 0.348 | HR1: 0.453/SR1: 0.432 | LR: 1.00e-04 [2026-04-17 09:48:55] Epoch 1 | Step 3710 | Loss: 0.3288 | LM: 0.3160 | LB: 1.1273 | CL0: 2.9 | CL1: 2.2 | HR0: 0.347/SR0: 0.348 | HR1: 0.452/SR1: 0.432 | LR: 1.00e-04 [2026-04-17 09:49:01] Epoch 1 | Step 3720 | Loss: 0.3288 | LM: 0.3160 | LB: 1.1273 | CL0: 2.9 | CL1: 2.2 | HR0: 0.347/SR0: 0.348 | HR1: 0.452/SR1: 0.432 | LR: 1.00e-04 [2026-04-17 09:49:08] Epoch 1 | Step 3730 | Loss: 0.3288 | LM: 0.3159 | LB: 1.1272 | CL0: 2.9 | CL1: 2.2 | HR0: 0.347/SR0: 0.348 | HR1: 0.452/SR1: 0.432 | LR: 1.00e-04 [2026-04-17 09:49:14] Epoch 1 | Step 3740 | Loss: 0.3286 | LM: 0.3157 | LB: 1.1271 | CL0: 2.9 | CL1: 2.2 | HR0: 0.347/SR0: 0.348 | HR1: 0.452/SR1: 0.432 | LR: 1.00e-04 [2026-04-17 09:49:20] Epoch 1 | Step 3750 | Loss: 0.3286 | LM: 0.3157 | LB: 1.1271 | CL0: 2.9 | CL1: 2.2 | HR0: 0.347/SR0: 0.348 | HR1: 0.452/SR1: 0.431 | LR: 1.00e-04 [2026-04-17 09:49:26] Epoch 1 | Step 3760 | Loss: 0.3286 | LM: 0.3156 | LB: 1.1270 | CL0: 2.9 | CL1: 2.2 | HR0: 0.347/SR0: 0.348 | HR1: 0.452/SR1: 0.431 | LR: 1.00e-04 [2026-04-17 09:49:33] Epoch 1 | Step 3770 | Loss: 0.3286 | LM: 0.3158 | LB: 1.1270 | CL0: 2.9 | CL1: 2.2 | HR0: 0.347/SR0: 0.348 | HR1: 0.452/SR1: 0.431 | LR: 1.00e-04 [2026-04-17 09:49:40] Epoch 1 | Step 3780 | Loss: 0.3286 | LM: 0.3158 | LB: 1.1269 | CL0: 2.9 | CL1: 2.2 | HR0: 0.347/SR0: 0.348 | HR1: 0.452/SR1: 0.431 | LR: 1.00e-04 [2026-04-17 09:49:47] Epoch 1 | Step 3790 | Loss: 0.3285 | LM: 0.3157 | LB: 1.1268 | CL0: 2.9 | CL1: 2.2 | HR0: 0.347/SR0: 0.348 | HR1: 0.452/SR1: 0.431 | LR: 1.00e-04 [2026-04-17 09:49:54] Epoch 1 | Step 3800 | Loss: 0.3285 | LM: 0.3156 | LB: 1.1267 | CL0: 2.9 | CL1: 2.2 | HR0: 0.347/SR0: 0.348 | HR1: 0.452/SR1: 0.431 | LR: 1.00e-04 [2026-04-17 09:50:01] Epoch 1 | Step 3810 | Loss: 0.3284 | LM: 0.3154 | LB: 1.1267 | CL0: 2.9 | CL1: 2.2 | HR0: 0.347/SR0: 0.348 | HR1: 0.452/SR1: 0.431 | LR: 1.00e-04 [2026-04-17 09:50:08] Epoch 1 | Step 3820 | Loss: 0.3284 | LM: 0.3156 | LB: 1.1266 | CL0: 2.9 | CL1: 2.2 | HR0: 0.347/SR0: 0.348 | HR1: 0.452/SR1: 0.431 | LR: 1.00e-04 [2026-04-17 09:50:15] Epoch 1 | Step 3830 | Loss: 0.3284 | LM: 0.3157 | LB: 1.1265 | CL0: 2.9 | CL1: 2.2 | HR0: 0.347/SR0: 0.348 | HR1: 0.452/SR1: 0.431 | LR: 1.00e-04 [2026-04-17 09:50:22] Epoch 1 | Step 3840 | Loss: 0.3285 | LM: 0.3157 | LB: 1.1264 | CL0: 2.9 | CL1: 2.2 | HR0: 0.347/SR0: 0.348 | HR1: 0.452/SR1: 0.431 | LR: 1.00e-04 [2026-04-17 09:50:29] Epoch 1 | Step 3850 | Loss: 0.3285 | LM: 0.3156 | LB: 1.1264 | CL0: 2.9 | CL1: 2.2 | HR0: 0.347/SR0: 0.348 | HR1: 0.452/SR1: 0.431 | LR: 1.00e-04 [2026-04-17 09:50:37] Epoch 1 | Step 3860 | Loss: 0.3285 | LM: 0.3158 | LB: 1.1263 | CL0: 2.9 | CL1: 2.2 | HR0: 0.347/SR0: 0.348 | HR1: 0.451/SR1: 0.431 | LR: 1.00e-04 [2026-04-17 09:50:44] Epoch 1 | Step 3870 | Loss: 0.3286 | LM: 0.3158 | LB: 1.1262 | CL0: 2.9 | CL1: 2.2 | HR0: 0.347/SR0: 0.348 | HR1: 0.451/SR1: 0.431 | LR: 1.00e-04 [2026-04-17 09:50:51] Epoch 1 | Step 3880 | Loss: 0.3286 | LM: 0.3160 | LB: 1.1262 | CL0: 2.9 | CL1: 2.2 | HR0: 0.347/SR0: 0.348 | HR1: 0.451/SR1: 0.430 | LR: 1.00e-04 [2026-04-17 09:50:59] Epoch 1 | Step 3890 | Loss: 0.3285 | LM: 0.3159 | LB: 1.1261 | CL0: 2.9 | CL1: 2.2 | HR0: 0.347/SR0: 0.348 | HR1: 0.451/SR1: 0.430 | LR: 1.00e-04 [2026-04-17 09:51:06] Epoch 1 | Step 3900 | Loss: 0.3284 | LM: 0.3159 | LB: 1.1260 | CL0: 2.9 | CL1: 2.2 | HR0: 0.347/SR0: 0.348 | HR1: 0.451/SR1: 0.430 | LR: 1.00e-04 [2026-04-17 09:51:13] Epoch 1 | Step 3910 | Loss: 0.3284 | LM: 0.3159 | LB: 1.1260 | CL0: 2.9 | CL1: 2.2 | HR0: 0.347/SR0: 0.348 | HR1: 0.451/SR1: 0.430 | LR: 1.00e-04 [2026-04-17 09:51:21] Epoch 1 | Step 3920 | Loss: 0.3283 | LM: 0.3159 | LB: 1.1259 | CL0: 2.9 | CL1: 2.2 | HR0: 0.347/SR0: 0.348 | HR1: 0.451/SR1: 0.430 | LR: 1.00e-04 [2026-04-17 09:51:28] Epoch 1 | Step 3930 | Loss: 0.3284 | LM: 0.3159 | LB: 1.1259 | CL0: 2.9 | CL1: 2.2 | HR0: 0.347/SR0: 0.348 | HR1: 0.451/SR1: 0.430 | LR: 1.00e-04 [2026-04-17 09:51:35] Epoch 1 | Step 3940 | Loss: 0.3282 | LM: 0.3157 | LB: 1.1258 | CL0: 2.9 | CL1: 2.2 | HR0: 0.347/SR0: 0.348 | HR1: 0.451/SR1: 0.430 | LR: 1.00e-04 [2026-04-17 09:51:43] Epoch 1 | Step 3950 | Loss: 0.3282 | LM: 0.3156 | LB: 1.1257 | CL0: 2.9 | CL1: 2.2 | HR0: 0.347/SR0: 0.348 | HR1: 0.451/SR1: 0.430 | LR: 1.00e-04 [2026-04-17 09:51:50] Epoch 1 | Step 3960 | Loss: 0.3282 | LM: 0.3154 | LB: 1.1257 | CL0: 2.9 | CL1: 2.2 | HR0: 0.347/SR0: 0.348 | HR1: 0.451/SR1: 0.430 | LR: 1.00e-04 [2026-04-17 09:51:58] Epoch 1 | Step 3970 | Loss: 0.3281 | LM: 0.3153 | LB: 1.1256 | CL0: 2.9 | CL1: 2.2 | HR0: 0.347/SR0: 0.348 | HR1: 0.451/SR1: 0.430 | LR: 1.00e-04 [2026-04-17 09:52:05] Epoch 1 | Step 3980 | Loss: 0.3281 | LM: 0.3156 | LB: 1.1256 | CL0: 2.9 | CL1: 2.2 | HR0: 0.347/SR0: 0.348 | HR1: 0.451/SR1: 0.430 | LR: 1.00e-04 [2026-04-17 09:52:13] Epoch 1 | Step 3990 | Loss: 0.3282 | LM: 0.3157 | LB: 1.1255 | CL0: 2.9 | CL1: 2.2 | HR0: 0.347/SR0: 0.348 | HR1: 0.451/SR1: 0.430 | LR: 1.00e-04 [2026-04-17 09:52:20] Epoch 1 | Step 4000 | Loss: 0.3282 | LM: 0.3156 | LB: 1.1254 | CL0: 2.9 | CL1: 2.2 | HR0: 0.347/SR0: 0.348 | HR1: 0.451/SR1: 0.430 | LR: 1.00e-04 [2026-04-17 09:52:21] Validation | Batch 10/784 | Loss: 0.3339 | LM_LOSS: 0.3229 | LB_LOSS: 1.0990 [2026-04-17 09:52:23] Validation | Batch 20/784 | Loss: 0.3456 | LM_LOSS: 0.3346 | LB_LOSS: 1.0991 [2026-04-17 09:52:24] Validation | Batch 30/784 | Loss: 0.3298 | LM_LOSS: 0.3188 | LB_LOSS: 1.0981 [2026-04-17 09:52:26] Validation | Batch 40/784 | Loss: 0.3302 | LM_LOSS: 0.3192 | LB_LOSS: 1.0979 [2026-04-17 09:52:27] Validation | Batch 50/784 | Loss: 0.3257 | LM_LOSS: 0.3147 | LB_LOSS: 1.0971 [2026-04-17 09:52:29] Validation | Batch 60/784 | Loss: 0.3264 | LM_LOSS: 0.3154 | LB_LOSS: 1.0967 [2026-04-17 09:52:30] Validation | Batch 70/784 | Loss: 0.3236 | LM_LOSS: 0.3127 | LB_LOSS: 1.0960 [2026-04-17 09:52:31] Validation | Batch 80/784 | Loss: 0.3191 | LM_LOSS: 0.3082 | LB_LOSS: 1.0954 [2026-04-17 09:52:33] Validation | Batch 90/784 | Loss: 0.3181 | LM_LOSS: 0.3071 | LB_LOSS: 1.0960 [2026-04-17 09:52:34] Validation | Batch 100/784 | Loss: 0.3197 | LM_LOSS: 0.3087 | LB_LOSS: 1.0965 [2026-04-17 09:52:35] Validation | Batch 110/784 | Loss: 0.3154 | LM_LOSS: 0.3044 | LB_LOSS: 1.0966 [2026-04-17 09:52:37] Validation | Batch 120/784 | Loss: 0.3182 | LM_LOSS: 0.3073 | LB_LOSS: 1.0965 [2026-04-17 09:52:38] Validation | Batch 130/784 | Loss: 0.3206 | LM_LOSS: 0.3097 | LB_LOSS: 1.0965 [2026-04-17 09:52:40] Validation | Batch 140/784 | Loss: 0.3199 | LM_LOSS: 0.3090 | LB_LOSS: 1.0962 [2026-04-17 09:52:41] Validation | Batch 150/784 | Loss: 0.3158 | LM_LOSS: 0.3048 | LB_LOSS: 1.0966 [2026-04-17 09:52:43] Validation | Batch 160/784 | Loss: 0.3163 | LM_LOSS: 0.3053 | LB_LOSS: 1.0963 [2026-04-17 09:52:44] Validation | Batch 170/784 | Loss: 0.3169 | LM_LOSS: 0.3059 | LB_LOSS: 1.0959 [2026-04-17 09:52:45] Validation | Batch 180/784 | Loss: 0.3144 | LM_LOSS: 0.3034 | LB_LOSS: 1.0960 [2026-04-17 09:52:47] Validation | Batch 190/784 | Loss: 0.3159 | LM_LOSS: 0.3049 | LB_LOSS: 1.0964 [2026-04-17 09:52:48] Validation | Batch 200/784 | Loss: 0.3159 | LM_LOSS: 0.3049 | LB_LOSS: 1.0965 [2026-04-17 09:52:50] Validation | Batch 210/784 | Loss: 0.3150 | LM_LOSS: 0.3040 | LB_LOSS: 1.0964 [2026-04-17 09:52:51] Validation | Batch 220/784 | Loss: 0.3157 | LM_LOSS: 0.3047 | LB_LOSS: 1.0964 [2026-04-17 09:52:53] Validation | Batch 230/784 | Loss: 0.3162 | LM_LOSS: 0.3052 | LB_LOSS: 1.0963 [2026-04-17 09:52:54] Validation | Batch 240/784 | Loss: 0.3167 | LM_LOSS: 0.3058 | LB_LOSS: 1.0967 [2026-04-17 09:52:55] Validation | Batch 250/784 | Loss: 0.3166 | LM_LOSS: 0.3057 | LB_LOSS: 1.0965 [2026-04-17 09:52:57] Validation | Batch 260/784 | Loss: 0.3167 | LM_LOSS: 0.3058 | LB_LOSS: 1.0968 [2026-04-17 09:52:58] Validation | Batch 270/784 | Loss: 0.3162 | LM_LOSS: 0.3052 | LB_LOSS: 1.0968 [2026-04-17 09:53:00] Validation | Batch 280/784 | Loss: 0.3169 | LM_LOSS: 0.3059 | LB_LOSS: 1.0970 [2026-04-17 09:53:01] Validation | Batch 290/784 | Loss: 0.3178 | LM_LOSS: 0.3068 | LB_LOSS: 1.0971 [2026-04-17 09:53:02] Validation | Batch 300/784 | Loss: 0.3183 | LM_LOSS: 0.3074 | LB_LOSS: 1.0971 [2026-04-17 09:53:04] Validation | Batch 310/784 | Loss: 0.3178 | LM_LOSS: 0.3068 | LB_LOSS: 1.0971 [2026-04-17 09:53:05] Validation | Batch 320/784 | Loss: 0.3192 | LM_LOSS: 0.3082 | LB_LOSS: 1.0971 [2026-04-17 09:53:07] Validation | Batch 330/784 | Loss: 0.3190 | LM_LOSS: 0.3080 | LB_LOSS: 1.0971 [2026-04-17 09:53:08] Validation | Batch 340/784 | Loss: 0.3180 | LM_LOSS: 0.3070 | LB_LOSS: 1.0972 [2026-04-17 09:53:09] Validation | Batch 350/784 | Loss: 0.3182 | LM_LOSS: 0.3072 | LB_LOSS: 1.0974 [2026-04-17 09:53:11] Validation | Batch 360/784 | Loss: 0.3178 | LM_LOSS: 0.3069 | LB_LOSS: 1.0974 [2026-04-17 09:53:12] Validation | Batch 370/784 | Loss: 0.3182 | LM_LOSS: 0.3072 | LB_LOSS: 1.0973 [2026-04-17 09:53:13] Validation | Batch 380/784 | Loss: 0.3180 | LM_LOSS: 0.3070 | LB_LOSS: 1.0974 [2026-04-17 09:53:15] Validation | Batch 390/784 | Loss: 0.3179 | LM_LOSS: 0.3069 | LB_LOSS: 1.0974 [2026-04-17 09:53:16] Validation | Batch 400/784 | Loss: 0.3180 | LM_LOSS: 0.3070 | LB_LOSS: 1.0974 [2026-04-17 09:53:17] Validation | Batch 410/784 | Loss: 0.3181 | LM_LOSS: 0.3072 | LB_LOSS: 1.0975 [2026-04-17 09:53:18] Validation | Batch 420/784 | Loss: 0.3182 | LM_LOSS: 0.3072 | LB_LOSS: 1.0975 [2026-04-17 09:53:20] Validation | Batch 430/784 | Loss: 0.3182 | LM_LOSS: 0.3072 | LB_LOSS: 1.0974 [2026-04-17 09:53:21] Validation | Batch 440/784 | Loss: 0.3177 | LM_LOSS: 0.3068 | LB_LOSS: 1.0975 [2026-04-17 09:53:22] Validation | Batch 450/784 | Loss: 0.3172 | LM_LOSS: 0.3062 | LB_LOSS: 1.0975 [2026-04-17 09:53:24] Validation | Batch 460/784 | Loss: 0.3177 | LM_LOSS: 0.3067 | LB_LOSS: 1.0976 [2026-04-17 09:53:25] Validation | Batch 470/784 | Loss: 0.3168 | LM_LOSS: 0.3059 | LB_LOSS: 1.0975 [2026-04-17 09:53:26] Validation | Batch 480/784 | Loss: 0.3172 | LM_LOSS: 0.3062 | LB_LOSS: 1.0975 [2026-04-17 09:53:28] Validation | Batch 490/784 | Loss: 0.3166 | LM_LOSS: 0.3056 | LB_LOSS: 1.0974 [2026-04-17 09:53:29] Validation | Batch 500/784 | Loss: 0.3171 | LM_LOSS: 0.3062 | LB_LOSS: 1.0973 [2026-04-17 09:53:31] Validation | Batch 510/784 | Loss: 0.3168 | LM_LOSS: 0.3059 | LB_LOSS: 1.0973 [2026-04-17 09:53:32] Validation | Batch 520/784 | Loss: 0.3169 | LM_LOSS: 0.3059 | LB_LOSS: 1.0972 [2026-04-17 09:53:33] Validation | Batch 530/784 | Loss: 0.3177 | LM_LOSS: 0.3067 | LB_LOSS: 1.0972 [2026-04-17 09:53:35] Validation | Batch 540/784 | Loss: 0.3181 | LM_LOSS: 0.3072 | LB_LOSS: 1.0972 [2026-04-17 09:53:36] Validation | Batch 550/784 | Loss: 0.3195 | LM_LOSS: 0.3086 | LB_LOSS: 1.0972 [2026-04-17 09:53:38] Validation | Batch 560/784 | Loss: 0.3196 | LM_LOSS: 0.3086 | LB_LOSS: 1.0972 [2026-04-17 09:53:39] Validation | Batch 570/784 | Loss: 0.3191 | LM_LOSS: 0.3082 | LB_LOSS: 1.0971 [2026-04-17 09:53:40] Validation | Batch 580/784 | Loss: 0.3186 | LM_LOSS: 0.3076 | LB_LOSS: 1.0972 [2026-04-17 09:53:42] Validation | Batch 590/784 | Loss: 0.3189 | LM_LOSS: 0.3080 | LB_LOSS: 1.0971 [2026-04-17 09:53:43] Validation | Batch 600/784 | Loss: 0.3188 | LM_LOSS: 0.3078 | LB_LOSS: 1.0970 [2026-04-17 09:53:45] Validation | Batch 610/784 | Loss: 0.3189 | LM_LOSS: 0.3079 | LB_LOSS: 1.0970 [2026-04-17 09:53:46] Validation | Batch 620/784 | Loss: 0.3189 | LM_LOSS: 0.3079 | LB_LOSS: 1.0970 [2026-04-17 09:53:47] Validation | Batch 630/784 | Loss: 0.3195 | LM_LOSS: 0.3085 | LB_LOSS: 1.0970 [2026-04-17 09:53:49] Validation | Batch 640/784 | Loss: 0.3196 | LM_LOSS: 0.3087 | LB_LOSS: 1.0970 [2026-04-17 09:53:51] Validation | Batch 650/784 | Loss: 0.3196 | LM_LOSS: 0.3086 | LB_LOSS: 1.0971 [2026-04-17 09:53:52] Validation | Batch 660/784 | Loss: 0.3199 | LM_LOSS: 0.3089 | LB_LOSS: 1.0970 [2026-04-17 09:53:54] Validation | Batch 670/784 | Loss: 0.3203 | LM_LOSS: 0.3093 | LB_LOSS: 1.0971 [2026-04-17 09:53:55] Validation | Batch 680/784 | Loss: 0.3201 | LM_LOSS: 0.3091 | LB_LOSS: 1.0971 [2026-04-17 09:53:56] Validation | Batch 690/784 | Loss: 0.3203 | LM_LOSS: 0.3093 | LB_LOSS: 1.0970 [2026-04-17 09:53:58] Validation | Batch 700/784 | Loss: 0.3203 | LM_LOSS: 0.3093 | LB_LOSS: 1.0970 [2026-04-17 09:53:59] Validation | Batch 710/784 | Loss: 0.3200 | LM_LOSS: 0.3090 | LB_LOSS: 1.0969 [2026-04-17 09:54:01] Validation | Batch 720/784 | Loss: 0.3197 | LM_LOSS: 0.3087 | LB_LOSS: 1.0968 [2026-04-17 09:54:02] Validation | Batch 730/784 | Loss: 0.3192 | LM_LOSS: 0.3082 | LB_LOSS: 1.0968 [2026-04-17 09:54:03] Validation | Batch 740/784 | Loss: 0.3192 | LM_LOSS: 0.3083 | LB_LOSS: 1.0969 [2026-04-17 09:54:04] Validation | Batch 750/784 | Loss: 0.3186 | LM_LOSS: 0.3076 | LB_LOSS: 1.0969 [2026-04-17 09:54:06] Validation | Batch 760/784 | Loss: 0.3187 | LM_LOSS: 0.3078 | LB_LOSS: 1.0969 [2026-04-17 09:54:07] Validation | Batch 770/784 | Loss: 0.3189 | LM_LOSS: 0.3079 | LB_LOSS: 1.0969 [2026-04-17 09:54:09] Validation | Batch 780/784 | Loss: 0.3191 | LM_LOSS: 0.3081 | LB_LOSS: 1.0969 [2026-04-17 09:54:09] Validation | Batch 784/784 | Loss: 0.3193 | LM_LOSS: 0.3083 | LB_LOSS: 1.0969 [2026-04-17 09:54:12] Validation | Loss: 0.3193 | LM_LOSS: 0.3083 | LB_LOSS: 1.0969 | PPL: 1.36 | Time: 109.04s [2026-04-17 09:54:16] New best model saved! Val loss: 0.3193 [2026-04-17 09:54:23] Epoch 1 | Step 4010 | Loss: 0.3282 | LM: 0.3157 | LB: 1.1254 | CL0: 2.9 | CL1: 2.2 | HR0: 0.347/SR0: 0.348 | HR1: 0.451/SR1: 0.430 | LR: 1.00e-04 [2026-04-17 09:54:30] Epoch 1 | Step 4020 | Loss: 0.3282 | LM: 0.3160 | LB: 1.1253 | CL0: 2.9 | CL1: 2.2 | HR0: 0.347/SR0: 0.348 | HR1: 0.451/SR1: 0.430 | LR: 1.00e-04 [2026-04-17 09:54:37] Epoch 1 | Step 4030 | Loss: 0.3281 | LM: 0.3157 | LB: 1.1253 | CL0: 2.9 | CL1: 2.2 | HR0: 0.347/SR0: 0.348 | HR1: 0.450/SR1: 0.430 | LR: 1.00e-04 [2026-04-17 09:54:43] Epoch 1 | Step 4040 | Loss: 0.3281 | LM: 0.3156 | LB: 1.1252 | CL0: 2.9 | CL1: 2.2 | HR0: 0.347/SR0: 0.348 | HR1: 0.450/SR1: 0.430 | LR: 1.00e-04 [2026-04-17 09:54:50] Epoch 1 | Step 4050 | Loss: 0.3280 | LM: 0.3155 | LB: 1.1252 | CL0: 2.9 | CL1: 2.2 | HR0: 0.347/SR0: 0.348 | HR1: 0.450/SR1: 0.430 | LR: 1.00e-04 [2026-04-17 09:54:57] Epoch 1 | Step 4060 | Loss: 0.3279 | LM: 0.3155 | LB: 1.1251 | CL0: 2.9 | CL1: 2.2 | HR0: 0.347/SR0: 0.348 | HR1: 0.450/SR1: 0.429 | LR: 1.00e-04 [2026-04-17 09:55:03] Epoch 1 | Step 4070 | Loss: 0.3278 | LM: 0.3153 | LB: 1.1250 | CL0: 2.9 | CL1: 2.2 | HR0: 0.347/SR0: 0.348 | HR1: 0.450/SR1: 0.429 | LR: 1.00e-04 [2026-04-17 09:55:10] Epoch 1 | Step 4080 | Loss: 0.3278 | LM: 0.3154 | LB: 1.1250 | CL0: 2.9 | CL1: 2.2 | HR0: 0.347/SR0: 0.348 | HR1: 0.450/SR1: 0.429 | LR: 1.00e-04 [2026-04-17 09:55:16] Epoch 1 | Step 4090 | Loss: 0.3276 | LM: 0.3154 | LB: 1.1250 | CL0: 2.9 | CL1: 2.2 | HR0: 0.347/SR0: 0.348 | HR1: 0.450/SR1: 0.429 | LR: 1.00e-04 [2026-04-17 09:55:23] Epoch 1 | Step 4100 | Loss: 0.3276 | LM: 0.3154 | LB: 1.1249 | CL0: 2.9 | CL1: 2.2 | HR0: 0.347/SR0: 0.348 | HR1: 0.450/SR1: 0.429 | LR: 1.00e-04 [2026-04-17 09:55:30] Epoch 1 | Step 4110 | Loss: 0.3277 | LM: 0.3155 | LB: 1.1249 | CL0: 2.9 | CL1: 2.2 | HR0: 0.347/SR0: 0.348 | HR1: 0.450/SR1: 0.429 | LR: 1.00e-04 [2026-04-17 09:55:37] Epoch 1 | Step 4120 | Loss: 0.3276 | LM: 0.3155 | LB: 1.1248 | CL0: 2.9 | CL1: 2.2 | HR0: 0.347/SR0: 0.348 | HR1: 0.450/SR1: 0.429 | LR: 1.00e-04 [2026-04-17 09:55:43] Epoch 1 | Step 4130 | Loss: 0.3275 | LM: 0.3155 | LB: 1.1247 | CL0: 2.9 | CL1: 2.2 | HR0: 0.347/SR0: 0.348 | HR1: 0.450/SR1: 0.429 | LR: 1.00e-04 [2026-04-17 09:55:50] Epoch 1 | Step 4140 | Loss: 0.3275 | LM: 0.3154 | LB: 1.1247 | CL0: 2.9 | CL1: 2.2 | HR0: 0.347/SR0: 0.348 | HR1: 0.450/SR1: 0.429 | LR: 1.00e-04 [2026-04-17 09:55:57] Epoch 1 | Step 4150 | Loss: 0.3273 | LM: 0.3154 | LB: 1.1246 | CL0: 2.9 | CL1: 2.2 | HR0: 0.347/SR0: 0.348 | HR1: 0.450/SR1: 0.429 | LR: 1.00e-04 [2026-04-17 09:56:03] Epoch 1 | Step 4160 | Loss: 0.3273 | LM: 0.3153 | LB: 1.1246 | CL0: 2.9 | CL1: 2.2 | HR0: 0.347/SR0: 0.348 | HR1: 0.450/SR1: 0.429 | LR: 1.00e-04 [2026-04-17 09:56:10] Epoch 1 | Step 4170 | Loss: 0.3273 | LM: 0.3153 | LB: 1.1245 | CL0: 2.9 | CL1: 2.2 | HR0: 0.347/SR0: 0.348 | HR1: 0.450/SR1: 0.429 | LR: 1.00e-04 [2026-04-17 09:56:17] Epoch 1 | Step 4180 | Loss: 0.3273 | LM: 0.3152 | LB: 1.1245 | CL0: 2.9 | CL1: 2.2 | HR0: 0.347/SR0: 0.348 | HR1: 0.450/SR1: 0.429 | LR: 1.00e-04 [2026-04-17 09:56:23] Epoch 1 | Step 4190 | Loss: 0.3273 | LM: 0.3152 | LB: 1.1244 | CL0: 2.9 | CL1: 2.2 | HR0: 0.347/SR0: 0.348 | HR1: 0.450/SR1: 0.429 | LR: 1.00e-04 [2026-04-17 09:56:30] Epoch 1 | Step 4200 | Loss: 0.3273 | LM: 0.3152 | LB: 1.1243 | CL0: 2.9 | CL1: 2.2 | HR0: 0.347/SR0: 0.348 | HR1: 0.450/SR1: 0.429 | LR: 1.00e-04 [2026-04-17 09:56:37] Epoch 1 | Step 4210 | Loss: 0.3273 | LM: 0.3152 | LB: 1.1243 | CL0: 2.9 | CL1: 2.2 | HR0: 0.347/SR0: 0.348 | HR1: 0.450/SR1: 0.429 | LR: 1.00e-04 [2026-04-17 09:56:44] Epoch 1 | Step 4220 | Loss: 0.3272 | LM: 0.3149 | LB: 1.1243 | CL0: 2.9 | CL1: 2.2 | HR0: 0.347/SR0: 0.348 | HR1: 0.449/SR1: 0.429 | LR: 1.00e-04 [2026-04-17 09:56:50] Epoch 1 | Step 4230 | Loss: 0.3270 | LM: 0.3148 | LB: 1.1241 | CL0: 2.9 | CL1: 2.2 | HR0: 0.347/SR0: 0.348 | HR1: 0.449/SR1: 0.428 | LR: 1.00e-04 [2026-04-17 09:56:57] Epoch 1 | Step 4240 | Loss: 0.3270 | LM: 0.3147 | LB: 1.1241 | CL0: 2.9 | CL1: 2.2 | HR0: 0.347/SR0: 0.348 | HR1: 0.449/SR1: 0.428 | LR: 1.00e-04 [2026-04-17 09:57:03] Epoch 1 | Step 4250 | Loss: 0.3269 | LM: 0.3146 | LB: 1.1240 | CL0: 2.9 | CL1: 2.2 | HR0: 0.347/SR0: 0.348 | HR1: 0.449/SR1: 0.428 | LR: 1.00e-04 [2026-04-17 09:57:10] Epoch 1 | Step 4260 | Loss: 0.3269 | LM: 0.3146 | LB: 1.1239 | CL0: 2.9 | CL1: 2.2 | HR0: 0.347/SR0: 0.348 | HR1: 0.449/SR1: 0.428 | LR: 1.00e-04 [2026-04-17 09:57:17] Epoch 1 | Step 4270 | Loss: 0.3270 | LM: 0.3147 | LB: 1.1239 | CL0: 2.9 | CL1: 2.2 | HR0: 0.347/SR0: 0.348 | HR1: 0.449/SR1: 0.428 | LR: 1.00e-04 [2026-04-17 09:57:23] Epoch 1 | Step 4280 | Loss: 0.3270 | LM: 0.3146 | LB: 1.1239 | CL0: 2.9 | CL1: 2.2 | HR0: 0.347/SR0: 0.348 | HR1: 0.449/SR1: 0.428 | LR: 1.00e-04 [2026-04-17 09:57:30] Epoch 1 | Step 4290 | Loss: 0.3269 | LM: 0.3144 | LB: 1.1238 | CL0: 2.9 | CL1: 2.2 | HR0: 0.347/SR0: 0.348 | HR1: 0.449/SR1: 0.428 | LR: 1.00e-04 [2026-04-17 09:57:36] Epoch 1 | Step 4300 | Loss: 0.3268 | LM: 0.3145 | LB: 1.1238 | CL0: 2.9 | CL1: 2.2 | HR0: 0.347/SR0: 0.348 | HR1: 0.449/SR1: 0.428 | LR: 1.00e-04 [2026-04-17 09:57:43] Epoch 1 | Step 4310 | Loss: 0.3267 | LM: 0.3144 | LB: 1.1238 | CL0: 2.9 | CL1: 2.2 | HR0: 0.347/SR0: 0.348 | HR1: 0.449/SR1: 0.428 | LR: 1.00e-04 [2026-04-17 09:57:49] Epoch 1 | Step 4320 | Loss: 0.3266 | LM: 0.3142 | LB: 1.1238 | CL0: 2.9 | CL1: 2.2 | HR0: 0.347/SR0: 0.348 | HR1: 0.449/SR1: 0.428 | LR: 1.00e-04 [2026-04-17 09:57:56] Epoch 1 | Step 4330 | Loss: 0.3266 | LM: 0.3143 | LB: 1.1237 | CL0: 2.9 | CL1: 2.2 | HR0: 0.347/SR0: 0.348 | HR1: 0.449/SR1: 0.428 | LR: 1.00e-04 [2026-04-17 09:58:03] Epoch 1 | Step 4340 | Loss: 0.3266 | LM: 0.3143 | LB: 1.1236 | CL0: 2.9 | CL1: 2.2 | HR0: 0.347/SR0: 0.348 | HR1: 0.449/SR1: 0.428 | LR: 1.00e-04 [2026-04-17 09:58:09] Epoch 1 | Step 4350 | Loss: 0.3266 | LM: 0.3144 | LB: 1.1236 | CL0: 2.9 | CL1: 2.2 | HR0: 0.347/SR0: 0.348 | HR1: 0.449/SR1: 0.428 | LR: 1.00e-04 [2026-04-17 09:58:16] Epoch 1 | Step 4360 | Loss: 0.3265 | LM: 0.3144 | LB: 1.1236 | CL0: 2.9 | CL1: 2.2 | HR0: 0.347/SR0: 0.348 | HR1: 0.449/SR1: 0.428 | LR: 1.00e-04 [2026-04-17 09:58:23] Epoch 1 | Step 4370 | Loss: 0.3265 | LM: 0.3144 | LB: 1.1236 | CL0: 2.9 | CL1: 2.2 | HR0: 0.347/SR0: 0.348 | HR1: 0.449/SR1: 0.428 | LR: 1.00e-04 [2026-04-17 09:58:29] Epoch 1 | Step 4380 | Loss: 0.3263 | LM: 0.3147 | LB: 1.1235 | CL0: 2.9 | CL1: 2.2 | HR0: 0.347/SR0: 0.348 | HR1: 0.449/SR1: 0.428 | LR: 1.00e-04 [2026-04-17 09:58:36] Epoch 1 | Step 4390 | Loss: 0.3262 | LM: 0.3145 | LB: 1.1234 | CL0: 2.9 | CL1: 2.2 | HR0: 0.347/SR0: 0.348 | HR1: 0.449/SR1: 0.428 | LR: 1.00e-04 [2026-04-17 09:58:43] Epoch 1 | Step 4400 | Loss: 0.3262 | LM: 0.3145 | LB: 1.1234 | CL0: 2.9 | CL1: 2.2 | HR0: 0.347/SR0: 0.348 | HR1: 0.449/SR1: 0.428 | LR: 1.00e-04 [2026-04-17 09:58:49] Epoch 1 | Step 4410 | Loss: 0.3262 | LM: 0.3146 | LB: 1.1234 | CL0: 2.9 | CL1: 2.2 | HR0: 0.347/SR0: 0.348 | HR1: 0.449/SR1: 0.428 | LR: 1.00e-04 [2026-04-17 09:58:56] Epoch 1 | Step 4420 | Loss: 0.3262 | LM: 0.3147 | LB: 1.1234 | CL0: 2.9 | CL1: 2.2 | HR0: 0.347/SR0: 0.348 | HR1: 0.449/SR1: 0.428 | LR: 1.00e-04 [2026-04-17 09:59:03] Epoch 1 | Step 4430 | Loss: 0.3261 | LM: 0.3147 | LB: 1.1233 | CL0: 2.9 | CL1: 2.2 | HR0: 0.347/SR0: 0.348 | HR1: 0.448/SR1: 0.427 | LR: 1.00e-04 [2026-04-17 09:59:10] Epoch 1 | Step 4440 | Loss: 0.3261 | LM: 0.3146 | LB: 1.1233 | CL0: 2.9 | CL1: 2.2 | HR0: 0.347/SR0: 0.348 | HR1: 0.448/SR1: 0.427 | LR: 1.00e-04 [2026-04-17 09:59:18] Epoch 1 | Step 4450 | Loss: 0.3260 | LM: 0.3146 | LB: 1.1232 | CL0: 2.9 | CL1: 2.2 | HR0: 0.347/SR0: 0.348 | HR1: 0.448/SR1: 0.427 | LR: 1.00e-04 [2026-04-17 09:59:25] Epoch 1 | Step 4460 | Loss: 0.3260 | LM: 0.3146 | LB: 1.1231 | CL0: 2.9 | CL1: 2.2 | HR0: 0.347/SR0: 0.348 | HR1: 0.448/SR1: 0.427 | LR: 1.00e-04 [2026-04-17 09:59:32] Epoch 1 | Step 4470 | Loss: 0.3260 | LM: 0.3144 | LB: 1.1231 | CL0: 2.9 | CL1: 2.2 | HR0: 0.347/SR0: 0.348 | HR1: 0.448/SR1: 0.427 | LR: 1.00e-04 [2026-04-17 09:59:39] Epoch 1 | Step 4480 | Loss: 0.3259 | LM: 0.3142 | LB: 1.1231 | CL0: 2.9 | CL1: 2.2 | HR0: 0.347/SR0: 0.348 | HR1: 0.448/SR1: 0.427 | LR: 1.00e-04 [2026-04-17 09:59:46] Epoch 1 | Step 4490 | Loss: 0.3260 | LM: 0.3143 | LB: 1.1230 | CL0: 2.9 | CL1: 2.2 | HR0: 0.347/SR0: 0.348 | HR1: 0.448/SR1: 0.427 | LR: 1.00e-04 [2026-04-17 09:59:52] Epoch 1 | Step 4500 | Loss: 0.3260 | LM: 0.3142 | LB: 1.1230 | CL0: 2.9 | CL1: 2.2 | HR0: 0.347/SR0: 0.348 | HR1: 0.448/SR1: 0.427 | LR: 1.00e-04 [2026-04-17 09:59:59] Epoch 1 | Step 4510 | Loss: 0.3259 | LM: 0.3144 | LB: 1.1229 | CL0: 2.9 | CL1: 2.2 | HR0: 0.347/SR0: 0.348 | HR1: 0.448/SR1: 0.427 | LR: 1.00e-04 [2026-04-17 10:00:06] Epoch 1 | Step 4520 | Loss: 0.3259 | LM: 0.3149 | LB: 1.1229 | CL0: 2.9 | CL1: 2.2 | HR0: 0.347/SR0: 0.348 | HR1: 0.448/SR1: 0.427 | LR: 1.00e-04 [2026-04-17 10:00:13] Epoch 1 | Step 4530 | Loss: 0.3259 | LM: 0.3148 | LB: 1.1229 | CL0: 2.9 | CL1: 2.2 | HR0: 0.347/SR0: 0.348 | HR1: 0.448/SR1: 0.427 | LR: 1.00e-04 [2026-04-17 10:00:19] Epoch 1 | Step 4540 | Loss: 0.3258 | LM: 0.3147 | LB: 1.1228 | CL0: 2.9 | CL1: 2.2 | HR0: 0.347/SR0: 0.348 | HR1: 0.448/SR1: 0.427 | LR: 1.00e-04 [2026-04-17 10:00:26] Epoch 1 | Step 4550 | Loss: 0.3258 | LM: 0.3148 | LB: 1.1228 | CL0: 2.9 | CL1: 2.2 | HR0: 0.347/SR0: 0.348 | HR1: 0.448/SR1: 0.427 | LR: 1.00e-04 [2026-04-17 10:00:32] Epoch 1 | Step 4560 | Loss: 0.3257 | LM: 0.3146 | LB: 1.1227 | CL0: 2.9 | CL1: 2.2 | HR0: 0.347/SR0: 0.348 | HR1: 0.448/SR1: 0.427 | LR: 1.00e-04 [2026-04-17 10:00:39] Epoch 1 | Step 4570 | Loss: 0.3257 | LM: 0.3145 | LB: 1.1227 | CL0: 2.9 | CL1: 2.2 | HR0: 0.347/SR0: 0.348 | HR1: 0.448/SR1: 0.427 | LR: 1.00e-04 [2026-04-17 10:00:46] Epoch 1 | Step 4580 | Loss: 0.3256 | LM: 0.3144 | LB: 1.1226 | CL0: 2.9 | CL1: 2.2 | HR0: 0.347/SR0: 0.348 | HR1: 0.448/SR1: 0.427 | LR: 1.00e-04 [2026-04-17 10:00:52] Epoch 1 | Step 4590 | Loss: 0.3255 | LM: 0.3145 | LB: 1.1226 | CL0: 2.9 | CL1: 2.2 | HR0: 0.347/SR0: 0.348 | HR1: 0.448/SR1: 0.427 | LR: 1.00e-04 [2026-04-17 10:01:00] Epoch 1 | Step 4600 | Loss: 0.3255 | LM: 0.3144 | LB: 1.1226 | CL0: 2.9 | CL1: 2.2 | HR0: 0.347/SR0: 0.348 | HR1: 0.448/SR1: 0.427 | LR: 1.00e-04 [2026-04-17 10:01:06] Epoch 1 | Step 4610 | Loss: 0.3255 | LM: 0.3145 | LB: 1.1225 | CL0: 2.9 | CL1: 2.2 | HR0: 0.347/SR0: 0.348 | HR1: 0.448/SR1: 0.427 | LR: 1.00e-04 [2026-04-17 10:01:13] Epoch 1 | Step 4620 | Loss: 0.3255 | LM: 0.3144 | LB: 1.1225 | CL0: 2.9 | CL1: 2.2 | HR0: 0.347/SR0: 0.348 | HR1: 0.448/SR1: 0.427 | LR: 1.00e-04 [2026-04-17 10:01:20] Epoch 1 | Step 4630 | Loss: 0.3256 | LM: 0.3146 | LB: 1.1224 | CL0: 2.9 | CL1: 2.2 | HR0: 0.347/SR0: 0.348 | HR1: 0.448/SR1: 0.427 | LR: 1.00e-04 [2026-04-17 10:01:26] Epoch 1 | Step 4640 | Loss: 0.3255 | LM: 0.3144 | LB: 1.1224 | CL0: 2.9 | CL1: 2.2 | HR0: 0.347/SR0: 0.348 | HR1: 0.448/SR1: 0.427 | LR: 1.00e-04 [2026-04-17 10:01:33] Epoch 1 | Step 4650 | Loss: 0.3255 | LM: 0.3143 | LB: 1.1223 | CL0: 2.9 | CL1: 2.2 | HR0: 0.347/SR0: 0.348 | HR1: 0.448/SR1: 0.426 | LR: 1.00e-04 [2026-04-17 10:01:39] Epoch 1 | Step 4660 | Loss: 0.3255 | LM: 0.3144 | LB: 1.1223 | CL0: 2.9 | CL1: 2.2 | HR0: 0.347/SR0: 0.348 | HR1: 0.448/SR1: 0.426 | LR: 1.00e-04 [2026-04-17 10:01:46] Epoch 1 | Step 4670 | Loss: 0.3254 | LM: 0.3144 | LB: 1.1222 | CL0: 2.9 | CL1: 2.2 | HR0: 0.347/SR0: 0.348 | HR1: 0.448/SR1: 0.426 | LR: 1.00e-04 [2026-04-17 10:01:53] Epoch 1 | Step 4680 | Loss: 0.3254 | LM: 0.3144 | LB: 1.1222 | CL0: 2.9 | CL1: 2.2 | HR0: 0.347/SR0: 0.348 | HR1: 0.448/SR1: 0.426 | LR: 1.00e-04 [2026-04-17 10:02:00] Epoch 1 | Step 4690 | Loss: 0.3254 | LM: 0.3145 | LB: 1.1221 | CL0: 2.9 | CL1: 2.2 | HR0: 0.347/SR0: 0.348 | HR1: 0.447/SR1: 0.426 | LR: 1.00e-04 [2026-04-17 10:02:06] Epoch 1 | Step 4700 | Loss: 0.3253 | LM: 0.3144 | LB: 1.1221 | CL0: 2.9 | CL1: 2.2 | HR0: 0.347/SR0: 0.348 | HR1: 0.447/SR1: 0.426 | LR: 1.00e-04 [2026-04-17 10:02:13] Epoch 1 | Step 4710 | Loss: 0.3253 | LM: 0.3143 | LB: 1.1220 | CL0: 2.9 | CL1: 2.2 | HR0: 0.347/SR0: 0.348 | HR1: 0.447/SR1: 0.426 | LR: 1.00e-04 [2026-04-17 10:02:19] Epoch 1 | Step 4720 | Loss: 0.3253 | LM: 0.3142 | LB: 1.1219 | CL0: 2.9 | CL1: 2.2 | HR0: 0.347/SR0: 0.348 | HR1: 0.447/SR1: 0.426 | LR: 1.00e-04 [2026-04-17 10:02:26] Epoch 1 | Step 4730 | Loss: 0.3253 | LM: 0.3142 | LB: 1.1218 | CL0: 2.9 | CL1: 2.2 | HR0: 0.347/SR0: 0.348 | HR1: 0.447/SR1: 0.426 | LR: 1.00e-04 [2026-04-17 10:02:33] Epoch 1 | Step 4740 | Loss: 0.3253 | LM: 0.3143 | LB: 1.1218 | CL0: 2.9 | CL1: 2.2 | HR0: 0.347/SR0: 0.348 | HR1: 0.447/SR1: 0.426 | LR: 1.00e-04 [2026-04-17 10:02:39] Epoch 1 | Step 4750 | Loss: 0.3253 | LM: 0.3143 | LB: 1.1217 | CL0: 2.9 | CL1: 2.2 | HR0: 0.347/SR0: 0.348 | HR1: 0.447/SR1: 0.426 | LR: 1.00e-04 [2026-04-17 10:02:46] Epoch 1 | Step 4760 | Loss: 0.3252 | LM: 0.3142 | LB: 1.1217 | CL0: 2.9 | CL1: 2.2 | HR0: 0.347/SR0: 0.348 | HR1: 0.447/SR1: 0.426 | LR: 1.00e-04 [2026-04-17 10:02:52] Epoch 1 | Step 4770 | Loss: 0.3253 | LM: 0.3143 | LB: 1.1217 | CL0: 2.9 | CL1: 2.2 | HR0: 0.347/SR0: 0.348 | HR1: 0.447/SR1: 0.426 | LR: 1.00e-04 [2026-04-17 10:02:59] Epoch 1 | Step 4780 | Loss: 0.3252 | LM: 0.3143 | LB: 1.1216 | CL0: 2.9 | CL1: 2.2 | HR0: 0.347/SR0: 0.348 | HR1: 0.447/SR1: 0.426 | LR: 1.00e-04 [2026-04-17 10:03:06] Epoch 1 | Step 4790 | Loss: 0.3252 | LM: 0.3142 | LB: 1.1215 | CL0: 2.9 | CL1: 2.2 | HR0: 0.347/SR0: 0.348 | HR1: 0.447/SR1: 0.426 | LR: 1.00e-04 [2026-04-17 10:03:12] Epoch 1 | Step 4800 | Loss: 0.3252 | LM: 0.3144 | LB: 1.1215 | CL0: 2.9 | CL1: 2.2 | HR0: 0.347/SR0: 0.348 | HR1: 0.447/SR1: 0.426 | LR: 1.00e-04 [2026-04-17 10:03:19] Epoch 1 | Step 4810 | Loss: 0.3253 | LM: 0.3145 | LB: 1.1215 | CL0: 2.9 | CL1: 2.2 | HR0: 0.347/SR0: 0.348 | HR1: 0.447/SR1: 0.426 | LR: 1.00e-04 [2026-04-17 10:03:26] Epoch 1 | Step 4820 | Loss: 0.3253 | LM: 0.3145 | LB: 1.1214 | CL0: 2.9 | CL1: 2.2 | HR0: 0.347/SR0: 0.348 | HR1: 0.447/SR1: 0.426 | LR: 1.00e-04 [2026-04-17 10:03:32] Epoch 1 | Step 4830 | Loss: 0.3253 | LM: 0.3146 | LB: 1.1213 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.348 | HR1: 0.447/SR1: 0.425 | LR: 1.00e-04 [2026-04-17 10:03:39] Epoch 1 | Step 4840 | Loss: 0.3254 | LM: 0.3148 | LB: 1.1213 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.348 | HR1: 0.447/SR1: 0.425 | LR: 1.00e-04 [2026-04-17 10:03:46] Epoch 1 | Step 4850 | Loss: 0.3253 | LM: 0.3147 | LB: 1.1213 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.348 | HR1: 0.447/SR1: 0.425 | LR: 1.00e-04 [2026-04-17 10:03:53] Epoch 1 | Step 4860 | Loss: 0.3252 | LM: 0.3146 | LB: 1.1212 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.348 | HR1: 0.447/SR1: 0.425 | LR: 1.00e-04 [2026-04-17 10:04:00] Epoch 1 | Step 4870 | Loss: 0.3252 | LM: 0.3146 | LB: 1.1212 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.348 | HR1: 0.447/SR1: 0.425 | LR: 1.00e-04 [2026-04-17 10:04:06] Epoch 1 | Step 4880 | Loss: 0.3252 | LM: 0.3148 | LB: 1.1212 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.348 | HR1: 0.447/SR1: 0.425 | LR: 1.00e-04 [2026-04-17 10:04:13] Epoch 1 | Step 4890 | Loss: 0.3252 | LM: 0.3146 | LB: 1.1211 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.348 | HR1: 0.447/SR1: 0.425 | LR: 1.00e-04 [2026-04-17 10:04:20] Epoch 1 | Step 4900 | Loss: 0.3251 | LM: 0.3145 | LB: 1.1211 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.348 | HR1: 0.446/SR1: 0.425 | LR: 1.00e-04 [2026-04-17 10:04:27] Epoch 1 | Step 4910 | Loss: 0.3251 | LM: 0.3144 | LB: 1.1210 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.348 | HR1: 0.446/SR1: 0.425 | LR: 1.00e-04 [2026-04-17 10:04:34] Epoch 1 | Step 4920 | Loss: 0.3251 | LM: 0.3142 | LB: 1.1210 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.348 | HR1: 0.446/SR1: 0.425 | LR: 1.00e-04 [2026-04-17 10:04:41] Epoch 1 | Step 4930 | Loss: 0.3250 | LM: 0.3141 | LB: 1.1209 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.348 | HR1: 0.446/SR1: 0.425 | LR: 1.00e-04 [2026-04-17 10:04:47] Epoch 1 | Step 4940 | Loss: 0.3250 | LM: 0.3142 | LB: 1.1209 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.348 | HR1: 0.446/SR1: 0.425 | LR: 1.00e-04 [2026-04-17 10:04:54] Epoch 1 | Step 4950 | Loss: 0.3251 | LM: 0.3140 | LB: 1.1208 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.348 | HR1: 0.446/SR1: 0.425 | LR: 1.00e-04 [2026-04-17 10:05:01] Epoch 1 | Step 4960 | Loss: 0.3250 | LM: 0.3140 | LB: 1.1208 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.348 | HR1: 0.446/SR1: 0.425 | LR: 1.00e-04 [2026-04-17 10:05:07] Epoch 1 | Step 4970 | Loss: 0.3249 | LM: 0.3139 | LB: 1.1207 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.348 | HR1: 0.446/SR1: 0.425 | LR: 1.00e-04 [2026-04-17 10:05:14] Epoch 1 | Step 4980 | Loss: 0.3250 | LM: 0.3139 | LB: 1.1207 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.348 | HR1: 0.446/SR1: 0.425 | LR: 1.00e-04 [2026-04-17 10:05:21] Epoch 1 | Step 4990 | Loss: 0.3249 | LM: 0.3138 | LB: 1.1206 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.348 | HR1: 0.446/SR1: 0.425 | LR: 1.00e-04 [2026-04-17 10:05:28] Epoch 1 | Step 5000 | Loss: 0.3249 | LM: 0.3137 | LB: 1.1206 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.348 | HR1: 0.446/SR1: 0.425 | LR: 1.00e-04 [2026-04-17 10:05:29] Validation | Batch 10/784 | Loss: 0.3294 | LM_LOSS: 0.3184 | LB_LOSS: 1.0973 [2026-04-17 10:05:31] Validation | Batch 20/784 | Loss: 0.3395 | LM_LOSS: 0.3286 | LB_LOSS: 1.0975 [2026-04-17 10:05:32] Validation | Batch 30/784 | Loss: 0.3245 | LM_LOSS: 0.3136 | LB_LOSS: 1.0963 [2026-04-17 10:05:34] Validation | Batch 40/784 | Loss: 0.3262 | LM_LOSS: 0.3153 | LB_LOSS: 1.0960 [2026-04-17 10:05:35] Validation | Batch 50/784 | Loss: 0.3229 | LM_LOSS: 0.3120 | LB_LOSS: 1.0953 [2026-04-17 10:05:36] Validation | Batch 60/784 | Loss: 0.3237 | LM_LOSS: 0.3128 | LB_LOSS: 1.0949 [2026-04-17 10:05:38] Validation | Batch 70/784 | Loss: 0.3210 | LM_LOSS: 0.3101 | LB_LOSS: 1.0942 [2026-04-17 10:05:39] Validation | Batch 80/784 | Loss: 0.3167 | LM_LOSS: 0.3058 | LB_LOSS: 1.0937 [2026-04-17 10:05:40] Validation | Batch 90/784 | Loss: 0.3153 | LM_LOSS: 0.3044 | LB_LOSS: 1.0942 [2026-04-17 10:05:42] Validation | Batch 100/784 | Loss: 0.3169 | LM_LOSS: 0.3060 | LB_LOSS: 1.0947 [2026-04-17 10:05:43] Validation | Batch 110/784 | Loss: 0.3127 | LM_LOSS: 0.3018 | LB_LOSS: 1.0949 [2026-04-17 10:05:44] Validation | Batch 120/784 | Loss: 0.3158 | LM_LOSS: 0.3048 | LB_LOSS: 1.0947 [2026-04-17 10:05:46] Validation | Batch 130/784 | Loss: 0.3180 | LM_LOSS: 0.3071 | LB_LOSS: 1.0947 [2026-04-17 10:05:47] Validation | Batch 140/784 | Loss: 0.3175 | LM_LOSS: 0.3066 | LB_LOSS: 1.0945 [2026-04-17 10:05:49] Validation | Batch 150/784 | Loss: 0.3136 | LM_LOSS: 0.3026 | LB_LOSS: 1.0948 [2026-04-17 10:05:50] Validation | Batch 160/784 | Loss: 0.3139 | LM_LOSS: 0.3029 | LB_LOSS: 1.0944 [2026-04-17 10:05:52] Validation | Batch 170/784 | Loss: 0.3146 | LM_LOSS: 0.3036 | LB_LOSS: 1.0941 [2026-04-17 10:05:53] Validation | Batch 180/784 | Loss: 0.3119 | LM_LOSS: 0.3010 | LB_LOSS: 1.0942 [2026-04-17 10:05:54] Validation | Batch 190/784 | Loss: 0.3136 | LM_LOSS: 0.3026 | LB_LOSS: 1.0946 [2026-04-17 10:05:55] Validation | Batch 200/784 | Loss: 0.3138 | LM_LOSS: 0.3028 | LB_LOSS: 1.0947 [2026-04-17 10:05:57] Validation | Batch 210/784 | Loss: 0.3127 | LM_LOSS: 0.3018 | LB_LOSS: 1.0946 [2026-04-17 10:05:58] Validation | Batch 220/784 | Loss: 0.3135 | LM_LOSS: 0.3025 | LB_LOSS: 1.0946 [2026-04-17 10:06:00] Validation | Batch 230/784 | Loss: 0.3138 | LM_LOSS: 0.3029 | LB_LOSS: 1.0945 [2026-04-17 10:06:01] Validation | Batch 240/784 | Loss: 0.3141 | LM_LOSS: 0.3032 | LB_LOSS: 1.0949 [2026-04-17 10:06:03] Validation | Batch 250/784 | Loss: 0.3140 | LM_LOSS: 0.3030 | LB_LOSS: 1.0947 [2026-04-17 10:06:04] Validation | Batch 260/784 | Loss: 0.3140 | LM_LOSS: 0.3030 | LB_LOSS: 1.0950 [2026-04-17 10:06:06] Validation | Batch 270/784 | Loss: 0.3134 | LM_LOSS: 0.3025 | LB_LOSS: 1.0950 [2026-04-17 10:06:07] Validation | Batch 280/784 | Loss: 0.3140 | LM_LOSS: 0.3031 | LB_LOSS: 1.0952 [2026-04-17 10:06:08] Validation | Batch 290/784 | Loss: 0.3149 | LM_LOSS: 0.3040 | LB_LOSS: 1.0953 [2026-04-17 10:06:10] Validation | Batch 300/784 | Loss: 0.3155 | LM_LOSS: 0.3046 | LB_LOSS: 1.0953 [2026-04-17 10:06:11] Validation | Batch 310/784 | Loss: 0.3150 | LM_LOSS: 0.3041 | LB_LOSS: 1.0953 [2026-04-17 10:06:13] Validation | Batch 320/784 | Loss: 0.3164 | LM_LOSS: 0.3054 | LB_LOSS: 1.0953 [2026-04-17 10:06:14] Validation | Batch 330/784 | Loss: 0.3162 | LM_LOSS: 0.3052 | LB_LOSS: 1.0953 [2026-04-17 10:06:15] Validation | Batch 340/784 | Loss: 0.3153 | LM_LOSS: 0.3043 | LB_LOSS: 1.0954 [2026-04-17 10:06:17] Validation | Batch 350/784 | Loss: 0.3154 | LM_LOSS: 0.3045 | LB_LOSS: 1.0956 [2026-04-17 10:06:18] Validation | Batch 360/784 | Loss: 0.3151 | LM_LOSS: 0.3042 | LB_LOSS: 1.0956 [2026-04-17 10:06:19] Validation | Batch 370/784 | Loss: 0.3155 | LM_LOSS: 0.3046 | LB_LOSS: 1.0955 [2026-04-17 10:06:20] Validation | Batch 380/784 | Loss: 0.3153 | LM_LOSS: 0.3043 | LB_LOSS: 1.0956 [2026-04-17 10:06:22] Validation | Batch 390/784 | Loss: 0.3152 | LM_LOSS: 0.3042 | LB_LOSS: 1.0956 [2026-04-17 10:06:23] Validation | Batch 400/784 | Loss: 0.3153 | LM_LOSS: 0.3044 | LB_LOSS: 1.0956 [2026-04-17 10:06:24] Validation | Batch 410/784 | Loss: 0.3155 | LM_LOSS: 0.3046 | LB_LOSS: 1.0957 [2026-04-17 10:06:25] Validation | Batch 420/784 | Loss: 0.3156 | LM_LOSS: 0.3046 | LB_LOSS: 1.0957 [2026-04-17 10:06:27] Validation | Batch 430/784 | Loss: 0.3156 | LM_LOSS: 0.3046 | LB_LOSS: 1.0956 [2026-04-17 10:06:28] Validation | Batch 440/784 | Loss: 0.3152 | LM_LOSS: 0.3043 | LB_LOSS: 1.0957 [2026-04-17 10:06:30] Validation | Batch 450/784 | Loss: 0.3147 | LM_LOSS: 0.3037 | LB_LOSS: 1.0957 [2026-04-17 10:06:31] Validation | Batch 460/784 | Loss: 0.3151 | LM_LOSS: 0.3041 | LB_LOSS: 1.0958 [2026-04-17 10:06:32] Validation | Batch 470/784 | Loss: 0.3142 | LM_LOSS: 0.3033 | LB_LOSS: 1.0957 [2026-04-17 10:06:34] Validation | Batch 480/784 | Loss: 0.3147 | LM_LOSS: 0.3037 | LB_LOSS: 1.0957 [2026-04-17 10:06:35] Validation | Batch 490/784 | Loss: 0.3141 | LM_LOSS: 0.3032 | LB_LOSS: 1.0956 [2026-04-17 10:06:36] Validation | Batch 500/784 | Loss: 0.3146 | LM_LOSS: 0.3037 | LB_LOSS: 1.0955 [2026-04-17 10:06:38] Validation | Batch 510/784 | Loss: 0.3144 | LM_LOSS: 0.3034 | LB_LOSS: 1.0955 [2026-04-17 10:06:39] Validation | Batch 520/784 | Loss: 0.3145 | LM_LOSS: 0.3035 | LB_LOSS: 1.0954 [2026-04-17 10:06:41] Validation | Batch 530/784 | Loss: 0.3153 | LM_LOSS: 0.3043 | LB_LOSS: 1.0954 [2026-04-17 10:06:42] Validation | Batch 540/784 | Loss: 0.3157 | LM_LOSS: 0.3047 | LB_LOSS: 1.0954 [2026-04-17 10:06:43] Validation | Batch 550/784 | Loss: 0.3171 | LM_LOSS: 0.3061 | LB_LOSS: 1.0954 [2026-04-17 10:06:45] Validation | Batch 560/784 | Loss: 0.3171 | LM_LOSS: 0.3062 | LB_LOSS: 1.0954 [2026-04-17 10:06:46] Validation | Batch 570/784 | Loss: 0.3167 | LM_LOSS: 0.3058 | LB_LOSS: 1.0953 [2026-04-17 10:06:47] Validation | Batch 580/784 | Loss: 0.3162 | LM_LOSS: 0.3052 | LB_LOSS: 1.0954 [2026-04-17 10:06:49] Validation | Batch 590/784 | Loss: 0.3165 | LM_LOSS: 0.3055 | LB_LOSS: 1.0953 [2026-04-17 10:06:50] Validation | Batch 600/784 | Loss: 0.3164 | LM_LOSS: 0.3055 | LB_LOSS: 1.0952 [2026-04-17 10:06:51] Validation | Batch 610/784 | Loss: 0.3165 | LM_LOSS: 0.3056 | LB_LOSS: 1.0952 [2026-04-17 10:06:53] Validation | Batch 620/784 | Loss: 0.3165 | LM_LOSS: 0.3056 | LB_LOSS: 1.0952 [2026-04-17 10:06:54] Validation | Batch 630/784 | Loss: 0.3171 | LM_LOSS: 0.3062 | LB_LOSS: 1.0952 [2026-04-17 10:06:56] Validation | Batch 640/784 | Loss: 0.3173 | LM_LOSS: 0.3063 | LB_LOSS: 1.0952 [2026-04-17 10:06:58] Validation | Batch 650/784 | Loss: 0.3172 | LM_LOSS: 0.3062 | LB_LOSS: 1.0953 [2026-04-17 10:06:59] Validation | Batch 660/784 | Loss: 0.3174 | LM_LOSS: 0.3065 | LB_LOSS: 1.0953 [2026-04-17 10:07:00] Validation | Batch 670/784 | Loss: 0.3179 | LM_LOSS: 0.3069 | LB_LOSS: 1.0953 [2026-04-17 10:07:02] Validation | Batch 680/784 | Loss: 0.3176 | LM_LOSS: 0.3067 | LB_LOSS: 1.0953 [2026-04-17 10:07:03] Validation | Batch 690/784 | Loss: 0.3179 | LM_LOSS: 0.3069 | LB_LOSS: 1.0952 [2026-04-17 10:07:05] Validation | Batch 700/784 | Loss: 0.3179 | LM_LOSS: 0.3070 | LB_LOSS: 1.0952 [2026-04-17 10:07:06] Validation | Batch 710/784 | Loss: 0.3177 | LM_LOSS: 0.3067 | LB_LOSS: 1.0951 [2026-04-17 10:07:07] Validation | Batch 720/784 | Loss: 0.3174 | LM_LOSS: 0.3064 | LB_LOSS: 1.0951 [2026-04-17 10:07:09] Validation | Batch 730/784 | Loss: 0.3168 | LM_LOSS: 0.3059 | LB_LOSS: 1.0950 [2026-04-17 10:07:10] Validation | Batch 740/784 | Loss: 0.3170 | LM_LOSS: 0.3060 | LB_LOSS: 1.0951 [2026-04-17 10:07:11] Validation | Batch 750/784 | Loss: 0.3164 | LM_LOSS: 0.3054 | LB_LOSS: 1.0951 [2026-04-17 10:07:12] Validation | Batch 760/784 | Loss: 0.3165 | LM_LOSS: 0.3056 | LB_LOSS: 1.0951 [2026-04-17 10:07:14] Validation | Batch 770/784 | Loss: 0.3167 | LM_LOSS: 0.3058 | LB_LOSS: 1.0952 [2026-04-17 10:07:15] Validation | Batch 780/784 | Loss: 0.3169 | LM_LOSS: 0.3060 | LB_LOSS: 1.0951 [2026-04-17 10:07:16] Validation | Batch 784/784 | Loss: 0.3171 | LM_LOSS: 0.3062 | LB_LOSS: 1.0951 [2026-04-17 10:07:19] Validation | Loss: 0.3171 | LM_LOSS: 0.3062 | LB_LOSS: 1.0951 | PPL: 1.36 | Time: 107.82s [2026-04-17 10:07:25] New best model saved! Val loss: 0.3171 [2026-04-17 10:07:32] Epoch 1 | Step 5010 | Loss: 0.3248 | LM: 0.3136 | LB: 1.1206 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.348 | HR1: 0.446/SR1: 0.425 | LR: 1.00e-04 [2026-04-17 10:07:39] Epoch 1 | Step 5020 | Loss: 0.3247 | LM: 0.3135 | LB: 1.1205 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.348 | HR1: 0.446/SR1: 0.425 | LR: 1.00e-04 [2026-04-17 10:07:46] Epoch 1 | Step 5030 | Loss: 0.3247 | LM: 0.3134 | LB: 1.1205 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.348 | HR1: 0.446/SR1: 0.425 | LR: 1.00e-04 [2026-04-17 10:07:52] Epoch 1 | Step 5040 | Loss: 0.3246 | LM: 0.3133 | LB: 1.1204 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.348 | HR1: 0.446/SR1: 0.424 | LR: 1.00e-04 [2026-04-17 10:07:59] Epoch 1 | Step 5050 | Loss: 0.3245 | LM: 0.3133 | LB: 1.1204 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.348 | HR1: 0.446/SR1: 0.424 | LR: 1.00e-04 [2026-04-17 10:08:05] Epoch 1 | Step 5060 | Loss: 0.3245 | LM: 0.3132 | LB: 1.1203 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.348 | HR1: 0.446/SR1: 0.424 | LR: 1.00e-04 [2026-04-17 10:08:12] Epoch 1 | Step 5070 | Loss: 0.3243 | LM: 0.3132 | LB: 1.1202 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.348 | HR1: 0.446/SR1: 0.424 | LR: 1.00e-04 [2026-04-17 10:08:19] Epoch 1 | Step 5080 | Loss: 0.3243 | LM: 0.3131 | LB: 1.1202 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.348 | HR1: 0.446/SR1: 0.424 | LR: 1.00e-04 [2026-04-17 10:08:26] Epoch 1 | Step 5090 | Loss: 0.3243 | LM: 0.3131 | LB: 1.1202 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.348 | HR1: 0.446/SR1: 0.424 | LR: 1.00e-04 [2026-04-17 10:08:32] Epoch 1 | Step 5100 | Loss: 0.3243 | LM: 0.3132 | LB: 1.1201 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.348 | HR1: 0.446/SR1: 0.424 | LR: 1.00e-04 [2026-04-17 10:08:39] Epoch 1 | Step 5110 | Loss: 0.3243 | LM: 0.3131 | LB: 1.1201 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.348 | HR1: 0.446/SR1: 0.424 | LR: 1.00e-04 [2026-04-17 10:08:47] Epoch 1 | Step 5120 | Loss: 0.3243 | LM: 0.3130 | LB: 1.1200 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.348 | HR1: 0.446/SR1: 0.424 | LR: 1.00e-04 [2026-04-17 10:08:54] Epoch 1 | Step 5130 | Loss: 0.3242 | LM: 0.3130 | LB: 1.1200 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.348 | HR1: 0.445/SR1: 0.424 | LR: 1.00e-04 [2026-04-17 10:09:01] Epoch 1 | Step 5140 | Loss: 0.3241 | LM: 0.3127 | LB: 1.1200 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.348 | HR1: 0.445/SR1: 0.424 | LR: 1.00e-04 [2026-04-17 10:09:07] Epoch 1 | Step 5150 | Loss: 0.3241 | LM: 0.3128 | LB: 1.1199 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.348 | HR1: 0.445/SR1: 0.424 | LR: 1.00e-04 [2026-04-17 10:09:14] Epoch 1 | Step 5160 | Loss: 0.3240 | LM: 0.3127 | LB: 1.1199 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.348 | HR1: 0.445/SR1: 0.424 | LR: 1.00e-04 [2026-04-17 10:09:20] Epoch 1 | Step 5170 | Loss: 0.3240 | LM: 0.3124 | LB: 1.1199 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.348 | HR1: 0.445/SR1: 0.424 | LR: 1.00e-04 [2026-04-17 10:09:27] Epoch 1 | Step 5180 | Loss: 0.3240 | LM: 0.3124 | LB: 1.1198 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.348 | HR1: 0.445/SR1: 0.424 | LR: 1.00e-04 [2026-04-17 10:09:34] Epoch 1 | Step 5190 | Loss: 0.3240 | LM: 0.3124 | LB: 1.1198 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.348 | HR1: 0.445/SR1: 0.424 | LR: 1.00e-04 [2026-04-17 10:09:41] Epoch 1 | Step 5200 | Loss: 0.3239 | LM: 0.3123 | LB: 1.1198 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.348 | HR1: 0.445/SR1: 0.424 | LR: 1.00e-04 [2026-04-17 10:09:47] Epoch 1 | Step 5210 | Loss: 0.3238 | LM: 0.3122 | LB: 1.1197 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.348 | HR1: 0.445/SR1: 0.424 | LR: 1.00e-04 [2026-04-17 10:09:54] Epoch 1 | Step 5220 | Loss: 0.3238 | LM: 0.3122 | LB: 1.1197 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.348 | HR1: 0.445/SR1: 0.424 | LR: 1.00e-04 [2026-04-17 10:10:01] Epoch 1 | Step 5230 | Loss: 0.3239 | LM: 0.3122 | LB: 1.1197 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.348 | HR1: 0.445/SR1: 0.424 | LR: 1.00e-04 [2026-04-17 10:10:07] Epoch 1 | Step 5240 | Loss: 0.3239 | LM: 0.3122 | LB: 1.1196 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.348 | HR1: 0.445/SR1: 0.424 | LR: 1.00e-04 [2026-04-17 10:10:14] Epoch 1 | Step 5250 | Loss: 0.3238 | LM: 0.3120 | LB: 1.1195 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.348 | HR1: 0.445/SR1: 0.424 | LR: 1.00e-04 [2026-04-17 10:10:20] Epoch 1 | Step 5260 | Loss: 0.3238 | LM: 0.3119 | LB: 1.1195 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.348 | HR1: 0.445/SR1: 0.424 | LR: 1.00e-04 [2026-04-17 10:10:27] Epoch 1 | Step 5270 | Loss: 0.3237 | LM: 0.3119 | LB: 1.1194 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.348 | HR1: 0.445/SR1: 0.423 | LR: 1.00e-04 [2026-04-17 10:10:34] Epoch 1 | Step 5280 | Loss: 0.3237 | LM: 0.3119 | LB: 1.1194 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.348 | HR1: 0.445/SR1: 0.423 | LR: 1.00e-04 [2026-04-17 10:10:40] Epoch 1 | Step 5290 | Loss: 0.3238 | LM: 0.3120 | LB: 1.1194 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.348 | HR1: 0.445/SR1: 0.423 | LR: 1.00e-04 [2026-04-17 10:10:46] Epoch 1 | Step 5300 | Loss: 0.3237 | LM: 0.3118 | LB: 1.1194 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.348 | HR1: 0.445/SR1: 0.423 | LR: 1.00e-04 [2026-04-17 10:10:53] Epoch 1 | Step 5310 | Loss: 0.3237 | LM: 0.3118 | LB: 1.1193 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.348 | HR1: 0.445/SR1: 0.423 | LR: 1.00e-04 [2026-04-17 10:10:59] Epoch 1 | Step 5320 | Loss: 0.3237 | LM: 0.3118 | LB: 1.1193 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.348 | HR1: 0.445/SR1: 0.423 | LR: 1.00e-04 [2026-04-17 10:11:06] Epoch 1 | Step 5330 | Loss: 0.3237 | LM: 0.3118 | LB: 1.1192 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.348 | HR1: 0.445/SR1: 0.423 | LR: 1.00e-04 [2026-04-17 10:11:13] Epoch 1 | Step 5340 | Loss: 0.3236 | LM: 0.3117 | LB: 1.1192 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.348 | HR1: 0.445/SR1: 0.423 | LR: 1.00e-04 [2026-04-17 10:11:19] Epoch 1 | Step 5350 | Loss: 0.3237 | LM: 0.3117 | LB: 1.1192 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.348 | HR1: 0.445/SR1: 0.423 | LR: 1.00e-04 [2026-04-17 10:11:25] Epoch 1 | Step 5360 | Loss: 0.3237 | LM: 0.3117 | LB: 1.1191 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.348 | HR1: 0.445/SR1: 0.423 | LR: 1.00e-04 [2026-04-17 10:11:32] Epoch 1 | Step 5370 | Loss: 0.3237 | LM: 0.3117 | LB: 1.1191 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.348 | HR1: 0.445/SR1: 0.423 | LR: 1.00e-04 [2026-04-17 10:11:39] Epoch 1 | Step 5380 | Loss: 0.3236 | LM: 0.3118 | LB: 1.1191 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.348 | HR1: 0.445/SR1: 0.423 | LR: 1.00e-04 [2026-04-17 10:11:46] Epoch 1 | Step 5390 | Loss: 0.3236 | LM: 0.3117 | LB: 1.1191 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.348 | HR1: 0.445/SR1: 0.423 | LR: 1.00e-04 [2026-04-17 10:11:53] Epoch 1 | Step 5400 | Loss: 0.3235 | LM: 0.3118 | LB: 1.1191 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.348 | HR1: 0.445/SR1: 0.423 | LR: 1.00e-04 [2026-04-17 10:12:00] Epoch 1 | Step 5410 | Loss: 0.3235 | LM: 0.3120 | LB: 1.1190 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.348 | HR1: 0.445/SR1: 0.423 | LR: 1.00e-04 [2026-04-17 10:12:06] Epoch 1 | Step 5420 | Loss: 0.3235 | LM: 0.3119 | LB: 1.1190 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.348 | HR1: 0.445/SR1: 0.423 | LR: 1.00e-04 [2026-04-17 10:12:13] Epoch 1 | Step 5430 | Loss: 0.3236 | LM: 0.3120 | LB: 1.1190 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.348 | HR1: 0.445/SR1: 0.423 | LR: 1.00e-04 [2026-04-17 10:12:19] Epoch 1 | Step 5440 | Loss: 0.3235 | LM: 0.3120 | LB: 1.1189 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.348 | HR1: 0.445/SR1: 0.423 | LR: 1.00e-04 [2026-04-17 10:12:26] Epoch 1 | Step 5450 | Loss: 0.3235 | LM: 0.3119 | LB: 1.1189 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.348 | HR1: 0.445/SR1: 0.423 | LR: 1.00e-04 [2026-04-17 10:12:32] Epoch 1 | Step 5460 | Loss: 0.3235 | LM: 0.3120 | LB: 1.1189 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.348 | HR1: 0.445/SR1: 0.423 | LR: 1.00e-04 [2026-04-17 10:12:39] Epoch 1 | Step 5470 | Loss: 0.3234 | LM: 0.3120 | LB: 1.1188 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.348 | HR1: 0.444/SR1: 0.423 | LR: 1.00e-04 [2026-04-17 10:12:45] Epoch 1 | Step 5480 | Loss: 0.3233 | LM: 0.3119 | LB: 1.1188 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.348 | HR1: 0.444/SR1: 0.423 | LR: 1.00e-04 [2026-04-17 10:12:52] Epoch 1 | Step 5490 | Loss: 0.3233 | LM: 0.3118 | LB: 1.1187 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.348 | HR1: 0.444/SR1: 0.423 | LR: 1.00e-04 [2026-04-17 10:12:58] Epoch 1 | Step 5500 | Loss: 0.3233 | LM: 0.3119 | LB: 1.1187 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.348 | HR1: 0.444/SR1: 0.423 | LR: 1.00e-04 [2026-04-17 10:13:05] Epoch 1 | Step 5510 | Loss: 0.3232 | LM: 0.3119 | LB: 1.1187 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.348 | HR1: 0.444/SR1: 0.423 | LR: 1.00e-04 [2026-04-17 10:13:11] Epoch 1 | Step 5520 | Loss: 0.3232 | LM: 0.3120 | LB: 1.1186 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.348 | HR1: 0.444/SR1: 0.423 | LR: 1.00e-04 [2026-04-17 10:13:18] Epoch 1 | Step 5530 | Loss: 0.3232 | LM: 0.3120 | LB: 1.1186 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.348 | HR1: 0.444/SR1: 0.423 | LR: 1.00e-04 [2026-04-17 10:13:24] Epoch 1 | Step 5540 | Loss: 0.3232 | LM: 0.3120 | LB: 1.1186 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.348 | HR1: 0.444/SR1: 0.423 | LR: 1.00e-04 [2026-04-17 10:13:31] Epoch 1 | Step 5550 | Loss: 0.3232 | LM: 0.3120 | LB: 1.1185 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.444/SR1: 0.423 | LR: 1.00e-04 [2026-04-17 10:13:37] Epoch 1 | Step 5560 | Loss: 0.3231 | LM: 0.3120 | LB: 1.1185 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.444/SR1: 0.422 | LR: 1.00e-04 [2026-04-17 10:13:43] Epoch 1 | Step 5570 | Loss: 0.3231 | LM: 0.3119 | LB: 1.1184 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.444/SR1: 0.422 | LR: 1.00e-04 [2026-04-17 10:13:50] Epoch 1 | Step 5580 | Loss: 0.3231 | LM: 0.3120 | LB: 1.1184 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.444/SR1: 0.422 | LR: 1.00e-04 [2026-04-17 10:13:56] Epoch 1 | Step 5590 | Loss: 0.3231 | LM: 0.3120 | LB: 1.1184 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.444/SR1: 0.422 | LR: 1.00e-04 [2026-04-17 10:14:03] Epoch 1 | Step 5600 | Loss: 0.3230 | LM: 0.3120 | LB: 1.1184 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.444/SR1: 0.422 | LR: 1.00e-04 [2026-04-17 10:14:09] Epoch 1 | Step 5610 | Loss: 0.3230 | LM: 0.3120 | LB: 1.1183 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.444/SR1: 0.422 | LR: 1.00e-04 [2026-04-17 10:14:16] Epoch 1 | Step 5620 | Loss: 0.3230 | LM: 0.3120 | LB: 1.1183 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.444/SR1: 0.422 | LR: 1.00e-04 [2026-04-17 10:14:22] Epoch 1 | Step 5630 | Loss: 0.3230 | LM: 0.3120 | LB: 1.1182 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.444/SR1: 0.422 | LR: 1.00e-04 [2026-04-17 10:14:28] Epoch 1 | Step 5640 | Loss: 0.3230 | LM: 0.3120 | LB: 1.1182 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.444/SR1: 0.422 | LR: 1.00e-04 [2026-04-17 10:14:35] Epoch 1 | Step 5650 | Loss: 0.3230 | LM: 0.3119 | LB: 1.1182 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.444/SR1: 0.422 | LR: 1.00e-04 [2026-04-17 10:14:41] Epoch 1 | Step 5660 | Loss: 0.3230 | LM: 0.3118 | LB: 1.1181 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.444/SR1: 0.422 | LR: 1.00e-04 [2026-04-17 10:14:48] Epoch 1 | Step 5670 | Loss: 0.3229 | LM: 0.3118 | LB: 1.1181 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.444/SR1: 0.422 | LR: 1.00e-04 [2026-04-17 10:14:54] Epoch 1 | Step 5680 | Loss: 0.3231 | LM: 0.3119 | LB: 1.1181 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.444/SR1: 0.422 | LR: 1.00e-04 [2026-04-17 10:15:01] Epoch 1 | Step 5690 | Loss: 0.3229 | LM: 0.3119 | LB: 1.1180 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.444/SR1: 0.422 | LR: 1.00e-04 [2026-04-17 10:15:07] Epoch 1 | Step 5700 | Loss: 0.3230 | LM: 0.3118 | LB: 1.1180 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.444/SR1: 0.422 | LR: 1.00e-04 [2026-04-17 10:15:14] Epoch 1 | Step 5710 | Loss: 0.3228 | LM: 0.3117 | LB: 1.1179 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.444/SR1: 0.422 | LR: 1.00e-04 [2026-04-17 10:15:20] Epoch 1 | Step 5720 | Loss: 0.3228 | LM: 0.3116 | LB: 1.1179 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.444/SR1: 0.422 | LR: 1.00e-04 [2026-04-17 10:15:26] Epoch 1 | Step 5730 | Loss: 0.3227 | LM: 0.3114 | LB: 1.1178 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.444/SR1: 0.422 | LR: 1.00e-04 [2026-04-17 10:15:33] Epoch 1 | Step 5740 | Loss: 0.3226 | LM: 0.3114 | LB: 1.1178 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.444/SR1: 0.422 | LR: 1.00e-04 [2026-04-17 10:15:39] Epoch 1 | Step 5750 | Loss: 0.3227 | LM: 0.3115 | LB: 1.1178 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.444/SR1: 0.422 | LR: 1.00e-04 [2026-04-17 10:15:45] Epoch 1 | Step 5760 | Loss: 0.3227 | LM: 0.3115 | LB: 1.1178 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.444/SR1: 0.422 | LR: 1.00e-04 [2026-04-17 10:15:52] Epoch 1 | Step 5770 | Loss: 0.3226 | LM: 0.3113 | LB: 1.1177 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.444/SR1: 0.422 | LR: 1.00e-04 [2026-04-17 10:15:58] Epoch 1 | Step 5780 | Loss: 0.3227 | LM: 0.3115 | LB: 1.1177 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.444/SR1: 0.422 | LR: 1.00e-04 [2026-04-17 10:16:05] Epoch 1 | Step 5790 | Loss: 0.3227 | LM: 0.3113 | LB: 1.1177 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.444/SR1: 0.422 | LR: 1.00e-04 [2026-04-17 10:16:11] Epoch 1 | Step 5800 | Loss: 0.3226 | LM: 0.3113 | LB: 1.1176 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.443/SR1: 0.422 | LR: 1.00e-04 [2026-04-17 10:16:17] Epoch 1 | Step 5810 | Loss: 0.3225 | LM: 0.3112 | LB: 1.1176 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.443/SR1: 0.421 | LR: 1.00e-04 [2026-04-17 10:16:24] Epoch 1 | Step 5820 | Loss: 0.3224 | LM: 0.3112 | LB: 1.1175 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.443/SR1: 0.421 | LR: 1.00e-04 [2026-04-17 10:16:30] Epoch 1 | Step 5830 | Loss: 0.3224 | LM: 0.3112 | LB: 1.1175 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.443/SR1: 0.421 | LR: 1.00e-04 [2026-04-17 10:16:37] Epoch 1 | Step 5840 | Loss: 0.3224 | LM: 0.3112 | LB: 1.1174 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.443/SR1: 0.421 | LR: 1.00e-04 [2026-04-17 10:16:43] Epoch 1 | Step 5850 | Loss: 0.3224 | LM: 0.3111 | LB: 1.1174 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.443/SR1: 0.421 | LR: 1.00e-04 [2026-04-17 10:16:50] Epoch 1 | Step 5860 | Loss: 0.3224 | LM: 0.3111 | LB: 1.1174 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.443/SR1: 0.421 | LR: 1.00e-04 [2026-04-17 10:16:56] Epoch 1 | Step 5870 | Loss: 0.3223 | LM: 0.3109 | LB: 1.1173 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.443/SR1: 0.421 | LR: 1.00e-04 [2026-04-17 10:17:02] Epoch 1 | Step 5880 | Loss: 0.3223 | LM: 0.3109 | LB: 1.1173 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.443/SR1: 0.421 | LR: 1.00e-04 [2026-04-17 10:17:09] Epoch 1 | Step 5890 | Loss: 0.3222 | LM: 0.3108 | LB: 1.1173 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.443/SR1: 0.421 | LR: 1.00e-04 [2026-04-17 10:17:15] Epoch 1 | Step 5900 | Loss: 0.3222 | LM: 0.3108 | LB: 1.1172 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.443/SR1: 0.421 | LR: 1.00e-04 [2026-04-17 10:17:22] Epoch 1 | Step 5910 | Loss: 0.3222 | LM: 0.3107 | LB: 1.1172 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.443/SR1: 0.421 | LR: 1.00e-04 [2026-04-17 10:17:28] Epoch 1 | Step 5920 | Loss: 0.3221 | LM: 0.3106 | LB: 1.1172 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.443/SR1: 0.421 | LR: 1.00e-04 [2026-04-17 10:17:35] Epoch 1 | Step 5930 | Loss: 0.3221 | LM: 0.3104 | LB: 1.1171 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.443/SR1: 0.421 | LR: 1.00e-04 [2026-04-17 10:17:41] Epoch 1 | Step 5940 | Loss: 0.3222 | LM: 0.3105 | LB: 1.1171 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.443/SR1: 0.421 | LR: 1.00e-04 [2026-04-17 10:17:48] Epoch 1 | Step 5950 | Loss: 0.3222 | LM: 0.3106 | LB: 1.1171 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.443/SR1: 0.421 | LR: 1.00e-04 [2026-04-17 10:17:54] Epoch 1 | Step 5960 | Loss: 0.3221 | LM: 0.3108 | LB: 1.1170 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.443/SR1: 0.421 | LR: 1.00e-04 [2026-04-17 10:18:01] Epoch 1 | Step 5970 | Loss: 0.3221 | LM: 0.3107 | LB: 1.1170 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.443/SR1: 0.421 | LR: 1.00e-04 [2026-04-17 10:18:07] Epoch 1 | Step 5980 | Loss: 0.3221 | LM: 0.3107 | LB: 1.1170 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.443/SR1: 0.421 | LR: 1.00e-04 [2026-04-17 10:18:14] Epoch 1 | Step 5990 | Loss: 0.3221 | LM: 0.3106 | LB: 1.1169 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.443/SR1: 0.421 | LR: 1.00e-04 [2026-04-17 10:18:20] Epoch 1 | Step 6000 | Loss: 0.3220 | LM: 0.3106 | LB: 1.1169 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.443/SR1: 0.421 | LR: 1.00e-04 [2026-04-17 10:18:30] Checkpoint saved: outputs/2026-04-17/08-57-56/checkpoints/checkpoint_step_6000.pt [2026-04-17 10:18:46] Validation | Batch 10/784 | Loss: 0.3271 | LM_LOSS: 0.3162 | LB_LOSS: 1.0937 [2026-04-17 10:18:48] Validation | Batch 20/784 | Loss: 0.3365 | LM_LOSS: 0.3256 | LB_LOSS: 1.0937 [2026-04-17 10:18:49] Validation | Batch 30/784 | Loss: 0.3220 | LM_LOSS: 0.3111 | LB_LOSS: 1.0927 [2026-04-17 10:18:50] Validation | Batch 40/784 | Loss: 0.3245 | LM_LOSS: 0.3135 | LB_LOSS: 1.0925 [2026-04-17 10:18:52] Validation | Batch 50/784 | Loss: 0.3211 | LM_LOSS: 0.3102 | LB_LOSS: 1.0918 [2026-04-17 10:18:53] Validation | Batch 60/784 | Loss: 0.3221 | LM_LOSS: 0.3112 | LB_LOSS: 1.0913 [2026-04-17 10:18:54] Validation | Batch 70/784 | Loss: 0.3199 | LM_LOSS: 0.3090 | LB_LOSS: 1.0906 [2026-04-17 10:18:56] Validation | Batch 80/784 | Loss: 0.3154 | LM_LOSS: 0.3045 | LB_LOSS: 1.0901 [2026-04-17 10:18:57] Validation | Batch 90/784 | Loss: 0.3140 | LM_LOSS: 0.3031 | LB_LOSS: 1.0907 [2026-04-17 10:18:58] Validation | Batch 100/784 | Loss: 0.3158 | LM_LOSS: 0.3049 | LB_LOSS: 1.0912 [2026-04-17 10:19:00] Validation | Batch 110/784 | Loss: 0.3116 | LM_LOSS: 0.3007 | LB_LOSS: 1.0913 [2026-04-17 10:19:01] Validation | Batch 120/784 | Loss: 0.3148 | LM_LOSS: 0.3039 | LB_LOSS: 1.0912 [2026-04-17 10:19:03] Validation | Batch 130/784 | Loss: 0.3170 | LM_LOSS: 0.3061 | LB_LOSS: 1.0912 [2026-04-17 10:19:04] Validation | Batch 140/784 | Loss: 0.3164 | LM_LOSS: 0.3055 | LB_LOSS: 1.0909 [2026-04-17 10:19:05] Validation | Batch 150/784 | Loss: 0.3123 | LM_LOSS: 0.3014 | LB_LOSS: 1.0913 [2026-04-17 10:19:07] Validation | Batch 160/784 | Loss: 0.3127 | LM_LOSS: 0.3018 | LB_LOSS: 1.0909 [2026-04-17 10:19:08] Validation | Batch 170/784 | Loss: 0.3132 | LM_LOSS: 0.3023 | LB_LOSS: 1.0906 [2026-04-17 10:19:10] Validation | Batch 180/784 | Loss: 0.3106 | LM_LOSS: 0.2997 | LB_LOSS: 1.0906 [2026-04-17 10:19:11] Validation | Batch 190/784 | Loss: 0.3122 | LM_LOSS: 0.3013 | LB_LOSS: 1.0911 [2026-04-17 10:19:12] Validation | Batch 200/784 | Loss: 0.3124 | LM_LOSS: 0.3015 | LB_LOSS: 1.0912 [2026-04-17 10:19:14] Validation | Batch 210/784 | Loss: 0.3115 | LM_LOSS: 0.3006 | LB_LOSS: 1.0911 [2026-04-17 10:19:15] Validation | Batch 220/784 | Loss: 0.3123 | LM_LOSS: 0.3014 | LB_LOSS: 1.0911 [2026-04-17 10:19:17] Validation | Batch 230/784 | Loss: 0.3127 | LM_LOSS: 0.3018 | LB_LOSS: 1.0910 [2026-04-17 10:19:18] Validation | Batch 240/784 | Loss: 0.3130 | LM_LOSS: 0.3021 | LB_LOSS: 1.0914 [2026-04-17 10:19:19] Validation | Batch 250/784 | Loss: 0.3129 | LM_LOSS: 0.3020 | LB_LOSS: 1.0912 [2026-04-17 10:19:21] Validation | Batch 260/784 | Loss: 0.3128 | LM_LOSS: 0.3019 | LB_LOSS: 1.0915 [2026-04-17 10:19:23] Validation | Batch 270/784 | Loss: 0.3123 | LM_LOSS: 0.3014 | LB_LOSS: 1.0915 [2026-04-17 10:19:24] Validation | Batch 280/784 | Loss: 0.3130 | LM_LOSS: 0.3021 | LB_LOSS: 1.0917 [2026-04-17 10:19:25] Validation | Batch 290/784 | Loss: 0.3138 | LM_LOSS: 0.3029 | LB_LOSS: 1.0918 [2026-04-17 10:19:26] Validation | Batch 300/784 | Loss: 0.3143 | LM_LOSS: 0.3034 | LB_LOSS: 1.0919 [2026-04-17 10:19:28] Validation | Batch 310/784 | Loss: 0.3138 | LM_LOSS: 0.3029 | LB_LOSS: 1.0918 [2026-04-17 10:19:29] Validation | Batch 320/784 | Loss: 0.3153 | LM_LOSS: 0.3044 | LB_LOSS: 1.0918 [2026-04-17 10:19:31] Validation | Batch 330/784 | Loss: 0.3152 | LM_LOSS: 0.3043 | LB_LOSS: 1.0918 [2026-04-17 10:19:32] Validation | Batch 340/784 | Loss: 0.3142 | LM_LOSS: 0.3033 | LB_LOSS: 1.0919 [2026-04-17 10:19:33] Validation | Batch 350/784 | Loss: 0.3143 | LM_LOSS: 0.3034 | LB_LOSS: 1.0921 [2026-04-17 10:19:34] Validation | Batch 360/784 | Loss: 0.3141 | LM_LOSS: 0.3032 | LB_LOSS: 1.0921 [2026-04-17 10:19:36] Validation | Batch 370/784 | Loss: 0.3145 | LM_LOSS: 0.3036 | LB_LOSS: 1.0920 [2026-04-17 10:19:37] Validation | Batch 380/784 | Loss: 0.3143 | LM_LOSS: 0.3034 | LB_LOSS: 1.0921 [2026-04-17 10:19:38] Validation | Batch 390/784 | Loss: 0.3141 | LM_LOSS: 0.3032 | LB_LOSS: 1.0921 [2026-04-17 10:19:40] Validation | Batch 400/784 | Loss: 0.3142 | LM_LOSS: 0.3033 | LB_LOSS: 1.0921 [2026-04-17 10:19:41] Validation | Batch 410/784 | Loss: 0.3145 | LM_LOSS: 0.3035 | LB_LOSS: 1.0922 [2026-04-17 10:19:42] Validation | Batch 420/784 | Loss: 0.3145 | LM_LOSS: 0.3036 | LB_LOSS: 1.0922 [2026-04-17 10:19:43] Validation | Batch 430/784 | Loss: 0.3144 | LM_LOSS: 0.3035 | LB_LOSS: 1.0921 [2026-04-17 10:19:45] Validation | Batch 440/784 | Loss: 0.3140 | LM_LOSS: 0.3031 | LB_LOSS: 1.0922 [2026-04-17 10:19:46] Validation | Batch 450/784 | Loss: 0.3135 | LM_LOSS: 0.3026 | LB_LOSS: 1.0922 [2026-04-17 10:19:47] Validation | Batch 460/784 | Loss: 0.3140 | LM_LOSS: 0.3031 | LB_LOSS: 1.0922 [2026-04-17 10:19:49] Validation | Batch 470/784 | Loss: 0.3131 | LM_LOSS: 0.3022 | LB_LOSS: 1.0922 [2026-04-17 10:19:50] Validation | Batch 480/784 | Loss: 0.3135 | LM_LOSS: 0.3026 | LB_LOSS: 1.0921 [2026-04-17 10:19:51] Validation | Batch 490/784 | Loss: 0.3130 | LM_LOSS: 0.3021 | LB_LOSS: 1.0921 [2026-04-17 10:19:53] Validation | Batch 500/784 | Loss: 0.3136 | LM_LOSS: 0.3027 | LB_LOSS: 1.0920 [2026-04-17 10:19:54] Validation | Batch 510/784 | Loss: 0.3133 | LM_LOSS: 0.3024 | LB_LOSS: 1.0920 [2026-04-17 10:19:56] Validation | Batch 520/784 | Loss: 0.3134 | LM_LOSS: 0.3024 | LB_LOSS: 1.0919 [2026-04-17 10:19:57] Validation | Batch 530/784 | Loss: 0.3142 | LM_LOSS: 0.3033 | LB_LOSS: 1.0919 [2026-04-17 10:19:58] Validation | Batch 540/784 | Loss: 0.3146 | LM_LOSS: 0.3037 | LB_LOSS: 1.0919 [2026-04-17 10:20:00] Validation | Batch 550/784 | Loss: 0.3160 | LM_LOSS: 0.3051 | LB_LOSS: 1.0918 [2026-04-17 10:20:01] Validation | Batch 560/784 | Loss: 0.3161 | LM_LOSS: 0.3051 | LB_LOSS: 1.0919 [2026-04-17 10:20:03] Validation | Batch 570/784 | Loss: 0.3156 | LM_LOSS: 0.3047 | LB_LOSS: 1.0918 [2026-04-17 10:20:04] Validation | Batch 580/784 | Loss: 0.3151 | LM_LOSS: 0.3042 | LB_LOSS: 1.0919 [2026-04-17 10:20:05] Validation | Batch 590/784 | Loss: 0.3153 | LM_LOSS: 0.3044 | LB_LOSS: 1.0918 [2026-04-17 10:20:07] Validation | Batch 600/784 | Loss: 0.3153 | LM_LOSS: 0.3044 | LB_LOSS: 1.0917 [2026-04-17 10:20:08] Validation | Batch 610/784 | Loss: 0.3153 | LM_LOSS: 0.3044 | LB_LOSS: 1.0917 [2026-04-17 10:20:10] Validation | Batch 620/784 | Loss: 0.3153 | LM_LOSS: 0.3044 | LB_LOSS: 1.0917 [2026-04-17 10:20:11] Validation | Batch 630/784 | Loss: 0.3159 | LM_LOSS: 0.3050 | LB_LOSS: 1.0917 [2026-04-17 10:20:13] Validation | Batch 640/784 | Loss: 0.3161 | LM_LOSS: 0.3051 | LB_LOSS: 1.0917 [2026-04-17 10:20:14] Validation | Batch 650/784 | Loss: 0.3160 | LM_LOSS: 0.3051 | LB_LOSS: 1.0918 [2026-04-17 10:20:16] Validation | Batch 660/784 | Loss: 0.3163 | LM_LOSS: 0.3054 | LB_LOSS: 1.0918 [2026-04-17 10:20:17] Validation | Batch 670/784 | Loss: 0.3167 | LM_LOSS: 0.3058 | LB_LOSS: 1.0918 [2026-04-17 10:20:18] Validation | Batch 680/784 | Loss: 0.3165 | LM_LOSS: 0.3055 | LB_LOSS: 1.0918 [2026-04-17 10:20:20] Validation | Batch 690/784 | Loss: 0.3168 | LM_LOSS: 0.3058 | LB_LOSS: 1.0917 [2026-04-17 10:20:21] Validation | Batch 700/784 | Loss: 0.3168 | LM_LOSS: 0.3059 | LB_LOSS: 1.0917 [2026-04-17 10:20:23] Validation | Batch 710/784 | Loss: 0.3165 | LM_LOSS: 0.3056 | LB_LOSS: 1.0916 [2026-04-17 10:20:24] Validation | Batch 720/784 | Loss: 0.3162 | LM_LOSS: 0.3053 | LB_LOSS: 1.0916 [2026-04-17 10:20:25] Validation | Batch 730/784 | Loss: 0.3157 | LM_LOSS: 0.3048 | LB_LOSS: 1.0915 [2026-04-17 10:20:27] Validation | Batch 740/784 | Loss: 0.3158 | LM_LOSS: 0.3049 | LB_LOSS: 1.0916 [2026-04-17 10:20:28] Validation | Batch 750/784 | Loss: 0.3152 | LM_LOSS: 0.3043 | LB_LOSS: 1.0916 [2026-04-17 10:20:29] Validation | Batch 760/784 | Loss: 0.3153 | LM_LOSS: 0.3044 | LB_LOSS: 1.0916 [2026-04-17 10:20:30] Validation | Batch 770/784 | Loss: 0.3155 | LM_LOSS: 0.3046 | LB_LOSS: 1.0917 [2026-04-17 10:20:32] Validation | Batch 780/784 | Loss: 0.3158 | LM_LOSS: 0.3048 | LB_LOSS: 1.0916 [2026-04-17 10:20:32] Validation | Batch 784/784 | Loss: 0.3159 | LM_LOSS: 0.3050 | LB_LOSS: 1.0916 [2026-04-17 10:20:35] Validation | Loss: 0.3159 | LM_LOSS: 0.3050 | LB_LOSS: 1.0916 | PPL: 1.36 | Time: 107.57s [2026-04-17 10:20:39] New best model saved! Val loss: 0.3159 [2026-04-17 10:20:45] Epoch 1 | Step 6010 | Loss: 0.3220 | LM: 0.3107 | LB: 1.1169 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.443/SR1: 0.421 | LR: 1.00e-04 [2026-04-17 10:20:52] Epoch 1 | Step 6020 | Loss: 0.3219 | LM: 0.3106 | LB: 1.1168 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.443/SR1: 0.421 | LR: 1.00e-04 [2026-04-17 10:20:58] Epoch 1 | Step 6030 | Loss: 0.3219 | LM: 0.3108 | LB: 1.1168 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.443/SR1: 0.421 | LR: 1.00e-04 [2026-04-17 10:21:05] Epoch 1 | Step 6040 | Loss: 0.3219 | LM: 0.3108 | LB: 1.1168 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.443/SR1: 0.421 | LR: 1.00e-04 [2026-04-17 10:21:11] Epoch 1 | Step 6050 | Loss: 0.3219 | LM: 0.3108 | LB: 1.1167 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.443/SR1: 0.421 | LR: 1.00e-04 [2026-04-17 10:21:18] Epoch 1 | Step 6060 | Loss: 0.3219 | LM: 0.3109 | LB: 1.1167 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.443/SR1: 0.421 | LR: 1.00e-04 [2026-04-17 10:21:24] Epoch 1 | Step 6070 | Loss: 0.3219 | LM: 0.3109 | LB: 1.1167 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.443/SR1: 0.420 | LR: 1.00e-04 [2026-04-17 10:21:31] Epoch 1 | Step 6080 | Loss: 0.3219 | LM: 0.3109 | LB: 1.1166 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.443/SR1: 0.420 | LR: 1.00e-04 [2026-04-17 10:21:37] Epoch 1 | Step 6090 | Loss: 0.3218 | LM: 0.3109 | LB: 1.1166 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.443/SR1: 0.420 | LR: 1.00e-04 [2026-04-17 10:21:44] Epoch 1 | Step 6100 | Loss: 0.3217 | LM: 0.3107 | LB: 1.1166 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.443/SR1: 0.420 | LR: 1.00e-04 [2026-04-17 10:21:50] Epoch 1 | Step 6110 | Loss: 0.3217 | LM: 0.3107 | LB: 1.1166 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.442/SR1: 0.420 | LR: 1.00e-04 [2026-04-17 10:21:57] Epoch 1 | Step 6120 | Loss: 0.3217 | LM: 0.3106 | LB: 1.1166 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.442/SR1: 0.420 | LR: 1.00e-04 [2026-04-17 10:22:03] Epoch 1 | Step 6130 | Loss: 0.3217 | LM: 0.3106 | LB: 1.1165 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.442/SR1: 0.420 | LR: 1.00e-04 [2026-04-17 10:22:10] Epoch 1 | Step 6140 | Loss: 0.3217 | LM: 0.3106 | LB: 1.1165 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.442/SR1: 0.420 | LR: 1.00e-04 [2026-04-17 10:22:16] Epoch 1 | Step 6150 | Loss: 0.3216 | LM: 0.3106 | LB: 1.1165 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.442/SR1: 0.420 | LR: 1.00e-04 [2026-04-17 10:22:23] Epoch 1 | Step 6160 | Loss: 0.3216 | LM: 0.3106 | LB: 1.1164 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.442/SR1: 0.420 | LR: 1.00e-04 [2026-04-17 10:22:30] Epoch 1 | Step 6170 | Loss: 0.3216 | LM: 0.3106 | LB: 1.1164 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.442/SR1: 0.420 | LR: 1.00e-04 [2026-04-17 10:22:36] Epoch 1 | Step 6180 | Loss: 0.3216 | LM: 0.3106 | LB: 1.1163 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.442/SR1: 0.420 | LR: 1.00e-04 [2026-04-17 10:22:42] Epoch 1 | Step 6190 | Loss: 0.3215 | LM: 0.3105 | LB: 1.1163 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.442/SR1: 0.420 | LR: 1.00e-04 [2026-04-17 10:22:49] Epoch 1 | Step 6200 | Loss: 0.3215 | LM: 0.3105 | LB: 1.1163 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.442/SR1: 0.420 | LR: 1.00e-04 [2026-04-17 10:22:56] Epoch 1 | Step 6210 | Loss: 0.3214 | LM: 0.3105 | LB: 1.1163 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.442/SR1: 0.420 | LR: 1.00e-04 [2026-04-17 10:23:02] Epoch 1 | Step 6220 | Loss: 0.3215 | LM: 0.3104 | LB: 1.1162 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.442/SR1: 0.420 | LR: 1.00e-04 [2026-04-17 10:23:09] Epoch 1 | Step 6230 | Loss: 0.3215 | LM: 0.3105 | LB: 1.1162 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.442/SR1: 0.420 | LR: 1.00e-04 [2026-04-17 10:23:16] Epoch 1 | Step 6240 | Loss: 0.3216 | LM: 0.3105 | LB: 1.1161 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.442/SR1: 0.420 | LR: 1.00e-04 [2026-04-17 10:23:22] Epoch 1 | Step 6250 | Loss: 0.3216 | LM: 0.3104 | LB: 1.1161 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.442/SR1: 0.420 | LR: 1.00e-04 [2026-04-17 10:23:29] Epoch 1 | Step 6260 | Loss: 0.3216 | LM: 0.3105 | LB: 1.1160 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.442/SR1: 0.420 | LR: 1.00e-04 [2026-04-17 10:23:35] Epoch 1 | Step 6270 | Loss: 0.3215 | LM: 0.3104 | LB: 1.1160 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.442/SR1: 0.420 | LR: 1.00e-04 [2026-04-17 10:23:42] Epoch 1 | Step 6280 | Loss: 0.3215 | LM: 0.3103 | LB: 1.1160 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.442/SR1: 0.420 | LR: 1.00e-04 [2026-04-17 10:23:48] Epoch 1 | Step 6290 | Loss: 0.3216 | LM: 0.3104 | LB: 1.1160 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.442/SR1: 0.420 | LR: 1.00e-04 [2026-04-17 10:23:55] Epoch 1 | Step 6300 | Loss: 0.3215 | LM: 0.3104 | LB: 1.1159 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.442/SR1: 0.420 | LR: 1.00e-04 [2026-04-17 10:24:01] Epoch 1 | Step 6310 | Loss: 0.3215 | LM: 0.3103 | LB: 1.1159 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.442/SR1: 0.420 | LR: 1.00e-04 [2026-04-17 10:24:08] Epoch 1 | Step 6320 | Loss: 0.3216 | LM: 0.3105 | LB: 1.1159 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.442/SR1: 0.419 | LR: 1.00e-04 [2026-04-17 10:24:14] Epoch 1 | Step 6330 | Loss: 0.3216 | LM: 0.3106 | LB: 1.1158 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.442/SR1: 0.419 | LR: 1.00e-04 [2026-04-17 10:24:20] Epoch 1 | Step 6340 | Loss: 0.3216 | LM: 0.3105 | LB: 1.1158 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.442/SR1: 0.419 | LR: 1.00e-04 [2026-04-17 10:24:27] Epoch 1 | Step 6350 | Loss: 0.3216 | LM: 0.3105 | LB: 1.1158 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.442/SR1: 0.419 | LR: 1.00e-04 [2026-04-17 10:24:33] Epoch 1 | Step 6360 | Loss: 0.3216 | LM: 0.3105 | LB: 1.1158 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.442/SR1: 0.419 | LR: 1.00e-04 [2026-04-17 10:24:40] Epoch 1 | Step 6370 | Loss: 0.3215 | LM: 0.3103 | LB: 1.1157 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.442/SR1: 0.419 | LR: 1.00e-04 [2026-04-17 10:24:46] Epoch 1 | Step 6380 | Loss: 0.3215 | LM: 0.3103 | LB: 1.1157 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.442/SR1: 0.419 | LR: 1.00e-04 [2026-04-17 10:24:52] Epoch 1 | Step 6390 | Loss: 0.3216 | LM: 0.3102 | LB: 1.1157 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.442/SR1: 0.419 | LR: 1.00e-04 [2026-04-17 10:24:59] Epoch 1 | Step 6400 | Loss: 0.3215 | LM: 0.3101 | LB: 1.1156 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.442/SR1: 0.419 | LR: 1.00e-04 [2026-04-17 10:25:05] Epoch 1 | Step 6410 | Loss: 0.3215 | LM: 0.3100 | LB: 1.1156 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.442/SR1: 0.419 | LR: 1.00e-04 [2026-04-17 10:25:12] Epoch 1 | Step 6420 | Loss: 0.3215 | LM: 0.3100 | LB: 1.1156 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.442/SR1: 0.419 | LR: 1.00e-04 [2026-04-17 10:25:18] Epoch 1 | Step 6430 | Loss: 0.3214 | LM: 0.3099 | LB: 1.1156 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.442/SR1: 0.419 | LR: 1.00e-04 [2026-04-17 10:25:25] Epoch 1 | Step 6440 | Loss: 0.3214 | LM: 0.3099 | LB: 1.1156 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.442/SR1: 0.419 | LR: 1.00e-04 [2026-04-17 10:25:32] Epoch 1 | Step 6450 | Loss: 0.3213 | LM: 0.3099 | LB: 1.1155 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.442/SR1: 0.419 | LR: 1.00e-04 [2026-04-17 10:25:38] Epoch 1 | Step 6460 | Loss: 0.3214 | LM: 0.3098 | LB: 1.1155 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.441/SR1: 0.419 | LR: 1.00e-04 [2026-04-17 10:25:45] Epoch 1 | Step 6470 | Loss: 0.3213 | LM: 0.3099 | LB: 1.1154 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.441/SR1: 0.419 | LR: 1.00e-04 [2026-04-17 10:25:51] Epoch 1 | Step 6480 | Loss: 0.3213 | LM: 0.3098 | LB: 1.1154 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.441/SR1: 0.419 | LR: 1.00e-04 [2026-04-17 10:25:58] Epoch 1 | Step 6490 | Loss: 0.3213 | LM: 0.3097 | LB: 1.1154 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.441/SR1: 0.419 | LR: 1.00e-04 [2026-04-17 10:26:04] Epoch 1 | Step 6500 | Loss: 0.3213 | LM: 0.3097 | LB: 1.1154 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.441/SR1: 0.419 | LR: 1.00e-04 [2026-04-17 10:26:11] Epoch 1 | Step 6510 | Loss: 0.3213 | LM: 0.3097 | LB: 1.1153 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.441/SR1: 0.419 | LR: 1.00e-04 [2026-04-17 10:26:17] Epoch 1 | Step 6520 | Loss: 0.3213 | LM: 0.3097 | LB: 1.1153 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.441/SR1: 0.419 | LR: 1.00e-04 [2026-04-17 10:26:24] Epoch 1 | Step 6530 | Loss: 0.3213 | LM: 0.3098 | LB: 1.1153 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.441/SR1: 0.419 | LR: 1.00e-04 [2026-04-17 10:26:30] Epoch 1 | Step 6540 | Loss: 0.3213 | LM: 0.3098 | LB: 1.1153 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.441/SR1: 0.419 | LR: 1.00e-04 [2026-04-17 10:26:37] Epoch 1 | Step 6550 | Loss: 0.3212 | LM: 0.3098 | LB: 1.1152 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.441/SR1: 0.419 | LR: 1.00e-04 [2026-04-17 10:26:43] Epoch 1 | Step 6560 | Loss: 0.3212 | LM: 0.3097 | LB: 1.1152 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.441/SR1: 0.419 | LR: 1.00e-04 [2026-04-17 10:26:50] Epoch 1 | Step 6570 | Loss: 0.3212 | LM: 0.3098 | LB: 1.1151 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.441/SR1: 0.419 | LR: 1.00e-04 [2026-04-17 10:26:56] Epoch 1 | Step 6580 | Loss: 0.3213 | LM: 0.3097 | LB: 1.1151 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.441/SR1: 0.419 | LR: 1.00e-04 [2026-04-17 10:27:03] Epoch 1 | Step 6590 | Loss: 0.3213 | LM: 0.3096 | LB: 1.1151 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.441/SR1: 0.419 | LR: 1.00e-04 [2026-04-17 10:27:09] Epoch 1 | Step 6600 | Loss: 0.3212 | LM: 0.3095 | LB: 1.1150 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.441/SR1: 0.419 | LR: 1.00e-04 [2026-04-17 10:27:15] Epoch 1 | Step 6610 | Loss: 0.3212 | LM: 0.3095 | LB: 1.1150 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.441/SR1: 0.419 | LR: 1.00e-04 [2026-04-17 10:27:22] Epoch 1 | Step 6620 | Loss: 0.3213 | LM: 0.3094 | LB: 1.1150 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.441/SR1: 0.418 | LR: 1.00e-04 [2026-04-17 10:27:28] Epoch 1 | Step 6630 | Loss: 0.3212 | LM: 0.3093 | LB: 1.1149 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.441/SR1: 0.418 | LR: 1.00e-04 [2026-04-17 10:27:35] Epoch 1 | Step 6640 | Loss: 0.3212 | LM: 0.3092 | LB: 1.1149 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.441/SR1: 0.418 | LR: 1.00e-04 [2026-04-17 10:27:41] Epoch 1 | Step 6650 | Loss: 0.3212 | LM: 0.3093 | LB: 1.1149 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.441/SR1: 0.418 | LR: 1.00e-04 [2026-04-17 10:27:48] Epoch 1 | Step 6660 | Loss: 0.3212 | LM: 0.3094 | LB: 1.1148 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.441/SR1: 0.418 | LR: 1.00e-04 [2026-04-17 10:27:54] Epoch 1 | Step 6670 | Loss: 0.3213 | LM: 0.3095 | LB: 1.1148 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.441/SR1: 0.418 | LR: 1.00e-04 [2026-04-17 10:28:00] Epoch 1 | Step 6680 | Loss: 0.3212 | LM: 0.3095 | LB: 1.1148 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.441/SR1: 0.418 | LR: 1.00e-04 [2026-04-17 10:28:06] Epoch 1 | Step 6690 | Loss: 0.3212 | LM: 0.3094 | LB: 1.1147 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.441/SR1: 0.418 | LR: 1.00e-04 [2026-04-17 10:28:13] Epoch 1 | Step 6700 | Loss: 0.3211 | LM: 0.3094 | LB: 1.1147 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.441/SR1: 0.418 | LR: 1.00e-04 [2026-04-17 10:28:19] Epoch 1 | Step 6710 | Loss: 0.3211 | LM: 0.3093 | LB: 1.1147 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.441/SR1: 0.418 | LR: 1.00e-04 [2026-04-17 10:28:26] Epoch 1 | Step 6720 | Loss: 0.3210 | LM: 0.3092 | LB: 1.1146 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.441/SR1: 0.418 | LR: 1.00e-04 [2026-04-17 10:28:32] Epoch 1 | Step 6730 | Loss: 0.3210 | LM: 0.3092 | LB: 1.1146 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.441/SR1: 0.418 | LR: 1.00e-04 [2026-04-17 10:28:38] Epoch 1 | Step 6740 | Loss: 0.3210 | LM: 0.3092 | LB: 1.1146 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.441/SR1: 0.418 | LR: 1.00e-04 [2026-04-17 10:28:45] Epoch 1 | Step 6750 | Loss: 0.3210 | LM: 0.3092 | LB: 1.1145 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.441/SR1: 0.418 | LR: 1.00e-04 [2026-04-17 10:28:51] Epoch 1 | Step 6760 | Loss: 0.3209 | LM: 0.3091 | LB: 1.1145 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.441/SR1: 0.418 | LR: 1.00e-04 [2026-04-17 10:28:57] Epoch 1 | Step 6770 | Loss: 0.3210 | LM: 0.3092 | LB: 1.1145 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.441/SR1: 0.418 | LR: 1.00e-04 [2026-04-17 10:29:04] Epoch 1 | Step 6780 | Loss: 0.3210 | LM: 0.3093 | LB: 1.1145 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.441/SR1: 0.418 | LR: 1.00e-04 [2026-04-17 10:29:10] Epoch 1 | Step 6790 | Loss: 0.3209 | LM: 0.3091 | LB: 1.1144 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.441/SR1: 0.418 | LR: 1.00e-04 [2026-04-17 10:29:17] Epoch 1 | Step 6800 | Loss: 0.3209 | LM: 0.3092 | LB: 1.1144 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.441/SR1: 0.418 | LR: 1.00e-04 [2026-04-17 10:29:23] Epoch 1 | Step 6810 | Loss: 0.3209 | LM: 0.3092 | LB: 1.1144 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.440/SR1: 0.418 | LR: 1.00e-04 [2026-04-17 10:29:29] Epoch 1 | Step 6820 | Loss: 0.3208 | LM: 0.3092 | LB: 1.1143 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.440/SR1: 0.418 | LR: 1.00e-04 [2026-04-17 10:29:36] Epoch 1 | Step 6830 | Loss: 0.3208 | LM: 0.3091 | LB: 1.1143 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.440/SR1: 0.418 | LR: 1.00e-04 [2026-04-17 10:29:42] Epoch 1 | Step 6840 | Loss: 0.3208 | LM: 0.3092 | LB: 1.1143 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.440/SR1: 0.418 | LR: 1.00e-04 [2026-04-17 10:29:49] Epoch 1 | Step 6850 | Loss: 0.3208 | LM: 0.3091 | LB: 1.1142 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.440/SR1: 0.418 | LR: 1.00e-04 [2026-04-17 10:29:55] Epoch 1 | Step 6860 | Loss: 0.3208 | LM: 0.3091 | LB: 1.1142 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.440/SR1: 0.418 | LR: 1.00e-04 [2026-04-17 10:30:01] Epoch 1 | Step 6870 | Loss: 0.3207 | LM: 0.3091 | LB: 1.1142 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.440/SR1: 0.418 | LR: 1.00e-04 [2026-04-17 10:30:08] Epoch 1 | Step 6880 | Loss: 0.3207 | LM: 0.3092 | LB: 1.1142 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.440/SR1: 0.418 | LR: 1.00e-04 [2026-04-17 10:30:14] Epoch 1 | Step 6890 | Loss: 0.3208 | LM: 0.3091 | LB: 1.1141 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.440/SR1: 0.417 | LR: 1.00e-04 [2026-04-17 10:30:20] Epoch 1 | Step 6900 | Loss: 0.3207 | LM: 0.3090 | LB: 1.1141 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.440/SR1: 0.417 | LR: 1.00e-04 [2026-04-17 10:30:27] Epoch 1 | Step 6910 | Loss: 0.3207 | LM: 0.3091 | LB: 1.1141 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.440/SR1: 0.417 | LR: 1.00e-04 [2026-04-17 10:30:33] Epoch 1 | Step 6920 | Loss: 0.3207 | LM: 0.3092 | LB: 1.1140 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.440/SR1: 0.417 | LR: 1.00e-04 [2026-04-17 10:30:39] Epoch 1 | Step 6930 | Loss: 0.3206 | LM: 0.3091 | LB: 1.1140 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.440/SR1: 0.417 | LR: 1.00e-04 [2026-04-17 10:30:46] Epoch 1 | Step 6940 | Loss: 0.3207 | LM: 0.3090 | LB: 1.1140 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.440/SR1: 0.417 | LR: 1.00e-04 [2026-04-17 10:30:52] Epoch 1 | Step 6950 | Loss: 0.3206 | LM: 0.3091 | LB: 1.1140 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.440/SR1: 0.417 | LR: 1.00e-04 [2026-04-17 10:30:59] Epoch 1 | Step 6960 | Loss: 0.3206 | LM: 0.3092 | LB: 1.1140 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.440/SR1: 0.417 | LR: 1.00e-04 [2026-04-17 10:31:05] Epoch 1 | Step 6970 | Loss: 0.3206 | LM: 0.3091 | LB: 1.1139 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.440/SR1: 0.417 | LR: 1.00e-04 [2026-04-17 10:31:11] Epoch 1 | Step 6980 | Loss: 0.3205 | LM: 0.3090 | LB: 1.1139 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.440/SR1: 0.417 | LR: 1.00e-04 [2026-04-17 10:31:18] Epoch 1 | Step 6990 | Loss: 0.3205 | LM: 0.3089 | LB: 1.1139 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.440/SR1: 0.417 | LR: 1.00e-04 [2026-04-17 10:31:24] Epoch 1 | Step 7000 | Loss: 0.3204 | LM: 0.3089 | LB: 1.1138 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.440/SR1: 0.417 | LR: 1.00e-04 [2026-04-17 10:31:25] Validation | Batch 10/784 | Loss: 0.3273 | LM_LOSS: 0.3164 | LB_LOSS: 1.0907 [2026-04-17 10:31:27] Validation | Batch 20/784 | Loss: 0.3359 | LM_LOSS: 0.3249 | LB_LOSS: 1.0908 [2026-04-17 10:31:28] Validation | Batch 30/784 | Loss: 0.3214 | LM_LOSS: 0.3105 | LB_LOSS: 1.0900 [2026-04-17 10:31:30] Validation | Batch 40/784 | Loss: 0.3235 | LM_LOSS: 0.3126 | LB_LOSS: 1.0898 [2026-04-17 10:31:31] Validation | Batch 50/784 | Loss: 0.3197 | LM_LOSS: 0.3088 | LB_LOSS: 1.0891 [2026-04-17 10:31:32] Validation | Batch 60/784 | Loss: 0.3212 | LM_LOSS: 0.3103 | LB_LOSS: 1.0887 [2026-04-17 10:31:34] Validation | Batch 70/784 | Loss: 0.3186 | LM_LOSS: 0.3077 | LB_LOSS: 1.0880 [2026-04-17 10:31:35] Validation | Batch 80/784 | Loss: 0.3143 | LM_LOSS: 0.3034 | LB_LOSS: 1.0875 [2026-04-17 10:31:36] Validation | Batch 90/784 | Loss: 0.3127 | LM_LOSS: 0.3018 | LB_LOSS: 1.0881 [2026-04-17 10:31:38] Validation | Batch 100/784 | Loss: 0.3146 | LM_LOSS: 0.3037 | LB_LOSS: 1.0886 [2026-04-17 10:31:39] Validation | Batch 110/784 | Loss: 0.3101 | LM_LOSS: 0.2992 | LB_LOSS: 1.0887 [2026-04-17 10:31:40] Validation | Batch 120/784 | Loss: 0.3133 | LM_LOSS: 0.3024 | LB_LOSS: 1.0886 [2026-04-17 10:31:42] Validation | Batch 130/784 | Loss: 0.3157 | LM_LOSS: 0.3048 | LB_LOSS: 1.0886 [2026-04-17 10:31:43] Validation | Batch 140/784 | Loss: 0.3151 | LM_LOSS: 0.3042 | LB_LOSS: 1.0884 [2026-04-17 10:31:45] Validation | Batch 150/784 | Loss: 0.3110 | LM_LOSS: 0.3001 | LB_LOSS: 1.0887 [2026-04-17 10:31:46] Validation | Batch 160/784 | Loss: 0.3117 | LM_LOSS: 0.3008 | LB_LOSS: 1.0884 [2026-04-17 10:31:48] Validation | Batch 170/784 | Loss: 0.3123 | LM_LOSS: 0.3014 | LB_LOSS: 1.0881 [2026-04-17 10:31:49] Validation | Batch 180/784 | Loss: 0.3097 | LM_LOSS: 0.2988 | LB_LOSS: 1.0881 [2026-04-17 10:31:50] Validation | Batch 190/784 | Loss: 0.3113 | LM_LOSS: 0.3004 | LB_LOSS: 1.0885 [2026-04-17 10:31:51] Validation | Batch 200/784 | Loss: 0.3118 | LM_LOSS: 0.3009 | LB_LOSS: 1.0886 [2026-04-17 10:31:53] Validation | Batch 210/784 | Loss: 0.3108 | LM_LOSS: 0.2999 | LB_LOSS: 1.0885 [2026-04-17 10:31:54] Validation | Batch 220/784 | Loss: 0.3116 | LM_LOSS: 0.3007 | LB_LOSS: 1.0886 [2026-04-17 10:31:56] Validation | Batch 230/784 | Loss: 0.3119 | LM_LOSS: 0.3010 | LB_LOSS: 1.0885 [2026-04-17 10:31:57] Validation | Batch 240/784 | Loss: 0.3122 | LM_LOSS: 0.3013 | LB_LOSS: 1.0888 [2026-04-17 10:31:58] Validation | Batch 250/784 | Loss: 0.3121 | LM_LOSS: 0.3012 | LB_LOSS: 1.0887 [2026-04-17 10:32:00] Validation | Batch 260/784 | Loss: 0.3121 | LM_LOSS: 0.3012 | LB_LOSS: 1.0889 [2026-04-17 10:32:02] Validation | Batch 270/784 | Loss: 0.3115 | LM_LOSS: 0.3006 | LB_LOSS: 1.0889 [2026-04-17 10:32:03] Validation | Batch 280/784 | Loss: 0.3121 | LM_LOSS: 0.3012 | LB_LOSS: 1.0891 [2026-04-17 10:32:04] Validation | Batch 290/784 | Loss: 0.3131 | LM_LOSS: 0.3022 | LB_LOSS: 1.0892 [2026-04-17 10:32:05] Validation | Batch 300/784 | Loss: 0.3137 | LM_LOSS: 0.3028 | LB_LOSS: 1.0893 [2026-04-17 10:32:07] Validation | Batch 310/784 | Loss: 0.3132 | LM_LOSS: 0.3023 | LB_LOSS: 1.0892 [2026-04-17 10:32:08] Validation | Batch 320/784 | Loss: 0.3146 | LM_LOSS: 0.3038 | LB_LOSS: 1.0892 [2026-04-17 10:32:10] Validation | Batch 330/784 | Loss: 0.3145 | LM_LOSS: 0.3036 | LB_LOSS: 1.0892 [2026-04-17 10:32:11] Validation | Batch 340/784 | Loss: 0.3135 | LM_LOSS: 0.3026 | LB_LOSS: 1.0893 [2026-04-17 10:32:12] Validation | Batch 350/784 | Loss: 0.3136 | LM_LOSS: 0.3027 | LB_LOSS: 1.0895 [2026-04-17 10:32:13] Validation | Batch 360/784 | Loss: 0.3133 | LM_LOSS: 0.3024 | LB_LOSS: 1.0895 [2026-04-17 10:32:15] Validation | Batch 370/784 | Loss: 0.3138 | LM_LOSS: 0.3029 | LB_LOSS: 1.0894 [2026-04-17 10:32:16] Validation | Batch 380/784 | Loss: 0.3135 | LM_LOSS: 0.3026 | LB_LOSS: 1.0894 [2026-04-17 10:32:17] Validation | Batch 390/784 | Loss: 0.3134 | LM_LOSS: 0.3025 | LB_LOSS: 1.0895 [2026-04-17 10:32:19] Validation | Batch 400/784 | Loss: 0.3135 | LM_LOSS: 0.3026 | LB_LOSS: 1.0895 [2026-04-17 10:32:20] Validation | Batch 410/784 | Loss: 0.3138 | LM_LOSS: 0.3029 | LB_LOSS: 1.0895 [2026-04-17 10:32:21] Validation | Batch 420/784 | Loss: 0.3139 | LM_LOSS: 0.3030 | LB_LOSS: 1.0896 [2026-04-17 10:32:22] Validation | Batch 430/784 | Loss: 0.3138 | LM_LOSS: 0.3029 | LB_LOSS: 1.0895 [2026-04-17 10:32:23] Validation | Batch 440/784 | Loss: 0.3134 | LM_LOSS: 0.3025 | LB_LOSS: 1.0896 [2026-04-17 10:32:25] Validation | Batch 450/784 | Loss: 0.3129 | LM_LOSS: 0.3020 | LB_LOSS: 1.0895 [2026-04-17 10:32:26] Validation | Batch 460/784 | Loss: 0.3133 | LM_LOSS: 0.3024 | LB_LOSS: 1.0896 [2026-04-17 10:32:28] Validation | Batch 470/784 | Loss: 0.3125 | LM_LOSS: 0.3016 | LB_LOSS: 1.0896 [2026-04-17 10:32:29] Validation | Batch 480/784 | Loss: 0.3129 | LM_LOSS: 0.3020 | LB_LOSS: 1.0895 [2026-04-17 10:32:30] Validation | Batch 490/784 | Loss: 0.3123 | LM_LOSS: 0.3014 | LB_LOSS: 1.0894 [2026-04-17 10:32:32] Validation | Batch 500/784 | Loss: 0.3128 | LM_LOSS: 0.3019 | LB_LOSS: 1.0894 [2026-04-17 10:32:33] Validation | Batch 510/784 | Loss: 0.3125 | LM_LOSS: 0.3016 | LB_LOSS: 1.0894 [2026-04-17 10:32:34] Validation | Batch 520/784 | Loss: 0.3125 | LM_LOSS: 0.3016 | LB_LOSS: 1.0893 [2026-04-17 10:32:36] Validation | Batch 530/784 | Loss: 0.3133 | LM_LOSS: 0.3024 | LB_LOSS: 1.0892 [2026-04-17 10:32:37] Validation | Batch 540/784 | Loss: 0.3137 | LM_LOSS: 0.3028 | LB_LOSS: 1.0893 [2026-04-17 10:32:39] Validation | Batch 550/784 | Loss: 0.3151 | LM_LOSS: 0.3042 | LB_LOSS: 1.0892 [2026-04-17 10:32:40] Validation | Batch 560/784 | Loss: 0.3151 | LM_LOSS: 0.3042 | LB_LOSS: 1.0893 [2026-04-17 10:32:41] Validation | Batch 570/784 | Loss: 0.3147 | LM_LOSS: 0.3038 | LB_LOSS: 1.0892 [2026-04-17 10:32:43] Validation | Batch 580/784 | Loss: 0.3141 | LM_LOSS: 0.3032 | LB_LOSS: 1.0892 [2026-04-17 10:32:44] Validation | Batch 590/784 | Loss: 0.3144 | LM_LOSS: 0.3035 | LB_LOSS: 1.0892 [2026-04-17 10:32:45] Validation | Batch 600/784 | Loss: 0.3143 | LM_LOSS: 0.3034 | LB_LOSS: 1.0891 [2026-04-17 10:32:47] Validation | Batch 610/784 | Loss: 0.3144 | LM_LOSS: 0.3035 | LB_LOSS: 1.0891 [2026-04-17 10:32:48] Validation | Batch 620/784 | Loss: 0.3143 | LM_LOSS: 0.3034 | LB_LOSS: 1.0891 [2026-04-17 10:32:50] Validation | Batch 630/784 | Loss: 0.3149 | LM_LOSS: 0.3040 | LB_LOSS: 1.0891 [2026-04-17 10:32:51] Validation | Batch 640/784 | Loss: 0.3151 | LM_LOSS: 0.3042 | LB_LOSS: 1.0891 [2026-04-17 10:32:53] Validation | Batch 650/784 | Loss: 0.3150 | LM_LOSS: 0.3041 | LB_LOSS: 1.0892 [2026-04-17 10:32:54] Validation | Batch 660/784 | Loss: 0.3153 | LM_LOSS: 0.3044 | LB_LOSS: 1.0891 [2026-04-17 10:32:56] Validation | Batch 670/784 | Loss: 0.3156 | LM_LOSS: 0.3048 | LB_LOSS: 1.0892 [2026-04-17 10:32:57] Validation | Batch 680/784 | Loss: 0.3154 | LM_LOSS: 0.3045 | LB_LOSS: 1.0892 [2026-04-17 10:32:58] Validation | Batch 690/784 | Loss: 0.3156 | LM_LOSS: 0.3047 | LB_LOSS: 1.0891 [2026-04-17 10:33:00] Validation | Batch 700/784 | Loss: 0.3156 | LM_LOSS: 0.3047 | LB_LOSS: 1.0891 [2026-04-17 10:33:01] Validation | Batch 710/784 | Loss: 0.3154 | LM_LOSS: 0.3045 | LB_LOSS: 1.0890 [2026-04-17 10:33:03] Validation | Batch 720/784 | Loss: 0.3151 | LM_LOSS: 0.3042 | LB_LOSS: 1.0889 [2026-04-17 10:33:04] Validation | Batch 730/784 | Loss: 0.3146 | LM_LOSS: 0.3037 | LB_LOSS: 1.0889 [2026-04-17 10:33:05] Validation | Batch 740/784 | Loss: 0.3147 | LM_LOSS: 0.3038 | LB_LOSS: 1.0890 [2026-04-17 10:33:06] Validation | Batch 750/784 | Loss: 0.3140 | LM_LOSS: 0.3031 | LB_LOSS: 1.0890 [2026-04-17 10:33:08] Validation | Batch 760/784 | Loss: 0.3142 | LM_LOSS: 0.3033 | LB_LOSS: 1.0890 [2026-04-17 10:33:09] Validation | Batch 770/784 | Loss: 0.3144 | LM_LOSS: 0.3035 | LB_LOSS: 1.0891 [2026-04-17 10:33:10] Validation | Batch 780/784 | Loss: 0.3146 | LM_LOSS: 0.3037 | LB_LOSS: 1.0890 [2026-04-17 10:33:11] Validation | Batch 784/784 | Loss: 0.3148 | LM_LOSS: 0.3039 | LB_LOSS: 1.0890 [2026-04-17 10:33:14] Validation | Loss: 0.3148 | LM_LOSS: 0.3039 | LB_LOSS: 1.0890 | PPL: 1.36 | Time: 106.74s [2026-04-17 10:33:17] New best model saved! Val loss: 0.3148 [2026-04-17 10:33:24] Epoch 1 | Step 7010 | Loss: 0.3204 | LM: 0.3088 | LB: 1.1138 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.440/SR1: 0.417 | LR: 1.00e-04 [2026-04-17 10:33:30] Epoch 1 | Step 7020 | Loss: 0.3204 | LM: 0.3090 | LB: 1.1138 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.440/SR1: 0.417 | LR: 1.00e-04 [2026-04-17 10:33:37] Epoch 1 | Step 7030 | Loss: 0.3205 | LM: 0.3090 | LB: 1.1138 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.440/SR1: 0.417 | LR: 1.00e-04 [2026-04-17 10:33:43] Epoch 1 | Step 7040 | Loss: 0.3205 | LM: 0.3091 | LB: 1.1137 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.440/SR1: 0.417 | LR: 1.00e-04 [2026-04-17 10:33:50] Epoch 1 | Step 7050 | Loss: 0.3205 | LM: 0.3091 | LB: 1.1137 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.440/SR1: 0.417 | LR: 1.00e-04 [2026-04-17 10:33:56] Epoch 1 | Step 7060 | Loss: 0.3205 | LM: 0.3091 | LB: 1.1137 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.440/SR1: 0.417 | LR: 1.00e-04 [2026-04-17 10:34:02] Epoch 1 | Step 7070 | Loss: 0.3205 | LM: 0.3091 | LB: 1.1136 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.440/SR1: 0.417 | LR: 1.00e-04 [2026-04-17 10:34:09] Epoch 1 | Step 7080 | Loss: 0.3205 | LM: 0.3091 | LB: 1.1136 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.440/SR1: 0.417 | LR: 1.00e-04 [2026-04-17 10:34:15] Epoch 1 | Step 7090 | Loss: 0.3205 | LM: 0.3090 | LB: 1.1136 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.440/SR1: 0.417 | LR: 1.00e-04 [2026-04-17 10:34:21] Epoch 1 | Step 7100 | Loss: 0.3204 | LM: 0.3089 | LB: 1.1135 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.440/SR1: 0.417 | LR: 1.00e-04 [2026-04-17 10:34:27] Epoch 1 | Step 7110 | Loss: 0.3203 | LM: 0.3088 | LB: 1.1135 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.440/SR1: 0.417 | LR: 1.00e-04 [2026-04-17 10:34:34] Epoch 1 | Step 7120 | Loss: 0.3203 | LM: 0.3089 | LB: 1.1135 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.440/SR1: 0.417 | LR: 1.00e-04 [2026-04-17 10:34:40] Epoch 1 | Step 7130 | Loss: 0.3203 | LM: 0.3089 | LB: 1.1134 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.440/SR1: 0.417 | LR: 1.00e-04 [2026-04-17 10:34:46] Epoch 1 | Step 7140 | Loss: 0.3203 | LM: 0.3088 | LB: 1.1134 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.440/SR1: 0.417 | LR: 1.00e-04 [2026-04-17 10:34:52] Epoch 1 | Step 7150 | Loss: 0.3203 | LM: 0.3088 | LB: 1.1134 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.440/SR1: 0.417 | LR: 1.00e-04 [2026-04-17 10:34:59] Epoch 1 | Step 7160 | Loss: 0.3203 | LM: 0.3088 | LB: 1.1133 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.439/SR1: 0.417 | LR: 1.00e-04 [2026-04-17 10:35:05] Epoch 1 | Step 7170 | Loss: 0.3202 | LM: 0.3086 | LB: 1.1133 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.439/SR1: 0.417 | LR: 1.00e-04 [2026-04-17 10:35:12] Epoch 1 | Step 7180 | Loss: 0.3202 | LM: 0.3087 | LB: 1.1133 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.439/SR1: 0.417 | LR: 1.00e-04 [2026-04-17 10:35:18] Epoch 1 | Step 7190 | Loss: 0.3202 | LM: 0.3087 | LB: 1.1132 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.439/SR1: 0.416 | LR: 1.00e-04 [2026-04-17 10:35:24] Epoch 1 | Step 7200 | Loss: 0.3201 | LM: 0.3087 | LB: 1.1132 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.439/SR1: 0.416 | LR: 1.00e-04 [2026-04-17 10:35:30] Epoch 1 | Step 7210 | Loss: 0.3201 | LM: 0.3086 | LB: 1.1132 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.439/SR1: 0.416 | LR: 1.00e-04 [2026-04-17 10:35:36] Epoch 1 | Step 7220 | Loss: 0.3201 | LM: 0.3086 | LB: 1.1132 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.439/SR1: 0.416 | LR: 1.00e-04 [2026-04-17 10:35:42] Epoch 1 | Step 7230 | Loss: 0.3201 | LM: 0.3085 | LB: 1.1131 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.439/SR1: 0.416 | LR: 1.00e-04 [2026-04-17 10:35:49] Epoch 1 | Step 7240 | Loss: 0.3201 | LM: 0.3086 | LB: 1.1131 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.439/SR1: 0.416 | LR: 1.00e-04 [2026-04-17 10:35:55] Epoch 1 | Step 7250 | Loss: 0.3201 | LM: 0.3085 | LB: 1.1131 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.439/SR1: 0.416 | LR: 1.00e-04 [2026-04-17 10:36:01] Epoch 1 | Step 7260 | Loss: 0.3201 | LM: 0.3086 | LB: 1.1131 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.439/SR1: 0.416 | LR: 1.00e-04 [2026-04-17 10:36:08] Epoch 1 | Step 7270 | Loss: 0.3201 | LM: 0.3086 | LB: 1.1130 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.439/SR1: 0.416 | LR: 1.00e-04 [2026-04-17 10:36:14] Epoch 1 | Step 7280 | Loss: 0.3201 | LM: 0.3086 | LB: 1.1130 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.439/SR1: 0.416 | LR: 1.00e-04 [2026-04-17 10:36:20] Epoch 1 | Step 7290 | Loss: 0.3200 | LM: 0.3085 | LB: 1.1129 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.439/SR1: 0.416 | LR: 1.00e-04 [2026-04-17 10:36:26] Epoch 1 | Step 7300 | Loss: 0.3200 | LM: 0.3085 | LB: 1.1129 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.439/SR1: 0.416 | LR: 1.00e-04 [2026-04-17 10:36:33] Epoch 1 | Step 7310 | Loss: 0.3200 | LM: 0.3086 | LB: 1.1129 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.439/SR1: 0.416 | LR: 1.00e-04 [2026-04-17 10:36:39] Epoch 1 | Step 7320 | Loss: 0.3200 | LM: 0.3085 | LB: 1.1129 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.439/SR1: 0.416 | LR: 1.00e-04 [2026-04-17 10:36:45] Epoch 1 | Step 7330 | Loss: 0.3199 | LM: 0.3084 | LB: 1.1128 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.439/SR1: 0.416 | LR: 1.00e-04 [2026-04-17 10:36:51] Epoch 1 | Step 7340 | Loss: 0.3199 | LM: 0.3083 | LB: 1.1128 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.439/SR1: 0.416 | LR: 1.00e-04 [2026-04-17 10:36:58] Epoch 1 | Step 7350 | Loss: 0.3199 | LM: 0.3083 | LB: 1.1128 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.439/SR1: 0.416 | LR: 1.00e-04 [2026-04-17 10:37:04] Epoch 1 | Step 7360 | Loss: 0.3199 | LM: 0.3084 | LB: 1.1128 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.439/SR1: 0.416 | LR: 1.00e-04 [2026-04-17 10:37:10] Epoch 1 | Step 7370 | Loss: 0.3199 | LM: 0.3085 | LB: 1.1127 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.439/SR1: 0.416 | LR: 1.00e-04 [2026-04-17 10:37:16] Epoch 1 | Step 7380 | Loss: 0.3199 | LM: 0.3086 | LB: 1.1127 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.439/SR1: 0.416 | LR: 1.00e-04 [2026-04-17 10:37:22] Epoch 1 | Step 7390 | Loss: 0.3199 | LM: 0.3085 | LB: 1.1127 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.439/SR1: 0.416 | LR: 1.00e-04 [2026-04-17 10:37:29] Epoch 1 | Step 7400 | Loss: 0.3198 | LM: 0.3086 | LB: 1.1126 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.439/SR1: 0.416 | LR: 1.00e-04 [2026-04-17 10:37:35] Epoch 1 | Step 7410 | Loss: 0.3198 | LM: 0.3085 | LB: 1.1126 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.439/SR1: 0.416 | LR: 1.00e-04 [2026-04-17 10:37:41] Epoch 1 | Step 7420 | Loss: 0.3198 | LM: 0.3085 | LB: 1.1126 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.439/SR1: 0.416 | LR: 1.00e-04 [2026-04-17 10:37:47] Epoch 1 | Step 7430 | Loss: 0.3197 | LM: 0.3084 | LB: 1.1126 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.439/SR1: 0.416 | LR: 1.00e-04 [2026-04-17 10:37:54] Epoch 1 | Step 7440 | Loss: 0.3197 | LM: 0.3085 | LB: 1.1125 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.439/SR1: 0.416 | LR: 1.00e-04 [2026-04-17 10:38:00] Epoch 1 | Step 7450 | Loss: 0.3196 | LM: 0.3086 | LB: 1.1125 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.439/SR1: 0.416 | LR: 1.00e-04 [2026-04-17 10:38:07] Epoch 1 | Step 7460 | Loss: 0.3196 | LM: 0.3085 | LB: 1.1125 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.439/SR1: 0.416 | LR: 1.00e-04 [2026-04-17 10:38:13] Epoch 1 | Step 7470 | Loss: 0.3196 | LM: 0.3084 | LB: 1.1124 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.439/SR1: 0.416 | LR: 1.00e-04 [2026-04-17 10:38:19] Epoch 1 | Step 7480 | Loss: 0.3196 | LM: 0.3084 | LB: 1.1124 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.439/SR1: 0.416 | LR: 1.00e-04 [2026-04-17 10:38:26] Epoch 1 | Step 7490 | Loss: 0.3195 | LM: 0.3083 | LB: 1.1124 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.439/SR1: 0.416 | LR: 1.00e-04 [2026-04-17 10:38:32] Epoch 1 | Step 7500 | Loss: 0.3195 | LM: 0.3083 | LB: 1.1123 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.439/SR1: 0.416 | LR: 1.00e-04 [2026-04-17 10:38:38] Epoch 1 | Step 7510 | Loss: 0.3195 | LM: 0.3082 | LB: 1.1123 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.439/SR1: 0.415 | LR: 1.00e-04 [2026-04-17 10:38:45] Epoch 1 | Step 7520 | Loss: 0.3196 | LM: 0.3082 | LB: 1.1123 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.439/SR1: 0.415 | LR: 1.00e-04 [2026-04-17 10:38:51] Epoch 1 | Step 7530 | Loss: 0.3195 | LM: 0.3082 | LB: 1.1123 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.439/SR1: 0.415 | LR: 1.00e-04 [2026-04-17 10:38:58] Epoch 1 | Step 7540 | Loss: 0.3195 | LM: 0.3081 | LB: 1.1122 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.439/SR1: 0.415 | LR: 1.00e-04 [2026-04-17 10:39:04] Epoch 1 | Step 7550 | Loss: 0.3195 | LM: 0.3079 | LB: 1.1122 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.439/SR1: 0.415 | LR: 1.00e-04 [2026-04-17 10:39:10] Epoch 1 | Step 7560 | Loss: 0.3195 | LM: 0.3079 | LB: 1.1122 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.439/SR1: 0.415 | LR: 1.00e-04 [2026-04-17 10:39:17] Epoch 1 | Step 7570 | Loss: 0.3195 | LM: 0.3079 | LB: 1.1121 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.438/SR1: 0.415 | LR: 1.00e-04 [2026-04-17 10:39:23] Epoch 1 | Step 7580 | Loss: 0.3195 | LM: 0.3079 | LB: 1.1121 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.438/SR1: 0.415 | LR: 1.00e-04 [2026-04-17 10:39:30] Epoch 1 | Step 7590 | Loss: 0.3195 | LM: 0.3080 | LB: 1.1121 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.438/SR1: 0.415 | LR: 1.00e-04 [2026-04-17 10:39:36] Epoch 1 | Step 7600 | Loss: 0.3195 | LM: 0.3080 | LB: 1.1120 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.438/SR1: 0.415 | LR: 1.00e-04 [2026-04-17 10:39:42] Epoch 1 | Step 7610 | Loss: 0.3195 | LM: 0.3080 | LB: 1.1120 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.438/SR1: 0.415 | LR: 1.00e-04 [2026-04-17 10:39:49] Epoch 1 | Step 7620 | Loss: 0.3194 | LM: 0.3079 | LB: 1.1120 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.438/SR1: 0.415 | LR: 1.00e-04 [2026-04-17 10:39:55] Epoch 1 | Step 7630 | Loss: 0.3195 | LM: 0.3080 | LB: 1.1119 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.438/SR1: 0.415 | LR: 1.00e-04 [2026-04-17 10:40:01] Epoch 1 | Step 7640 | Loss: 0.3194 | LM: 0.3080 | LB: 1.1119 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.438/SR1: 0.415 | LR: 1.00e-04 [2026-04-17 10:40:08] Epoch 1 | Step 7650 | Loss: 0.3193 | LM: 0.3079 | LB: 1.1119 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.438/SR1: 0.415 | LR: 1.00e-04 [2026-04-17 10:40:14] Epoch 1 | Step 7660 | Loss: 0.3194 | LM: 0.3079 | LB: 1.1119 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.438/SR1: 0.415 | LR: 1.00e-04 [2026-04-17 10:40:21] Epoch 1 | Step 7670 | Loss: 0.3193 | LM: 0.3079 | LB: 1.1118 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.438/SR1: 0.415 | LR: 1.00e-04 [2026-04-17 10:40:27] Epoch 1 | Step 7680 | Loss: 0.3193 | LM: 0.3079 | LB: 1.1118 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.438/SR1: 0.415 | LR: 1.00e-04 [2026-04-17 10:40:33] Epoch 1 | Step 7690 | Loss: 0.3194 | LM: 0.3080 | LB: 1.1118 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.438/SR1: 0.415 | LR: 1.00e-04 [2026-04-17 10:40:40] Epoch 1 | Step 7700 | Loss: 0.3193 | LM: 0.3079 | LB: 1.1117 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.438/SR1: 0.415 | LR: 1.00e-04 [2026-04-17 10:40:46] Epoch 1 | Step 7710 | Loss: 0.3193 | LM: 0.3078 | LB: 1.1117 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.438/SR1: 0.415 | LR: 1.00e-04 [2026-04-17 10:40:52] Epoch 1 | Step 7720 | Loss: 0.3193 | LM: 0.3078 | LB: 1.1117 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.438/SR1: 0.415 | LR: 1.00e-04 [2026-04-17 10:40:59] Epoch 1 | Step 7730 | Loss: 0.3192 | LM: 0.3077 | LB: 1.1116 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.438/SR1: 0.415 | LR: 1.00e-04 [2026-04-17 10:41:05] Epoch 1 | Step 7740 | Loss: 0.3192 | LM: 0.3076 | LB: 1.1116 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.438/SR1: 0.415 | LR: 1.00e-04 [2026-04-17 10:41:11] Epoch 1 | Step 7750 | Loss: 0.3192 | LM: 0.3077 | LB: 1.1116 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.438/SR1: 0.415 | LR: 1.00e-04 [2026-04-17 10:41:18] Epoch 1 | Step 7760 | Loss: 0.3192 | LM: 0.3077 | LB: 1.1116 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.438/SR1: 0.415 | LR: 1.00e-04 [2026-04-17 10:41:24] Epoch 1 | Step 7770 | Loss: 0.3191 | LM: 0.3076 | LB: 1.1115 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.438/SR1: 0.415 | LR: 1.00e-04 [2026-04-17 10:41:30] Epoch 1 | Step 7780 | Loss: 0.3192 | LM: 0.3076 | LB: 1.1115 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.438/SR1: 0.415 | LR: 1.00e-04 [2026-04-17 10:41:37] Epoch 1 | Step 7790 | Loss: 0.3192 | LM: 0.3075 | LB: 1.1115 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.438/SR1: 0.415 | LR: 1.00e-04 [2026-04-17 10:41:43] Epoch 1 | Step 7800 | Loss: 0.3192 | LM: 0.3077 | LB: 1.1114 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.438/SR1: 0.414 | LR: 1.00e-04 [2026-04-17 10:41:49] Epoch 1 | Step 7810 | Loss: 0.3192 | LM: 0.3077 | LB: 1.1114 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.438/SR1: 0.414 | LR: 1.00e-04 [2026-04-17 10:41:56] Epoch 1 | Step 7820 | Loss: 0.3192 | LM: 0.3076 | LB: 1.1114 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.438/SR1: 0.414 | LR: 1.00e-04 [2026-04-17 10:42:02] Epoch 1 | Step 7830 | Loss: 0.3192 | LM: 0.3075 | LB: 1.1114 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.438/SR1: 0.414 | LR: 1.00e-04 [2026-04-17 10:42:08] Epoch 1 | Step 7840 | Loss: 0.3192 | LM: 0.3075 | LB: 1.1113 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.438/SR1: 0.414 | LR: 1.00e-04 [2026-04-17 10:42:15] Epoch 1 | Step 7850 | Loss: 0.3192 | LM: 0.3074 | LB: 1.1113 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.438/SR1: 0.414 | LR: 1.00e-04 [2026-04-17 10:42:21] Epoch 1 | Step 7860 | Loss: 0.3192 | LM: 0.3073 | LB: 1.1113 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.438/SR1: 0.414 | LR: 1.00e-04 [2026-04-17 10:42:28] Epoch 1 | Step 7870 | Loss: 0.3192 | LM: 0.3073 | LB: 1.1112 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.438/SR1: 0.414 | LR: 1.00e-04 [2026-04-17 10:42:34] Epoch 1 | Step 7880 | Loss: 0.3192 | LM: 0.3073 | LB: 1.1112 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.438/SR1: 0.414 | LR: 1.00e-04 [2026-04-17 10:42:40] Epoch 1 | Step 7890 | Loss: 0.3192 | LM: 0.3073 | LB: 1.1112 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.438/SR1: 0.414 | LR: 1.00e-04 [2026-04-17 10:42:47] Epoch 1 | Step 7900 | Loss: 0.3191 | LM: 0.3073 | LB: 1.1112 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.438/SR1: 0.414 | LR: 1.00e-04 [2026-04-17 10:42:53] Epoch 1 | Step 7910 | Loss: 0.3191 | LM: 0.3072 | LB: 1.1111 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.438/SR1: 0.414 | LR: 1.00e-04 [2026-04-17 10:43:00] Epoch 1 | Step 7920 | Loss: 0.3191 | LM: 0.3071 | LB: 1.1111 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.437/SR1: 0.414 | LR: 1.00e-04 [2026-04-17 10:43:06] Epoch 1 | Step 7930 | Loss: 0.3191 | LM: 0.3070 | LB: 1.1111 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.437/SR1: 0.414 | LR: 1.00e-04 [2026-04-17 10:43:12] Epoch 1 | Step 7940 | Loss: 0.3190 | LM: 0.3069 | LB: 1.1110 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.437/SR1: 0.414 | LR: 1.00e-04 [2026-04-17 10:43:19] Epoch 1 | Step 7950 | Loss: 0.3190 | LM: 0.3069 | LB: 1.1110 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.437/SR1: 0.414 | LR: 1.00e-04 [2026-04-17 10:43:25] Epoch 1 | Step 7960 | Loss: 0.3190 | LM: 0.3069 | LB: 1.1110 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.437/SR1: 0.414 | LR: 1.00e-04 [2026-04-17 10:43:31] Epoch 1 | Step 7970 | Loss: 0.3191 | LM: 0.3069 | LB: 1.1110 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.437/SR1: 0.414 | LR: 1.00e-04 [2026-04-17 10:43:38] Epoch 1 | Step 7980 | Loss: 0.3190 | LM: 0.3069 | LB: 1.1109 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.437/SR1: 0.414 | LR: 1.00e-04 [2026-04-17 10:43:44] Epoch 1 | Step 7990 | Loss: 0.3190 | LM: 0.3068 | LB: 1.1109 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.437/SR1: 0.414 | LR: 1.00e-04 [2026-04-17 10:43:51] Epoch 1 | Step 8000 | Loss: 0.3189 | LM: 0.3067 | LB: 1.1109 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.437/SR1: 0.414 | LR: 1.00e-04 [2026-04-17 10:43:52] Validation | Batch 10/784 | Loss: 0.3275 | LM_LOSS: 0.3167 | LB_LOSS: 1.0853 [2026-04-17 10:43:54] Validation | Batch 20/784 | Loss: 0.3384 | LM_LOSS: 0.3275 | LB_LOSS: 1.0856 [2026-04-17 10:43:55] Validation | Batch 30/784 | Loss: 0.3238 | LM_LOSS: 0.3130 | LB_LOSS: 1.0848 [2026-04-17 10:43:56] Validation | Batch 40/784 | Loss: 0.3249 | LM_LOSS: 0.3141 | LB_LOSS: 1.0847 [2026-04-17 10:43:58] Validation | Batch 50/784 | Loss: 0.3209 | LM_LOSS: 0.3100 | LB_LOSS: 1.0841 [2026-04-17 10:43:59] Validation | Batch 60/784 | Loss: 0.3215 | LM_LOSS: 0.3107 | LB_LOSS: 1.0836 [2026-04-17 10:44:00] Validation | Batch 70/784 | Loss: 0.3186 | LM_LOSS: 0.3077 | LB_LOSS: 1.0830 [2026-04-17 10:44:01] Validation | Batch 80/784 | Loss: 0.3143 | LM_LOSS: 0.3035 | LB_LOSS: 1.0825 [2026-04-17 10:44:03] Validation | Batch 90/784 | Loss: 0.3124 | LM_LOSS: 0.3016 | LB_LOSS: 1.0830 [2026-04-17 10:44:04] Validation | Batch 100/784 | Loss: 0.3140 | LM_LOSS: 0.3032 | LB_LOSS: 1.0835 [2026-04-17 10:44:06] Validation | Batch 110/784 | Loss: 0.3094 | LM_LOSS: 0.2985 | LB_LOSS: 1.0836 [2026-04-17 10:44:07] Validation | Batch 120/784 | Loss: 0.3125 | LM_LOSS: 0.3016 | LB_LOSS: 1.0835 [2026-04-17 10:44:08] Validation | Batch 130/784 | Loss: 0.3149 | LM_LOSS: 0.3041 | LB_LOSS: 1.0835 [2026-04-17 10:44:10] Validation | Batch 140/784 | Loss: 0.3143 | LM_LOSS: 0.3035 | LB_LOSS: 1.0833 [2026-04-17 10:44:11] Validation | Batch 150/784 | Loss: 0.3104 | LM_LOSS: 0.2996 | LB_LOSS: 1.0836 [2026-04-17 10:44:12] Validation | Batch 160/784 | Loss: 0.3112 | LM_LOSS: 0.3004 | LB_LOSS: 1.0832 [2026-04-17 10:44:14] Validation | Batch 170/784 | Loss: 0.3118 | LM_LOSS: 0.3010 | LB_LOSS: 1.0829 [2026-04-17 10:44:15] Validation | Batch 180/784 | Loss: 0.3093 | LM_LOSS: 0.2985 | LB_LOSS: 1.0830 [2026-04-17 10:44:17] Validation | Batch 190/784 | Loss: 0.3110 | LM_LOSS: 0.3001 | LB_LOSS: 1.0834 [2026-04-17 10:44:18] Validation | Batch 200/784 | Loss: 0.3113 | LM_LOSS: 0.3005 | LB_LOSS: 1.0835 [2026-04-17 10:44:19] Validation | Batch 210/784 | Loss: 0.3102 | LM_LOSS: 0.2994 | LB_LOSS: 1.0834 [2026-04-17 10:44:21] Validation | Batch 220/784 | Loss: 0.3110 | LM_LOSS: 0.3001 | LB_LOSS: 1.0834 [2026-04-17 10:44:22] Validation | Batch 230/784 | Loss: 0.3113 | LM_LOSS: 0.3005 | LB_LOSS: 1.0833 [2026-04-17 10:44:23] Validation | Batch 240/784 | Loss: 0.3116 | LM_LOSS: 0.3008 | LB_LOSS: 1.0837 [2026-04-17 10:44:25] Validation | Batch 250/784 | Loss: 0.3115 | LM_LOSS: 0.3007 | LB_LOSS: 1.0835 [2026-04-17 10:44:26] Validation | Batch 260/784 | Loss: 0.3116 | LM_LOSS: 0.3007 | LB_LOSS: 1.0837 [2026-04-17 10:44:28] Validation | Batch 270/784 | Loss: 0.3111 | LM_LOSS: 0.3002 | LB_LOSS: 1.0838 [2026-04-17 10:44:29] Validation | Batch 280/784 | Loss: 0.3117 | LM_LOSS: 0.3009 | LB_LOSS: 1.0839 [2026-04-17 10:44:30] Validation | Batch 290/784 | Loss: 0.3126 | LM_LOSS: 0.3017 | LB_LOSS: 1.0840 [2026-04-17 10:44:32] Validation | Batch 300/784 | Loss: 0.3132 | LM_LOSS: 0.3023 | LB_LOSS: 1.0841 [2026-04-17 10:44:33] Validation | Batch 310/784 | Loss: 0.3127 | LM_LOSS: 0.3019 | LB_LOSS: 1.0840 [2026-04-17 10:44:35] Validation | Batch 320/784 | Loss: 0.3141 | LM_LOSS: 0.3033 | LB_LOSS: 1.0840 [2026-04-17 10:44:36] Validation | Batch 330/784 | Loss: 0.3140 | LM_LOSS: 0.3032 | LB_LOSS: 1.0840 [2026-04-17 10:44:37] Validation | Batch 340/784 | Loss: 0.3130 | LM_LOSS: 0.3022 | LB_LOSS: 1.0841 [2026-04-17 10:44:38] Validation | Batch 350/784 | Loss: 0.3130 | LM_LOSS: 0.3022 | LB_LOSS: 1.0843 [2026-04-17 10:44:40] Validation | Batch 360/784 | Loss: 0.3128 | LM_LOSS: 0.3019 | LB_LOSS: 1.0843 [2026-04-17 10:44:41] Validation | Batch 370/784 | Loss: 0.3131 | LM_LOSS: 0.3023 | LB_LOSS: 1.0842 [2026-04-17 10:44:42] Validation | Batch 380/784 | Loss: 0.3129 | LM_LOSS: 0.3021 | LB_LOSS: 1.0843 [2026-04-17 10:44:44] Validation | Batch 390/784 | Loss: 0.3127 | LM_LOSS: 0.3019 | LB_LOSS: 1.0843 [2026-04-17 10:44:45] Validation | Batch 400/784 | Loss: 0.3128 | LM_LOSS: 0.3020 | LB_LOSS: 1.0843 [2026-04-17 10:44:46] Validation | Batch 410/784 | Loss: 0.3130 | LM_LOSS: 0.3022 | LB_LOSS: 1.0844 [2026-04-17 10:44:47] Validation | Batch 420/784 | Loss: 0.3132 | LM_LOSS: 0.3023 | LB_LOSS: 1.0844 [2026-04-17 10:44:49] Validation | Batch 430/784 | Loss: 0.3131 | LM_LOSS: 0.3023 | LB_LOSS: 1.0843 [2026-04-17 10:44:50] Validation | Batch 440/784 | Loss: 0.3127 | LM_LOSS: 0.3019 | LB_LOSS: 1.0844 [2026-04-17 10:44:51] Validation | Batch 450/784 | Loss: 0.3122 | LM_LOSS: 0.3013 | LB_LOSS: 1.0843 [2026-04-17 10:44:52] Validation | Batch 460/784 | Loss: 0.3126 | LM_LOSS: 0.3018 | LB_LOSS: 1.0844 [2026-04-17 10:44:54] Validation | Batch 470/784 | Loss: 0.3119 | LM_LOSS: 0.3011 | LB_LOSS: 1.0844 [2026-04-17 10:44:55] Validation | Batch 480/784 | Loss: 0.3123 | LM_LOSS: 0.3015 | LB_LOSS: 1.0843 [2026-04-17 10:44:57] Validation | Batch 490/784 | Loss: 0.3117 | LM_LOSS: 0.3009 | LB_LOSS: 1.0843 [2026-04-17 10:44:58] Validation | Batch 500/784 | Loss: 0.3122 | LM_LOSS: 0.3014 | LB_LOSS: 1.0842 [2026-04-17 10:44:59] Validation | Batch 510/784 | Loss: 0.3118 | LM_LOSS: 0.3010 | LB_LOSS: 1.0842 [2026-04-17 10:45:01] Validation | Batch 520/784 | Loss: 0.3119 | LM_LOSS: 0.3011 | LB_LOSS: 1.0841 [2026-04-17 10:45:02] Validation | Batch 530/784 | Loss: 0.3127 | LM_LOSS: 0.3019 | LB_LOSS: 1.0841 [2026-04-17 10:45:03] Validation | Batch 540/784 | Loss: 0.3131 | LM_LOSS: 0.3023 | LB_LOSS: 1.0841 [2026-04-17 10:45:05] Validation | Batch 550/784 | Loss: 0.3145 | LM_LOSS: 0.3036 | LB_LOSS: 1.0841 [2026-04-17 10:45:06] Validation | Batch 560/784 | Loss: 0.3145 | LM_LOSS: 0.3037 | LB_LOSS: 1.0841 [2026-04-17 10:45:08] Validation | Batch 570/784 | Loss: 0.3141 | LM_LOSS: 0.3032 | LB_LOSS: 1.0840 [2026-04-17 10:45:09] Validation | Batch 580/784 | Loss: 0.3135 | LM_LOSS: 0.3027 | LB_LOSS: 1.0841 [2026-04-17 10:45:10] Validation | Batch 590/784 | Loss: 0.3139 | LM_LOSS: 0.3030 | LB_LOSS: 1.0840 [2026-04-17 10:45:12] Validation | Batch 600/784 | Loss: 0.3138 | LM_LOSS: 0.3029 | LB_LOSS: 1.0840 [2026-04-17 10:45:13] Validation | Batch 610/784 | Loss: 0.3138 | LM_LOSS: 0.3030 | LB_LOSS: 1.0840 [2026-04-17 10:45:14] Validation | Batch 620/784 | Loss: 0.3138 | LM_LOSS: 0.3029 | LB_LOSS: 1.0839 [2026-04-17 10:45:16] Validation | Batch 630/784 | Loss: 0.3144 | LM_LOSS: 0.3035 | LB_LOSS: 1.0840 [2026-04-17 10:45:17] Validation | Batch 640/784 | Loss: 0.3145 | LM_LOSS: 0.3037 | LB_LOSS: 1.0839 [2026-04-17 10:45:19] Validation | Batch 650/784 | Loss: 0.3144 | LM_LOSS: 0.3036 | LB_LOSS: 1.0840 [2026-04-17 10:45:20] Validation | Batch 660/784 | Loss: 0.3147 | LM_LOSS: 0.3038 | LB_LOSS: 1.0840 [2026-04-17 10:45:22] Validation | Batch 670/784 | Loss: 0.3151 | LM_LOSS: 0.3042 | LB_LOSS: 1.0841 [2026-04-17 10:45:23] Validation | Batch 680/784 | Loss: 0.3148 | LM_LOSS: 0.3040 | LB_LOSS: 1.0841 [2026-04-17 10:45:24] Validation | Batch 690/784 | Loss: 0.3150 | LM_LOSS: 0.3042 | LB_LOSS: 1.0840 [2026-04-17 10:45:26] Validation | Batch 700/784 | Loss: 0.3151 | LM_LOSS: 0.3042 | LB_LOSS: 1.0840 [2026-04-17 10:45:27] Validation | Batch 710/784 | Loss: 0.3148 | LM_LOSS: 0.3040 | LB_LOSS: 1.0839 [2026-04-17 10:45:29] Validation | Batch 720/784 | Loss: 0.3145 | LM_LOSS: 0.3037 | LB_LOSS: 1.0838 [2026-04-17 10:45:30] Validation | Batch 730/784 | Loss: 0.3141 | LM_LOSS: 0.3032 | LB_LOSS: 1.0838 [2026-04-17 10:45:31] Validation | Batch 740/784 | Loss: 0.3141 | LM_LOSS: 0.3033 | LB_LOSS: 1.0839 [2026-04-17 10:45:32] Validation | Batch 750/784 | Loss: 0.3135 | LM_LOSS: 0.3026 | LB_LOSS: 1.0839 [2026-04-17 10:45:34] Validation | Batch 760/784 | Loss: 0.3136 | LM_LOSS: 0.3028 | LB_LOSS: 1.0839 [2026-04-17 10:45:35] Validation | Batch 770/784 | Loss: 0.3138 | LM_LOSS: 0.3029 | LB_LOSS: 1.0839 [2026-04-17 10:45:36] Validation | Batch 780/784 | Loss: 0.3140 | LM_LOSS: 0.3032 | LB_LOSS: 1.0839 [2026-04-17 10:45:37] Validation | Batch 784/784 | Loss: 0.3142 | LM_LOSS: 0.3034 | LB_LOSS: 1.0839 [2026-04-17 10:45:40] Validation | Loss: 0.3142 | LM_LOSS: 0.3034 | LB_LOSS: 1.0839 | PPL: 1.36 | Time: 106.09s [2026-04-17 10:45:44] New best model saved! Val loss: 0.3142 [2026-04-17 10:45:51] Epoch 1 | Step 8010 | Loss: 0.3189 | LM: 0.3066 | LB: 1.1108 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.437/SR1: 0.414 | LR: 1.00e-04 [2026-04-17 10:45:57] Epoch 1 | Step 8020 | Loss: 0.3189 | LM: 0.3066 | LB: 1.1108 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.437/SR1: 0.414 | LR: 1.00e-04 [2026-04-17 10:46:03] Epoch 1 | Step 8030 | Loss: 0.3189 | LM: 0.3067 | LB: 1.1108 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.437/SR1: 0.414 | LR: 1.00e-04 [2026-04-17 10:46:10] Epoch 1 | Step 8040 | Loss: 0.3188 | LM: 0.3066 | LB: 1.1108 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.437/SR1: 0.414 | LR: 1.00e-04 [2026-04-17 10:46:17] Epoch 1 | Step 8050 | Loss: 0.3188 | LM: 0.3067 | LB: 1.1107 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.437/SR1: 0.414 | LR: 1.00e-04 [2026-04-17 10:46:23] Epoch 1 | Step 8060 | Loss: 0.3187 | LM: 0.3067 | LB: 1.1107 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.437/SR1: 0.414 | LR: 1.00e-04 [2026-04-17 10:46:29] Epoch 1 | Step 8070 | Loss: 0.3188 | LM: 0.3067 | LB: 1.1107 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.437/SR1: 0.414 | LR: 1.00e-04 [2026-04-17 10:46:36] Epoch 1 | Step 8080 | Loss: 0.3188 | LM: 0.3067 | LB: 1.1106 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.437/SR1: 0.413 | LR: 1.00e-04 [2026-04-17 10:46:42] Epoch 1 | Step 8090 | Loss: 0.3188 | LM: 0.3066 | LB: 1.1106 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.437/SR1: 0.413 | LR: 1.00e-04 [2026-04-17 10:46:49] Epoch 1 | Step 8100 | Loss: 0.3188 | LM: 0.3066 | LB: 1.1106 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.437/SR1: 0.413 | LR: 1.00e-04 [2026-04-17 10:46:55] Epoch 1 | Step 8110 | Loss: 0.3188 | LM: 0.3065 | LB: 1.1106 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.437/SR1: 0.413 | LR: 1.00e-04 [2026-04-17 10:47:02] Epoch 1 | Step 8120 | Loss: 0.3188 | LM: 0.3065 | LB: 1.1105 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.437/SR1: 0.413 | LR: 1.00e-04 [2026-04-17 10:47:08] Epoch 1 | Step 8130 | Loss: 0.3187 | LM: 0.3064 | LB: 1.1105 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.437/SR1: 0.413 | LR: 1.00e-04 [2026-04-17 10:47:14] Epoch 1 | Step 8140 | Loss: 0.3187 | LM: 0.3065 | LB: 1.1105 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.437/SR1: 0.413 | LR: 1.00e-04 [2026-04-17 10:47:21] Epoch 1 | Step 8150 | Loss: 0.3187 | LM: 0.3064 | LB: 1.1105 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.437/SR1: 0.413 | LR: 1.00e-04 [2026-04-17 10:47:27] Epoch 1 | Step 8160 | Loss: 0.3187 | LM: 0.3064 | LB: 1.1104 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.437/SR1: 0.413 | LR: 1.00e-04 [2026-04-17 10:47:33] Epoch 1 | Step 8170 | Loss: 0.3187 | LM: 0.3062 | LB: 1.1104 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.437/SR1: 0.413 | LR: 1.00e-04 [2026-04-17 10:47:40] Epoch 1 | Step 8180 | Loss: 0.3187 | LM: 0.3062 | LB: 1.1104 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.437/SR1: 0.413 | LR: 1.00e-04 [2026-04-17 10:47:46] Epoch 1 | Step 8190 | Loss: 0.3187 | LM: 0.3062 | LB: 1.1104 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.437/SR1: 0.413 | LR: 1.00e-04 [2026-04-17 10:47:53] Epoch 1 | Step 8200 | Loss: 0.3187 | LM: 0.3061 | LB: 1.1103 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.437/SR1: 0.413 | LR: 1.00e-04 [2026-04-17 10:47:59] Epoch 1 | Step 8210 | Loss: 0.3186 | LM: 0.3061 | LB: 1.1103 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.437/SR1: 0.413 | LR: 1.00e-04 [2026-04-17 10:48:05] Epoch 1 | Step 8220 | Loss: 0.3186 | LM: 0.3061 | LB: 1.1103 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.437/SR1: 0.413 | LR: 1.00e-04 [2026-04-17 10:48:12] Epoch 1 | Step 8230 | Loss: 0.3186 | LM: 0.3061 | LB: 1.1102 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.437/SR1: 0.413 | LR: 1.00e-04 [2026-04-17 10:48:18] Epoch 1 | Step 8240 | Loss: 0.3186 | LM: 0.3062 | LB: 1.1102 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.437/SR1: 0.413 | LR: 1.00e-04 [2026-04-17 10:48:25] Epoch 1 | Step 8250 | Loss: 0.3185 | LM: 0.3061 | LB: 1.1102 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.437/SR1: 0.413 | LR: 1.00e-04 [2026-04-17 10:48:31] Epoch 1 | Step 8260 | Loss: 0.3186 | LM: 0.3062 | LB: 1.1102 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.437/SR1: 0.413 | LR: 1.00e-04 [2026-04-17 10:48:38] Epoch 1 | Step 8270 | Loss: 0.3186 | LM: 0.3062 | LB: 1.1101 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.437/SR1: 0.413 | LR: 1.00e-04 [2026-04-17 10:48:44] Epoch 1 | Step 8280 | Loss: 0.3186 | LM: 0.3062 | LB: 1.1101 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.437/SR1: 0.413 | LR: 1.00e-04 [2026-04-17 10:48:50] Epoch 1 | Step 8290 | Loss: 0.3185 | LM: 0.3061 | LB: 1.1101 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.437/SR1: 0.413 | LR: 1.00e-04 [2026-04-17 10:48:57] Epoch 1 | Step 8300 | Loss: 0.3185 | LM: 0.3062 | LB: 1.1100 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.436/SR1: 0.413 | LR: 1.00e-04 [2026-04-17 10:49:03] Epoch 1 | Step 8310 | Loss: 0.3185 | LM: 0.3061 | LB: 1.1100 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.436/SR1: 0.413 | LR: 1.00e-04 [2026-04-17 10:49:09] Epoch 1 | Step 8320 | Loss: 0.3185 | LM: 0.3062 | LB: 1.1100 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.436/SR1: 0.413 | LR: 1.00e-04 [2026-04-17 10:49:16] Epoch 1 | Step 8330 | Loss: 0.3185 | LM: 0.3061 | LB: 1.1100 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.436/SR1: 0.413 | LR: 1.00e-04 [2026-04-17 10:49:22] Epoch 1 | Step 8340 | Loss: 0.3185 | LM: 0.3060 | LB: 1.1099 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.436/SR1: 0.413 | LR: 1.00e-04 [2026-04-17 10:49:29] Epoch 1 | Step 8350 | Loss: 0.3185 | LM: 0.3060 | LB: 1.1099 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.436/SR1: 0.413 | LR: 1.00e-04 [2026-04-17 10:49:35] Epoch 1 | Step 8360 | Loss: 0.3185 | LM: 0.3060 | LB: 1.1099 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.436/SR1: 0.413 | LR: 1.00e-04 [2026-04-17 10:49:41] Epoch 1 | Step 8370 | Loss: 0.3185 | LM: 0.3059 | LB: 1.1099 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.436/SR1: 0.413 | LR: 1.00e-04 [2026-04-17 10:49:48] Epoch 1 | Step 8380 | Loss: 0.3185 | LM: 0.3058 | LB: 1.1098 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.436/SR1: 0.413 | LR: 1.00e-04 [2026-04-17 10:49:54] Epoch 1 | Step 8390 | Loss: 0.3185 | LM: 0.3058 | LB: 1.1098 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.436/SR1: 0.412 | LR: 1.00e-04 [2026-04-17 10:50:01] Epoch 1 | Step 8400 | Loss: 0.3185 | LM: 0.3058 | LB: 1.1098 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.436/SR1: 0.412 | LR: 1.00e-04 [2026-04-17 10:50:07] Epoch 1 | Step 8410 | Loss: 0.3184 | LM: 0.3058 | LB: 1.1098 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.436/SR1: 0.412 | LR: 1.00e-04 [2026-04-17 10:50:13] Epoch 1 | Step 8420 | Loss: 0.3184 | LM: 0.3057 | LB: 1.1097 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.436/SR1: 0.412 | LR: 1.00e-04 [2026-04-17 10:50:20] Epoch 1 | Step 8430 | Loss: 0.3184 | LM: 0.3056 | LB: 1.1097 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.436/SR1: 0.412 | LR: 1.00e-04 [2026-04-17 10:50:26] Epoch 1 | Step 8440 | Loss: 0.3183 | LM: 0.3057 | LB: 1.1097 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.436/SR1: 0.412 | LR: 1.00e-04 [2026-04-17 10:50:32] Epoch 1 | Step 8450 | Loss: 0.3183 | LM: 0.3056 | LB: 1.1097 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.436/SR1: 0.412 | LR: 1.00e-04 [2026-04-17 10:50:39] Epoch 1 | Step 8460 | Loss: 0.3183 | LM: 0.3056 | LB: 1.1097 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.436/SR1: 0.412 | LR: 1.00e-04 [2026-04-17 10:50:45] Epoch 1 | Step 8470 | Loss: 0.3182 | LM: 0.3056 | LB: 1.1096 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.436/SR1: 0.412 | LR: 1.00e-04 [2026-04-17 10:50:52] Epoch 1 | Step 8480 | Loss: 0.3182 | LM: 0.3055 | LB: 1.1096 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.436/SR1: 0.412 | LR: 1.00e-04 [2026-04-17 10:50:58] Epoch 1 | Step 8490 | Loss: 0.3181 | LM: 0.3055 | LB: 1.1096 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.436/SR1: 0.412 | LR: 1.00e-04 [2026-04-17 10:51:04] Epoch 1 | Step 8500 | Loss: 0.3181 | LM: 0.3056 | LB: 1.1096 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.436/SR1: 0.412 | LR: 1.00e-04 [2026-04-17 10:51:11] Epoch 1 | Step 8510 | Loss: 0.3181 | LM: 0.3056 | LB: 1.1095 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.436/SR1: 0.412 | LR: 1.00e-04 [2026-04-17 10:51:17] Epoch 1 | Step 8520 | Loss: 0.3182 | LM: 0.3057 | LB: 1.1095 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.436/SR1: 0.412 | LR: 1.00e-04 [2026-04-17 10:51:24] Epoch 1 | Step 8530 | Loss: 0.3182 | LM: 0.3058 | LB: 1.1095 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.436/SR1: 0.412 | LR: 1.00e-04 [2026-04-17 10:51:30] Epoch 1 | Step 8540 | Loss: 0.3182 | LM: 0.3059 | LB: 1.1095 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.436/SR1: 0.412 | LR: 1.00e-04 [2026-04-17 10:51:36] Epoch 1 | Step 8550 | Loss: 0.3182 | LM: 0.3059 | LB: 1.1095 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.436/SR1: 0.412 | LR: 1.00e-04 [2026-04-17 10:51:42] Epoch 1 | Step 8560 | Loss: 0.3181 | LM: 0.3058 | LB: 1.1094 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.436/SR1: 0.412 | LR: 1.00e-04 [2026-04-17 10:51:49] Epoch 1 | Step 8570 | Loss: 0.3181 | LM: 0.3057 | LB: 1.1094 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.436/SR1: 0.412 | LR: 1.00e-04 [2026-04-17 10:51:55] Epoch 1 | Step 8580 | Loss: 0.3180 | LM: 0.3057 | LB: 1.1094 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.436/SR1: 0.412 | LR: 1.00e-04 [2026-04-17 10:52:01] Epoch 1 | Step 8590 | Loss: 0.3181 | LM: 0.3057 | LB: 1.1093 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.436/SR1: 0.412 | LR: 1.00e-04 [2026-04-17 10:52:08] Epoch 1 | Step 8600 | Loss: 0.3180 | LM: 0.3056 | LB: 1.1093 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.436/SR1: 0.412 | LR: 1.00e-04 [2026-04-17 10:52:14] Epoch 1 | Step 8610 | Loss: 0.3180 | LM: 0.3057 | LB: 1.1093 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.436/SR1: 0.412 | LR: 1.00e-04 [2026-04-17 10:52:21] Epoch 1 | Step 8620 | Loss: 0.3179 | LM: 0.3056 | LB: 1.1093 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.436/SR1: 0.412 | LR: 1.00e-04 [2026-04-17 10:52:28] Epoch 1 | Step 8630 | Loss: 0.3179 | LM: 0.3056 | LB: 1.1093 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.436/SR1: 0.412 | LR: 1.00e-04 [2026-04-17 10:52:34] Epoch 1 | Step 8640 | Loss: 0.3179 | LM: 0.3055 | LB: 1.1093 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.436/SR1: 0.412 | LR: 1.00e-04 [2026-04-17 10:52:41] Epoch 1 | Step 8650 | Loss: 0.3178 | LM: 0.3055 | LB: 1.1092 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.436/SR1: 0.412 | LR: 1.00e-04 [2026-04-17 10:52:47] Epoch 1 | Step 8660 | Loss: 0.3178 | LM: 0.3054 | LB: 1.1092 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.436/SR1: 0.412 | LR: 1.00e-04 [2026-04-17 10:52:54] Epoch 1 | Step 8670 | Loss: 0.3178 | LM: 0.3054 | LB: 1.1092 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.436/SR1: 0.412 | LR: 1.00e-04 [2026-04-17 10:53:00] Epoch 1 | Step 8680 | Loss: 0.3178 | LM: 0.3054 | LB: 1.1092 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.436/SR1: 0.412 | LR: 1.00e-04 [2026-04-17 10:53:07] Epoch 1 | Step 8690 | Loss: 0.3177 | LM: 0.3053 | LB: 1.1091 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.436/SR1: 0.412 | LR: 1.00e-04 [2026-04-17 10:53:13] Epoch 1 | Step 8700 | Loss: 0.3178 | LM: 0.3053 | LB: 1.1091 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.436/SR1: 0.412 | LR: 1.00e-04 [2026-04-17 10:53:20] Epoch 1 | Step 8710 | Loss: 0.3178 | LM: 0.3053 | LB: 1.1091 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.436/SR1: 0.412 | LR: 1.00e-04 [2026-04-17 10:53:26] Epoch 1 | Step 8720 | Loss: 0.3178 | LM: 0.3053 | LB: 1.1091 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.436/SR1: 0.412 | LR: 1.00e-04 [2026-04-17 10:53:32] Epoch 1 | Step 8730 | Loss: 0.3178 | LM: 0.3052 | LB: 1.1090 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.436/SR1: 0.412 | LR: 1.00e-04 [2026-04-17 10:53:39] Epoch 1 | Step 8740 | Loss: 0.3179 | LM: 0.3054 | LB: 1.1090 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.436/SR1: 0.412 | LR: 1.00e-04 [2026-04-17 10:53:45] Epoch 1 | Step 8750 | Loss: 0.3179 | LM: 0.3054 | LB: 1.1090 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.436/SR1: 0.412 | LR: 1.00e-04 [2026-04-17 10:53:52] Epoch 1 | Step 8760 | Loss: 0.3179 | LM: 0.3054 | LB: 1.1090 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.436/SR1: 0.412 | LR: 1.00e-04 [2026-04-17 10:53:58] Epoch 1 | Step 8770 | Loss: 0.3178 | LM: 0.3055 | LB: 1.1090 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.436/SR1: 0.412 | LR: 1.00e-04 [2026-04-17 10:54:05] Epoch 1 | Step 8780 | Loss: 0.3178 | LM: 0.3054 | LB: 1.1090 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.435/SR1: 0.411 | LR: 1.00e-04 [2026-04-17 10:54:11] Epoch 1 | Step 8790 | Loss: 0.3178 | LM: 0.3054 | LB: 1.1089 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.435/SR1: 0.411 | LR: 1.00e-04 [2026-04-17 10:54:17] Epoch 1 | Step 8800 | Loss: 0.3178 | LM: 0.3054 | LB: 1.1089 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.435/SR1: 0.411 | LR: 1.00e-04 [2026-04-17 10:54:23] Epoch 1 | Step 8810 | Loss: 0.3178 | LM: 0.3054 | LB: 1.1089 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.435/SR1: 0.411 | LR: 1.00e-04 [2026-04-17 10:54:30] Epoch 1 | Step 8820 | Loss: 0.3178 | LM: 0.3053 | LB: 1.1089 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.435/SR1: 0.411 | LR: 1.00e-04 [2026-04-17 10:54:36] Epoch 1 | Step 8830 | Loss: 0.3178 | LM: 0.3054 | LB: 1.1089 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.435/SR1: 0.411 | LR: 1.00e-04 [2026-04-17 10:54:42] Epoch 1 | Step 8840 | Loss: 0.3177 | LM: 0.3053 | LB: 1.1088 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.435/SR1: 0.411 | LR: 1.00e-04 [2026-04-17 10:54:49] Epoch 1 | Step 8850 | Loss: 0.3177 | LM: 0.3054 | LB: 1.1088 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.435/SR1: 0.411 | LR: 1.00e-04 [2026-04-17 10:54:55] Epoch 1 | Step 8860 | Loss: 0.3177 | LM: 0.3054 | LB: 1.1088 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.435/SR1: 0.411 | LR: 1.00e-04 [2026-04-17 10:55:02] Epoch 1 | Step 8870 | Loss: 0.3177 | LM: 0.3055 | LB: 1.1088 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.435/SR1: 0.411 | LR: 1.00e-04 [2026-04-17 10:55:08] Epoch 1 | Step 8880 | Loss: 0.3177 | LM: 0.3056 | LB: 1.1088 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.435/SR1: 0.411 | LR: 1.00e-04 [2026-04-17 10:55:15] Epoch 1 | Step 8890 | Loss: 0.3177 | LM: 0.3056 | LB: 1.1088 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.435/SR1: 0.411 | LR: 1.00e-04 [2026-04-17 10:55:21] Epoch 1 | Step 8900 | Loss: 0.3176 | LM: 0.3056 | LB: 1.1087 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.435/SR1: 0.411 | LR: 1.00e-04 [2026-04-17 10:55:27] Epoch 1 | Step 8910 | Loss: 0.3176 | LM: 0.3057 | LB: 1.1087 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.435/SR1: 0.411 | LR: 1.00e-04 [2026-04-17 10:55:34] Epoch 1 | Step 8920 | Loss: 0.3176 | LM: 0.3056 | LB: 1.1087 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.435/SR1: 0.411 | LR: 1.00e-04 [2026-04-17 10:55:40] Epoch 1 | Step 8930 | Loss: 0.3175 | LM: 0.3056 | LB: 1.1087 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.435/SR1: 0.411 | LR: 1.00e-04 [2026-04-17 10:55:47] Epoch 1 | Step 8940 | Loss: 0.3175 | LM: 0.3056 | LB: 1.1087 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.435/SR1: 0.411 | LR: 1.00e-04 [2026-04-17 10:55:53] Epoch 1 | Step 8950 | Loss: 0.3175 | LM: 0.3055 | LB: 1.1086 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.435/SR1: 0.411 | LR: 1.00e-04 [2026-04-17 10:55:59] Epoch 1 | Step 8960 | Loss: 0.3175 | LM: 0.3055 | LB: 1.1086 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.435/SR1: 0.411 | LR: 1.00e-04 [2026-04-17 10:56:06] Epoch 1 | Step 8970 | Loss: 0.3174 | LM: 0.3055 | LB: 1.1086 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.435/SR1: 0.411 | LR: 1.00e-04 [2026-04-17 10:56:12] Epoch 1 | Step 8980 | Loss: 0.3174 | LM: 0.3054 | LB: 1.1086 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.435/SR1: 0.411 | LR: 1.00e-04 [2026-04-17 10:56:19] Epoch 1 | Step 8990 | Loss: 0.3174 | LM: 0.3055 | LB: 1.1086 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.435/SR1: 0.411 | LR: 1.00e-04 [2026-04-17 10:56:25] Epoch 1 | Step 9000 | Loss: 0.3174 | LM: 0.3055 | LB: 1.1085 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.435/SR1: 0.411 | LR: 1.00e-04 [2026-04-17 10:56:34] Checkpoint saved: outputs/2026-04-17/08-57-56/checkpoints/checkpoint_step_9000.pt [2026-04-17 10:56:50] Validation | Batch 10/784 | Loss: 0.3217 | LM_LOSS: 0.3109 | LB_LOSS: 1.0872 [2026-04-17 10:56:52] Validation | Batch 20/784 | Loss: 0.3333 | LM_LOSS: 0.3224 | LB_LOSS: 1.0874 [2026-04-17 10:56:53] Validation | Batch 30/784 | Loss: 0.3195 | LM_LOSS: 0.3086 | LB_LOSS: 1.0866 [2026-04-17 10:56:55] Validation | Batch 40/784 | Loss: 0.3210 | LM_LOSS: 0.3101 | LB_LOSS: 1.0866 [2026-04-17 10:56:56] Validation | Batch 50/784 | Loss: 0.3173 | LM_LOSS: 0.3064 | LB_LOSS: 1.0859 [2026-04-17 10:56:57] Validation | Batch 60/784 | Loss: 0.3177 | LM_LOSS: 0.3069 | LB_LOSS: 1.0854 [2026-04-17 10:56:59] Validation | Batch 70/784 | Loss: 0.3149 | LM_LOSS: 0.3041 | LB_LOSS: 1.0848 [2026-04-17 10:57:00] Validation | Batch 80/784 | Loss: 0.3111 | LM_LOSS: 0.3002 | LB_LOSS: 1.0843 [2026-04-17 10:57:01] Validation | Batch 90/784 | Loss: 0.3095 | LM_LOSS: 0.2987 | LB_LOSS: 1.0849 [2026-04-17 10:57:02] Validation | Batch 100/784 | Loss: 0.3113 | LM_LOSS: 0.3004 | LB_LOSS: 1.0853 [2026-04-17 10:57:04] Validation | Batch 110/784 | Loss: 0.3068 | LM_LOSS: 0.2960 | LB_LOSS: 1.0855 [2026-04-17 10:57:05] Validation | Batch 120/784 | Loss: 0.3102 | LM_LOSS: 0.2993 | LB_LOSS: 1.0854 [2026-04-17 10:57:06] Validation | Batch 130/784 | Loss: 0.3127 | LM_LOSS: 0.3018 | LB_LOSS: 1.0853 [2026-04-17 10:57:08] Validation | Batch 140/784 | Loss: 0.3122 | LM_LOSS: 0.3013 | LB_LOSS: 1.0851 [2026-04-17 10:57:09] Validation | Batch 150/784 | Loss: 0.3084 | LM_LOSS: 0.2976 | LB_LOSS: 1.0854 [2026-04-17 10:57:11] Validation | Batch 160/784 | Loss: 0.3090 | LM_LOSS: 0.2982 | LB_LOSS: 1.0851 [2026-04-17 10:57:12] Validation | Batch 170/784 | Loss: 0.3097 | LM_LOSS: 0.2988 | LB_LOSS: 1.0848 [2026-04-17 10:57:14] Validation | Batch 180/784 | Loss: 0.3072 | LM_LOSS: 0.2964 | LB_LOSS: 1.0848 [2026-04-17 10:57:15] Validation | Batch 190/784 | Loss: 0.3089 | LM_LOSS: 0.2981 | LB_LOSS: 1.0852 [2026-04-17 10:57:16] Validation | Batch 200/784 | Loss: 0.3092 | LM_LOSS: 0.2983 | LB_LOSS: 1.0853 [2026-04-17 10:57:18] Validation | Batch 210/784 | Loss: 0.3080 | LM_LOSS: 0.2972 | LB_LOSS: 1.0852 [2026-04-17 10:57:19] Validation | Batch 220/784 | Loss: 0.3088 | LM_LOSS: 0.2979 | LB_LOSS: 1.0852 [2026-04-17 10:57:20] Validation | Batch 230/784 | Loss: 0.3092 | LM_LOSS: 0.2984 | LB_LOSS: 1.0852 [2026-04-17 10:57:22] Validation | Batch 240/784 | Loss: 0.3095 | LM_LOSS: 0.2987 | LB_LOSS: 1.0855 [2026-04-17 10:57:23] Validation | Batch 250/784 | Loss: 0.3094 | LM_LOSS: 0.2986 | LB_LOSS: 1.0854 [2026-04-17 10:57:25] Validation | Batch 260/784 | Loss: 0.3095 | LM_LOSS: 0.2986 | LB_LOSS: 1.0856 [2026-04-17 10:57:26] Validation | Batch 270/784 | Loss: 0.3090 | LM_LOSS: 0.2981 | LB_LOSS: 1.0856 [2026-04-17 10:57:27] Validation | Batch 280/784 | Loss: 0.3096 | LM_LOSS: 0.2988 | LB_LOSS: 1.0858 [2026-04-17 10:57:29] Validation | Batch 290/784 | Loss: 0.3106 | LM_LOSS: 0.2997 | LB_LOSS: 1.0859 [2026-04-17 10:57:30] Validation | Batch 300/784 | Loss: 0.3112 | LM_LOSS: 0.3003 | LB_LOSS: 1.0860 [2026-04-17 10:57:31] Validation | Batch 310/784 | Loss: 0.3107 | LM_LOSS: 0.2998 | LB_LOSS: 1.0859 [2026-04-17 10:57:33] Validation | Batch 320/784 | Loss: 0.3121 | LM_LOSS: 0.3012 | LB_LOSS: 1.0859 [2026-04-17 10:57:34] Validation | Batch 330/784 | Loss: 0.3119 | LM_LOSS: 0.3011 | LB_LOSS: 1.0859 [2026-04-17 10:57:35] Validation | Batch 340/784 | Loss: 0.3109 | LM_LOSS: 0.3001 | LB_LOSS: 1.0860 [2026-04-17 10:57:37] Validation | Batch 350/784 | Loss: 0.3110 | LM_LOSS: 0.3001 | LB_LOSS: 1.0862 [2026-04-17 10:57:38] Validation | Batch 360/784 | Loss: 0.3107 | LM_LOSS: 0.2999 | LB_LOSS: 1.0862 [2026-04-17 10:57:39] Validation | Batch 370/784 | Loss: 0.3112 | LM_LOSS: 0.3004 | LB_LOSS: 1.0861 [2026-04-17 10:57:41] Validation | Batch 380/784 | Loss: 0.3111 | LM_LOSS: 0.3002 | LB_LOSS: 1.0861 [2026-04-17 10:57:42] Validation | Batch 390/784 | Loss: 0.3109 | LM_LOSS: 0.3001 | LB_LOSS: 1.0862 [2026-04-17 10:57:43] Validation | Batch 400/784 | Loss: 0.3112 | LM_LOSS: 0.3003 | LB_LOSS: 1.0862 [2026-04-17 10:57:44] Validation | Batch 410/784 | Loss: 0.3114 | LM_LOSS: 0.3006 | LB_LOSS: 1.0862 [2026-04-17 10:57:46] Validation | Batch 420/784 | Loss: 0.3116 | LM_LOSS: 0.3007 | LB_LOSS: 1.0863 [2026-04-17 10:57:47] Validation | Batch 430/784 | Loss: 0.3115 | LM_LOSS: 0.3006 | LB_LOSS: 1.0862 [2026-04-17 10:57:48] Validation | Batch 440/784 | Loss: 0.3111 | LM_LOSS: 0.3002 | LB_LOSS: 1.0863 [2026-04-17 10:57:50] Validation | Batch 450/784 | Loss: 0.3105 | LM_LOSS: 0.2997 | LB_LOSS: 1.0862 [2026-04-17 10:57:51] Validation | Batch 460/784 | Loss: 0.3109 | LM_LOSS: 0.3001 | LB_LOSS: 1.0863 [2026-04-17 10:57:52] Validation | Batch 470/784 | Loss: 0.3102 | LM_LOSS: 0.2993 | LB_LOSS: 1.0863 [2026-04-17 10:57:54] Validation | Batch 480/784 | Loss: 0.3105 | LM_LOSS: 0.2997 | LB_LOSS: 1.0862 [2026-04-17 10:57:55] Validation | Batch 490/784 | Loss: 0.3100 | LM_LOSS: 0.2991 | LB_LOSS: 1.0862 [2026-04-17 10:57:56] Validation | Batch 500/784 | Loss: 0.3104 | LM_LOSS: 0.2996 | LB_LOSS: 1.0861 [2026-04-17 10:57:58] Validation | Batch 510/784 | Loss: 0.3101 | LM_LOSS: 0.2992 | LB_LOSS: 1.0861 [2026-04-17 10:57:59] Validation | Batch 520/784 | Loss: 0.3101 | LM_LOSS: 0.2993 | LB_LOSS: 1.0860 [2026-04-17 10:58:00] Validation | Batch 530/784 | Loss: 0.3110 | LM_LOSS: 0.3001 | LB_LOSS: 1.0860 [2026-04-17 10:58:02] Validation | Batch 540/784 | Loss: 0.3114 | LM_LOSS: 0.3005 | LB_LOSS: 1.0860 [2026-04-17 10:58:03] Validation | Batch 550/784 | Loss: 0.3128 | LM_LOSS: 0.3019 | LB_LOSS: 1.0859 [2026-04-17 10:58:05] Validation | Batch 560/784 | Loss: 0.3128 | LM_LOSS: 0.3020 | LB_LOSS: 1.0860 [2026-04-17 10:58:06] Validation | Batch 570/784 | Loss: 0.3124 | LM_LOSS: 0.3016 | LB_LOSS: 1.0859 [2026-04-17 10:58:07] Validation | Batch 580/784 | Loss: 0.3118 | LM_LOSS: 0.3010 | LB_LOSS: 1.0859 [2026-04-17 10:58:09] Validation | Batch 590/784 | Loss: 0.3121 | LM_LOSS: 0.3012 | LB_LOSS: 1.0859 [2026-04-17 10:58:10] Validation | Batch 600/784 | Loss: 0.3120 | LM_LOSS: 0.3012 | LB_LOSS: 1.0858 [2026-04-17 10:58:11] Validation | Batch 610/784 | Loss: 0.3122 | LM_LOSS: 0.3013 | LB_LOSS: 1.0858 [2026-04-17 10:58:13] Validation | Batch 620/784 | Loss: 0.3121 | LM_LOSS: 0.3012 | LB_LOSS: 1.0858 [2026-04-17 10:58:14] Validation | Batch 630/784 | Loss: 0.3127 | LM_LOSS: 0.3019 | LB_LOSS: 1.0858 [2026-04-17 10:58:16] Validation | Batch 640/784 | Loss: 0.3128 | LM_LOSS: 0.3019 | LB_LOSS: 1.0858 [2026-04-17 10:58:17] Validation | Batch 650/784 | Loss: 0.3127 | LM_LOSS: 0.3019 | LB_LOSS: 1.0859 [2026-04-17 10:58:19] Validation | Batch 660/784 | Loss: 0.3130 | LM_LOSS: 0.3021 | LB_LOSS: 1.0859 [2026-04-17 10:58:20] Validation | Batch 670/784 | Loss: 0.3134 | LM_LOSS: 0.3025 | LB_LOSS: 1.0859 [2026-04-17 10:58:22] Validation | Batch 680/784 | Loss: 0.3131 | LM_LOSS: 0.3023 | LB_LOSS: 1.0859 [2026-04-17 10:58:23] Validation | Batch 690/784 | Loss: 0.3133 | LM_LOSS: 0.3025 | LB_LOSS: 1.0859 [2026-04-17 10:58:24] Validation | Batch 700/784 | Loss: 0.3134 | LM_LOSS: 0.3025 | LB_LOSS: 1.0858 [2026-04-17 10:58:26] Validation | Batch 710/784 | Loss: 0.3131 | LM_LOSS: 0.3023 | LB_LOSS: 1.0858 [2026-04-17 10:58:27] Validation | Batch 720/784 | Loss: 0.3129 | LM_LOSS: 0.3020 | LB_LOSS: 1.0857 [2026-04-17 10:58:28] Validation | Batch 730/784 | Loss: 0.3124 | LM_LOSS: 0.3015 | LB_LOSS: 1.0857 [2026-04-17 10:58:30] Validation | Batch 740/784 | Loss: 0.3124 | LM_LOSS: 0.3016 | LB_LOSS: 1.0857 [2026-04-17 10:58:31] Validation | Batch 750/784 | Loss: 0.3118 | LM_LOSS: 0.3009 | LB_LOSS: 1.0857 [2026-04-17 10:58:32] Validation | Batch 760/784 | Loss: 0.3119 | LM_LOSS: 0.3011 | LB_LOSS: 1.0857 [2026-04-17 10:58:33] Validation | Batch 770/784 | Loss: 0.3120 | LM_LOSS: 0.3012 | LB_LOSS: 1.0858 [2026-04-17 10:58:35] Validation | Batch 780/784 | Loss: 0.3124 | LM_LOSS: 0.3015 | LB_LOSS: 1.0858 [2026-04-17 10:58:35] Validation | Batch 784/784 | Loss: 0.3125 | LM_LOSS: 0.3017 | LB_LOSS: 1.0858 [2026-04-17 10:58:39] Validation | Loss: 0.3125 | LM_LOSS: 0.3017 | LB_LOSS: 1.0858 | PPL: 1.35 | Time: 106.37s [2026-04-17 10:58:44] New best model saved! Val loss: 0.3125 [2026-04-17 10:58:50] Epoch 1 | Step 9010 | Loss: 0.3174 | LM: 0.3055 | LB: 1.1085 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.435/SR1: 0.411 | LR: 1.00e-04 [2026-04-17 10:58:57] Epoch 1 | Step 9020 | Loss: 0.3173 | LM: 0.3055 | LB: 1.1085 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.435/SR1: 0.411 | LR: 1.00e-04 [2026-04-17 10:59:03] Epoch 1 | Step 9030 | Loss: 0.3173 | LM: 0.3055 | LB: 1.1085 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.435/SR1: 0.411 | LR: 1.00e-04 [2026-04-17 10:59:09] Epoch 1 | Step 9040 | Loss: 0.3173 | LM: 0.3056 | LB: 1.1085 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.435/SR1: 0.411 | LR: 1.00e-04 [2026-04-17 10:59:16] Epoch 1 | Step 9050 | Loss: 0.3173 | LM: 0.3057 | LB: 1.1084 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.435/SR1: 0.411 | LR: 1.00e-04 [2026-04-17 10:59:22] Epoch 1 | Step 9060 | Loss: 0.3173 | LM: 0.3056 | LB: 1.1084 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.435/SR1: 0.411 | LR: 1.00e-04 [2026-04-17 10:59:29] Epoch 1 | Step 9070 | Loss: 0.3172 | LM: 0.3056 | LB: 1.1084 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.435/SR1: 0.411 | LR: 1.00e-04 [2026-04-17 10:59:35] Epoch 1 | Step 9080 | Loss: 0.3172 | LM: 0.3056 | LB: 1.1084 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.435/SR1: 0.411 | LR: 1.00e-04 [2026-04-17 10:59:41] Epoch 1 | Step 9090 | Loss: 0.3172 | LM: 0.3055 | LB: 1.1083 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.435/SR1: 0.411 | LR: 1.00e-04 [2026-04-17 10:59:48] Epoch 1 | Step 9100 | Loss: 0.3172 | LM: 0.3055 | LB: 1.1083 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.435/SR1: 0.411 | LR: 1.00e-04 [2026-04-17 10:59:54] Epoch 1 | Step 9110 | Loss: 0.3172 | LM: 0.3055 | LB: 1.1083 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.435/SR1: 0.411 | LR: 1.00e-04 [2026-04-17 11:00:00] Epoch 1 | Step 9120 | Loss: 0.3172 | LM: 0.3056 | LB: 1.1083 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.435/SR1: 0.411 | LR: 1.00e-04 [2026-04-17 11:00:07] Epoch 1 | Step 9130 | Loss: 0.3171 | LM: 0.3055 | LB: 1.1083 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.435/SR1: 0.411 | LR: 1.00e-04 [2026-04-17 11:00:13] Epoch 1 | Step 9140 | Loss: 0.3171 | LM: 0.3054 | LB: 1.1082 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.435/SR1: 0.411 | LR: 1.00e-04 [2026-04-17 11:00:20] Epoch 1 | Step 9150 | Loss: 0.3171 | LM: 0.3055 | LB: 1.1082 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.435/SR1: 0.411 | LR: 1.00e-04 [2026-04-17 11:00:26] Epoch 1 | Step 9160 | Loss: 0.3171 | LM: 0.3054 | LB: 1.1082 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.435/SR1: 0.411 | LR: 1.00e-04 [2026-04-17 11:00:33] Epoch 1 | Step 9170 | Loss: 0.3170 | LM: 0.3054 | LB: 1.1082 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.435/SR1: 0.411 | LR: 1.00e-04 [2026-04-17 11:00:39] Epoch 1 | Step 9180 | Loss: 0.3170 | LM: 0.3053 | LB: 1.1081 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.435/SR1: 0.410 | LR: 1.00e-04 [2026-04-17 11:00:45] Epoch 1 | Step 9190 | Loss: 0.3170 | LM: 0.3053 | LB: 1.1081 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.435/SR1: 0.410 | LR: 1.00e-04 [2026-04-17 11:00:52] Epoch 1 | Step 9200 | Loss: 0.3170 | LM: 0.3053 | LB: 1.1081 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.435/SR1: 0.410 | LR: 1.00e-04 [2026-04-17 11:00:58] Epoch 1 | Step 9210 | Loss: 0.3170 | LM: 0.3054 | LB: 1.1081 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.435/SR1: 0.410 | LR: 1.00e-04 [2026-04-17 11:01:04] Epoch 1 | Step 9220 | Loss: 0.3170 | LM: 0.3052 | LB: 1.1081 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.435/SR1: 0.410 | LR: 1.00e-04 [2026-04-17 11:01:11] Epoch 1 | Step 9230 | Loss: 0.3171 | LM: 0.3052 | LB: 1.1080 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.435/SR1: 0.410 | LR: 1.00e-04 [2026-04-17 11:01:17] Epoch 1 | Step 9240 | Loss: 0.3170 | LM: 0.3053 | LB: 1.1080 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.435/SR1: 0.410 | LR: 1.00e-04 [2026-04-17 11:01:23] Epoch 1 | Step 9250 | Loss: 0.3170 | LM: 0.3052 | LB: 1.1080 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.435/SR1: 0.410 | LR: 1.00e-04 [2026-04-17 11:01:30] Epoch 1 | Step 9260 | Loss: 0.3170 | LM: 0.3052 | LB: 1.1080 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.435/SR1: 0.410 | LR: 1.00e-04 [2026-04-17 11:01:36] Epoch 1 | Step 9270 | Loss: 0.3170 | LM: 0.3052 | LB: 1.1080 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.435/SR1: 0.410 | LR: 1.00e-04 [2026-04-17 11:01:43] Epoch 1 | Step 9280 | Loss: 0.3169 | LM: 0.3051 | LB: 1.1079 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.435/SR1: 0.410 | LR: 1.00e-04 [2026-04-17 11:01:49] Epoch 1 | Step 9290 | Loss: 0.3169 | LM: 0.3051 | LB: 1.1079 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.435/SR1: 0.410 | LR: 1.00e-04 [2026-04-17 11:01:55] Epoch 1 | Step 9300 | Loss: 0.3169 | LM: 0.3051 | LB: 1.1079 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.435/SR1: 0.410 | LR: 1.00e-04 [2026-04-17 11:02:02] Epoch 1 | Step 9310 | Loss: 0.3169 | LM: 0.3051 | LB: 1.1079 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.435/SR1: 0.410 | LR: 1.00e-04 [2026-04-17 11:02:08] Epoch 1 | Step 9320 | Loss: 0.3169 | LM: 0.3050 | LB: 1.1078 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.434/SR1: 0.410 | LR: 1.00e-04 [2026-04-17 11:02:14] Epoch 1 | Step 9330 | Loss: 0.3170 | LM: 0.3051 | LB: 1.1078 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.434/SR1: 0.410 | LR: 1.00e-04 [2026-04-17 11:02:21] Epoch 1 | Step 9340 | Loss: 0.3169 | LM: 0.3051 | LB: 1.1078 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.434/SR1: 0.410 | LR: 1.00e-04 [2026-04-17 11:02:27] Epoch 1 | Step 9350 | Loss: 0.3169 | LM: 0.3051 | LB: 1.1078 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.434/SR1: 0.410 | LR: 1.00e-04 [2026-04-17 11:02:34] Epoch 1 | Step 9360 | Loss: 0.3169 | LM: 0.3051 | LB: 1.1078 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.434/SR1: 0.410 | LR: 1.00e-04 [2026-04-17 11:02:40] Epoch 1 | Step 9370 | Loss: 0.3169 | LM: 0.3051 | LB: 1.1078 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.434/SR1: 0.410 | LR: 1.00e-04 [2026-04-17 11:02:46] Epoch 1 | Step 9380 | Loss: 0.3169 | LM: 0.3051 | LB: 1.1078 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.434/SR1: 0.410 | LR: 1.00e-04 [2026-04-17 11:02:53] Epoch 1 | Step 9390 | Loss: 0.3168 | LM: 0.3051 | LB: 1.1077 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.434/SR1: 0.410 | LR: 1.00e-04 [2026-04-17 11:02:59] Epoch 1 | Step 9400 | Loss: 0.3168 | LM: 0.3050 | LB: 1.1077 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.434/SR1: 0.410 | LR: 1.00e-04 [2026-04-17 11:03:05] Epoch 1 | Step 9410 | Loss: 0.3168 | LM: 0.3050 | LB: 1.1077 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.434/SR1: 0.410 | LR: 1.00e-04 [2026-04-17 11:03:12] Epoch 1 | Step 9420 | Loss: 0.3168 | LM: 0.3051 | LB: 1.1077 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.434/SR1: 0.410 | LR: 1.00e-04 [2026-04-17 11:03:18] Epoch 1 | Step 9430 | Loss: 0.3168 | LM: 0.3051 | LB: 1.1077 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.434/SR1: 0.410 | LR: 1.00e-04 [2026-04-17 11:03:25] Epoch 1 | Step 9440 | Loss: 0.3167 | LM: 0.3050 | LB: 1.1076 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.434/SR1: 0.410 | LR: 1.00e-04 [2026-04-17 11:03:31] Epoch 1 | Step 9450 | Loss: 0.3167 | LM: 0.3050 | LB: 1.1076 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.434/SR1: 0.410 | LR: 1.00e-04 [2026-04-17 11:03:38] Epoch 1 | Step 9460 | Loss: 0.3167 | LM: 0.3050 | LB: 1.1076 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.434/SR1: 0.410 | LR: 1.00e-04 [2026-04-17 11:03:44] Epoch 1 | Step 9470 | Loss: 0.3166 | LM: 0.3049 | LB: 1.1076 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.434/SR1: 0.410 | LR: 1.00e-04 [2026-04-17 11:03:51] Epoch 1 | Step 9480 | Loss: 0.3166 | LM: 0.3048 | LB: 1.1076 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.434/SR1: 0.410 | LR: 1.00e-04 [2026-04-17 11:03:57] Epoch 1 | Step 9490 | Loss: 0.3166 | LM: 0.3047 | LB: 1.1075 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.434/SR1: 0.410 | LR: 1.00e-04 [2026-04-17 11:04:03] Epoch 1 | Step 9500 | Loss: 0.3166 | LM: 0.3048 | LB: 1.1075 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.434/SR1: 0.410 | LR: 1.00e-04 [2026-04-17 11:04:10] Epoch 1 | Step 9510 | Loss: 0.3166 | LM: 0.3048 | LB: 1.1075 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.434/SR1: 0.410 | LR: 1.00e-04 [2026-04-17 11:04:16] Epoch 1 | Step 9520 | Loss: 0.3166 | LM: 0.3048 | LB: 1.1075 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.434/SR1: 0.410 | LR: 1.00e-04 [2026-04-17 11:04:22] Epoch 1 | Step 9530 | Loss: 0.3166 | LM: 0.3047 | LB: 1.1074 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.434/SR1: 0.410 | LR: 1.00e-04 [2026-04-17 11:04:29] Epoch 1 | Step 9540 | Loss: 0.3166 | LM: 0.3049 | LB: 1.1074 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.434/SR1: 0.410 | LR: 1.00e-04 [2026-04-17 11:04:35] Epoch 1 | Step 9550 | Loss: 0.3166 | LM: 0.3049 | LB: 1.1074 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.434/SR1: 0.410 | LR: 1.00e-04 [2026-04-17 11:04:42] Epoch 1 | Step 9560 | Loss: 0.3166 | LM: 0.3049 | LB: 1.1074 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.434/SR1: 0.410 | LR: 1.00e-04 [2026-04-17 11:04:48] Epoch 1 | Step 9570 | Loss: 0.3167 | LM: 0.3049 | LB: 1.1074 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.434/SR1: 0.410 | LR: 1.00e-04 [2026-04-17 11:04:55] Epoch 1 | Step 9580 | Loss: 0.3167 | LM: 0.3049 | LB: 1.1074 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.434/SR1: 0.409 | LR: 1.00e-04 [2026-04-17 11:05:01] Epoch 1 | Step 9590 | Loss: 0.3167 | LM: 0.3049 | LB: 1.1073 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.434/SR1: 0.409 | LR: 1.00e-04 [2026-04-17 11:05:08] Epoch 1 | Step 9600 | Loss: 0.3167 | LM: 0.3049 | LB: 1.1073 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.434/SR1: 0.409 | LR: 1.00e-04 [2026-04-17 11:05:14] Epoch 1 | Step 9610 | Loss: 0.3167 | LM: 0.3049 | LB: 1.1073 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.434/SR1: 0.409 | LR: 1.00e-04 [2026-04-17 11:05:20] Epoch 1 | Step 9620 | Loss: 0.3167 | LM: 0.3049 | LB: 1.1073 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.434/SR1: 0.409 | LR: 1.00e-04 [2026-04-17 11:05:27] Epoch 1 | Step 9630 | Loss: 0.3167 | LM: 0.3050 | LB: 1.1073 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.434/SR1: 0.409 | LR: 1.00e-04 [2026-04-17 11:05:33] Epoch 1 | Step 9640 | Loss: 0.3167 | LM: 0.3050 | LB: 1.1072 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.434/SR1: 0.409 | LR: 1.00e-04 [2026-04-17 11:05:39] Epoch 1 | Step 9650 | Loss: 0.3167 | LM: 0.3049 | LB: 1.1072 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.434/SR1: 0.409 | LR: 1.00e-04 [2026-04-17 11:05:46] Epoch 1 | Step 9660 | Loss: 0.3167 | LM: 0.3049 | LB: 1.1072 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.434/SR1: 0.409 | LR: 1.00e-04 [2026-04-17 11:05:52] Epoch 1 | Step 9670 | Loss: 0.3167 | LM: 0.3049 | LB: 1.1072 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.434/SR1: 0.409 | LR: 1.00e-04 [2026-04-17 11:05:59] Epoch 1 | Step 9680 | Loss: 0.3167 | LM: 0.3049 | LB: 1.1072 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.434/SR1: 0.409 | LR: 1.00e-04 [2026-04-17 11:06:05] Epoch 1 | Step 9690 | Loss: 0.3166 | LM: 0.3049 | LB: 1.1071 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.434/SR1: 0.409 | LR: 1.00e-04 [2026-04-17 11:06:11] Epoch 1 | Step 9700 | Loss: 0.3166 | LM: 0.3048 | LB: 1.1071 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.434/SR1: 0.409 | LR: 1.00e-04 [2026-04-17 11:06:18] Epoch 1 | Step 9710 | Loss: 0.3166 | LM: 0.3047 | LB: 1.1071 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.434/SR1: 0.409 | LR: 1.00e-04 [2026-04-17 11:06:24] Epoch 1 | Step 9720 | Loss: 0.3166 | LM: 0.3048 | LB: 1.1071 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.434/SR1: 0.409 | LR: 1.00e-04 [2026-04-17 11:06:31] Epoch 1 | Step 9730 | Loss: 0.3166 | LM: 0.3047 | LB: 1.1071 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.434/SR1: 0.409 | LR: 1.00e-04 [2026-04-17 11:06:37] Epoch 1 | Step 9740 | Loss: 0.3166 | LM: 0.3047 | LB: 1.1070 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.434/SR1: 0.409 | LR: 1.00e-04 [2026-04-17 11:06:44] Epoch 1 | Step 9750 | Loss: 0.3165 | LM: 0.3047 | LB: 1.1070 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.434/SR1: 0.409 | LR: 1.00e-04 [2026-04-17 11:06:50] Epoch 1 | Step 9760 | Loss: 0.3165 | LM: 0.3047 | LB: 1.1070 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.434/SR1: 0.409 | LR: 1.00e-04 [2026-04-17 11:06:56] Epoch 1 | Step 9770 | Loss: 0.3165 | LM: 0.3047 | LB: 1.1070 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.434/SR1: 0.409 | LR: 1.00e-04 [2026-04-17 11:07:03] Epoch 1 | Step 9780 | Loss: 0.3165 | LM: 0.3048 | LB: 1.1070 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.434/SR1: 0.409 | LR: 1.00e-04 [2026-04-17 11:07:09] Epoch 1 | Step 9790 | Loss: 0.3165 | LM: 0.3048 | LB: 1.1069 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.434/SR1: 0.409 | LR: 1.00e-04 [2026-04-17 11:07:16] Epoch 1 | Step 9800 | Loss: 0.3164 | LM: 0.3048 | LB: 1.1069 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.434/SR1: 0.409 | LR: 1.00e-04 [2026-04-17 11:07:22] Epoch 1 | Step 9810 | Loss: 0.3165 | LM: 0.3049 | LB: 1.1069 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.434/SR1: 0.409 | LR: 1.00e-04 [2026-04-17 11:07:29] Epoch 1 | Step 9820 | Loss: 0.3164 | LM: 0.3048 | LB: 1.1069 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.434/SR1: 0.409 | LR: 1.00e-04 [2026-04-17 11:07:35] Epoch 1 | Step 9830 | Loss: 0.3164 | LM: 0.3048 | LB: 1.1069 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.434/SR1: 0.409 | LR: 1.00e-04 [2026-04-17 11:07:41] Epoch 1 | Step 9840 | Loss: 0.3164 | LM: 0.3049 | LB: 1.1068 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.434/SR1: 0.409 | LR: 1.00e-04 [2026-04-17 11:07:48] Epoch 1 | Step 9850 | Loss: 0.3164 | LM: 0.3048 | LB: 1.1068 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.434/SR1: 0.409 | LR: 1.00e-04 [2026-04-17 11:07:54] Epoch 1 | Step 9860 | Loss: 0.3164 | LM: 0.3048 | LB: 1.1068 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.434/SR1: 0.409 | LR: 1.00e-04 [2026-04-17 11:08:01] Epoch 1 | Step 9870 | Loss: 0.3164 | LM: 0.3048 | LB: 1.1068 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.434/SR1: 0.409 | LR: 1.00e-04 [2026-04-17 11:08:07] Epoch 1 | Step 9880 | Loss: 0.3164 | LM: 0.3047 | LB: 1.1068 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.434/SR1: 0.409 | LR: 1.00e-04 [2026-04-17 11:08:14] Epoch 1 | Step 9890 | Loss: 0.3164 | LM: 0.3047 | LB: 1.1067 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.434/SR1: 0.409 | LR: 1.00e-04 [2026-04-17 11:08:20] Epoch 1 | Step 9900 | Loss: 0.3164 | LM: 0.3047 | LB: 1.1067 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.434/SR1: 0.409 | LR: 1.00e-04 [2026-04-17 11:08:26] Epoch 1 | Step 9910 | Loss: 0.3164 | LM: 0.3046 | LB: 1.1067 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.434/SR1: 0.409 | LR: 1.00e-04 [2026-04-17 11:08:33] Epoch 1 | Step 9920 | Loss: 0.3164 | LM: 0.3046 | LB: 1.1067 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.434/SR1: 0.409 | LR: 1.00e-04 [2026-04-17 11:08:39] Epoch 1 | Step 9930 | Loss: 0.3164 | LM: 0.3046 | LB: 1.1067 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.434/SR1: 0.409 | LR: 1.00e-04 [2026-04-17 11:08:45] Epoch 1 | Step 9940 | Loss: 0.3163 | LM: 0.3045 | LB: 1.1066 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.433/SR1: 0.409 | LR: 1.00e-04 [2026-04-17 11:08:52] Epoch 1 | Step 9950 | Loss: 0.3163 | LM: 0.3046 | LB: 1.1066 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.433/SR1: 0.409 | LR: 1.00e-04 [2026-04-17 11:08:58] Epoch 1 | Step 9960 | Loss: 0.3163 | LM: 0.3046 | LB: 1.1066 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.433/SR1: 0.409 | LR: 1.00e-04 [2026-04-17 11:09:05] Epoch 1 | Step 9970 | Loss: 0.3163 | LM: 0.3046 | LB: 1.1066 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.433/SR1: 0.409 | LR: 1.00e-04 [2026-04-17 11:09:11] Epoch 1 | Step 9980 | Loss: 0.3163 | LM: 0.3046 | LB: 1.1066 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.433/SR1: 0.409 | LR: 1.00e-04 [2026-04-17 11:09:17] Epoch 1 | Step 9990 | Loss: 0.3163 | LM: 0.3046 | LB: 1.1065 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.433/SR1: 0.408 | LR: 1.00e-04 [2026-04-17 11:09:24] Epoch 1 | Step 10000 | Loss: 0.3163 | LM: 0.3045 | LB: 1.1065 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.433/SR1: 0.408 | LR: 1.00e-04 [2026-04-17 11:09:25] Validation | Batch 10/784 | Loss: 0.3200 | LM_LOSS: 0.3092 | LB_LOSS: 1.0833 [2026-04-17 11:09:27] Validation | Batch 20/784 | Loss: 0.3317 | LM_LOSS: 0.3209 | LB_LOSS: 1.0834 [2026-04-17 11:09:28] Validation | Batch 30/784 | Loss: 0.3171 | LM_LOSS: 0.3063 | LB_LOSS: 1.0826 [2026-04-17 11:09:29] Validation | Batch 40/784 | Loss: 0.3190 | LM_LOSS: 0.3081 | LB_LOSS: 1.0825 [2026-04-17 11:09:31] Validation | Batch 50/784 | Loss: 0.3158 | LM_LOSS: 0.3050 | LB_LOSS: 1.0818 [2026-04-17 11:09:32] Validation | Batch 60/784 | Loss: 0.3165 | LM_LOSS: 0.3057 | LB_LOSS: 1.0814 [2026-04-17 11:09:33] Validation | Batch 70/784 | Loss: 0.3140 | LM_LOSS: 0.3032 | LB_LOSS: 1.0807 [2026-04-17 11:09:34] Validation | Batch 80/784 | Loss: 0.3100 | LM_LOSS: 0.2992 | LB_LOSS: 1.0803 [2026-04-17 11:09:36] Validation | Batch 90/784 | Loss: 0.3085 | LM_LOSS: 0.2977 | LB_LOSS: 1.0808 [2026-04-17 11:09:37] Validation | Batch 100/784 | Loss: 0.3104 | LM_LOSS: 0.2996 | LB_LOSS: 1.0812 [2026-04-17 11:09:38] Validation | Batch 110/784 | Loss: 0.3060 | LM_LOSS: 0.2952 | LB_LOSS: 1.0813 [2026-04-17 11:09:40] Validation | Batch 120/784 | Loss: 0.3094 | LM_LOSS: 0.2986 | LB_LOSS: 1.0812 [2026-04-17 11:09:41] Validation | Batch 130/784 | Loss: 0.3121 | LM_LOSS: 0.3013 | LB_LOSS: 1.0812 [2026-04-17 11:09:43] Validation | Batch 140/784 | Loss: 0.3115 | LM_LOSS: 0.3007 | LB_LOSS: 1.0810 [2026-04-17 11:09:44] Validation | Batch 150/784 | Loss: 0.3076 | LM_LOSS: 0.2968 | LB_LOSS: 1.0813 [2026-04-17 11:09:45] Validation | Batch 160/784 | Loss: 0.3082 | LM_LOSS: 0.2974 | LB_LOSS: 1.0810 [2026-04-17 11:09:47] Validation | Batch 170/784 | Loss: 0.3090 | LM_LOSS: 0.2982 | LB_LOSS: 1.0807 [2026-04-17 11:09:48] Validation | Batch 180/784 | Loss: 0.3065 | LM_LOSS: 0.2957 | LB_LOSS: 1.0807 [2026-04-17 11:09:50] Validation | Batch 190/784 | Loss: 0.3083 | LM_LOSS: 0.2974 | LB_LOSS: 1.0811 [2026-04-17 11:09:51] Validation | Batch 200/784 | Loss: 0.3085 | LM_LOSS: 0.2977 | LB_LOSS: 1.0812 [2026-04-17 11:09:52] Validation | Batch 210/784 | Loss: 0.3073 | LM_LOSS: 0.2965 | LB_LOSS: 1.0811 [2026-04-17 11:09:54] Validation | Batch 220/784 | Loss: 0.3080 | LM_LOSS: 0.2972 | LB_LOSS: 1.0811 [2026-04-17 11:09:55] Validation | Batch 230/784 | Loss: 0.3085 | LM_LOSS: 0.2977 | LB_LOSS: 1.0810 [2026-04-17 11:09:56] Validation | Batch 240/784 | Loss: 0.3089 | LM_LOSS: 0.2981 | LB_LOSS: 1.0814 [2026-04-17 11:09:58] Validation | Batch 250/784 | Loss: 0.3087 | LM_LOSS: 0.2978 | LB_LOSS: 1.0812 [2026-04-17 11:09:59] Validation | Batch 260/784 | Loss: 0.3089 | LM_LOSS: 0.2981 | LB_LOSS: 1.0814 [2026-04-17 11:10:01] Validation | Batch 270/784 | Loss: 0.3084 | LM_LOSS: 0.2976 | LB_LOSS: 1.0815 [2026-04-17 11:10:02] Validation | Batch 280/784 | Loss: 0.3090 | LM_LOSS: 0.2982 | LB_LOSS: 1.0816 [2026-04-17 11:10:04] Validation | Batch 290/784 | Loss: 0.3099 | LM_LOSS: 0.2991 | LB_LOSS: 1.0818 [2026-04-17 11:10:05] Validation | Batch 300/784 | Loss: 0.3106 | LM_LOSS: 0.2997 | LB_LOSS: 1.0818 [2026-04-17 11:10:06] Validation | Batch 310/784 | Loss: 0.3101 | LM_LOSS: 0.2992 | LB_LOSS: 1.0817 [2026-04-17 11:10:08] Validation | Batch 320/784 | Loss: 0.3115 | LM_LOSS: 0.3007 | LB_LOSS: 1.0817 [2026-04-17 11:10:09] Validation | Batch 330/784 | Loss: 0.3114 | LM_LOSS: 0.3006 | LB_LOSS: 1.0817 [2026-04-17 11:10:10] Validation | Batch 340/784 | Loss: 0.3105 | LM_LOSS: 0.2997 | LB_LOSS: 1.0818 [2026-04-17 11:10:12] Validation | Batch 350/784 | Loss: 0.3106 | LM_LOSS: 0.2998 | LB_LOSS: 1.0820 [2026-04-17 11:10:13] Validation | Batch 360/784 | Loss: 0.3104 | LM_LOSS: 0.2995 | LB_LOSS: 1.0820 [2026-04-17 11:10:14] Validation | Batch 370/784 | Loss: 0.3108 | LM_LOSS: 0.3000 | LB_LOSS: 1.0819 [2026-04-17 11:10:15] Validation | Batch 380/784 | Loss: 0.3106 | LM_LOSS: 0.2998 | LB_LOSS: 1.0820 [2026-04-17 11:10:17] Validation | Batch 390/784 | Loss: 0.3105 | LM_LOSS: 0.2997 | LB_LOSS: 1.0820 [2026-04-17 11:10:18] Validation | Batch 400/784 | Loss: 0.3107 | LM_LOSS: 0.2999 | LB_LOSS: 1.0820 [2026-04-17 11:10:19] Validation | Batch 410/784 | Loss: 0.3110 | LM_LOSS: 0.3001 | LB_LOSS: 1.0820 [2026-04-17 11:10:20] Validation | Batch 420/784 | Loss: 0.3111 | LM_LOSS: 0.3003 | LB_LOSS: 1.0821 [2026-04-17 11:10:22] Validation | Batch 430/784 | Loss: 0.3110 | LM_LOSS: 0.3002 | LB_LOSS: 1.0820 [2026-04-17 11:10:23] Validation | Batch 440/784 | Loss: 0.3106 | LM_LOSS: 0.2998 | LB_LOSS: 1.0821 [2026-04-17 11:10:24] Validation | Batch 450/784 | Loss: 0.3101 | LM_LOSS: 0.2992 | LB_LOSS: 1.0820 [2026-04-17 11:10:26] Validation | Batch 460/784 | Loss: 0.3105 | LM_LOSS: 0.2997 | LB_LOSS: 1.0821 [2026-04-17 11:10:27] Validation | Batch 470/784 | Loss: 0.3098 | LM_LOSS: 0.2990 | LB_LOSS: 1.0820 [2026-04-17 11:10:28] Validation | Batch 480/784 | Loss: 0.3102 | LM_LOSS: 0.2993 | LB_LOSS: 1.0820 [2026-04-17 11:10:30] Validation | Batch 490/784 | Loss: 0.3096 | LM_LOSS: 0.2988 | LB_LOSS: 1.0820 [2026-04-17 11:10:31] Validation | Batch 500/784 | Loss: 0.3100 | LM_LOSS: 0.2992 | LB_LOSS: 1.0819 [2026-04-17 11:10:32] Validation | Batch 510/784 | Loss: 0.3096 | LM_LOSS: 0.2988 | LB_LOSS: 1.0819 [2026-04-17 11:10:34] Validation | Batch 520/784 | Loss: 0.3097 | LM_LOSS: 0.2988 | LB_LOSS: 1.0818 [2026-04-17 11:10:35] Validation | Batch 530/784 | Loss: 0.3105 | LM_LOSS: 0.2996 | LB_LOSS: 1.0818 [2026-04-17 11:10:36] Validation | Batch 540/784 | Loss: 0.3109 | LM_LOSS: 0.3001 | LB_LOSS: 1.0818 [2026-04-17 11:10:38] Validation | Batch 550/784 | Loss: 0.3122 | LM_LOSS: 0.3014 | LB_LOSS: 1.0817 [2026-04-17 11:10:39] Validation | Batch 560/784 | Loss: 0.3122 | LM_LOSS: 0.3014 | LB_LOSS: 1.0818 [2026-04-17 11:10:41] Validation | Batch 570/784 | Loss: 0.3118 | LM_LOSS: 0.3010 | LB_LOSS: 1.0817 [2026-04-17 11:10:42] Validation | Batch 580/784 | Loss: 0.3113 | LM_LOSS: 0.3004 | LB_LOSS: 1.0817 [2026-04-17 11:10:43] Validation | Batch 590/784 | Loss: 0.3115 | LM_LOSS: 0.3007 | LB_LOSS: 1.0817 [2026-04-17 11:10:44] Validation | Batch 600/784 | Loss: 0.3114 | LM_LOSS: 0.3006 | LB_LOSS: 1.0816 [2026-04-17 11:10:46] Validation | Batch 610/784 | Loss: 0.3115 | LM_LOSS: 0.3007 | LB_LOSS: 1.0816 [2026-04-17 11:10:47] Validation | Batch 620/784 | Loss: 0.3114 | LM_LOSS: 0.3006 | LB_LOSS: 1.0816 [2026-04-17 11:10:49] Validation | Batch 630/784 | Loss: 0.3120 | LM_LOSS: 0.3012 | LB_LOSS: 1.0817 [2026-04-17 11:10:50] Validation | Batch 640/784 | Loss: 0.3121 | LM_LOSS: 0.3013 | LB_LOSS: 1.0816 [2026-04-17 11:10:52] Validation | Batch 650/784 | Loss: 0.3120 | LM_LOSS: 0.3012 | LB_LOSS: 1.0817 [2026-04-17 11:10:53] Validation | Batch 660/784 | Loss: 0.3123 | LM_LOSS: 0.3015 | LB_LOSS: 1.0817 [2026-04-17 11:10:55] Validation | Batch 670/784 | Loss: 0.3127 | LM_LOSS: 0.3019 | LB_LOSS: 1.0818 [2026-04-17 11:10:56] Validation | Batch 680/784 | Loss: 0.3125 | LM_LOSS: 0.3016 | LB_LOSS: 1.0817 [2026-04-17 11:10:57] Validation | Batch 690/784 | Loss: 0.3127 | LM_LOSS: 0.3019 | LB_LOSS: 1.0817 [2026-04-17 11:10:59] Validation | Batch 700/784 | Loss: 0.3127 | LM_LOSS: 0.3019 | LB_LOSS: 1.0816 [2026-04-17 11:11:00] Validation | Batch 710/784 | Loss: 0.3125 | LM_LOSS: 0.3017 | LB_LOSS: 1.0816 [2026-04-17 11:11:02] Validation | Batch 720/784 | Loss: 0.3122 | LM_LOSS: 0.3014 | LB_LOSS: 1.0815 [2026-04-17 11:11:03] Validation | Batch 730/784 | Loss: 0.3118 | LM_LOSS: 0.3010 | LB_LOSS: 1.0815 [2026-04-17 11:11:04] Validation | Batch 740/784 | Loss: 0.3119 | LM_LOSS: 0.3011 | LB_LOSS: 1.0816 [2026-04-17 11:11:05] Validation | Batch 750/784 | Loss: 0.3112 | LM_LOSS: 0.3004 | LB_LOSS: 1.0816 [2026-04-17 11:11:07] Validation | Batch 760/784 | Loss: 0.3114 | LM_LOSS: 0.3006 | LB_LOSS: 1.0816 [2026-04-17 11:11:08] Validation | Batch 770/784 | Loss: 0.3116 | LM_LOSS: 0.3007 | LB_LOSS: 1.0816 [2026-04-17 11:11:09] Validation | Batch 780/784 | Loss: 0.3119 | LM_LOSS: 0.3011 | LB_LOSS: 1.0816 [2026-04-17 11:11:10] Validation | Batch 784/784 | Loss: 0.3121 | LM_LOSS: 0.3012 | LB_LOSS: 1.0816 [2026-04-17 11:11:13] Validation | Loss: 0.3121 | LM_LOSS: 0.3012 | LB_LOSS: 1.0816 | PPL: 1.35 | Time: 106.12s [2026-04-17 11:11:18] New best model saved! Val loss: 0.3121 [2026-04-17 11:11:24] Epoch 1 | Step 10010 | Loss: 0.3163 | LM: 0.3047 | LB: 1.1065 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.433/SR1: 0.408 | LR: 1.00e-04 [2026-04-17 11:11:31] Epoch 1 | Step 10020 | Loss: 0.3163 | LM: 0.3048 | LB: 1.1065 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.433/SR1: 0.408 | LR: 1.00e-04 [2026-04-17 11:11:37] Epoch 1 | Step 10030 | Loss: 0.3163 | LM: 0.3048 | LB: 1.1065 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.433/SR1: 0.408 | LR: 1.00e-04 [2026-04-17 11:11:43] Epoch 1 | Step 10040 | Loss: 0.3162 | LM: 0.3048 | LB: 1.1065 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.433/SR1: 0.408 | LR: 1.00e-04 [2026-04-17 11:11:50] Epoch 1 | Step 10050 | Loss: 0.3162 | LM: 0.3048 | LB: 1.1064 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.433/SR1: 0.408 | LR: 1.00e-04 [2026-04-17 11:11:56] Epoch 1 | Step 10060 | Loss: 0.3162 | LM: 0.3047 | LB: 1.1064 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.433/SR1: 0.408 | LR: 1.00e-04 [2026-04-17 11:12:02] Epoch 1 | Step 10070 | Loss: 0.3162 | LM: 0.3048 | LB: 1.1064 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.433/SR1: 0.408 | LR: 1.00e-04 [2026-04-17 11:12:09] Epoch 1 | Step 10080 | Loss: 0.3162 | LM: 0.3048 | LB: 1.1064 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.433/SR1: 0.408 | LR: 1.00e-04 [2026-04-17 11:12:16] Epoch 1 | Step 10090 | Loss: 0.3163 | LM: 0.3048 | LB: 1.1064 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.433/SR1: 0.408 | LR: 1.00e-04 [2026-04-17 11:12:22] Epoch 1 | Step 10100 | Loss: 0.3162 | LM: 0.3047 | LB: 1.1064 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.433/SR1: 0.408 | LR: 1.00e-04 [2026-04-17 11:12:28] Epoch 1 | Step 10110 | Loss: 0.3162 | LM: 0.3046 | LB: 1.1063 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.433/SR1: 0.408 | LR: 1.00e-04 [2026-04-17 11:12:35] Epoch 1 | Step 10120 | Loss: 0.3162 | LM: 0.3046 | LB: 1.1063 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.433/SR1: 0.408 | LR: 1.00e-04 [2026-04-17 11:12:41] Epoch 1 | Step 10130 | Loss: 0.3162 | LM: 0.3047 | LB: 1.1063 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.433/SR1: 0.408 | LR: 1.00e-04 [2026-04-17 11:12:47] Epoch 1 | Step 10140 | Loss: 0.3163 | LM: 0.3047 | LB: 1.1063 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.433/SR1: 0.408 | LR: 1.00e-04 [2026-04-17 11:12:54] Epoch 1 | Step 10150 | Loss: 0.3162 | LM: 0.3046 | LB: 1.1063 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.433/SR1: 0.408 | LR: 1.00e-04 [2026-04-17 11:13:00] Epoch 1 | Step 10160 | Loss: 0.3162 | LM: 0.3046 | LB: 1.1062 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.433/SR1: 0.408 | LR: 1.00e-04 [2026-04-17 11:13:07] Epoch 1 | Step 10170 | Loss: 0.3162 | LM: 0.3045 | LB: 1.1062 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.433/SR1: 0.408 | LR: 1.00e-04 [2026-04-17 11:13:13] Epoch 1 | Step 10180 | Loss: 0.3161 | LM: 0.3045 | LB: 1.1062 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.433/SR1: 0.408 | LR: 1.00e-04 [2026-04-17 11:13:20] Epoch 1 | Step 10190 | Loss: 0.3162 | LM: 0.3045 | LB: 1.1062 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.433/SR1: 0.408 | LR: 1.00e-04 [2026-04-17 11:13:26] Epoch 1 | Step 10200 | Loss: 0.3161 | LM: 0.3045 | LB: 1.1062 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.433/SR1: 0.408 | LR: 1.00e-04 [2026-04-17 11:13:32] Epoch 1 | Step 10210 | Loss: 0.3161 | LM: 0.3045 | LB: 1.1061 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.433/SR1: 0.408 | LR: 1.00e-04 [2026-04-17 11:13:38] Epoch 1 | Step 10220 | Loss: 0.3161 | LM: 0.3044 | LB: 1.1061 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.433/SR1: 0.408 | LR: 1.00e-04 [2026-04-17 11:13:45] Epoch 1 | Step 10230 | Loss: 0.3161 | LM: 0.3044 | LB: 1.1061 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.433/SR1: 0.408 | LR: 1.00e-04 [2026-04-17 11:13:51] Epoch 1 | Step 10240 | Loss: 0.3161 | LM: 0.3044 | LB: 1.1061 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.433/SR1: 0.408 | LR: 1.00e-04 [2026-04-17 11:13:58] Epoch 1 | Step 10250 | Loss: 0.3161 | LM: 0.3044 | LB: 1.1061 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.433/SR1: 0.408 | LR: 1.00e-04 [2026-04-17 11:14:04] Epoch 1 | Step 10260 | Loss: 0.3161 | LM: 0.3045 | LB: 1.1060 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.433/SR1: 0.408 | LR: 1.00e-04 [2026-04-17 11:14:11] Epoch 1 | Step 10270 | Loss: 0.3161 | LM: 0.3044 | LB: 1.1060 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.433/SR1: 0.408 | LR: 1.00e-04 [2026-04-17 11:14:17] Epoch 1 | Step 10280 | Loss: 0.3161 | LM: 0.3045 | LB: 1.1060 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.433/SR1: 0.408 | LR: 1.00e-04 [2026-04-17 11:14:23] Epoch 1 | Step 10290 | Loss: 0.3161 | LM: 0.3044 | LB: 1.1060 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.433/SR1: 0.408 | LR: 1.00e-04 [2026-04-17 11:14:30] Epoch 1 | Step 10300 | Loss: 0.3161 | LM: 0.3044 | LB: 1.1059 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.433/SR1: 0.408 | LR: 1.00e-04 [2026-04-17 11:14:36] Epoch 1 | Step 10310 | Loss: 0.3160 | LM: 0.3043 | LB: 1.1059 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.433/SR1: 0.408 | LR: 1.00e-04 [2026-04-17 11:14:42] Epoch 1 | Step 10320 | Loss: 0.3160 | LM: 0.3042 | LB: 1.1059 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.433/SR1: 0.408 | LR: 1.00e-04 [2026-04-17 11:14:48] Epoch 1 | Step 10330 | Loss: 0.3160 | LM: 0.3042 | LB: 1.1059 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.433/SR1: 0.408 | LR: 1.00e-04 [2026-04-17 11:14:55] Epoch 1 | Step 10340 | Loss: 0.3160 | LM: 0.3042 | LB: 1.1058 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.433/SR1: 0.408 | LR: 1.00e-04 [2026-04-17 11:15:01] Epoch 1 | Step 10350 | Loss: 0.3160 | LM: 0.3042 | LB: 1.1058 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.433/SR1: 0.408 | LR: 1.00e-04 [2026-04-17 11:15:08] Epoch 1 | Step 10360 | Loss: 0.3160 | LM: 0.3042 | LB: 1.1058 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.433/SR1: 0.408 | LR: 1.00e-04 [2026-04-17 11:15:14] Epoch 1 | Step 10370 | Loss: 0.3160 | LM: 0.3043 | LB: 1.1058 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.433/SR1: 0.408 | LR: 1.00e-04 [2026-04-17 11:15:20] Epoch 1 | Step 10380 | Loss: 0.3160 | LM: 0.3044 | LB: 1.1058 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.433/SR1: 0.407 | LR: 1.00e-04 [2026-04-17 11:15:27] Epoch 1 | Step 10390 | Loss: 0.3160 | LM: 0.3043 | LB: 1.1058 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.433/SR1: 0.407 | LR: 1.00e-04 [2026-04-17 11:15:33] Epoch 1 | Step 10400 | Loss: 0.3160 | LM: 0.3043 | LB: 1.1057 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.433/SR1: 0.407 | LR: 1.00e-04 [2026-04-17 11:15:39] Epoch 1 | Step 10410 | Loss: 0.3160 | LM: 0.3043 | LB: 1.1057 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.433/SR1: 0.407 | LR: 1.00e-04 [2026-04-17 11:15:46] Epoch 1 | Step 10420 | Loss: 0.3159 | LM: 0.3043 | LB: 1.1057 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.433/SR1: 0.407 | LR: 1.00e-04 [2026-04-17 11:15:52] Epoch 1 | Step 10430 | Loss: 0.3159 | LM: 0.3042 | LB: 1.1057 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.433/SR1: 0.407 | LR: 1.00e-04 [2026-04-17 11:15:59] Epoch 1 | Step 10440 | Loss: 0.3159 | LM: 0.3041 | LB: 1.1057 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.433/SR1: 0.407 | LR: 1.00e-04 [2026-04-17 11:16:05] Epoch 1 | Step 10450 | Loss: 0.3159 | LM: 0.3041 | LB: 1.1056 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.433/SR1: 0.407 | LR: 1.00e-04 [2026-04-17 11:16:11] Epoch 1 | Step 10460 | Loss: 0.3159 | LM: 0.3042 | LB: 1.1056 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.433/SR1: 0.407 | LR: 1.00e-04 [2026-04-17 11:16:18] Epoch 1 | Step 10470 | Loss: 0.3159 | LM: 0.3042 | LB: 1.1056 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.433/SR1: 0.407 | LR: 1.00e-04 [2026-04-17 11:16:24] Epoch 1 | Step 10480 | Loss: 0.3159 | LM: 0.3043 | LB: 1.1056 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.433/SR1: 0.407 | LR: 1.00e-04 [2026-04-17 11:16:31] Epoch 1 | Step 10490 | Loss: 0.3159 | LM: 0.3042 | LB: 1.1056 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.433/SR1: 0.407 | LR: 1.00e-04 [2026-04-17 11:16:37] Epoch 1 | Step 10500 | Loss: 0.3158 | LM: 0.3042 | LB: 1.1055 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.432/SR1: 0.407 | LR: 1.00e-04 [2026-04-17 11:16:43] Epoch 1 | Step 10510 | Loss: 0.3158 | LM: 0.3041 | LB: 1.1055 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.432/SR1: 0.407 | LR: 1.00e-04 [2026-04-17 11:16:50] Epoch 1 | Step 10520 | Loss: 0.3158 | LM: 0.3041 | LB: 1.1055 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.432/SR1: 0.407 | LR: 1.00e-04 [2026-04-17 11:16:56] Epoch 1 | Step 10530 | Loss: 0.3157 | LM: 0.3041 | LB: 1.1055 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.432/SR1: 0.407 | LR: 1.00e-04 [2026-04-17 11:17:03] Epoch 1 | Step 10540 | Loss: 0.3157 | LM: 0.3041 | LB: 1.1055 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.432/SR1: 0.407 | LR: 1.00e-04 [2026-04-17 11:17:09] Epoch 1 | Step 10550 | Loss: 0.3157 | LM: 0.3041 | LB: 1.1055 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.432/SR1: 0.407 | LR: 1.00e-04 [2026-04-17 11:17:15] Epoch 1 | Step 10560 | Loss: 0.3157 | LM: 0.3040 | LB: 1.1055 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.432/SR1: 0.407 | LR: 1.00e-04 [2026-04-17 11:17:22] Epoch 1 | Step 10570 | Loss: 0.3156 | LM: 0.3039 | LB: 1.1054 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.432/SR1: 0.407 | LR: 1.00e-04 [2026-04-17 11:17:28] Epoch 1 | Step 10580 | Loss: 0.3156 | LM: 0.3039 | LB: 1.1054 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.432/SR1: 0.407 | LR: 1.00e-04 [2026-04-17 11:17:35] Epoch 1 | Step 10590 | Loss: 0.3156 | LM: 0.3039 | LB: 1.1054 | CL0: 2.9 | CL1: 2.3 | HR0: 0.347/SR0: 0.347 | HR1: 0.432/SR1: 0.407 | LR: 1.00e-04 [2026-04-17 11:17:36] Epoch 1 completed in 8200.82s | Loss: 0.3156 | CL0: 2.9 | CL1: 2.3 [2026-04-17 11:17:45] Checkpoint saved: outputs/2026-04-17/08-57-56/checkpoints/checkpoint_step_10591.pt [2026-04-17 11:17:59] ============================================================ [2026-04-17 11:17:59] EPOCH 2/3 [2026-04-17 11:17:59] ============================================================ [2026-04-17 11:18:05] Epoch 2 | Step 10600 | Loss: 0.2437 | LM: 0.2207 | LB: 1.0879 | CL0: 3.0 | CL1: 2.4 | HR0: 0.338/SR0: 0.339 | HR1: 0.421/SR1: 0.390 | LR: 1.00e-04 [2026-04-17 11:18:12] Epoch 2 | Step 10610 | Loss: 0.2243 | LM: 0.1950 | LB: 1.0900 | CL0: 3.0 | CL1: 2.4 | HR0: 0.338/SR0: 0.339 | HR1: 0.423/SR1: 0.392 | LR: 1.00e-04 [2026-04-17 11:18:18] Epoch 2 | Step 10620 | Loss: 0.2218 | LM: 0.1985 | LB: 1.0911 | CL0: 2.9 | CL1: 2.4 | HR0: 0.341/SR0: 0.342 | HR1: 0.424/SR1: 0.391 | LR: 1.00e-04 [2026-04-17 11:18:24] Epoch 2 | Step 10630 | Loss: 0.2262 | LM: 0.2213 | LB: 1.0916 | CL0: 2.9 | CL1: 2.4 | HR0: 0.344/SR0: 0.344 | HR1: 0.423/SR1: 0.390 | LR: 1.00e-04 [2026-04-17 11:18:31] Epoch 2 | Step 10640 | Loss: 0.2262 | LM: 0.2167 | LB: 1.0893 | CL0: 2.9 | CL1: 2.4 | HR0: 0.346/SR0: 0.345 | HR1: 0.418/SR1: 0.387 | LR: 1.00e-04 [2026-04-17 11:18:38] Epoch 2 | Step 10650 | Loss: 0.2299 | LM: 0.2133 | LB: 1.0884 | CL0: 2.9 | CL1: 2.4 | HR0: 0.346/SR0: 0.345 | HR1: 0.418/SR1: 0.386 | LR: 1.00e-04 [2026-04-17 11:18:44] Epoch 2 | Step 10660 | Loss: 0.2275 | LM: 0.2090 | LB: 1.0878 | CL0: 2.9 | CL1: 2.4 | HR0: 0.345/SR0: 0.345 | HR1: 0.417/SR1: 0.386 | LR: 1.00e-04 [2026-04-17 11:18:50] Epoch 2 | Step 10670 | Loss: 0.2330 | LM: 0.2153 | LB: 1.0898 | CL0: 2.9 | CL1: 2.4 | HR0: 0.349/SR0: 0.347 | HR1: 0.415/SR1: 0.385 | LR: 1.00e-04 [2026-04-17 11:18:57] Epoch 2 | Step 10680 | Loss: 0.2343 | LM: 0.2122 | LB: 1.0900 | CL0: 2.9 | CL1: 2.4 | HR0: 0.349/SR0: 0.347 | HR1: 0.414/SR1: 0.384 | LR: 1.00e-04 [2026-04-17 11:19:03] Epoch 2 | Step 10690 | Loss: 0.2362 | LM: 0.2117 | LB: 1.0905 | CL0: 2.9 | CL1: 2.4 | HR0: 0.350/SR0: 0.348 | HR1: 0.415/SR1: 0.385 | LR: 1.00e-04 [2026-04-17 11:19:09] Epoch 2 | Step 10700 | Loss: 0.2365 | LM: 0.2194 | LB: 1.0901 | CL0: 2.9 | CL1: 2.4 | HR0: 0.349/SR0: 0.347 | HR1: 0.415/SR1: 0.385 | LR: 1.00e-04 [2026-04-17 11:19:16] Epoch 2 | Step 10710 | Loss: 0.2374 | LM: 0.2179 | LB: 1.0909 | CL0: 2.9 | CL1: 2.4 | HR0: 0.349/SR0: 0.347 | HR1: 0.416/SR1: 0.386 | LR: 1.00e-04 [2026-04-17 11:19:22] Epoch 2 | Step 10720 | Loss: 0.2396 | LM: 0.2175 | LB: 1.0900 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.386 | LR: 1.00e-04 [2026-04-17 11:19:28] Epoch 2 | Step 10730 | Loss: 0.2413 | LM: 0.2262 | LB: 1.0911 | CL0: 2.9 | CL1: 2.4 | HR0: 0.349/SR0: 0.348 | HR1: 0.416/SR1: 0.386 | LR: 1.00e-04 [2026-04-17 11:19:35] Epoch 2 | Step 10740 | Loss: 0.2403 | LM: 0.2263 | LB: 1.0907 | CL0: 2.9 | CL1: 2.4 | HR0: 0.349/SR0: 0.347 | HR1: 0.416/SR1: 0.386 | LR: 1.00e-04 [2026-04-17 11:19:41] Epoch 2 | Step 10750 | Loss: 0.2414 | LM: 0.2289 | LB: 1.0908 | CL0: 2.9 | CL1: 2.4 | HR0: 0.350/SR0: 0.348 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-04 [2026-04-17 11:19:48] Epoch 2 | Step 10760 | Loss: 0.2427 | LM: 0.2300 | LB: 1.0911 | CL0: 2.9 | CL1: 2.4 | HR0: 0.350/SR0: 0.348 | HR1: 0.416/SR1: 0.386 | LR: 1.00e-04 [2026-04-17 11:19:54] Epoch 2 | Step 10770 | Loss: 0.2422 | LM: 0.2321 | LB: 1.0913 | CL0: 2.9 | CL1: 2.4 | HR0: 0.349/SR0: 0.348 | HR1: 0.417/SR1: 0.386 | LR: 1.00e-04 [2026-04-17 11:20:00] Epoch 2 | Step 10780 | Loss: 0.2410 | LM: 0.2311 | LB: 1.0914 | CL0: 2.9 | CL1: 2.4 | HR0: 0.349/SR0: 0.348 | HR1: 0.417/SR1: 0.387 | LR: 1.00e-04 [2026-04-17 11:20:07] Epoch 2 | Step 10790 | Loss: 0.2406 | LM: 0.2300 | LB: 1.0910 | CL0: 2.9 | CL1: 2.4 | HR0: 0.349/SR0: 0.347 | HR1: 0.417/SR1: 0.387 | LR: 1.00e-04 [2026-04-17 11:20:13] Epoch 2 | Step 10800 | Loss: 0.2413 | LM: 0.2275 | LB: 1.0904 | CL0: 2.9 | CL1: 2.4 | HR0: 0.349/SR0: 0.347 | HR1: 0.416/SR1: 0.387 | LR: 1.00e-04 [2026-04-17 11:20:19] Epoch 2 | Step 10810 | Loss: 0.2424 | LM: 0.2280 | LB: 1.0910 | CL0: 2.9 | CL1: 2.4 | HR0: 0.349/SR0: 0.348 | HR1: 0.417/SR1: 0.387 | LR: 1.00e-04 [2026-04-17 11:20:26] Epoch 2 | Step 10820 | Loss: 0.2422 | LM: 0.2292 | LB: 1.0913 | CL0: 2.9 | CL1: 2.4 | HR0: 0.349/SR0: 0.348 | HR1: 0.417/SR1: 0.387 | LR: 1.00e-04 [2026-04-17 11:20:32] Epoch 2 | Step 10830 | Loss: 0.2412 | LM: 0.2281 | LB: 1.0912 | CL0: 2.9 | CL1: 2.4 | HR0: 0.349/SR0: 0.348 | HR1: 0.417/SR1: 0.387 | LR: 1.00e-04 [2026-04-17 11:20:39] Epoch 2 | Step 10840 | Loss: 0.2418 | LM: 0.2283 | LB: 1.0914 | CL0: 2.9 | CL1: 2.4 | HR0: 0.350/SR0: 0.348 | HR1: 0.417/SR1: 0.387 | LR: 1.00e-04 [2026-04-17 11:20:45] Epoch 2 | Step 10850 | Loss: 0.2404 | LM: 0.2260 | LB: 1.0912 | CL0: 2.9 | CL1: 2.4 | HR0: 0.349/SR0: 0.348 | HR1: 0.417/SR1: 0.387 | LR: 1.00e-04 [2026-04-17 11:20:52] Epoch 2 | Step 10860 | Loss: 0.2397 | LM: 0.2245 | LB: 1.0911 | CL0: 2.9 | CL1: 2.4 | HR0: 0.349/SR0: 0.348 | HR1: 0.417/SR1: 0.387 | LR: 1.00e-04 [2026-04-17 11:20:58] Epoch 2 | Step 10870 | Loss: 0.2401 | LM: 0.2252 | LB: 1.0911 | CL0: 2.9 | CL1: 2.4 | HR0: 0.349/SR0: 0.348 | HR1: 0.417/SR1: 0.387 | LR: 1.00e-04 [2026-04-17 11:21:04] Epoch 2 | Step 10880 | Loss: 0.2399 | LM: 0.2255 | LB: 1.0909 | CL0: 2.9 | CL1: 2.4 | HR0: 0.349/SR0: 0.348 | HR1: 0.417/SR1: 0.387 | LR: 1.00e-04 [2026-04-17 11:21:11] Epoch 2 | Step 10890 | Loss: 0.2406 | LM: 0.2260 | LB: 1.0912 | CL0: 2.9 | CL1: 2.4 | HR0: 0.350/SR0: 0.348 | HR1: 0.417/SR1: 0.387 | LR: 1.00e-04 [2026-04-17 11:21:17] Epoch 2 | Step 10900 | Loss: 0.2407 | LM: 0.2258 | LB: 1.0913 | CL0: 2.9 | CL1: 2.4 | HR0: 0.350/SR0: 0.348 | HR1: 0.417/SR1: 0.387 | LR: 1.00e-04 [2026-04-17 11:21:23] Epoch 2 | Step 10910 | Loss: 0.2400 | LM: 0.2253 | LB: 1.0914 | CL0: 2.9 | CL1: 2.4 | HR0: 0.350/SR0: 0.348 | HR1: 0.417/SR1: 0.387 | LR: 1.00e-04 [2026-04-17 11:21:30] Epoch 2 | Step 10920 | Loss: 0.2406 | LM: 0.2262 | LB: 1.0917 | CL0: 2.9 | CL1: 2.4 | HR0: 0.350/SR0: 0.348 | HR1: 0.418/SR1: 0.388 | LR: 1.00e-04 [2026-04-17 11:21:36] Epoch 2 | Step 10930 | Loss: 0.2403 | LM: 0.2250 | LB: 1.0920 | CL0: 2.9 | CL1: 2.4 | HR0: 0.350/SR0: 0.348 | HR1: 0.418/SR1: 0.388 | LR: 1.00e-04 [2026-04-17 11:21:42] Epoch 2 | Step 10940 | Loss: 0.2401 | LM: 0.2250 | LB: 1.0919 | CL0: 2.9 | CL1: 2.4 | HR0: 0.350/SR0: 0.348 | HR1: 0.418/SR1: 0.388 | LR: 1.00e-04 [2026-04-17 11:21:48] Epoch 2 | Step 10950 | Loss: 0.2403 | LM: 0.2256 | LB: 1.0921 | CL0: 2.9 | CL1: 2.4 | HR0: 0.349/SR0: 0.348 | HR1: 0.419/SR1: 0.389 | LR: 1.00e-04 [2026-04-17 11:21:55] Epoch 2 | Step 10960 | Loss: 0.2402 | LM: 0.2258 | LB: 1.0920 | CL0: 2.9 | CL1: 2.4 | HR0: 0.349/SR0: 0.348 | HR1: 0.419/SR1: 0.389 | LR: 1.00e-04 [2026-04-17 11:22:01] Epoch 2 | Step 10970 | Loss: 0.2405 | LM: 0.2279 | LB: 1.0918 | CL0: 2.9 | CL1: 2.4 | HR0: 0.349/SR0: 0.348 | HR1: 0.418/SR1: 0.388 | LR: 1.00e-04 [2026-04-17 11:22:08] Epoch 2 | Step 10980 | Loss: 0.2404 | LM: 0.2282 | LB: 1.0917 | CL0: 2.9 | CL1: 2.4 | HR0: 0.349/SR0: 0.348 | HR1: 0.418/SR1: 0.388 | LR: 1.00e-04 [2026-04-17 11:22:14] Epoch 2 | Step 10990 | Loss: 0.2396 | LM: 0.2275 | LB: 1.0919 | CL0: 2.9 | CL1: 2.4 | HR0: 0.349/SR0: 0.348 | HR1: 0.418/SR1: 0.388 | LR: 1.00e-04 [2026-04-17 11:22:20] Epoch 2 | Step 11000 | Loss: 0.2397 | LM: 0.2286 | LB: 1.0918 | CL0: 2.9 | CL1: 2.4 | HR0: 0.349/SR0: 0.348 | HR1: 0.418/SR1: 0.388 | LR: 1.00e-04 [2026-04-17 11:22:21] Validation | Batch 10/784 | Loss: 0.3278 | LM_LOSS: 0.3169 | LB_LOSS: 1.0879 [2026-04-17 11:22:23] Validation | Batch 20/784 | Loss: 0.3408 | LM_LOSS: 0.3299 | LB_LOSS: 1.0879 [2026-04-17 11:22:24] Validation | Batch 30/784 | Loss: 0.3263 | LM_LOSS: 0.3154 | LB_LOSS: 1.0870 [2026-04-17 11:22:26] Validation | Batch 40/784 | Loss: 0.3273 | LM_LOSS: 0.3164 | LB_LOSS: 1.0870 [2026-04-17 11:22:27] Validation | Batch 50/784 | Loss: 0.3240 | LM_LOSS: 0.3131 | LB_LOSS: 1.0863 [2026-04-17 11:22:28] Validation | Batch 60/784 | Loss: 0.3250 | LM_LOSS: 0.3141 | LB_LOSS: 1.0858 [2026-04-17 11:22:30] Validation | Batch 70/784 | Loss: 0.3227 | LM_LOSS: 0.3118 | LB_LOSS: 1.0851 [2026-04-17 11:22:31] Validation | Batch 80/784 | Loss: 0.3188 | LM_LOSS: 0.3080 | LB_LOSS: 1.0847 [2026-04-17 11:22:32] Validation | Batch 90/784 | Loss: 0.3177 | LM_LOSS: 0.3068 | LB_LOSS: 1.0852 [2026-04-17 11:22:34] Validation | Batch 100/784 | Loss: 0.3202 | LM_LOSS: 0.3093 | LB_LOSS: 1.0857 [2026-04-17 11:22:35] Validation | Batch 110/784 | Loss: 0.3151 | LM_LOSS: 0.3042 | LB_LOSS: 1.0858 [2026-04-17 11:22:36] Validation | Batch 120/784 | Loss: 0.3185 | LM_LOSS: 0.3077 | LB_LOSS: 1.0857 [2026-04-17 11:22:38] Validation | Batch 130/784 | Loss: 0.3209 | LM_LOSS: 0.3101 | LB_LOSS: 1.0857 [2026-04-17 11:22:39] Validation | Batch 140/784 | Loss: 0.3202 | LM_LOSS: 0.3093 | LB_LOSS: 1.0855 [2026-04-17 11:22:41] Validation | Batch 150/784 | Loss: 0.3162 | LM_LOSS: 0.3053 | LB_LOSS: 1.0858 [2026-04-17 11:22:42] Validation | Batch 160/784 | Loss: 0.3171 | LM_LOSS: 0.3062 | LB_LOSS: 1.0855 [2026-04-17 11:22:44] Validation | Batch 170/784 | Loss: 0.3175 | LM_LOSS: 0.3066 | LB_LOSS: 1.0852 [2026-04-17 11:22:45] Validation | Batch 180/784 | Loss: 0.3150 | LM_LOSS: 0.3042 | LB_LOSS: 1.0852 [2026-04-17 11:22:46] Validation | Batch 190/784 | Loss: 0.3171 | LM_LOSS: 0.3062 | LB_LOSS: 1.0856 [2026-04-17 11:22:47] Validation | Batch 200/784 | Loss: 0.3174 | LM_LOSS: 0.3065 | LB_LOSS: 1.0857 [2026-04-17 11:22:49] Validation | Batch 210/784 | Loss: 0.3162 | LM_LOSS: 0.3053 | LB_LOSS: 1.0856 [2026-04-17 11:22:50] Validation | Batch 220/784 | Loss: 0.3173 | LM_LOSS: 0.3065 | LB_LOSS: 1.0856 [2026-04-17 11:22:52] Validation | Batch 230/784 | Loss: 0.3178 | LM_LOSS: 0.3070 | LB_LOSS: 1.0855 [2026-04-17 11:22:53] Validation | Batch 240/784 | Loss: 0.3182 | LM_LOSS: 0.3073 | LB_LOSS: 1.0859 [2026-04-17 11:22:55] Validation | Batch 250/784 | Loss: 0.3181 | LM_LOSS: 0.3072 | LB_LOSS: 1.0858 [2026-04-17 11:22:56] Validation | Batch 260/784 | Loss: 0.3182 | LM_LOSS: 0.3073 | LB_LOSS: 1.0860 [2026-04-17 11:22:58] Validation | Batch 270/784 | Loss: 0.3178 | LM_LOSS: 0.3069 | LB_LOSS: 1.0860 [2026-04-17 11:22:59] Validation | Batch 280/784 | Loss: 0.3184 | LM_LOSS: 0.3075 | LB_LOSS: 1.0862 [2026-04-17 11:23:00] Validation | Batch 290/784 | Loss: 0.3194 | LM_LOSS: 0.3085 | LB_LOSS: 1.0863 [2026-04-17 11:23:02] Validation | Batch 300/784 | Loss: 0.3201 | LM_LOSS: 0.3093 | LB_LOSS: 1.0864 [2026-04-17 11:23:03] Validation | Batch 310/784 | Loss: 0.3196 | LM_LOSS: 0.3087 | LB_LOSS: 1.0863 [2026-04-17 11:23:04] Validation | Batch 320/784 | Loss: 0.3211 | LM_LOSS: 0.3103 | LB_LOSS: 1.0863 [2026-04-17 11:23:06] Validation | Batch 330/784 | Loss: 0.3211 | LM_LOSS: 0.3102 | LB_LOSS: 1.0863 [2026-04-17 11:23:07] Validation | Batch 340/784 | Loss: 0.3199 | LM_LOSS: 0.3091 | LB_LOSS: 1.0864 [2026-04-17 11:23:08] Validation | Batch 350/784 | Loss: 0.3200 | LM_LOSS: 0.3091 | LB_LOSS: 1.0866 [2026-04-17 11:23:09] Validation | Batch 360/784 | Loss: 0.3197 | LM_LOSS: 0.3089 | LB_LOSS: 1.0866 [2026-04-17 11:23:11] Validation | Batch 370/784 | Loss: 0.3201 | LM_LOSS: 0.3092 | LB_LOSS: 1.0865 [2026-04-17 11:23:12] Validation | Batch 380/784 | Loss: 0.3199 | LM_LOSS: 0.3090 | LB_LOSS: 1.0866 [2026-04-17 11:23:13] Validation | Batch 390/784 | Loss: 0.3197 | LM_LOSS: 0.3089 | LB_LOSS: 1.0866 [2026-04-17 11:23:15] Validation | Batch 400/784 | Loss: 0.3200 | LM_LOSS: 0.3091 | LB_LOSS: 1.0866 [2026-04-17 11:23:16] Validation | Batch 410/784 | Loss: 0.3203 | LM_LOSS: 0.3094 | LB_LOSS: 1.0866 [2026-04-17 11:23:17] Validation | Batch 420/784 | Loss: 0.3204 | LM_LOSS: 0.3096 | LB_LOSS: 1.0867 [2026-04-17 11:23:18] Validation | Batch 430/784 | Loss: 0.3205 | LM_LOSS: 0.3096 | LB_LOSS: 1.0866 [2026-04-17 11:23:20] Validation | Batch 440/784 | Loss: 0.3201 | LM_LOSS: 0.3093 | LB_LOSS: 1.0867 [2026-04-17 11:23:21] Validation | Batch 450/784 | Loss: 0.3194 | LM_LOSS: 0.3085 | LB_LOSS: 1.0866 [2026-04-17 11:23:22] Validation | Batch 460/784 | Loss: 0.3199 | LM_LOSS: 0.3090 | LB_LOSS: 1.0867 [2026-04-17 11:23:24] Validation | Batch 470/784 | Loss: 0.3192 | LM_LOSS: 0.3083 | LB_LOSS: 1.0866 [2026-04-17 11:23:25] Validation | Batch 480/784 | Loss: 0.3196 | LM_LOSS: 0.3088 | LB_LOSS: 1.0866 [2026-04-17 11:23:27] Validation | Batch 490/784 | Loss: 0.3191 | LM_LOSS: 0.3082 | LB_LOSS: 1.0865 [2026-04-17 11:23:28] Validation | Batch 500/784 | Loss: 0.3195 | LM_LOSS: 0.3086 | LB_LOSS: 1.0865 [2026-04-17 11:23:29] Validation | Batch 510/784 | Loss: 0.3191 | LM_LOSS: 0.3082 | LB_LOSS: 1.0865 [2026-04-17 11:23:31] Validation | Batch 520/784 | Loss: 0.3192 | LM_LOSS: 0.3083 | LB_LOSS: 1.0864 [2026-04-17 11:23:32] Validation | Batch 530/784 | Loss: 0.3201 | LM_LOSS: 0.3092 | LB_LOSS: 1.0863 [2026-04-17 11:23:33] Validation | Batch 540/784 | Loss: 0.3204 | LM_LOSS: 0.3096 | LB_LOSS: 1.0863 [2026-04-17 11:23:35] Validation | Batch 550/784 | Loss: 0.3218 | LM_LOSS: 0.3109 | LB_LOSS: 1.0863 [2026-04-17 11:23:36] Validation | Batch 560/784 | Loss: 0.3218 | LM_LOSS: 0.3110 | LB_LOSS: 1.0864 [2026-04-17 11:23:38] Validation | Batch 570/784 | Loss: 0.3213 | LM_LOSS: 0.3105 | LB_LOSS: 1.0863 [2026-04-17 11:23:39] Validation | Batch 580/784 | Loss: 0.3208 | LM_LOSS: 0.3100 | LB_LOSS: 1.0863 [2026-04-17 11:23:40] Validation | Batch 590/784 | Loss: 0.3211 | LM_LOSS: 0.3103 | LB_LOSS: 1.0862 [2026-04-17 11:23:42] Validation | Batch 600/784 | Loss: 0.3210 | LM_LOSS: 0.3101 | LB_LOSS: 1.0862 [2026-04-17 11:23:43] Validation | Batch 610/784 | Loss: 0.3211 | LM_LOSS: 0.3103 | LB_LOSS: 1.0862 [2026-04-17 11:23:44] Validation | Batch 620/784 | Loss: 0.3210 | LM_LOSS: 0.3102 | LB_LOSS: 1.0862 [2026-04-17 11:23:46] Validation | Batch 630/784 | Loss: 0.3217 | LM_LOSS: 0.3109 | LB_LOSS: 1.0862 [2026-04-17 11:23:48] Validation | Batch 640/784 | Loss: 0.3217 | LM_LOSS: 0.3108 | LB_LOSS: 1.0862 [2026-04-17 11:23:49] Validation | Batch 650/784 | Loss: 0.3216 | LM_LOSS: 0.3107 | LB_LOSS: 1.0863 [2026-04-17 11:23:51] Validation | Batch 660/784 | Loss: 0.3218 | LM_LOSS: 0.3110 | LB_LOSS: 1.0862 [2026-04-17 11:23:52] Validation | Batch 670/784 | Loss: 0.3223 | LM_LOSS: 0.3114 | LB_LOSS: 1.0863 [2026-04-17 11:23:53] Validation | Batch 680/784 | Loss: 0.3220 | LM_LOSS: 0.3112 | LB_LOSS: 1.0863 [2026-04-17 11:23:55] Validation | Batch 690/784 | Loss: 0.3222 | LM_LOSS: 0.3114 | LB_LOSS: 1.0862 [2026-04-17 11:23:56] Validation | Batch 700/784 | Loss: 0.3223 | LM_LOSS: 0.3114 | LB_LOSS: 1.0862 [2026-04-17 11:23:57] Validation | Batch 710/784 | Loss: 0.3221 | LM_LOSS: 0.3112 | LB_LOSS: 1.0861 [2026-04-17 11:23:59] Validation | Batch 720/784 | Loss: 0.3218 | LM_LOSS: 0.3109 | LB_LOSS: 1.0861 [2026-04-17 11:24:00] Validation | Batch 730/784 | Loss: 0.3212 | LM_LOSS: 0.3104 | LB_LOSS: 1.0860 [2026-04-17 11:24:02] Validation | Batch 740/784 | Loss: 0.3213 | LM_LOSS: 0.3105 | LB_LOSS: 1.0861 [2026-04-17 11:24:03] Validation | Batch 750/784 | Loss: 0.3207 | LM_LOSS: 0.3098 | LB_LOSS: 1.0861 [2026-04-17 11:24:04] Validation | Batch 760/784 | Loss: 0.3208 | LM_LOSS: 0.3100 | LB_LOSS: 1.0861 [2026-04-17 11:24:05] Validation | Batch 770/784 | Loss: 0.3210 | LM_LOSS: 0.3102 | LB_LOSS: 1.0862 [2026-04-17 11:24:07] Validation | Batch 780/784 | Loss: 0.3213 | LM_LOSS: 0.3105 | LB_LOSS: 1.0861 [2026-04-17 11:24:07] Validation | Batch 784/784 | Loss: 0.3215 | LM_LOSS: 0.3107 | LB_LOSS: 1.0861 [2026-04-17 11:24:10] Validation | Loss: 0.3215 | LM_LOSS: 0.3107 | LB_LOSS: 1.0861 | PPL: 1.36 | Time: 107.07s [2026-04-17 11:24:16] Epoch 2 | Step 11010 | Loss: 0.2393 | LM: 0.2287 | LB: 1.0915 | CL0: 2.9 | CL1: 2.4 | HR0: 0.349/SR0: 0.348 | HR1: 0.418/SR1: 0.388 | LR: 1.00e-04 [2026-04-17 11:24:23] Epoch 2 | Step 11020 | Loss: 0.2393 | LM: 0.2272 | LB: 1.0915 | CL0: 2.9 | CL1: 2.4 | HR0: 0.349/SR0: 0.348 | HR1: 0.418/SR1: 0.388 | LR: 1.00e-04 [2026-04-17 11:24:29] Epoch 2 | Step 11030 | Loss: 0.2390 | LM: 0.2268 | LB: 1.0916 | CL0: 2.9 | CL1: 2.4 | HR0: 0.349/SR0: 0.348 | HR1: 0.418/SR1: 0.388 | LR: 1.00e-04 [2026-04-17 11:24:36] Epoch 2 | Step 11040 | Loss: 0.2395 | LM: 0.2272 | LB: 1.0914 | CL0: 2.9 | CL1: 2.4 | HR0: 0.349/SR0: 0.348 | HR1: 0.418/SR1: 0.388 | LR: 1.00e-04 [2026-04-17 11:24:42] Epoch 2 | Step 11050 | Loss: 0.2391 | LM: 0.2260 | LB: 1.0914 | CL0: 2.9 | CL1: 2.4 | HR0: 0.349/SR0: 0.348 | HR1: 0.418/SR1: 0.388 | LR: 1.00e-04 [2026-04-17 11:24:49] Epoch 2 | Step 11060 | Loss: 0.2391 | LM: 0.2249 | LB: 1.0913 | CL0: 2.9 | CL1: 2.4 | HR0: 0.349/SR0: 0.348 | HR1: 0.418/SR1: 0.388 | LR: 1.00e-04 [2026-04-17 11:24:55] Epoch 2 | Step 11070 | Loss: 0.2392 | LM: 0.2253 | LB: 1.0915 | CL0: 2.9 | CL1: 2.4 | HR0: 0.349/SR0: 0.348 | HR1: 0.418/SR1: 0.388 | LR: 1.00e-04 [2026-04-17 11:25:01] Epoch 2 | Step 11080 | Loss: 0.2393 | LM: 0.2246 | LB: 1.0913 | CL0: 2.9 | CL1: 2.4 | HR0: 0.349/SR0: 0.348 | HR1: 0.417/SR1: 0.388 | LR: 1.00e-04 [2026-04-17 11:25:07] Epoch 2 | Step 11090 | Loss: 0.2389 | LM: 0.2240 | LB: 1.0913 | CL0: 2.9 | CL1: 2.4 | HR0: 0.349/SR0: 0.348 | HR1: 0.418/SR1: 0.388 | LR: 1.00e-04 [2026-04-17 11:25:14] Epoch 2 | Step 11100 | Loss: 0.2386 | LM: 0.2253 | LB: 1.0912 | CL0: 2.9 | CL1: 2.4 | HR0: 0.349/SR0: 0.348 | HR1: 0.418/SR1: 0.388 | LR: 1.00e-04 [2026-04-17 11:25:20] Epoch 2 | Step 11110 | Loss: 0.2396 | LM: 0.2264 | LB: 1.0912 | CL0: 2.9 | CL1: 2.4 | HR0: 0.349/SR0: 0.348 | HR1: 0.418/SR1: 0.388 | LR: 1.00e-04 [2026-04-17 11:25:27] Epoch 2 | Step 11120 | Loss: 0.2394 | LM: 0.2251 | LB: 1.0912 | CL0: 2.9 | CL1: 2.4 | HR0: 0.349/SR0: 0.348 | HR1: 0.418/SR1: 0.388 | LR: 1.00e-04 [2026-04-17 11:25:33] Epoch 2 | Step 11130 | Loss: 0.2390 | LM: 0.2245 | LB: 1.0908 | CL0: 2.9 | CL1: 2.4 | HR0: 0.349/SR0: 0.348 | HR1: 0.417/SR1: 0.388 | LR: 1.00e-04 [2026-04-17 11:25:40] Epoch 2 | Step 11140 | Loss: 0.2383 | LM: 0.2244 | LB: 1.0908 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.417/SR1: 0.388 | LR: 1.00e-04 [2026-04-17 11:25:46] Epoch 2 | Step 11150 | Loss: 0.2388 | LM: 0.2237 | LB: 1.0908 | CL0: 2.9 | CL1: 2.4 | HR0: 0.349/SR0: 0.347 | HR1: 0.417/SR1: 0.388 | LR: 1.00e-04 [2026-04-17 11:25:52] Epoch 2 | Step 11160 | Loss: 0.2384 | LM: 0.2226 | LB: 1.0907 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.417/SR1: 0.388 | LR: 1.00e-04 [2026-04-17 11:25:59] Epoch 2 | Step 11170 | Loss: 0.2384 | LM: 0.2238 | LB: 1.0907 | CL0: 2.9 | CL1: 2.4 | HR0: 0.349/SR0: 0.347 | HR1: 0.417/SR1: 0.388 | LR: 1.00e-04 [2026-04-17 11:26:05] Epoch 2 | Step 11180 | Loss: 0.2381 | LM: 0.2225 | LB: 1.0907 | CL0: 2.9 | CL1: 2.4 | HR0: 0.349/SR0: 0.348 | HR1: 0.417/SR1: 0.388 | LR: 1.00e-04 [2026-04-17 11:26:12] Epoch 2 | Step 11190 | Loss: 0.2378 | LM: 0.2232 | LB: 1.0908 | CL0: 2.9 | CL1: 2.4 | HR0: 0.349/SR0: 0.348 | HR1: 0.417/SR1: 0.388 | LR: 1.00e-04 [2026-04-17 11:26:18] Epoch 2 | Step 11200 | Loss: 0.2374 | LM: 0.2236 | LB: 1.0909 | CL0: 2.9 | CL1: 2.4 | HR0: 0.349/SR0: 0.348 | HR1: 0.417/SR1: 0.388 | LR: 1.00e-04 [2026-04-17 11:26:24] Epoch 2 | Step 11210 | Loss: 0.2377 | LM: 0.2237 | LB: 1.0909 | CL0: 2.9 | CL1: 2.4 | HR0: 0.349/SR0: 0.348 | HR1: 0.417/SR1: 0.388 | LR: 1.00e-04 [2026-04-17 11:26:31] Epoch 2 | Step 11220 | Loss: 0.2378 | LM: 0.2234 | LB: 1.0912 | CL0: 2.9 | CL1: 2.4 | HR0: 0.349/SR0: 0.348 | HR1: 0.418/SR1: 0.388 | LR: 1.00e-04 [2026-04-17 11:26:37] Epoch 2 | Step 11230 | Loss: 0.2381 | LM: 0.2229 | LB: 1.0913 | CL0: 2.9 | CL1: 2.4 | HR0: 0.349/SR0: 0.348 | HR1: 0.418/SR1: 0.388 | LR: 1.00e-04 [2026-04-17 11:26:43] Epoch 2 | Step 11240 | Loss: 0.2383 | LM: 0.2226 | LB: 1.0912 | CL0: 2.9 | CL1: 2.4 | HR0: 0.349/SR0: 0.348 | HR1: 0.418/SR1: 0.388 | LR: 1.00e-04 [2026-04-17 11:26:50] Epoch 2 | Step 11250 | Loss: 0.2379 | LM: 0.2225 | LB: 1.0912 | CL0: 2.9 | CL1: 2.4 | HR0: 0.349/SR0: 0.348 | HR1: 0.418/SR1: 0.388 | LR: 1.00e-04 [2026-04-17 11:26:56] Epoch 2 | Step 11260 | Loss: 0.2380 | LM: 0.2229 | LB: 1.0914 | CL0: 2.9 | CL1: 2.4 | HR0: 0.349/SR0: 0.348 | HR1: 0.418/SR1: 0.388 | LR: 1.00e-04 [2026-04-17 11:27:03] Epoch 2 | Step 11270 | Loss: 0.2379 | LM: 0.2211 | LB: 1.0913 | CL0: 2.9 | CL1: 2.4 | HR0: 0.349/SR0: 0.348 | HR1: 0.418/SR1: 0.388 | LR: 1.00e-04 [2026-04-17 11:27:09] Epoch 2 | Step 11280 | Loss: 0.2370 | LM: 0.2205 | LB: 1.0913 | CL0: 2.9 | CL1: 2.4 | HR0: 0.349/SR0: 0.348 | HR1: 0.418/SR1: 0.388 | LR: 1.00e-04 [2026-04-17 11:27:15] Epoch 2 | Step 11290 | Loss: 0.2372 | LM: 0.2197 | LB: 1.0913 | CL0: 2.9 | CL1: 2.4 | HR0: 0.349/SR0: 0.348 | HR1: 0.418/SR1: 0.388 | LR: 1.00e-04 [2026-04-17 11:27:22] Epoch 2 | Step 11300 | Loss: 0.2372 | LM: 0.2203 | LB: 1.0914 | CL0: 2.9 | CL1: 2.4 | HR0: 0.349/SR0: 0.348 | HR1: 0.418/SR1: 0.388 | LR: 1.00e-04 [2026-04-17 11:27:28] Epoch 2 | Step 11310 | Loss: 0.2372 | LM: 0.2201 | LB: 1.0915 | CL0: 2.9 | CL1: 2.4 | HR0: 0.349/SR0: 0.348 | HR1: 0.418/SR1: 0.388 | LR: 1.00e-04 [2026-04-17 11:27:34] Epoch 2 | Step 11320 | Loss: 0.2368 | LM: 0.2205 | LB: 1.0915 | CL0: 2.9 | CL1: 2.4 | HR0: 0.349/SR0: 0.348 | HR1: 0.418/SR1: 0.388 | LR: 1.00e-04 [2026-04-17 11:27:40] Epoch 2 | Step 11330 | Loss: 0.2370 | LM: 0.2208 | LB: 1.0914 | CL0: 2.9 | CL1: 2.4 | HR0: 0.349/SR0: 0.348 | HR1: 0.418/SR1: 0.388 | LR: 1.00e-04 [2026-04-17 11:27:47] Epoch 2 | Step 11340 | Loss: 0.2372 | LM: 0.2206 | LB: 1.0913 | CL0: 2.9 | CL1: 2.4 | HR0: 0.349/SR0: 0.348 | HR1: 0.418/SR1: 0.388 | LR: 1.00e-04 [2026-04-17 11:27:53] Epoch 2 | Step 11350 | Loss: 0.2381 | LM: 0.2203 | LB: 1.0914 | CL0: 2.9 | CL1: 2.4 | HR0: 0.349/SR0: 0.348 | HR1: 0.418/SR1: 0.388 | LR: 1.00e-04 [2026-04-17 11:27:59] Epoch 2 | Step 11360 | Loss: 0.2377 | LM: 0.2199 | LB: 1.0912 | CL0: 2.9 | CL1: 2.4 | HR0: 0.349/SR0: 0.348 | HR1: 0.418/SR1: 0.388 | LR: 1.00e-04 [2026-04-17 11:28:06] Epoch 2 | Step 11370 | Loss: 0.2378 | LM: 0.2192 | LB: 1.0912 | CL0: 2.9 | CL1: 2.4 | HR0: 0.349/SR0: 0.348 | HR1: 0.418/SR1: 0.388 | LR: 1.00e-04 [2026-04-17 11:28:12] Epoch 2 | Step 11380 | Loss: 0.2378 | LM: 0.2195 | LB: 1.0913 | CL0: 2.9 | CL1: 2.4 | HR0: 0.349/SR0: 0.348 | HR1: 0.418/SR1: 0.388 | LR: 1.00e-04 [2026-04-17 11:28:19] Epoch 2 | Step 11390 | Loss: 0.2383 | LM: 0.2196 | LB: 1.0912 | CL0: 2.9 | CL1: 2.4 | HR0: 0.349/SR0: 0.348 | HR1: 0.418/SR1: 0.388 | LR: 1.00e-04 [2026-04-17 11:28:25] Epoch 2 | Step 11400 | Loss: 0.2383 | LM: 0.2189 | LB: 1.0914 | CL0: 2.9 | CL1: 2.4 | HR0: 0.349/SR0: 0.348 | HR1: 0.418/SR1: 0.389 | LR: 1.00e-04 [2026-04-17 11:28:32] Epoch 2 | Step 11410 | Loss: 0.2384 | LM: 0.2202 | LB: 1.0913 | CL0: 2.9 | CL1: 2.4 | HR0: 0.349/SR0: 0.348 | HR1: 0.418/SR1: 0.389 | LR: 1.00e-04 [2026-04-17 11:28:38] Epoch 2 | Step 11420 | Loss: 0.2383 | LM: 0.2203 | LB: 1.0913 | CL0: 2.9 | CL1: 2.4 | HR0: 0.349/SR0: 0.348 | HR1: 0.418/SR1: 0.389 | LR: 1.00e-04 [2026-04-17 11:28:44] Epoch 2 | Step 11430 | Loss: 0.2385 | LM: 0.2213 | LB: 1.0912 | CL0: 2.9 | CL1: 2.4 | HR0: 0.349/SR0: 0.348 | HR1: 0.418/SR1: 0.388 | LR: 1.00e-04 [2026-04-17 11:28:51] Epoch 2 | Step 11440 | Loss: 0.2387 | LM: 0.2213 | LB: 1.0912 | CL0: 2.9 | CL1: 2.4 | HR0: 0.349/SR0: 0.348 | HR1: 0.418/SR1: 0.389 | LR: 1.00e-04 [2026-04-17 11:28:57] Epoch 2 | Step 11450 | Loss: 0.2390 | LM: 0.2215 | LB: 1.0911 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.418/SR1: 0.389 | LR: 1.00e-04 [2026-04-17 11:29:03] Epoch 2 | Step 11460 | Loss: 0.2393 | LM: 0.2226 | LB: 1.0911 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.418/SR1: 0.388 | LR: 1.00e-04 [2026-04-17 11:29:10] Epoch 2 | Step 11470 | Loss: 0.2392 | LM: 0.2222 | LB: 1.0911 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.418/SR1: 0.388 | LR: 1.00e-04 [2026-04-17 11:29:16] Epoch 2 | Step 11480 | Loss: 0.2393 | LM: 0.2219 | LB: 1.0911 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.418/SR1: 0.388 | LR: 1.00e-04 [2026-04-17 11:29:22] Epoch 2 | Step 11490 | Loss: 0.2389 | LM: 0.2212 | LB: 1.0911 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.418/SR1: 0.389 | LR: 1.00e-04 [2026-04-17 11:29:29] Epoch 2 | Step 11500 | Loss: 0.2390 | LM: 0.2218 | LB: 1.0912 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.418/SR1: 0.389 | LR: 1.00e-04 [2026-04-17 11:29:35] Epoch 2 | Step 11510 | Loss: 0.2387 | LM: 0.2219 | LB: 1.0912 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.418/SR1: 0.389 | LR: 1.00e-04 [2026-04-17 11:29:42] Epoch 2 | Step 11520 | Loss: 0.2394 | LM: 0.2225 | LB: 1.0912 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.418/SR1: 0.389 | LR: 1.00e-04 [2026-04-17 11:29:48] Epoch 2 | Step 11530 | Loss: 0.2393 | LM: 0.2232 | LB: 1.0912 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.418/SR1: 0.389 | LR: 1.00e-04 [2026-04-17 11:29:55] Epoch 2 | Step 11540 | Loss: 0.2392 | LM: 0.2236 | LB: 1.0912 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.418/SR1: 0.389 | LR: 1.00e-04 [2026-04-17 11:30:01] Epoch 2 | Step 11550 | Loss: 0.2397 | LM: 0.2247 | LB: 1.0912 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.418/SR1: 0.389 | LR: 1.00e-04 [2026-04-17 11:30:08] Epoch 2 | Step 11560 | Loss: 0.2394 | LM: 0.2246 | LB: 1.0912 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.418/SR1: 0.389 | LR: 1.00e-04 [2026-04-17 11:30:14] Epoch 2 | Step 11570 | Loss: 0.2394 | LM: 0.2248 | LB: 1.0911 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.418/SR1: 0.389 | LR: 1.00e-04 [2026-04-17 11:30:20] Epoch 2 | Step 11580 | Loss: 0.2395 | LM: 0.2255 | LB: 1.0911 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.418/SR1: 0.389 | LR: 1.00e-04 [2026-04-17 11:30:27] Epoch 2 | Step 11590 | Loss: 0.2393 | LM: 0.2260 | LB: 1.0910 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.418/SR1: 0.388 | LR: 1.00e-04 [2026-04-17 11:30:33] Epoch 2 | Step 11600 | Loss: 0.2397 | LM: 0.2265 | LB: 1.0911 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.418/SR1: 0.389 | LR: 1.00e-04 [2026-04-17 11:30:39] Epoch 2 | Step 11610 | Loss: 0.2399 | LM: 0.2270 | LB: 1.0912 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.418/SR1: 0.389 | LR: 1.00e-04 [2026-04-17 11:30:45] Epoch 2 | Step 11620 | Loss: 0.2395 | LM: 0.2268 | LB: 1.0911 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.418/SR1: 0.389 | LR: 1.00e-04 [2026-04-17 11:30:51] Epoch 2 | Step 11630 | Loss: 0.2393 | LM: 0.2265 | LB: 1.0911 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.348 | HR1: 0.418/SR1: 0.389 | LR: 1.00e-04 [2026-04-17 11:30:58] Epoch 2 | Step 11640 | Loss: 0.2387 | LM: 0.2259 | LB: 1.0911 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.418/SR1: 0.388 | LR: 1.00e-04 [2026-04-17 11:31:04] Epoch 2 | Step 11650 | Loss: 0.2386 | LM: 0.2253 | LB: 1.0910 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.418/SR1: 0.388 | LR: 1.00e-04 [2026-04-17 11:31:10] Epoch 2 | Step 11660 | Loss: 0.2386 | LM: 0.2250 | LB: 1.0910 | CL0: 2.9 | CL1: 2.4 | HR0: 0.349/SR0: 0.348 | HR1: 0.418/SR1: 0.388 | LR: 1.00e-04 [2026-04-17 11:31:17] Epoch 2 | Step 11670 | Loss: 0.2389 | LM: 0.2256 | LB: 1.0911 | CL0: 2.9 | CL1: 2.4 | HR0: 0.349/SR0: 0.348 | HR1: 0.418/SR1: 0.388 | LR: 1.00e-04 [2026-04-17 11:31:23] Epoch 2 | Step 11680 | Loss: 0.2388 | LM: 0.2252 | LB: 1.0910 | CL0: 2.9 | CL1: 2.4 | HR0: 0.349/SR0: 0.348 | HR1: 0.418/SR1: 0.388 | LR: 1.00e-04 [2026-04-17 11:31:29] Epoch 2 | Step 11690 | Loss: 0.2387 | LM: 0.2248 | LB: 1.0908 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.348 | HR1: 0.418/SR1: 0.388 | LR: 1.00e-04 [2026-04-17 11:31:35] Epoch 2 | Step 11700 | Loss: 0.2387 | LM: 0.2245 | LB: 1.0909 | CL0: 2.9 | CL1: 2.4 | HR0: 0.349/SR0: 0.348 | HR1: 0.418/SR1: 0.388 | LR: 1.00e-04 [2026-04-17 11:31:42] Epoch 2 | Step 11710 | Loss: 0.2387 | LM: 0.2243 | LB: 1.0909 | CL0: 2.9 | CL1: 2.4 | HR0: 0.349/SR0: 0.348 | HR1: 0.418/SR1: 0.388 | LR: 1.00e-04 [2026-04-17 11:31:48] Epoch 2 | Step 11720 | Loss: 0.2388 | LM: 0.2245 | LB: 1.0909 | CL0: 2.9 | CL1: 2.4 | HR0: 0.349/SR0: 0.348 | HR1: 0.418/SR1: 0.388 | LR: 1.00e-04 [2026-04-17 11:31:54] Epoch 2 | Step 11730 | Loss: 0.2387 | LM: 0.2244 | LB: 1.0909 | CL0: 2.9 | CL1: 2.4 | HR0: 0.349/SR0: 0.348 | HR1: 0.418/SR1: 0.388 | LR: 1.00e-04 [2026-04-17 11:32:00] Epoch 2 | Step 11740 | Loss: 0.2389 | LM: 0.2241 | LB: 1.0909 | CL0: 2.9 | CL1: 2.4 | HR0: 0.349/SR0: 0.348 | HR1: 0.418/SR1: 0.388 | LR: 1.00e-04 [2026-04-17 11:32:06] Epoch 2 | Step 11750 | Loss: 0.2389 | LM: 0.2239 | LB: 1.0910 | CL0: 2.9 | CL1: 2.4 | HR0: 0.349/SR0: 0.348 | HR1: 0.418/SR1: 0.388 | LR: 1.00e-04 [2026-04-17 11:32:12] Epoch 2 | Step 11760 | Loss: 0.2390 | LM: 0.2241 | LB: 1.0910 | CL0: 2.9 | CL1: 2.4 | HR0: 0.349/SR0: 0.348 | HR1: 0.418/SR1: 0.388 | LR: 1.00e-04 [2026-04-17 11:32:18] Epoch 2 | Step 11770 | Loss: 0.2388 | LM: 0.2242 | LB: 1.0910 | CL0: 2.9 | CL1: 2.4 | HR0: 0.349/SR0: 0.348 | HR1: 0.418/SR1: 0.388 | LR: 1.00e-04 [2026-04-17 11:32:25] Epoch 2 | Step 11780 | Loss: 0.2390 | LM: 0.2242 | LB: 1.0911 | CL0: 2.9 | CL1: 2.4 | HR0: 0.349/SR0: 0.348 | HR1: 0.418/SR1: 0.388 | LR: 1.00e-04 [2026-04-17 11:32:31] Epoch 2 | Step 11790 | Loss: 0.2391 | LM: 0.2241 | LB: 1.0911 | CL0: 2.9 | CL1: 2.4 | HR0: 0.349/SR0: 0.348 | HR1: 0.418/SR1: 0.388 | LR: 1.00e-04 [2026-04-17 11:32:38] Epoch 2 | Step 11800 | Loss: 0.2390 | LM: 0.2239 | LB: 1.0910 | CL0: 2.9 | CL1: 2.4 | HR0: 0.349/SR0: 0.348 | HR1: 0.418/SR1: 0.388 | LR: 1.00e-04 [2026-04-17 11:32:44] Epoch 2 | Step 11810 | Loss: 0.2389 | LM: 0.2239 | LB: 1.0910 | CL0: 2.9 | CL1: 2.4 | HR0: 0.349/SR0: 0.348 | HR1: 0.418/SR1: 0.388 | LR: 1.00e-04 [2026-04-17 11:32:50] Epoch 2 | Step 11820 | Loss: 0.2393 | LM: 0.2248 | LB: 1.0910 | CL0: 2.9 | CL1: 2.4 | HR0: 0.349/SR0: 0.348 | HR1: 0.418/SR1: 0.388 | LR: 1.00e-04 [2026-04-17 11:32:57] Epoch 2 | Step 11830 | Loss: 0.2393 | LM: 0.2246 | LB: 1.0909 | CL0: 2.9 | CL1: 2.4 | HR0: 0.349/SR0: 0.348 | HR1: 0.418/SR1: 0.388 | LR: 1.00e-04 [2026-04-17 11:33:03] Epoch 2 | Step 11840 | Loss: 0.2392 | LM: 0.2245 | LB: 1.0909 | CL0: 2.9 | CL1: 2.4 | HR0: 0.349/SR0: 0.348 | HR1: 0.418/SR1: 0.388 | LR: 1.00e-04 [2026-04-17 11:33:09] Epoch 2 | Step 11850 | Loss: 0.2391 | LM: 0.2243 | LB: 1.0910 | CL0: 2.9 | CL1: 2.4 | HR0: 0.349/SR0: 0.348 | HR1: 0.418/SR1: 0.388 | LR: 1.00e-04 [2026-04-17 11:33:16] Epoch 2 | Step 11860 | Loss: 0.2388 | LM: 0.2243 | LB: 1.0910 | CL0: 2.9 | CL1: 2.4 | HR0: 0.349/SR0: 0.348 | HR1: 0.418/SR1: 0.388 | LR: 1.00e-04 [2026-04-17 11:33:22] Epoch 2 | Step 11870 | Loss: 0.2388 | LM: 0.2241 | LB: 1.0910 | CL0: 2.9 | CL1: 2.4 | HR0: 0.349/SR0: 0.348 | HR1: 0.418/SR1: 0.388 | LR: 1.00e-04 [2026-04-17 11:33:28] Epoch 2 | Step 11880 | Loss: 0.2388 | LM: 0.2241 | LB: 1.0910 | CL0: 2.9 | CL1: 2.4 | HR0: 0.349/SR0: 0.348 | HR1: 0.418/SR1: 0.388 | LR: 1.00e-04 [2026-04-17 11:33:35] Epoch 2 | Step 11890 | Loss: 0.2390 | LM: 0.2239 | LB: 1.0909 | CL0: 2.9 | CL1: 2.4 | HR0: 0.349/SR0: 0.348 | HR1: 0.418/SR1: 0.388 | LR: 1.00e-04 [2026-04-17 11:33:41] Epoch 2 | Step 11900 | Loss: 0.2390 | LM: 0.2240 | LB: 1.0908 | CL0: 2.9 | CL1: 2.4 | HR0: 0.349/SR0: 0.348 | HR1: 0.418/SR1: 0.388 | LR: 1.00e-04 [2026-04-17 11:33:48] Epoch 2 | Step 11910 | Loss: 0.2390 | LM: 0.2243 | LB: 1.0907 | CL0: 2.9 | CL1: 2.4 | HR0: 0.349/SR0: 0.348 | HR1: 0.418/SR1: 0.388 | LR: 1.00e-04 [2026-04-17 11:33:54] Epoch 2 | Step 11920 | Loss: 0.2391 | LM: 0.2244 | LB: 1.0906 | CL0: 2.9 | CL1: 2.4 | HR0: 0.349/SR0: 0.348 | HR1: 0.417/SR1: 0.388 | LR: 1.00e-04 [2026-04-17 11:34:00] Epoch 2 | Step 11930 | Loss: 0.2391 | LM: 0.2243 | LB: 1.0907 | CL0: 2.9 | CL1: 2.4 | HR0: 0.349/SR0: 0.348 | HR1: 0.417/SR1: 0.388 | LR: 1.00e-04 [2026-04-17 11:34:07] Epoch 2 | Step 11940 | Loss: 0.2389 | LM: 0.2244 | LB: 1.0907 | CL0: 2.9 | CL1: 2.4 | HR0: 0.349/SR0: 0.348 | HR1: 0.418/SR1: 0.388 | LR: 1.00e-04 [2026-04-17 11:34:13] Epoch 2 | Step 11950 | Loss: 0.2391 | LM: 0.2248 | LB: 1.0907 | CL0: 2.9 | CL1: 2.4 | HR0: 0.349/SR0: 0.348 | HR1: 0.418/SR1: 0.388 | LR: 1.00e-04 [2026-04-17 11:34:19] Epoch 2 | Step 11960 | Loss: 0.2393 | LM: 0.2256 | LB: 1.0907 | CL0: 2.9 | CL1: 2.4 | HR0: 0.349/SR0: 0.348 | HR1: 0.418/SR1: 0.388 | LR: 1.00e-04 [2026-04-17 11:34:26] Epoch 2 | Step 11970 | Loss: 0.2391 | LM: 0.2251 | LB: 1.0906 | CL0: 2.9 | CL1: 2.4 | HR0: 0.349/SR0: 0.348 | HR1: 0.417/SR1: 0.388 | LR: 1.00e-04 [2026-04-17 11:34:32] Epoch 2 | Step 11980 | Loss: 0.2393 | LM: 0.2257 | LB: 1.0906 | CL0: 2.9 | CL1: 2.4 | HR0: 0.349/SR0: 0.348 | HR1: 0.417/SR1: 0.388 | LR: 1.00e-04 [2026-04-17 11:34:38] Epoch 2 | Step 11990 | Loss: 0.2393 | LM: 0.2254 | LB: 1.0906 | CL0: 2.9 | CL1: 2.4 | HR0: 0.349/SR0: 0.348 | HR1: 0.417/SR1: 0.388 | LR: 1.00e-04 [2026-04-17 11:34:45] Epoch 2 | Step 12000 | Loss: 0.2397 | LM: 0.2254 | LB: 1.0906 | CL0: 2.9 | CL1: 2.4 | HR0: 0.349/SR0: 0.348 | HR1: 0.417/SR1: 0.388 | LR: 1.00e-04 [2026-04-17 11:34:54] Checkpoint saved: outputs/2026-04-17/08-57-56/checkpoints/checkpoint_step_12000.pt [2026-04-17 11:35:09] Validation | Batch 10/784 | Loss: 0.3275 | LM_LOSS: 0.3166 | LB_LOSS: 1.0849 [2026-04-17 11:35:10] Validation | Batch 20/784 | Loss: 0.3425 | LM_LOSS: 0.3316 | LB_LOSS: 1.0851 [2026-04-17 11:35:11] Validation | Batch 30/784 | Loss: 0.3281 | LM_LOSS: 0.3173 | LB_LOSS: 1.0844 [2026-04-17 11:35:13] Validation | Batch 40/784 | Loss: 0.3313 | LM_LOSS: 0.3204 | LB_LOSS: 1.0843 [2026-04-17 11:35:14] Validation | Batch 50/784 | Loss: 0.3279 | LM_LOSS: 0.3171 | LB_LOSS: 1.0837 [2026-04-17 11:35:15] Validation | Batch 60/784 | Loss: 0.3284 | LM_LOSS: 0.3176 | LB_LOSS: 1.0832 [2026-04-17 11:35:17] Validation | Batch 70/784 | Loss: 0.3268 | LM_LOSS: 0.3160 | LB_LOSS: 1.0826 [2026-04-17 11:35:18] Validation | Batch 80/784 | Loss: 0.3227 | LM_LOSS: 0.3119 | LB_LOSS: 1.0821 [2026-04-17 11:35:19] Validation | Batch 90/784 | Loss: 0.3220 | LM_LOSS: 0.3112 | LB_LOSS: 1.0827 [2026-04-17 11:35:21] Validation | Batch 100/784 | Loss: 0.3236 | LM_LOSS: 0.3128 | LB_LOSS: 1.0831 [2026-04-17 11:35:22] Validation | Batch 110/784 | Loss: 0.3184 | LM_LOSS: 0.3076 | LB_LOSS: 1.0833 [2026-04-17 11:35:23] Validation | Batch 120/784 | Loss: 0.3217 | LM_LOSS: 0.3109 | LB_LOSS: 1.0832 [2026-04-17 11:35:25] Validation | Batch 130/784 | Loss: 0.3241 | LM_LOSS: 0.3133 | LB_LOSS: 1.0831 [2026-04-17 11:35:26] Validation | Batch 140/784 | Loss: 0.3235 | LM_LOSS: 0.3127 | LB_LOSS: 1.0830 [2026-04-17 11:35:28] Validation | Batch 150/784 | Loss: 0.3198 | LM_LOSS: 0.3089 | LB_LOSS: 1.0833 [2026-04-17 11:35:29] Validation | Batch 160/784 | Loss: 0.3210 | LM_LOSS: 0.3102 | LB_LOSS: 1.0830 [2026-04-17 11:35:30] Validation | Batch 170/784 | Loss: 0.3219 | LM_LOSS: 0.3110 | LB_LOSS: 1.0827 [2026-04-17 11:35:32] Validation | Batch 180/784 | Loss: 0.3192 | LM_LOSS: 0.3084 | LB_LOSS: 1.0827 [2026-04-17 11:35:33] Validation | Batch 190/784 | Loss: 0.3214 | LM_LOSS: 0.3106 | LB_LOSS: 1.0831 [2026-04-17 11:35:34] Validation | Batch 200/784 | Loss: 0.3218 | LM_LOSS: 0.3110 | LB_LOSS: 1.0832 [2026-04-17 11:35:36] Validation | Batch 210/784 | Loss: 0.3205 | LM_LOSS: 0.3097 | LB_LOSS: 1.0831 [2026-04-17 11:35:37] Validation | Batch 220/784 | Loss: 0.3216 | LM_LOSS: 0.3108 | LB_LOSS: 1.0831 [2026-04-17 11:35:39] Validation | Batch 230/784 | Loss: 0.3222 | LM_LOSS: 0.3113 | LB_LOSS: 1.0830 [2026-04-17 11:35:40] Validation | Batch 240/784 | Loss: 0.3226 | LM_LOSS: 0.3118 | LB_LOSS: 1.0834 [2026-04-17 11:35:41] Validation | Batch 250/784 | Loss: 0.3226 | LM_LOSS: 0.3118 | LB_LOSS: 1.0832 [2026-04-17 11:35:43] Validation | Batch 260/784 | Loss: 0.3227 | LM_LOSS: 0.3119 | LB_LOSS: 1.0834 [2026-04-17 11:35:44] Validation | Batch 270/784 | Loss: 0.3224 | LM_LOSS: 0.3115 | LB_LOSS: 1.0835 [2026-04-17 11:35:46] Validation | Batch 280/784 | Loss: 0.3228 | LM_LOSS: 0.3119 | LB_LOSS: 1.0836 [2026-04-17 11:35:47] Validation | Batch 290/784 | Loss: 0.3236 | LM_LOSS: 0.3127 | LB_LOSS: 1.0838 [2026-04-17 11:35:48] Validation | Batch 300/784 | Loss: 0.3242 | LM_LOSS: 0.3133 | LB_LOSS: 1.0838 [2026-04-17 11:35:50] Validation | Batch 310/784 | Loss: 0.3235 | LM_LOSS: 0.3127 | LB_LOSS: 1.0838 [2026-04-17 11:35:51] Validation | Batch 320/784 | Loss: 0.3250 | LM_LOSS: 0.3142 | LB_LOSS: 1.0837 [2026-04-17 11:35:53] Validation | Batch 330/784 | Loss: 0.3249 | LM_LOSS: 0.3141 | LB_LOSS: 1.0837 [2026-04-17 11:35:54] Validation | Batch 340/784 | Loss: 0.3238 | LM_LOSS: 0.3130 | LB_LOSS: 1.0838 [2026-04-17 11:35:55] Validation | Batch 350/784 | Loss: 0.3239 | LM_LOSS: 0.3131 | LB_LOSS: 1.0840 [2026-04-17 11:35:56] Validation | Batch 360/784 | Loss: 0.3236 | LM_LOSS: 0.3127 | LB_LOSS: 1.0840 [2026-04-17 11:35:58] Validation | Batch 370/784 | Loss: 0.3239 | LM_LOSS: 0.3131 | LB_LOSS: 1.0839 [2026-04-17 11:35:59] Validation | Batch 380/784 | Loss: 0.3237 | LM_LOSS: 0.3128 | LB_LOSS: 1.0840 [2026-04-17 11:36:00] Validation | Batch 390/784 | Loss: 0.3235 | LM_LOSS: 0.3127 | LB_LOSS: 1.0841 [2026-04-17 11:36:01] Validation | Batch 400/784 | Loss: 0.3238 | LM_LOSS: 0.3130 | LB_LOSS: 1.0840 [2026-04-17 11:36:03] Validation | Batch 410/784 | Loss: 0.3242 | LM_LOSS: 0.3134 | LB_LOSS: 1.0841 [2026-04-17 11:36:04] Validation | Batch 420/784 | Loss: 0.3244 | LM_LOSS: 0.3136 | LB_LOSS: 1.0841 [2026-04-17 11:36:05] Validation | Batch 430/784 | Loss: 0.3246 | LM_LOSS: 0.3138 | LB_LOSS: 1.0840 [2026-04-17 11:36:06] Validation | Batch 440/784 | Loss: 0.3243 | LM_LOSS: 0.3135 | LB_LOSS: 1.0841 [2026-04-17 11:36:08] Validation | Batch 450/784 | Loss: 0.3235 | LM_LOSS: 0.3127 | LB_LOSS: 1.0840 [2026-04-17 11:36:09] Validation | Batch 460/784 | Loss: 0.3240 | LM_LOSS: 0.3132 | LB_LOSS: 1.0841 [2026-04-17 11:36:11] Validation | Batch 470/784 | Loss: 0.3233 | LM_LOSS: 0.3124 | LB_LOSS: 1.0841 [2026-04-17 11:36:12] Validation | Batch 480/784 | Loss: 0.3238 | LM_LOSS: 0.3129 | LB_LOSS: 1.0840 [2026-04-17 11:36:13] Validation | Batch 490/784 | Loss: 0.3232 | LM_LOSS: 0.3124 | LB_LOSS: 1.0840 [2026-04-17 11:36:14] Validation | Batch 500/784 | Loss: 0.3237 | LM_LOSS: 0.3128 | LB_LOSS: 1.0839 [2026-04-17 11:36:16] Validation | Batch 510/784 | Loss: 0.3235 | LM_LOSS: 0.3127 | LB_LOSS: 1.0839 [2026-04-17 11:36:17] Validation | Batch 520/784 | Loss: 0.3236 | LM_LOSS: 0.3128 | LB_LOSS: 1.0838 [2026-04-17 11:36:19] Validation | Batch 530/784 | Loss: 0.3245 | LM_LOSS: 0.3136 | LB_LOSS: 1.0838 [2026-04-17 11:36:20] Validation | Batch 540/784 | Loss: 0.3248 | LM_LOSS: 0.3140 | LB_LOSS: 1.0838 [2026-04-17 11:36:21] Validation | Batch 550/784 | Loss: 0.3261 | LM_LOSS: 0.3153 | LB_LOSS: 1.0837 [2026-04-17 11:36:23] Validation | Batch 560/784 | Loss: 0.3262 | LM_LOSS: 0.3154 | LB_LOSS: 1.0838 [2026-04-17 11:36:24] Validation | Batch 570/784 | Loss: 0.3256 | LM_LOSS: 0.3148 | LB_LOSS: 1.0837 [2026-04-17 11:36:25] Validation | Batch 580/784 | Loss: 0.3251 | LM_LOSS: 0.3143 | LB_LOSS: 1.0838 [2026-04-17 11:36:27] Validation | Batch 590/784 | Loss: 0.3254 | LM_LOSS: 0.3145 | LB_LOSS: 1.0837 [2026-04-17 11:36:28] Validation | Batch 600/784 | Loss: 0.3252 | LM_LOSS: 0.3144 | LB_LOSS: 1.0836 [2026-04-17 11:36:30] Validation | Batch 610/784 | Loss: 0.3254 | LM_LOSS: 0.3145 | LB_LOSS: 1.0836 [2026-04-17 11:36:31] Validation | Batch 620/784 | Loss: 0.3253 | LM_LOSS: 0.3145 | LB_LOSS: 1.0836 [2026-04-17 11:36:32] Validation | Batch 630/784 | Loss: 0.3260 | LM_LOSS: 0.3152 | LB_LOSS: 1.0837 [2026-04-17 11:36:34] Validation | Batch 640/784 | Loss: 0.3260 | LM_LOSS: 0.3152 | LB_LOSS: 1.0836 [2026-04-17 11:36:36] Validation | Batch 650/784 | Loss: 0.3259 | LM_LOSS: 0.3150 | LB_LOSS: 1.0837 [2026-04-17 11:36:37] Validation | Batch 660/784 | Loss: 0.3262 | LM_LOSS: 0.3154 | LB_LOSS: 1.0837 [2026-04-17 11:36:38] Validation | Batch 670/784 | Loss: 0.3266 | LM_LOSS: 0.3158 | LB_LOSS: 1.0837 [2026-04-17 11:36:40] Validation | Batch 680/784 | Loss: 0.3263 | LM_LOSS: 0.3155 | LB_LOSS: 1.0837 [2026-04-17 11:36:41] Validation | Batch 690/784 | Loss: 0.3265 | LM_LOSS: 0.3157 | LB_LOSS: 1.0837 [2026-04-17 11:36:43] Validation | Batch 700/784 | Loss: 0.3265 | LM_LOSS: 0.3157 | LB_LOSS: 1.0836 [2026-04-17 11:36:44] Validation | Batch 710/784 | Loss: 0.3263 | LM_LOSS: 0.3155 | LB_LOSS: 1.0836 [2026-04-17 11:36:45] Validation | Batch 720/784 | Loss: 0.3261 | LM_LOSS: 0.3152 | LB_LOSS: 1.0835 [2026-04-17 11:36:47] Validation | Batch 730/784 | Loss: 0.3255 | LM_LOSS: 0.3147 | LB_LOSS: 1.0835 [2026-04-17 11:36:48] Validation | Batch 740/784 | Loss: 0.3256 | LM_LOSS: 0.3148 | LB_LOSS: 1.0836 [2026-04-17 11:36:49] Validation | Batch 750/784 | Loss: 0.3250 | LM_LOSS: 0.3141 | LB_LOSS: 1.0836 [2026-04-17 11:36:50] Validation | Batch 760/784 | Loss: 0.3252 | LM_LOSS: 0.3143 | LB_LOSS: 1.0836 [2026-04-17 11:36:52] Validation | Batch 770/784 | Loss: 0.3253 | LM_LOSS: 0.3145 | LB_LOSS: 1.0836 [2026-04-17 11:36:53] Validation | Batch 780/784 | Loss: 0.3256 | LM_LOSS: 0.3147 | LB_LOSS: 1.0836 [2026-04-17 11:36:54] Validation | Batch 784/784 | Loss: 0.3257 | LM_LOSS: 0.3149 | LB_LOSS: 1.0836 [2026-04-17 11:36:57] Validation | Loss: 0.3257 | LM_LOSS: 0.3149 | LB_LOSS: 1.0836 | PPL: 1.37 | Time: 106.23s [2026-04-17 11:37:03] Epoch 2 | Step 12010 | Loss: 0.2398 | LM: 0.2258 | LB: 1.0906 | CL0: 2.9 | CL1: 2.4 | HR0: 0.349/SR0: 0.348 | HR1: 0.417/SR1: 0.388 | LR: 1.00e-04 [2026-04-17 11:37:09] Epoch 2 | Step 12020 | Loss: 0.2398 | LM: 0.2260 | LB: 1.0906 | CL0: 2.9 | CL1: 2.4 | HR0: 0.349/SR0: 0.348 | HR1: 0.417/SR1: 0.388 | LR: 1.00e-04 [2026-04-17 11:37:16] Epoch 2 | Step 12030 | Loss: 0.2396 | LM: 0.2257 | LB: 1.0906 | CL0: 2.9 | CL1: 2.4 | HR0: 0.349/SR0: 0.348 | HR1: 0.417/SR1: 0.388 | LR: 1.00e-04 [2026-04-17 11:37:22] Epoch 2 | Step 12040 | Loss: 0.2396 | LM: 0.2259 | LB: 1.0906 | CL0: 2.9 | CL1: 2.4 | HR0: 0.349/SR0: 0.348 | HR1: 0.418/SR1: 0.388 | LR: 1.00e-04 [2026-04-17 11:37:29] Epoch 2 | Step 12050 | Loss: 0.2397 | LM: 0.2260 | LB: 1.0906 | CL0: 2.9 | CL1: 2.4 | HR0: 0.349/SR0: 0.348 | HR1: 0.417/SR1: 0.388 | LR: 1.00e-04 [2026-04-17 11:37:35] Epoch 2 | Step 12060 | Loss: 0.2397 | LM: 0.2261 | LB: 1.0905 | CL0: 2.9 | CL1: 2.4 | HR0: 0.349/SR0: 0.348 | HR1: 0.417/SR1: 0.388 | LR: 1.00e-04 [2026-04-17 11:37:41] Epoch 2 | Step 12070 | Loss: 0.2399 | LM: 0.2266 | LB: 1.0906 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.348 | HR1: 0.417/SR1: 0.388 | LR: 1.00e-04 [2026-04-17 11:37:48] Epoch 2 | Step 12080 | Loss: 0.2400 | LM: 0.2270 | LB: 1.0906 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.348 | HR1: 0.418/SR1: 0.388 | LR: 1.00e-04 [2026-04-17 11:37:54] Epoch 2 | Step 12090 | Loss: 0.2400 | LM: 0.2268 | LB: 1.0906 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.348 | HR1: 0.418/SR1: 0.388 | LR: 1.00e-04 [2026-04-17 11:38:00] Epoch 2 | Step 12100 | Loss: 0.2400 | LM: 0.2270 | LB: 1.0906 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.348 | HR1: 0.418/SR1: 0.388 | LR: 1.00e-04 [2026-04-17 11:38:07] Epoch 2 | Step 12110 | Loss: 0.2397 | LM: 0.2267 | LB: 1.0906 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.348 | HR1: 0.418/SR1: 0.388 | LR: 1.00e-04 [2026-04-17 11:38:13] Epoch 2 | Step 12120 | Loss: 0.2397 | LM: 0.2268 | LB: 1.0906 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.418/SR1: 0.388 | LR: 1.00e-04 [2026-04-17 11:38:20] Epoch 2 | Step 12130 | Loss: 0.2398 | LM: 0.2269 | LB: 1.0906 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.418/SR1: 0.388 | LR: 1.00e-04 [2026-04-17 11:38:26] Epoch 2 | Step 12140 | Loss: 0.2397 | LM: 0.2271 | LB: 1.0905 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.418/SR1: 0.388 | LR: 1.00e-04 [2026-04-17 11:38:33] Epoch 2 | Step 12150 | Loss: 0.2397 | LM: 0.2274 | LB: 1.0905 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.418/SR1: 0.388 | LR: 1.00e-04 [2026-04-17 11:38:39] Epoch 2 | Step 12160 | Loss: 0.2397 | LM: 0.2274 | LB: 1.0905 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.418/SR1: 0.388 | LR: 1.00e-04 [2026-04-17 11:38:45] Epoch 2 | Step 12170 | Loss: 0.2396 | LM: 0.2272 | LB: 1.0905 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.418/SR1: 0.388 | LR: 1.00e-04 [2026-04-17 11:38:52] Epoch 2 | Step 12180 | Loss: 0.2395 | LM: 0.2272 | LB: 1.0905 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.418/SR1: 0.388 | LR: 1.00e-04 [2026-04-17 11:38:58] Epoch 2 | Step 12190 | Loss: 0.2396 | LM: 0.2271 | LB: 1.0906 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.418/SR1: 0.388 | LR: 1.00e-04 [2026-04-17 11:39:05] Epoch 2 | Step 12200 | Loss: 0.2396 | LM: 0.2273 | LB: 1.0906 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.418/SR1: 0.388 | LR: 1.00e-04 [2026-04-17 11:39:11] Epoch 2 | Step 12210 | Loss: 0.2396 | LM: 0.2275 | LB: 1.0906 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.348 | HR1: 0.418/SR1: 0.388 | LR: 1.00e-04 [2026-04-17 11:39:17] Epoch 2 | Step 12220 | Loss: 0.2395 | LM: 0.2272 | LB: 1.0906 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.348 | HR1: 0.418/SR1: 0.388 | LR: 1.00e-04 [2026-04-17 11:39:24] Epoch 2 | Step 12230 | Loss: 0.2395 | LM: 0.2274 | LB: 1.0906 | CL0: 2.9 | CL1: 2.4 | HR0: 0.349/SR0: 0.348 | HR1: 0.418/SR1: 0.388 | LR: 1.00e-04 [2026-04-17 11:39:30] Epoch 2 | Step 12240 | Loss: 0.2395 | LM: 0.2272 | LB: 1.0907 | CL0: 2.9 | CL1: 2.4 | HR0: 0.349/SR0: 0.348 | HR1: 0.418/SR1: 0.388 | LR: 1.00e-04 [2026-04-17 11:39:36] Epoch 2 | Step 12250 | Loss: 0.2395 | LM: 0.2273 | LB: 1.0906 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.348 | HR1: 0.418/SR1: 0.388 | LR: 1.00e-04 [2026-04-17 11:39:43] Epoch 2 | Step 12260 | Loss: 0.2395 | LM: 0.2276 | LB: 1.0906 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.348 | HR1: 0.418/SR1: 0.388 | LR: 1.00e-04 [2026-04-17 11:39:49] Epoch 2 | Step 12270 | Loss: 0.2396 | LM: 0.2271 | LB: 1.0906 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.348 | HR1: 0.418/SR1: 0.388 | LR: 1.00e-04 [2026-04-17 11:39:55] Epoch 2 | Step 12280 | Loss: 0.2397 | LM: 0.2269 | LB: 1.0906 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.348 | HR1: 0.418/SR1: 0.388 | LR: 1.00e-04 [2026-04-17 11:40:02] Epoch 2 | Step 12290 | Loss: 0.2397 | LM: 0.2269 | LB: 1.0906 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.418/SR1: 0.388 | LR: 1.00e-04 [2026-04-17 11:40:08] Epoch 2 | Step 12300 | Loss: 0.2396 | LM: 0.2274 | LB: 1.0905 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.418/SR1: 0.388 | LR: 1.00e-04 [2026-04-17 11:40:14] Epoch 2 | Step 12310 | Loss: 0.2396 | LM: 0.2270 | LB: 1.0905 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.418/SR1: 0.388 | LR: 1.00e-04 [2026-04-17 11:40:21] Epoch 2 | Step 12320 | Loss: 0.2394 | LM: 0.2269 | LB: 1.0905 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.418/SR1: 0.388 | LR: 1.00e-04 [2026-04-17 11:40:27] Epoch 2 | Step 12330 | Loss: 0.2395 | LM: 0.2270 | LB: 1.0905 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.418/SR1: 0.388 | LR: 1.00e-04 [2026-04-17 11:40:34] Epoch 2 | Step 12340 | Loss: 0.2396 | LM: 0.2269 | LB: 1.0905 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.418/SR1: 0.388 | LR: 1.00e-04 [2026-04-17 11:40:40] Epoch 2 | Step 12350 | Loss: 0.2395 | LM: 0.2267 | LB: 1.0904 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.418/SR1: 0.388 | LR: 1.00e-04 [2026-04-17 11:40:46] Epoch 2 | Step 12360 | Loss: 0.2396 | LM: 0.2268 | LB: 1.0905 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.418/SR1: 0.388 | LR: 1.00e-04 [2026-04-17 11:40:53] Epoch 2 | Step 12370 | Loss: 0.2394 | LM: 0.2265 | LB: 1.0904 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.418/SR1: 0.388 | LR: 1.00e-04 [2026-04-17 11:40:59] Epoch 2 | Step 12380 | Loss: 0.2394 | LM: 0.2265 | LB: 1.0905 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.418/SR1: 0.388 | LR: 1.00e-04 [2026-04-17 11:41:05] Epoch 2 | Step 12390 | Loss: 0.2394 | LM: 0.2263 | LB: 1.0905 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.418/SR1: 0.388 | LR: 1.00e-04 [2026-04-17 11:41:12] Epoch 2 | Step 12400 | Loss: 0.2394 | LM: 0.2260 | LB: 1.0904 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.418/SR1: 0.388 | LR: 1.00e-04 [2026-04-17 11:41:18] Epoch 2 | Step 12410 | Loss: 0.2394 | LM: 0.2259 | LB: 1.0905 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.418/SR1: 0.388 | LR: 1.00e-04 [2026-04-17 11:41:25] Epoch 2 | Step 12420 | Loss: 0.2394 | LM: 0.2261 | LB: 1.0905 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.418/SR1: 0.388 | LR: 1.00e-04 [2026-04-17 11:41:31] Epoch 2 | Step 12430 | Loss: 0.2394 | LM: 0.2261 | LB: 1.0905 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.418/SR1: 0.388 | LR: 1.00e-04 [2026-04-17 11:41:37] Epoch 2 | Step 12440 | Loss: 0.2393 | LM: 0.2260 | LB: 1.0905 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.418/SR1: 0.388 | LR: 1.00e-04 [2026-04-17 11:41:43] Epoch 2 | Step 12450 | Loss: 0.2394 | LM: 0.2261 | LB: 1.0904 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.418/SR1: 0.388 | LR: 1.00e-04 [2026-04-17 11:41:50] Epoch 2 | Step 12460 | Loss: 0.2393 | LM: 0.2260 | LB: 1.0904 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.418/SR1: 0.388 | LR: 1.00e-04 [2026-04-17 11:41:56] Epoch 2 | Step 12470 | Loss: 0.2395 | LM: 0.2263 | LB: 1.0904 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.418/SR1: 0.388 | LR: 1.00e-04 [2026-04-17 11:42:03] Epoch 2 | Step 12480 | Loss: 0.2394 | LM: 0.2263 | LB: 1.0903 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.418/SR1: 0.388 | LR: 1.00e-04 [2026-04-17 11:42:09] Epoch 2 | Step 12490 | Loss: 0.2393 | LM: 0.2262 | LB: 1.0903 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.418/SR1: 0.388 | LR: 1.00e-04 [2026-04-17 11:42:16] Epoch 2 | Step 12500 | Loss: 0.2395 | LM: 0.2268 | LB: 1.0903 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.418/SR1: 0.388 | LR: 1.00e-04 [2026-04-17 11:42:22] Epoch 2 | Step 12510 | Loss: 0.2397 | LM: 0.2274 | LB: 1.0903 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.418/SR1: 0.388 | LR: 1.00e-04 [2026-04-17 11:42:28] Epoch 2 | Step 12520 | Loss: 0.2398 | LM: 0.2276 | LB: 1.0903 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.418/SR1: 0.388 | LR: 1.00e-04 [2026-04-17 11:42:35] Epoch 2 | Step 12530 | Loss: 0.2398 | LM: 0.2277 | LB: 1.0903 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.417/SR1: 0.388 | LR: 1.00e-04 [2026-04-17 11:42:41] Epoch 2 | Step 12540 | Loss: 0.2396 | LM: 0.2277 | LB: 1.0903 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.417/SR1: 0.388 | LR: 1.00e-04 [2026-04-17 11:42:47] Epoch 2 | Step 12550 | Loss: 0.2394 | LM: 0.2277 | LB: 1.0903 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.417/SR1: 0.388 | LR: 1.00e-04 [2026-04-17 11:42:54] Epoch 2 | Step 12560 | Loss: 0.2394 | LM: 0.2279 | LB: 1.0903 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.417/SR1: 0.388 | LR: 1.00e-04 [2026-04-17 11:43:00] Epoch 2 | Step 12570 | Loss: 0.2396 | LM: 0.2280 | LB: 1.0904 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.418/SR1: 0.388 | LR: 1.00e-04 [2026-04-17 11:43:06] Epoch 2 | Step 12580 | Loss: 0.2394 | LM: 0.2279 | LB: 1.0903 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.418/SR1: 0.388 | LR: 1.00e-04 [2026-04-17 11:43:13] Epoch 2 | Step 12590 | Loss: 0.2392 | LM: 0.2277 | LB: 1.0903 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.417/SR1: 0.388 | LR: 1.00e-04 [2026-04-17 11:43:19] Epoch 2 | Step 12600 | Loss: 0.2394 | LM: 0.2282 | LB: 1.0902 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.417/SR1: 0.388 | LR: 1.00e-04 [2026-04-17 11:43:26] Epoch 2 | Step 12610 | Loss: 0.2393 | LM: 0.2279 | LB: 1.0903 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.418/SR1: 0.388 | LR: 1.00e-04 [2026-04-17 11:43:32] Epoch 2 | Step 12620 | Loss: 0.2392 | LM: 0.2276 | LB: 1.0903 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.418/SR1: 0.388 | LR: 1.00e-04 [2026-04-17 11:43:38] Epoch 2 | Step 12630 | Loss: 0.2392 | LM: 0.2274 | LB: 1.0902 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.417/SR1: 0.388 | LR: 1.00e-04 [2026-04-17 11:43:45] Epoch 2 | Step 12640 | Loss: 0.2391 | LM: 0.2271 | LB: 1.0902 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.417/SR1: 0.388 | LR: 1.00e-04 [2026-04-17 11:43:51] Epoch 2 | Step 12650 | Loss: 0.2394 | LM: 0.2275 | LB: 1.0901 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.417/SR1: 0.388 | LR: 1.00e-04 [2026-04-17 11:43:57] Epoch 2 | Step 12660 | Loss: 0.2395 | LM: 0.2275 | LB: 1.0901 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.417/SR1: 0.388 | LR: 1.00e-04 [2026-04-17 11:44:04] Epoch 2 | Step 12670 | Loss: 0.2395 | LM: 0.2277 | LB: 1.0901 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.417/SR1: 0.388 | LR: 1.00e-04 [2026-04-17 11:44:10] Epoch 2 | Step 12680 | Loss: 0.2397 | LM: 0.2279 | LB: 1.0900 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.417/SR1: 0.388 | LR: 1.00e-04 [2026-04-17 11:44:16] Epoch 2 | Step 12690 | Loss: 0.2397 | LM: 0.2277 | LB: 1.0900 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.417/SR1: 0.388 | LR: 1.00e-04 [2026-04-17 11:44:23] Epoch 2 | Step 12700 | Loss: 0.2399 | LM: 0.2280 | LB: 1.0900 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.417/SR1: 0.388 | LR: 1.00e-04 [2026-04-17 11:44:29] Epoch 2 | Step 12710 | Loss: 0.2398 | LM: 0.2279 | LB: 1.0899 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.417/SR1: 0.387 | LR: 1.00e-04 [2026-04-17 11:44:36] Epoch 2 | Step 12720 | Loss: 0.2397 | LM: 0.2279 | LB: 1.0899 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.417/SR1: 0.387 | LR: 1.00e-04 [2026-04-17 11:44:42] Epoch 2 | Step 12730 | Loss: 0.2396 | LM: 0.2279 | LB: 1.0899 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.417/SR1: 0.387 | LR: 1.00e-04 [2026-04-17 11:44:48] Epoch 2 | Step 12740 | Loss: 0.2395 | LM: 0.2279 | LB: 1.0899 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.417/SR1: 0.387 | LR: 1.00e-04 [2026-04-17 11:44:54] Epoch 2 | Step 12750 | Loss: 0.2396 | LM: 0.2277 | LB: 1.0899 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.417/SR1: 0.387 | LR: 1.00e-04 [2026-04-17 11:45:01] Epoch 2 | Step 12760 | Loss: 0.2398 | LM: 0.2281 | LB: 1.0898 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.417/SR1: 0.387 | LR: 9.99e-05 [2026-04-17 11:45:07] Epoch 2 | Step 12770 | Loss: 0.2397 | LM: 0.2280 | LB: 1.0898 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.417/SR1: 0.387 | LR: 9.99e-05 [2026-04-17 11:45:14] Epoch 2 | Step 12780 | Loss: 0.2397 | LM: 0.2281 | LB: 1.0898 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.417/SR1: 0.387 | LR: 9.99e-05 [2026-04-17 11:45:20] Epoch 2 | Step 12790 | Loss: 0.2397 | LM: 0.2280 | LB: 1.0897 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.417/SR1: 0.387 | LR: 9.99e-05 [2026-04-17 11:45:26] Epoch 2 | Step 12800 | Loss: 0.2399 | LM: 0.2284 | LB: 1.0897 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.417/SR1: 0.387 | LR: 9.98e-05 [2026-04-17 11:45:33] Epoch 2 | Step 12810 | Loss: 0.2400 | LM: 0.2286 | LB: 1.0897 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.417/SR1: 0.387 | LR: 9.98e-05 [2026-04-17 11:45:39] Epoch 2 | Step 12820 | Loss: 0.2401 | LM: 0.2286 | LB: 1.0897 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.417/SR1: 0.387 | LR: 9.97e-05 [2026-04-17 11:45:45] Epoch 2 | Step 12830 | Loss: 0.2402 | LM: 0.2293 | LB: 1.0896 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.417/SR1: 0.387 | LR: 9.97e-05 [2026-04-17 11:45:52] Epoch 2 | Step 12840 | Loss: 0.2403 | LM: 0.2290 | LB: 1.0896 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.417/SR1: 0.387 | LR: 9.96e-05 [2026-04-17 11:45:58] Epoch 2 | Step 12850 | Loss: 0.2402 | LM: 0.2291 | LB: 1.0896 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.417/SR1: 0.387 | LR: 9.96e-05 [2026-04-17 11:46:04] Epoch 2 | Step 12860 | Loss: 0.2403 | LM: 0.2288 | LB: 1.0896 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.417/SR1: 0.387 | LR: 9.95e-05 [2026-04-17 11:46:11] Epoch 2 | Step 12870 | Loss: 0.2403 | LM: 0.2288 | LB: 1.0896 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.417/SR1: 0.387 | LR: 9.94e-05 [2026-04-17 11:46:17] Epoch 2 | Step 12880 | Loss: 0.2402 | LM: 0.2288 | LB: 1.0896 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.417/SR1: 0.387 | LR: 9.94e-05 [2026-04-17 11:46:24] Epoch 2 | Step 12890 | Loss: 0.2403 | LM: 0.2288 | LB: 1.0896 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.417/SR1: 0.387 | LR: 9.93e-05 [2026-04-17 11:46:30] Epoch 2 | Step 12900 | Loss: 0.2404 | LM: 0.2289 | LB: 1.0896 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.417/SR1: 0.387 | LR: 9.92e-05 [2026-04-17 11:46:36] Epoch 2 | Step 12910 | Loss: 0.2404 | LM: 0.2289 | LB: 1.0896 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.417/SR1: 0.387 | LR: 9.91e-05 [2026-04-17 11:46:43] Epoch 2 | Step 12920 | Loss: 0.2404 | LM: 0.2289 | LB: 1.0897 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.417/SR1: 0.387 | LR: 9.90e-05 [2026-04-17 11:46:49] Epoch 2 | Step 12930 | Loss: 0.2402 | LM: 0.2287 | LB: 1.0896 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.417/SR1: 0.387 | LR: 9.89e-05 [2026-04-17 11:46:56] Epoch 2 | Step 12940 | Loss: 0.2403 | LM: 0.2289 | LB: 1.0896 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.417/SR1: 0.387 | LR: 9.88e-05 [2026-04-17 11:47:02] Epoch 2 | Step 12950 | Loss: 0.2403 | LM: 0.2288 | LB: 1.0896 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.417/SR1: 0.387 | LR: 9.87e-05 [2026-04-17 11:47:08] Epoch 2 | Step 12960 | Loss: 0.2402 | LM: 0.2286 | LB: 1.0896 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.417/SR1: 0.387 | LR: 9.86e-05 [2026-04-17 11:47:15] Epoch 2 | Step 12970 | Loss: 0.2402 | LM: 0.2285 | LB: 1.0896 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.417/SR1: 0.387 | LR: 9.85e-05 [2026-04-17 11:47:21] Epoch 2 | Step 12980 | Loss: 0.2400 | LM: 0.2282 | LB: 1.0896 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.417/SR1: 0.387 | LR: 9.84e-05 [2026-04-17 11:47:28] Epoch 2 | Step 12990 | Loss: 0.2401 | LM: 0.2281 | LB: 1.0896 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.417/SR1: 0.387 | LR: 9.83e-05 [2026-04-17 11:47:34] Epoch 2 | Step 13000 | Loss: 0.2401 | LM: 0.2280 | LB: 1.0895 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.417/SR1: 0.387 | LR: 9.82e-05 [2026-04-17 11:47:35] Validation | Batch 10/784 | Loss: 0.3281 | LM_LOSS: 0.3173 | LB_LOSS: 1.0855 [2026-04-17 11:47:37] Validation | Batch 20/784 | Loss: 0.3387 | LM_LOSS: 0.3279 | LB_LOSS: 1.0858 [2026-04-17 11:47:38] Validation | Batch 30/784 | Loss: 0.3241 | LM_LOSS: 0.3133 | LB_LOSS: 1.0850 [2026-04-17 11:47:39] Validation | Batch 40/784 | Loss: 0.3258 | LM_LOSS: 0.3150 | LB_LOSS: 1.0849 [2026-04-17 11:47:41] Validation | Batch 50/784 | Loss: 0.3230 | LM_LOSS: 0.3122 | LB_LOSS: 1.0842 [2026-04-17 11:47:42] Validation | Batch 60/784 | Loss: 0.3244 | LM_LOSS: 0.3136 | LB_LOSS: 1.0838 [2026-04-17 11:47:43] Validation | Batch 70/784 | Loss: 0.3221 | LM_LOSS: 0.3113 | LB_LOSS: 1.0832 [2026-04-17 11:47:45] Validation | Batch 80/784 | Loss: 0.3181 | LM_LOSS: 0.3073 | LB_LOSS: 1.0827 [2026-04-17 11:47:46] Validation | Batch 90/784 | Loss: 0.3170 | LM_LOSS: 0.3062 | LB_LOSS: 1.0832 [2026-04-17 11:47:48] Validation | Batch 100/784 | Loss: 0.3190 | LM_LOSS: 0.3082 | LB_LOSS: 1.0837 [2026-04-17 11:47:49] Validation | Batch 110/784 | Loss: 0.3140 | LM_LOSS: 0.3032 | LB_LOSS: 1.0838 [2026-04-17 11:47:50] Validation | Batch 120/784 | Loss: 0.3176 | LM_LOSS: 0.3067 | LB_LOSS: 1.0838 [2026-04-17 11:47:52] Validation | Batch 130/784 | Loss: 0.3205 | LM_LOSS: 0.3097 | LB_LOSS: 1.0837 [2026-04-17 11:47:53] Validation | Batch 140/784 | Loss: 0.3198 | LM_LOSS: 0.3089 | LB_LOSS: 1.0835 [2026-04-17 11:47:54] Validation | Batch 150/784 | Loss: 0.3158 | LM_LOSS: 0.3050 | LB_LOSS: 1.0838 [2026-04-17 11:47:55] Validation | Batch 160/784 | Loss: 0.3167 | LM_LOSS: 0.3059 | LB_LOSS: 1.0835 [2026-04-17 11:47:57] Validation | Batch 170/784 | Loss: 0.3172 | LM_LOSS: 0.3064 | LB_LOSS: 1.0832 [2026-04-17 11:47:58] Validation | Batch 180/784 | Loss: 0.3149 | LM_LOSS: 0.3041 | LB_LOSS: 1.0832 [2026-04-17 11:47:59] Validation | Batch 190/784 | Loss: 0.3169 | LM_LOSS: 0.3060 | LB_LOSS: 1.0837 [2026-04-17 11:48:01] Validation | Batch 200/784 | Loss: 0.3174 | LM_LOSS: 0.3066 | LB_LOSS: 1.0837 [2026-04-17 11:48:02] Validation | Batch 210/784 | Loss: 0.3162 | LM_LOSS: 0.3054 | LB_LOSS: 1.0836 [2026-04-17 11:48:03] Validation | Batch 220/784 | Loss: 0.3170 | LM_LOSS: 0.3062 | LB_LOSS: 1.0837 [2026-04-17 11:48:05] Validation | Batch 230/784 | Loss: 0.3176 | LM_LOSS: 0.3067 | LB_LOSS: 1.0836 [2026-04-17 11:48:06] Validation | Batch 240/784 | Loss: 0.3181 | LM_LOSS: 0.3072 | LB_LOSS: 1.0840 [2026-04-17 11:48:08] Validation | Batch 250/784 | Loss: 0.3180 | LM_LOSS: 0.3072 | LB_LOSS: 1.0838 [2026-04-17 11:48:09] Validation | Batch 260/784 | Loss: 0.3182 | LM_LOSS: 0.3073 | LB_LOSS: 1.0840 [2026-04-17 11:48:11] Validation | Batch 270/784 | Loss: 0.3179 | LM_LOSS: 0.3071 | LB_LOSS: 1.0841 [2026-04-17 11:48:12] Validation | Batch 280/784 | Loss: 0.3184 | LM_LOSS: 0.3075 | LB_LOSS: 1.0842 [2026-04-17 11:48:13] Validation | Batch 290/784 | Loss: 0.3193 | LM_LOSS: 0.3085 | LB_LOSS: 1.0843 [2026-04-17 11:48:15] Validation | Batch 300/784 | Loss: 0.3201 | LM_LOSS: 0.3093 | LB_LOSS: 1.0844 [2026-04-17 11:48:16] Validation | Batch 310/784 | Loss: 0.3195 | LM_LOSS: 0.3087 | LB_LOSS: 1.0843 [2026-04-17 11:48:17] Validation | Batch 320/784 | Loss: 0.3211 | LM_LOSS: 0.3103 | LB_LOSS: 1.0843 [2026-04-17 11:48:19] Validation | Batch 330/784 | Loss: 0.3210 | LM_LOSS: 0.3101 | LB_LOSS: 1.0843 [2026-04-17 11:48:20] Validation | Batch 340/784 | Loss: 0.3198 | LM_LOSS: 0.3090 | LB_LOSS: 1.0844 [2026-04-17 11:48:21] Validation | Batch 350/784 | Loss: 0.3200 | LM_LOSS: 0.3091 | LB_LOSS: 1.0846 [2026-04-17 11:48:22] Validation | Batch 360/784 | Loss: 0.3197 | LM_LOSS: 0.3089 | LB_LOSS: 1.0846 [2026-04-17 11:48:24] Validation | Batch 370/784 | Loss: 0.3202 | LM_LOSS: 0.3094 | LB_LOSS: 1.0845 [2026-04-17 11:48:25] Validation | Batch 380/784 | Loss: 0.3201 | LM_LOSS: 0.3092 | LB_LOSS: 1.0846 [2026-04-17 11:48:27] Validation | Batch 390/784 | Loss: 0.3199 | LM_LOSS: 0.3091 | LB_LOSS: 1.0847 [2026-04-17 11:48:28] Validation | Batch 400/784 | Loss: 0.3202 | LM_LOSS: 0.3093 | LB_LOSS: 1.0846 [2026-04-17 11:48:29] Validation | Batch 410/784 | Loss: 0.3205 | LM_LOSS: 0.3096 | LB_LOSS: 1.0847 [2026-04-17 11:48:30] Validation | Batch 420/784 | Loss: 0.3207 | LM_LOSS: 0.3099 | LB_LOSS: 1.0847 [2026-04-17 11:48:32] Validation | Batch 430/784 | Loss: 0.3208 | LM_LOSS: 0.3099 | LB_LOSS: 1.0846 [2026-04-17 11:48:33] Validation | Batch 440/784 | Loss: 0.3204 | LM_LOSS: 0.3096 | LB_LOSS: 1.0847 [2026-04-17 11:48:34] Validation | Batch 450/784 | Loss: 0.3198 | LM_LOSS: 0.3089 | LB_LOSS: 1.0846 [2026-04-17 11:48:36] Validation | Batch 460/784 | Loss: 0.3202 | LM_LOSS: 0.3094 | LB_LOSS: 1.0847 [2026-04-17 11:48:37] Validation | Batch 470/784 | Loss: 0.3195 | LM_LOSS: 0.3086 | LB_LOSS: 1.0847 [2026-04-17 11:48:38] Validation | Batch 480/784 | Loss: 0.3200 | LM_LOSS: 0.3091 | LB_LOSS: 1.0846 [2026-04-17 11:48:40] Validation | Batch 490/784 | Loss: 0.3194 | LM_LOSS: 0.3085 | LB_LOSS: 1.0846 [2026-04-17 11:48:41] Validation | Batch 500/784 | Loss: 0.3198 | LM_LOSS: 0.3090 | LB_LOSS: 1.0845 [2026-04-17 11:48:42] Validation | Batch 510/784 | Loss: 0.3195 | LM_LOSS: 0.3087 | LB_LOSS: 1.0845 [2026-04-17 11:48:44] Validation | Batch 520/784 | Loss: 0.3197 | LM_LOSS: 0.3088 | LB_LOSS: 1.0844 [2026-04-17 11:48:45] Validation | Batch 530/784 | Loss: 0.3205 | LM_LOSS: 0.3097 | LB_LOSS: 1.0844 [2026-04-17 11:48:46] Validation | Batch 540/784 | Loss: 0.3209 | LM_LOSS: 0.3100 | LB_LOSS: 1.0844 [2026-04-17 11:48:48] Validation | Batch 550/784 | Loss: 0.3221 | LM_LOSS: 0.3113 | LB_LOSS: 1.0843 [2026-04-17 11:48:49] Validation | Batch 560/784 | Loss: 0.3222 | LM_LOSS: 0.3114 | LB_LOSS: 1.0844 [2026-04-17 11:48:51] Validation | Batch 570/784 | Loss: 0.3217 | LM_LOSS: 0.3109 | LB_LOSS: 1.0843 [2026-04-17 11:48:52] Validation | Batch 580/784 | Loss: 0.3212 | LM_LOSS: 0.3104 | LB_LOSS: 1.0844 [2026-04-17 11:48:53] Validation | Batch 590/784 | Loss: 0.3214 | LM_LOSS: 0.3106 | LB_LOSS: 1.0843 [2026-04-17 11:48:55] Validation | Batch 600/784 | Loss: 0.3213 | LM_LOSS: 0.3105 | LB_LOSS: 1.0842 [2026-04-17 11:48:56] Validation | Batch 610/784 | Loss: 0.3214 | LM_LOSS: 0.3106 | LB_LOSS: 1.0842 [2026-04-17 11:48:58] Validation | Batch 620/784 | Loss: 0.3213 | LM_LOSS: 0.3104 | LB_LOSS: 1.0842 [2026-04-17 11:48:59] Validation | Batch 630/784 | Loss: 0.3220 | LM_LOSS: 0.3111 | LB_LOSS: 1.0843 [2026-04-17 11:49:01] Validation | Batch 640/784 | Loss: 0.3220 | LM_LOSS: 0.3112 | LB_LOSS: 1.0842 [2026-04-17 11:49:02] Validation | Batch 650/784 | Loss: 0.3218 | LM_LOSS: 0.3110 | LB_LOSS: 1.0843 [2026-04-17 11:49:03] Validation | Batch 660/784 | Loss: 0.3222 | LM_LOSS: 0.3114 | LB_LOSS: 1.0843 [2026-04-17 11:49:05] Validation | Batch 670/784 | Loss: 0.3227 | LM_LOSS: 0.3118 | LB_LOSS: 1.0843 [2026-04-17 11:49:06] Validation | Batch 680/784 | Loss: 0.3224 | LM_LOSS: 0.3115 | LB_LOSS: 1.0843 [2026-04-17 11:49:08] Validation | Batch 690/784 | Loss: 0.3225 | LM_LOSS: 0.3117 | LB_LOSS: 1.0843 [2026-04-17 11:49:09] Validation | Batch 700/784 | Loss: 0.3226 | LM_LOSS: 0.3118 | LB_LOSS: 1.0842 [2026-04-17 11:49:10] Validation | Batch 710/784 | Loss: 0.3224 | LM_LOSS: 0.3116 | LB_LOSS: 1.0842 [2026-04-17 11:49:12] Validation | Batch 720/784 | Loss: 0.3221 | LM_LOSS: 0.3113 | LB_LOSS: 1.0841 [2026-04-17 11:49:13] Validation | Batch 730/784 | Loss: 0.3216 | LM_LOSS: 0.3108 | LB_LOSS: 1.0841 [2026-04-17 11:49:14] Validation | Batch 740/784 | Loss: 0.3217 | LM_LOSS: 0.3109 | LB_LOSS: 1.0842 [2026-04-17 11:49:15] Validation | Batch 750/784 | Loss: 0.3210 | LM_LOSS: 0.3102 | LB_LOSS: 1.0842 [2026-04-17 11:49:17] Validation | Batch 760/784 | Loss: 0.3212 | LM_LOSS: 0.3104 | LB_LOSS: 1.0841 [2026-04-17 11:49:18] Validation | Batch 770/784 | Loss: 0.3214 | LM_LOSS: 0.3105 | LB_LOSS: 1.0842 [2026-04-17 11:49:19] Validation | Batch 780/784 | Loss: 0.3217 | LM_LOSS: 0.3109 | LB_LOSS: 1.0842 [2026-04-17 11:49:20] Validation | Batch 784/784 | Loss: 0.3220 | LM_LOSS: 0.3111 | LB_LOSS: 1.0842 [2026-04-17 11:49:23] Validation | Loss: 0.3220 | LM_LOSS: 0.3111 | LB_LOSS: 1.0842 | PPL: 1.37 | Time: 106.06s [2026-04-17 11:49:29] Epoch 2 | Step 13010 | Loss: 0.2401 | LM: 0.2281 | LB: 1.0895 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.417/SR1: 0.387 | LR: 9.80e-05 [2026-04-17 11:49:35] Epoch 2 | Step 13020 | Loss: 0.2401 | LM: 0.2283 | LB: 1.0896 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.417/SR1: 0.387 | LR: 9.79e-05 [2026-04-17 11:49:41] Epoch 2 | Step 13030 | Loss: 0.2401 | LM: 0.2282 | LB: 1.0895 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.417/SR1: 0.387 | LR: 9.78e-05 [2026-04-17 11:49:48] Epoch 2 | Step 13040 | Loss: 0.2401 | LM: 0.2283 | LB: 1.0895 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.417/SR1: 0.387 | LR: 9.76e-05 [2026-04-17 11:49:54] Epoch 2 | Step 13050 | Loss: 0.2400 | LM: 0.2282 | LB: 1.0895 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.417/SR1: 0.387 | LR: 9.75e-05 [2026-04-17 11:50:01] Epoch 2 | Step 13060 | Loss: 0.2401 | LM: 0.2283 | LB: 1.0895 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.417/SR1: 0.387 | LR: 9.73e-05 [2026-04-17 11:50:07] Epoch 2 | Step 13070 | Loss: 0.2401 | LM: 0.2281 | LB: 1.0895 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.417/SR1: 0.387 | LR: 9.72e-05 [2026-04-17 11:50:14] Epoch 2 | Step 13080 | Loss: 0.2400 | LM: 0.2281 | LB: 1.0895 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.417/SR1: 0.387 | LR: 9.70e-05 [2026-04-17 11:50:20] Epoch 2 | Step 13090 | Loss: 0.2402 | LM: 0.2284 | LB: 1.0895 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.417/SR1: 0.387 | LR: 9.69e-05 [2026-04-17 11:50:26] Epoch 2 | Step 13100 | Loss: 0.2403 | LM: 0.2283 | LB: 1.0895 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.417/SR1: 0.387 | LR: 9.67e-05 [2026-04-17 11:50:33] Epoch 2 | Step 13110 | Loss: 0.2403 | LM: 0.2284 | LB: 1.0894 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.417/SR1: 0.387 | LR: 9.65e-05 [2026-04-17 11:50:39] Epoch 2 | Step 13120 | Loss: 0.2402 | LM: 0.2284 | LB: 1.0895 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.417/SR1: 0.387 | LR: 9.64e-05 [2026-04-17 11:50:46] Epoch 2 | Step 13130 | Loss: 0.2400 | LM: 0.2285 | LB: 1.0894 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.417/SR1: 0.387 | LR: 9.62e-05 [2026-04-17 11:50:52] Epoch 2 | Step 13140 | Loss: 0.2400 | LM: 0.2284 | LB: 1.0894 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.417/SR1: 0.387 | LR: 9.60e-05 [2026-04-17 11:50:58] Epoch 2 | Step 13150 | Loss: 0.2401 | LM: 0.2282 | LB: 1.0894 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.417/SR1: 0.387 | LR: 9.58e-05 [2026-04-17 11:51:05] Epoch 2 | Step 13160 | Loss: 0.2403 | LM: 0.2284 | LB: 1.0894 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.417/SR1: 0.387 | LR: 9.56e-05 [2026-04-17 11:51:11] Epoch 2 | Step 13170 | Loss: 0.2402 | LM: 0.2282 | LB: 1.0894 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.417/SR1: 0.387 | LR: 9.54e-05 [2026-04-17 11:51:18] Epoch 2 | Step 13180 | Loss: 0.2403 | LM: 0.2282 | LB: 1.0894 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.417/SR1: 0.387 | LR: 9.52e-05 [2026-04-17 11:51:24] Epoch 2 | Step 13190 | Loss: 0.2403 | LM: 0.2282 | LB: 1.0894 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.417/SR1: 0.387 | LR: 9.50e-05 [2026-04-17 11:51:30] Epoch 2 | Step 13200 | Loss: 0.2402 | LM: 0.2281 | LB: 1.0893 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.417/SR1: 0.387 | LR: 9.48e-05 [2026-04-17 11:51:37] Epoch 2 | Step 13210 | Loss: 0.2404 | LM: 0.2285 | LB: 1.0894 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.417/SR1: 0.387 | LR: 9.46e-05 [2026-04-17 11:51:44] Epoch 2 | Step 13220 | Loss: 0.2404 | LM: 0.2286 | LB: 1.0893 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.417/SR1: 0.387 | LR: 9.44e-05 [2026-04-17 11:51:50] Epoch 2 | Step 13230 | Loss: 0.2404 | LM: 0.2287 | LB: 1.0893 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.417/SR1: 0.387 | LR: 9.42e-05 [2026-04-17 11:51:57] Epoch 2 | Step 13240 | Loss: 0.2403 | LM: 0.2289 | LB: 1.0893 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.417/SR1: 0.387 | LR: 9.40e-05 [2026-04-17 11:52:03] Epoch 2 | Step 13250 | Loss: 0.2403 | LM: 0.2290 | LB: 1.0893 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.417/SR1: 0.387 | LR: 9.37e-05 [2026-04-17 11:52:09] Epoch 2 | Step 13260 | Loss: 0.2402 | LM: 0.2290 | LB: 1.0893 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.417/SR1: 0.387 | LR: 9.35e-05 [2026-04-17 11:52:16] Epoch 2 | Step 13270 | Loss: 0.2402 | LM: 0.2288 | LB: 1.0893 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.417/SR1: 0.387 | LR: 9.33e-05 [2026-04-17 11:52:22] Epoch 2 | Step 13280 | Loss: 0.2401 | LM: 0.2289 | LB: 1.0893 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.417/SR1: 0.387 | LR: 9.30e-05 [2026-04-17 11:52:28] Epoch 2 | Step 13290 | Loss: 0.2401 | LM: 0.2288 | LB: 1.0893 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.417/SR1: 0.387 | LR: 9.28e-05 [2026-04-17 11:52:35] Epoch 2 | Step 13300 | Loss: 0.2401 | LM: 0.2287 | LB: 1.0893 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.417/SR1: 0.387 | LR: 9.26e-05 [2026-04-17 11:52:41] Epoch 2 | Step 13310 | Loss: 0.2400 | LM: 0.2286 | LB: 1.0893 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.417/SR1: 0.387 | LR: 9.23e-05 [2026-04-17 11:52:47] Epoch 2 | Step 13320 | Loss: 0.2401 | LM: 0.2288 | LB: 1.0892 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.417/SR1: 0.387 | LR: 9.21e-05 [2026-04-17 11:52:53] Epoch 2 | Step 13330 | Loss: 0.2400 | LM: 0.2286 | LB: 1.0893 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.417/SR1: 0.387 | LR: 9.18e-05 [2026-04-17 11:53:00] Epoch 2 | Step 13340 | Loss: 0.2399 | LM: 0.2284 | LB: 1.0892 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.417/SR1: 0.387 | LR: 9.15e-05 [2026-04-17 11:53:06] Epoch 2 | Step 13350 | Loss: 0.2400 | LM: 0.2284 | LB: 1.0892 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.417/SR1: 0.387 | LR: 9.13e-05 [2026-04-17 11:53:13] Epoch 2 | Step 13360 | Loss: 0.2398 | LM: 0.2281 | LB: 1.0893 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.417/SR1: 0.387 | LR: 9.10e-05 [2026-04-17 11:53:19] Epoch 2 | Step 13370 | Loss: 0.2397 | LM: 0.2281 | LB: 1.0892 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.417/SR1: 0.387 | LR: 9.08e-05 [2026-04-17 11:53:25] Epoch 2 | Step 13380 | Loss: 0.2397 | LM: 0.2280 | LB: 1.0892 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.417/SR1: 0.387 | LR: 9.05e-05 [2026-04-17 11:53:31] Epoch 2 | Step 13390 | Loss: 0.2399 | LM: 0.2284 | LB: 1.0893 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.417/SR1: 0.387 | LR: 9.02e-05 [2026-04-17 11:53:37] Epoch 2 | Step 13400 | Loss: 0.2400 | LM: 0.2286 | LB: 1.0892 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.417/SR1: 0.387 | LR: 8.99e-05 [2026-04-17 11:53:44] Epoch 2 | Step 13410 | Loss: 0.2401 | LM: 0.2285 | LB: 1.0893 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.417/SR1: 0.387 | LR: 8.96e-05 [2026-04-17 11:53:50] Epoch 2 | Step 13420 | Loss: 0.2401 | LM: 0.2286 | LB: 1.0893 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.417/SR1: 0.387 | LR: 8.94e-05 [2026-04-17 11:53:56] Epoch 2 | Step 13430 | Loss: 0.2401 | LM: 0.2284 | LB: 1.0892 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.387 | LR: 8.91e-05 [2026-04-17 11:54:03] Epoch 2 | Step 13440 | Loss: 0.2402 | LM: 0.2286 | LB: 1.0892 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.387 | LR: 8.88e-05 [2026-04-17 11:54:09] Epoch 2 | Step 13450 | Loss: 0.2402 | LM: 0.2286 | LB: 1.0892 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.387 | LR: 8.85e-05 [2026-04-17 11:54:15] Epoch 2 | Step 13460 | Loss: 0.2401 | LM: 0.2283 | LB: 1.0892 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.387 | LR: 8.82e-05 [2026-04-17 11:54:21] Epoch 2 | Step 13470 | Loss: 0.2400 | LM: 0.2282 | LB: 1.0892 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.387 | LR: 8.79e-05 [2026-04-17 11:54:28] Epoch 2 | Step 13480 | Loss: 0.2400 | LM: 0.2282 | LB: 1.0893 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.387 | LR: 8.76e-05 [2026-04-17 11:54:34] Epoch 2 | Step 13490 | Loss: 0.2401 | LM: 0.2282 | LB: 1.0893 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.387 | LR: 8.73e-05 [2026-04-17 11:54:40] Epoch 2 | Step 13500 | Loss: 0.2400 | LM: 0.2283 | LB: 1.0893 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.417/SR1: 0.387 | LR: 8.70e-05 [2026-04-17 11:54:46] Epoch 2 | Step 13510 | Loss: 0.2401 | LM: 0.2282 | LB: 1.0892 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.387 | LR: 8.66e-05 [2026-04-17 11:54:53] Epoch 2 | Step 13520 | Loss: 0.2402 | LM: 0.2285 | LB: 1.0892 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.417/SR1: 0.387 | LR: 8.63e-05 [2026-04-17 11:54:59] Epoch 2 | Step 13530 | Loss: 0.2401 | LM: 0.2285 | LB: 1.0892 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.417/SR1: 0.387 | LR: 8.60e-05 [2026-04-17 11:55:06] Epoch 2 | Step 13540 | Loss: 0.2401 | LM: 0.2284 | LB: 1.0892 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.417/SR1: 0.387 | LR: 8.57e-05 [2026-04-17 11:55:12] Epoch 2 | Step 13550 | Loss: 0.2400 | LM: 0.2284 | LB: 1.0891 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.387 | LR: 8.54e-05 [2026-04-17 11:55:18] Epoch 2 | Step 13560 | Loss: 0.2399 | LM: 0.2287 | LB: 1.0892 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.417/SR1: 0.387 | LR: 8.50e-05 [2026-04-17 11:55:24] Epoch 2 | Step 13570 | Loss: 0.2399 | LM: 0.2288 | LB: 1.0892 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.417/SR1: 0.387 | LR: 8.47e-05 [2026-04-17 11:55:31] Epoch 2 | Step 13580 | Loss: 0.2400 | LM: 0.2287 | LB: 1.0891 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.387 | LR: 8.44e-05 [2026-04-17 11:55:37] Epoch 2 | Step 13590 | Loss: 0.2399 | LM: 0.2284 | LB: 1.0892 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.417/SR1: 0.387 | LR: 8.40e-05 [2026-04-17 11:55:43] Epoch 2 | Step 13600 | Loss: 0.2399 | LM: 0.2286 | LB: 1.0892 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.417/SR1: 0.387 | LR: 8.37e-05 [2026-04-17 11:55:49] Epoch 2 | Step 13610 | Loss: 0.2398 | LM: 0.2284 | LB: 1.0891 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.417/SR1: 0.387 | LR: 8.33e-05 [2026-04-17 11:55:55] Epoch 2 | Step 13620 | Loss: 0.2398 | LM: 0.2283 | LB: 1.0891 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.417/SR1: 0.387 | LR: 8.30e-05 [2026-04-17 11:56:02] Epoch 2 | Step 13630 | Loss: 0.2397 | LM: 0.2280 | LB: 1.0891 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.417/SR1: 0.387 | LR: 8.26e-05 [2026-04-17 11:56:08] Epoch 2 | Step 13640 | Loss: 0.2397 | LM: 0.2280 | LB: 1.0891 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.417/SR1: 0.387 | LR: 8.23e-05 [2026-04-17 11:56:14] Epoch 2 | Step 13650 | Loss: 0.2397 | LM: 0.2279 | LB: 1.0891 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.417/SR1: 0.387 | LR: 8.19e-05 [2026-04-17 11:56:20] Epoch 2 | Step 13660 | Loss: 0.2397 | LM: 0.2278 | LB: 1.0891 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.417/SR1: 0.387 | LR: 8.16e-05 [2026-04-17 11:56:27] Epoch 2 | Step 13670 | Loss: 0.2397 | LM: 0.2277 | LB: 1.0891 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.417/SR1: 0.387 | LR: 8.12e-05 [2026-04-17 11:56:33] Epoch 2 | Step 13680 | Loss: 0.2396 | LM: 0.2275 | LB: 1.0891 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.417/SR1: 0.387 | LR: 8.08e-05 [2026-04-17 11:56:39] Epoch 2 | Step 13690 | Loss: 0.2396 | LM: 0.2278 | LB: 1.0891 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.417/SR1: 0.387 | LR: 8.05e-05 [2026-04-17 11:56:45] Epoch 2 | Step 13700 | Loss: 0.2396 | LM: 0.2279 | LB: 1.0891 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.417/SR1: 0.387 | LR: 8.01e-05 [2026-04-17 11:56:52] Epoch 2 | Step 13710 | Loss: 0.2396 | LM: 0.2280 | LB: 1.0891 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.417/SR1: 0.387 | LR: 7.97e-05 [2026-04-17 11:56:58] Epoch 2 | Step 13720 | Loss: 0.2396 | LM: 0.2281 | LB: 1.0891 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.417/SR1: 0.387 | LR: 7.94e-05 [2026-04-17 11:57:04] Epoch 2 | Step 13730 | Loss: 0.2396 | LM: 0.2280 | LB: 1.0891 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.417/SR1: 0.387 | LR: 7.90e-05 [2026-04-17 11:57:11] Epoch 2 | Step 13740 | Loss: 0.2397 | LM: 0.2279 | LB: 1.0890 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.417/SR1: 0.387 | LR: 7.86e-05 [2026-04-17 11:57:17] Epoch 2 | Step 13750 | Loss: 0.2398 | LM: 0.2279 | LB: 1.0891 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.417/SR1: 0.387 | LR: 7.82e-05 [2026-04-17 11:57:23] Epoch 2 | Step 13760 | Loss: 0.2397 | LM: 0.2277 | LB: 1.0890 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.417/SR1: 0.387 | LR: 7.79e-05 [2026-04-17 11:57:29] Epoch 2 | Step 13770 | Loss: 0.2397 | LM: 0.2280 | LB: 1.0890 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.417/SR1: 0.387 | LR: 7.75e-05 [2026-04-17 11:57:36] Epoch 2 | Step 13780 | Loss: 0.2396 | LM: 0.2280 | LB: 1.0891 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.417/SR1: 0.387 | LR: 7.71e-05 [2026-04-17 11:57:42] Epoch 2 | Step 13790 | Loss: 0.2396 | LM: 0.2280 | LB: 1.0891 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.417/SR1: 0.387 | LR: 7.67e-05 [2026-04-17 11:57:48] Epoch 2 | Step 13800 | Loss: 0.2396 | LM: 0.2281 | LB: 1.0891 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.417/SR1: 0.387 | LR: 7.63e-05 [2026-04-17 11:57:54] Epoch 2 | Step 13810 | Loss: 0.2395 | LM: 0.2282 | LB: 1.0891 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.417/SR1: 0.387 | LR: 7.59e-05 [2026-04-17 11:58:01] Epoch 2 | Step 13820 | Loss: 0.2395 | LM: 0.2281 | LB: 1.0890 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.417/SR1: 0.387 | LR: 7.55e-05 [2026-04-17 11:58:07] Epoch 2 | Step 13830 | Loss: 0.2396 | LM: 0.2281 | LB: 1.0890 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.417/SR1: 0.387 | LR: 7.51e-05 [2026-04-17 11:58:13] Epoch 2 | Step 13840 | Loss: 0.2395 | LM: 0.2283 | LB: 1.0890 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.387 | LR: 7.47e-05 [2026-04-17 11:58:19] Epoch 2 | Step 13850 | Loss: 0.2397 | LM: 0.2286 | LB: 1.0890 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.387 | LR: 7.43e-05 [2026-04-17 11:58:26] Epoch 2 | Step 13860 | Loss: 0.2397 | LM: 0.2286 | LB: 1.0890 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.417/SR1: 0.387 | LR: 7.39e-05 [2026-04-17 11:58:32] Epoch 2 | Step 13870 | Loss: 0.2397 | LM: 0.2285 | LB: 1.0890 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.387 | LR: 7.35e-05 [2026-04-17 11:58:38] Epoch 2 | Step 13880 | Loss: 0.2399 | LM: 0.2288 | LB: 1.0890 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.387 | LR: 7.31e-05 [2026-04-17 11:58:45] Epoch 2 | Step 13890 | Loss: 0.2399 | LM: 0.2290 | LB: 1.0890 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.387 | LR: 7.27e-05 [2026-04-17 11:58:51] Epoch 2 | Step 13900 | Loss: 0.2400 | LM: 0.2292 | LB: 1.0890 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.387 | LR: 7.23e-05 [2026-04-17 11:58:57] Epoch 2 | Step 13910 | Loss: 0.2399 | LM: 0.2289 | LB: 1.0890 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.386 | LR: 7.19e-05 [2026-04-17 11:59:04] Epoch 2 | Step 13920 | Loss: 0.2399 | LM: 0.2289 | LB: 1.0890 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.387 | LR: 7.15e-05 [2026-04-17 11:59:10] Epoch 2 | Step 13930 | Loss: 0.2400 | LM: 0.2288 | LB: 1.0890 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.387 | LR: 7.10e-05 [2026-04-17 11:59:17] Epoch 2 | Step 13940 | Loss: 0.2400 | LM: 0.2290 | LB: 1.0890 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.387 | LR: 7.06e-05 [2026-04-17 11:59:23] Epoch 2 | Step 13950 | Loss: 0.2401 | LM: 0.2289 | LB: 1.0890 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.386 | LR: 7.02e-05 [2026-04-17 11:59:30] Epoch 2 | Step 13960 | Loss: 0.2401 | LM: 0.2290 | LB: 1.0890 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.386 | LR: 6.98e-05 [2026-04-17 11:59:36] Epoch 2 | Step 13970 | Loss: 0.2401 | LM: 0.2291 | LB: 1.0890 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.386 | LR: 6.94e-05 [2026-04-17 11:59:42] Epoch 2 | Step 13980 | Loss: 0.2400 | LM: 0.2290 | LB: 1.0890 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.387 | LR: 6.89e-05 [2026-04-17 11:59:49] Epoch 2 | Step 13990 | Loss: 0.2400 | LM: 0.2290 | LB: 1.0890 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.387 | LR: 6.85e-05 [2026-04-17 11:59:55] Epoch 2 | Step 14000 | Loss: 0.2402 | LM: 0.2290 | LB: 1.0890 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.386 | LR: 6.81e-05 [2026-04-17 11:59:56] Validation | Batch 10/784 | Loss: 0.3260 | LM_LOSS: 0.3152 | LB_LOSS: 1.0846 [2026-04-17 11:59:58] Validation | Batch 20/784 | Loss: 0.3384 | LM_LOSS: 0.3275 | LB_LOSS: 1.0849 [2026-04-17 11:59:59] Validation | Batch 30/784 | Loss: 0.3235 | LM_LOSS: 0.3126 | LB_LOSS: 1.0842 [2026-04-17 12:00:00] Validation | Batch 40/784 | Loss: 0.3257 | LM_LOSS: 0.3149 | LB_LOSS: 1.0842 [2026-04-17 12:00:02] Validation | Batch 50/784 | Loss: 0.3229 | LM_LOSS: 0.3120 | LB_LOSS: 1.0835 [2026-04-17 12:00:03] Validation | Batch 60/784 | Loss: 0.3245 | LM_LOSS: 0.3137 | LB_LOSS: 1.0830 [2026-04-17 12:00:04] Validation | Batch 70/784 | Loss: 0.3216 | LM_LOSS: 0.3108 | LB_LOSS: 1.0824 [2026-04-17 12:00:06] Validation | Batch 80/784 | Loss: 0.3178 | LM_LOSS: 0.3070 | LB_LOSS: 1.0819 [2026-04-17 12:00:07] Validation | Batch 90/784 | Loss: 0.3164 | LM_LOSS: 0.3056 | LB_LOSS: 1.0825 [2026-04-17 12:00:08] Validation | Batch 100/784 | Loss: 0.3185 | LM_LOSS: 0.3077 | LB_LOSS: 1.0829 [2026-04-17 12:00:10] Validation | Batch 110/784 | Loss: 0.3134 | LM_LOSS: 0.3026 | LB_LOSS: 1.0831 [2026-04-17 12:00:11] Validation | Batch 120/784 | Loss: 0.3170 | LM_LOSS: 0.3061 | LB_LOSS: 1.0830 [2026-04-17 12:00:12] Validation | Batch 130/784 | Loss: 0.3199 | LM_LOSS: 0.3091 | LB_LOSS: 1.0829 [2026-04-17 12:00:14] Validation | Batch 140/784 | Loss: 0.3193 | LM_LOSS: 0.3084 | LB_LOSS: 1.0827 [2026-04-17 12:00:15] Validation | Batch 150/784 | Loss: 0.3153 | LM_LOSS: 0.3044 | LB_LOSS: 1.0831 [2026-04-17 12:00:17] Validation | Batch 160/784 | Loss: 0.3160 | LM_LOSS: 0.3052 | LB_LOSS: 1.0827 [2026-04-17 12:00:18] Validation | Batch 170/784 | Loss: 0.3163 | LM_LOSS: 0.3054 | LB_LOSS: 1.0825 [2026-04-17 12:00:19] Validation | Batch 180/784 | Loss: 0.3139 | LM_LOSS: 0.3030 | LB_LOSS: 1.0825 [2026-04-17 12:00:21] Validation | Batch 190/784 | Loss: 0.3158 | LM_LOSS: 0.3050 | LB_LOSS: 1.0829 [2026-04-17 12:00:22] Validation | Batch 200/784 | Loss: 0.3163 | LM_LOSS: 0.3054 | LB_LOSS: 1.0830 [2026-04-17 12:00:23] Validation | Batch 210/784 | Loss: 0.3153 | LM_LOSS: 0.3045 | LB_LOSS: 1.0829 [2026-04-17 12:00:25] Validation | Batch 220/784 | Loss: 0.3160 | LM_LOSS: 0.3051 | LB_LOSS: 1.0829 [2026-04-17 12:00:26] Validation | Batch 230/784 | Loss: 0.3165 | LM_LOSS: 0.3057 | LB_LOSS: 1.0828 [2026-04-17 12:00:27] Validation | Batch 240/784 | Loss: 0.3169 | LM_LOSS: 0.3061 | LB_LOSS: 1.0832 [2026-04-17 12:00:29] Validation | Batch 250/784 | Loss: 0.3168 | LM_LOSS: 0.3060 | LB_LOSS: 1.0830 [2026-04-17 12:00:30] Validation | Batch 260/784 | Loss: 0.3169 | LM_LOSS: 0.3061 | LB_LOSS: 1.0832 [2026-04-17 12:00:32] Validation | Batch 270/784 | Loss: 0.3167 | LM_LOSS: 0.3058 | LB_LOSS: 1.0833 [2026-04-17 12:00:33] Validation | Batch 280/784 | Loss: 0.3172 | LM_LOSS: 0.3063 | LB_LOSS: 1.0834 [2026-04-17 12:00:34] Validation | Batch 290/784 | Loss: 0.3183 | LM_LOSS: 0.3075 | LB_LOSS: 1.0836 [2026-04-17 12:00:36] Validation | Batch 300/784 | Loss: 0.3191 | LM_LOSS: 0.3082 | LB_LOSS: 1.0836 [2026-04-17 12:00:37] Validation | Batch 310/784 | Loss: 0.3185 | LM_LOSS: 0.3077 | LB_LOSS: 1.0835 [2026-04-17 12:00:39] Validation | Batch 320/784 | Loss: 0.3201 | LM_LOSS: 0.3092 | LB_LOSS: 1.0835 [2026-04-17 12:00:40] Validation | Batch 330/784 | Loss: 0.3199 | LM_LOSS: 0.3090 | LB_LOSS: 1.0835 [2026-04-17 12:00:41] Validation | Batch 340/784 | Loss: 0.3187 | LM_LOSS: 0.3079 | LB_LOSS: 1.0836 [2026-04-17 12:00:43] Validation | Batch 350/784 | Loss: 0.3189 | LM_LOSS: 0.3080 | LB_LOSS: 1.0838 [2026-04-17 12:00:44] Validation | Batch 360/784 | Loss: 0.3186 | LM_LOSS: 0.3078 | LB_LOSS: 1.0838 [2026-04-17 12:00:45] Validation | Batch 370/784 | Loss: 0.3191 | LM_LOSS: 0.3082 | LB_LOSS: 1.0837 [2026-04-17 12:00:46] Validation | Batch 380/784 | Loss: 0.3189 | LM_LOSS: 0.3081 | LB_LOSS: 1.0838 [2026-04-17 12:00:48] Validation | Batch 390/784 | Loss: 0.3188 | LM_LOSS: 0.3080 | LB_LOSS: 1.0839 [2026-04-17 12:00:49] Validation | Batch 400/784 | Loss: 0.3190 | LM_LOSS: 0.3082 | LB_LOSS: 1.0838 [2026-04-17 12:00:50] Validation | Batch 410/784 | Loss: 0.3193 | LM_LOSS: 0.3085 | LB_LOSS: 1.0839 [2026-04-17 12:00:51] Validation | Batch 420/784 | Loss: 0.3195 | LM_LOSS: 0.3087 | LB_LOSS: 1.0839 [2026-04-17 12:00:53] Validation | Batch 430/784 | Loss: 0.3195 | LM_LOSS: 0.3087 | LB_LOSS: 1.0838 [2026-04-17 12:00:54] Validation | Batch 440/784 | Loss: 0.3192 | LM_LOSS: 0.3083 | LB_LOSS: 1.0839 [2026-04-17 12:00:55] Validation | Batch 450/784 | Loss: 0.3185 | LM_LOSS: 0.3077 | LB_LOSS: 1.0838 [2026-04-17 12:00:56] Validation | Batch 460/784 | Loss: 0.3190 | LM_LOSS: 0.3081 | LB_LOSS: 1.0839 [2026-04-17 12:00:58] Validation | Batch 470/784 | Loss: 0.3182 | LM_LOSS: 0.3073 | LB_LOSS: 1.0839 [2026-04-17 12:00:59] Validation | Batch 480/784 | Loss: 0.3186 | LM_LOSS: 0.3078 | LB_LOSS: 1.0838 [2026-04-17 12:01:01] Validation | Batch 490/784 | Loss: 0.3180 | LM_LOSS: 0.3071 | LB_LOSS: 1.0838 [2026-04-17 12:01:02] Validation | Batch 500/784 | Loss: 0.3183 | LM_LOSS: 0.3075 | LB_LOSS: 1.0837 [2026-04-17 12:01:03] Validation | Batch 510/784 | Loss: 0.3180 | LM_LOSS: 0.3072 | LB_LOSS: 1.0837 [2026-04-17 12:01:05] Validation | Batch 520/784 | Loss: 0.3182 | LM_LOSS: 0.3074 | LB_LOSS: 1.0836 [2026-04-17 12:01:06] Validation | Batch 530/784 | Loss: 0.3191 | LM_LOSS: 0.3082 | LB_LOSS: 1.0836 [2026-04-17 12:01:07] Validation | Batch 540/784 | Loss: 0.3194 | LM_LOSS: 0.3086 | LB_LOSS: 1.0836 [2026-04-17 12:01:09] Validation | Batch 550/784 | Loss: 0.3207 | LM_LOSS: 0.3099 | LB_LOSS: 1.0835 [2026-04-17 12:01:10] Validation | Batch 560/784 | Loss: 0.3208 | LM_LOSS: 0.3100 | LB_LOSS: 1.0836 [2026-04-17 12:01:12] Validation | Batch 570/784 | Loss: 0.3204 | LM_LOSS: 0.3095 | LB_LOSS: 1.0835 [2026-04-17 12:01:13] Validation | Batch 580/784 | Loss: 0.3199 | LM_LOSS: 0.3090 | LB_LOSS: 1.0835 [2026-04-17 12:01:14] Validation | Batch 590/784 | Loss: 0.3201 | LM_LOSS: 0.3093 | LB_LOSS: 1.0835 [2026-04-17 12:01:16] Validation | Batch 600/784 | Loss: 0.3200 | LM_LOSS: 0.3091 | LB_LOSS: 1.0834 [2026-04-17 12:01:17] Validation | Batch 610/784 | Loss: 0.3201 | LM_LOSS: 0.3092 | LB_LOSS: 1.0834 [2026-04-17 12:01:18] Validation | Batch 620/784 | Loss: 0.3199 | LM_LOSS: 0.3091 | LB_LOSS: 1.0834 [2026-04-17 12:01:20] Validation | Batch 630/784 | Loss: 0.3206 | LM_LOSS: 0.3098 | LB_LOSS: 1.0834 [2026-04-17 12:01:22] Validation | Batch 640/784 | Loss: 0.3207 | LM_LOSS: 0.3099 | LB_LOSS: 1.0834 [2026-04-17 12:01:23] Validation | Batch 650/784 | Loss: 0.3206 | LM_LOSS: 0.3098 | LB_LOSS: 1.0835 [2026-04-17 12:01:24] Validation | Batch 660/784 | Loss: 0.3209 | LM_LOSS: 0.3101 | LB_LOSS: 1.0835 [2026-04-17 12:01:26] Validation | Batch 670/784 | Loss: 0.3213 | LM_LOSS: 0.3105 | LB_LOSS: 1.0835 [2026-04-17 12:01:27] Validation | Batch 680/784 | Loss: 0.3210 | LM_LOSS: 0.3102 | LB_LOSS: 1.0835 [2026-04-17 12:01:29] Validation | Batch 690/784 | Loss: 0.3212 | LM_LOSS: 0.3104 | LB_LOSS: 1.0835 [2026-04-17 12:01:30] Validation | Batch 700/784 | Loss: 0.3213 | LM_LOSS: 0.3104 | LB_LOSS: 1.0834 [2026-04-17 12:01:31] Validation | Batch 710/784 | Loss: 0.3210 | LM_LOSS: 0.3102 | LB_LOSS: 1.0834 [2026-04-17 12:01:33] Validation | Batch 720/784 | Loss: 0.3208 | LM_LOSS: 0.3099 | LB_LOSS: 1.0833 [2026-04-17 12:01:34] Validation | Batch 730/784 | Loss: 0.3203 | LM_LOSS: 0.3094 | LB_LOSS: 1.0833 [2026-04-17 12:01:35] Validation | Batch 740/784 | Loss: 0.3204 | LM_LOSS: 0.3095 | LB_LOSS: 1.0833 [2026-04-17 12:01:36] Validation | Batch 750/784 | Loss: 0.3197 | LM_LOSS: 0.3089 | LB_LOSS: 1.0833 [2026-04-17 12:01:38] Validation | Batch 760/784 | Loss: 0.3198 | LM_LOSS: 0.3090 | LB_LOSS: 1.0833 [2026-04-17 12:01:39] Validation | Batch 770/784 | Loss: 0.3200 | LM_LOSS: 0.3092 | LB_LOSS: 1.0834 [2026-04-17 12:01:41] Validation | Batch 780/784 | Loss: 0.3204 | LM_LOSS: 0.3096 | LB_LOSS: 1.0834 [2026-04-17 12:01:41] Validation | Batch 784/784 | Loss: 0.3206 | LM_LOSS: 0.3098 | LB_LOSS: 1.0834 [2026-04-17 12:01:44] Validation | Loss: 0.3206 | LM_LOSS: 0.3098 | LB_LOSS: 1.0834 | PPL: 1.36 | Time: 106.22s [2026-04-17 12:01:50] Epoch 2 | Step 14010 | Loss: 0.2401 | LM: 0.2291 | LB: 1.0890 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.386 | LR: 6.77e-05 [2026-04-17 12:01:57] Epoch 2 | Step 14020 | Loss: 0.2400 | LM: 0.2289 | LB: 1.0890 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.386 | LR: 6.72e-05 [2026-04-17 12:02:03] Epoch 2 | Step 14030 | Loss: 0.2399 | LM: 0.2287 | LB: 1.0890 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.386 | LR: 6.68e-05 [2026-04-17 12:02:10] Epoch 2 | Step 14040 | Loss: 0.2398 | LM: 0.2287 | LB: 1.0890 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.386 | LR: 6.64e-05 [2026-04-17 12:02:16] Epoch 2 | Step 14050 | Loss: 0.2399 | LM: 0.2287 | LB: 1.0889 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.386 | LR: 6.60e-05 [2026-04-17 12:02:22] Epoch 2 | Step 14060 | Loss: 0.2398 | LM: 0.2287 | LB: 1.0889 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.386 | LR: 6.55e-05 [2026-04-17 12:02:29] Epoch 2 | Step 14070 | Loss: 0.2399 | LM: 0.2289 | LB: 1.0889 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.386 | LR: 6.51e-05 [2026-04-17 12:02:35] Epoch 2 | Step 14080 | Loss: 0.2398 | LM: 0.2289 | LB: 1.0890 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.386 | LR: 6.47e-05 [2026-04-17 12:02:42] Epoch 2 | Step 14090 | Loss: 0.2400 | LM: 0.2293 | LB: 1.0889 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.386 | LR: 6.42e-05 [2026-04-17 12:02:48] Epoch 2 | Step 14100 | Loss: 0.2401 | LM: 0.2294 | LB: 1.0889 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.386 | LR: 6.38e-05 [2026-04-17 12:02:54] Epoch 2 | Step 14110 | Loss: 0.2400 | LM: 0.2292 | LB: 1.0889 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.386 | LR: 6.33e-05 [2026-04-17 12:03:01] Epoch 2 | Step 14120 | Loss: 0.2399 | LM: 0.2291 | LB: 1.0889 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.386 | LR: 6.29e-05 [2026-04-17 12:03:07] Epoch 2 | Step 14130 | Loss: 0.2401 | LM: 0.2291 | LB: 1.0889 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.386 | LR: 6.25e-05 [2026-04-17 12:03:13] Epoch 2 | Step 14140 | Loss: 0.2400 | LM: 0.2288 | LB: 1.0889 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.386 | LR: 6.20e-05 [2026-04-17 12:03:20] Epoch 2 | Step 14150 | Loss: 0.2401 | LM: 0.2288 | LB: 1.0889 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.386 | LR: 6.16e-05 [2026-04-17 12:03:26] Epoch 2 | Step 14160 | Loss: 0.2401 | LM: 0.2288 | LB: 1.0888 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.386 | LR: 6.12e-05 [2026-04-17 12:03:33] Epoch 2 | Step 14170 | Loss: 0.2402 | LM: 0.2289 | LB: 1.0888 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.386 | LR: 6.07e-05 [2026-04-17 12:03:39] Epoch 2 | Step 14180 | Loss: 0.2402 | LM: 0.2290 | LB: 1.0889 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.386 | LR: 6.03e-05 [2026-04-17 12:03:45] Epoch 2 | Step 14190 | Loss: 0.2403 | LM: 0.2293 | LB: 1.0889 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.386 | LR: 5.98e-05 [2026-04-17 12:03:52] Epoch 2 | Step 14200 | Loss: 0.2402 | LM: 0.2291 | LB: 1.0889 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.386 | LR: 5.94e-05 [2026-04-17 12:03:58] Epoch 2 | Step 14210 | Loss: 0.2401 | LM: 0.2290 | LB: 1.0889 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.386 | LR: 5.89e-05 [2026-04-17 12:04:04] Epoch 2 | Step 14220 | Loss: 0.2401 | LM: 0.2291 | LB: 1.0889 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.386 | LR: 5.85e-05 [2026-04-17 12:04:11] Epoch 2 | Step 14230 | Loss: 0.2402 | LM: 0.2292 | LB: 1.0889 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.386 | LR: 5.81e-05 [2026-04-17 12:04:17] Epoch 2 | Step 14240 | Loss: 0.2401 | LM: 0.2290 | LB: 1.0889 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.386 | LR: 5.76e-05 [2026-04-17 12:04:23] Epoch 2 | Step 14250 | Loss: 0.2401 | LM: 0.2290 | LB: 1.0889 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.386 | LR: 5.72e-05 [2026-04-17 12:04:30] Epoch 2 | Step 14260 | Loss: 0.2401 | LM: 0.2289 | LB: 1.0889 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.386 | LR: 5.67e-05 [2026-04-17 12:04:36] Epoch 2 | Step 14270 | Loss: 0.2401 | LM: 0.2288 | LB: 1.0889 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.386 | LR: 5.63e-05 [2026-04-17 12:04:43] Epoch 2 | Step 14280 | Loss: 0.2402 | LM: 0.2287 | LB: 1.0889 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.386 | LR: 5.58e-05 [2026-04-17 12:04:49] Epoch 2 | Step 14290 | Loss: 0.2402 | LM: 0.2286 | LB: 1.0889 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.386 | LR: 5.54e-05 [2026-04-17 12:04:56] Epoch 2 | Step 14300 | Loss: 0.2402 | LM: 0.2286 | LB: 1.0889 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.386 | LR: 5.49e-05 [2026-04-17 12:05:02] Epoch 2 | Step 14310 | Loss: 0.2401 | LM: 0.2286 | LB: 1.0889 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.386 | LR: 5.45e-05 [2026-04-17 12:05:09] Epoch 2 | Step 14320 | Loss: 0.2401 | LM: 0.2284 | LB: 1.0889 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.386 | LR: 5.41e-05 [2026-04-17 12:05:15] Epoch 2 | Step 14330 | Loss: 0.2401 | LM: 0.2282 | LB: 1.0889 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.386 | LR: 5.36e-05 [2026-04-17 12:05:21] Epoch 2 | Step 14340 | Loss: 0.2401 | LM: 0.2283 | LB: 1.0889 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.386 | LR: 5.32e-05 [2026-04-17 12:05:28] Epoch 2 | Step 14350 | Loss: 0.2402 | LM: 0.2283 | LB: 1.0889 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.386 | LR: 5.27e-05 [2026-04-17 12:05:34] Epoch 2 | Step 14360 | Loss: 0.2401 | LM: 0.2284 | LB: 1.0889 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.386 | LR: 5.23e-05 [2026-04-17 12:05:40] Epoch 2 | Step 14370 | Loss: 0.2401 | LM: 0.2285 | LB: 1.0889 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.386 | LR: 5.18e-05 [2026-04-17 12:05:47] Epoch 2 | Step 14380 | Loss: 0.2401 | LM: 0.2285 | LB: 1.0889 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.386 | LR: 5.14e-05 [2026-04-17 12:05:53] Epoch 2 | Step 14390 | Loss: 0.2400 | LM: 0.2283 | LB: 1.0889 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.386 | LR: 5.09e-05 [2026-04-17 12:06:00] Epoch 2 | Step 14400 | Loss: 0.2400 | LM: 0.2283 | LB: 1.0889 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.386 | LR: 5.05e-05 [2026-04-17 12:06:06] Epoch 2 | Step 14410 | Loss: 0.2399 | LM: 0.2281 | LB: 1.0888 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.386 | LR: 5.01e-05 [2026-04-17 12:06:12] Epoch 2 | Step 14420 | Loss: 0.2399 | LM: 0.2280 | LB: 1.0889 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.386 | LR: 4.96e-05 [2026-04-17 12:06:19] Epoch 2 | Step 14430 | Loss: 0.2398 | LM: 0.2280 | LB: 1.0889 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.386 | LR: 4.92e-05 [2026-04-17 12:06:25] Epoch 2 | Step 14440 | Loss: 0.2398 | LM: 0.2280 | LB: 1.0889 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.386 | LR: 4.87e-05 [2026-04-17 12:06:31] Epoch 2 | Step 14450 | Loss: 0.2398 | LM: 0.2281 | LB: 1.0889 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.386 | LR: 4.83e-05 [2026-04-17 12:06:38] Epoch 2 | Step 14460 | Loss: 0.2398 | LM: 0.2280 | LB: 1.0889 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.386 | LR: 4.79e-05 [2026-04-17 12:06:44] Epoch 2 | Step 14470 | Loss: 0.2399 | LM: 0.2284 | LB: 1.0889 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.386 | LR: 4.74e-05 [2026-04-17 12:06:51] Epoch 2 | Step 14480 | Loss: 0.2399 | LM: 0.2283 | LB: 1.0889 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.386 | LR: 4.70e-05 [2026-04-17 12:06:57] Epoch 2 | Step 14490 | Loss: 0.2399 | LM: 0.2283 | LB: 1.0889 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.386 | LR: 4.65e-05 [2026-04-17 12:07:04] Epoch 2 | Step 14500 | Loss: 0.2399 | LM: 0.2284 | LB: 1.0888 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.386 | LR: 4.61e-05 [2026-04-17 12:07:10] Epoch 2 | Step 14510 | Loss: 0.2399 | LM: 0.2284 | LB: 1.0888 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.386 | LR: 4.57e-05 [2026-04-17 12:07:16] Epoch 2 | Step 14520 | Loss: 0.2399 | LM: 0.2285 | LB: 1.0888 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.386 | LR: 4.52e-05 [2026-04-17 12:07:23] Epoch 2 | Step 14530 | Loss: 0.2399 | LM: 0.2283 | LB: 1.0888 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.386 | LR: 4.48e-05 [2026-04-17 12:07:29] Epoch 2 | Step 14540 | Loss: 0.2399 | LM: 0.2283 | LB: 1.0888 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.386 | LR: 4.44e-05 [2026-04-17 12:07:35] Epoch 2 | Step 14550 | Loss: 0.2400 | LM: 0.2281 | LB: 1.0888 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.386 | LR: 4.39e-05 [2026-04-17 12:07:42] Epoch 2 | Step 14560 | Loss: 0.2400 | LM: 0.2280 | LB: 1.0888 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.386 | LR: 4.35e-05 [2026-04-17 12:07:48] Epoch 2 | Step 14570 | Loss: 0.2400 | LM: 0.2281 | LB: 1.0888 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.386 | LR: 4.31e-05 [2026-04-17 12:07:55] Epoch 2 | Step 14580 | Loss: 0.2400 | LM: 0.2278 | LB: 1.0888 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.386 | LR: 4.26e-05 [2026-04-17 12:08:01] Epoch 2 | Step 14590 | Loss: 0.2400 | LM: 0.2278 | LB: 1.0888 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.386 | LR: 4.22e-05 [2026-04-17 12:08:07] Epoch 2 | Step 14600 | Loss: 0.2399 | LM: 0.2279 | LB: 1.0888 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.386 | LR: 4.18e-05 [2026-04-17 12:08:14] Epoch 2 | Step 14610 | Loss: 0.2399 | LM: 0.2279 | LB: 1.0888 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.386 | LR: 4.14e-05 [2026-04-17 12:08:20] Epoch 2 | Step 14620 | Loss: 0.2398 | LM: 0.2277 | LB: 1.0888 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.386 | LR: 4.09e-05 [2026-04-17 12:08:27] Epoch 2 | Step 14630 | Loss: 0.2398 | LM: 0.2277 | LB: 1.0888 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.386 | LR: 4.05e-05 [2026-04-17 12:08:33] Epoch 2 | Step 14640 | Loss: 0.2399 | LM: 0.2276 | LB: 1.0888 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.386 | LR: 4.01e-05 [2026-04-17 12:08:39] Epoch 2 | Step 14650 | Loss: 0.2399 | LM: 0.2276 | LB: 1.0887 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.386 | LR: 3.97e-05 [2026-04-17 12:08:46] Epoch 2 | Step 14660 | Loss: 0.2399 | LM: 0.2278 | LB: 1.0888 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.386 | LR: 3.93e-05 [2026-04-17 12:08:52] Epoch 2 | Step 14670 | Loss: 0.2400 | LM: 0.2279 | LB: 1.0887 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.386 | LR: 3.89e-05 [2026-04-17 12:08:59] Epoch 2 | Step 14680 | Loss: 0.2399 | LM: 0.2278 | LB: 1.0887 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.386 | LR: 3.84e-05 [2026-04-17 12:09:05] Epoch 2 | Step 14690 | Loss: 0.2399 | LM: 0.2276 | LB: 1.0887 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.386 | LR: 3.80e-05 [2026-04-17 12:09:12] Epoch 2 | Step 14700 | Loss: 0.2399 | LM: 0.2276 | LB: 1.0887 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.386 | LR: 3.76e-05 [2026-04-17 12:09:18] Epoch 2 | Step 14710 | Loss: 0.2399 | LM: 0.2276 | LB: 1.0887 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.386 | LR: 3.72e-05 [2026-04-17 12:09:24] Epoch 2 | Step 14720 | Loss: 0.2399 | LM: 0.2275 | LB: 1.0887 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.386 | LR: 3.68e-05 [2026-04-17 12:09:31] Epoch 2 | Step 14730 | Loss: 0.2399 | LM: 0.2275 | LB: 1.0887 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.386 | LR: 3.64e-05 [2026-04-17 12:09:37] Epoch 2 | Step 14740 | Loss: 0.2398 | LM: 0.2276 | LB: 1.0887 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.386 | LR: 3.60e-05 [2026-04-17 12:09:43] Epoch 2 | Step 14750 | Loss: 0.2399 | LM: 0.2276 | LB: 1.0887 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.386 | LR: 3.56e-05 [2026-04-17 12:09:50] Epoch 2 | Step 14760 | Loss: 0.2399 | LM: 0.2275 | LB: 1.0887 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.386 | LR: 3.52e-05 [2026-04-17 12:09:56] Epoch 2 | Step 14770 | Loss: 0.2399 | LM: 0.2275 | LB: 1.0887 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.386 | LR: 3.48e-05 [2026-04-17 12:10:03] Epoch 2 | Step 14780 | Loss: 0.2399 | LM: 0.2276 | LB: 1.0887 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.386 | LR: 3.44e-05 [2026-04-17 12:10:09] Epoch 2 | Step 14790 | Loss: 0.2399 | LM: 0.2277 | LB: 1.0887 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.386 | LR: 3.40e-05 [2026-04-17 12:10:16] Epoch 2 | Step 14800 | Loss: 0.2398 | LM: 0.2276 | LB: 1.0887 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.386 | LR: 3.36e-05 [2026-04-17 12:10:22] Epoch 2 | Step 14810 | Loss: 0.2398 | LM: 0.2277 | LB: 1.0887 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.386 | LR: 3.32e-05 [2026-04-17 12:10:28] Epoch 2 | Step 14820 | Loss: 0.2398 | LM: 0.2276 | LB: 1.0887 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.386 | LR: 3.28e-05 [2026-04-17 12:10:35] Epoch 2 | Step 14830 | Loss: 0.2398 | LM: 0.2277 | LB: 1.0887 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.386 | LR: 3.24e-05 [2026-04-17 12:10:41] Epoch 2 | Step 14840 | Loss: 0.2397 | LM: 0.2275 | LB: 1.0886 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.386 | LR: 3.21e-05 [2026-04-17 12:10:48] Epoch 2 | Step 14850 | Loss: 0.2397 | LM: 0.2275 | LB: 1.0886 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.386 | LR: 3.17e-05 [2026-04-17 12:10:54] Epoch 2 | Step 14860 | Loss: 0.2397 | LM: 0.2273 | LB: 1.0886 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.386 | LR: 3.13e-05 [2026-04-17 12:11:00] Epoch 2 | Step 14870 | Loss: 0.2396 | LM: 0.2275 | LB: 1.0886 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.386 | LR: 3.09e-05 [2026-04-17 12:11:07] Epoch 2 | Step 14880 | Loss: 0.2397 | LM: 0.2275 | LB: 1.0886 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.386 | LR: 3.05e-05 [2026-04-17 12:11:13] Epoch 2 | Step 14890 | Loss: 0.2396 | LM: 0.2273 | LB: 1.0886 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.386 | LR: 3.02e-05 [2026-04-17 12:11:20] Epoch 2 | Step 14900 | Loss: 0.2396 | LM: 0.2272 | LB: 1.0885 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.386 | LR: 2.98e-05 [2026-04-17 12:11:26] Epoch 2 | Step 14910 | Loss: 0.2396 | LM: 0.2272 | LB: 1.0885 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.386 | LR: 2.94e-05 [2026-04-17 12:11:33] Epoch 2 | Step 14920 | Loss: 0.2396 | LM: 0.2271 | LB: 1.0885 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.386 | LR: 2.91e-05 [2026-04-17 12:11:39] Epoch 2 | Step 14930 | Loss: 0.2395 | LM: 0.2271 | LB: 1.0885 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.386 | LR: 2.87e-05 [2026-04-17 12:11:45] Epoch 2 | Step 14940 | Loss: 0.2395 | LM: 0.2271 | LB: 1.0885 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.386 | LR: 2.83e-05 [2026-04-17 12:11:51] Epoch 2 | Step 14950 | Loss: 0.2396 | LM: 0.2272 | LB: 1.0885 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.386 | LR: 2.80e-05 [2026-04-17 12:11:58] Epoch 2 | Step 14960 | Loss: 0.2396 | LM: 0.2271 | LB: 1.0885 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.386 | LR: 2.76e-05 [2026-04-17 12:12:04] Epoch 2 | Step 14970 | Loss: 0.2396 | LM: 0.2272 | LB: 1.0885 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.386 | LR: 2.73e-05 [2026-04-17 12:12:11] Epoch 2 | Step 14980 | Loss: 0.2396 | LM: 0.2272 | LB: 1.0885 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.386 | LR: 2.69e-05 [2026-04-17 12:12:17] Epoch 2 | Step 14990 | Loss: 0.2396 | LM: 0.2270 | LB: 1.0885 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.386 | LR: 2.66e-05 [2026-04-17 12:12:24] Epoch 2 | Step 15000 | Loss: 0.2396 | LM: 0.2271 | LB: 1.0885 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.386 | LR: 2.62e-05 [2026-04-17 12:12:32] Checkpoint saved: outputs/2026-04-17/08-57-56/checkpoints/checkpoint_step_15000.pt [2026-04-17 12:12:48] Validation | Batch 10/784 | Loss: 0.3280 | LM_LOSS: 0.3171 | LB_LOSS: 1.0842 [2026-04-17 12:12:50] Validation | Batch 20/784 | Loss: 0.3381 | LM_LOSS: 0.3272 | LB_LOSS: 1.0844 [2026-04-17 12:12:51] Validation | Batch 30/784 | Loss: 0.3232 | LM_LOSS: 0.3123 | LB_LOSS: 1.0837 [2026-04-17 12:12:53] Validation | Batch 40/784 | Loss: 0.3255 | LM_LOSS: 0.3146 | LB_LOSS: 1.0836 [2026-04-17 12:12:54] Validation | Batch 50/784 | Loss: 0.3225 | LM_LOSS: 0.3117 | LB_LOSS: 1.0829 [2026-04-17 12:12:55] Validation | Batch 60/784 | Loss: 0.3243 | LM_LOSS: 0.3135 | LB_LOSS: 1.0825 [2026-04-17 12:12:57] Validation | Batch 70/784 | Loss: 0.3215 | LM_LOSS: 0.3107 | LB_LOSS: 1.0818 [2026-04-17 12:12:58] Validation | Batch 80/784 | Loss: 0.3176 | LM_LOSS: 0.3067 | LB_LOSS: 1.0814 [2026-04-17 12:12:59] Validation | Batch 90/784 | Loss: 0.3164 | LM_LOSS: 0.3056 | LB_LOSS: 1.0819 [2026-04-17 12:13:00] Validation | Batch 100/784 | Loss: 0.3182 | LM_LOSS: 0.3074 | LB_LOSS: 1.0824 [2026-04-17 12:13:02] Validation | Batch 110/784 | Loss: 0.3130 | LM_LOSS: 0.3022 | LB_LOSS: 1.0825 [2026-04-17 12:13:03] Validation | Batch 120/784 | Loss: 0.3166 | LM_LOSS: 0.3058 | LB_LOSS: 1.0824 [2026-04-17 12:13:05] Validation | Batch 130/784 | Loss: 0.3196 | LM_LOSS: 0.3087 | LB_LOSS: 1.0824 [2026-04-17 12:13:06] Validation | Batch 140/784 | Loss: 0.3189 | LM_LOSS: 0.3080 | LB_LOSS: 1.0822 [2026-04-17 12:13:07] Validation | Batch 150/784 | Loss: 0.3149 | LM_LOSS: 0.3041 | LB_LOSS: 1.0825 [2026-04-17 12:13:08] Validation | Batch 160/784 | Loss: 0.3156 | LM_LOSS: 0.3048 | LB_LOSS: 1.0822 [2026-04-17 12:13:10] Validation | Batch 170/784 | Loss: 0.3158 | LM_LOSS: 0.3050 | LB_LOSS: 1.0819 [2026-04-17 12:13:11] Validation | Batch 180/784 | Loss: 0.3135 | LM_LOSS: 0.3026 | LB_LOSS: 1.0819 [2026-04-17 12:13:12] Validation | Batch 190/784 | Loss: 0.3155 | LM_LOSS: 0.3047 | LB_LOSS: 1.0824 [2026-04-17 12:13:14] Validation | Batch 200/784 | Loss: 0.3160 | LM_LOSS: 0.3051 | LB_LOSS: 1.0824 [2026-04-17 12:13:15] Validation | Batch 210/784 | Loss: 0.3149 | LM_LOSS: 0.3041 | LB_LOSS: 1.0823 [2026-04-17 12:13:16] Validation | Batch 220/784 | Loss: 0.3156 | LM_LOSS: 0.3048 | LB_LOSS: 1.0824 [2026-04-17 12:13:18] Validation | Batch 230/784 | Loss: 0.3162 | LM_LOSS: 0.3053 | LB_LOSS: 1.0823 [2026-04-17 12:13:19] Validation | Batch 240/784 | Loss: 0.3165 | LM_LOSS: 0.3057 | LB_LOSS: 1.0826 [2026-04-17 12:13:21] Validation | Batch 250/784 | Loss: 0.3164 | LM_LOSS: 0.3056 | LB_LOSS: 1.0825 [2026-04-17 12:13:22] Validation | Batch 260/784 | Loss: 0.3166 | LM_LOSS: 0.3058 | LB_LOSS: 1.0827 [2026-04-17 12:13:24] Validation | Batch 270/784 | Loss: 0.3164 | LM_LOSS: 0.3056 | LB_LOSS: 1.0827 [2026-04-17 12:13:25] Validation | Batch 280/784 | Loss: 0.3169 | LM_LOSS: 0.3060 | LB_LOSS: 1.0829 [2026-04-17 12:13:26] Validation | Batch 290/784 | Loss: 0.3179 | LM_LOSS: 0.3071 | LB_LOSS: 1.0830 [2026-04-17 12:13:28] Validation | Batch 300/784 | Loss: 0.3187 | LM_LOSS: 0.3079 | LB_LOSS: 1.0830 [2026-04-17 12:13:29] Validation | Batch 310/784 | Loss: 0.3181 | LM_LOSS: 0.3073 | LB_LOSS: 1.0830 [2026-04-17 12:13:31] Validation | Batch 320/784 | Loss: 0.3197 | LM_LOSS: 0.3089 | LB_LOSS: 1.0830 [2026-04-17 12:13:32] Validation | Batch 330/784 | Loss: 0.3195 | LM_LOSS: 0.3087 | LB_LOSS: 1.0830 [2026-04-17 12:13:33] Validation | Batch 340/784 | Loss: 0.3184 | LM_LOSS: 0.3075 | LB_LOSS: 1.0831 [2026-04-17 12:13:34] Validation | Batch 350/784 | Loss: 0.3185 | LM_LOSS: 0.3077 | LB_LOSS: 1.0832 [2026-04-17 12:13:36] Validation | Batch 360/784 | Loss: 0.3183 | LM_LOSS: 0.3075 | LB_LOSS: 1.0833 [2026-04-17 12:13:37] Validation | Batch 370/784 | Loss: 0.3188 | LM_LOSS: 0.3080 | LB_LOSS: 1.0832 [2026-04-17 12:13:38] Validation | Batch 380/784 | Loss: 0.3186 | LM_LOSS: 0.3078 | LB_LOSS: 1.0832 [2026-04-17 12:13:40] Validation | Batch 390/784 | Loss: 0.3185 | LM_LOSS: 0.3077 | LB_LOSS: 1.0833 [2026-04-17 12:13:41] Validation | Batch 400/784 | Loss: 0.3188 | LM_LOSS: 0.3079 | LB_LOSS: 1.0833 [2026-04-17 12:13:42] Validation | Batch 410/784 | Loss: 0.3191 | LM_LOSS: 0.3082 | LB_LOSS: 1.0833 [2026-04-17 12:13:44] Validation | Batch 420/784 | Loss: 0.3193 | LM_LOSS: 0.3085 | LB_LOSS: 1.0834 [2026-04-17 12:13:45] Validation | Batch 430/784 | Loss: 0.3194 | LM_LOSS: 0.3085 | LB_LOSS: 1.0833 [2026-04-17 12:13:46] Validation | Batch 440/784 | Loss: 0.3190 | LM_LOSS: 0.3082 | LB_LOSS: 1.0833 [2026-04-17 12:13:48] Validation | Batch 450/784 | Loss: 0.3183 | LM_LOSS: 0.3075 | LB_LOSS: 1.0833 [2026-04-17 12:13:49] Validation | Batch 460/784 | Loss: 0.3188 | LM_LOSS: 0.3080 | LB_LOSS: 1.0834 [2026-04-17 12:13:50] Validation | Batch 470/784 | Loss: 0.3180 | LM_LOSS: 0.3072 | LB_LOSS: 1.0833 [2026-04-17 12:13:52] Validation | Batch 480/784 | Loss: 0.3184 | LM_LOSS: 0.3076 | LB_LOSS: 1.0833 [2026-04-17 12:13:53] Validation | Batch 490/784 | Loss: 0.3178 | LM_LOSS: 0.3069 | LB_LOSS: 1.0832 [2026-04-17 12:13:54] Validation | Batch 500/784 | Loss: 0.3182 | LM_LOSS: 0.3073 | LB_LOSS: 1.0832 [2026-04-17 12:13:56] Validation | Batch 510/784 | Loss: 0.3179 | LM_LOSS: 0.3070 | LB_LOSS: 1.0831 [2026-04-17 12:13:57] Validation | Batch 520/784 | Loss: 0.3180 | LM_LOSS: 0.3072 | LB_LOSS: 1.0830 [2026-04-17 12:13:59] Validation | Batch 530/784 | Loss: 0.3189 | LM_LOSS: 0.3080 | LB_LOSS: 1.0830 [2026-04-17 12:14:00] Validation | Batch 540/784 | Loss: 0.3192 | LM_LOSS: 0.3084 | LB_LOSS: 1.0830 [2026-04-17 12:14:01] Validation | Batch 550/784 | Loss: 0.3205 | LM_LOSS: 0.3097 | LB_LOSS: 1.0830 [2026-04-17 12:14:03] Validation | Batch 560/784 | Loss: 0.3206 | LM_LOSS: 0.3097 | LB_LOSS: 1.0830 [2026-04-17 12:14:04] Validation | Batch 570/784 | Loss: 0.3201 | LM_LOSS: 0.3093 | LB_LOSS: 1.0829 [2026-04-17 12:14:05] Validation | Batch 580/784 | Loss: 0.3196 | LM_LOSS: 0.3088 | LB_LOSS: 1.0830 [2026-04-17 12:14:07] Validation | Batch 590/784 | Loss: 0.3199 | LM_LOSS: 0.3090 | LB_LOSS: 1.0829 [2026-04-17 12:14:08] Validation | Batch 600/784 | Loss: 0.3197 | LM_LOSS: 0.3089 | LB_LOSS: 1.0829 [2026-04-17 12:14:10] Validation | Batch 610/784 | Loss: 0.3198 | LM_LOSS: 0.3090 | LB_LOSS: 1.0829 [2026-04-17 12:14:11] Validation | Batch 620/784 | Loss: 0.3197 | LM_LOSS: 0.3089 | LB_LOSS: 1.0829 [2026-04-17 12:14:12] Validation | Batch 630/784 | Loss: 0.3204 | LM_LOSS: 0.3096 | LB_LOSS: 1.0829 [2026-04-17 12:14:14] Validation | Batch 640/784 | Loss: 0.3205 | LM_LOSS: 0.3097 | LB_LOSS: 1.0829 [2026-04-17 12:14:15] Validation | Batch 650/784 | Loss: 0.3204 | LM_LOSS: 0.3095 | LB_LOSS: 1.0829 [2026-04-17 12:14:16] Validation | Batch 660/784 | Loss: 0.3207 | LM_LOSS: 0.3099 | LB_LOSS: 1.0829 [2026-04-17 12:14:18] Validation | Batch 670/784 | Loss: 0.3211 | LM_LOSS: 0.3103 | LB_LOSS: 1.0830 [2026-04-17 12:14:19] Validation | Batch 680/784 | Loss: 0.3208 | LM_LOSS: 0.3100 | LB_LOSS: 1.0830 [2026-04-17 12:14:21] Validation | Batch 690/784 | Loss: 0.3210 | LM_LOSS: 0.3102 | LB_LOSS: 1.0829 [2026-04-17 12:14:22] Validation | Batch 700/784 | Loss: 0.3210 | LM_LOSS: 0.3102 | LB_LOSS: 1.0829 [2026-04-17 12:14:23] Validation | Batch 710/784 | Loss: 0.3208 | LM_LOSS: 0.3100 | LB_LOSS: 1.0828 [2026-04-17 12:14:25] Validation | Batch 720/784 | Loss: 0.3205 | LM_LOSS: 0.3097 | LB_LOSS: 1.0828 [2026-04-17 12:14:26] Validation | Batch 730/784 | Loss: 0.3201 | LM_LOSS: 0.3092 | LB_LOSS: 1.0827 [2026-04-17 12:14:27] Validation | Batch 740/784 | Loss: 0.3201 | LM_LOSS: 0.3093 | LB_LOSS: 1.0828 [2026-04-17 12:14:28] Validation | Batch 750/784 | Loss: 0.3195 | LM_LOSS: 0.3086 | LB_LOSS: 1.0828 [2026-04-17 12:14:30] Validation | Batch 760/784 | Loss: 0.3196 | LM_LOSS: 0.3088 | LB_LOSS: 1.0828 [2026-04-17 12:14:31] Validation | Batch 770/784 | Loss: 0.3198 | LM_LOSS: 0.3090 | LB_LOSS: 1.0828 [2026-04-17 12:14:32] Validation | Batch 780/784 | Loss: 0.3201 | LM_LOSS: 0.3093 | LB_LOSS: 1.0828 [2026-04-17 12:14:33] Validation | Batch 784/784 | Loss: 0.3203 | LM_LOSS: 0.3095 | LB_LOSS: 1.0828 [2026-04-17 12:14:36] Validation | Loss: 0.3203 | LM_LOSS: 0.3095 | LB_LOSS: 1.0828 | PPL: 1.36 | Time: 106.01s [2026-04-17 12:14:42] Epoch 2 | Step 15010 | Loss: 0.2397 | LM: 0.2273 | LB: 1.0885 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.386 | LR: 2.59e-05 [2026-04-17 12:14:49] Epoch 2 | Step 15020 | Loss: 0.2397 | LM: 0.2274 | LB: 1.0884 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.386 | LR: 2.56e-05 [2026-04-17 12:14:55] Epoch 2 | Step 15030 | Loss: 0.2398 | LM: 0.2274 | LB: 1.0885 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.386 | LR: 2.52e-05 [2026-04-17 12:15:02] Epoch 2 | Step 15040 | Loss: 0.2398 | LM: 0.2273 | LB: 1.0884 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.386 | LR: 2.49e-05 [2026-04-17 12:15:08] Epoch 2 | Step 15050 | Loss: 0.2398 | LM: 0.2272 | LB: 1.0884 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.386 | LR: 2.46e-05 [2026-04-17 12:15:15] Epoch 2 | Step 15060 | Loss: 0.2399 | LM: 0.2273 | LB: 1.0884 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.386 | LR: 2.42e-05 [2026-04-17 12:15:21] Epoch 2 | Step 15070 | Loss: 0.2399 | LM: 0.2274 | LB: 1.0885 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.386 | LR: 2.39e-05 [2026-04-17 12:15:27] Epoch 2 | Step 15080 | Loss: 0.2399 | LM: 0.2273 | LB: 1.0885 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.386 | LR: 2.36e-05 [2026-04-17 12:15:34] Epoch 2 | Step 15090 | Loss: 0.2400 | LM: 0.2274 | LB: 1.0885 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.386 | LR: 2.33e-05 [2026-04-17 12:15:40] Epoch 2 | Step 15100 | Loss: 0.2399 | LM: 0.2274 | LB: 1.0885 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.386 | LR: 2.30e-05 [2026-04-17 12:15:47] Epoch 2 | Step 15110 | Loss: 0.2400 | LM: 0.2276 | LB: 1.0885 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.386 | LR: 2.27e-05 [2026-04-17 12:15:53] Epoch 2 | Step 15120 | Loss: 0.2401 | LM: 0.2277 | LB: 1.0885 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.386 | LR: 2.23e-05 [2026-04-17 12:16:00] Epoch 2 | Step 15130 | Loss: 0.2402 | LM: 0.2278 | LB: 1.0885 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.386 | LR: 2.20e-05 [2026-04-17 12:16:06] Epoch 2 | Step 15140 | Loss: 0.2402 | LM: 0.2277 | LB: 1.0885 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.386 | LR: 2.17e-05 [2026-04-17 12:16:13] Epoch 2 | Step 15150 | Loss: 0.2402 | LM: 0.2276 | LB: 1.0885 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.386 | LR: 2.14e-05 [2026-04-17 12:16:19] Epoch 2 | Step 15160 | Loss: 0.2401 | LM: 0.2276 | LB: 1.0885 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.386 | LR: 2.11e-05 [2026-04-17 12:16:26] Epoch 2 | Step 15170 | Loss: 0.2402 | LM: 0.2278 | LB: 1.0885 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.386 | LR: 2.09e-05 [2026-04-17 12:16:32] Epoch 2 | Step 15180 | Loss: 0.2401 | LM: 0.2277 | LB: 1.0885 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.386 | LR: 2.06e-05 [2026-04-17 12:16:38] Epoch 2 | Step 15190 | Loss: 0.2402 | LM: 0.2276 | LB: 1.0885 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.386 | LR: 2.03e-05 [2026-04-17 12:16:45] Epoch 2 | Step 15200 | Loss: 0.2402 | LM: 0.2278 | LB: 1.0885 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.386 | LR: 2.00e-05 [2026-04-17 12:16:51] Epoch 2 | Step 15210 | Loss: 0.2402 | LM: 0.2279 | LB: 1.0885 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.386 | LR: 1.97e-05 [2026-04-17 12:16:58] Epoch 2 | Step 15220 | Loss: 0.2401 | LM: 0.2278 | LB: 1.0885 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.386 | LR: 1.94e-05 [2026-04-17 12:17:04] Epoch 2 | Step 15230 | Loss: 0.2401 | LM: 0.2278 | LB: 1.0885 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.386 | LR: 1.92e-05 [2026-04-17 12:17:10] Epoch 2 | Step 15240 | Loss: 0.2401 | LM: 0.2280 | LB: 1.0885 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.386 | LR: 1.89e-05 [2026-04-17 12:17:17] Epoch 2 | Step 15250 | Loss: 0.2401 | LM: 0.2280 | LB: 1.0885 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.386 | LR: 1.86e-05 [2026-04-17 12:17:23] Epoch 2 | Step 15260 | Loss: 0.2401 | LM: 0.2281 | LB: 1.0885 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.386 | LR: 1.84e-05 [2026-04-17 12:17:30] Epoch 2 | Step 15270 | Loss: 0.2402 | LM: 0.2281 | LB: 1.0885 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.386 | LR: 1.81e-05 [2026-04-17 12:17:36] Epoch 2 | Step 15280 | Loss: 0.2401 | LM: 0.2280 | LB: 1.0884 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.386 | LR: 1.79e-05 [2026-04-17 12:17:42] Epoch 2 | Step 15290 | Loss: 0.2401 | LM: 0.2280 | LB: 1.0884 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.386 | LR: 1.76e-05 [2026-04-17 12:17:49] Epoch 2 | Step 15300 | Loss: 0.2401 | LM: 0.2281 | LB: 1.0884 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.386 | LR: 1.74e-05 [2026-04-17 12:17:55] Epoch 2 | Step 15310 | Loss: 0.2401 | LM: 0.2281 | LB: 1.0884 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.386 | LR: 1.71e-05 [2026-04-17 12:18:01] Epoch 2 | Step 15320 | Loss: 0.2401 | LM: 0.2280 | LB: 1.0884 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.386 | LR: 1.69e-05 [2026-04-17 12:18:08] Epoch 2 | Step 15330 | Loss: 0.2402 | LM: 0.2280 | LB: 1.0884 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.386 | LR: 1.67e-05 [2026-04-17 12:18:14] Epoch 2 | Step 15340 | Loss: 0.2402 | LM: 0.2280 | LB: 1.0884 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.386 | LR: 1.64e-05 [2026-04-17 12:18:20] Epoch 2 | Step 15350 | Loss: 0.2401 | LM: 0.2279 | LB: 1.0884 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.386 | LR: 1.62e-05 [2026-04-17 12:18:26] Epoch 2 | Step 15360 | Loss: 0.2401 | LM: 0.2279 | LB: 1.0884 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.386 | LR: 1.60e-05 [2026-04-17 12:18:33] Epoch 2 | Step 15370 | Loss: 0.2402 | LM: 0.2281 | LB: 1.0884 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.386 | LR: 1.58e-05 [2026-04-17 12:18:39] Epoch 2 | Step 15380 | Loss: 0.2403 | LM: 0.2280 | LB: 1.0884 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.386 | LR: 1.55e-05 [2026-04-17 12:18:46] Epoch 2 | Step 15390 | Loss: 0.2403 | LM: 0.2282 | LB: 1.0884 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.386 | LR: 1.53e-05 [2026-04-17 12:18:52] Epoch 2 | Step 15400 | Loss: 0.2403 | LM: 0.2283 | LB: 1.0884 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.386 | LR: 1.51e-05 [2026-04-17 12:18:59] Epoch 2 | Step 15410 | Loss: 0.2403 | LM: 0.2282 | LB: 1.0884 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.386 | LR: 1.49e-05 [2026-04-17 12:19:05] Epoch 2 | Step 15420 | Loss: 0.2402 | LM: 0.2281 | LB: 1.0884 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.386 | LR: 1.47e-05 [2026-04-17 12:19:11] Epoch 2 | Step 15430 | Loss: 0.2402 | LM: 0.2281 | LB: 1.0884 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.386 | LR: 1.45e-05 [2026-04-17 12:19:18] Epoch 2 | Step 15440 | Loss: 0.2404 | LM: 0.2283 | LB: 1.0884 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.386 | LR: 1.43e-05 [2026-04-17 12:19:24] Epoch 2 | Step 15450 | Loss: 0.2403 | LM: 0.2283 | LB: 1.0884 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.386 | LR: 1.41e-05 [2026-04-17 12:19:30] Epoch 2 | Step 15460 | Loss: 0.2404 | LM: 0.2282 | LB: 1.0884 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.386 | LR: 1.40e-05 [2026-04-17 12:19:37] Epoch 2 | Step 15470 | Loss: 0.2404 | LM: 0.2282 | LB: 1.0885 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.386 | LR: 1.38e-05 [2026-04-17 12:19:43] Epoch 2 | Step 15480 | Loss: 0.2403 | LM: 0.2281 | LB: 1.0884 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.386 | LR: 1.36e-05 [2026-04-17 12:19:49] Epoch 2 | Step 15490 | Loss: 0.2402 | LM: 0.2282 | LB: 1.0884 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.386 | LR: 1.34e-05 [2026-04-17 12:19:56] Epoch 2 | Step 15500 | Loss: 0.2403 | LM: 0.2281 | LB: 1.0884 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.386 | LR: 1.33e-05 [2026-04-17 12:20:02] Epoch 2 | Step 15510 | Loss: 0.2402 | LM: 0.2281 | LB: 1.0884 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.386 | LR: 1.31e-05 [2026-04-17 12:20:09] Epoch 2 | Step 15520 | Loss: 0.2402 | LM: 0.2281 | LB: 1.0884 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.386 | LR: 1.29e-05 [2026-04-17 12:20:15] Epoch 2 | Step 15530 | Loss: 0.2402 | LM: 0.2280 | LB: 1.0884 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.386 | LR: 1.28e-05 [2026-04-17 12:20:21] Epoch 2 | Step 15540 | Loss: 0.2401 | LM: 0.2280 | LB: 1.0884 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.386 | LR: 1.26e-05 [2026-04-17 12:20:28] Epoch 2 | Step 15550 | Loss: 0.2401 | LM: 0.2280 | LB: 1.0884 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.386 | LR: 1.25e-05 [2026-04-17 12:20:34] Epoch 2 | Step 15560 | Loss: 0.2401 | LM: 0.2278 | LB: 1.0884 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.386 | LR: 1.23e-05 [2026-04-17 12:20:41] Epoch 2 | Step 15570 | Loss: 0.2401 | LM: 0.2278 | LB: 1.0884 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.386 | LR: 1.22e-05 [2026-04-17 12:20:47] Epoch 2 | Step 15580 | Loss: 0.2402 | LM: 0.2280 | LB: 1.0884 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.386 | LR: 1.21e-05 [2026-04-17 12:20:54] Epoch 2 | Step 15590 | Loss: 0.2402 | LM: 0.2279 | LB: 1.0884 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.386 | LR: 1.19e-05 [2026-04-17 12:21:00] Epoch 2 | Step 15600 | Loss: 0.2402 | LM: 0.2279 | LB: 1.0884 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.386 | LR: 1.18e-05 [2026-04-17 12:21:07] Epoch 2 | Step 15610 | Loss: 0.2401 | LM: 0.2277 | LB: 1.0884 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.386 | LR: 1.17e-05 [2026-04-17 12:21:13] Epoch 2 | Step 15620 | Loss: 0.2401 | LM: 0.2276 | LB: 1.0883 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.386 | LR: 1.16e-05 [2026-04-17 12:21:20] Epoch 2 | Step 15630 | Loss: 0.2401 | LM: 0.2276 | LB: 1.0883 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.386 | LR: 1.15e-05 [2026-04-17 12:21:26] Epoch 2 | Step 15640 | Loss: 0.2401 | LM: 0.2276 | LB: 1.0883 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.386 | LR: 1.13e-05 [2026-04-17 12:21:33] Epoch 2 | Step 15650 | Loss: 0.2401 | LM: 0.2275 | LB: 1.0883 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.386 | LR: 1.12e-05 [2026-04-17 12:21:39] Epoch 2 | Step 15660 | Loss: 0.2401 | LM: 0.2274 | LB: 1.0884 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.386 | LR: 1.11e-05 [2026-04-17 12:21:46] Epoch 2 | Step 15670 | Loss: 0.2401 | LM: 0.2272 | LB: 1.0884 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.386 | LR: 1.10e-05 [2026-04-17 12:21:52] Epoch 2 | Step 15680 | Loss: 0.2401 | LM: 0.2274 | LB: 1.0884 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.386 | LR: 1.09e-05 [2026-04-17 12:21:59] Epoch 2 | Step 15690 | Loss: 0.2400 | LM: 0.2272 | LB: 1.0883 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.386 | LR: 1.09e-05 [2026-04-17 12:22:05] Epoch 2 | Step 15700 | Loss: 0.2400 | LM: 0.2272 | LB: 1.0883 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.386 | LR: 1.08e-05 [2026-04-17 12:22:12] Epoch 2 | Step 15710 | Loss: 0.2400 | LM: 0.2272 | LB: 1.0883 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.386 | LR: 1.07e-05 [2026-04-17 12:22:18] Epoch 2 | Step 15720 | Loss: 0.2400 | LM: 0.2272 | LB: 1.0883 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.386 | LR: 1.06e-05 [2026-04-17 12:22:25] Epoch 2 | Step 15730 | Loss: 0.2401 | LM: 0.2272 | LB: 1.0884 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.386 | LR: 1.05e-05 [2026-04-17 12:22:31] Epoch 2 | Step 15740 | Loss: 0.2402 | LM: 0.2274 | LB: 1.0883 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.386 | LR: 1.05e-05 [2026-04-17 12:22:37] Epoch 2 | Step 15750 | Loss: 0.2402 | LM: 0.2273 | LB: 1.0883 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.386 | LR: 1.04e-05 [2026-04-17 12:22:44] Epoch 2 | Step 15760 | Loss: 0.2401 | LM: 0.2272 | LB: 1.0883 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.386 | LR: 1.04e-05 [2026-04-17 12:22:50] Epoch 2 | Step 15770 | Loss: 0.2401 | LM: 0.2273 | LB: 1.0883 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.386 | LR: 1.03e-05 [2026-04-17 12:22:57] Epoch 2 | Step 15780 | Loss: 0.2401 | LM: 0.2272 | LB: 1.0883 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.386 | LR: 1.03e-05 [2026-04-17 12:23:03] Epoch 2 | Step 15790 | Loss: 0.2401 | LM: 0.2273 | LB: 1.0883 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.386 | LR: 1.02e-05 [2026-04-17 12:23:10] Epoch 2 | Step 15800 | Loss: 0.2401 | LM: 0.2273 | LB: 1.0883 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.386 | LR: 1.02e-05 [2026-04-17 12:23:16] Epoch 2 | Step 15810 | Loss: 0.2401 | LM: 0.2273 | LB: 1.0883 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.386 | LR: 1.01e-05 [2026-04-17 12:23:22] Epoch 2 | Step 15820 | Loss: 0.2401 | LM: 0.2273 | LB: 1.0883 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.386 | LR: 1.01e-05 [2026-04-17 12:23:29] Epoch 2 | Step 15830 | Loss: 0.2400 | LM: 0.2272 | LB: 1.0883 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.386 | LR: 1.01e-05 [2026-04-17 12:23:35] Epoch 2 | Step 15840 | Loss: 0.2401 | LM: 0.2271 | LB: 1.0883 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.386 | LR: 1.00e-05 [2026-04-17 12:23:41] Epoch 2 | Step 15850 | Loss: 0.2400 | LM: 0.2272 | LB: 1.0883 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.386 | LR: 1.00e-05 [2026-04-17 12:23:48] Epoch 2 | Step 15860 | Loss: 0.2400 | LM: 0.2271 | LB: 1.0883 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.386 | LR: 1.00e-05 [2026-04-17 12:23:54] Epoch 2 | Step 15870 | Loss: 0.2400 | LM: 0.2272 | LB: 1.0882 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.386 | LR: 1.00e-05 [2026-04-17 12:24:00] Epoch 2 | Step 15880 | Loss: 0.2400 | LM: 0.2272 | LB: 1.0882 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.386 | LR: 1.00e-05 [2026-04-17 12:24:07] Epoch 2 | Step 15890 | Loss: 0.2400 | LM: 0.2273 | LB: 1.0882 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.386 | LR: 1.00e-05 [2026-04-17 12:24:13] Epoch 2 | Step 15900 | Loss: 0.2400 | LM: 0.2273 | LB: 1.0882 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.386 | LR: 1.00e-05 [2026-04-17 12:24:20] Epoch 2 | Step 15910 | Loss: 0.2399 | LM: 0.2272 | LB: 1.0882 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.386 | LR: 1.00e-05 [2026-04-17 12:24:26] Epoch 2 | Step 15920 | Loss: 0.2399 | LM: 0.2274 | LB: 1.0882 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.386 | LR: 1.00e-05 [2026-04-17 12:24:33] Epoch 2 | Step 15930 | Loss: 0.2399 | LM: 0.2274 | LB: 1.0882 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.386 | LR: 1.00e-05 [2026-04-17 12:24:39] Epoch 2 | Step 15940 | Loss: 0.2399 | LM: 0.2274 | LB: 1.0882 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.386 | LR: 1.00e-05 [2026-04-17 12:24:45] Epoch 2 | Step 15950 | Loss: 0.2399 | LM: 0.2273 | LB: 1.0882 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.386 | LR: 1.00e-05 [2026-04-17 12:24:52] Epoch 2 | Step 15960 | Loss: 0.2398 | LM: 0.2273 | LB: 1.0883 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.386 | LR: 1.00e-05 [2026-04-17 12:24:58] Epoch 2 | Step 15970 | Loss: 0.2399 | LM: 0.2272 | LB: 1.0882 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.386 | LR: 1.00e-05 [2026-04-17 12:25:04] Epoch 2 | Step 15980 | Loss: 0.2399 | LM: 0.2273 | LB: 1.0882 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 12:25:11] Epoch 2 | Step 15990 | Loss: 0.2399 | LM: 0.2275 | LB: 1.0882 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 12:25:17] Epoch 2 | Step 16000 | Loss: 0.2399 | LM: 0.2275 | LB: 1.0882 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 12:25:18] Validation | Batch 10/784 | Loss: 0.3263 | LM_LOSS: 0.3154 | LB_LOSS: 1.0842 [2026-04-17 12:25:20] Validation | Batch 20/784 | Loss: 0.3371 | LM_LOSS: 0.3262 | LB_LOSS: 1.0845 [2026-04-17 12:25:21] Validation | Batch 30/784 | Loss: 0.3227 | LM_LOSS: 0.3119 | LB_LOSS: 1.0837 [2026-04-17 12:25:23] Validation | Batch 40/784 | Loss: 0.3250 | LM_LOSS: 0.3142 | LB_LOSS: 1.0837 [2026-04-17 12:25:24] Validation | Batch 50/784 | Loss: 0.3221 | LM_LOSS: 0.3113 | LB_LOSS: 1.0830 [2026-04-17 12:25:25] Validation | Batch 60/784 | Loss: 0.3239 | LM_LOSS: 0.3131 | LB_LOSS: 1.0826 [2026-04-17 12:25:27] Validation | Batch 70/784 | Loss: 0.3212 | LM_LOSS: 0.3103 | LB_LOSS: 1.0819 [2026-04-17 12:25:28] Validation | Batch 80/784 | Loss: 0.3173 | LM_LOSS: 0.3065 | LB_LOSS: 1.0815 [2026-04-17 12:25:29] Validation | Batch 90/784 | Loss: 0.3162 | LM_LOSS: 0.3053 | LB_LOSS: 1.0820 [2026-04-17 12:25:30] Validation | Batch 100/784 | Loss: 0.3180 | LM_LOSS: 0.3071 | LB_LOSS: 1.0825 [2026-04-17 12:25:32] Validation | Batch 110/784 | Loss: 0.3128 | LM_LOSS: 0.3019 | LB_LOSS: 1.0826 [2026-04-17 12:25:33] Validation | Batch 120/784 | Loss: 0.3163 | LM_LOSS: 0.3055 | LB_LOSS: 1.0825 [2026-04-17 12:25:35] Validation | Batch 130/784 | Loss: 0.3193 | LM_LOSS: 0.3085 | LB_LOSS: 1.0825 [2026-04-17 12:25:36] Validation | Batch 140/784 | Loss: 0.3186 | LM_LOSS: 0.3078 | LB_LOSS: 1.0823 [2026-04-17 12:25:37] Validation | Batch 150/784 | Loss: 0.3147 | LM_LOSS: 0.3039 | LB_LOSS: 1.0826 [2026-04-17 12:25:39] Validation | Batch 160/784 | Loss: 0.3154 | LM_LOSS: 0.3046 | LB_LOSS: 1.0823 [2026-04-17 12:25:41] Validation | Batch 170/784 | Loss: 0.3157 | LM_LOSS: 0.3048 | LB_LOSS: 1.0820 [2026-04-17 12:25:42] Validation | Batch 180/784 | Loss: 0.3133 | LM_LOSS: 0.3025 | LB_LOSS: 1.0820 [2026-04-17 12:25:43] Validation | Batch 190/784 | Loss: 0.3153 | LM_LOSS: 0.3045 | LB_LOSS: 1.0824 [2026-04-17 12:25:44] Validation | Batch 200/784 | Loss: 0.3157 | LM_LOSS: 0.3049 | LB_LOSS: 1.0825 [2026-04-17 12:25:46] Validation | Batch 210/784 | Loss: 0.3146 | LM_LOSS: 0.3038 | LB_LOSS: 1.0824 [2026-04-17 12:25:47] Validation | Batch 220/784 | Loss: 0.3154 | LM_LOSS: 0.3046 | LB_LOSS: 1.0824 [2026-04-17 12:25:49] Validation | Batch 230/784 | Loss: 0.3159 | LM_LOSS: 0.3051 | LB_LOSS: 1.0823 [2026-04-17 12:25:50] Validation | Batch 240/784 | Loss: 0.3163 | LM_LOSS: 0.3055 | LB_LOSS: 1.0827 [2026-04-17 12:25:51] Validation | Batch 250/784 | Loss: 0.3162 | LM_LOSS: 0.3054 | LB_LOSS: 1.0825 [2026-04-17 12:25:53] Validation | Batch 260/784 | Loss: 0.3164 | LM_LOSS: 0.3056 | LB_LOSS: 1.0827 [2026-04-17 12:25:55] Validation | Batch 270/784 | Loss: 0.3162 | LM_LOSS: 0.3054 | LB_LOSS: 1.0828 [2026-04-17 12:25:56] Validation | Batch 280/784 | Loss: 0.3166 | LM_LOSS: 0.3058 | LB_LOSS: 1.0830 [2026-04-17 12:25:57] Validation | Batch 290/784 | Loss: 0.3177 | LM_LOSS: 0.3069 | LB_LOSS: 1.0831 [2026-04-17 12:25:58] Validation | Batch 300/784 | Loss: 0.3184 | LM_LOSS: 0.3076 | LB_LOSS: 1.0831 [2026-04-17 12:26:00] Validation | Batch 310/784 | Loss: 0.3179 | LM_LOSS: 0.3071 | LB_LOSS: 1.0831 [2026-04-17 12:26:01] Validation | Batch 320/784 | Loss: 0.3195 | LM_LOSS: 0.3086 | LB_LOSS: 1.0831 [2026-04-17 12:26:03] Validation | Batch 330/784 | Loss: 0.3193 | LM_LOSS: 0.3084 | LB_LOSS: 1.0830 [2026-04-17 12:26:04] Validation | Batch 340/784 | Loss: 0.3181 | LM_LOSS: 0.3073 | LB_LOSS: 1.0831 [2026-04-17 12:26:05] Validation | Batch 350/784 | Loss: 0.3183 | LM_LOSS: 0.3074 | LB_LOSS: 1.0833 [2026-04-17 12:26:06] Validation | Batch 360/784 | Loss: 0.3180 | LM_LOSS: 0.3072 | LB_LOSS: 1.0833 [2026-04-17 12:26:08] Validation | Batch 370/784 | Loss: 0.3185 | LM_LOSS: 0.3077 | LB_LOSS: 1.0833 [2026-04-17 12:26:09] Validation | Batch 380/784 | Loss: 0.3183 | LM_LOSS: 0.3075 | LB_LOSS: 1.0833 [2026-04-17 12:26:10] Validation | Batch 390/784 | Loss: 0.3182 | LM_LOSS: 0.3074 | LB_LOSS: 1.0834 [2026-04-17 12:26:12] Validation | Batch 400/784 | Loss: 0.3185 | LM_LOSS: 0.3076 | LB_LOSS: 1.0834 [2026-04-17 12:26:13] Validation | Batch 410/784 | Loss: 0.3188 | LM_LOSS: 0.3079 | LB_LOSS: 1.0834 [2026-04-17 12:26:14] Validation | Batch 420/784 | Loss: 0.3190 | LM_LOSS: 0.3082 | LB_LOSS: 1.0834 [2026-04-17 12:26:15] Validation | Batch 430/784 | Loss: 0.3191 | LM_LOSS: 0.3082 | LB_LOSS: 1.0834 [2026-04-17 12:26:17] Validation | Batch 440/784 | Loss: 0.3187 | LM_LOSS: 0.3079 | LB_LOSS: 1.0834 [2026-04-17 12:26:18] Validation | Batch 450/784 | Loss: 0.3180 | LM_LOSS: 0.3072 | LB_LOSS: 1.0834 [2026-04-17 12:26:19] Validation | Batch 460/784 | Loss: 0.3185 | LM_LOSS: 0.3076 | LB_LOSS: 1.0834 [2026-04-17 12:26:21] Validation | Batch 470/784 | Loss: 0.3177 | LM_LOSS: 0.3068 | LB_LOSS: 1.0834 [2026-04-17 12:26:22] Validation | Batch 480/784 | Loss: 0.3181 | LM_LOSS: 0.3073 | LB_LOSS: 1.0834 [2026-04-17 12:26:23] Validation | Batch 490/784 | Loss: 0.3175 | LM_LOSS: 0.3066 | LB_LOSS: 1.0833 [2026-04-17 12:26:25] Validation | Batch 500/784 | Loss: 0.3179 | LM_LOSS: 0.3070 | LB_LOSS: 1.0832 [2026-04-17 12:26:26] Validation | Batch 510/784 | Loss: 0.3176 | LM_LOSS: 0.3067 | LB_LOSS: 1.0832 [2026-04-17 12:26:27] Validation | Batch 520/784 | Loss: 0.3177 | LM_LOSS: 0.3069 | LB_LOSS: 1.0831 [2026-04-17 12:26:29] Validation | Batch 530/784 | Loss: 0.3186 | LM_LOSS: 0.3078 | LB_LOSS: 1.0831 [2026-04-17 12:26:30] Validation | Batch 540/784 | Loss: 0.3189 | LM_LOSS: 0.3081 | LB_LOSS: 1.0831 [2026-04-17 12:26:32] Validation | Batch 550/784 | Loss: 0.3202 | LM_LOSS: 0.3094 | LB_LOSS: 1.0831 [2026-04-17 12:26:33] Validation | Batch 560/784 | Loss: 0.3203 | LM_LOSS: 0.3095 | LB_LOSS: 1.0831 [2026-04-17 12:26:34] Validation | Batch 570/784 | Loss: 0.3198 | LM_LOSS: 0.3090 | LB_LOSS: 1.0830 [2026-04-17 12:26:36] Validation | Batch 580/784 | Loss: 0.3193 | LM_LOSS: 0.3085 | LB_LOSS: 1.0831 [2026-04-17 12:26:37] Validation | Batch 590/784 | Loss: 0.3196 | LM_LOSS: 0.3087 | LB_LOSS: 1.0830 [2026-04-17 12:26:38] Validation | Batch 600/784 | Loss: 0.3194 | LM_LOSS: 0.3086 | LB_LOSS: 1.0830 [2026-04-17 12:26:40] Validation | Batch 610/784 | Loss: 0.3196 | LM_LOSS: 0.3087 | LB_LOSS: 1.0830 [2026-04-17 12:26:41] Validation | Batch 620/784 | Loss: 0.3194 | LM_LOSS: 0.3086 | LB_LOSS: 1.0830 [2026-04-17 12:26:43] Validation | Batch 630/784 | Loss: 0.3202 | LM_LOSS: 0.3093 | LB_LOSS: 1.0830 [2026-04-17 12:26:44] Validation | Batch 640/784 | Loss: 0.3202 | LM_LOSS: 0.3094 | LB_LOSS: 1.0830 [2026-04-17 12:26:46] Validation | Batch 650/784 | Loss: 0.3201 | LM_LOSS: 0.3092 | LB_LOSS: 1.0830 [2026-04-17 12:26:47] Validation | Batch 660/784 | Loss: 0.3204 | LM_LOSS: 0.3096 | LB_LOSS: 1.0830 [2026-04-17 12:26:49] Validation | Batch 670/784 | Loss: 0.3208 | LM_LOSS: 0.3100 | LB_LOSS: 1.0831 [2026-04-17 12:26:50] Validation | Batch 680/784 | Loss: 0.3205 | LM_LOSS: 0.3097 | LB_LOSS: 1.0831 [2026-04-17 12:26:51] Validation | Batch 690/784 | Loss: 0.3207 | LM_LOSS: 0.3099 | LB_LOSS: 1.0830 [2026-04-17 12:26:53] Validation | Batch 700/784 | Loss: 0.3208 | LM_LOSS: 0.3099 | LB_LOSS: 1.0830 [2026-04-17 12:26:54] Validation | Batch 710/784 | Loss: 0.3205 | LM_LOSS: 0.3097 | LB_LOSS: 1.0829 [2026-04-17 12:26:56] Validation | Batch 720/784 | Loss: 0.3203 | LM_LOSS: 0.3094 | LB_LOSS: 1.0828 [2026-04-17 12:26:57] Validation | Batch 730/784 | Loss: 0.3198 | LM_LOSS: 0.3089 | LB_LOSS: 1.0828 [2026-04-17 12:26:58] Validation | Batch 740/784 | Loss: 0.3198 | LM_LOSS: 0.3090 | LB_LOSS: 1.0829 [2026-04-17 12:26:59] Validation | Batch 750/784 | Loss: 0.3192 | LM_LOSS: 0.3083 | LB_LOSS: 1.0829 [2026-04-17 12:27:01] Validation | Batch 760/784 | Loss: 0.3193 | LM_LOSS: 0.3085 | LB_LOSS: 1.0829 [2026-04-17 12:27:02] Validation | Batch 770/784 | Loss: 0.3195 | LM_LOSS: 0.3087 | LB_LOSS: 1.0829 [2026-04-17 12:27:03] Validation | Batch 780/784 | Loss: 0.3198 | LM_LOSS: 0.3090 | LB_LOSS: 1.0829 [2026-04-17 12:27:04] Validation | Batch 784/784 | Loss: 0.3200 | LM_LOSS: 0.3092 | LB_LOSS: 1.0829 [2026-04-17 12:27:06] Validation | Loss: 0.3200 | LM_LOSS: 0.3092 | LB_LOSS: 1.0829 | PPL: 1.36 | Time: 106.74s [2026-04-17 12:27:13] Epoch 2 | Step 16010 | Loss: 0.2400 | LM: 0.2277 | LB: 1.0882 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 12:27:19] Epoch 2 | Step 16020 | Loss: 0.2400 | LM: 0.2276 | LB: 1.0882 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.386 | LR: 1.00e-05 [2026-04-17 12:27:25] Epoch 2 | Step 16030 | Loss: 0.2399 | LM: 0.2276 | LB: 1.0883 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.386 | LR: 1.00e-05 [2026-04-17 12:27:31] Epoch 2 | Step 16040 | Loss: 0.2400 | LM: 0.2276 | LB: 1.0883 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.386 | LR: 1.00e-05 [2026-04-17 12:27:38] Epoch 2 | Step 16050 | Loss: 0.2399 | LM: 0.2276 | LB: 1.0883 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.386 | LR: 1.00e-05 [2026-04-17 12:27:44] Epoch 2 | Step 16060 | Loss: 0.2399 | LM: 0.2275 | LB: 1.0883 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.386 | LR: 1.00e-05 [2026-04-17 12:27:51] Epoch 2 | Step 16070 | Loss: 0.2399 | LM: 0.2275 | LB: 1.0883 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.386 | LR: 1.00e-05 [2026-04-17 12:27:57] Epoch 2 | Step 16080 | Loss: 0.2399 | LM: 0.2275 | LB: 1.0883 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.386 | LR: 1.00e-05 [2026-04-17 12:28:04] Epoch 2 | Step 16090 | Loss: 0.2399 | LM: 0.2274 | LB: 1.0883 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.386 | LR: 1.00e-05 [2026-04-17 12:28:10] Epoch 2 | Step 16100 | Loss: 0.2399 | LM: 0.2274 | LB: 1.0883 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.386 | LR: 1.00e-05 [2026-04-17 12:28:17] Epoch 2 | Step 16110 | Loss: 0.2399 | LM: 0.2275 | LB: 1.0884 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.386 | LR: 1.00e-05 [2026-04-17 12:28:23] Epoch 2 | Step 16120 | Loss: 0.2398 | LM: 0.2274 | LB: 1.0883 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.386 | LR: 1.00e-05 [2026-04-17 12:28:29] Epoch 2 | Step 16130 | Loss: 0.2398 | LM: 0.2274 | LB: 1.0883 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.386 | LR: 1.00e-05 [2026-04-17 12:28:36] Epoch 2 | Step 16140 | Loss: 0.2398 | LM: 0.2274 | LB: 1.0883 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.386 | LR: 1.00e-05 [2026-04-17 12:28:42] Epoch 2 | Step 16150 | Loss: 0.2398 | LM: 0.2274 | LB: 1.0883 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.386 | LR: 1.00e-05 [2026-04-17 12:28:48] Epoch 2 | Step 16160 | Loss: 0.2398 | LM: 0.2274 | LB: 1.0883 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.386 | LR: 1.00e-05 [2026-04-17 12:28:55] Epoch 2 | Step 16170 | Loss: 0.2397 | LM: 0.2273 | LB: 1.0883 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.386 | LR: 1.00e-05 [2026-04-17 12:29:01] Epoch 2 | Step 16180 | Loss: 0.2397 | LM: 0.2273 | LB: 1.0883 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.386 | LR: 1.00e-05 [2026-04-17 12:29:07] Epoch 2 | Step 16190 | Loss: 0.2397 | LM: 0.2273 | LB: 1.0883 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.386 | LR: 1.00e-05 [2026-04-17 12:29:13] Epoch 2 | Step 16200 | Loss: 0.2397 | LM: 0.2273 | LB: 1.0883 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.386 | LR: 1.00e-05 [2026-04-17 12:29:19] Epoch 2 | Step 16210 | Loss: 0.2397 | LM: 0.2272 | LB: 1.0882 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.386 | LR: 1.00e-05 [2026-04-17 12:29:26] Epoch 2 | Step 16220 | Loss: 0.2397 | LM: 0.2272 | LB: 1.0882 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 12:29:32] Epoch 2 | Step 16230 | Loss: 0.2397 | LM: 0.2273 | LB: 1.0882 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 12:29:38] Epoch 2 | Step 16240 | Loss: 0.2397 | LM: 0.2272 | LB: 1.0882 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 12:29:45] Epoch 2 | Step 16250 | Loss: 0.2397 | LM: 0.2273 | LB: 1.0882 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 12:29:51] Epoch 2 | Step 16260 | Loss: 0.2396 | LM: 0.2272 | LB: 1.0882 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 12:29:58] Epoch 2 | Step 16270 | Loss: 0.2397 | LM: 0.2273 | LB: 1.0882 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.386 | LR: 1.00e-05 [2026-04-17 12:30:04] Epoch 2 | Step 16280 | Loss: 0.2396 | LM: 0.2272 | LB: 1.0882 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.386 | LR: 1.00e-05 [2026-04-17 12:30:10] Epoch 2 | Step 16290 | Loss: 0.2396 | LM: 0.2271 | LB: 1.0882 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 12:30:17] Epoch 2 | Step 16300 | Loss: 0.2396 | LM: 0.2273 | LB: 1.0882 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 12:30:23] Epoch 2 | Step 16310 | Loss: 0.2397 | LM: 0.2272 | LB: 1.0882 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 12:30:30] Epoch 2 | Step 16320 | Loss: 0.2397 | LM: 0.2273 | LB: 1.0882 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 12:30:36] Epoch 2 | Step 16330 | Loss: 0.2397 | LM: 0.2272 | LB: 1.0882 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 12:30:42] Epoch 2 | Step 16340 | Loss: 0.2397 | LM: 0.2272 | LB: 1.0882 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 12:30:49] Epoch 2 | Step 16350 | Loss: 0.2398 | LM: 0.2272 | LB: 1.0882 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 12:30:56] Epoch 2 | Step 16360 | Loss: 0.2398 | LM: 0.2271 | LB: 1.0882 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 12:31:02] Epoch 2 | Step 16370 | Loss: 0.2397 | LM: 0.2270 | LB: 1.0882 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 12:31:08] Epoch 2 | Step 16380 | Loss: 0.2397 | LM: 0.2269 | LB: 1.0882 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 12:31:15] Epoch 2 | Step 16390 | Loss: 0.2397 | LM: 0.2268 | LB: 1.0882 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 12:31:21] Epoch 2 | Step 16400 | Loss: 0.2397 | LM: 0.2269 | LB: 1.0882 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 12:31:27] Epoch 2 | Step 16410 | Loss: 0.2397 | LM: 0.2268 | LB: 1.0882 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 12:31:34] Epoch 2 | Step 16420 | Loss: 0.2397 | LM: 0.2268 | LB: 1.0882 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 12:31:40] Epoch 2 | Step 16430 | Loss: 0.2397 | LM: 0.2267 | LB: 1.0882 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 12:31:46] Epoch 2 | Step 16440 | Loss: 0.2397 | LM: 0.2267 | LB: 1.0882 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 12:31:53] Epoch 2 | Step 16450 | Loss: 0.2396 | LM: 0.2267 | LB: 1.0882 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 12:31:59] Epoch 2 | Step 16460 | Loss: 0.2396 | LM: 0.2268 | LB: 1.0882 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 12:32:06] Epoch 2 | Step 16470 | Loss: 0.2396 | LM: 0.2269 | LB: 1.0882 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 12:32:12] Epoch 2 | Step 16480 | Loss: 0.2397 | LM: 0.2268 | LB: 1.0882 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 12:32:19] Epoch 2 | Step 16490 | Loss: 0.2397 | LM: 0.2269 | LB: 1.0882 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 12:32:25] Epoch 2 | Step 16500 | Loss: 0.2397 | LM: 0.2270 | LB: 1.0882 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 12:32:31] Epoch 2 | Step 16510 | Loss: 0.2397 | LM: 0.2270 | LB: 1.0882 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 12:32:38] Epoch 2 | Step 16520 | Loss: 0.2397 | LM: 0.2269 | LB: 1.0881 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 12:32:44] Epoch 2 | Step 16530 | Loss: 0.2397 | LM: 0.2269 | LB: 1.0882 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 12:32:50] Epoch 2 | Step 16540 | Loss: 0.2398 | LM: 0.2270 | LB: 1.0881 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 12:32:57] Epoch 2 | Step 16550 | Loss: 0.2397 | LM: 0.2268 | LB: 1.0882 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 12:33:03] Epoch 2 | Step 16560 | Loss: 0.2397 | LM: 0.2267 | LB: 1.0882 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 12:33:10] Epoch 2 | Step 16570 | Loss: 0.2398 | LM: 0.2268 | LB: 1.0881 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 12:33:16] Epoch 2 | Step 16580 | Loss: 0.2398 | LM: 0.2269 | LB: 1.0881 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 12:33:23] Epoch 2 | Step 16590 | Loss: 0.2399 | LM: 0.2269 | LB: 1.0881 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 12:33:29] Epoch 2 | Step 16600 | Loss: 0.2399 | LM: 0.2269 | LB: 1.0881 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 12:33:36] Epoch 2 | Step 16610 | Loss: 0.2398 | LM: 0.2267 | LB: 1.0881 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 12:33:42] Epoch 2 | Step 16620 | Loss: 0.2398 | LM: 0.2267 | LB: 1.0881 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 12:33:49] Epoch 2 | Step 16630 | Loss: 0.2398 | LM: 0.2268 | LB: 1.0881 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 12:33:55] Epoch 2 | Step 16640 | Loss: 0.2398 | LM: 0.2268 | LB: 1.0881 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 12:34:01] Epoch 2 | Step 16650 | Loss: 0.2398 | LM: 0.2268 | LB: 1.0880 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 12:34:08] Epoch 2 | Step 16660 | Loss: 0.2399 | LM: 0.2269 | LB: 1.0881 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 12:34:14] Epoch 2 | Step 16670 | Loss: 0.2398 | LM: 0.2268 | LB: 1.0880 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 12:34:20] Epoch 2 | Step 16680 | Loss: 0.2399 | LM: 0.2269 | LB: 1.0880 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 12:34:27] Epoch 2 | Step 16690 | Loss: 0.2399 | LM: 0.2270 | LB: 1.0880 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 12:34:33] Epoch 2 | Step 16700 | Loss: 0.2400 | LM: 0.2270 | LB: 1.0880 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 12:34:39] Epoch 2 | Step 16710 | Loss: 0.2399 | LM: 0.2270 | LB: 1.0880 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 12:34:46] Epoch 2 | Step 16720 | Loss: 0.2399 | LM: 0.2269 | LB: 1.0880 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 12:34:52] Epoch 2 | Step 16730 | Loss: 0.2399 | LM: 0.2270 | LB: 1.0880 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 12:34:59] Epoch 2 | Step 16740 | Loss: 0.2398 | LM: 0.2269 | LB: 1.0880 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 12:35:05] Epoch 2 | Step 16750 | Loss: 0.2399 | LM: 0.2269 | LB: 1.0880 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 12:35:11] Epoch 2 | Step 16760 | Loss: 0.2399 | LM: 0.2269 | LB: 1.0880 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 12:35:18] Epoch 2 | Step 16770 | Loss: 0.2399 | LM: 0.2268 | LB: 1.0880 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 12:35:24] Epoch 2 | Step 16780 | Loss: 0.2399 | LM: 0.2268 | LB: 1.0880 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 12:35:30] Epoch 2 | Step 16790 | Loss: 0.2399 | LM: 0.2269 | LB: 1.0880 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 12:35:37] Epoch 2 | Step 16800 | Loss: 0.2398 | LM: 0.2269 | LB: 1.0880 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 12:35:43] Epoch 2 | Step 16810 | Loss: 0.2398 | LM: 0.2269 | LB: 1.0880 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 12:35:50] Epoch 2 | Step 16820 | Loss: 0.2398 | LM: 0.2269 | LB: 1.0880 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 12:35:56] Epoch 2 | Step 16830 | Loss: 0.2398 | LM: 0.2268 | LB: 1.0880 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 12:36:02] Epoch 2 | Step 16840 | Loss: 0.2398 | LM: 0.2269 | LB: 1.0880 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 12:36:09] Epoch 2 | Step 16850 | Loss: 0.2399 | LM: 0.2269 | LB: 1.0880 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 12:36:15] Epoch 2 | Step 16860 | Loss: 0.2399 | LM: 0.2270 | LB: 1.0880 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 12:36:22] Epoch 2 | Step 16870 | Loss: 0.2399 | LM: 0.2269 | LB: 1.0880 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 12:36:28] Epoch 2 | Step 16880 | Loss: 0.2399 | LM: 0.2268 | LB: 1.0880 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 12:36:34] Epoch 2 | Step 16890 | Loss: 0.2398 | LM: 0.2268 | LB: 1.0879 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 12:36:41] Epoch 2 | Step 16900 | Loss: 0.2398 | LM: 0.2268 | LB: 1.0880 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 12:36:47] Epoch 2 | Step 16910 | Loss: 0.2398 | LM: 0.2268 | LB: 1.0880 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 12:36:53] Epoch 2 | Step 16920 | Loss: 0.2398 | LM: 0.2268 | LB: 1.0880 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 12:37:00] Epoch 2 | Step 16930 | Loss: 0.2398 | LM: 0.2268 | LB: 1.0879 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 12:37:06] Epoch 2 | Step 16940 | Loss: 0.2398 | LM: 0.2269 | LB: 1.0879 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 12:37:13] Epoch 2 | Step 16950 | Loss: 0.2398 | LM: 0.2270 | LB: 1.0879 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 12:37:19] Epoch 2 | Step 16960 | Loss: 0.2398 | LM: 0.2270 | LB: 1.0879 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 12:37:25] Epoch 2 | Step 16970 | Loss: 0.2398 | LM: 0.2271 | LB: 1.0879 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 12:37:32] Epoch 2 | Step 16980 | Loss: 0.2398 | LM: 0.2271 | LB: 1.0879 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 12:37:38] Epoch 2 | Step 16990 | Loss: 0.2398 | LM: 0.2271 | LB: 1.0879 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 12:37:44] Epoch 2 | Step 17000 | Loss: 0.2399 | LM: 0.2272 | LB: 1.0879 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 12:37:45] Validation | Batch 10/784 | Loss: 0.3263 | LM_LOSS: 0.3155 | LB_LOSS: 1.0843 [2026-04-17 12:37:47] Validation | Batch 20/784 | Loss: 0.3369 | LM_LOSS: 0.3261 | LB_LOSS: 1.0845 [2026-04-17 12:37:48] Validation | Batch 30/784 | Loss: 0.3225 | LM_LOSS: 0.3117 | LB_LOSS: 1.0838 [2026-04-17 12:37:50] Validation | Batch 40/784 | Loss: 0.3248 | LM_LOSS: 0.3139 | LB_LOSS: 1.0837 [2026-04-17 12:37:51] Validation | Batch 50/784 | Loss: 0.3218 | LM_LOSS: 0.3110 | LB_LOSS: 1.0831 [2026-04-17 12:37:52] Validation | Batch 60/784 | Loss: 0.3236 | LM_LOSS: 0.3128 | LB_LOSS: 1.0826 [2026-04-17 12:37:54] Validation | Batch 70/784 | Loss: 0.3208 | LM_LOSS: 0.3100 | LB_LOSS: 1.0820 [2026-04-17 12:37:55] Validation | Batch 80/784 | Loss: 0.3171 | LM_LOSS: 0.3063 | LB_LOSS: 1.0815 [2026-04-17 12:37:56] Validation | Batch 90/784 | Loss: 0.3159 | LM_LOSS: 0.3051 | LB_LOSS: 1.0821 [2026-04-17 12:37:58] Validation | Batch 100/784 | Loss: 0.3177 | LM_LOSS: 0.3069 | LB_LOSS: 1.0825 [2026-04-17 12:37:59] Validation | Batch 110/784 | Loss: 0.3126 | LM_LOSS: 0.3017 | LB_LOSS: 1.0827 [2026-04-17 12:38:00] Validation | Batch 120/784 | Loss: 0.3161 | LM_LOSS: 0.3053 | LB_LOSS: 1.0826 [2026-04-17 12:38:02] Validation | Batch 130/784 | Loss: 0.3191 | LM_LOSS: 0.3082 | LB_LOSS: 1.0825 [2026-04-17 12:38:03] Validation | Batch 140/784 | Loss: 0.3184 | LM_LOSS: 0.3075 | LB_LOSS: 1.0824 [2026-04-17 12:38:05] Validation | Batch 150/784 | Loss: 0.3144 | LM_LOSS: 0.3036 | LB_LOSS: 1.0827 [2026-04-17 12:38:06] Validation | Batch 160/784 | Loss: 0.3152 | LM_LOSS: 0.3044 | LB_LOSS: 1.0823 [2026-04-17 12:38:08] Validation | Batch 170/784 | Loss: 0.3154 | LM_LOSS: 0.3046 | LB_LOSS: 1.0821 [2026-04-17 12:38:09] Validation | Batch 180/784 | Loss: 0.3130 | LM_LOSS: 0.3022 | LB_LOSS: 1.0821 [2026-04-17 12:38:10] Validation | Batch 190/784 | Loss: 0.3150 | LM_LOSS: 0.3042 | LB_LOSS: 1.0825 [2026-04-17 12:38:11] Validation | Batch 200/784 | Loss: 0.3155 | LM_LOSS: 0.3046 | LB_LOSS: 1.0826 [2026-04-17 12:38:13] Validation | Batch 210/784 | Loss: 0.3143 | LM_LOSS: 0.3035 | LB_LOSS: 1.0825 [2026-04-17 12:38:14] Validation | Batch 220/784 | Loss: 0.3151 | LM_LOSS: 0.3043 | LB_LOSS: 1.0825 [2026-04-17 12:38:16] Validation | Batch 230/784 | Loss: 0.3157 | LM_LOSS: 0.3048 | LB_LOSS: 1.0824 [2026-04-17 12:38:17] Validation | Batch 240/784 | Loss: 0.3160 | LM_LOSS: 0.3052 | LB_LOSS: 1.0828 [2026-04-17 12:38:18] Validation | Batch 250/784 | Loss: 0.3159 | LM_LOSS: 0.3051 | LB_LOSS: 1.0826 [2026-04-17 12:38:20] Validation | Batch 260/784 | Loss: 0.3161 | LM_LOSS: 0.3053 | LB_LOSS: 1.0828 [2026-04-17 12:38:22] Validation | Batch 270/784 | Loss: 0.3159 | LM_LOSS: 0.3051 | LB_LOSS: 1.0829 [2026-04-17 12:38:23] Validation | Batch 280/784 | Loss: 0.3164 | LM_LOSS: 0.3055 | LB_LOSS: 1.0830 [2026-04-17 12:38:24] Validation | Batch 290/784 | Loss: 0.3174 | LM_LOSS: 0.3066 | LB_LOSS: 1.0831 [2026-04-17 12:38:25] Validation | Batch 300/784 | Loss: 0.3182 | LM_LOSS: 0.3074 | LB_LOSS: 1.0832 [2026-04-17 12:38:27] Validation | Batch 310/784 | Loss: 0.3176 | LM_LOSS: 0.3068 | LB_LOSS: 1.0831 [2026-04-17 12:38:28] Validation | Batch 320/784 | Loss: 0.3192 | LM_LOSS: 0.3084 | LB_LOSS: 1.0831 [2026-04-17 12:38:30] Validation | Batch 330/784 | Loss: 0.3190 | LM_LOSS: 0.3082 | LB_LOSS: 1.0831 [2026-04-17 12:38:31] Validation | Batch 340/784 | Loss: 0.3179 | LM_LOSS: 0.3070 | LB_LOSS: 1.0832 [2026-04-17 12:38:32] Validation | Batch 350/784 | Loss: 0.3180 | LM_LOSS: 0.3072 | LB_LOSS: 1.0834 [2026-04-17 12:38:33] Validation | Batch 360/784 | Loss: 0.3178 | LM_LOSS: 0.3069 | LB_LOSS: 1.0834 [2026-04-17 12:38:35] Validation | Batch 370/784 | Loss: 0.3183 | LM_LOSS: 0.3074 | LB_LOSS: 1.0833 [2026-04-17 12:38:36] Validation | Batch 380/784 | Loss: 0.3181 | LM_LOSS: 0.3072 | LB_LOSS: 1.0834 [2026-04-17 12:38:37] Validation | Batch 390/784 | Loss: 0.3180 | LM_LOSS: 0.3071 | LB_LOSS: 1.0834 [2026-04-17 12:38:39] Validation | Batch 400/784 | Loss: 0.3182 | LM_LOSS: 0.3074 | LB_LOSS: 1.0834 [2026-04-17 12:38:40] Validation | Batch 410/784 | Loss: 0.3185 | LM_LOSS: 0.3077 | LB_LOSS: 1.0834 [2026-04-17 12:38:41] Validation | Batch 420/784 | Loss: 0.3187 | LM_LOSS: 0.3079 | LB_LOSS: 1.0835 [2026-04-17 12:38:42] Validation | Batch 430/784 | Loss: 0.3188 | LM_LOSS: 0.3080 | LB_LOSS: 1.0834 [2026-04-17 12:38:43] Validation | Batch 440/784 | Loss: 0.3185 | LM_LOSS: 0.3076 | LB_LOSS: 1.0834 [2026-04-17 12:38:45] Validation | Batch 450/784 | Loss: 0.3177 | LM_LOSS: 0.3069 | LB_LOSS: 1.0834 [2026-04-17 12:38:46] Validation | Batch 460/784 | Loss: 0.3182 | LM_LOSS: 0.3074 | LB_LOSS: 1.0835 [2026-04-17 12:38:48] Validation | Batch 470/784 | Loss: 0.3174 | LM_LOSS: 0.3066 | LB_LOSS: 1.0835 [2026-04-17 12:38:49] Validation | Batch 480/784 | Loss: 0.3179 | LM_LOSS: 0.3070 | LB_LOSS: 1.0834 [2026-04-17 12:38:50] Validation | Batch 490/784 | Loss: 0.3172 | LM_LOSS: 0.3064 | LB_LOSS: 1.0834 [2026-04-17 12:38:52] Validation | Batch 500/784 | Loss: 0.3176 | LM_LOSS: 0.3068 | LB_LOSS: 1.0833 [2026-04-17 12:38:53] Validation | Batch 510/784 | Loss: 0.3173 | LM_LOSS: 0.3065 | LB_LOSS: 1.0833 [2026-04-17 12:38:54] Validation | Batch 520/784 | Loss: 0.3175 | LM_LOSS: 0.3067 | LB_LOSS: 1.0832 [2026-04-17 12:38:56] Validation | Batch 530/784 | Loss: 0.3183 | LM_LOSS: 0.3075 | LB_LOSS: 1.0831 [2026-04-17 12:38:57] Validation | Batch 540/784 | Loss: 0.3187 | LM_LOSS: 0.3079 | LB_LOSS: 1.0832 [2026-04-17 12:38:59] Validation | Batch 550/784 | Loss: 0.3199 | LM_LOSS: 0.3091 | LB_LOSS: 1.0831 [2026-04-17 12:39:00] Validation | Batch 560/784 | Loss: 0.3200 | LM_LOSS: 0.3092 | LB_LOSS: 1.0832 [2026-04-17 12:39:01] Validation | Batch 570/784 | Loss: 0.3196 | LM_LOSS: 0.3088 | LB_LOSS: 1.0831 [2026-04-17 12:39:03] Validation | Batch 580/784 | Loss: 0.3191 | LM_LOSS: 0.3082 | LB_LOSS: 1.0831 [2026-04-17 12:39:04] Validation | Batch 590/784 | Loss: 0.3193 | LM_LOSS: 0.3085 | LB_LOSS: 1.0831 [2026-04-17 12:39:05] Validation | Batch 600/784 | Loss: 0.3192 | LM_LOSS: 0.3083 | LB_LOSS: 1.0830 [2026-04-17 12:39:07] Validation | Batch 610/784 | Loss: 0.3193 | LM_LOSS: 0.3084 | LB_LOSS: 1.0830 [2026-04-17 12:39:08] Validation | Batch 620/784 | Loss: 0.3191 | LM_LOSS: 0.3083 | LB_LOSS: 1.0830 [2026-04-17 12:39:10] Validation | Batch 630/784 | Loss: 0.3199 | LM_LOSS: 0.3090 | LB_LOSS: 1.0830 [2026-04-17 12:39:11] Validation | Batch 640/784 | Loss: 0.3199 | LM_LOSS: 0.3091 | LB_LOSS: 1.0830 [2026-04-17 12:39:13] Validation | Batch 650/784 | Loss: 0.3198 | LM_LOSS: 0.3090 | LB_LOSS: 1.0831 [2026-04-17 12:39:14] Validation | Batch 660/784 | Loss: 0.3202 | LM_LOSS: 0.3093 | LB_LOSS: 1.0831 [2026-04-17 12:39:16] Validation | Batch 670/784 | Loss: 0.3206 | LM_LOSS: 0.3097 | LB_LOSS: 1.0831 [2026-04-17 12:39:17] Validation | Batch 680/784 | Loss: 0.3203 | LM_LOSS: 0.3094 | LB_LOSS: 1.0831 [2026-04-17 12:39:18] Validation | Batch 690/784 | Loss: 0.3204 | LM_LOSS: 0.3096 | LB_LOSS: 1.0831 [2026-04-17 12:39:20] Validation | Batch 700/784 | Loss: 0.3205 | LM_LOSS: 0.3097 | LB_LOSS: 1.0830 [2026-04-17 12:39:21] Validation | Batch 710/784 | Loss: 0.3203 | LM_LOSS: 0.3094 | LB_LOSS: 1.0830 [2026-04-17 12:39:23] Validation | Batch 720/784 | Loss: 0.3200 | LM_LOSS: 0.3092 | LB_LOSS: 1.0829 [2026-04-17 12:39:24] Validation | Batch 730/784 | Loss: 0.3195 | LM_LOSS: 0.3087 | LB_LOSS: 1.0829 [2026-04-17 12:39:25] Validation | Batch 740/784 | Loss: 0.3196 | LM_LOSS: 0.3087 | LB_LOSS: 1.0829 [2026-04-17 12:39:26] Validation | Batch 750/784 | Loss: 0.3189 | LM_LOSS: 0.3081 | LB_LOSS: 1.0829 [2026-04-17 12:39:27] Validation | Batch 760/784 | Loss: 0.3190 | LM_LOSS: 0.3082 | LB_LOSS: 1.0829 [2026-04-17 12:39:29] Validation | Batch 770/784 | Loss: 0.3192 | LM_LOSS: 0.3084 | LB_LOSS: 1.0830 [2026-04-17 12:39:30] Validation | Batch 780/784 | Loss: 0.3196 | LM_LOSS: 0.3087 | LB_LOSS: 1.0830 [2026-04-17 12:39:31] Validation | Batch 784/784 | Loss: 0.3198 | LM_LOSS: 0.3089 | LB_LOSS: 1.0829 [2026-04-17 12:39:34] Validation | Loss: 0.3198 | LM_LOSS: 0.3089 | LB_LOSS: 1.0829 | PPL: 1.36 | Time: 106.60s [2026-04-17 12:39:40] Epoch 2 | Step 17010 | Loss: 0.2399 | LM: 0.2272 | LB: 1.0879 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 12:39:47] Epoch 2 | Step 17020 | Loss: 0.2399 | LM: 0.2272 | LB: 1.0879 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 12:39:53] Epoch 2 | Step 17030 | Loss: 0.2399 | LM: 0.2272 | LB: 1.0879 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 12:40:00] Epoch 2 | Step 17040 | Loss: 0.2398 | LM: 0.2271 | LB: 1.0879 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 12:40:06] Epoch 2 | Step 17050 | Loss: 0.2397 | LM: 0.2270 | LB: 1.0879 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 12:40:13] Epoch 2 | Step 17060 | Loss: 0.2396 | LM: 0.2271 | LB: 1.0879 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 12:40:19] Epoch 2 | Step 17070 | Loss: 0.2397 | LM: 0.2271 | LB: 1.0879 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 12:40:26] Epoch 2 | Step 17080 | Loss: 0.2397 | LM: 0.2271 | LB: 1.0879 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 12:40:32] Epoch 2 | Step 17090 | Loss: 0.2397 | LM: 0.2272 | LB: 1.0879 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 12:40:39] Epoch 2 | Step 17100 | Loss: 0.2397 | LM: 0.2272 | LB: 1.0879 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 12:40:45] Epoch 2 | Step 17110 | Loss: 0.2397 | LM: 0.2272 | LB: 1.0879 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 12:40:52] Epoch 2 | Step 17120 | Loss: 0.2397 | LM: 0.2272 | LB: 1.0879 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 12:40:58] Epoch 2 | Step 17130 | Loss: 0.2397 | LM: 0.2272 | LB: 1.0879 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 12:41:05] Epoch 2 | Step 17140 | Loss: 0.2398 | LM: 0.2273 | LB: 1.0879 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 12:41:11] Epoch 2 | Step 17150 | Loss: 0.2398 | LM: 0.2274 | LB: 1.0879 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 12:41:17] Epoch 2 | Step 17160 | Loss: 0.2398 | LM: 0.2274 | LB: 1.0879 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 12:41:24] Epoch 2 | Step 17170 | Loss: 0.2398 | LM: 0.2272 | LB: 1.0879 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 12:41:30] Epoch 2 | Step 17180 | Loss: 0.2398 | LM: 0.2274 | LB: 1.0879 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 12:41:37] Epoch 2 | Step 17190 | Loss: 0.2399 | LM: 0.2274 | LB: 1.0879 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 12:41:43] Epoch 2 | Step 17200 | Loss: 0.2398 | LM: 0.2274 | LB: 1.0879 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 12:41:50] Epoch 2 | Step 17210 | Loss: 0.2398 | LM: 0.2274 | LB: 1.0879 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 12:41:56] Epoch 2 | Step 17220 | Loss: 0.2398 | LM: 0.2274 | LB: 1.0879 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 12:42:02] Epoch 2 | Step 17230 | Loss: 0.2398 | LM: 0.2274 | LB: 1.0879 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 12:42:09] Epoch 2 | Step 17240 | Loss: 0.2398 | LM: 0.2275 | LB: 1.0879 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 12:42:15] Epoch 2 | Step 17250 | Loss: 0.2398 | LM: 0.2275 | LB: 1.0879 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 12:42:22] Epoch 2 | Step 17260 | Loss: 0.2398 | LM: 0.2275 | LB: 1.0879 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 12:42:28] Epoch 2 | Step 17270 | Loss: 0.2398 | LM: 0.2276 | LB: 1.0879 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 12:42:35] Epoch 2 | Step 17280 | Loss: 0.2398 | LM: 0.2275 | LB: 1.0879 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 12:42:41] Epoch 2 | Step 17290 | Loss: 0.2399 | LM: 0.2275 | LB: 1.0879 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 12:42:47] Epoch 2 | Step 17300 | Loss: 0.2399 | LM: 0.2277 | LB: 1.0879 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 12:42:54] Epoch 2 | Step 17310 | Loss: 0.2399 | LM: 0.2277 | LB: 1.0879 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 12:43:00] Epoch 2 | Step 17320 | Loss: 0.2399 | LM: 0.2277 | LB: 1.0879 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 12:43:07] Epoch 2 | Step 17330 | Loss: 0.2398 | LM: 0.2276 | LB: 1.0879 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 12:43:13] Epoch 2 | Step 17340 | Loss: 0.2398 | LM: 0.2276 | LB: 1.0879 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 12:43:19] Epoch 2 | Step 17350 | Loss: 0.2398 | LM: 0.2276 | LB: 1.0879 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 12:43:26] Epoch 2 | Step 17360 | Loss: 0.2398 | LM: 0.2275 | LB: 1.0879 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 12:43:32] Epoch 2 | Step 17370 | Loss: 0.2398 | LM: 0.2276 | LB: 1.0879 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 12:43:38] Epoch 2 | Step 17380 | Loss: 0.2398 | LM: 0.2276 | LB: 1.0879 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 12:43:45] Epoch 2 | Step 17390 | Loss: 0.2398 | LM: 0.2278 | LB: 1.0879 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 12:43:51] Epoch 2 | Step 17400 | Loss: 0.2398 | LM: 0.2278 | LB: 1.0879 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 12:43:57] Epoch 2 | Step 17410 | Loss: 0.2398 | LM: 0.2277 | LB: 1.0879 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 12:44:04] Epoch 2 | Step 17420 | Loss: 0.2398 | LM: 0.2278 | LB: 1.0879 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 12:44:10] Epoch 2 | Step 17430 | Loss: 0.2398 | LM: 0.2279 | LB: 1.0879 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 12:44:16] Epoch 2 | Step 17440 | Loss: 0.2398 | LM: 0.2279 | LB: 1.0879 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 12:44:23] Epoch 2 | Step 17450 | Loss: 0.2399 | LM: 0.2280 | LB: 1.0879 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 12:44:29] Epoch 2 | Step 17460 | Loss: 0.2399 | LM: 0.2281 | LB: 1.0879 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 12:44:35] Epoch 2 | Step 17470 | Loss: 0.2399 | LM: 0.2281 | LB: 1.0879 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 12:44:42] Epoch 2 | Step 17480 | Loss: 0.2400 | LM: 0.2281 | LB: 1.0879 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 12:44:48] Epoch 2 | Step 17490 | Loss: 0.2399 | LM: 0.2280 | LB: 1.0879 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 12:44:55] Epoch 2 | Step 17500 | Loss: 0.2399 | LM: 0.2281 | LB: 1.0879 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 12:45:01] Epoch 2 | Step 17510 | Loss: 0.2399 | LM: 0.2280 | LB: 1.0879 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 12:45:07] Epoch 2 | Step 17520 | Loss: 0.2399 | LM: 0.2280 | LB: 1.0879 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 12:45:14] Epoch 2 | Step 17530 | Loss: 0.2399 | LM: 0.2280 | LB: 1.0879 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 12:45:20] Epoch 2 | Step 17540 | Loss: 0.2398 | LM: 0.2280 | LB: 1.0879 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 12:45:27] Epoch 2 | Step 17550 | Loss: 0.2399 | LM: 0.2280 | LB: 1.0879 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 12:45:33] Epoch 2 | Step 17560 | Loss: 0.2399 | LM: 0.2279 | LB: 1.0879 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 12:45:40] Epoch 2 | Step 17570 | Loss: 0.2399 | LM: 0.2280 | LB: 1.0879 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 12:45:46] Epoch 2 | Step 17580 | Loss: 0.2399 | LM: 0.2280 | LB: 1.0879 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 12:45:52] Epoch 2 | Step 17590 | Loss: 0.2399 | LM: 0.2281 | LB: 1.0879 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 12:45:59] Epoch 2 | Step 17600 | Loss: 0.2400 | LM: 0.2281 | LB: 1.0879 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 12:46:05] Epoch 2 | Step 17610 | Loss: 0.2399 | LM: 0.2281 | LB: 1.0879 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 12:46:11] Epoch 2 | Step 17620 | Loss: 0.2399 | LM: 0.2280 | LB: 1.0879 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 12:46:18] Epoch 2 | Step 17630 | Loss: 0.2399 | LM: 0.2280 | LB: 1.0879 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 12:46:24] Epoch 2 | Step 17640 | Loss: 0.2398 | LM: 0.2279 | LB: 1.0879 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 12:46:30] Epoch 2 | Step 17650 | Loss: 0.2399 | LM: 0.2279 | LB: 1.0879 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 12:46:37] Epoch 2 | Step 17660 | Loss: 0.2399 | LM: 0.2279 | LB: 1.0879 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 12:46:43] Epoch 2 | Step 17670 | Loss: 0.2399 | LM: 0.2279 | LB: 1.0879 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 12:46:49] Epoch 2 | Step 17680 | Loss: 0.2400 | LM: 0.2280 | LB: 1.0879 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 12:46:56] Epoch 2 | Step 17690 | Loss: 0.2400 | LM: 0.2280 | LB: 1.0879 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 12:47:02] Epoch 2 | Step 17700 | Loss: 0.2401 | LM: 0.2281 | LB: 1.0879 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 12:47:08] Epoch 2 | Step 17710 | Loss: 0.2401 | LM: 0.2280 | LB: 1.0879 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 12:47:15] Epoch 2 | Step 17720 | Loss: 0.2401 | LM: 0.2280 | LB: 1.0879 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 12:47:21] Epoch 2 | Step 17730 | Loss: 0.2401 | LM: 0.2281 | LB: 1.0879 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 12:47:27] Epoch 2 | Step 17740 | Loss: 0.2401 | LM: 0.2281 | LB: 1.0879 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 12:47:34] Epoch 2 | Step 17750 | Loss: 0.2401 | LM: 0.2282 | LB: 1.0879 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 12:47:40] Epoch 2 | Step 17760 | Loss: 0.2401 | LM: 0.2282 | LB: 1.0879 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 12:47:47] Epoch 2 | Step 17770 | Loss: 0.2400 | LM: 0.2281 | LB: 1.0879 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 12:47:53] Epoch 2 | Step 17780 | Loss: 0.2401 | LM: 0.2281 | LB: 1.0879 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 12:47:59] Epoch 2 | Step 17790 | Loss: 0.2401 | LM: 0.2282 | LB: 1.0879 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 12:48:06] Epoch 2 | Step 17800 | Loss: 0.2401 | LM: 0.2282 | LB: 1.0879 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 12:48:12] Epoch 2 | Step 17810 | Loss: 0.2401 | LM: 0.2282 | LB: 1.0879 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 12:48:18] Epoch 2 | Step 17820 | Loss: 0.2401 | LM: 0.2282 | LB: 1.0879 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 12:48:25] Epoch 2 | Step 17830 | Loss: 0.2400 | LM: 0.2283 | LB: 1.0878 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.415/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 12:48:31] Epoch 2 | Step 17840 | Loss: 0.2400 | LM: 0.2283 | LB: 1.0878 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.415/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 12:48:37] Epoch 2 | Step 17850 | Loss: 0.2400 | LM: 0.2282 | LB: 1.0878 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.415/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 12:48:44] Epoch 2 | Step 17860 | Loss: 0.2400 | LM: 0.2282 | LB: 1.0878 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.415/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 12:48:50] Epoch 2 | Step 17870 | Loss: 0.2399 | LM: 0.2282 | LB: 1.0878 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.415/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 12:48:57] Epoch 2 | Step 17880 | Loss: 0.2400 | LM: 0.2281 | LB: 1.0878 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.415/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 12:49:03] Epoch 2 | Step 17890 | Loss: 0.2399 | LM: 0.2281 | LB: 1.0878 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.415/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 12:49:09] Epoch 2 | Step 17900 | Loss: 0.2399 | LM: 0.2281 | LB: 1.0878 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.415/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 12:49:15] Epoch 2 | Step 17910 | Loss: 0.2398 | LM: 0.2280 | LB: 1.0878 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.415/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 12:49:22] Epoch 2 | Step 17920 | Loss: 0.2398 | LM: 0.2280 | LB: 1.0878 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.415/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 12:49:28] Epoch 2 | Step 17930 | Loss: 0.2398 | LM: 0.2280 | LB: 1.0878 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.415/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 12:49:34] Epoch 2 | Step 17940 | Loss: 0.2398 | LM: 0.2279 | LB: 1.0878 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.415/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 12:49:41] Epoch 2 | Step 17950 | Loss: 0.2398 | LM: 0.2279 | LB: 1.0878 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.415/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 12:49:47] Epoch 2 | Step 17960 | Loss: 0.2398 | LM: 0.2279 | LB: 1.0878 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.415/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 12:49:54] Epoch 2 | Step 17970 | Loss: 0.2398 | LM: 0.2280 | LB: 1.0878 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.415/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 12:50:00] Epoch 2 | Step 17980 | Loss: 0.2398 | LM: 0.2280 | LB: 1.0878 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.415/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 12:50:06] Epoch 2 | Step 17990 | Loss: 0.2398 | LM: 0.2279 | LB: 1.0878 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.415/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 12:50:12] Epoch 2 | Step 18000 | Loss: 0.2398 | LM: 0.2279 | LB: 1.0878 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.415/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 12:50:21] Checkpoint saved: outputs/2026-04-17/08-57-56/checkpoints/checkpoint_step_18000.pt [2026-04-17 12:50:37] Validation | Batch 10/784 | Loss: 0.3263 | LM_LOSS: 0.3155 | LB_LOSS: 1.0844 [2026-04-17 12:50:39] Validation | Batch 20/784 | Loss: 0.3371 | LM_LOSS: 0.3263 | LB_LOSS: 1.0846 [2026-04-17 12:50:40] Validation | Batch 30/784 | Loss: 0.3226 | LM_LOSS: 0.3118 | LB_LOSS: 1.0839 [2026-04-17 12:50:41] Validation | Batch 40/784 | Loss: 0.3249 | LM_LOSS: 0.3141 | LB_LOSS: 1.0838 [2026-04-17 12:50:43] Validation | Batch 50/784 | Loss: 0.3220 | LM_LOSS: 0.3111 | LB_LOSS: 1.0831 [2026-04-17 12:50:44] Validation | Batch 60/784 | Loss: 0.3238 | LM_LOSS: 0.3130 | LB_LOSS: 1.0827 [2026-04-17 12:50:45] Validation | Batch 70/784 | Loss: 0.3211 | LM_LOSS: 0.3102 | LB_LOSS: 1.0820 [2026-04-17 12:50:47] Validation | Batch 80/784 | Loss: 0.3172 | LM_LOSS: 0.3064 | LB_LOSS: 1.0816 [2026-04-17 12:50:48] Validation | Batch 90/784 | Loss: 0.3161 | LM_LOSS: 0.3053 | LB_LOSS: 1.0821 [2026-04-17 12:50:49] Validation | Batch 100/784 | Loss: 0.3179 | LM_LOSS: 0.3071 | LB_LOSS: 1.0826 [2026-04-17 12:50:51] Validation | Batch 110/784 | Loss: 0.3127 | LM_LOSS: 0.3018 | LB_LOSS: 1.0827 [2026-04-17 12:50:52] Validation | Batch 120/784 | Loss: 0.3162 | LM_LOSS: 0.3054 | LB_LOSS: 1.0826 [2026-04-17 12:50:53] Validation | Batch 130/784 | Loss: 0.3192 | LM_LOSS: 0.3084 | LB_LOSS: 1.0826 [2026-04-17 12:50:55] Validation | Batch 140/784 | Loss: 0.3185 | LM_LOSS: 0.3077 | LB_LOSS: 1.0824 [2026-04-17 12:50:56] Validation | Batch 150/784 | Loss: 0.3146 | LM_LOSS: 0.3038 | LB_LOSS: 1.0827 [2026-04-17 12:50:58] Validation | Batch 160/784 | Loss: 0.3153 | LM_LOSS: 0.3045 | LB_LOSS: 1.0824 [2026-04-17 12:50:59] Validation | Batch 170/784 | Loss: 0.3156 | LM_LOSS: 0.3047 | LB_LOSS: 1.0821 [2026-04-17 12:51:00] Validation | Batch 180/784 | Loss: 0.3132 | LM_LOSS: 0.3023 | LB_LOSS: 1.0821 [2026-04-17 12:51:02] Validation | Batch 190/784 | Loss: 0.3152 | LM_LOSS: 0.3044 | LB_LOSS: 1.0826 [2026-04-17 12:51:03] Validation | Batch 200/784 | Loss: 0.3156 | LM_LOSS: 0.3048 | LB_LOSS: 1.0826 [2026-04-17 12:51:04] Validation | Batch 210/784 | Loss: 0.3145 | LM_LOSS: 0.3037 | LB_LOSS: 1.0826 [2026-04-17 12:51:06] Validation | Batch 220/784 | Loss: 0.3153 | LM_LOSS: 0.3045 | LB_LOSS: 1.0826 [2026-04-17 12:51:07] Validation | Batch 230/784 | Loss: 0.3158 | LM_LOSS: 0.3050 | LB_LOSS: 1.0825 [2026-04-17 12:51:09] Validation | Batch 240/784 | Loss: 0.3162 | LM_LOSS: 0.3054 | LB_LOSS: 1.0828 [2026-04-17 12:51:10] Validation | Batch 250/784 | Loss: 0.3161 | LM_LOSS: 0.3053 | LB_LOSS: 1.0827 [2026-04-17 12:51:11] Validation | Batch 260/784 | Loss: 0.3163 | LM_LOSS: 0.3055 | LB_LOSS: 1.0829 [2026-04-17 12:51:13] Validation | Batch 270/784 | Loss: 0.3161 | LM_LOSS: 0.3053 | LB_LOSS: 1.0829 [2026-04-17 12:51:14] Validation | Batch 280/784 | Loss: 0.3166 | LM_LOSS: 0.3057 | LB_LOSS: 1.0831 [2026-04-17 12:51:16] Validation | Batch 290/784 | Loss: 0.3176 | LM_LOSS: 0.3068 | LB_LOSS: 1.0832 [2026-04-17 12:51:17] Validation | Batch 300/784 | Loss: 0.3184 | LM_LOSS: 0.3075 | LB_LOSS: 1.0833 [2026-04-17 12:51:18] Validation | Batch 310/784 | Loss: 0.3178 | LM_LOSS: 0.3070 | LB_LOSS: 1.0832 [2026-04-17 12:51:20] Validation | Batch 320/784 | Loss: 0.3194 | LM_LOSS: 0.3086 | LB_LOSS: 1.0832 [2026-04-17 12:51:21] Validation | Batch 330/784 | Loss: 0.3192 | LM_LOSS: 0.3084 | LB_LOSS: 1.0832 [2026-04-17 12:51:22] Validation | Batch 340/784 | Loss: 0.3181 | LM_LOSS: 0.3072 | LB_LOSS: 1.0833 [2026-04-17 12:51:24] Validation | Batch 350/784 | Loss: 0.3182 | LM_LOSS: 0.3074 | LB_LOSS: 1.0835 [2026-04-17 12:51:25] Validation | Batch 360/784 | Loss: 0.3180 | LM_LOSS: 0.3071 | LB_LOSS: 1.0835 [2026-04-17 12:51:26] Validation | Batch 370/784 | Loss: 0.3184 | LM_LOSS: 0.3076 | LB_LOSS: 1.0834 [2026-04-17 12:51:27] Validation | Batch 380/784 | Loss: 0.3183 | LM_LOSS: 0.3074 | LB_LOSS: 1.0834 [2026-04-17 12:51:29] Validation | Batch 390/784 | Loss: 0.3182 | LM_LOSS: 0.3073 | LB_LOSS: 1.0835 [2026-04-17 12:51:30] Validation | Batch 400/784 | Loss: 0.3184 | LM_LOSS: 0.3076 | LB_LOSS: 1.0835 [2026-04-17 12:51:31] Validation | Batch 410/784 | Loss: 0.3187 | LM_LOSS: 0.3079 | LB_LOSS: 1.0835 [2026-04-17 12:51:32] Validation | Batch 420/784 | Loss: 0.3189 | LM_LOSS: 0.3081 | LB_LOSS: 1.0836 [2026-04-17 12:51:34] Validation | Batch 430/784 | Loss: 0.3190 | LM_LOSS: 0.3082 | LB_LOSS: 1.0835 [2026-04-17 12:51:35] Validation | Batch 440/784 | Loss: 0.3187 | LM_LOSS: 0.3078 | LB_LOSS: 1.0835 [2026-04-17 12:51:37] Validation | Batch 450/784 | Loss: 0.3180 | LM_LOSS: 0.3071 | LB_LOSS: 1.0835 [2026-04-17 12:51:38] Validation | Batch 460/784 | Loss: 0.3184 | LM_LOSS: 0.3076 | LB_LOSS: 1.0836 [2026-04-17 12:51:39] Validation | Batch 470/784 | Loss: 0.3176 | LM_LOSS: 0.3068 | LB_LOSS: 1.0835 [2026-04-17 12:51:41] Validation | Batch 480/784 | Loss: 0.3181 | LM_LOSS: 0.3073 | LB_LOSS: 1.0835 [2026-04-17 12:51:42] Validation | Batch 490/784 | Loss: 0.3174 | LM_LOSS: 0.3066 | LB_LOSS: 1.0834 [2026-04-17 12:51:43] Validation | Batch 500/784 | Loss: 0.3178 | LM_LOSS: 0.3070 | LB_LOSS: 1.0834 [2026-04-17 12:51:45] Validation | Batch 510/784 | Loss: 0.3175 | LM_LOSS: 0.3067 | LB_LOSS: 1.0834 [2026-04-17 12:51:46] Validation | Batch 520/784 | Loss: 0.3177 | LM_LOSS: 0.3069 | LB_LOSS: 1.0833 [2026-04-17 12:51:47] Validation | Batch 530/784 | Loss: 0.3186 | LM_LOSS: 0.3077 | LB_LOSS: 1.0832 [2026-04-17 12:51:49] Validation | Batch 540/784 | Loss: 0.3189 | LM_LOSS: 0.3081 | LB_LOSS: 1.0832 [2026-04-17 12:51:50] Validation | Batch 550/784 | Loss: 0.3202 | LM_LOSS: 0.3093 | LB_LOSS: 1.0832 [2026-04-17 12:51:52] Validation | Batch 560/784 | Loss: 0.3203 | LM_LOSS: 0.3094 | LB_LOSS: 1.0833 [2026-04-17 12:51:53] Validation | Batch 570/784 | Loss: 0.3198 | LM_LOSS: 0.3090 | LB_LOSS: 1.0832 [2026-04-17 12:51:54] Validation | Batch 580/784 | Loss: 0.3193 | LM_LOSS: 0.3085 | LB_LOSS: 1.0832 [2026-04-17 12:51:56] Validation | Batch 590/784 | Loss: 0.3196 | LM_LOSS: 0.3087 | LB_LOSS: 1.0831 [2026-04-17 12:51:57] Validation | Batch 600/784 | Loss: 0.3194 | LM_LOSS: 0.3086 | LB_LOSS: 1.0831 [2026-04-17 12:51:59] Validation | Batch 610/784 | Loss: 0.3195 | LM_LOSS: 0.3087 | LB_LOSS: 1.0831 [2026-04-17 12:52:00] Validation | Batch 620/784 | Loss: 0.3194 | LM_LOSS: 0.3086 | LB_LOSS: 1.0831 [2026-04-17 12:52:01] Validation | Batch 630/784 | Loss: 0.3201 | LM_LOSS: 0.3093 | LB_LOSS: 1.0831 [2026-04-17 12:52:03] Validation | Batch 640/784 | Loss: 0.3202 | LM_LOSS: 0.3094 | LB_LOSS: 1.0831 [2026-04-17 12:52:05] Validation | Batch 650/784 | Loss: 0.3201 | LM_LOSS: 0.3092 | LB_LOSS: 1.0832 [2026-04-17 12:52:06] Validation | Batch 660/784 | Loss: 0.3204 | LM_LOSS: 0.3096 | LB_LOSS: 1.0831 [2026-04-17 12:52:07] Validation | Batch 670/784 | Loss: 0.3208 | LM_LOSS: 0.3100 | LB_LOSS: 1.0832 [2026-04-17 12:52:09] Validation | Batch 680/784 | Loss: 0.3205 | LM_LOSS: 0.3097 | LB_LOSS: 1.0832 [2026-04-17 12:52:10] Validation | Batch 690/784 | Loss: 0.3207 | LM_LOSS: 0.3099 | LB_LOSS: 1.0831 [2026-04-17 12:52:12] Validation | Batch 700/784 | Loss: 0.3207 | LM_LOSS: 0.3099 | LB_LOSS: 1.0831 [2026-04-17 12:52:13] Validation | Batch 710/784 | Loss: 0.3205 | LM_LOSS: 0.3097 | LB_LOSS: 1.0831 [2026-04-17 12:52:14] Validation | Batch 720/784 | Loss: 0.3202 | LM_LOSS: 0.3094 | LB_LOSS: 1.0830 [2026-04-17 12:52:16] Validation | Batch 730/784 | Loss: 0.3197 | LM_LOSS: 0.3089 | LB_LOSS: 1.0830 [2026-04-17 12:52:17] Validation | Batch 740/784 | Loss: 0.3198 | LM_LOSS: 0.3090 | LB_LOSS: 1.0830 [2026-04-17 12:52:19] Validation | Batch 750/784 | Loss: 0.3191 | LM_LOSS: 0.3083 | LB_LOSS: 1.0830 [2026-04-17 12:52:20] Validation | Batch 760/784 | Loss: 0.3193 | LM_LOSS: 0.3085 | LB_LOSS: 1.0830 [2026-04-17 12:52:22] Validation | Batch 770/784 | Loss: 0.3195 | LM_LOSS: 0.3086 | LB_LOSS: 1.0831 [2026-04-17 12:52:23] Validation | Batch 780/784 | Loss: 0.3198 | LM_LOSS: 0.3090 | LB_LOSS: 1.0830 [2026-04-17 12:52:24] Validation | Batch 784/784 | Loss: 0.3200 | LM_LOSS: 0.3092 | LB_LOSS: 1.0830 [2026-04-17 12:52:26] Validation | Loss: 0.3200 | LM_LOSS: 0.3092 | LB_LOSS: 1.0830 | PPL: 1.36 | Time: 107.98s [2026-04-17 12:52:33] Epoch 2 | Step 18010 | Loss: 0.2398 | LM: 0.2279 | LB: 1.0878 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.415/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 12:52:39] Epoch 2 | Step 18020 | Loss: 0.2397 | LM: 0.2278 | LB: 1.0878 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.415/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 12:52:45] Epoch 2 | Step 18030 | Loss: 0.2397 | LM: 0.2278 | LB: 1.0878 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.415/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 12:52:52] Epoch 2 | Step 18040 | Loss: 0.2398 | LM: 0.2278 | LB: 1.0878 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.415/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 12:52:58] Epoch 2 | Step 18050 | Loss: 0.2397 | LM: 0.2278 | LB: 1.0878 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.415/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 12:53:05] Epoch 2 | Step 18060 | Loss: 0.2398 | LM: 0.2278 | LB: 1.0878 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.415/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 12:53:11] Epoch 2 | Step 18070 | Loss: 0.2398 | LM: 0.2278 | LB: 1.0878 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.415/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 12:53:18] Epoch 2 | Step 18080 | Loss: 0.2398 | LM: 0.2278 | LB: 1.0878 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.415/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 12:53:24] Epoch 2 | Step 18090 | Loss: 0.2399 | LM: 0.2279 | LB: 1.0878 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.415/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 12:53:31] Epoch 2 | Step 18100 | Loss: 0.2398 | LM: 0.2279 | LB: 1.0878 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.415/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 12:53:37] Epoch 2 | Step 18110 | Loss: 0.2399 | LM: 0.2279 | LB: 1.0878 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.415/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 12:53:43] Epoch 2 | Step 18120 | Loss: 0.2398 | LM: 0.2279 | LB: 1.0878 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.415/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 12:53:50] Epoch 2 | Step 18130 | Loss: 0.2398 | LM: 0.2280 | LB: 1.0878 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.415/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 12:53:56] Epoch 2 | Step 18140 | Loss: 0.2398 | LM: 0.2279 | LB: 1.0878 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.415/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 12:54:03] Epoch 2 | Step 18150 | Loss: 0.2398 | LM: 0.2280 | LB: 1.0878 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.415/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 12:54:09] Epoch 2 | Step 18160 | Loss: 0.2398 | LM: 0.2281 | LB: 1.0878 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.415/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 12:54:15] Epoch 2 | Step 18170 | Loss: 0.2398 | LM: 0.2281 | LB: 1.0878 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.415/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 12:54:22] Epoch 2 | Step 18180 | Loss: 0.2398 | LM: 0.2280 | LB: 1.0878 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.415/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 12:54:28] Epoch 2 | Step 18190 | Loss: 0.2398 | LM: 0.2280 | LB: 1.0878 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.415/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 12:54:34] Epoch 2 | Step 18200 | Loss: 0.2398 | LM: 0.2281 | LB: 1.0878 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.415/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 12:54:41] Epoch 2 | Step 18210 | Loss: 0.2398 | LM: 0.2281 | LB: 1.0878 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.415/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 12:54:47] Epoch 2 | Step 18220 | Loss: 0.2398 | LM: 0.2280 | LB: 1.0878 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.415/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 12:54:54] Epoch 2 | Step 18230 | Loss: 0.2398 | LM: 0.2280 | LB: 1.0878 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.415/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 12:55:00] Epoch 2 | Step 18240 | Loss: 0.2397 | LM: 0.2280 | LB: 1.0878 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.415/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 12:55:06] Epoch 2 | Step 18250 | Loss: 0.2398 | LM: 0.2280 | LB: 1.0878 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.415/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 12:55:13] Epoch 2 | Step 18260 | Loss: 0.2397 | LM: 0.2280 | LB: 1.0878 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.415/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 12:55:19] Epoch 2 | Step 18270 | Loss: 0.2398 | LM: 0.2281 | LB: 1.0878 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.415/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 12:55:25] Epoch 2 | Step 18280 | Loss: 0.2398 | LM: 0.2280 | LB: 1.0878 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.415/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 12:55:32] Epoch 2 | Step 18290 | Loss: 0.2398 | LM: 0.2281 | LB: 1.0878 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.415/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 12:55:38] Epoch 2 | Step 18300 | Loss: 0.2398 | LM: 0.2281 | LB: 1.0878 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 12:55:45] Epoch 2 | Step 18310 | Loss: 0.2398 | LM: 0.2282 | LB: 1.0878 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 12:55:51] Epoch 2 | Step 18320 | Loss: 0.2398 | LM: 0.2282 | LB: 1.0878 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 12:55:57] Epoch 2 | Step 18330 | Loss: 0.2398 | LM: 0.2281 | LB: 1.0878 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 12:56:04] Epoch 2 | Step 18340 | Loss: 0.2398 | LM: 0.2281 | LB: 1.0878 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.415/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 12:56:09] Epoch 2 | Step 18350 | Loss: 0.2398 | LM: 0.2281 | LB: 1.0878 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.415/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 12:56:16] Epoch 2 | Step 18360 | Loss: 0.2397 | LM: 0.2281 | LB: 1.0878 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.415/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 12:56:22] Epoch 2 | Step 18370 | Loss: 0.2397 | LM: 0.2281 | LB: 1.0878 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.415/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 12:56:29] Epoch 2 | Step 18380 | Loss: 0.2398 | LM: 0.2280 | LB: 1.0878 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.415/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 12:56:35] Epoch 2 | Step 18390 | Loss: 0.2398 | LM: 0.2281 | LB: 1.0878 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.415/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 12:56:41] Epoch 2 | Step 18400 | Loss: 0.2398 | LM: 0.2281 | LB: 1.0878 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.415/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 12:56:47] Epoch 2 | Step 18410 | Loss: 0.2398 | LM: 0.2281 | LB: 1.0878 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.415/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 12:56:54] Epoch 2 | Step 18420 | Loss: 0.2398 | LM: 0.2281 | LB: 1.0878 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.415/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 12:57:00] Epoch 2 | Step 18430 | Loss: 0.2399 | LM: 0.2281 | LB: 1.0878 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.415/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 12:57:07] Epoch 2 | Step 18440 | Loss: 0.2399 | LM: 0.2281 | LB: 1.0878 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.415/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 12:57:14] Epoch 2 | Step 18450 | Loss: 0.2398 | LM: 0.2280 | LB: 1.0878 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.415/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 12:57:20] Epoch 2 | Step 18460 | Loss: 0.2398 | LM: 0.2280 | LB: 1.0878 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.415/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 12:57:27] Epoch 2 | Step 18470 | Loss: 0.2398 | LM: 0.2280 | LB: 1.0878 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.415/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 12:57:33] Epoch 2 | Step 18480 | Loss: 0.2398 | LM: 0.2281 | LB: 1.0878 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.415/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 12:57:39] Epoch 2 | Step 18490 | Loss: 0.2398 | LM: 0.2281 | LB: 1.0878 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 12:57:45] Epoch 2 | Step 18500 | Loss: 0.2398 | LM: 0.2281 | LB: 1.0878 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 12:57:52] Epoch 2 | Step 18510 | Loss: 0.2398 | LM: 0.2281 | LB: 1.0878 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 12:57:58] Epoch 2 | Step 18520 | Loss: 0.2398 | LM: 0.2282 | LB: 1.0878 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 12:58:04] Epoch 2 | Step 18530 | Loss: 0.2397 | LM: 0.2281 | LB: 1.0878 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 12:58:11] Epoch 2 | Step 18540 | Loss: 0.2397 | LM: 0.2281 | LB: 1.0878 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 12:58:17] Epoch 2 | Step 18550 | Loss: 0.2397 | LM: 0.2280 | LB: 1.0878 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 12:58:24] Epoch 2 | Step 18560 | Loss: 0.2398 | LM: 0.2282 | LB: 1.0878 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 12:58:30] Epoch 2 | Step 18570 | Loss: 0.2398 | LM: 0.2283 | LB: 1.0878 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 12:58:37] Epoch 2 | Step 18580 | Loss: 0.2398 | LM: 0.2282 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 12:58:43] Epoch 2 | Step 18590 | Loss: 0.2398 | LM: 0.2282 | LB: 1.0878 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 12:58:50] Epoch 2 | Step 18600 | Loss: 0.2398 | LM: 0.2282 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.415/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 12:58:56] Epoch 2 | Step 18610 | Loss: 0.2398 | LM: 0.2281 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.415/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 12:59:02] Epoch 2 | Step 18620 | Loss: 0.2398 | LM: 0.2281 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.415/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 12:59:09] Epoch 2 | Step 18630 | Loss: 0.2397 | LM: 0.2280 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.415/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 12:59:15] Epoch 2 | Step 18640 | Loss: 0.2398 | LM: 0.2279 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.415/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 12:59:22] Epoch 2 | Step 18650 | Loss: 0.2398 | LM: 0.2279 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.415/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 12:59:28] Epoch 2 | Step 18660 | Loss: 0.2398 | LM: 0.2280 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.415/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 12:59:35] Epoch 2 | Step 18670 | Loss: 0.2398 | LM: 0.2280 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.415/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 12:59:41] Epoch 2 | Step 18680 | Loss: 0.2398 | LM: 0.2280 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.415/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 12:59:47] Epoch 2 | Step 18690 | Loss: 0.2398 | LM: 0.2281 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.415/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 12:59:54] Epoch 2 | Step 18700 | Loss: 0.2398 | LM: 0.2280 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.415/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 13:00:00] Epoch 2 | Step 18710 | Loss: 0.2398 | LM: 0.2280 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.415/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 13:00:07] Epoch 2 | Step 18720 | Loss: 0.2398 | LM: 0.2280 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 13:00:13] Epoch 2 | Step 18730 | Loss: 0.2397 | LM: 0.2280 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 13:00:19] Epoch 2 | Step 18740 | Loss: 0.2397 | LM: 0.2281 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 13:00:26] Epoch 2 | Step 18750 | Loss: 0.2397 | LM: 0.2280 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 13:00:32] Epoch 2 | Step 18760 | Loss: 0.2396 | LM: 0.2279 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 13:00:39] Epoch 2 | Step 18770 | Loss: 0.2396 | LM: 0.2280 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 13:00:45] Epoch 2 | Step 18780 | Loss: 0.2396 | LM: 0.2281 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 13:00:51] Epoch 2 | Step 18790 | Loss: 0.2396 | LM: 0.2281 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 13:00:58] Epoch 2 | Step 18800 | Loss: 0.2396 | LM: 0.2280 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 13:01:04] Epoch 2 | Step 18810 | Loss: 0.2396 | LM: 0.2280 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 13:01:10] Epoch 2 | Step 18820 | Loss: 0.2396 | LM: 0.2281 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 13:01:17] Epoch 2 | Step 18830 | Loss: 0.2396 | LM: 0.2280 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 13:01:23] Epoch 2 | Step 18840 | Loss: 0.2396 | LM: 0.2280 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 13:01:30] Epoch 2 | Step 18850 | Loss: 0.2397 | LM: 0.2281 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 13:01:36] Epoch 2 | Step 18860 | Loss: 0.2397 | LM: 0.2281 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 13:01:42] Epoch 2 | Step 18870 | Loss: 0.2397 | LM: 0.2282 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 13:01:49] Epoch 2 | Step 18880 | Loss: 0.2397 | LM: 0.2282 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 13:01:55] Epoch 2 | Step 18890 | Loss: 0.2397 | LM: 0.2281 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 13:02:02] Epoch 2 | Step 18900 | Loss: 0.2397 | LM: 0.2281 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 13:02:08] Epoch 2 | Step 18910 | Loss: 0.2397 | LM: 0.2282 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 13:02:15] Epoch 2 | Step 18920 | Loss: 0.2398 | LM: 0.2282 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 13:02:21] Epoch 2 | Step 18930 | Loss: 0.2398 | LM: 0.2282 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 13:02:27] Epoch 2 | Step 18940 | Loss: 0.2398 | LM: 0.2282 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 13:02:34] Epoch 2 | Step 18950 | Loss: 0.2398 | LM: 0.2282 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 13:02:40] Epoch 2 | Step 18960 | Loss: 0.2398 | LM: 0.2282 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 13:02:46] Epoch 2 | Step 18970 | Loss: 0.2398 | LM: 0.2282 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 13:02:53] Epoch 2 | Step 18980 | Loss: 0.2398 | LM: 0.2282 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.415/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 13:02:59] Epoch 2 | Step 18990 | Loss: 0.2398 | LM: 0.2282 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.415/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 13:03:05] Epoch 2 | Step 19000 | Loss: 0.2398 | LM: 0.2282 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.415/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 13:03:07] Validation | Batch 10/784 | Loss: 0.3261 | LM_LOSS: 0.3152 | LB_LOSS: 1.0843 [2026-04-17 13:03:08] Validation | Batch 20/784 | Loss: 0.3367 | LM_LOSS: 0.3258 | LB_LOSS: 1.0846 [2026-04-17 13:03:09] Validation | Batch 30/784 | Loss: 0.3222 | LM_LOSS: 0.3114 | LB_LOSS: 1.0838 [2026-04-17 13:03:11] Validation | Batch 40/784 | Loss: 0.3246 | LM_LOSS: 0.3138 | LB_LOSS: 1.0838 [2026-04-17 13:03:12] Validation | Batch 50/784 | Loss: 0.3217 | LM_LOSS: 0.3109 | LB_LOSS: 1.0831 [2026-04-17 13:03:14] Validation | Batch 60/784 | Loss: 0.3235 | LM_LOSS: 0.3126 | LB_LOSS: 1.0827 [2026-04-17 13:03:15] Validation | Batch 70/784 | Loss: 0.3208 | LM_LOSS: 0.3100 | LB_LOSS: 1.0820 [2026-04-17 13:03:16] Validation | Batch 80/784 | Loss: 0.3170 | LM_LOSS: 0.3062 | LB_LOSS: 1.0816 [2026-04-17 13:03:18] Validation | Batch 90/784 | Loss: 0.3159 | LM_LOSS: 0.3051 | LB_LOSS: 1.0821 [2026-04-17 13:03:19] Validation | Batch 100/784 | Loss: 0.3177 | LM_LOSS: 0.3069 | LB_LOSS: 1.0826 [2026-04-17 13:03:20] Validation | Batch 110/784 | Loss: 0.3125 | LM_LOSS: 0.3017 | LB_LOSS: 1.0827 [2026-04-17 13:03:22] Validation | Batch 120/784 | Loss: 0.3161 | LM_LOSS: 0.3053 | LB_LOSS: 1.0826 [2026-04-17 13:03:23] Validation | Batch 130/784 | Loss: 0.3190 | LM_LOSS: 0.3082 | LB_LOSS: 1.0826 [2026-04-17 13:03:24] Validation | Batch 140/784 | Loss: 0.3183 | LM_LOSS: 0.3075 | LB_LOSS: 1.0824 [2026-04-17 13:03:26] Validation | Batch 150/784 | Loss: 0.3144 | LM_LOSS: 0.3036 | LB_LOSS: 1.0827 [2026-04-17 13:03:27] Validation | Batch 160/784 | Loss: 0.3152 | LM_LOSS: 0.3044 | LB_LOSS: 1.0824 [2026-04-17 13:03:29] Validation | Batch 170/784 | Loss: 0.3154 | LM_LOSS: 0.3046 | LB_LOSS: 1.0821 [2026-04-17 13:03:30] Validation | Batch 180/784 | Loss: 0.3130 | LM_LOSS: 0.3022 | LB_LOSS: 1.0821 [2026-04-17 13:03:31] Validation | Batch 190/784 | Loss: 0.3150 | LM_LOSS: 0.3042 | LB_LOSS: 1.0826 [2026-04-17 13:03:33] Validation | Batch 200/784 | Loss: 0.3154 | LM_LOSS: 0.3046 | LB_LOSS: 1.0826 [2026-04-17 13:03:34] Validation | Batch 210/784 | Loss: 0.3143 | LM_LOSS: 0.3035 | LB_LOSS: 1.0826 [2026-04-17 13:03:35] Validation | Batch 220/784 | Loss: 0.3151 | LM_LOSS: 0.3043 | LB_LOSS: 1.0826 [2026-04-17 13:03:37] Validation | Batch 230/784 | Loss: 0.3156 | LM_LOSS: 0.3048 | LB_LOSS: 1.0825 [2026-04-17 13:03:38] Validation | Batch 240/784 | Loss: 0.3160 | LM_LOSS: 0.3052 | LB_LOSS: 1.0828 [2026-04-17 13:03:39] Validation | Batch 250/784 | Loss: 0.3159 | LM_LOSS: 0.3051 | LB_LOSS: 1.0827 [2026-04-17 13:03:41] Validation | Batch 260/784 | Loss: 0.3161 | LM_LOSS: 0.3053 | LB_LOSS: 1.0829 [2026-04-17 13:03:43] Validation | Batch 270/784 | Loss: 0.3159 | LM_LOSS: 0.3051 | LB_LOSS: 1.0830 [2026-04-17 13:03:44] Validation | Batch 280/784 | Loss: 0.3164 | LM_LOSS: 0.3055 | LB_LOSS: 1.0831 [2026-04-17 13:03:45] Validation | Batch 290/784 | Loss: 0.3174 | LM_LOSS: 0.3066 | LB_LOSS: 1.0832 [2026-04-17 13:03:46] Validation | Batch 300/784 | Loss: 0.3181 | LM_LOSS: 0.3073 | LB_LOSS: 1.0833 [2026-04-17 13:03:48] Validation | Batch 310/784 | Loss: 0.3176 | LM_LOSS: 0.3067 | LB_LOSS: 1.0832 [2026-04-17 13:03:49] Validation | Batch 320/784 | Loss: 0.3192 | LM_LOSS: 0.3083 | LB_LOSS: 1.0832 [2026-04-17 13:03:51] Validation | Batch 330/784 | Loss: 0.3190 | LM_LOSS: 0.3082 | LB_LOSS: 1.0832 [2026-04-17 13:03:52] Validation | Batch 340/784 | Loss: 0.3178 | LM_LOSS: 0.3070 | LB_LOSS: 1.0833 [2026-04-17 13:03:53] Validation | Batch 350/784 | Loss: 0.3180 | LM_LOSS: 0.3072 | LB_LOSS: 1.0835 [2026-04-17 13:03:54] Validation | Batch 360/784 | Loss: 0.3178 | LM_LOSS: 0.3069 | LB_LOSS: 1.0835 [2026-04-17 13:03:55] Validation | Batch 370/784 | Loss: 0.3182 | LM_LOSS: 0.3074 | LB_LOSS: 1.0834 [2026-04-17 13:03:57] Validation | Batch 380/784 | Loss: 0.3181 | LM_LOSS: 0.3072 | LB_LOSS: 1.0835 [2026-04-17 13:03:58] Validation | Batch 390/784 | Loss: 0.3180 | LM_LOSS: 0.3071 | LB_LOSS: 1.0835 [2026-04-17 13:03:59] Validation | Batch 400/784 | Loss: 0.3182 | LM_LOSS: 0.3074 | LB_LOSS: 1.0835 [2026-04-17 13:04:00] Validation | Batch 410/784 | Loss: 0.3185 | LM_LOSS: 0.3077 | LB_LOSS: 1.0835 [2026-04-17 13:04:02] Validation | Batch 420/784 | Loss: 0.3187 | LM_LOSS: 0.3079 | LB_LOSS: 1.0836 [2026-04-17 13:04:03] Validation | Batch 430/784 | Loss: 0.3188 | LM_LOSS: 0.3080 | LB_LOSS: 1.0835 [2026-04-17 13:04:04] Validation | Batch 440/784 | Loss: 0.3185 | LM_LOSS: 0.3076 | LB_LOSS: 1.0835 [2026-04-17 13:04:06] Validation | Batch 450/784 | Loss: 0.3178 | LM_LOSS: 0.3069 | LB_LOSS: 1.0835 [2026-04-17 13:04:07] Validation | Batch 460/784 | Loss: 0.3182 | LM_LOSS: 0.3074 | LB_LOSS: 1.0836 [2026-04-17 13:04:08] Validation | Batch 470/784 | Loss: 0.3174 | LM_LOSS: 0.3066 | LB_LOSS: 1.0835 [2026-04-17 13:04:10] Validation | Batch 480/784 | Loss: 0.3179 | LM_LOSS: 0.3071 | LB_LOSS: 1.0835 [2026-04-17 13:04:11] Validation | Batch 490/784 | Loss: 0.3172 | LM_LOSS: 0.3064 | LB_LOSS: 1.0834 [2026-04-17 13:04:12] Validation | Batch 500/784 | Loss: 0.3176 | LM_LOSS: 0.3068 | LB_LOSS: 1.0834 [2026-04-17 13:04:14] Validation | Batch 510/784 | Loss: 0.3173 | LM_LOSS: 0.3065 | LB_LOSS: 1.0834 [2026-04-17 13:04:15] Validation | Batch 520/784 | Loss: 0.3175 | LM_LOSS: 0.3067 | LB_LOSS: 1.0833 [2026-04-17 13:04:17] Validation | Batch 530/784 | Loss: 0.3184 | LM_LOSS: 0.3075 | LB_LOSS: 1.0832 [2026-04-17 13:04:18] Validation | Batch 540/784 | Loss: 0.3187 | LM_LOSS: 0.3079 | LB_LOSS: 1.0833 [2026-04-17 13:04:19] Validation | Batch 550/784 | Loss: 0.3200 | LM_LOSS: 0.3092 | LB_LOSS: 1.0832 [2026-04-17 13:04:21] Validation | Batch 560/784 | Loss: 0.3201 | LM_LOSS: 0.3093 | LB_LOSS: 1.0833 [2026-04-17 13:04:22] Validation | Batch 570/784 | Loss: 0.3196 | LM_LOSS: 0.3088 | LB_LOSS: 1.0832 [2026-04-17 13:04:24] Validation | Batch 580/784 | Loss: 0.3191 | LM_LOSS: 0.3083 | LB_LOSS: 1.0832 [2026-04-17 13:04:25] Validation | Batch 590/784 | Loss: 0.3194 | LM_LOSS: 0.3085 | LB_LOSS: 1.0831 [2026-04-17 13:04:26] Validation | Batch 600/784 | Loss: 0.3192 | LM_LOSS: 0.3084 | LB_LOSS: 1.0831 [2026-04-17 13:04:28] Validation | Batch 610/784 | Loss: 0.3193 | LM_LOSS: 0.3085 | LB_LOSS: 1.0831 [2026-04-17 13:04:29] Validation | Batch 620/784 | Loss: 0.3192 | LM_LOSS: 0.3084 | LB_LOSS: 1.0831 [2026-04-17 13:04:31] Validation | Batch 630/784 | Loss: 0.3199 | LM_LOSS: 0.3091 | LB_LOSS: 1.0831 [2026-04-17 13:04:32] Validation | Batch 640/784 | Loss: 0.3200 | LM_LOSS: 0.3092 | LB_LOSS: 1.0831 [2026-04-17 13:04:34] Validation | Batch 650/784 | Loss: 0.3199 | LM_LOSS: 0.3090 | LB_LOSS: 1.0832 [2026-04-17 13:04:35] Validation | Batch 660/784 | Loss: 0.3202 | LM_LOSS: 0.3094 | LB_LOSS: 1.0832 [2026-04-17 13:04:37] Validation | Batch 670/784 | Loss: 0.3206 | LM_LOSS: 0.3098 | LB_LOSS: 1.0832 [2026-04-17 13:04:38] Validation | Batch 680/784 | Loss: 0.3203 | LM_LOSS: 0.3095 | LB_LOSS: 1.0832 [2026-04-17 13:04:39] Validation | Batch 690/784 | Loss: 0.3205 | LM_LOSS: 0.3097 | LB_LOSS: 1.0832 [2026-04-17 13:04:41] Validation | Batch 700/784 | Loss: 0.3206 | LM_LOSS: 0.3097 | LB_LOSS: 1.0831 [2026-04-17 13:04:42] Validation | Batch 710/784 | Loss: 0.3203 | LM_LOSS: 0.3095 | LB_LOSS: 1.0831 [2026-04-17 13:04:44] Validation | Batch 720/784 | Loss: 0.3200 | LM_LOSS: 0.3092 | LB_LOSS: 1.0830 [2026-04-17 13:04:45] Validation | Batch 730/784 | Loss: 0.3196 | LM_LOSS: 0.3087 | LB_LOSS: 1.0830 [2026-04-17 13:04:46] Validation | Batch 740/784 | Loss: 0.3196 | LM_LOSS: 0.3088 | LB_LOSS: 1.0830 [2026-04-17 13:04:47] Validation | Batch 750/784 | Loss: 0.3190 | LM_LOSS: 0.3081 | LB_LOSS: 1.0830 [2026-04-17 13:04:48] Validation | Batch 760/784 | Loss: 0.3191 | LM_LOSS: 0.3083 | LB_LOSS: 1.0830 [2026-04-17 13:04:50] Validation | Batch 770/784 | Loss: 0.3193 | LM_LOSS: 0.3085 | LB_LOSS: 1.0831 [2026-04-17 13:04:51] Validation | Batch 780/784 | Loss: 0.3196 | LM_LOSS: 0.3088 | LB_LOSS: 1.0830 [2026-04-17 13:04:52] Validation | Batch 784/784 | Loss: 0.3198 | LM_LOSS: 0.3090 | LB_LOSS: 1.0830 [2026-04-17 13:04:55] Validation | Loss: 0.3198 | LM_LOSS: 0.3090 | LB_LOSS: 1.0830 | PPL: 1.36 | Time: 106.43s [2026-04-17 13:05:01] Epoch 2 | Step 19010 | Loss: 0.2398 | LM: 0.2282 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 13:05:08] Epoch 2 | Step 19020 | Loss: 0.2398 | LM: 0.2282 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 13:05:14] Epoch 2 | Step 19030 | Loss: 0.2398 | LM: 0.2282 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.415/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 13:05:21] Epoch 2 | Step 19040 | Loss: 0.2398 | LM: 0.2282 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.415/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 13:05:27] Epoch 2 | Step 19050 | Loss: 0.2398 | LM: 0.2281 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 13:05:34] Epoch 2 | Step 19060 | Loss: 0.2397 | LM: 0.2281 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 13:05:40] Epoch 2 | Step 19070 | Loss: 0.2398 | LM: 0.2281 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 13:05:46] Epoch 2 | Step 19080 | Loss: 0.2397 | LM: 0.2280 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 13:05:53] Epoch 2 | Step 19090 | Loss: 0.2397 | LM: 0.2280 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 13:05:59] Epoch 2 | Step 19100 | Loss: 0.2397 | LM: 0.2279 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 13:06:05] Epoch 2 | Step 19110 | Loss: 0.2397 | LM: 0.2279 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 13:06:12] Epoch 2 | Step 19120 | Loss: 0.2396 | LM: 0.2278 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 13:06:18] Epoch 2 | Step 19130 | Loss: 0.2396 | LM: 0.2278 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 13:06:24] Epoch 2 | Step 19140 | Loss: 0.2396 | LM: 0.2277 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 13:06:31] Epoch 2 | Step 19150 | Loss: 0.2396 | LM: 0.2277 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 13:06:37] Epoch 2 | Step 19160 | Loss: 0.2396 | LM: 0.2277 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 13:06:44] Epoch 2 | Step 19170 | Loss: 0.2396 | LM: 0.2278 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 13:06:49] Epoch 2 | Step 19180 | Loss: 0.2396 | LM: 0.2278 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 13:06:56] Epoch 2 | Step 19190 | Loss: 0.2396 | LM: 0.2278 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 13:07:02] Epoch 2 | Step 19200 | Loss: 0.2396 | LM: 0.2279 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 13:07:08] Epoch 2 | Step 19210 | Loss: 0.2396 | LM: 0.2279 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 13:07:15] Epoch 2 | Step 19220 | Loss: 0.2396 | LM: 0.2279 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 13:07:21] Epoch 2 | Step 19230 | Loss: 0.2396 | LM: 0.2278 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 13:07:28] Epoch 2 | Step 19240 | Loss: 0.2396 | LM: 0.2277 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 13:07:34] Epoch 2 | Step 19250 | Loss: 0.2396 | LM: 0.2278 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 13:07:41] Epoch 2 | Step 19260 | Loss: 0.2396 | LM: 0.2278 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 13:07:47] Epoch 2 | Step 19270 | Loss: 0.2396 | LM: 0.2278 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 13:07:53] Epoch 2 | Step 19280 | Loss: 0.2396 | LM: 0.2278 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 13:08:00] Epoch 2 | Step 19290 | Loss: 0.2396 | LM: 0.2277 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 13:08:06] Epoch 2 | Step 19300 | Loss: 0.2396 | LM: 0.2278 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 13:08:13] Epoch 2 | Step 19310 | Loss: 0.2396 | LM: 0.2279 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 13:08:19] Epoch 2 | Step 19320 | Loss: 0.2396 | LM: 0.2278 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 13:08:26] Epoch 2 | Step 19330 | Loss: 0.2396 | LM: 0.2278 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 13:08:32] Epoch 2 | Step 19340 | Loss: 0.2396 | LM: 0.2279 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 13:08:38] Epoch 2 | Step 19350 | Loss: 0.2396 | LM: 0.2280 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 13:08:44] Epoch 2 | Step 19360 | Loss: 0.2396 | LM: 0.2279 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 13:08:51] Epoch 2 | Step 19370 | Loss: 0.2396 | LM: 0.2279 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 13:08:57] Epoch 2 | Step 19380 | Loss: 0.2396 | LM: 0.2279 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 13:09:03] Epoch 2 | Step 19390 | Loss: 0.2396 | LM: 0.2279 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 13:09:09] Epoch 2 | Step 19400 | Loss: 0.2396 | LM: 0.2279 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 13:09:16] Epoch 2 | Step 19410 | Loss: 0.2396 | LM: 0.2280 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 13:09:22] Epoch 2 | Step 19420 | Loss: 0.2396 | LM: 0.2280 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 13:09:28] Epoch 2 | Step 19430 | Loss: 0.2396 | LM: 0.2281 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 13:09:35] Epoch 2 | Step 19440 | Loss: 0.2396 | LM: 0.2281 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 13:09:41] Epoch 2 | Step 19450 | Loss: 0.2396 | LM: 0.2282 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 13:09:47] Epoch 2 | Step 19460 | Loss: 0.2396 | LM: 0.2281 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 13:09:54] Epoch 2 | Step 19470 | Loss: 0.2396 | LM: 0.2280 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 13:10:00] Epoch 2 | Step 19480 | Loss: 0.2396 | LM: 0.2280 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 13:10:06] Epoch 2 | Step 19490 | Loss: 0.2396 | LM: 0.2281 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 13:10:13] Epoch 2 | Step 19500 | Loss: 0.2396 | LM: 0.2281 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 13:10:19] Epoch 2 | Step 19510 | Loss: 0.2396 | LM: 0.2281 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 13:10:25] Epoch 2 | Step 19520 | Loss: 0.2395 | LM: 0.2281 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 13:10:32] Epoch 2 | Step 19530 | Loss: 0.2395 | LM: 0.2281 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 13:10:38] Epoch 2 | Step 19540 | Loss: 0.2395 | LM: 0.2281 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 13:10:45] Epoch 2 | Step 19550 | Loss: 0.2396 | LM: 0.2282 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 13:10:51] Epoch 2 | Step 19560 | Loss: 0.2396 | LM: 0.2282 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 13:10:57] Epoch 2 | Step 19570 | Loss: 0.2396 | LM: 0.2282 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 13:11:04] Epoch 2 | Step 19580 | Loss: 0.2397 | LM: 0.2282 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 13:11:10] Epoch 2 | Step 19590 | Loss: 0.2396 | LM: 0.2282 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 13:11:16] Epoch 2 | Step 19600 | Loss: 0.2396 | LM: 0.2282 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 13:11:22] Epoch 2 | Step 19610 | Loss: 0.2396 | LM: 0.2281 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 13:11:29] Epoch 2 | Step 19620 | Loss: 0.2396 | LM: 0.2281 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 13:11:35] Epoch 2 | Step 19630 | Loss: 0.2396 | LM: 0.2282 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 13:11:41] Epoch 2 | Step 19640 | Loss: 0.2396 | LM: 0.2281 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 13:11:47] Epoch 2 | Step 19650 | Loss: 0.2395 | LM: 0.2281 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 13:11:54] Epoch 2 | Step 19660 | Loss: 0.2395 | LM: 0.2281 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 13:12:00] Epoch 2 | Step 19670 | Loss: 0.2396 | LM: 0.2282 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 13:12:06] Epoch 2 | Step 19680 | Loss: 0.2396 | LM: 0.2282 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 13:12:12] Epoch 2 | Step 19690 | Loss: 0.2396 | LM: 0.2281 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 13:12:18] Epoch 2 | Step 19700 | Loss: 0.2396 | LM: 0.2281 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 13:12:25] Epoch 2 | Step 19710 | Loss: 0.2396 | LM: 0.2281 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 13:12:31] Epoch 2 | Step 19720 | Loss: 0.2396 | LM: 0.2281 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 13:12:37] Epoch 2 | Step 19730 | Loss: 0.2396 | LM: 0.2280 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 13:12:43] Epoch 2 | Step 19740 | Loss: 0.2396 | LM: 0.2280 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 13:12:50] Epoch 2 | Step 19750 | Loss: 0.2396 | LM: 0.2281 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 13:12:56] Epoch 2 | Step 19760 | Loss: 0.2396 | LM: 0.2281 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 13:13:02] Epoch 2 | Step 19770 | Loss: 0.2396 | LM: 0.2281 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 13:13:09] Epoch 2 | Step 19780 | Loss: 0.2396 | LM: 0.2281 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 13:13:15] Epoch 2 | Step 19790 | Loss: 0.2396 | LM: 0.2280 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 13:13:21] Epoch 2 | Step 19800 | Loss: 0.2396 | LM: 0.2280 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 13:13:27] Epoch 2 | Step 19810 | Loss: 0.2396 | LM: 0.2280 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 13:13:34] Epoch 2 | Step 19820 | Loss: 0.2396 | LM: 0.2280 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 13:13:40] Epoch 2 | Step 19830 | Loss: 0.2395 | LM: 0.2280 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 13:13:47] Epoch 2 | Step 19840 | Loss: 0.2396 | LM: 0.2280 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 13:13:53] Epoch 2 | Step 19850 | Loss: 0.2396 | LM: 0.2280 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 13:13:59] Epoch 2 | Step 19860 | Loss: 0.2395 | LM: 0.2280 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 13:14:06] Epoch 2 | Step 19870 | Loss: 0.2395 | LM: 0.2279 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 13:14:12] Epoch 2 | Step 19880 | Loss: 0.2395 | LM: 0.2279 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 13:14:18] Epoch 2 | Step 19890 | Loss: 0.2395 | LM: 0.2280 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 13:14:25] Epoch 2 | Step 19900 | Loss: 0.2395 | LM: 0.2279 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 13:14:31] Epoch 2 | Step 19910 | Loss: 0.2395 | LM: 0.2279 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 13:14:37] Epoch 2 | Step 19920 | Loss: 0.2395 | LM: 0.2279 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 13:14:44] Epoch 2 | Step 19930 | Loss: 0.2395 | LM: 0.2279 | LB: 1.0878 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 13:14:50] Epoch 2 | Step 19940 | Loss: 0.2395 | LM: 0.2280 | LB: 1.0878 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 13:14:57] Epoch 2 | Step 19950 | Loss: 0.2396 | LM: 0.2279 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 13:15:03] Epoch 2 | Step 19960 | Loss: 0.2396 | LM: 0.2280 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 13:15:09] Epoch 2 | Step 19970 | Loss: 0.2396 | LM: 0.2280 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 13:15:16] Epoch 2 | Step 19980 | Loss: 0.2396 | LM: 0.2281 | LB: 1.0878 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 13:15:22] Epoch 2 | Step 19990 | Loss: 0.2396 | LM: 0.2281 | LB: 1.0878 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 13:15:29] Epoch 2 | Step 20000 | Loss: 0.2396 | LM: 0.2280 | LB: 1.0878 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 13:15:30] Validation | Batch 10/784 | Loss: 0.3253 | LM_LOSS: 0.3145 | LB_LOSS: 1.0845 [2026-04-17 13:15:31] Validation | Batch 20/784 | Loss: 0.3360 | LM_LOSS: 0.3252 | LB_LOSS: 1.0847 [2026-04-17 13:15:33] Validation | Batch 30/784 | Loss: 0.3217 | LM_LOSS: 0.3108 | LB_LOSS: 1.0840 [2026-04-17 13:15:34] Validation | Batch 40/784 | Loss: 0.3241 | LM_LOSS: 0.3132 | LB_LOSS: 1.0839 [2026-04-17 13:15:35] Validation | Batch 50/784 | Loss: 0.3212 | LM_LOSS: 0.3104 | LB_LOSS: 1.0833 [2026-04-17 13:15:37] Validation | Batch 60/784 | Loss: 0.3230 | LM_LOSS: 0.3121 | LB_LOSS: 1.0828 [2026-04-17 13:15:38] Validation | Batch 70/784 | Loss: 0.3203 | LM_LOSS: 0.3094 | LB_LOSS: 1.0822 [2026-04-17 13:15:39] Validation | Batch 80/784 | Loss: 0.3165 | LM_LOSS: 0.3057 | LB_LOSS: 1.0817 [2026-04-17 13:15:41] Validation | Batch 90/784 | Loss: 0.3154 | LM_LOSS: 0.3046 | LB_LOSS: 1.0823 [2026-04-17 13:15:42] Validation | Batch 100/784 | Loss: 0.3172 | LM_LOSS: 0.3064 | LB_LOSS: 1.0827 [2026-04-17 13:15:43] Validation | Batch 110/784 | Loss: 0.3120 | LM_LOSS: 0.3012 | LB_LOSS: 1.0829 [2026-04-17 13:15:45] Validation | Batch 120/784 | Loss: 0.3156 | LM_LOSS: 0.3048 | LB_LOSS: 1.0828 [2026-04-17 13:15:46] Validation | Batch 130/784 | Loss: 0.3186 | LM_LOSS: 0.3078 | LB_LOSS: 1.0827 [2026-04-17 13:15:47] Validation | Batch 140/784 | Loss: 0.3179 | LM_LOSS: 0.3071 | LB_LOSS: 1.0826 [2026-04-17 13:15:49] Validation | Batch 150/784 | Loss: 0.3140 | LM_LOSS: 0.3031 | LB_LOSS: 1.0829 [2026-04-17 13:15:50] Validation | Batch 160/784 | Loss: 0.3148 | LM_LOSS: 0.3040 | LB_LOSS: 1.0826 [2026-04-17 13:15:52] Validation | Batch 170/784 | Loss: 0.3150 | LM_LOSS: 0.3042 | LB_LOSS: 1.0823 [2026-04-17 13:15:53] Validation | Batch 180/784 | Loss: 0.3126 | LM_LOSS: 0.3018 | LB_LOSS: 1.0823 [2026-04-17 13:15:54] Validation | Batch 190/784 | Loss: 0.3146 | LM_LOSS: 0.3038 | LB_LOSS: 1.0827 [2026-04-17 13:15:56] Validation | Batch 200/784 | Loss: 0.3150 | LM_LOSS: 0.3042 | LB_LOSS: 1.0828 [2026-04-17 13:15:57] Validation | Batch 210/784 | Loss: 0.3139 | LM_LOSS: 0.3031 | LB_LOSS: 1.0827 [2026-04-17 13:15:58] Validation | Batch 220/784 | Loss: 0.3147 | LM_LOSS: 0.3039 | LB_LOSS: 1.0827 [2026-04-17 13:16:00] Validation | Batch 230/784 | Loss: 0.3153 | LM_LOSS: 0.3045 | LB_LOSS: 1.0826 [2026-04-17 13:16:01] Validation | Batch 240/784 | Loss: 0.3157 | LM_LOSS: 0.3049 | LB_LOSS: 1.0830 [2026-04-17 13:16:03] Validation | Batch 250/784 | Loss: 0.3156 | LM_LOSS: 0.3048 | LB_LOSS: 1.0828 [2026-04-17 13:16:04] Validation | Batch 260/784 | Loss: 0.3158 | LM_LOSS: 0.3049 | LB_LOSS: 1.0830 [2026-04-17 13:16:06] Validation | Batch 270/784 | Loss: 0.3156 | LM_LOSS: 0.3048 | LB_LOSS: 1.0831 [2026-04-17 13:16:07] Validation | Batch 280/784 | Loss: 0.3160 | LM_LOSS: 0.3052 | LB_LOSS: 1.0833 [2026-04-17 13:16:08] Validation | Batch 290/784 | Loss: 0.3171 | LM_LOSS: 0.3063 | LB_LOSS: 1.0834 [2026-04-17 13:16:10] Validation | Batch 300/784 | Loss: 0.3179 | LM_LOSS: 0.3070 | LB_LOSS: 1.0834 [2026-04-17 13:16:11] Validation | Batch 310/784 | Loss: 0.3173 | LM_LOSS: 0.3065 | LB_LOSS: 1.0833 [2026-04-17 13:16:13] Validation | Batch 320/784 | Loss: 0.3189 | LM_LOSS: 0.3080 | LB_LOSS: 1.0833 [2026-04-17 13:16:14] Validation | Batch 330/784 | Loss: 0.3187 | LM_LOSS: 0.3079 | LB_LOSS: 1.0833 [2026-04-17 13:16:15] Validation | Batch 340/784 | Loss: 0.3176 | LM_LOSS: 0.3067 | LB_LOSS: 1.0834 [2026-04-17 13:16:17] Validation | Batch 350/784 | Loss: 0.3177 | LM_LOSS: 0.3069 | LB_LOSS: 1.0836 [2026-04-17 13:16:18] Validation | Batch 360/784 | Loss: 0.3175 | LM_LOSS: 0.3066 | LB_LOSS: 1.0836 [2026-04-17 13:16:19] Validation | Batch 370/784 | Loss: 0.3180 | LM_LOSS: 0.3071 | LB_LOSS: 1.0835 [2026-04-17 13:16:20] Validation | Batch 380/784 | Loss: 0.3178 | LM_LOSS: 0.3069 | LB_LOSS: 1.0836 [2026-04-17 13:16:22] Validation | Batch 390/784 | Loss: 0.3177 | LM_LOSS: 0.3068 | LB_LOSS: 1.0837 [2026-04-17 13:16:23] Validation | Batch 400/784 | Loss: 0.3179 | LM_LOSS: 0.3071 | LB_LOSS: 1.0837 [2026-04-17 13:16:24] Validation | Batch 410/784 | Loss: 0.3182 | LM_LOSS: 0.3074 | LB_LOSS: 1.0837 [2026-04-17 13:16:25] Validation | Batch 420/784 | Loss: 0.3185 | LM_LOSS: 0.3076 | LB_LOSS: 1.0837 [2026-04-17 13:16:27] Validation | Batch 430/784 | Loss: 0.3185 | LM_LOSS: 0.3077 | LB_LOSS: 1.0837 [2026-04-17 13:16:28] Validation | Batch 440/784 | Loss: 0.3182 | LM_LOSS: 0.3073 | LB_LOSS: 1.0837 [2026-04-17 13:16:29] Validation | Batch 450/784 | Loss: 0.3174 | LM_LOSS: 0.3066 | LB_LOSS: 1.0836 [2026-04-17 13:16:31] Validation | Batch 460/784 | Loss: 0.3179 | LM_LOSS: 0.3071 | LB_LOSS: 1.0837 [2026-04-17 13:16:32] Validation | Batch 470/784 | Loss: 0.3171 | LM_LOSS: 0.3063 | LB_LOSS: 1.0837 [2026-04-17 13:16:33] Validation | Batch 480/784 | Loss: 0.3176 | LM_LOSS: 0.3067 | LB_LOSS: 1.0837 [2026-04-17 13:16:35] Validation | Batch 490/784 | Loss: 0.3169 | LM_LOSS: 0.3061 | LB_LOSS: 1.0836 [2026-04-17 13:16:36] Validation | Batch 500/784 | Loss: 0.3173 | LM_LOSS: 0.3065 | LB_LOSS: 1.0835 [2026-04-17 13:16:37] Validation | Batch 510/784 | Loss: 0.3170 | LM_LOSS: 0.3062 | LB_LOSS: 1.0835 [2026-04-17 13:16:39] Validation | Batch 520/784 | Loss: 0.3172 | LM_LOSS: 0.3064 | LB_LOSS: 1.0834 [2026-04-17 13:16:40] Validation | Batch 530/784 | Loss: 0.3180 | LM_LOSS: 0.3072 | LB_LOSS: 1.0834 [2026-04-17 13:16:41] Validation | Batch 540/784 | Loss: 0.3184 | LM_LOSS: 0.3076 | LB_LOSS: 1.0834 [2026-04-17 13:16:43] Validation | Batch 550/784 | Loss: 0.3197 | LM_LOSS: 0.3088 | LB_LOSS: 1.0834 [2026-04-17 13:16:44] Validation | Batch 560/784 | Loss: 0.3198 | LM_LOSS: 0.3089 | LB_LOSS: 1.0834 [2026-04-17 13:16:46] Validation | Batch 570/784 | Loss: 0.3193 | LM_LOSS: 0.3085 | LB_LOSS: 1.0833 [2026-04-17 13:16:47] Validation | Batch 580/784 | Loss: 0.3188 | LM_LOSS: 0.3080 | LB_LOSS: 1.0834 [2026-04-17 13:16:48] Validation | Batch 590/784 | Loss: 0.3190 | LM_LOSS: 0.3082 | LB_LOSS: 1.0833 [2026-04-17 13:16:50] Validation | Batch 600/784 | Loss: 0.3189 | LM_LOSS: 0.3081 | LB_LOSS: 1.0832 [2026-04-17 13:16:51] Validation | Batch 610/784 | Loss: 0.3190 | LM_LOSS: 0.3082 | LB_LOSS: 1.0832 [2026-04-17 13:16:52] Validation | Batch 620/784 | Loss: 0.3189 | LM_LOSS: 0.3080 | LB_LOSS: 1.0832 [2026-04-17 13:16:54] Validation | Batch 630/784 | Loss: 0.3196 | LM_LOSS: 0.3088 | LB_LOSS: 1.0833 [2026-04-17 13:16:55] Validation | Batch 640/784 | Loss: 0.3197 | LM_LOSS: 0.3088 | LB_LOSS: 1.0832 [2026-04-17 13:16:57] Validation | Batch 650/784 | Loss: 0.3195 | LM_LOSS: 0.3087 | LB_LOSS: 1.0833 [2026-04-17 13:16:58] Validation | Batch 660/784 | Loss: 0.3199 | LM_LOSS: 0.3091 | LB_LOSS: 1.0833 [2026-04-17 13:17:00] Validation | Batch 670/784 | Loss: 0.3203 | LM_LOSS: 0.3095 | LB_LOSS: 1.0834 [2026-04-17 13:17:01] Validation | Batch 680/784 | Loss: 0.3200 | LM_LOSS: 0.3092 | LB_LOSS: 1.0834 [2026-04-17 13:17:03] Validation | Batch 690/784 | Loss: 0.3202 | LM_LOSS: 0.3093 | LB_LOSS: 1.0833 [2026-04-17 13:17:04] Validation | Batch 700/784 | Loss: 0.3202 | LM_LOSS: 0.3094 | LB_LOSS: 1.0832 [2026-04-17 13:17:05] Validation | Batch 710/784 | Loss: 0.3200 | LM_LOSS: 0.3092 | LB_LOSS: 1.0832 [2026-04-17 13:17:07] Validation | Batch 720/784 | Loss: 0.3197 | LM_LOSS: 0.3089 | LB_LOSS: 1.0831 [2026-04-17 13:17:08] Validation | Batch 730/784 | Loss: 0.3192 | LM_LOSS: 0.3084 | LB_LOSS: 1.0831 [2026-04-17 13:17:09] Validation | Batch 740/784 | Loss: 0.3193 | LM_LOSS: 0.3085 | LB_LOSS: 1.0832 [2026-04-17 13:17:10] Validation | Batch 750/784 | Loss: 0.3186 | LM_LOSS: 0.3078 | LB_LOSS: 1.0832 [2026-04-17 13:17:12] Validation | Batch 760/784 | Loss: 0.3188 | LM_LOSS: 0.3080 | LB_LOSS: 1.0832 [2026-04-17 13:17:13] Validation | Batch 770/784 | Loss: 0.3190 | LM_LOSS: 0.3081 | LB_LOSS: 1.0832 [2026-04-17 13:17:15] Validation | Batch 780/784 | Loss: 0.3193 | LM_LOSS: 0.3085 | LB_LOSS: 1.0832 [2026-04-17 13:17:15] Validation | Batch 784/784 | Loss: 0.3195 | LM_LOSS: 0.3087 | LB_LOSS: 1.0832 [2026-04-17 13:17:18] Validation | Loss: 0.3195 | LM_LOSS: 0.3087 | LB_LOSS: 1.0832 | PPL: 1.36 | Time: 106.63s [2026-04-17 13:17:25] Epoch 2 | Step 20010 | Loss: 0.2396 | LM: 0.2280 | LB: 1.0878 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 13:17:31] Epoch 2 | Step 20020 | Loss: 0.2396 | LM: 0.2279 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 13:17:37] Epoch 2 | Step 20030 | Loss: 0.2396 | LM: 0.2280 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 13:17:44] Epoch 2 | Step 20040 | Loss: 0.2396 | LM: 0.2280 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 13:17:50] Epoch 2 | Step 20050 | Loss: 0.2396 | LM: 0.2280 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 13:17:56] Epoch 2 | Step 20060 | Loss: 0.2396 | LM: 0.2279 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 13:18:03] Epoch 2 | Step 20070 | Loss: 0.2396 | LM: 0.2280 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 13:18:09] Epoch 2 | Step 20080 | Loss: 0.2395 | LM: 0.2279 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 13:18:16] Epoch 2 | Step 20090 | Loss: 0.2395 | LM: 0.2280 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 13:18:22] Epoch 2 | Step 20100 | Loss: 0.2395 | LM: 0.2280 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 13:18:28] Epoch 2 | Step 20110 | Loss: 0.2395 | LM: 0.2279 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 13:18:35] Epoch 2 | Step 20120 | Loss: 0.2395 | LM: 0.2279 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 13:18:41] Epoch 2 | Step 20130 | Loss: 0.2395 | LM: 0.2279 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 13:18:48] Epoch 2 | Step 20140 | Loss: 0.2395 | LM: 0.2278 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 13:18:54] Epoch 2 | Step 20150 | Loss: 0.2395 | LM: 0.2279 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 13:19:00] Epoch 2 | Step 20160 | Loss: 0.2395 | LM: 0.2279 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 13:19:06] Epoch 2 | Step 20170 | Loss: 0.2395 | LM: 0.2279 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 13:19:13] Epoch 2 | Step 20180 | Loss: 0.2395 | LM: 0.2279 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 13:19:19] Epoch 2 | Step 20190 | Loss: 0.2395 | LM: 0.2280 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 13:19:26] Epoch 2 | Step 20200 | Loss: 0.2395 | LM: 0.2280 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 13:19:32] Epoch 2 | Step 20210 | Loss: 0.2395 | LM: 0.2280 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 13:19:39] Epoch 2 | Step 20220 | Loss: 0.2395 | LM: 0.2281 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 13:19:45] Epoch 2 | Step 20230 | Loss: 0.2396 | LM: 0.2281 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 13:19:51] Epoch 2 | Step 20240 | Loss: 0.2395 | LM: 0.2280 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 13:19:58] Epoch 2 | Step 20250 | Loss: 0.2395 | LM: 0.2281 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 13:20:04] Epoch 2 | Step 20260 | Loss: 0.2395 | LM: 0.2281 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 13:20:11] Epoch 2 | Step 20270 | Loss: 0.2395 | LM: 0.2281 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 13:20:16] Epoch 2 | Step 20280 | Loss: 0.2396 | LM: 0.2281 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 13:20:23] Epoch 2 | Step 20290 | Loss: 0.2395 | LM: 0.2281 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 13:20:29] Epoch 2 | Step 20300 | Loss: 0.2395 | LM: 0.2281 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 13:20:35] Epoch 2 | Step 20310 | Loss: 0.2395 | LM: 0.2281 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 13:20:42] Epoch 2 | Step 20320 | Loss: 0.2395 | LM: 0.2280 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 13:20:48] Epoch 2 | Step 20330 | Loss: 0.2395 | LM: 0.2280 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 13:20:55] Epoch 2 | Step 20340 | Loss: 0.2396 | LM: 0.2281 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 13:21:01] Epoch 2 | Step 20350 | Loss: 0.2395 | LM: 0.2281 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 13:21:07] Epoch 2 | Step 20360 | Loss: 0.2395 | LM: 0.2280 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 13:21:14] Epoch 2 | Step 20370 | Loss: 0.2395 | LM: 0.2281 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 13:21:20] Epoch 2 | Step 20380 | Loss: 0.2395 | LM: 0.2280 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 13:21:27] Epoch 2 | Step 20390 | Loss: 0.2395 | LM: 0.2280 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 13:21:33] Epoch 2 | Step 20400 | Loss: 0.2395 | LM: 0.2280 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 13:21:40] Epoch 2 | Step 20410 | Loss: 0.2395 | LM: 0.2281 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 13:21:46] Epoch 2 | Step 20420 | Loss: 0.2395 | LM: 0.2281 | LB: 1.0878 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 13:21:53] Epoch 2 | Step 20430 | Loss: 0.2395 | LM: 0.2281 | LB: 1.0878 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 13:21:59] Epoch 2 | Step 20440 | Loss: 0.2395 | LM: 0.2282 | LB: 1.0878 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 13:22:06] Epoch 2 | Step 20450 | Loss: 0.2395 | LM: 0.2282 | LB: 1.0878 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 13:22:12] Epoch 2 | Step 20460 | Loss: 0.2395 | LM: 0.2282 | LB: 1.0878 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 13:22:19] Epoch 2 | Step 20470 | Loss: 0.2395 | LM: 0.2281 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 13:22:25] Epoch 2 | Step 20480 | Loss: 0.2395 | LM: 0.2282 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 13:22:32] Epoch 2 | Step 20490 | Loss: 0.2395 | LM: 0.2281 | LB: 1.0878 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 13:22:38] Epoch 2 | Step 20500 | Loss: 0.2395 | LM: 0.2281 | LB: 1.0878 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 13:22:45] Epoch 2 | Step 20510 | Loss: 0.2395 | LM: 0.2282 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 13:22:51] Epoch 2 | Step 20520 | Loss: 0.2395 | LM: 0.2283 | LB: 1.0878 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 13:22:58] Epoch 2 | Step 20530 | Loss: 0.2395 | LM: 0.2282 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 13:23:04] Epoch 2 | Step 20540 | Loss: 0.2395 | LM: 0.2282 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 13:23:11] Epoch 2 | Step 20550 | Loss: 0.2395 | LM: 0.2282 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 13:23:17] Epoch 2 | Step 20560 | Loss: 0.2395 | LM: 0.2282 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 13:23:23] Epoch 2 | Step 20570 | Loss: 0.2395 | LM: 0.2283 | LB: 1.0878 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 13:23:30] Epoch 2 | Step 20580 | Loss: 0.2395 | LM: 0.2284 | LB: 1.0878 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 13:23:36] Epoch 2 | Step 20590 | Loss: 0.2395 | LM: 0.2284 | LB: 1.0878 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 13:23:43] Epoch 2 | Step 20600 | Loss: 0.2396 | LM: 0.2284 | LB: 1.0878 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 13:23:49] Epoch 2 | Step 20610 | Loss: 0.2395 | LM: 0.2284 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 13:23:55] Epoch 2 | Step 20620 | Loss: 0.2395 | LM: 0.2284 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 13:24:02] Epoch 2 | Step 20630 | Loss: 0.2395 | LM: 0.2284 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 13:24:08] Epoch 2 | Step 20640 | Loss: 0.2395 | LM: 0.2285 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 13:24:14] Epoch 2 | Step 20650 | Loss: 0.2396 | LM: 0.2286 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 13:24:21] Epoch 2 | Step 20660 | Loss: 0.2396 | LM: 0.2286 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 13:24:27] Epoch 2 | Step 20670 | Loss: 0.2395 | LM: 0.2286 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 13:24:34] Epoch 2 | Step 20680 | Loss: 0.2395 | LM: 0.2285 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 13:24:41] Epoch 2 | Step 20690 | Loss: 0.2395 | LM: 0.2285 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 13:24:47] Epoch 2 | Step 20700 | Loss: 0.2396 | LM: 0.2284 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 13:24:53] Epoch 2 | Step 20710 | Loss: 0.2395 | LM: 0.2283 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 13:25:00] Epoch 2 | Step 20720 | Loss: 0.2395 | LM: 0.2284 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 13:25:06] Epoch 2 | Step 20730 | Loss: 0.2395 | LM: 0.2283 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 13:25:12] Epoch 2 | Step 20740 | Loss: 0.2395 | LM: 0.2283 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 13:25:19] Epoch 2 | Step 20750 | Loss: 0.2396 | LM: 0.2282 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 13:25:25] Epoch 2 | Step 20760 | Loss: 0.2395 | LM: 0.2283 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 13:25:32] Epoch 2 | Step 20770 | Loss: 0.2395 | LM: 0.2283 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 13:25:38] Epoch 2 | Step 20780 | Loss: 0.2395 | LM: 0.2283 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 13:25:45] Epoch 2 | Step 20790 | Loss: 0.2395 | LM: 0.2282 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 13:25:51] Epoch 2 | Step 20800 | Loss: 0.2395 | LM: 0.2283 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 13:25:58] Epoch 2 | Step 20810 | Loss: 0.2395 | LM: 0.2283 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 13:26:04] Epoch 2 | Step 20820 | Loss: 0.2395 | LM: 0.2283 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 13:26:10] Epoch 2 | Step 20830 | Loss: 0.2395 | LM: 0.2283 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 13:26:16] Epoch 2 | Step 20840 | Loss: 0.2395 | LM: 0.2283 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 13:26:23] Epoch 2 | Step 20850 | Loss: 0.2395 | LM: 0.2283 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 13:26:29] Epoch 2 | Step 20860 | Loss: 0.2394 | LM: 0.2283 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 13:26:35] Epoch 2 | Step 20870 | Loss: 0.2394 | LM: 0.2283 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 13:26:41] Epoch 2 | Step 20880 | Loss: 0.2394 | LM: 0.2283 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 13:26:48] Epoch 2 | Step 20890 | Loss: 0.2395 | LM: 0.2283 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 13:26:54] Epoch 2 | Step 20900 | Loss: 0.2394 | LM: 0.2284 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 13:27:00] Epoch 2 | Step 20910 | Loss: 0.2394 | LM: 0.2283 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 13:27:07] Epoch 2 | Step 20920 | Loss: 0.2394 | LM: 0.2283 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 13:27:13] Epoch 2 | Step 20930 | Loss: 0.2394 | LM: 0.2283 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 13:27:19] Epoch 2 | Step 20940 | Loss: 0.2394 | LM: 0.2284 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 13:27:26] Epoch 2 | Step 20950 | Loss: 0.2394 | LM: 0.2284 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 13:27:32] Epoch 2 | Step 20960 | Loss: 0.2394 | LM: 0.2284 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 13:27:39] Epoch 2 | Step 20970 | Loss: 0.2394 | LM: 0.2284 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 13:27:45] Epoch 2 | Step 20980 | Loss: 0.2394 | LM: 0.2284 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 13:27:51] Epoch 2 | Step 20990 | Loss: 0.2394 | LM: 0.2284 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 13:27:57] Epoch 2 | Step 21000 | Loss: 0.2394 | LM: 0.2284 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 13:28:06] Checkpoint saved: outputs/2026-04-17/08-57-56/checkpoints/checkpoint_step_21000.pt [2026-04-17 13:28:22] Validation | Batch 10/784 | Loss: 0.3257 | LM_LOSS: 0.3149 | LB_LOSS: 1.0846 [2026-04-17 13:28:24] Validation | Batch 20/784 | Loss: 0.3363 | LM_LOSS: 0.3255 | LB_LOSS: 1.0848 [2026-04-17 13:28:25] Validation | Batch 30/784 | Loss: 0.3220 | LM_LOSS: 0.3111 | LB_LOSS: 1.0841 [2026-04-17 13:28:27] Validation | Batch 40/784 | Loss: 0.3244 | LM_LOSS: 0.3136 | LB_LOSS: 1.0840 [2026-04-17 13:28:28] Validation | Batch 50/784 | Loss: 0.3214 | LM_LOSS: 0.3106 | LB_LOSS: 1.0833 [2026-04-17 13:28:30] Validation | Batch 60/784 | Loss: 0.3231 | LM_LOSS: 0.3123 | LB_LOSS: 1.0829 [2026-04-17 13:28:31] Validation | Batch 70/784 | Loss: 0.3204 | LM_LOSS: 0.3095 | LB_LOSS: 1.0823 [2026-04-17 13:28:32] Validation | Batch 80/784 | Loss: 0.3166 | LM_LOSS: 0.3058 | LB_LOSS: 1.0818 [2026-04-17 13:28:33] Validation | Batch 90/784 | Loss: 0.3155 | LM_LOSS: 0.3047 | LB_LOSS: 1.0824 [2026-04-17 13:28:35] Validation | Batch 100/784 | Loss: 0.3173 | LM_LOSS: 0.3065 | LB_LOSS: 1.0828 [2026-04-17 13:28:36] Validation | Batch 110/784 | Loss: 0.3122 | LM_LOSS: 0.3013 | LB_LOSS: 1.0829 [2026-04-17 13:28:37] Validation | Batch 120/784 | Loss: 0.3157 | LM_LOSS: 0.3049 | LB_LOSS: 1.0829 [2026-04-17 13:28:39] Validation | Batch 130/784 | Loss: 0.3187 | LM_LOSS: 0.3079 | LB_LOSS: 1.0828 [2026-04-17 13:28:40] Validation | Batch 140/784 | Loss: 0.3180 | LM_LOSS: 0.3072 | LB_LOSS: 1.0826 [2026-04-17 13:28:42] Validation | Batch 150/784 | Loss: 0.3140 | LM_LOSS: 0.3032 | LB_LOSS: 1.0829 [2026-04-17 13:28:43] Validation | Batch 160/784 | Loss: 0.3148 | LM_LOSS: 0.3040 | LB_LOSS: 1.0826 [2026-04-17 13:28:45] Validation | Batch 170/784 | Loss: 0.3150 | LM_LOSS: 0.3042 | LB_LOSS: 1.0823 [2026-04-17 13:28:46] Validation | Batch 180/784 | Loss: 0.3126 | LM_LOSS: 0.3018 | LB_LOSS: 1.0823 [2026-04-17 13:28:47] Validation | Batch 190/784 | Loss: 0.3147 | LM_LOSS: 0.3039 | LB_LOSS: 1.0828 [2026-04-17 13:28:48] Validation | Batch 200/784 | Loss: 0.3151 | LM_LOSS: 0.3042 | LB_LOSS: 1.0828 [2026-04-17 13:28:49] Validation | Batch 210/784 | Loss: 0.3140 | LM_LOSS: 0.3031 | LB_LOSS: 1.0828 [2026-04-17 13:28:51] Validation | Batch 220/784 | Loss: 0.3147 | LM_LOSS: 0.3039 | LB_LOSS: 1.0828 [2026-04-17 13:28:52] Validation | Batch 230/784 | Loss: 0.3153 | LM_LOSS: 0.3045 | LB_LOSS: 1.0827 [2026-04-17 13:28:54] Validation | Batch 240/784 | Loss: 0.3157 | LM_LOSS: 0.3049 | LB_LOSS: 1.0831 [2026-04-17 13:28:55] Validation | Batch 250/784 | Loss: 0.3156 | LM_LOSS: 0.3048 | LB_LOSS: 1.0829 [2026-04-17 13:28:56] Validation | Batch 260/784 | Loss: 0.3158 | LM_LOSS: 0.3050 | LB_LOSS: 1.0831 [2026-04-17 13:28:58] Validation | Batch 270/784 | Loss: 0.3156 | LM_LOSS: 0.3048 | LB_LOSS: 1.0832 [2026-04-17 13:28:59] Validation | Batch 280/784 | Loss: 0.3161 | LM_LOSS: 0.3052 | LB_LOSS: 1.0833 [2026-04-17 13:29:01] Validation | Batch 290/784 | Loss: 0.3171 | LM_LOSS: 0.3063 | LB_LOSS: 1.0834 [2026-04-17 13:29:02] Validation | Batch 300/784 | Loss: 0.3179 | LM_LOSS: 0.3071 | LB_LOSS: 1.0835 [2026-04-17 13:29:03] Validation | Batch 310/784 | Loss: 0.3173 | LM_LOSS: 0.3065 | LB_LOSS: 1.0834 [2026-04-17 13:29:05] Validation | Batch 320/784 | Loss: 0.3189 | LM_LOSS: 0.3081 | LB_LOSS: 1.0834 [2026-04-17 13:29:06] Validation | Batch 330/784 | Loss: 0.3187 | LM_LOSS: 0.3079 | LB_LOSS: 1.0834 [2026-04-17 13:29:07] Validation | Batch 340/784 | Loss: 0.3176 | LM_LOSS: 0.3067 | LB_LOSS: 1.0835 [2026-04-17 13:29:09] Validation | Batch 350/784 | Loss: 0.3177 | LM_LOSS: 0.3069 | LB_LOSS: 1.0837 [2026-04-17 13:29:10] Validation | Batch 360/784 | Loss: 0.3175 | LM_LOSS: 0.3066 | LB_LOSS: 1.0837 [2026-04-17 13:29:11] Validation | Batch 370/784 | Loss: 0.3179 | LM_LOSS: 0.3071 | LB_LOSS: 1.0836 [2026-04-17 13:29:12] Validation | Batch 380/784 | Loss: 0.3178 | LM_LOSS: 0.3069 | LB_LOSS: 1.0837 [2026-04-17 13:29:14] Validation | Batch 390/784 | Loss: 0.3177 | LM_LOSS: 0.3068 | LB_LOSS: 1.0837 [2026-04-17 13:29:15] Validation | Batch 400/784 | Loss: 0.3179 | LM_LOSS: 0.3071 | LB_LOSS: 1.0837 [2026-04-17 13:29:16] Validation | Batch 410/784 | Loss: 0.3182 | LM_LOSS: 0.3074 | LB_LOSS: 1.0837 [2026-04-17 13:29:17] Validation | Batch 420/784 | Loss: 0.3185 | LM_LOSS: 0.3076 | LB_LOSS: 1.0838 [2026-04-17 13:29:19] Validation | Batch 430/784 | Loss: 0.3185 | LM_LOSS: 0.3077 | LB_LOSS: 1.0837 [2026-04-17 13:29:21] Validation | Batch 440/784 | Loss: 0.3182 | LM_LOSS: 0.3073 | LB_LOSS: 1.0838 [2026-04-17 13:29:22] Validation | Batch 450/784 | Loss: 0.3175 | LM_LOSS: 0.3066 | LB_LOSS: 1.0837 [2026-04-17 13:29:24] Validation | Batch 460/784 | Loss: 0.3180 | LM_LOSS: 0.3071 | LB_LOSS: 1.0838 [2026-04-17 13:29:25] Validation | Batch 470/784 | Loss: 0.3172 | LM_LOSS: 0.3063 | LB_LOSS: 1.0838 [2026-04-17 13:29:26] Validation | Batch 480/784 | Loss: 0.3176 | LM_LOSS: 0.3068 | LB_LOSS: 1.0837 [2026-04-17 13:29:28] Validation | Batch 490/784 | Loss: 0.3170 | LM_LOSS: 0.3061 | LB_LOSS: 1.0837 [2026-04-17 13:29:29] Validation | Batch 500/784 | Loss: 0.3173 | LM_LOSS: 0.3065 | LB_LOSS: 1.0836 [2026-04-17 13:29:30] Validation | Batch 510/784 | Loss: 0.3170 | LM_LOSS: 0.3062 | LB_LOSS: 1.0836 [2026-04-17 13:29:32] Validation | Batch 520/784 | Loss: 0.3172 | LM_LOSS: 0.3064 | LB_LOSS: 1.0835 [2026-04-17 13:29:33] Validation | Batch 530/784 | Loss: 0.3181 | LM_LOSS: 0.3072 | LB_LOSS: 1.0835 [2026-04-17 13:29:34] Validation | Batch 540/784 | Loss: 0.3184 | LM_LOSS: 0.3076 | LB_LOSS: 1.0835 [2026-04-17 13:29:36] Validation | Batch 550/784 | Loss: 0.3197 | LM_LOSS: 0.3088 | LB_LOSS: 1.0834 [2026-04-17 13:29:37] Validation | Batch 560/784 | Loss: 0.3198 | LM_LOSS: 0.3089 | LB_LOSS: 1.0835 [2026-04-17 13:29:39] Validation | Batch 570/784 | Loss: 0.3193 | LM_LOSS: 0.3085 | LB_LOSS: 1.0834 [2026-04-17 13:29:40] Validation | Batch 580/784 | Loss: 0.3188 | LM_LOSS: 0.3080 | LB_LOSS: 1.0834 [2026-04-17 13:29:41] Validation | Batch 590/784 | Loss: 0.3191 | LM_LOSS: 0.3082 | LB_LOSS: 1.0834 [2026-04-17 13:29:43] Validation | Batch 600/784 | Loss: 0.3189 | LM_LOSS: 0.3081 | LB_LOSS: 1.0833 [2026-04-17 13:29:44] Validation | Batch 610/784 | Loss: 0.3190 | LM_LOSS: 0.3082 | LB_LOSS: 1.0833 [2026-04-17 13:29:46] Validation | Batch 620/784 | Loss: 0.3189 | LM_LOSS: 0.3081 | LB_LOSS: 1.0833 [2026-04-17 13:29:47] Validation | Batch 630/784 | Loss: 0.3196 | LM_LOSS: 0.3088 | LB_LOSS: 1.0833 [2026-04-17 13:29:49] Validation | Batch 640/784 | Loss: 0.3197 | LM_LOSS: 0.3089 | LB_LOSS: 1.0833 [2026-04-17 13:29:50] Validation | Batch 650/784 | Loss: 0.3196 | LM_LOSS: 0.3087 | LB_LOSS: 1.0834 [2026-04-17 13:29:52] Validation | Batch 660/784 | Loss: 0.3199 | LM_LOSS: 0.3091 | LB_LOSS: 1.0834 [2026-04-17 13:29:53] Validation | Batch 670/784 | Loss: 0.3203 | LM_LOSS: 0.3095 | LB_LOSS: 1.0834 [2026-04-17 13:29:54] Validation | Batch 680/784 | Loss: 0.3200 | LM_LOSS: 0.3092 | LB_LOSS: 1.0834 [2026-04-17 13:29:56] Validation | Batch 690/784 | Loss: 0.3202 | LM_LOSS: 0.3094 | LB_LOSS: 1.0834 [2026-04-17 13:29:57] Validation | Batch 700/784 | Loss: 0.3202 | LM_LOSS: 0.3094 | LB_LOSS: 1.0833 [2026-04-17 13:29:58] Validation | Batch 710/784 | Loss: 0.3200 | LM_LOSS: 0.3092 | LB_LOSS: 1.0833 [2026-04-17 13:30:00] Validation | Batch 720/784 | Loss: 0.3197 | LM_LOSS: 0.3089 | LB_LOSS: 1.0832 [2026-04-17 13:30:01] Validation | Batch 730/784 | Loss: 0.3192 | LM_LOSS: 0.3084 | LB_LOSS: 1.0832 [2026-04-17 13:30:02] Validation | Batch 740/784 | Loss: 0.3193 | LM_LOSS: 0.3085 | LB_LOSS: 1.0832 [2026-04-17 13:30:04] Validation | Batch 750/784 | Loss: 0.3187 | LM_LOSS: 0.3078 | LB_LOSS: 1.0832 [2026-04-17 13:30:05] Validation | Batch 760/784 | Loss: 0.3188 | LM_LOSS: 0.3080 | LB_LOSS: 1.0832 [2026-04-17 13:30:06] Validation | Batch 770/784 | Loss: 0.3190 | LM_LOSS: 0.3082 | LB_LOSS: 1.0833 [2026-04-17 13:30:08] Validation | Batch 780/784 | Loss: 0.3193 | LM_LOSS: 0.3085 | LB_LOSS: 1.0833 [2026-04-17 13:30:08] Validation | Batch 784/784 | Loss: 0.3195 | LM_LOSS: 0.3087 | LB_LOSS: 1.0833 [2026-04-17 13:30:11] Validation | Loss: 0.3195 | LM_LOSS: 0.3087 | LB_LOSS: 1.0833 | PPL: 1.36 | Time: 107.07s [2026-04-17 13:30:17] Epoch 2 | Step 21010 | Loss: 0.2394 | LM: 0.2284 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 13:30:24] Epoch 2 | Step 21020 | Loss: 0.2394 | LM: 0.2284 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 13:30:31] Epoch 2 | Step 21030 | Loss: 0.2394 | LM: 0.2284 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 13:30:38] Epoch 2 | Step 21040 | Loss: 0.2394 | LM: 0.2285 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 13:30:44] Epoch 2 | Step 21050 | Loss: 0.2394 | LM: 0.2285 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 13:30:50] Epoch 2 | Step 21060 | Loss: 0.2394 | LM: 0.2284 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 13:30:57] Epoch 2 | Step 21070 | Loss: 0.2394 | LM: 0.2285 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 13:31:03] Epoch 2 | Step 21080 | Loss: 0.2394 | LM: 0.2286 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 13:31:09] Epoch 2 | Step 21090 | Loss: 0.2394 | LM: 0.2286 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 13:31:16] Epoch 2 | Step 21100 | Loss: 0.2394 | LM: 0.2286 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 13:31:22] Epoch 2 | Step 21110 | Loss: 0.2394 | LM: 0.2288 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 13:31:28] Epoch 2 | Step 21120 | Loss: 0.2394 | LM: 0.2288 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 13:31:35] Epoch 2 | Step 21130 | Loss: 0.2394 | LM: 0.2288 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 13:31:41] Epoch 2 | Step 21140 | Loss: 0.2394 | LM: 0.2287 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 13:31:48] Epoch 2 | Step 21150 | Loss: 0.2394 | LM: 0.2288 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 13:31:54] Epoch 2 | Step 21160 | Loss: 0.2394 | LM: 0.2287 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 13:32:00] Epoch 2 | Step 21170 | Loss: 0.2394 | LM: 0.2288 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 13:32:07] Epoch 2 | Step 21180 | Loss: 0.2394 | LM: 0.2288 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 13:32:08] Epoch 2 completed in 8049.06s | Loss: 0.2394 | CL0: 2.9 | CL1: 2.4 [2026-04-17 13:32:17] Checkpoint saved: outputs/2026-04-17/08-57-56/checkpoints/checkpoint_step_21182.pt [2026-04-17 13:32:32] ============================================================ [2026-04-17 13:32:32] EPOCH 3/3 [2026-04-17 13:32:32] ============================================================ [2026-04-17 13:32:37] Epoch 3 | Step 21190 | Loss: 0.1821 | LM: 0.1604 | LB: 1.0846 | CL0: 2.9 | CL1: 2.4 | HR0: 0.342/SR0: 0.343 | HR1: 0.417/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 13:32:43] Epoch 3 | Step 21200 | Loss: 0.1960 | LM: 0.1804 | LB: 1.0844 | CL0: 2.9 | CL1: 2.5 | HR0: 0.343/SR0: 0.343 | HR1: 0.412/SR1: 0.381 | LR: 1.00e-05 [2026-04-17 13:32:50] Epoch 3 | Step 21210 | Loss: 0.2002 | LM: 0.1857 | LB: 1.0853 | CL0: 2.9 | CL1: 2.4 | HR0: 0.345/SR0: 0.345 | HR1: 0.412/SR1: 0.382 | LR: 1.00e-05 [2026-04-17 13:32:56] Epoch 3 | Step 21220 | Loss: 0.2059 | LM: 0.1942 | LB: 1.0853 | CL0: 2.9 | CL1: 2.4 | HR0: 0.344/SR0: 0.344 | HR1: 0.414/SR1: 0.383 | LR: 1.00e-05 [2026-04-17 13:33:02] Epoch 3 | Step 21230 | Loss: 0.2052 | LM: 0.1862 | LB: 1.0862 | CL0: 2.9 | CL1: 2.4 | HR0: 0.345/SR0: 0.344 | HR1: 0.415/SR1: 0.383 | LR: 1.00e-05 [2026-04-17 13:33:09] Epoch 3 | Step 21240 | Loss: 0.2116 | LM: 0.1919 | LB: 1.0858 | CL0: 2.9 | CL1: 2.4 | HR0: 0.344/SR0: 0.344 | HR1: 0.414/SR1: 0.383 | LR: 1.00e-05 [2026-04-17 13:33:15] Epoch 3 | Step 21250 | Loss: 0.2065 | LM: 0.1907 | LB: 1.0851 | CL0: 2.9 | CL1: 2.4 | HR0: 0.344/SR0: 0.343 | HR1: 0.415/SR1: 0.383 | LR: 1.00e-05 [2026-04-17 13:33:21] Epoch 3 | Step 21260 | Loss: 0.2117 | LM: 0.1991 | LB: 1.0853 | CL0: 2.9 | CL1: 2.4 | HR0: 0.345/SR0: 0.344 | HR1: 0.414/SR1: 0.383 | LR: 1.00e-05 [2026-04-17 13:33:28] Epoch 3 | Step 21270 | Loss: 0.2178 | LM: 0.2075 | LB: 1.0866 | CL0: 2.9 | CL1: 2.4 | HR0: 0.345/SR0: 0.345 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 13:33:34] Epoch 3 | Step 21280 | Loss: 0.2135 | LM: 0.2051 | LB: 1.0878 | CL0: 2.9 | CL1: 2.4 | HR0: 0.346/SR0: 0.345 | HR1: 0.417/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 13:33:40] Epoch 3 | Step 21290 | Loss: 0.2116 | LM: 0.2054 | LB: 1.0878 | CL0: 2.9 | CL1: 2.4 | HR0: 0.345/SR0: 0.345 | HR1: 0.417/SR1: 0.386 | LR: 1.00e-05 [2026-04-17 13:33:47] Epoch 3 | Step 21300 | Loss: 0.2097 | LM: 0.2029 | LB: 1.0880 | CL0: 2.9 | CL1: 2.4 | HR0: 0.345/SR0: 0.345 | HR1: 0.418/SR1: 0.386 | LR: 1.00e-05 [2026-04-17 13:33:53] Epoch 3 | Step 21310 | Loss: 0.2097 | LM: 0.1995 | LB: 1.0872 | CL0: 2.9 | CL1: 2.4 | HR0: 0.346/SR0: 0.345 | HR1: 0.417/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 13:33:59] Epoch 3 | Step 21320 | Loss: 0.2075 | LM: 0.1955 | LB: 1.0873 | CL0: 2.9 | CL1: 2.4 | HR0: 0.345/SR0: 0.345 | HR1: 0.417/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 13:34:06] Epoch 3 | Step 21330 | Loss: 0.2079 | LM: 0.1920 | LB: 1.0874 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.346 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 13:34:12] Epoch 3 | Step 21340 | Loss: 0.2089 | LM: 0.1965 | LB: 1.0881 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 13:34:19] Epoch 3 | Step 21350 | Loss: 0.2091 | LM: 0.1941 | LB: 1.0875 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 13:34:25] Epoch 3 | Step 21360 | Loss: 0.2107 | LM: 0.1920 | LB: 1.0874 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.415/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 13:34:31] Epoch 3 | Step 21370 | Loss: 0.2119 | LM: 0.1902 | LB: 1.0870 | CL0: 2.9 | CL1: 2.4 | HR0: 0.346/SR0: 0.346 | HR1: 0.415/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 13:34:37] Epoch 3 | Step 21380 | Loss: 0.2126 | LM: 0.1914 | LB: 1.0867 | CL0: 2.9 | CL1: 2.4 | HR0: 0.346/SR0: 0.346 | HR1: 0.415/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 13:34:43] Epoch 3 | Step 21390 | Loss: 0.2140 | LM: 0.1952 | LB: 1.0865 | CL0: 2.9 | CL1: 2.4 | HR0: 0.346/SR0: 0.345 | HR1: 0.415/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 13:34:50] Epoch 3 | Step 21400 | Loss: 0.2138 | LM: 0.1949 | LB: 1.0871 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.346 | HR1: 0.415/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 13:34:56] Epoch 3 | Step 21410 | Loss: 0.2130 | LM: 0.1928 | LB: 1.0869 | CL0: 2.9 | CL1: 2.4 | HR0: 0.346/SR0: 0.346 | HR1: 0.415/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 13:35:03] Epoch 3 | Step 21420 | Loss: 0.2128 | LM: 0.1925 | LB: 1.0869 | CL0: 2.9 | CL1: 2.4 | HR0: 0.346/SR0: 0.346 | HR1: 0.415/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 13:35:09] Epoch 3 | Step 21430 | Loss: 0.2133 | LM: 0.1903 | LB: 1.0869 | CL0: 2.9 | CL1: 2.4 | HR0: 0.346/SR0: 0.345 | HR1: 0.415/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 13:35:15] Epoch 3 | Step 21440 | Loss: 0.2142 | LM: 0.1909 | LB: 1.0870 | CL0: 2.9 | CL1: 2.4 | HR0: 0.346/SR0: 0.345 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 13:35:21] Epoch 3 | Step 21450 | Loss: 0.2130 | LM: 0.1893 | LB: 1.0869 | CL0: 2.9 | CL1: 2.4 | HR0: 0.346/SR0: 0.346 | HR1: 0.415/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 13:35:27] Epoch 3 | Step 21460 | Loss: 0.2123 | LM: 0.1874 | LB: 1.0865 | CL0: 2.9 | CL1: 2.4 | HR0: 0.346/SR0: 0.345 | HR1: 0.415/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 13:35:34] Epoch 3 | Step 21470 | Loss: 0.2119 | LM: 0.1883 | LB: 1.0865 | CL0: 2.9 | CL1: 2.4 | HR0: 0.346/SR0: 0.345 | HR1: 0.415/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 13:35:40] Epoch 3 | Step 21480 | Loss: 0.2109 | LM: 0.1884 | LB: 1.0864 | CL0: 2.9 | CL1: 2.4 | HR0: 0.346/SR0: 0.345 | HR1: 0.415/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 13:35:47] Epoch 3 | Step 21490 | Loss: 0.2095 | LM: 0.1873 | LB: 1.0865 | CL0: 2.9 | CL1: 2.4 | HR0: 0.346/SR0: 0.345 | HR1: 0.415/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 13:35:53] Epoch 3 | Step 21500 | Loss: 0.2094 | LM: 0.1882 | LB: 1.0865 | CL0: 2.9 | CL1: 2.4 | HR0: 0.346/SR0: 0.345 | HR1: 0.415/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 13:35:59] Epoch 3 | Step 21510 | Loss: 0.2110 | LM: 0.1906 | LB: 1.0866 | CL0: 2.9 | CL1: 2.4 | HR0: 0.346/SR0: 0.345 | HR1: 0.415/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 13:36:06] Epoch 3 | Step 21520 | Loss: 0.2110 | LM: 0.1910 | LB: 1.0865 | CL0: 2.9 | CL1: 2.4 | HR0: 0.346/SR0: 0.345 | HR1: 0.415/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 13:36:12] Epoch 3 | Step 21530 | Loss: 0.2101 | LM: 0.1912 | LB: 1.0863 | CL0: 2.9 | CL1: 2.4 | HR0: 0.346/SR0: 0.345 | HR1: 0.415/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 13:36:19] Epoch 3 | Step 21540 | Loss: 0.2104 | LM: 0.1902 | LB: 1.0862 | CL0: 2.9 | CL1: 2.4 | HR0: 0.346/SR0: 0.345 | HR1: 0.415/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 13:36:25] Epoch 3 | Step 21550 | Loss: 0.2105 | LM: 0.1905 | LB: 1.0862 | CL0: 2.9 | CL1: 2.4 | HR0: 0.346/SR0: 0.345 | HR1: 0.415/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 13:36:32] Epoch 3 | Step 21560 | Loss: 0.2105 | LM: 0.1908 | LB: 1.0857 | CL0: 2.9 | CL1: 2.4 | HR0: 0.345/SR0: 0.345 | HR1: 0.415/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 13:36:38] Epoch 3 | Step 21570 | Loss: 0.2092 | LM: 0.1899 | LB: 1.0857 | CL0: 2.9 | CL1: 2.4 | HR0: 0.345/SR0: 0.345 | HR1: 0.415/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 13:36:44] Epoch 3 | Step 21580 | Loss: 0.2093 | LM: 0.1895 | LB: 1.0859 | CL0: 2.9 | CL1: 2.4 | HR0: 0.345/SR0: 0.345 | HR1: 0.415/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 13:36:51] Epoch 3 | Step 21590 | Loss: 0.2094 | LM: 0.1894 | LB: 1.0857 | CL0: 2.9 | CL1: 2.4 | HR0: 0.345/SR0: 0.344 | HR1: 0.415/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 13:36:58] Epoch 3 | Step 21600 | Loss: 0.2090 | LM: 0.1891 | LB: 1.0860 | CL0: 2.9 | CL1: 2.4 | HR0: 0.345/SR0: 0.344 | HR1: 0.415/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 13:37:04] Epoch 3 | Step 21610 | Loss: 0.2091 | LM: 0.1905 | LB: 1.0861 | CL0: 2.9 | CL1: 2.4 | HR0: 0.345/SR0: 0.344 | HR1: 0.415/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 13:37:11] Epoch 3 | Step 21620 | Loss: 0.2092 | LM: 0.1906 | LB: 1.0858 | CL0: 2.9 | CL1: 2.4 | HR0: 0.345/SR0: 0.344 | HR1: 0.415/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 13:37:17] Epoch 3 | Step 21630 | Loss: 0.2089 | LM: 0.1891 | LB: 1.0858 | CL0: 2.9 | CL1: 2.4 | HR0: 0.345/SR0: 0.344 | HR1: 0.415/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 13:37:23] Epoch 3 | Step 21640 | Loss: 0.2090 | LM: 0.1888 | LB: 1.0860 | CL0: 2.9 | CL1: 2.4 | HR0: 0.345/SR0: 0.344 | HR1: 0.415/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 13:37:30] Epoch 3 | Step 21650 | Loss: 0.2087 | LM: 0.1897 | LB: 1.0861 | CL0: 2.9 | CL1: 2.4 | HR0: 0.345/SR0: 0.344 | HR1: 0.415/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 13:37:36] Epoch 3 | Step 21660 | Loss: 0.2086 | LM: 0.1896 | LB: 1.0858 | CL0: 2.9 | CL1: 2.4 | HR0: 0.345/SR0: 0.344 | HR1: 0.415/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 13:37:43] Epoch 3 | Step 21670 | Loss: 0.2085 | LM: 0.1909 | LB: 1.0860 | CL0: 2.9 | CL1: 2.4 | HR0: 0.345/SR0: 0.344 | HR1: 0.415/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 13:37:49] Epoch 3 | Step 21680 | Loss: 0.2086 | LM: 0.1915 | LB: 1.0861 | CL0: 2.9 | CL1: 2.4 | HR0: 0.345/SR0: 0.344 | HR1: 0.415/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 13:37:55] Epoch 3 | Step 21690 | Loss: 0.2081 | LM: 0.1908 | LB: 1.0860 | CL0: 2.9 | CL1: 2.4 | HR0: 0.345/SR0: 0.344 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 13:38:02] Epoch 3 | Step 21700 | Loss: 0.2076 | LM: 0.1908 | LB: 1.0859 | CL0: 2.9 | CL1: 2.4 | HR0: 0.345/SR0: 0.344 | HR1: 0.415/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 13:38:08] Epoch 3 | Step 21710 | Loss: 0.2075 | LM: 0.1911 | LB: 1.0860 | CL0: 2.9 | CL1: 2.4 | HR0: 0.345/SR0: 0.344 | HR1: 0.415/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 13:38:14] Epoch 3 | Step 21720 | Loss: 0.2074 | LM: 0.1896 | LB: 1.0859 | CL0: 2.9 | CL1: 2.4 | HR0: 0.345/SR0: 0.344 | HR1: 0.415/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 13:38:20] Epoch 3 | Step 21730 | Loss: 0.2079 | LM: 0.1893 | LB: 1.0859 | CL0: 2.9 | CL1: 2.4 | HR0: 0.345/SR0: 0.344 | HR1: 0.415/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 13:38:27] Epoch 3 | Step 21740 | Loss: 0.2082 | LM: 0.1891 | LB: 1.0860 | CL0: 2.9 | CL1: 2.4 | HR0: 0.345/SR0: 0.344 | HR1: 0.415/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 13:38:34] Epoch 3 | Step 21750 | Loss: 0.2075 | LM: 0.1880 | LB: 1.0859 | CL0: 2.9 | CL1: 2.4 | HR0: 0.345/SR0: 0.344 | HR1: 0.415/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 13:38:40] Epoch 3 | Step 21760 | Loss: 0.2075 | LM: 0.1886 | LB: 1.0859 | CL0: 2.9 | CL1: 2.4 | HR0: 0.345/SR0: 0.344 | HR1: 0.415/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 13:38:46] Epoch 3 | Step 21770 | Loss: 0.2068 | LM: 0.1883 | LB: 1.0861 | CL0: 2.9 | CL1: 2.4 | HR0: 0.345/SR0: 0.344 | HR1: 0.415/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 13:38:53] Epoch 3 | Step 21780 | Loss: 0.2065 | LM: 0.1879 | LB: 1.0862 | CL0: 2.9 | CL1: 2.4 | HR0: 0.345/SR0: 0.344 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 13:38:59] Epoch 3 | Step 21790 | Loss: 0.2072 | LM: 0.1900 | LB: 1.0866 | CL0: 2.9 | CL1: 2.4 | HR0: 0.345/SR0: 0.345 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 13:39:06] Epoch 3 | Step 21800 | Loss: 0.2074 | LM: 0.1909 | LB: 1.0868 | CL0: 2.9 | CL1: 2.4 | HR0: 0.345/SR0: 0.345 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 13:39:12] Epoch 3 | Step 21810 | Loss: 0.2078 | LM: 0.1915 | LB: 1.0868 | CL0: 2.9 | CL1: 2.4 | HR0: 0.346/SR0: 0.345 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 13:39:18] Epoch 3 | Step 21820 | Loss: 0.2073 | LM: 0.1914 | LB: 1.0869 | CL0: 2.9 | CL1: 2.4 | HR0: 0.346/SR0: 0.345 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 13:39:25] Epoch 3 | Step 21830 | Loss: 0.2079 | LM: 0.1932 | LB: 1.0871 | CL0: 2.9 | CL1: 2.4 | HR0: 0.346/SR0: 0.345 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 13:39:31] Epoch 3 | Step 21840 | Loss: 0.2081 | LM: 0.1937 | LB: 1.0871 | CL0: 2.9 | CL1: 2.4 | HR0: 0.346/SR0: 0.345 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 13:39:37] Epoch 3 | Step 21850 | Loss: 0.2083 | LM: 0.1940 | LB: 1.0870 | CL0: 2.9 | CL1: 2.4 | HR0: 0.346/SR0: 0.345 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 13:39:44] Epoch 3 | Step 21860 | Loss: 0.2080 | LM: 0.1939 | LB: 1.0868 | CL0: 2.9 | CL1: 2.4 | HR0: 0.346/SR0: 0.345 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 13:39:50] Epoch 3 | Step 21870 | Loss: 0.2085 | LM: 0.1946 | LB: 1.0869 | CL0: 2.9 | CL1: 2.4 | HR0: 0.346/SR0: 0.345 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 13:39:57] Epoch 3 | Step 21880 | Loss: 0.2087 | LM: 0.1947 | LB: 1.0871 | CL0: 2.9 | CL1: 2.4 | HR0: 0.346/SR0: 0.345 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 13:40:03] Epoch 3 | Step 21890 | Loss: 0.2095 | LM: 0.1958 | LB: 1.0872 | CL0: 2.9 | CL1: 2.4 | HR0: 0.346/SR0: 0.345 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 13:40:10] Epoch 3 | Step 21900 | Loss: 0.2089 | LM: 0.1956 | LB: 1.0873 | CL0: 2.9 | CL1: 2.4 | HR0: 0.346/SR0: 0.345 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 13:40:16] Epoch 3 | Step 21910 | Loss: 0.2088 | LM: 0.1950 | LB: 1.0872 | CL0: 2.9 | CL1: 2.4 | HR0: 0.346/SR0: 0.345 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 13:40:22] Epoch 3 | Step 21920 | Loss: 0.2091 | LM: 0.1954 | LB: 1.0872 | CL0: 2.9 | CL1: 2.4 | HR0: 0.346/SR0: 0.345 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 13:40:29] Epoch 3 | Step 21930 | Loss: 0.2093 | LM: 0.1954 | LB: 1.0874 | CL0: 2.9 | CL1: 2.4 | HR0: 0.346/SR0: 0.345 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 13:40:35] Epoch 3 | Step 21940 | Loss: 0.2091 | LM: 0.1950 | LB: 1.0874 | CL0: 2.9 | CL1: 2.4 | HR0: 0.346/SR0: 0.345 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 13:40:41] Epoch 3 | Step 21950 | Loss: 0.2088 | LM: 0.1946 | LB: 1.0876 | CL0: 2.9 | CL1: 2.4 | HR0: 0.346/SR0: 0.345 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 13:40:48] Epoch 3 | Step 21960 | Loss: 0.2087 | LM: 0.1945 | LB: 1.0875 | CL0: 2.9 | CL1: 2.4 | HR0: 0.346/SR0: 0.345 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 13:40:54] Epoch 3 | Step 21970 | Loss: 0.2089 | LM: 0.1953 | LB: 1.0875 | CL0: 2.9 | CL1: 2.4 | HR0: 0.346/SR0: 0.346 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 13:41:01] Epoch 3 | Step 21980 | Loss: 0.2090 | LM: 0.1956 | LB: 1.0876 | CL0: 2.9 | CL1: 2.4 | HR0: 0.346/SR0: 0.346 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 13:41:07] Epoch 3 | Step 21990 | Loss: 0.2088 | LM: 0.1953 | LB: 1.0876 | CL0: 2.9 | CL1: 2.4 | HR0: 0.346/SR0: 0.346 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 13:41:13] Epoch 3 | Step 22000 | Loss: 0.2087 | LM: 0.1951 | LB: 1.0876 | CL0: 2.9 | CL1: 2.4 | HR0: 0.346/SR0: 0.346 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 13:41:15] Validation | Batch 10/784 | Loss: 0.3327 | LM_LOSS: 0.3218 | LB_LOSS: 1.0847 [2026-04-17 13:41:16] Validation | Batch 20/784 | Loss: 0.3438 | LM_LOSS: 0.3330 | LB_LOSS: 1.0849 [2026-04-17 13:41:17] Validation | Batch 30/784 | Loss: 0.3294 | LM_LOSS: 0.3186 | LB_LOSS: 1.0842 [2026-04-17 13:41:19] Validation | Batch 40/784 | Loss: 0.3322 | LM_LOSS: 0.3213 | LB_LOSS: 1.0841 [2026-04-17 13:41:20] Validation | Batch 50/784 | Loss: 0.3295 | LM_LOSS: 0.3186 | LB_LOSS: 1.0834 [2026-04-17 13:41:21] Validation | Batch 60/784 | Loss: 0.3315 | LM_LOSS: 0.3206 | LB_LOSS: 1.0830 [2026-04-17 13:41:23] Validation | Batch 70/784 | Loss: 0.3288 | LM_LOSS: 0.3180 | LB_LOSS: 1.0823 [2026-04-17 13:41:24] Validation | Batch 80/784 | Loss: 0.3250 | LM_LOSS: 0.3142 | LB_LOSS: 1.0819 [2026-04-17 13:41:25] Validation | Batch 90/784 | Loss: 0.3239 | LM_LOSS: 0.3131 | LB_LOSS: 1.0824 [2026-04-17 13:41:27] Validation | Batch 100/784 | Loss: 0.3259 | LM_LOSS: 0.3150 | LB_LOSS: 1.0829 [2026-04-17 13:41:28] Validation | Batch 110/784 | Loss: 0.3204 | LM_LOSS: 0.3096 | LB_LOSS: 1.0830 [2026-04-17 13:41:29] Validation | Batch 120/784 | Loss: 0.3240 | LM_LOSS: 0.3132 | LB_LOSS: 1.0829 [2026-04-17 13:41:31] Validation | Batch 130/784 | Loss: 0.3271 | LM_LOSS: 0.3162 | LB_LOSS: 1.0829 [2026-04-17 13:41:32] Validation | Batch 140/784 | Loss: 0.3264 | LM_LOSS: 0.3156 | LB_LOSS: 1.0827 [2026-04-17 13:41:33] Validation | Batch 150/784 | Loss: 0.3224 | LM_LOSS: 0.3116 | LB_LOSS: 1.0830 [2026-04-17 13:41:35] Validation | Batch 160/784 | Loss: 0.3232 | LM_LOSS: 0.3124 | LB_LOSS: 1.0827 [2026-04-17 13:41:36] Validation | Batch 170/784 | Loss: 0.3234 | LM_LOSS: 0.3125 | LB_LOSS: 1.0824 [2026-04-17 13:41:38] Validation | Batch 180/784 | Loss: 0.3209 | LM_LOSS: 0.3101 | LB_LOSS: 1.0824 [2026-04-17 13:41:39] Validation | Batch 190/784 | Loss: 0.3231 | LM_LOSS: 0.3123 | LB_LOSS: 1.0829 [2026-04-17 13:41:40] Validation | Batch 200/784 | Loss: 0.3235 | LM_LOSS: 0.3127 | LB_LOSS: 1.0829 [2026-04-17 13:41:42] Validation | Batch 210/784 | Loss: 0.3223 | LM_LOSS: 0.3115 | LB_LOSS: 1.0828 [2026-04-17 13:41:43] Validation | Batch 220/784 | Loss: 0.3232 | LM_LOSS: 0.3124 | LB_LOSS: 1.0829 [2026-04-17 13:41:45] Validation | Batch 230/784 | Loss: 0.3238 | LM_LOSS: 0.3130 | LB_LOSS: 1.0828 [2026-04-17 13:41:46] Validation | Batch 240/784 | Loss: 0.3242 | LM_LOSS: 0.3134 | LB_LOSS: 1.0831 [2026-04-17 13:41:47] Validation | Batch 250/784 | Loss: 0.3241 | LM_LOSS: 0.3133 | LB_LOSS: 1.0830 [2026-04-17 13:41:49] Validation | Batch 260/784 | Loss: 0.3244 | LM_LOSS: 0.3135 | LB_LOSS: 1.0832 [2026-04-17 13:41:51] Validation | Batch 270/784 | Loss: 0.3242 | LM_LOSS: 0.3134 | LB_LOSS: 1.0832 [2026-04-17 13:41:52] Validation | Batch 280/784 | Loss: 0.3246 | LM_LOSS: 0.3138 | LB_LOSS: 1.0834 [2026-04-17 13:41:53] Validation | Batch 290/784 | Loss: 0.3257 | LM_LOSS: 0.3149 | LB_LOSS: 1.0835 [2026-04-17 13:41:54] Validation | Batch 300/784 | Loss: 0.3266 | LM_LOSS: 0.3157 | LB_LOSS: 1.0836 [2026-04-17 13:41:56] Validation | Batch 310/784 | Loss: 0.3260 | LM_LOSS: 0.3151 | LB_LOSS: 1.0835 [2026-04-17 13:41:57] Validation | Batch 320/784 | Loss: 0.3276 | LM_LOSS: 0.3167 | LB_LOSS: 1.0835 [2026-04-17 13:41:59] Validation | Batch 330/784 | Loss: 0.3274 | LM_LOSS: 0.3165 | LB_LOSS: 1.0835 [2026-04-17 13:42:00] Validation | Batch 340/784 | Loss: 0.3262 | LM_LOSS: 0.3153 | LB_LOSS: 1.0836 [2026-04-17 13:42:01] Validation | Batch 350/784 | Loss: 0.3263 | LM_LOSS: 0.3155 | LB_LOSS: 1.0838 [2026-04-17 13:42:02] Validation | Batch 360/784 | Loss: 0.3261 | LM_LOSS: 0.3153 | LB_LOSS: 1.0838 [2026-04-17 13:42:04] Validation | Batch 370/784 | Loss: 0.3266 | LM_LOSS: 0.3158 | LB_LOSS: 1.0837 [2026-04-17 13:42:05] Validation | Batch 380/784 | Loss: 0.3264 | LM_LOSS: 0.3156 | LB_LOSS: 1.0838 [2026-04-17 13:42:06] Validation | Batch 390/784 | Loss: 0.3264 | LM_LOSS: 0.3155 | LB_LOSS: 1.0838 [2026-04-17 13:42:08] Validation | Batch 400/784 | Loss: 0.3266 | LM_LOSS: 0.3158 | LB_LOSS: 1.0838 [2026-04-17 13:42:09] Validation | Batch 410/784 | Loss: 0.3269 | LM_LOSS: 0.3161 | LB_LOSS: 1.0838 [2026-04-17 13:42:10] Validation | Batch 420/784 | Loss: 0.3272 | LM_LOSS: 0.3164 | LB_LOSS: 1.0839 [2026-04-17 13:42:11] Validation | Batch 430/784 | Loss: 0.3273 | LM_LOSS: 0.3164 | LB_LOSS: 1.0838 [2026-04-17 13:42:13] Validation | Batch 440/784 | Loss: 0.3270 | LM_LOSS: 0.3161 | LB_LOSS: 1.0838 [2026-04-17 13:42:14] Validation | Batch 450/784 | Loss: 0.3262 | LM_LOSS: 0.3154 | LB_LOSS: 1.0838 [2026-04-17 13:42:15] Validation | Batch 460/784 | Loss: 0.3267 | LM_LOSS: 0.3159 | LB_LOSS: 1.0839 [2026-04-17 13:42:17] Validation | Batch 470/784 | Loss: 0.3259 | LM_LOSS: 0.3150 | LB_LOSS: 1.0838 [2026-04-17 13:42:18] Validation | Batch 480/784 | Loss: 0.3264 | LM_LOSS: 0.3155 | LB_LOSS: 1.0838 [2026-04-17 13:42:19] Validation | Batch 490/784 | Loss: 0.3257 | LM_LOSS: 0.3149 | LB_LOSS: 1.0837 [2026-04-17 13:42:20] Validation | Batch 500/784 | Loss: 0.3261 | LM_LOSS: 0.3152 | LB_LOSS: 1.0837 [2026-04-17 13:42:22] Validation | Batch 510/784 | Loss: 0.3258 | LM_LOSS: 0.3149 | LB_LOSS: 1.0837 [2026-04-17 13:42:23] Validation | Batch 520/784 | Loss: 0.3260 | LM_LOSS: 0.3152 | LB_LOSS: 1.0836 [2026-04-17 13:42:25] Validation | Batch 530/784 | Loss: 0.3268 | LM_LOSS: 0.3160 | LB_LOSS: 1.0835 [2026-04-17 13:42:26] Validation | Batch 540/784 | Loss: 0.3272 | LM_LOSS: 0.3164 | LB_LOSS: 1.0836 [2026-04-17 13:42:27] Validation | Batch 550/784 | Loss: 0.3285 | LM_LOSS: 0.3177 | LB_LOSS: 1.0835 [2026-04-17 13:42:29] Validation | Batch 560/784 | Loss: 0.3286 | LM_LOSS: 0.3178 | LB_LOSS: 1.0836 [2026-04-17 13:42:30] Validation | Batch 570/784 | Loss: 0.3281 | LM_LOSS: 0.3173 | LB_LOSS: 1.0835 [2026-04-17 13:42:31] Validation | Batch 580/784 | Loss: 0.3276 | LM_LOSS: 0.3168 | LB_LOSS: 1.0835 [2026-04-17 13:42:33] Validation | Batch 590/784 | Loss: 0.3278 | LM_LOSS: 0.3170 | LB_LOSS: 1.0834 [2026-04-17 13:42:34] Validation | Batch 600/784 | Loss: 0.3277 | LM_LOSS: 0.3169 | LB_LOSS: 1.0834 [2026-04-17 13:42:36] Validation | Batch 610/784 | Loss: 0.3278 | LM_LOSS: 0.3170 | LB_LOSS: 1.0834 [2026-04-17 13:42:37] Validation | Batch 620/784 | Loss: 0.3277 | LM_LOSS: 0.3169 | LB_LOSS: 1.0834 [2026-04-17 13:42:38] Validation | Batch 630/784 | Loss: 0.3285 | LM_LOSS: 0.3177 | LB_LOSS: 1.0834 [2026-04-17 13:42:40] Validation | Batch 640/784 | Loss: 0.3285 | LM_LOSS: 0.3177 | LB_LOSS: 1.0834 [2026-04-17 13:42:42] Validation | Batch 650/784 | Loss: 0.3284 | LM_LOSS: 0.3176 | LB_LOSS: 1.0835 [2026-04-17 13:42:43] Validation | Batch 660/784 | Loss: 0.3288 | LM_LOSS: 0.3180 | LB_LOSS: 1.0835 [2026-04-17 13:42:44] Validation | Batch 670/784 | Loss: 0.3292 | LM_LOSS: 0.3184 | LB_LOSS: 1.0835 [2026-04-17 13:42:46] Validation | Batch 680/784 | Loss: 0.3289 | LM_LOSS: 0.3181 | LB_LOSS: 1.0835 [2026-04-17 13:42:47] Validation | Batch 690/784 | Loss: 0.3291 | LM_LOSS: 0.3182 | LB_LOSS: 1.0835 [2026-04-17 13:42:49] Validation | Batch 700/784 | Loss: 0.3291 | LM_LOSS: 0.3183 | LB_LOSS: 1.0834 [2026-04-17 13:42:50] Validation | Batch 710/784 | Loss: 0.3289 | LM_LOSS: 0.3181 | LB_LOSS: 1.0834 [2026-04-17 13:42:52] Validation | Batch 720/784 | Loss: 0.3286 | LM_LOSS: 0.3178 | LB_LOSS: 1.0833 [2026-04-17 13:42:53] Validation | Batch 730/784 | Loss: 0.3281 | LM_LOSS: 0.3173 | LB_LOSS: 1.0833 [2026-04-17 13:42:54] Validation | Batch 740/784 | Loss: 0.3282 | LM_LOSS: 0.3173 | LB_LOSS: 1.0833 [2026-04-17 13:42:55] Validation | Batch 750/784 | Loss: 0.3275 | LM_LOSS: 0.3167 | LB_LOSS: 1.0833 [2026-04-17 13:42:57] Validation | Batch 760/784 | Loss: 0.3276 | LM_LOSS: 0.3168 | LB_LOSS: 1.0833 [2026-04-17 13:42:58] Validation | Batch 770/784 | Loss: 0.3278 | LM_LOSS: 0.3170 | LB_LOSS: 1.0834 [2026-04-17 13:42:59] Validation | Batch 780/784 | Loss: 0.3282 | LM_LOSS: 0.3173 | LB_LOSS: 1.0833 [2026-04-17 13:43:00] Validation | Batch 784/784 | Loss: 0.3284 | LM_LOSS: 0.3175 | LB_LOSS: 1.0833 [2026-04-17 13:43:03] Validation | Loss: 0.3284 | LM_LOSS: 0.3175 | LB_LOSS: 1.0833 | PPL: 1.37 | Time: 106.65s [2026-04-17 13:43:10] Epoch 3 | Step 22010 | Loss: 0.2085 | LM: 0.1953 | LB: 1.0876 | CL0: 2.9 | CL1: 2.4 | HR0: 0.346/SR0: 0.346 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 13:43:16] Epoch 3 | Step 22020 | Loss: 0.2084 | LM: 0.1955 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.346/SR0: 0.346 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 13:43:23] Epoch 3 | Step 22030 | Loss: 0.2078 | LM: 0.1948 | LB: 1.0878 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.346 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 13:43:29] Epoch 3 | Step 22040 | Loss: 0.2080 | LM: 0.1945 | LB: 1.0878 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.346 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 13:43:36] Epoch 3 | Step 22050 | Loss: 0.2080 | LM: 0.1943 | LB: 1.0878 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.346 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 13:43:42] Epoch 3 | Step 22060 | Loss: 0.2084 | LM: 0.1945 | LB: 1.0878 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.346 | HR1: 0.417/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 13:43:48] Epoch 3 | Step 22070 | Loss: 0.2084 | LM: 0.1951 | LB: 1.0878 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.346 | HR1: 0.417/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 13:43:55] Epoch 3 | Step 22080 | Loss: 0.2085 | LM: 0.1952 | LB: 1.0878 | CL0: 2.9 | CL1: 2.4 | HR0: 0.346/SR0: 0.346 | HR1: 0.417/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 13:44:01] Epoch 3 | Step 22090 | Loss: 0.2080 | LM: 0.1948 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.346/SR0: 0.346 | HR1: 0.417/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 13:44:08] Epoch 3 | Step 22100 | Loss: 0.2080 | LM: 0.1950 | LB: 1.0878 | CL0: 2.9 | CL1: 2.4 | HR0: 0.346/SR0: 0.346 | HR1: 0.417/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 13:44:14] Epoch 3 | Step 22110 | Loss: 0.2085 | LM: 0.1961 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.346/SR0: 0.346 | HR1: 0.417/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 13:44:21] Epoch 3 | Step 22120 | Loss: 0.2081 | LM: 0.1957 | LB: 1.0878 | CL0: 2.9 | CL1: 2.4 | HR0: 0.346/SR0: 0.346 | HR1: 0.417/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 13:44:27] Epoch 3 | Step 22130 | Loss: 0.2078 | LM: 0.1957 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.346/SR0: 0.346 | HR1: 0.417/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 13:44:34] Epoch 3 | Step 22140 | Loss: 0.2074 | LM: 0.1953 | LB: 1.0876 | CL0: 2.9 | CL1: 2.4 | HR0: 0.346/SR0: 0.346 | HR1: 0.417/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 13:44:40] Epoch 3 | Step 22150 | Loss: 0.2071 | LM: 0.1949 | LB: 1.0876 | CL0: 2.9 | CL1: 2.4 | HR0: 0.346/SR0: 0.346 | HR1: 0.417/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 13:44:46] Epoch 3 | Step 22160 | Loss: 0.2074 | LM: 0.1957 | LB: 1.0876 | CL0: 2.9 | CL1: 2.4 | HR0: 0.346/SR0: 0.346 | HR1: 0.417/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 13:44:53] Epoch 3 | Step 22170 | Loss: 0.2071 | LM: 0.1952 | LB: 1.0876 | CL0: 2.9 | CL1: 2.4 | HR0: 0.346/SR0: 0.346 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 13:44:59] Epoch 3 | Step 22180 | Loss: 0.2072 | LM: 0.1952 | LB: 1.0875 | CL0: 2.9 | CL1: 2.4 | HR0: 0.346/SR0: 0.346 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 13:45:06] Epoch 3 | Step 22190 | Loss: 0.2077 | LM: 0.1958 | LB: 1.0875 | CL0: 2.9 | CL1: 2.4 | HR0: 0.346/SR0: 0.346 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 13:45:12] Epoch 3 | Step 22200 | Loss: 0.2074 | LM: 0.1955 | LB: 1.0874 | CL0: 2.9 | CL1: 2.4 | HR0: 0.346/SR0: 0.346 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 13:45:18] Epoch 3 | Step 22210 | Loss: 0.2075 | LM: 0.1951 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.346/SR0: 0.346 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 13:45:25] Epoch 3 | Step 22220 | Loss: 0.2072 | LM: 0.1948 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.346 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 13:45:31] Epoch 3 | Step 22230 | Loss: 0.2071 | LM: 0.1949 | LB: 1.0878 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.346 | HR1: 0.417/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 13:45:38] Epoch 3 | Step 22240 | Loss: 0.2071 | LM: 0.1946 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.346 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 13:45:44] Epoch 3 | Step 22250 | Loss: 0.2072 | LM: 0.1946 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.346 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 13:45:51] Epoch 3 | Step 22260 | Loss: 0.2072 | LM: 0.1945 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.346 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 13:45:57] Epoch 3 | Step 22270 | Loss: 0.2073 | LM: 0.1946 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.346 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 13:46:03] Epoch 3 | Step 22280 | Loss: 0.2071 | LM: 0.1949 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.346 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 13:46:10] Epoch 3 | Step 22290 | Loss: 0.2071 | LM: 0.1944 | LB: 1.0876 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.346 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 13:46:16] Epoch 3 | Step 22300 | Loss: 0.2069 | LM: 0.1943 | LB: 1.0875 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.346 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 13:46:22] Epoch 3 | Step 22310 | Loss: 0.2070 | LM: 0.1942 | LB: 1.0875 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.346 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 13:46:29] Epoch 3 | Step 22320 | Loss: 0.2067 | LM: 0.1938 | LB: 1.0874 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.346 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 13:46:35] Epoch 3 | Step 22330 | Loss: 0.2067 | LM: 0.1944 | LB: 1.0876 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.346 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 13:46:41] Epoch 3 | Step 22340 | Loss: 0.2067 | LM: 0.1945 | LB: 1.0875 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.346 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 13:46:48] Epoch 3 | Step 22350 | Loss: 0.2067 | LM: 0.1946 | LB: 1.0876 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.346 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 13:46:54] Epoch 3 | Step 22360 | Loss: 0.2067 | LM: 0.1947 | LB: 1.0875 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.346 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 13:47:00] Epoch 3 | Step 22370 | Loss: 0.2066 | LM: 0.1943 | LB: 1.0875 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.346 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 13:47:07] Epoch 3 | Step 22380 | Loss: 0.2067 | LM: 0.1944 | LB: 1.0875 | CL0: 2.9 | CL1: 2.4 | HR0: 0.346/SR0: 0.346 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 13:47:13] Epoch 3 | Step 22390 | Loss: 0.2067 | LM: 0.1948 | LB: 1.0875 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.346 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 13:47:20] Epoch 3 | Step 22400 | Loss: 0.2067 | LM: 0.1947 | LB: 1.0875 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.346 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 13:47:26] Epoch 3 | Step 22410 | Loss: 0.2066 | LM: 0.1944 | LB: 1.0875 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.346 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 13:47:32] Epoch 3 | Step 22420 | Loss: 0.2067 | LM: 0.1949 | LB: 1.0876 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.346 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 13:47:39] Epoch 3 | Step 22430 | Loss: 0.2069 | LM: 0.1953 | LB: 1.0875 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.346 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 13:47:45] Epoch 3 | Step 22440 | Loss: 0.2068 | LM: 0.1950 | LB: 1.0875 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.346 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 13:47:51] Epoch 3 | Step 22450 | Loss: 0.2066 | LM: 0.1951 | LB: 1.0876 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.346 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 13:47:58] Epoch 3 | Step 22460 | Loss: 0.2066 | LM: 0.1951 | LB: 1.0876 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.346 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 13:48:04] Epoch 3 | Step 22470 | Loss: 0.2065 | LM: 0.1948 | LB: 1.0875 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.346 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 13:48:10] Epoch 3 | Step 22480 | Loss: 0.2067 | LM: 0.1948 | LB: 1.0875 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.346 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 13:48:17] Epoch 3 | Step 22490 | Loss: 0.2068 | LM: 0.1950 | LB: 1.0875 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.346 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 13:48:23] Epoch 3 | Step 22500 | Loss: 0.2070 | LM: 0.1948 | LB: 1.0874 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.346 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 13:48:29] Epoch 3 | Step 22510 | Loss: 0.2071 | LM: 0.1946 | LB: 1.0875 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.346 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 13:48:36] Epoch 3 | Step 22520 | Loss: 0.2071 | LM: 0.1945 | LB: 1.0875 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.346 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 13:48:42] Epoch 3 | Step 22530 | Loss: 0.2070 | LM: 0.1945 | LB: 1.0875 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.346 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 13:48:48] Epoch 3 | Step 22540 | Loss: 0.2069 | LM: 0.1948 | LB: 1.0875 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.346 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 13:48:54] Epoch 3 | Step 22550 | Loss: 0.2069 | LM: 0.1948 | LB: 1.0874 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.346 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 13:49:01] Epoch 3 | Step 22560 | Loss: 0.2068 | LM: 0.1948 | LB: 1.0874 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.346 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 13:49:07] Epoch 3 | Step 22570 | Loss: 0.2066 | LM: 0.1948 | LB: 1.0873 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.346 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 13:49:13] Epoch 3 | Step 22580 | Loss: 0.2064 | LM: 0.1948 | LB: 1.0873 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.346 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 13:49:19] Epoch 3 | Step 22590 | Loss: 0.2065 | LM: 0.1946 | LB: 1.0873 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.346 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 13:49:26] Epoch 3 | Step 22600 | Loss: 0.2067 | LM: 0.1950 | LB: 1.0873 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.346 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 13:49:32] Epoch 3 | Step 22610 | Loss: 0.2067 | LM: 0.1951 | LB: 1.0873 | CL0: 2.9 | CL1: 2.4 | HR0: 0.346/SR0: 0.346 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 13:49:38] Epoch 3 | Step 22620 | Loss: 0.2066 | LM: 0.1947 | LB: 1.0872 | CL0: 2.9 | CL1: 2.4 | HR0: 0.346/SR0: 0.346 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 13:49:44] Epoch 3 | Step 22630 | Loss: 0.2066 | LM: 0.1947 | LB: 1.0872 | CL0: 2.9 | CL1: 2.4 | HR0: 0.346/SR0: 0.346 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 13:49:51] Epoch 3 | Step 22640 | Loss: 0.2065 | LM: 0.1946 | LB: 1.0871 | CL0: 2.9 | CL1: 2.4 | HR0: 0.346/SR0: 0.346 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 13:49:58] Epoch 3 | Step 22650 | Loss: 0.2065 | LM: 0.1944 | LB: 1.0870 | CL0: 2.9 | CL1: 2.4 | HR0: 0.346/SR0: 0.346 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 13:50:04] Epoch 3 | Step 22660 | Loss: 0.2066 | LM: 0.1944 | LB: 1.0870 | CL0: 2.9 | CL1: 2.4 | HR0: 0.346/SR0: 0.346 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 13:50:10] Epoch 3 | Step 22670 | Loss: 0.2063 | LM: 0.1944 | LB: 1.0870 | CL0: 2.9 | CL1: 2.4 | HR0: 0.346/SR0: 0.346 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 13:50:16] Epoch 3 | Step 22680 | Loss: 0.2064 | LM: 0.1946 | LB: 1.0870 | CL0: 2.9 | CL1: 2.4 | HR0: 0.346/SR0: 0.346 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 13:50:22] Epoch 3 | Step 22690 | Loss: 0.2064 | LM: 0.1944 | LB: 1.0870 | CL0: 2.9 | CL1: 2.4 | HR0: 0.346/SR0: 0.346 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 13:50:28] Epoch 3 | Step 22700 | Loss: 0.2064 | LM: 0.1942 | LB: 1.0869 | CL0: 2.9 | CL1: 2.4 | HR0: 0.346/SR0: 0.346 | HR1: 0.415/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 13:50:35] Epoch 3 | Step 22710 | Loss: 0.2065 | LM: 0.1945 | LB: 1.0869 | CL0: 2.9 | CL1: 2.4 | HR0: 0.346/SR0: 0.346 | HR1: 0.415/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 13:50:41] Epoch 3 | Step 22720 | Loss: 0.2066 | LM: 0.1944 | LB: 1.0870 | CL0: 2.9 | CL1: 2.4 | HR0: 0.346/SR0: 0.346 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 13:50:47] Epoch 3 | Step 22730 | Loss: 0.2065 | LM: 0.1943 | LB: 1.0870 | CL0: 2.9 | CL1: 2.4 | HR0: 0.346/SR0: 0.346 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 13:50:54] Epoch 3 | Step 22740 | Loss: 0.2065 | LM: 0.1941 | LB: 1.0870 | CL0: 2.9 | CL1: 2.4 | HR0: 0.346/SR0: 0.346 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 13:51:00] Epoch 3 | Step 22750 | Loss: 0.2064 | LM: 0.1942 | LB: 1.0870 | CL0: 2.9 | CL1: 2.4 | HR0: 0.346/SR0: 0.346 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 13:51:06] Epoch 3 | Step 22760 | Loss: 0.2066 | LM: 0.1948 | LB: 1.0870 | CL0: 2.9 | CL1: 2.4 | HR0: 0.346/SR0: 0.346 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 13:51:12] Epoch 3 | Step 22770 | Loss: 0.2068 | LM: 0.1946 | LB: 1.0870 | CL0: 2.9 | CL1: 2.4 | HR0: 0.346/SR0: 0.346 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 13:51:18] Epoch 3 | Step 22780 | Loss: 0.2068 | LM: 0.1943 | LB: 1.0870 | CL0: 2.9 | CL1: 2.4 | HR0: 0.346/SR0: 0.346 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 13:51:24] Epoch 3 | Step 22790 | Loss: 0.2067 | LM: 0.1940 | LB: 1.0870 | CL0: 2.9 | CL1: 2.4 | HR0: 0.346/SR0: 0.346 | HR1: 0.415/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 13:51:31] Epoch 3 | Step 22800 | Loss: 0.2069 | LM: 0.1942 | LB: 1.0870 | CL0: 2.9 | CL1: 2.4 | HR0: 0.346/SR0: 0.346 | HR1: 0.415/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 13:51:37] Epoch 3 | Step 22810 | Loss: 0.2068 | LM: 0.1941 | LB: 1.0870 | CL0: 2.9 | CL1: 2.4 | HR0: 0.346/SR0: 0.346 | HR1: 0.415/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 13:51:43] Epoch 3 | Step 22820 | Loss: 0.2067 | LM: 0.1942 | LB: 1.0871 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.346 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 13:51:50] Epoch 3 | Step 22830 | Loss: 0.2070 | LM: 0.1946 | LB: 1.0871 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.346 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 13:51:56] Epoch 3 | Step 22840 | Loss: 0.2069 | LM: 0.1945 | LB: 1.0871 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.346 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 13:52:03] Epoch 3 | Step 22850 | Loss: 0.2068 | LM: 0.1946 | LB: 1.0871 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.346 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 13:52:09] Epoch 3 | Step 22860 | Loss: 0.2065 | LM: 0.1943 | LB: 1.0871 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.346 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 13:52:15] Epoch 3 | Step 22870 | Loss: 0.2067 | LM: 0.1944 | LB: 1.0871 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.346 | HR1: 0.415/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 13:52:21] Epoch 3 | Step 22880 | Loss: 0.2068 | LM: 0.1942 | LB: 1.0870 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.346 | HR1: 0.415/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 13:52:28] Epoch 3 | Step 22890 | Loss: 0.2068 | LM: 0.1942 | LB: 1.0870 | CL0: 2.9 | CL1: 2.4 | HR0: 0.346/SR0: 0.346 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 13:52:34] Epoch 3 | Step 22900 | Loss: 0.2071 | LM: 0.1947 | LB: 1.0870 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.346 | HR1: 0.415/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 13:52:40] Epoch 3 | Step 22910 | Loss: 0.2070 | LM: 0.1947 | LB: 1.0870 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.346 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 13:52:46] Epoch 3 | Step 22920 | Loss: 0.2070 | LM: 0.1947 | LB: 1.0871 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.346 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 13:52:52] Epoch 3 | Step 22930 | Loss: 0.2070 | LM: 0.1948 | LB: 1.0871 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.346 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 13:52:59] Epoch 3 | Step 22940 | Loss: 0.2071 | LM: 0.1949 | LB: 1.0871 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.346 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 13:53:06] Epoch 3 | Step 22950 | Loss: 0.2073 | LM: 0.1951 | LB: 1.0870 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.346 | HR1: 0.415/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 13:53:12] Epoch 3 | Step 22960 | Loss: 0.2073 | LM: 0.1948 | LB: 1.0871 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.346 | HR1: 0.415/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 13:53:18] Epoch 3 | Step 22970 | Loss: 0.2072 | LM: 0.1947 | LB: 1.0870 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.346 | HR1: 0.415/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 13:53:25] Epoch 3 | Step 22980 | Loss: 0.2071 | LM: 0.1945 | LB: 1.0870 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.346 | HR1: 0.415/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 13:53:31] Epoch 3 | Step 22990 | Loss: 0.2070 | LM: 0.1948 | LB: 1.0870 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.346 | HR1: 0.415/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 13:53:37] Epoch 3 | Step 23000 | Loss: 0.2071 | LM: 0.1950 | LB: 1.0870 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.346 | HR1: 0.415/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 13:53:38] Validation | Batch 10/784 | Loss: 0.3338 | LM_LOSS: 0.3230 | LB_LOSS: 1.0847 [2026-04-17 13:53:40] Validation | Batch 20/784 | Loss: 0.3454 | LM_LOSS: 0.3345 | LB_LOSS: 1.0849 [2026-04-17 13:53:41] Validation | Batch 30/784 | Loss: 0.3311 | LM_LOSS: 0.3202 | LB_LOSS: 1.0842 [2026-04-17 13:53:42] Validation | Batch 40/784 | Loss: 0.3337 | LM_LOSS: 0.3229 | LB_LOSS: 1.0841 [2026-04-17 13:53:44] Validation | Batch 50/784 | Loss: 0.3311 | LM_LOSS: 0.3202 | LB_LOSS: 1.0835 [2026-04-17 13:53:45] Validation | Batch 60/784 | Loss: 0.3331 | LM_LOSS: 0.3223 | LB_LOSS: 1.0830 [2026-04-17 13:53:46] Validation | Batch 70/784 | Loss: 0.3305 | LM_LOSS: 0.3196 | LB_LOSS: 1.0824 [2026-04-17 13:53:48] Validation | Batch 80/784 | Loss: 0.3267 | LM_LOSS: 0.3159 | LB_LOSS: 1.0819 [2026-04-17 13:53:49] Validation | Batch 90/784 | Loss: 0.3256 | LM_LOSS: 0.3147 | LB_LOSS: 1.0825 [2026-04-17 13:53:50] Validation | Batch 100/784 | Loss: 0.3276 | LM_LOSS: 0.3168 | LB_LOSS: 1.0829 [2026-04-17 13:53:52] Validation | Batch 110/784 | Loss: 0.3221 | LM_LOSS: 0.3113 | LB_LOSS: 1.0830 [2026-04-17 13:53:53] Validation | Batch 120/784 | Loss: 0.3258 | LM_LOSS: 0.3150 | LB_LOSS: 1.0830 [2026-04-17 13:53:54] Validation | Batch 130/784 | Loss: 0.3289 | LM_LOSS: 0.3180 | LB_LOSS: 1.0829 [2026-04-17 13:53:56] Validation | Batch 140/784 | Loss: 0.3282 | LM_LOSS: 0.3174 | LB_LOSS: 1.0827 [2026-04-17 13:53:57] Validation | Batch 150/784 | Loss: 0.3243 | LM_LOSS: 0.3135 | LB_LOSS: 1.0830 [2026-04-17 13:53:59] Validation | Batch 160/784 | Loss: 0.3251 | LM_LOSS: 0.3143 | LB_LOSS: 1.0827 [2026-04-17 13:54:00] Validation | Batch 170/784 | Loss: 0.3252 | LM_LOSS: 0.3144 | LB_LOSS: 1.0824 [2026-04-17 13:54:01] Validation | Batch 180/784 | Loss: 0.3227 | LM_LOSS: 0.3119 | LB_LOSS: 1.0825 [2026-04-17 13:54:03] Validation | Batch 190/784 | Loss: 0.3249 | LM_LOSS: 0.3141 | LB_LOSS: 1.0829 [2026-04-17 13:54:04] Validation | Batch 200/784 | Loss: 0.3254 | LM_LOSS: 0.3145 | LB_LOSS: 1.0830 [2026-04-17 13:54:05] Validation | Batch 210/784 | Loss: 0.3242 | LM_LOSS: 0.3134 | LB_LOSS: 1.0829 [2026-04-17 13:54:07] Validation | Batch 220/784 | Loss: 0.3251 | LM_LOSS: 0.3142 | LB_LOSS: 1.0829 [2026-04-17 13:54:08] Validation | Batch 230/784 | Loss: 0.3257 | LM_LOSS: 0.3148 | LB_LOSS: 1.0828 [2026-04-17 13:54:10] Validation | Batch 240/784 | Loss: 0.3261 | LM_LOSS: 0.3153 | LB_LOSS: 1.0832 [2026-04-17 13:54:11] Validation | Batch 250/784 | Loss: 0.3260 | LM_LOSS: 0.3152 | LB_LOSS: 1.0830 [2026-04-17 13:54:13] Validation | Batch 260/784 | Loss: 0.3263 | LM_LOSS: 0.3154 | LB_LOSS: 1.0832 [2026-04-17 13:54:14] Validation | Batch 270/784 | Loss: 0.3261 | LM_LOSS: 0.3153 | LB_LOSS: 1.0833 [2026-04-17 13:54:15] Validation | Batch 280/784 | Loss: 0.3266 | LM_LOSS: 0.3157 | LB_LOSS: 1.0834 [2026-04-17 13:54:17] Validation | Batch 290/784 | Loss: 0.3277 | LM_LOSS: 0.3168 | LB_LOSS: 1.0836 [2026-04-17 13:54:18] Validation | Batch 300/784 | Loss: 0.3285 | LM_LOSS: 0.3177 | LB_LOSS: 1.0836 [2026-04-17 13:54:19] Validation | Batch 310/784 | Loss: 0.3279 | LM_LOSS: 0.3170 | LB_LOSS: 1.0835 [2026-04-17 13:54:21] Validation | Batch 320/784 | Loss: 0.3295 | LM_LOSS: 0.3187 | LB_LOSS: 1.0835 [2026-04-17 13:54:22] Validation | Batch 330/784 | Loss: 0.3293 | LM_LOSS: 0.3184 | LB_LOSS: 1.0835 [2026-04-17 13:54:23] Validation | Batch 340/784 | Loss: 0.3281 | LM_LOSS: 0.3172 | LB_LOSS: 1.0836 [2026-04-17 13:54:25] Validation | Batch 350/784 | Loss: 0.3283 | LM_LOSS: 0.3174 | LB_LOSS: 1.0838 [2026-04-17 13:54:26] Validation | Batch 360/784 | Loss: 0.3281 | LM_LOSS: 0.3172 | LB_LOSS: 1.0838 [2026-04-17 13:54:27] Validation | Batch 370/784 | Loss: 0.3286 | LM_LOSS: 0.3177 | LB_LOSS: 1.0837 [2026-04-17 13:54:29] Validation | Batch 380/784 | Loss: 0.3284 | LM_LOSS: 0.3175 | LB_LOSS: 1.0838 [2026-04-17 13:54:30] Validation | Batch 390/784 | Loss: 0.3283 | LM_LOSS: 0.3175 | LB_LOSS: 1.0839 [2026-04-17 13:54:31] Validation | Batch 400/784 | Loss: 0.3286 | LM_LOSS: 0.3177 | LB_LOSS: 1.0838 [2026-04-17 13:54:32] Validation | Batch 410/784 | Loss: 0.3289 | LM_LOSS: 0.3181 | LB_LOSS: 1.0839 [2026-04-17 13:54:34] Validation | Batch 420/784 | Loss: 0.3292 | LM_LOSS: 0.3183 | LB_LOSS: 1.0839 [2026-04-17 13:54:35] Validation | Batch 430/784 | Loss: 0.3293 | LM_LOSS: 0.3184 | LB_LOSS: 1.0838 [2026-04-17 13:54:36] Validation | Batch 440/784 | Loss: 0.3289 | LM_LOSS: 0.3181 | LB_LOSS: 1.0839 [2026-04-17 13:54:38] Validation | Batch 450/784 | Loss: 0.3282 | LM_LOSS: 0.3173 | LB_LOSS: 1.0838 [2026-04-17 13:54:39] Validation | Batch 460/784 | Loss: 0.3287 | LM_LOSS: 0.3178 | LB_LOSS: 1.0839 [2026-04-17 13:54:40] Validation | Batch 470/784 | Loss: 0.3278 | LM_LOSS: 0.3170 | LB_LOSS: 1.0839 [2026-04-17 13:54:42] Validation | Batch 480/784 | Loss: 0.3283 | LM_LOSS: 0.3175 | LB_LOSS: 1.0839 [2026-04-17 13:54:43] Validation | Batch 490/784 | Loss: 0.3277 | LM_LOSS: 0.3168 | LB_LOSS: 1.0838 [2026-04-17 13:54:44] Validation | Batch 500/784 | Loss: 0.3280 | LM_LOSS: 0.3172 | LB_LOSS: 1.0837 [2026-04-17 13:54:46] Validation | Batch 510/784 | Loss: 0.3278 | LM_LOSS: 0.3169 | LB_LOSS: 1.0837 [2026-04-17 13:54:47] Validation | Batch 520/784 | Loss: 0.3280 | LM_LOSS: 0.3171 | LB_LOSS: 1.0836 [2026-04-17 13:54:49] Validation | Batch 530/784 | Loss: 0.3288 | LM_LOSS: 0.3180 | LB_LOSS: 1.0836 [2026-04-17 13:54:50] Validation | Batch 540/784 | Loss: 0.3292 | LM_LOSS: 0.3183 | LB_LOSS: 1.0836 [2026-04-17 13:54:51] Validation | Batch 550/784 | Loss: 0.3305 | LM_LOSS: 0.3196 | LB_LOSS: 1.0836 [2026-04-17 13:54:53] Validation | Batch 560/784 | Loss: 0.3306 | LM_LOSS: 0.3198 | LB_LOSS: 1.0836 [2026-04-17 13:54:54] Validation | Batch 570/784 | Loss: 0.3301 | LM_LOSS: 0.3193 | LB_LOSS: 1.0835 [2026-04-17 13:54:56] Validation | Batch 580/784 | Loss: 0.3296 | LM_LOSS: 0.3187 | LB_LOSS: 1.0836 [2026-04-17 13:54:57] Validation | Batch 590/784 | Loss: 0.3298 | LM_LOSS: 0.3190 | LB_LOSS: 1.0835 [2026-04-17 13:54:58] Validation | Batch 600/784 | Loss: 0.3297 | LM_LOSS: 0.3189 | LB_LOSS: 1.0834 [2026-04-17 13:55:00] Validation | Batch 610/784 | Loss: 0.3298 | LM_LOSS: 0.3190 | LB_LOSS: 1.0834 [2026-04-17 13:55:01] Validation | Batch 620/784 | Loss: 0.3297 | LM_LOSS: 0.3188 | LB_LOSS: 1.0834 [2026-04-17 13:55:03] Validation | Batch 630/784 | Loss: 0.3305 | LM_LOSS: 0.3196 | LB_LOSS: 1.0835 [2026-04-17 13:55:04] Validation | Batch 640/784 | Loss: 0.3305 | LM_LOSS: 0.3197 | LB_LOSS: 1.0834 [2026-04-17 13:55:06] Validation | Batch 650/784 | Loss: 0.3304 | LM_LOSS: 0.3195 | LB_LOSS: 1.0835 [2026-04-17 13:55:07] Validation | Batch 660/784 | Loss: 0.3308 | LM_LOSS: 0.3199 | LB_LOSS: 1.0835 [2026-04-17 13:55:09] Validation | Batch 670/784 | Loss: 0.3312 | LM_LOSS: 0.3204 | LB_LOSS: 1.0836 [2026-04-17 13:55:10] Validation | Batch 680/784 | Loss: 0.3309 | LM_LOSS: 0.3201 | LB_LOSS: 1.0835 [2026-04-17 13:55:11] Validation | Batch 690/784 | Loss: 0.3310 | LM_LOSS: 0.3202 | LB_LOSS: 1.0835 [2026-04-17 13:55:13] Validation | Batch 700/784 | Loss: 0.3311 | LM_LOSS: 0.3203 | LB_LOSS: 1.0834 [2026-04-17 13:55:14] Validation | Batch 710/784 | Loss: 0.3309 | LM_LOSS: 0.3200 | LB_LOSS: 1.0834 [2026-04-17 13:55:16] Validation | Batch 720/784 | Loss: 0.3306 | LM_LOSS: 0.3198 | LB_LOSS: 1.0833 [2026-04-17 13:55:17] Validation | Batch 730/784 | Loss: 0.3301 | LM_LOSS: 0.3193 | LB_LOSS: 1.0833 [2026-04-17 13:55:18] Validation | Batch 740/784 | Loss: 0.3301 | LM_LOSS: 0.3193 | LB_LOSS: 1.0834 [2026-04-17 13:55:19] Validation | Batch 750/784 | Loss: 0.3295 | LM_LOSS: 0.3186 | LB_LOSS: 1.0834 [2026-04-17 13:55:21] Validation | Batch 760/784 | Loss: 0.3296 | LM_LOSS: 0.3188 | LB_LOSS: 1.0834 [2026-04-17 13:55:22] Validation | Batch 770/784 | Loss: 0.3298 | LM_LOSS: 0.3190 | LB_LOSS: 1.0834 [2026-04-17 13:55:23] Validation | Batch 780/784 | Loss: 0.3301 | LM_LOSS: 0.3193 | LB_LOSS: 1.0834 [2026-04-17 13:55:24] Validation | Batch 784/784 | Loss: 0.3303 | LM_LOSS: 0.3195 | LB_LOSS: 1.0834 [2026-04-17 13:55:27] Validation | Loss: 0.3303 | LM_LOSS: 0.3195 | LB_LOSS: 1.0834 | PPL: 1.38 | Time: 106.80s [2026-04-17 13:55:33] Epoch 3 | Step 23010 | Loss: 0.2071 | LM: 0.1951 | LB: 1.0870 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.346 | HR1: 0.415/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 13:55:39] Epoch 3 | Step 23020 | Loss: 0.2071 | LM: 0.1956 | LB: 1.0870 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.346 | HR1: 0.415/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 13:55:46] Epoch 3 | Step 23030 | Loss: 0.2071 | LM: 0.1958 | LB: 1.0871 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.346 | HR1: 0.415/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 13:55:52] Epoch 3 | Step 23040 | Loss: 0.2071 | LM: 0.1962 | LB: 1.0870 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.346 | HR1: 0.415/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 13:55:59] Epoch 3 | Step 23050 | Loss: 0.2074 | LM: 0.1963 | LB: 1.0870 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.346 | HR1: 0.415/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 13:56:05] Epoch 3 | Step 23060 | Loss: 0.2075 | LM: 0.1963 | LB: 1.0870 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.346 | HR1: 0.415/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 13:56:12] Epoch 3 | Step 23070 | Loss: 0.2075 | LM: 0.1961 | LB: 1.0870 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.346 | HR1: 0.415/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 13:56:17] Epoch 3 | Step 23080 | Loss: 0.2074 | LM: 0.1964 | LB: 1.0870 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.346 | HR1: 0.415/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 13:56:24] Epoch 3 | Step 23090 | Loss: 0.2073 | LM: 0.1963 | LB: 1.0870 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.346 | HR1: 0.415/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 13:56:30] Epoch 3 | Step 23100 | Loss: 0.2072 | LM: 0.1962 | LB: 1.0870 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.346 | HR1: 0.415/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 13:56:37] Epoch 3 | Step 23110 | Loss: 0.2070 | LM: 0.1960 | LB: 1.0870 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.346 | HR1: 0.415/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 13:56:43] Epoch 3 | Step 23120 | Loss: 0.2071 | LM: 0.1959 | LB: 1.0870 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.346 | HR1: 0.415/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 13:56:50] Epoch 3 | Step 23130 | Loss: 0.2069 | LM: 0.1956 | LB: 1.0869 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.346 | HR1: 0.415/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 13:56:56] Epoch 3 | Step 23140 | Loss: 0.2070 | LM: 0.1958 | LB: 1.0870 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.346 | HR1: 0.415/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 13:57:03] Epoch 3 | Step 23150 | Loss: 0.2072 | LM: 0.1961 | LB: 1.0870 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.346 | HR1: 0.415/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 13:57:09] Epoch 3 | Step 23160 | Loss: 0.2071 | LM: 0.1962 | LB: 1.0870 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.346 | HR1: 0.415/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 13:57:16] Epoch 3 | Step 23170 | Loss: 0.2071 | LM: 0.1965 | LB: 1.0870 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.346 | HR1: 0.415/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 13:57:22] Epoch 3 | Step 23180 | Loss: 0.2071 | LM: 0.1964 | LB: 1.0870 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.346 | HR1: 0.415/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 13:57:28] Epoch 3 | Step 23190 | Loss: 0.2071 | LM: 0.1968 | LB: 1.0870 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.346 | HR1: 0.415/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 13:57:34] Epoch 3 | Step 23200 | Loss: 0.2070 | LM: 0.1967 | LB: 1.0870 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.346 | HR1: 0.415/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 13:57:41] Epoch 3 | Step 23210 | Loss: 0.2070 | LM: 0.1966 | LB: 1.0870 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.346 | HR1: 0.415/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 13:57:47] Epoch 3 | Step 23220 | Loss: 0.2070 | LM: 0.1965 | LB: 1.0870 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.346 | HR1: 0.415/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 13:57:53] Epoch 3 | Step 23230 | Loss: 0.2071 | LM: 0.1965 | LB: 1.0870 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.346 | HR1: 0.415/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 13:58:00] Epoch 3 | Step 23240 | Loss: 0.2070 | LM: 0.1965 | LB: 1.0870 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.346 | HR1: 0.415/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 13:58:06] Epoch 3 | Step 23250 | Loss: 0.2071 | LM: 0.1963 | LB: 1.0870 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.346 | HR1: 0.415/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 13:58:12] Epoch 3 | Step 23260 | Loss: 0.2072 | LM: 0.1965 | LB: 1.0870 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.346 | HR1: 0.415/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 13:58:19] Epoch 3 | Step 23270 | Loss: 0.2071 | LM: 0.1962 | LB: 1.0871 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.346 | HR1: 0.415/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 13:58:25] Epoch 3 | Step 23280 | Loss: 0.2072 | LM: 0.1960 | LB: 1.0870 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.346 | HR1: 0.415/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 13:58:31] Epoch 3 | Step 23290 | Loss: 0.2072 | LM: 0.1960 | LB: 1.0871 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.346 | HR1: 0.415/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 13:58:38] Epoch 3 | Step 23300 | Loss: 0.2074 | LM: 0.1961 | LB: 1.0871 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.346 | HR1: 0.415/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 13:58:44] Epoch 3 | Step 23310 | Loss: 0.2075 | LM: 0.1962 | LB: 1.0872 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.346 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 13:58:51] Epoch 3 | Step 23320 | Loss: 0.2076 | LM: 0.1965 | LB: 1.0871 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.346 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 13:58:57] Epoch 3 | Step 23330 | Loss: 0.2075 | LM: 0.1965 | LB: 1.0872 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.346 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 13:59:03] Epoch 3 | Step 23340 | Loss: 0.2076 | LM: 0.1971 | LB: 1.0872 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.346 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 13:59:10] Epoch 3 | Step 23350 | Loss: 0.2077 | LM: 0.1970 | LB: 1.0873 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.346 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 13:59:16] Epoch 3 | Step 23360 | Loss: 0.2077 | LM: 0.1971 | LB: 1.0873 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.346 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 13:59:23] Epoch 3 | Step 23370 | Loss: 0.2075 | LM: 0.1969 | LB: 1.0873 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.346 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 13:59:29] Epoch 3 | Step 23380 | Loss: 0.2076 | LM: 0.1967 | LB: 1.0873 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.346 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 13:59:36] Epoch 3 | Step 23390 | Loss: 0.2075 | LM: 0.1967 | LB: 1.0873 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.346 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 13:59:42] Epoch 3 | Step 23400 | Loss: 0.2075 | LM: 0.1967 | LB: 1.0873 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.346 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 13:59:49] Epoch 3 | Step 23410 | Loss: 0.2074 | LM: 0.1967 | LB: 1.0874 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.346 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 13:59:55] Epoch 3 | Step 23420 | Loss: 0.2074 | LM: 0.1969 | LB: 1.0874 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 14:00:02] Epoch 3 | Step 23430 | Loss: 0.2074 | LM: 0.1969 | LB: 1.0873 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.346 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 14:00:09] Epoch 3 | Step 23440 | Loss: 0.2074 | LM: 0.1967 | LB: 1.0873 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 14:00:16] Epoch 3 | Step 23450 | Loss: 0.2074 | LM: 0.1967 | LB: 1.0874 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 14:00:22] Epoch 3 | Step 23460 | Loss: 0.2074 | LM: 0.1965 | LB: 1.0874 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 14:00:29] Epoch 3 | Step 23470 | Loss: 0.2074 | LM: 0.1968 | LB: 1.0873 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 14:00:35] Epoch 3 | Step 23480 | Loss: 0.2074 | LM: 0.1968 | LB: 1.0873 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 14:00:41] Epoch 3 | Step 23490 | Loss: 0.2074 | LM: 0.1967 | LB: 1.0873 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 14:00:48] Epoch 3 | Step 23500 | Loss: 0.2074 | LM: 0.1967 | LB: 1.0874 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 14:00:54] Epoch 3 | Step 23510 | Loss: 0.2075 | LM: 0.1967 | LB: 1.0874 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 14:01:01] Epoch 3 | Step 23520 | Loss: 0.2075 | LM: 0.1963 | LB: 1.0874 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 14:01:08] Epoch 3 | Step 23530 | Loss: 0.2076 | LM: 0.1964 | LB: 1.0873 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 14:01:14] Epoch 3 | Step 23540 | Loss: 0.2075 | LM: 0.1961 | LB: 1.0874 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.415/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 14:01:21] Epoch 3 | Step 23550 | Loss: 0.2075 | LM: 0.1966 | LB: 1.0873 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.415/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 14:01:27] Epoch 3 | Step 23560 | Loss: 0.2074 | LM: 0.1963 | LB: 1.0873 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.415/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 14:01:33] Epoch 3 | Step 23570 | Loss: 0.2074 | LM: 0.1965 | LB: 1.0873 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.415/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 14:01:40] Epoch 3 | Step 23580 | Loss: 0.2073 | LM: 0.1965 | LB: 1.0873 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.415/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 14:01:46] Epoch 3 | Step 23590 | Loss: 0.2072 | LM: 0.1963 | LB: 1.0873 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.415/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 14:01:52] Epoch 3 | Step 23600 | Loss: 0.2072 | LM: 0.1963 | LB: 1.0873 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.415/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 14:01:59] Epoch 3 | Step 23610 | Loss: 0.2073 | LM: 0.1963 | LB: 1.0873 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.415/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 14:02:05] Epoch 3 | Step 23620 | Loss: 0.2072 | LM: 0.1964 | LB: 1.0874 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 14:02:12] Epoch 3 | Step 23630 | Loss: 0.2073 | LM: 0.1962 | LB: 1.0874 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 14:02:18] Epoch 3 | Step 23640 | Loss: 0.2073 | LM: 0.1962 | LB: 1.0874 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.415/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 14:02:25] Epoch 3 | Step 23650 | Loss: 0.2073 | LM: 0.1962 | LB: 1.0873 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.415/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 14:02:31] Epoch 3 | Step 23660 | Loss: 0.2072 | LM: 0.1960 | LB: 1.0873 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.415/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 14:02:38] Epoch 3 | Step 23670 | Loss: 0.2072 | LM: 0.1962 | LB: 1.0873 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.415/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 14:02:44] Epoch 3 | Step 23680 | Loss: 0.2072 | LM: 0.1961 | LB: 1.0873 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.415/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 14:02:50] Epoch 3 | Step 23690 | Loss: 0.2073 | LM: 0.1960 | LB: 1.0873 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.415/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 14:02:56] Epoch 3 | Step 23700 | Loss: 0.2072 | LM: 0.1959 | LB: 1.0873 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.415/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 14:03:03] Epoch 3 | Step 23710 | Loss: 0.2073 | LM: 0.1960 | LB: 1.0873 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.415/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 14:03:09] Epoch 3 | Step 23720 | Loss: 0.2072 | LM: 0.1958 | LB: 1.0873 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.415/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 14:03:16] Epoch 3 | Step 23730 | Loss: 0.2072 | LM: 0.1956 | LB: 1.0872 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.415/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 14:03:22] Epoch 3 | Step 23740 | Loss: 0.2072 | LM: 0.1956 | LB: 1.0872 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.415/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 14:03:29] Epoch 3 | Step 23750 | Loss: 0.2072 | LM: 0.1956 | LB: 1.0873 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 14:03:35] Epoch 3 | Step 23760 | Loss: 0.2071 | LM: 0.1957 | LB: 1.0873 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 14:03:41] Epoch 3 | Step 23770 | Loss: 0.2072 | LM: 0.1957 | LB: 1.0872 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.346 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 14:03:47] Epoch 3 | Step 23780 | Loss: 0.2072 | LM: 0.1956 | LB: 1.0873 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.346 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 14:03:54] Epoch 3 | Step 23790 | Loss: 0.2071 | LM: 0.1954 | LB: 1.0872 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.346 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 14:04:00] Epoch 3 | Step 23800 | Loss: 0.2070 | LM: 0.1953 | LB: 1.0873 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.346 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 14:04:07] Epoch 3 | Step 23810 | Loss: 0.2070 | LM: 0.1952 | LB: 1.0873 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 14:04:13] Epoch 3 | Step 23820 | Loss: 0.2070 | LM: 0.1951 | LB: 1.0873 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 14:04:19] Epoch 3 | Step 23830 | Loss: 0.2069 | LM: 0.1950 | LB: 1.0874 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 14:04:26] Epoch 3 | Step 23840 | Loss: 0.2070 | LM: 0.1953 | LB: 1.0874 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 14:04:33] Epoch 3 | Step 23850 | Loss: 0.2070 | LM: 0.1951 | LB: 1.0874 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 14:04:39] Epoch 3 | Step 23860 | Loss: 0.2070 | LM: 0.1951 | LB: 1.0874 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 14:04:46] Epoch 3 | Step 23870 | Loss: 0.2070 | LM: 0.1950 | LB: 1.0874 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 14:04:52] Epoch 3 | Step 23880 | Loss: 0.2070 | LM: 0.1952 | LB: 1.0874 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 14:04:59] Epoch 3 | Step 23890 | Loss: 0.2067 | LM: 0.1949 | LB: 1.0875 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 14:05:05] Epoch 3 | Step 23900 | Loss: 0.2067 | LM: 0.1951 | LB: 1.0875 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 14:05:12] Epoch 3 | Step 23910 | Loss: 0.2066 | LM: 0.1952 | LB: 1.0875 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 14:05:18] Epoch 3 | Step 23920 | Loss: 0.2067 | LM: 0.1952 | LB: 1.0875 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 14:05:25] Epoch 3 | Step 23930 | Loss: 0.2068 | LM: 0.1958 | LB: 1.0876 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 14:05:32] Epoch 3 | Step 23940 | Loss: 0.2068 | LM: 0.1958 | LB: 1.0876 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 14:05:38] Epoch 3 | Step 23950 | Loss: 0.2069 | LM: 0.1958 | LB: 1.0875 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 14:05:45] Epoch 3 | Step 23960 | Loss: 0.2069 | LM: 0.1959 | LB: 1.0875 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 14:05:51] Epoch 3 | Step 23970 | Loss: 0.2070 | LM: 0.1958 | LB: 1.0875 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 14:05:57] Epoch 3 | Step 23980 | Loss: 0.2070 | LM: 0.1959 | LB: 1.0875 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 14:06:04] Epoch 3 | Step 23990 | Loss: 0.2070 | LM: 0.1959 | LB: 1.0875 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 14:06:10] Epoch 3 | Step 24000 | Loss: 0.2070 | LM: 0.1958 | LB: 1.0875 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 14:06:19] Checkpoint saved: outputs/2026-04-17/08-57-56/checkpoints/checkpoint_step_24000.pt [2026-04-17 14:06:35] Validation | Batch 10/784 | Loss: 0.3345 | LM_LOSS: 0.3236 | LB_LOSS: 1.0846 [2026-04-17 14:06:37] Validation | Batch 20/784 | Loss: 0.3460 | LM_LOSS: 0.3351 | LB_LOSS: 1.0849 [2026-04-17 14:06:38] Validation | Batch 30/784 | Loss: 0.3317 | LM_LOSS: 0.3209 | LB_LOSS: 1.0841 [2026-04-17 14:06:39] Validation | Batch 40/784 | Loss: 0.3344 | LM_LOSS: 0.3236 | LB_LOSS: 1.0840 [2026-04-17 14:06:41] Validation | Batch 50/784 | Loss: 0.3319 | LM_LOSS: 0.3211 | LB_LOSS: 1.0834 [2026-04-17 14:06:42] Validation | Batch 60/784 | Loss: 0.3339 | LM_LOSS: 0.3231 | LB_LOSS: 1.0830 [2026-04-17 14:06:43] Validation | Batch 70/784 | Loss: 0.3312 | LM_LOSS: 0.3204 | LB_LOSS: 1.0823 [2026-04-17 14:06:45] Validation | Batch 80/784 | Loss: 0.3275 | LM_LOSS: 0.3167 | LB_LOSS: 1.0819 [2026-04-17 14:06:46] Validation | Batch 90/784 | Loss: 0.3264 | LM_LOSS: 0.3156 | LB_LOSS: 1.0824 [2026-04-17 14:06:47] Validation | Batch 100/784 | Loss: 0.3285 | LM_LOSS: 0.3176 | LB_LOSS: 1.0829 [2026-04-17 14:06:49] Validation | Batch 110/784 | Loss: 0.3230 | LM_LOSS: 0.3121 | LB_LOSS: 1.0830 [2026-04-17 14:06:50] Validation | Batch 120/784 | Loss: 0.3265 | LM_LOSS: 0.3157 | LB_LOSS: 1.0829 [2026-04-17 14:06:51] Validation | Batch 130/784 | Loss: 0.3296 | LM_LOSS: 0.3188 | LB_LOSS: 1.0829 [2026-04-17 14:06:53] Validation | Batch 140/784 | Loss: 0.3290 | LM_LOSS: 0.3182 | LB_LOSS: 1.0827 [2026-04-17 14:06:54] Validation | Batch 150/784 | Loss: 0.3250 | LM_LOSS: 0.3142 | LB_LOSS: 1.0830 [2026-04-17 14:06:56] Validation | Batch 160/784 | Loss: 0.3258 | LM_LOSS: 0.3150 | LB_LOSS: 1.0827 [2026-04-17 14:06:57] Validation | Batch 170/784 | Loss: 0.3259 | LM_LOSS: 0.3151 | LB_LOSS: 1.0824 [2026-04-17 14:06:59] Validation | Batch 180/784 | Loss: 0.3235 | LM_LOSS: 0.3126 | LB_LOSS: 1.0824 [2026-04-17 14:07:00] Validation | Batch 190/784 | Loss: 0.3257 | LM_LOSS: 0.3148 | LB_LOSS: 1.0829 [2026-04-17 14:07:01] Validation | Batch 200/784 | Loss: 0.3261 | LM_LOSS: 0.3153 | LB_LOSS: 1.0829 [2026-04-17 14:07:02] Validation | Batch 210/784 | Loss: 0.3250 | LM_LOSS: 0.3141 | LB_LOSS: 1.0828 [2026-04-17 14:07:04] Validation | Batch 220/784 | Loss: 0.3258 | LM_LOSS: 0.3150 | LB_LOSS: 1.0829 [2026-04-17 14:07:05] Validation | Batch 230/784 | Loss: 0.3264 | LM_LOSS: 0.3156 | LB_LOSS: 1.0828 [2026-04-17 14:07:07] Validation | Batch 240/784 | Loss: 0.3269 | LM_LOSS: 0.3161 | LB_LOSS: 1.0831 [2026-04-17 14:07:08] Validation | Batch 250/784 | Loss: 0.3268 | LM_LOSS: 0.3160 | LB_LOSS: 1.0830 [2026-04-17 14:07:09] Validation | Batch 260/784 | Loss: 0.3270 | LM_LOSS: 0.3162 | LB_LOSS: 1.0832 [2026-04-17 14:07:11] Validation | Batch 270/784 | Loss: 0.3269 | LM_LOSS: 0.3161 | LB_LOSS: 1.0832 [2026-04-17 14:07:12] Validation | Batch 280/784 | Loss: 0.3273 | LM_LOSS: 0.3165 | LB_LOSS: 1.0834 [2026-04-17 14:07:14] Validation | Batch 290/784 | Loss: 0.3284 | LM_LOSS: 0.3176 | LB_LOSS: 1.0835 [2026-04-17 14:07:15] Validation | Batch 300/784 | Loss: 0.3292 | LM_LOSS: 0.3184 | LB_LOSS: 1.0836 [2026-04-17 14:07:16] Validation | Batch 310/784 | Loss: 0.3286 | LM_LOSS: 0.3178 | LB_LOSS: 1.0835 [2026-04-17 14:07:18] Validation | Batch 320/784 | Loss: 0.3302 | LM_LOSS: 0.3194 | LB_LOSS: 1.0835 [2026-04-17 14:07:19] Validation | Batch 330/784 | Loss: 0.3300 | LM_LOSS: 0.3192 | LB_LOSS: 1.0835 [2026-04-17 14:07:20] Validation | Batch 340/784 | Loss: 0.3288 | LM_LOSS: 0.3180 | LB_LOSS: 1.0836 [2026-04-17 14:07:22] Validation | Batch 350/784 | Loss: 0.3290 | LM_LOSS: 0.3182 | LB_LOSS: 1.0838 [2026-04-17 14:07:23] Validation | Batch 360/784 | Loss: 0.3288 | LM_LOSS: 0.3180 | LB_LOSS: 1.0838 [2026-04-17 14:07:24] Validation | Batch 370/784 | Loss: 0.3293 | LM_LOSS: 0.3185 | LB_LOSS: 1.0837 [2026-04-17 14:07:25] Validation | Batch 380/784 | Loss: 0.3291 | LM_LOSS: 0.3183 | LB_LOSS: 1.0837 [2026-04-17 14:07:27] Validation | Batch 390/784 | Loss: 0.3290 | LM_LOSS: 0.3182 | LB_LOSS: 1.0838 [2026-04-17 14:07:28] Validation | Batch 400/784 | Loss: 0.3293 | LM_LOSS: 0.3185 | LB_LOSS: 1.0838 [2026-04-17 14:07:29] Validation | Batch 410/784 | Loss: 0.3296 | LM_LOSS: 0.3188 | LB_LOSS: 1.0838 [2026-04-17 14:07:30] Validation | Batch 420/784 | Loss: 0.3299 | LM_LOSS: 0.3191 | LB_LOSS: 1.0839 [2026-04-17 14:07:32] Validation | Batch 430/784 | Loss: 0.3300 | LM_LOSS: 0.3192 | LB_LOSS: 1.0838 [2026-04-17 14:07:33] Validation | Batch 440/784 | Loss: 0.3297 | LM_LOSS: 0.3188 | LB_LOSS: 1.0838 [2026-04-17 14:07:34] Validation | Batch 450/784 | Loss: 0.3289 | LM_LOSS: 0.3181 | LB_LOSS: 1.0838 [2026-04-17 14:07:36] Validation | Batch 460/784 | Loss: 0.3294 | LM_LOSS: 0.3186 | LB_LOSS: 1.0839 [2026-04-17 14:07:37] Validation | Batch 470/784 | Loss: 0.3286 | LM_LOSS: 0.3178 | LB_LOSS: 1.0838 [2026-04-17 14:07:38] Validation | Batch 480/784 | Loss: 0.3291 | LM_LOSS: 0.3183 | LB_LOSS: 1.0838 [2026-04-17 14:07:40] Validation | Batch 490/784 | Loss: 0.3284 | LM_LOSS: 0.3176 | LB_LOSS: 1.0837 [2026-04-17 14:07:41] Validation | Batch 500/784 | Loss: 0.3288 | LM_LOSS: 0.3180 | LB_LOSS: 1.0837 [2026-04-17 14:07:42] Validation | Batch 510/784 | Loss: 0.3285 | LM_LOSS: 0.3177 | LB_LOSS: 1.0836 [2026-04-17 14:07:44] Validation | Batch 520/784 | Loss: 0.3287 | LM_LOSS: 0.3179 | LB_LOSS: 1.0836 [2026-04-17 14:07:45] Validation | Batch 530/784 | Loss: 0.3296 | LM_LOSS: 0.3188 | LB_LOSS: 1.0835 [2026-04-17 14:07:46] Validation | Batch 540/784 | Loss: 0.3300 | LM_LOSS: 0.3192 | LB_LOSS: 1.0835 [2026-04-17 14:07:48] Validation | Batch 550/784 | Loss: 0.3313 | LM_LOSS: 0.3205 | LB_LOSS: 1.0835 [2026-04-17 14:07:49] Validation | Batch 560/784 | Loss: 0.3314 | LM_LOSS: 0.3206 | LB_LOSS: 1.0835 [2026-04-17 14:07:51] Validation | Batch 570/784 | Loss: 0.3309 | LM_LOSS: 0.3201 | LB_LOSS: 1.0835 [2026-04-17 14:07:52] Validation | Batch 580/784 | Loss: 0.3304 | LM_LOSS: 0.3195 | LB_LOSS: 1.0835 [2026-04-17 14:07:53] Validation | Batch 590/784 | Loss: 0.3306 | LM_LOSS: 0.3198 | LB_LOSS: 1.0834 [2026-04-17 14:07:55] Validation | Batch 600/784 | Loss: 0.3305 | LM_LOSS: 0.3197 | LB_LOSS: 1.0834 [2026-04-17 14:07:56] Validation | Batch 610/784 | Loss: 0.3306 | LM_LOSS: 0.3198 | LB_LOSS: 1.0834 [2026-04-17 14:07:57] Validation | Batch 620/784 | Loss: 0.3305 | LM_LOSS: 0.3196 | LB_LOSS: 1.0834 [2026-04-17 14:07:59] Validation | Batch 630/784 | Loss: 0.3313 | LM_LOSS: 0.3204 | LB_LOSS: 1.0834 [2026-04-17 14:08:01] Validation | Batch 640/784 | Loss: 0.3313 | LM_LOSS: 0.3205 | LB_LOSS: 1.0834 [2026-04-17 14:08:02] Validation | Batch 650/784 | Loss: 0.3312 | LM_LOSS: 0.3203 | LB_LOSS: 1.0835 [2026-04-17 14:08:03] Validation | Batch 660/784 | Loss: 0.3316 | LM_LOSS: 0.3207 | LB_LOSS: 1.0834 [2026-04-17 14:08:05] Validation | Batch 670/784 | Loss: 0.3320 | LM_LOSS: 0.3212 | LB_LOSS: 1.0835 [2026-04-17 14:08:06] Validation | Batch 680/784 | Loss: 0.3317 | LM_LOSS: 0.3209 | LB_LOSS: 1.0835 [2026-04-17 14:08:07] Validation | Batch 690/784 | Loss: 0.3318 | LM_LOSS: 0.3210 | LB_LOSS: 1.0834 [2026-04-17 14:08:09] Validation | Batch 700/784 | Loss: 0.3319 | LM_LOSS: 0.3211 | LB_LOSS: 1.0834 [2026-04-17 14:08:10] Validation | Batch 710/784 | Loss: 0.3317 | LM_LOSS: 0.3208 | LB_LOSS: 1.0834 [2026-04-17 14:08:12] Validation | Batch 720/784 | Loss: 0.3314 | LM_LOSS: 0.3205 | LB_LOSS: 1.0833 [2026-04-17 14:08:13] Validation | Batch 730/784 | Loss: 0.3309 | LM_LOSS: 0.3200 | LB_LOSS: 1.0832 [2026-04-17 14:08:14] Validation | Batch 740/784 | Loss: 0.3309 | LM_LOSS: 0.3201 | LB_LOSS: 1.0833 [2026-04-17 14:08:15] Validation | Batch 750/784 | Loss: 0.3302 | LM_LOSS: 0.3194 | LB_LOSS: 1.0833 [2026-04-17 14:08:17] Validation | Batch 760/784 | Loss: 0.3304 | LM_LOSS: 0.3196 | LB_LOSS: 1.0833 [2026-04-17 14:08:18] Validation | Batch 770/784 | Loss: 0.3306 | LM_LOSS: 0.3197 | LB_LOSS: 1.0834 [2026-04-17 14:08:19] Validation | Batch 780/784 | Loss: 0.3309 | LM_LOSS: 0.3201 | LB_LOSS: 1.0833 [2026-04-17 14:08:20] Validation | Batch 784/784 | Loss: 0.3311 | LM_LOSS: 0.3203 | LB_LOSS: 1.0833 [2026-04-17 14:08:22] Validation | Loss: 0.3311 | LM_LOSS: 0.3203 | LB_LOSS: 1.0833 | PPL: 1.38 | Time: 106.03s [2026-04-17 14:08:29] Epoch 3 | Step 24010 | Loss: 0.2070 | LM: 0.1959 | LB: 1.0875 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.346 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 14:08:35] Epoch 3 | Step 24020 | Loss: 0.2069 | LM: 0.1959 | LB: 1.0875 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 14:08:42] Epoch 3 | Step 24030 | Loss: 0.2069 | LM: 0.1958 | LB: 1.0875 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 14:08:48] Epoch 3 | Step 24040 | Loss: 0.2068 | LM: 0.1955 | LB: 1.0875 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 14:08:55] Epoch 3 | Step 24050 | Loss: 0.2067 | LM: 0.1955 | LB: 1.0875 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 14:09:01] Epoch 3 | Step 24060 | Loss: 0.2068 | LM: 0.1956 | LB: 1.0875 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 14:09:08] Epoch 3 | Step 24070 | Loss: 0.2067 | LM: 0.1954 | LB: 1.0875 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 14:09:14] Epoch 3 | Step 24080 | Loss: 0.2066 | LM: 0.1952 | LB: 1.0875 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 14:09:20] Epoch 3 | Step 24090 | Loss: 0.2064 | LM: 0.1950 | LB: 1.0875 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 14:09:27] Epoch 3 | Step 24100 | Loss: 0.2065 | LM: 0.1950 | LB: 1.0875 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 14:09:33] Epoch 3 | Step 24110 | Loss: 0.2066 | LM: 0.1952 | LB: 1.0875 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 14:09:40] Epoch 3 | Step 24120 | Loss: 0.2067 | LM: 0.1953 | LB: 1.0875 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 14:09:46] Epoch 3 | Step 24130 | Loss: 0.2068 | LM: 0.1952 | LB: 1.0875 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 14:09:53] Epoch 3 | Step 24140 | Loss: 0.2068 | LM: 0.1953 | LB: 1.0875 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 14:09:59] Epoch 3 | Step 24150 | Loss: 0.2068 | LM: 0.1954 | LB: 1.0875 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 14:10:05] Epoch 3 | Step 24160 | Loss: 0.2069 | LM: 0.1953 | LB: 1.0876 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 14:10:12] Epoch 3 | Step 24170 | Loss: 0.2069 | LM: 0.1954 | LB: 1.0876 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 14:10:18] Epoch 3 | Step 24180 | Loss: 0.2069 | LM: 0.1953 | LB: 1.0876 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 14:10:25] Epoch 3 | Step 24190 | Loss: 0.2070 | LM: 0.1953 | LB: 1.0876 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 14:10:31] Epoch 3 | Step 24200 | Loss: 0.2070 | LM: 0.1954 | LB: 1.0875 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 14:10:37] Epoch 3 | Step 24210 | Loss: 0.2070 | LM: 0.1952 | LB: 1.0875 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 14:10:44] Epoch 3 | Step 24220 | Loss: 0.2070 | LM: 0.1951 | LB: 1.0875 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 14:10:50] Epoch 3 | Step 24230 | Loss: 0.2071 | LM: 0.1952 | LB: 1.0875 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 14:10:56] Epoch 3 | Step 24240 | Loss: 0.2071 | LM: 0.1954 | LB: 1.0875 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 14:11:03] Epoch 3 | Step 24250 | Loss: 0.2071 | LM: 0.1953 | LB: 1.0875 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 14:11:10] Epoch 3 | Step 24260 | Loss: 0.2071 | LM: 0.1952 | LB: 1.0875 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 14:11:16] Epoch 3 | Step 24270 | Loss: 0.2070 | LM: 0.1953 | LB: 1.0875 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 14:11:22] Epoch 3 | Step 24280 | Loss: 0.2071 | LM: 0.1954 | LB: 1.0875 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 14:11:29] Epoch 3 | Step 24290 | Loss: 0.2071 | LM: 0.1956 | LB: 1.0875 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 14:11:35] Epoch 3 | Step 24300 | Loss: 0.2072 | LM: 0.1955 | LB: 1.0875 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 14:11:41] Epoch 3 | Step 24310 | Loss: 0.2072 | LM: 0.1955 | LB: 1.0875 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 14:11:48] Epoch 3 | Step 24320 | Loss: 0.2073 | LM: 0.1955 | LB: 1.0875 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 14:11:55] Epoch 3 | Step 24330 | Loss: 0.2073 | LM: 0.1956 | LB: 1.0876 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 14:12:01] Epoch 3 | Step 24340 | Loss: 0.2073 | LM: 0.1955 | LB: 1.0875 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 14:12:07] Epoch 3 | Step 24350 | Loss: 0.2073 | LM: 0.1956 | LB: 1.0875 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 14:12:14] Epoch 3 | Step 24360 | Loss: 0.2073 | LM: 0.1955 | LB: 1.0875 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 14:12:20] Epoch 3 | Step 24370 | Loss: 0.2073 | LM: 0.1955 | LB: 1.0876 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 14:12:26] Epoch 3 | Step 24380 | Loss: 0.2071 | LM: 0.1953 | LB: 1.0875 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 14:12:32] Epoch 3 | Step 24390 | Loss: 0.2071 | LM: 0.1955 | LB: 1.0875 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 14:12:39] Epoch 3 | Step 24400 | Loss: 0.2070 | LM: 0.1956 | LB: 1.0875 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 14:12:45] Epoch 3 | Step 24410 | Loss: 0.2070 | LM: 0.1956 | LB: 1.0875 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 14:12:52] Epoch 3 | Step 24420 | Loss: 0.2070 | LM: 0.1956 | LB: 1.0875 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 14:12:58] Epoch 3 | Step 24430 | Loss: 0.2069 | LM: 0.1956 | LB: 1.0875 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 14:13:05] Epoch 3 | Step 24440 | Loss: 0.2069 | LM: 0.1957 | LB: 1.0874 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 14:13:11] Epoch 3 | Step 24450 | Loss: 0.2069 | LM: 0.1957 | LB: 1.0875 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 14:13:18] Epoch 3 | Step 24460 | Loss: 0.2067 | LM: 0.1956 | LB: 1.0875 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 14:13:24] Epoch 3 | Step 24470 | Loss: 0.2068 | LM: 0.1956 | LB: 1.0875 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 14:13:30] Epoch 3 | Step 24480 | Loss: 0.2067 | LM: 0.1956 | LB: 1.0875 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 14:13:37] Epoch 3 | Step 24490 | Loss: 0.2067 | LM: 0.1958 | LB: 1.0875 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 14:13:43] Epoch 3 | Step 24500 | Loss: 0.2066 | LM: 0.1956 | LB: 1.0875 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 14:13:49] Epoch 3 | Step 24510 | Loss: 0.2066 | LM: 0.1955 | LB: 1.0875 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 14:13:56] Epoch 3 | Step 24520 | Loss: 0.2066 | LM: 0.1957 | LB: 1.0875 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 14:14:02] Epoch 3 | Step 24530 | Loss: 0.2066 | LM: 0.1957 | LB: 1.0875 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 14:14:08] Epoch 3 | Step 24540 | Loss: 0.2066 | LM: 0.1958 | LB: 1.0875 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 14:14:15] Epoch 3 | Step 24550 | Loss: 0.2066 | LM: 0.1957 | LB: 1.0875 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 14:14:21] Epoch 3 | Step 24560 | Loss: 0.2066 | LM: 0.1956 | LB: 1.0875 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 14:14:28] Epoch 3 | Step 24570 | Loss: 0.2067 | LM: 0.1957 | LB: 1.0875 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 14:14:34] Epoch 3 | Step 24580 | Loss: 0.2066 | LM: 0.1956 | LB: 1.0875 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 14:14:40] Epoch 3 | Step 24590 | Loss: 0.2066 | LM: 0.1958 | LB: 1.0875 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 14:14:46] Epoch 3 | Step 24600 | Loss: 0.2067 | LM: 0.1962 | LB: 1.0875 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 14:14:53] Epoch 3 | Step 24610 | Loss: 0.2066 | LM: 0.1960 | LB: 1.0874 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 14:14:59] Epoch 3 | Step 24620 | Loss: 0.2066 | LM: 0.1961 | LB: 1.0874 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 14:15:05] Epoch 3 | Step 24630 | Loss: 0.2065 | LM: 0.1958 | LB: 1.0874 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 14:15:12] Epoch 3 | Step 24640 | Loss: 0.2066 | LM: 0.1960 | LB: 1.0874 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 14:15:18] Epoch 3 | Step 24650 | Loss: 0.2065 | LM: 0.1959 | LB: 1.0874 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 14:15:24] Epoch 3 | Step 24660 | Loss: 0.2065 | LM: 0.1959 | LB: 1.0874 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 14:15:31] Epoch 3 | Step 24670 | Loss: 0.2065 | LM: 0.1959 | LB: 1.0874 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 14:15:37] Epoch 3 | Step 24680 | Loss: 0.2065 | LM: 0.1959 | LB: 1.0874 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 14:15:43] Epoch 3 | Step 24690 | Loss: 0.2064 | LM: 0.1958 | LB: 1.0874 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 14:15:49] Epoch 3 | Step 24700 | Loss: 0.2063 | LM: 0.1958 | LB: 1.0874 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 14:15:56] Epoch 3 | Step 24710 | Loss: 0.2063 | LM: 0.1956 | LB: 1.0874 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 14:16:02] Epoch 3 | Step 24720 | Loss: 0.2063 | LM: 0.1956 | LB: 1.0874 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 14:16:08] Epoch 3 | Step 24730 | Loss: 0.2063 | LM: 0.1956 | LB: 1.0874 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 14:16:15] Epoch 3 | Step 24740 | Loss: 0.2063 | LM: 0.1956 | LB: 1.0874 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 14:16:21] Epoch 3 | Step 24750 | Loss: 0.2062 | LM: 0.1957 | LB: 1.0874 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 14:16:28] Epoch 3 | Step 24760 | Loss: 0.2063 | LM: 0.1958 | LB: 1.0874 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 14:16:34] Epoch 3 | Step 24770 | Loss: 0.2063 | LM: 0.1957 | LB: 1.0874 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 14:16:40] Epoch 3 | Step 24780 | Loss: 0.2063 | LM: 0.1957 | LB: 1.0875 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 14:16:47] Epoch 3 | Step 24790 | Loss: 0.2062 | LM: 0.1955 | LB: 1.0874 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 14:16:53] Epoch 3 | Step 24800 | Loss: 0.2062 | LM: 0.1956 | LB: 1.0874 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 14:16:59] Epoch 3 | Step 24810 | Loss: 0.2062 | LM: 0.1955 | LB: 1.0874 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 14:17:06] Epoch 3 | Step 24820 | Loss: 0.2062 | LM: 0.1955 | LB: 1.0874 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 14:17:12] Epoch 3 | Step 24830 | Loss: 0.2062 | LM: 0.1956 | LB: 1.0874 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 14:17:18] Epoch 3 | Step 24840 | Loss: 0.2063 | LM: 0.1956 | LB: 1.0874 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 14:17:25] Epoch 3 | Step 24850 | Loss: 0.2064 | LM: 0.1957 | LB: 1.0874 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 14:17:32] Epoch 3 | Step 24860 | Loss: 0.2064 | LM: 0.1956 | LB: 1.0874 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 14:17:38] Epoch 3 | Step 24870 | Loss: 0.2065 | LM: 0.1957 | LB: 1.0874 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 14:17:44] Epoch 3 | Step 24880 | Loss: 0.2064 | LM: 0.1958 | LB: 1.0874 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 14:17:51] Epoch 3 | Step 24890 | Loss: 0.2064 | LM: 0.1958 | LB: 1.0874 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 14:17:57] Epoch 3 | Step 24900 | Loss: 0.2064 | LM: 0.1959 | LB: 1.0874 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 14:18:03] Epoch 3 | Step 24910 | Loss: 0.2065 | LM: 0.1958 | LB: 1.0874 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 14:18:10] Epoch 3 | Step 24920 | Loss: 0.2065 | LM: 0.1958 | LB: 1.0874 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.415/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 14:18:16] Epoch 3 | Step 24930 | Loss: 0.2065 | LM: 0.1958 | LB: 1.0875 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 14:18:23] Epoch 3 | Step 24940 | Loss: 0.2065 | LM: 0.1957 | LB: 1.0875 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 14:18:29] Epoch 3 | Step 24950 | Loss: 0.2064 | LM: 0.1955 | LB: 1.0874 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 14:18:35] Epoch 3 | Step 24960 | Loss: 0.2064 | LM: 0.1954 | LB: 1.0874 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 14:18:42] Epoch 3 | Step 24970 | Loss: 0.2063 | LM: 0.1953 | LB: 1.0874 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 14:18:48] Epoch 3 | Step 24980 | Loss: 0.2063 | LM: 0.1953 | LB: 1.0875 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 14:18:54] Epoch 3 | Step 24990 | Loss: 0.2062 | LM: 0.1951 | LB: 1.0875 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 14:19:01] Epoch 3 | Step 25000 | Loss: 0.2063 | LM: 0.1953 | LB: 1.0875 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 14:19:02] Validation | Batch 10/784 | Loss: 0.3348 | LM_LOSS: 0.3240 | LB_LOSS: 1.0847 [2026-04-17 14:19:04] Validation | Batch 20/784 | Loss: 0.3464 | LM_LOSS: 0.3356 | LB_LOSS: 1.0849 [2026-04-17 14:19:05] Validation | Batch 30/784 | Loss: 0.3322 | LM_LOSS: 0.3213 | LB_LOSS: 1.0842 [2026-04-17 14:19:07] Validation | Batch 40/784 | Loss: 0.3349 | LM_LOSS: 0.3241 | LB_LOSS: 1.0841 [2026-04-17 14:19:08] Validation | Batch 50/784 | Loss: 0.3324 | LM_LOSS: 0.3216 | LB_LOSS: 1.0834 [2026-04-17 14:19:09] Validation | Batch 60/784 | Loss: 0.3343 | LM_LOSS: 0.3235 | LB_LOSS: 1.0830 [2026-04-17 14:19:11] Validation | Batch 70/784 | Loss: 0.3316 | LM_LOSS: 0.3208 | LB_LOSS: 1.0823 [2026-04-17 14:19:12] Validation | Batch 80/784 | Loss: 0.3279 | LM_LOSS: 0.3171 | LB_LOSS: 1.0819 [2026-04-17 14:19:13] Validation | Batch 90/784 | Loss: 0.3268 | LM_LOSS: 0.3160 | LB_LOSS: 1.0824 [2026-04-17 14:19:15] Validation | Batch 100/784 | Loss: 0.3288 | LM_LOSS: 0.3180 | LB_LOSS: 1.0829 [2026-04-17 14:19:16] Validation | Batch 110/784 | Loss: 0.3233 | LM_LOSS: 0.3125 | LB_LOSS: 1.0830 [2026-04-17 14:19:17] Validation | Batch 120/784 | Loss: 0.3269 | LM_LOSS: 0.3161 | LB_LOSS: 1.0830 [2026-04-17 14:19:19] Validation | Batch 130/784 | Loss: 0.3300 | LM_LOSS: 0.3192 | LB_LOSS: 1.0829 [2026-04-17 14:19:20] Validation | Batch 140/784 | Loss: 0.3294 | LM_LOSS: 0.3186 | LB_LOSS: 1.0827 [2026-04-17 14:19:22] Validation | Batch 150/784 | Loss: 0.3254 | LM_LOSS: 0.3146 | LB_LOSS: 1.0830 [2026-04-17 14:19:23] Validation | Batch 160/784 | Loss: 0.3263 | LM_LOSS: 0.3154 | LB_LOSS: 1.0827 [2026-04-17 14:19:25] Validation | Batch 170/784 | Loss: 0.3264 | LM_LOSS: 0.3155 | LB_LOSS: 1.0824 [2026-04-17 14:19:26] Validation | Batch 180/784 | Loss: 0.3239 | LM_LOSS: 0.3131 | LB_LOSS: 1.0824 [2026-04-17 14:19:28] Validation | Batch 190/784 | Loss: 0.3261 | LM_LOSS: 0.3153 | LB_LOSS: 1.0829 [2026-04-17 14:19:29] Validation | Batch 200/784 | Loss: 0.3266 | LM_LOSS: 0.3158 | LB_LOSS: 1.0829 [2026-04-17 14:19:30] Validation | Batch 210/784 | Loss: 0.3254 | LM_LOSS: 0.3146 | LB_LOSS: 1.0828 [2026-04-17 14:19:32] Validation | Batch 220/784 | Loss: 0.3263 | LM_LOSS: 0.3155 | LB_LOSS: 1.0829 [2026-04-17 14:19:33] Validation | Batch 230/784 | Loss: 0.3269 | LM_LOSS: 0.3161 | LB_LOSS: 1.0828 [2026-04-17 14:19:34] Validation | Batch 240/784 | Loss: 0.3274 | LM_LOSS: 0.3165 | LB_LOSS: 1.0831 [2026-04-17 14:19:36] Validation | Batch 250/784 | Loss: 0.3273 | LM_LOSS: 0.3164 | LB_LOSS: 1.0830 [2026-04-17 14:19:37] Validation | Batch 260/784 | Loss: 0.3276 | LM_LOSS: 0.3167 | LB_LOSS: 1.0832 [2026-04-17 14:19:39] Validation | Batch 270/784 | Loss: 0.3274 | LM_LOSS: 0.3166 | LB_LOSS: 1.0832 [2026-04-17 14:19:40] Validation | Batch 280/784 | Loss: 0.3278 | LM_LOSS: 0.3170 | LB_LOSS: 1.0834 [2026-04-17 14:19:42] Validation | Batch 290/784 | Loss: 0.3289 | LM_LOSS: 0.3181 | LB_LOSS: 1.0835 [2026-04-17 14:19:43] Validation | Batch 300/784 | Loss: 0.3298 | LM_LOSS: 0.3189 | LB_LOSS: 1.0836 [2026-04-17 14:19:44] Validation | Batch 310/784 | Loss: 0.3291 | LM_LOSS: 0.3183 | LB_LOSS: 1.0835 [2026-04-17 14:19:46] Validation | Batch 320/784 | Loss: 0.3308 | LM_LOSS: 0.3199 | LB_LOSS: 1.0835 [2026-04-17 14:19:47] Validation | Batch 330/784 | Loss: 0.3306 | LM_LOSS: 0.3197 | LB_LOSS: 1.0835 [2026-04-17 14:19:48] Validation | Batch 340/784 | Loss: 0.3294 | LM_LOSS: 0.3185 | LB_LOSS: 1.0836 [2026-04-17 14:19:50] Validation | Batch 350/784 | Loss: 0.3296 | LM_LOSS: 0.3187 | LB_LOSS: 1.0838 [2026-04-17 14:19:51] Validation | Batch 360/784 | Loss: 0.3294 | LM_LOSS: 0.3185 | LB_LOSS: 1.0838 [2026-04-17 14:19:52] Validation | Batch 370/784 | Loss: 0.3299 | LM_LOSS: 0.3190 | LB_LOSS: 1.0837 [2026-04-17 14:19:53] Validation | Batch 380/784 | Loss: 0.3297 | LM_LOSS: 0.3188 | LB_LOSS: 1.0838 [2026-04-17 14:19:55] Validation | Batch 390/784 | Loss: 0.3296 | LM_LOSS: 0.3188 | LB_LOSS: 1.0838 [2026-04-17 14:19:56] Validation | Batch 400/784 | Loss: 0.3299 | LM_LOSS: 0.3191 | LB_LOSS: 1.0838 [2026-04-17 14:19:57] Validation | Batch 410/784 | Loss: 0.3302 | LM_LOSS: 0.3194 | LB_LOSS: 1.0838 [2026-04-17 14:19:58] Validation | Batch 420/784 | Loss: 0.3305 | LM_LOSS: 0.3196 | LB_LOSS: 1.0839 [2026-04-17 14:20:00] Validation | Batch 430/784 | Loss: 0.3306 | LM_LOSS: 0.3197 | LB_LOSS: 1.0838 [2026-04-17 14:20:01] Validation | Batch 440/784 | Loss: 0.3303 | LM_LOSS: 0.3194 | LB_LOSS: 1.0838 [2026-04-17 14:20:02] Validation | Batch 450/784 | Loss: 0.3295 | LM_LOSS: 0.3187 | LB_LOSS: 1.0838 [2026-04-17 14:20:04] Validation | Batch 460/784 | Loss: 0.3300 | LM_LOSS: 0.3192 | LB_LOSS: 1.0839 [2026-04-17 14:20:05] Validation | Batch 470/784 | Loss: 0.3292 | LM_LOSS: 0.3183 | LB_LOSS: 1.0838 [2026-04-17 14:20:06] Validation | Batch 480/784 | Loss: 0.3297 | LM_LOSS: 0.3189 | LB_LOSS: 1.0838 [2026-04-17 14:20:08] Validation | Batch 490/784 | Loss: 0.3290 | LM_LOSS: 0.3182 | LB_LOSS: 1.0837 [2026-04-17 14:20:09] Validation | Batch 500/784 | Loss: 0.3294 | LM_LOSS: 0.3186 | LB_LOSS: 1.0837 [2026-04-17 14:20:10] Validation | Batch 510/784 | Loss: 0.3291 | LM_LOSS: 0.3183 | LB_LOSS: 1.0837 [2026-04-17 14:20:12] Validation | Batch 520/784 | Loss: 0.3293 | LM_LOSS: 0.3185 | LB_LOSS: 1.0836 [2026-04-17 14:20:13] Validation | Batch 530/784 | Loss: 0.3302 | LM_LOSS: 0.3194 | LB_LOSS: 1.0835 [2026-04-17 14:20:14] Validation | Batch 540/784 | Loss: 0.3306 | LM_LOSS: 0.3197 | LB_LOSS: 1.0836 [2026-04-17 14:20:16] Validation | Batch 550/784 | Loss: 0.3319 | LM_LOSS: 0.3210 | LB_LOSS: 1.0835 [2026-04-17 14:20:17] Validation | Batch 560/784 | Loss: 0.3320 | LM_LOSS: 0.3212 | LB_LOSS: 1.0836 [2026-04-17 14:20:19] Validation | Batch 570/784 | Loss: 0.3315 | LM_LOSS: 0.3207 | LB_LOSS: 1.0835 [2026-04-17 14:20:20] Validation | Batch 580/784 | Loss: 0.3310 | LM_LOSS: 0.3201 | LB_LOSS: 1.0835 [2026-04-17 14:20:21] Validation | Batch 590/784 | Loss: 0.3312 | LM_LOSS: 0.3204 | LB_LOSS: 1.0834 [2026-04-17 14:20:23] Validation | Batch 600/784 | Loss: 0.3311 | LM_LOSS: 0.3203 | LB_LOSS: 1.0834 [2026-04-17 14:20:24] Validation | Batch 610/784 | Loss: 0.3312 | LM_LOSS: 0.3204 | LB_LOSS: 1.0834 [2026-04-17 14:20:25] Validation | Batch 620/784 | Loss: 0.3311 | LM_LOSS: 0.3202 | LB_LOSS: 1.0834 [2026-04-17 14:20:27] Validation | Batch 630/784 | Loss: 0.3319 | LM_LOSS: 0.3210 | LB_LOSS: 1.0834 [2026-04-17 14:20:29] Validation | Batch 640/784 | Loss: 0.3319 | LM_LOSS: 0.3211 | LB_LOSS: 1.0834 [2026-04-17 14:20:30] Validation | Batch 650/784 | Loss: 0.3318 | LM_LOSS: 0.3209 | LB_LOSS: 1.0835 [2026-04-17 14:20:32] Validation | Batch 660/784 | Loss: 0.3322 | LM_LOSS: 0.3213 | LB_LOSS: 1.0835 [2026-04-17 14:20:33] Validation | Batch 670/784 | Loss: 0.3326 | LM_LOSS: 0.3217 | LB_LOSS: 1.0835 [2026-04-17 14:20:34] Validation | Batch 680/784 | Loss: 0.3323 | LM_LOSS: 0.3214 | LB_LOSS: 1.0835 [2026-04-17 14:20:36] Validation | Batch 690/784 | Loss: 0.3324 | LM_LOSS: 0.3216 | LB_LOSS: 1.0835 [2026-04-17 14:20:37] Validation | Batch 700/784 | Loss: 0.3325 | LM_LOSS: 0.3217 | LB_LOSS: 1.0834 [2026-04-17 14:20:39] Validation | Batch 710/784 | Loss: 0.3323 | LM_LOSS: 0.3214 | LB_LOSS: 1.0834 [2026-04-17 14:20:40] Validation | Batch 720/784 | Loss: 0.3320 | LM_LOSS: 0.3211 | LB_LOSS: 1.0833 [2026-04-17 14:20:41] Validation | Batch 730/784 | Loss: 0.3315 | LM_LOSS: 0.3206 | LB_LOSS: 1.0833 [2026-04-17 14:20:43] Validation | Batch 740/784 | Loss: 0.3315 | LM_LOSS: 0.3207 | LB_LOSS: 1.0833 [2026-04-17 14:20:44] Validation | Batch 750/784 | Loss: 0.3308 | LM_LOSS: 0.3200 | LB_LOSS: 1.0833 [2026-04-17 14:20:45] Validation | Batch 760/784 | Loss: 0.3310 | LM_LOSS: 0.3201 | LB_LOSS: 1.0833 [2026-04-17 14:20:46] Validation | Batch 770/784 | Loss: 0.3312 | LM_LOSS: 0.3203 | LB_LOSS: 1.0834 [2026-04-17 14:20:48] Validation | Batch 780/784 | Loss: 0.3315 | LM_LOSS: 0.3207 | LB_LOSS: 1.0833 [2026-04-17 14:20:48] Validation | Batch 784/784 | Loss: 0.3317 | LM_LOSS: 0.3209 | LB_LOSS: 1.0833 [2026-04-17 14:20:51] Validation | Loss: 0.3317 | LM_LOSS: 0.3209 | LB_LOSS: 1.0833 | PPL: 1.38 | Time: 107.33s [2026-04-17 14:20:58] Epoch 3 | Step 25010 | Loss: 0.2064 | LM: 0.1952 | LB: 1.0875 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 14:21:04] Epoch 3 | Step 25020 | Loss: 0.2063 | LM: 0.1951 | LB: 1.0874 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 14:21:10] Epoch 3 | Step 25030 | Loss: 0.2063 | LM: 0.1950 | LB: 1.0874 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 14:21:16] Epoch 3 | Step 25040 | Loss: 0.2063 | LM: 0.1951 | LB: 1.0875 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 14:21:22] Epoch 3 | Step 25050 | Loss: 0.2064 | LM: 0.1952 | LB: 1.0875 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 14:21:29] Epoch 3 | Step 25060 | Loss: 0.2064 | LM: 0.1951 | LB: 1.0875 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 14:21:36] Epoch 3 | Step 25070 | Loss: 0.2064 | LM: 0.1953 | LB: 1.0875 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 14:21:42] Epoch 3 | Step 25080 | Loss: 0.2063 | LM: 0.1953 | LB: 1.0875 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 14:21:48] Epoch 3 | Step 25090 | Loss: 0.2063 | LM: 0.1952 | LB: 1.0875 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 14:21:54] Epoch 3 | Step 25100 | Loss: 0.2064 | LM: 0.1951 | LB: 1.0876 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 14:22:01] Epoch 3 | Step 25110 | Loss: 0.2064 | LM: 0.1951 | LB: 1.0876 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 14:22:07] Epoch 3 | Step 25120 | Loss: 0.2063 | LM: 0.1952 | LB: 1.0876 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 14:22:13] Epoch 3 | Step 25130 | Loss: 0.2063 | LM: 0.1951 | LB: 1.0876 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 14:22:19] Epoch 3 | Step 25140 | Loss: 0.2062 | LM: 0.1951 | LB: 1.0876 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 14:22:26] Epoch 3 | Step 25150 | Loss: 0.2064 | LM: 0.1951 | LB: 1.0876 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 14:22:32] Epoch 3 | Step 25160 | Loss: 0.2064 | LM: 0.1950 | LB: 1.0876 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 14:22:38] Epoch 3 | Step 25170 | Loss: 0.2064 | LM: 0.1951 | LB: 1.0876 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 14:22:45] Epoch 3 | Step 25180 | Loss: 0.2064 | LM: 0.1948 | LB: 1.0876 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 14:22:51] Epoch 3 | Step 25190 | Loss: 0.2065 | LM: 0.1950 | LB: 1.0876 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 14:22:58] Epoch 3 | Step 25200 | Loss: 0.2065 | LM: 0.1949 | LB: 1.0876 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 14:23:04] Epoch 3 | Step 25210 | Loss: 0.2066 | LM: 0.1950 | LB: 1.0876 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 14:23:10] Epoch 3 | Step 25220 | Loss: 0.2065 | LM: 0.1948 | LB: 1.0876 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 14:23:17] Epoch 3 | Step 25230 | Loss: 0.2065 | LM: 0.1948 | LB: 1.0876 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 14:23:23] Epoch 3 | Step 25240 | Loss: 0.2065 | LM: 0.1948 | LB: 1.0876 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 14:23:29] Epoch 3 | Step 25250 | Loss: 0.2064 | LM: 0.1947 | LB: 1.0876 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 14:23:36] Epoch 3 | Step 25260 | Loss: 0.2065 | LM: 0.1948 | LB: 1.0876 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 14:23:42] Epoch 3 | Step 25270 | Loss: 0.2066 | LM: 0.1949 | LB: 1.0876 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 14:23:49] Epoch 3 | Step 25280 | Loss: 0.2067 | LM: 0.1950 | LB: 1.0876 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 14:23:55] Epoch 3 | Step 25290 | Loss: 0.2067 | LM: 0.1949 | LB: 1.0876 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 14:24:01] Epoch 3 | Step 25300 | Loss: 0.2068 | LM: 0.1949 | LB: 1.0876 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 14:24:08] Epoch 3 | Step 25310 | Loss: 0.2068 | LM: 0.1951 | LB: 1.0876 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 14:24:14] Epoch 3 | Step 25320 | Loss: 0.2067 | LM: 0.1950 | LB: 1.0876 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 14:24:20] Epoch 3 | Step 25330 | Loss: 0.2067 | LM: 0.1950 | LB: 1.0876 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 14:24:27] Epoch 3 | Step 25340 | Loss: 0.2067 | LM: 0.1949 | LB: 1.0875 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 14:24:33] Epoch 3 | Step 25350 | Loss: 0.2066 | LM: 0.1949 | LB: 1.0875 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 14:24:39] Epoch 3 | Step 25360 | Loss: 0.2066 | LM: 0.1950 | LB: 1.0875 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 14:24:46] Epoch 3 | Step 25370 | Loss: 0.2067 | LM: 0.1950 | LB: 1.0875 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 14:24:52] Epoch 3 | Step 25380 | Loss: 0.2067 | LM: 0.1949 | LB: 1.0875 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 14:24:58] Epoch 3 | Step 25390 | Loss: 0.2068 | LM: 0.1951 | LB: 1.0875 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 14:25:05] Epoch 3 | Step 25400 | Loss: 0.2068 | LM: 0.1951 | LB: 1.0875 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 14:25:11] Epoch 3 | Step 25410 | Loss: 0.2068 | LM: 0.1952 | LB: 1.0875 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 14:25:17] Epoch 3 | Step 25420 | Loss: 0.2068 | LM: 0.1953 | LB: 1.0875 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 14:25:24] Epoch 3 | Step 25430 | Loss: 0.2068 | LM: 0.1953 | LB: 1.0875 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 14:25:30] Epoch 3 | Step 25440 | Loss: 0.2068 | LM: 0.1953 | LB: 1.0875 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 14:25:36] Epoch 3 | Step 25450 | Loss: 0.2068 | LM: 0.1952 | LB: 1.0876 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 14:25:42] Epoch 3 | Step 25460 | Loss: 0.2068 | LM: 0.1952 | LB: 1.0876 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 14:25:49] Epoch 3 | Step 25470 | Loss: 0.2068 | LM: 0.1952 | LB: 1.0876 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 14:25:56] Epoch 3 | Step 25480 | Loss: 0.2069 | LM: 0.1951 | LB: 1.0876 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 14:26:02] Epoch 3 | Step 25490 | Loss: 0.2069 | LM: 0.1953 | LB: 1.0875 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 14:26:09] Epoch 3 | Step 25500 | Loss: 0.2069 | LM: 0.1953 | LB: 1.0875 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 14:26:15] Epoch 3 | Step 25510 | Loss: 0.2069 | LM: 0.1953 | LB: 1.0876 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 14:26:21] Epoch 3 | Step 25520 | Loss: 0.2069 | LM: 0.1954 | LB: 1.0876 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 14:26:28] Epoch 3 | Step 25530 | Loss: 0.2069 | LM: 0.1953 | LB: 1.0875 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 14:26:34] Epoch 3 | Step 25540 | Loss: 0.2069 | LM: 0.1953 | LB: 1.0876 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 14:26:40] Epoch 3 | Step 25550 | Loss: 0.2069 | LM: 0.1953 | LB: 1.0875 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 14:26:47] Epoch 3 | Step 25560 | Loss: 0.2070 | LM: 0.1955 | LB: 1.0876 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 14:26:53] Epoch 3 | Step 25570 | Loss: 0.2069 | LM: 0.1953 | LB: 1.0876 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 14:27:00] Epoch 3 | Step 25580 | Loss: 0.2069 | LM: 0.1952 | LB: 1.0876 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 14:27:06] Epoch 3 | Step 25590 | Loss: 0.2069 | LM: 0.1951 | LB: 1.0876 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 14:27:13] Epoch 3 | Step 25600 | Loss: 0.2070 | LM: 0.1953 | LB: 1.0876 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 14:27:19] Epoch 3 | Step 25610 | Loss: 0.2069 | LM: 0.1952 | LB: 1.0876 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 14:27:25] Epoch 3 | Step 25620 | Loss: 0.2069 | LM: 0.1952 | LB: 1.0876 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 14:27:32] Epoch 3 | Step 25630 | Loss: 0.2069 | LM: 0.1953 | LB: 1.0876 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 14:27:38] Epoch 3 | Step 25640 | Loss: 0.2068 | LM: 0.1953 | LB: 1.0876 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 14:27:44] Epoch 3 | Step 25650 | Loss: 0.2068 | LM: 0.1953 | LB: 1.0876 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 14:27:51] Epoch 3 | Step 25660 | Loss: 0.2069 | LM: 0.1955 | LB: 1.0876 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 14:27:57] Epoch 3 | Step 25670 | Loss: 0.2068 | LM: 0.1953 | LB: 1.0876 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 14:28:03] Epoch 3 | Step 25680 | Loss: 0.2068 | LM: 0.1952 | LB: 1.0876 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 14:28:10] Epoch 3 | Step 25690 | Loss: 0.2068 | LM: 0.1953 | LB: 1.0876 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 14:28:16] Epoch 3 | Step 25700 | Loss: 0.2068 | LM: 0.1953 | LB: 1.0876 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 14:28:22] Epoch 3 | Step 25710 | Loss: 0.2068 | LM: 0.1952 | LB: 1.0876 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 14:28:29] Epoch 3 | Step 25720 | Loss: 0.2068 | LM: 0.1952 | LB: 1.0876 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 14:28:35] Epoch 3 | Step 25730 | Loss: 0.2068 | LM: 0.1953 | LB: 1.0876 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 14:28:42] Epoch 3 | Step 25740 | Loss: 0.2068 | LM: 0.1953 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 14:28:48] Epoch 3 | Step 25750 | Loss: 0.2068 | LM: 0.1953 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 14:28:54] Epoch 3 | Step 25760 | Loss: 0.2068 | LM: 0.1953 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 14:29:01] Epoch 3 | Step 25770 | Loss: 0.2068 | LM: 0.1954 | LB: 1.0876 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 14:29:07] Epoch 3 | Step 25780 | Loss: 0.2068 | LM: 0.1955 | LB: 1.0876 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 14:29:14] Epoch 3 | Step 25790 | Loss: 0.2068 | LM: 0.1954 | LB: 1.0876 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 14:29:20] Epoch 3 | Step 25800 | Loss: 0.2068 | LM: 0.1953 | LB: 1.0876 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 14:29:26] Epoch 3 | Step 25810 | Loss: 0.2067 | LM: 0.1954 | LB: 1.0876 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 14:29:33] Epoch 3 | Step 25820 | Loss: 0.2068 | LM: 0.1953 | LB: 1.0876 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 14:29:39] Epoch 3 | Step 25830 | Loss: 0.2067 | LM: 0.1953 | LB: 1.0876 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 14:29:46] Epoch 3 | Step 25840 | Loss: 0.2068 | LM: 0.1953 | LB: 1.0876 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 14:29:52] Epoch 3 | Step 25850 | Loss: 0.2068 | LM: 0.1952 | LB: 1.0876 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 14:29:58] Epoch 3 | Step 25860 | Loss: 0.2068 | LM: 0.1953 | LB: 1.0876 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 14:30:05] Epoch 3 | Step 25870 | Loss: 0.2068 | LM: 0.1954 | LB: 1.0876 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 14:30:11] Epoch 3 | Step 25880 | Loss: 0.2068 | LM: 0.1956 | LB: 1.0876 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 14:30:18] Epoch 3 | Step 25890 | Loss: 0.2069 | LM: 0.1957 | LB: 1.0876 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 14:30:24] Epoch 3 | Step 25900 | Loss: 0.2069 | LM: 0.1958 | LB: 1.0876 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 14:30:30] Epoch 3 | Step 25910 | Loss: 0.2070 | LM: 0.1957 | LB: 1.0876 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 14:30:37] Epoch 3 | Step 25920 | Loss: 0.2070 | LM: 0.1958 | LB: 1.0876 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 14:30:43] Epoch 3 | Step 25930 | Loss: 0.2070 | LM: 0.1960 | LB: 1.0876 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 14:30:50] Epoch 3 | Step 25940 | Loss: 0.2070 | LM: 0.1959 | LB: 1.0876 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 14:30:56] Epoch 3 | Step 25950 | Loss: 0.2071 | LM: 0.1960 | LB: 1.0876 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 14:31:03] Epoch 3 | Step 25960 | Loss: 0.2071 | LM: 0.1961 | LB: 1.0876 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 14:31:10] Epoch 3 | Step 25970 | Loss: 0.2071 | LM: 0.1960 | LB: 1.0876 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 14:31:16] Epoch 3 | Step 25980 | Loss: 0.2072 | LM: 0.1961 | LB: 1.0876 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 14:31:23] Epoch 3 | Step 25990 | Loss: 0.2071 | LM: 0.1960 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 14:31:29] Epoch 3 | Step 26000 | Loss: 0.2071 | LM: 0.1961 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 14:31:30] Validation | Batch 10/784 | Loss: 0.3353 | LM_LOSS: 0.3244 | LB_LOSS: 1.0847 [2026-04-17 14:31:32] Validation | Batch 20/784 | Loss: 0.3467 | LM_LOSS: 0.3359 | LB_LOSS: 1.0848 [2026-04-17 14:31:33] Validation | Batch 30/784 | Loss: 0.3325 | LM_LOSS: 0.3217 | LB_LOSS: 1.0841 [2026-04-17 14:31:35] Validation | Batch 40/784 | Loss: 0.3353 | LM_LOSS: 0.3244 | LB_LOSS: 1.0841 [2026-04-17 14:31:36] Validation | Batch 50/784 | Loss: 0.3327 | LM_LOSS: 0.3219 | LB_LOSS: 1.0834 [2026-04-17 14:31:37] Validation | Batch 60/784 | Loss: 0.3346 | LM_LOSS: 0.3238 | LB_LOSS: 1.0830 [2026-04-17 14:31:39] Validation | Batch 70/784 | Loss: 0.3319 | LM_LOSS: 0.3211 | LB_LOSS: 1.0823 [2026-04-17 14:31:40] Validation | Batch 80/784 | Loss: 0.3282 | LM_LOSS: 0.3174 | LB_LOSS: 1.0819 [2026-04-17 14:31:41] Validation | Batch 90/784 | Loss: 0.3270 | LM_LOSS: 0.3162 | LB_LOSS: 1.0824 [2026-04-17 14:31:43] Validation | Batch 100/784 | Loss: 0.3290 | LM_LOSS: 0.3182 | LB_LOSS: 1.0829 [2026-04-17 14:31:44] Validation | Batch 110/784 | Loss: 0.3235 | LM_LOSS: 0.3127 | LB_LOSS: 1.0830 [2026-04-17 14:31:45] Validation | Batch 120/784 | Loss: 0.3271 | LM_LOSS: 0.3163 | LB_LOSS: 1.0829 [2026-04-17 14:31:47] Validation | Batch 130/784 | Loss: 0.3302 | LM_LOSS: 0.3193 | LB_LOSS: 1.0829 [2026-04-17 14:31:48] Validation | Batch 140/784 | Loss: 0.3296 | LM_LOSS: 0.3187 | LB_LOSS: 1.0827 [2026-04-17 14:31:50] Validation | Batch 150/784 | Loss: 0.3256 | LM_LOSS: 0.3148 | LB_LOSS: 1.0830 [2026-04-17 14:31:51] Validation | Batch 160/784 | Loss: 0.3264 | LM_LOSS: 0.3156 | LB_LOSS: 1.0827 [2026-04-17 14:31:52] Validation | Batch 170/784 | Loss: 0.3265 | LM_LOSS: 0.3157 | LB_LOSS: 1.0824 [2026-04-17 14:31:54] Validation | Batch 180/784 | Loss: 0.3240 | LM_LOSS: 0.3132 | LB_LOSS: 1.0824 [2026-04-17 14:31:55] Validation | Batch 190/784 | Loss: 0.3263 | LM_LOSS: 0.3155 | LB_LOSS: 1.0829 [2026-04-17 14:31:56] Validation | Batch 200/784 | Loss: 0.3268 | LM_LOSS: 0.3160 | LB_LOSS: 1.0829 [2026-04-17 14:31:58] Validation | Batch 210/784 | Loss: 0.3256 | LM_LOSS: 0.3148 | LB_LOSS: 1.0828 [2026-04-17 14:31:59] Validation | Batch 220/784 | Loss: 0.3265 | LM_LOSS: 0.3157 | LB_LOSS: 1.0829 [2026-04-17 14:32:01] Validation | Batch 230/784 | Loss: 0.3271 | LM_LOSS: 0.3163 | LB_LOSS: 1.0828 [2026-04-17 14:32:02] Validation | Batch 240/784 | Loss: 0.3276 | LM_LOSS: 0.3168 | LB_LOSS: 1.0831 [2026-04-17 14:32:03] Validation | Batch 250/784 | Loss: 0.3275 | LM_LOSS: 0.3167 | LB_LOSS: 1.0830 [2026-04-17 14:32:05] Validation | Batch 260/784 | Loss: 0.3278 | LM_LOSS: 0.3169 | LB_LOSS: 1.0832 [2026-04-17 14:32:06] Validation | Batch 270/784 | Loss: 0.3276 | LM_LOSS: 0.3168 | LB_LOSS: 1.0832 [2026-04-17 14:32:08] Validation | Batch 280/784 | Loss: 0.3281 | LM_LOSS: 0.3172 | LB_LOSS: 1.0834 [2026-04-17 14:32:09] Validation | Batch 290/784 | Loss: 0.3292 | LM_LOSS: 0.3184 | LB_LOSS: 1.0835 [2026-04-17 14:32:10] Validation | Batch 300/784 | Loss: 0.3300 | LM_LOSS: 0.3192 | LB_LOSS: 1.0836 [2026-04-17 14:32:12] Validation | Batch 310/784 | Loss: 0.3294 | LM_LOSS: 0.3186 | LB_LOSS: 1.0835 [2026-04-17 14:32:13] Validation | Batch 320/784 | Loss: 0.3310 | LM_LOSS: 0.3202 | LB_LOSS: 1.0835 [2026-04-17 14:32:15] Validation | Batch 330/784 | Loss: 0.3308 | LM_LOSS: 0.3200 | LB_LOSS: 1.0835 [2026-04-17 14:32:16] Validation | Batch 340/784 | Loss: 0.3296 | LM_LOSS: 0.3188 | LB_LOSS: 1.0836 [2026-04-17 14:32:17] Validation | Batch 350/784 | Loss: 0.3298 | LM_LOSS: 0.3190 | LB_LOSS: 1.0838 [2026-04-17 14:32:18] Validation | Batch 360/784 | Loss: 0.3296 | LM_LOSS: 0.3188 | LB_LOSS: 1.0838 [2026-04-17 14:32:20] Validation | Batch 370/784 | Loss: 0.3301 | LM_LOSS: 0.3193 | LB_LOSS: 1.0837 [2026-04-17 14:32:21] Validation | Batch 380/784 | Loss: 0.3300 | LM_LOSS: 0.3191 | LB_LOSS: 1.0837 [2026-04-17 14:32:22] Validation | Batch 390/784 | Loss: 0.3299 | LM_LOSS: 0.3190 | LB_LOSS: 1.0838 [2026-04-17 14:32:23] Validation | Batch 400/784 | Loss: 0.3302 | LM_LOSS: 0.3193 | LB_LOSS: 1.0838 [2026-04-17 14:32:25] Validation | Batch 410/784 | Loss: 0.3305 | LM_LOSS: 0.3197 | LB_LOSS: 1.0838 [2026-04-17 14:32:26] Validation | Batch 420/784 | Loss: 0.3307 | LM_LOSS: 0.3199 | LB_LOSS: 1.0839 [2026-04-17 14:32:27] Validation | Batch 430/784 | Loss: 0.3309 | LM_LOSS: 0.3200 | LB_LOSS: 1.0838 [2026-04-17 14:32:28] Validation | Batch 440/784 | Loss: 0.3305 | LM_LOSS: 0.3197 | LB_LOSS: 1.0838 [2026-04-17 14:32:30] Validation | Batch 450/784 | Loss: 0.3298 | LM_LOSS: 0.3189 | LB_LOSS: 1.0838 [2026-04-17 14:32:31] Validation | Batch 460/784 | Loss: 0.3303 | LM_LOSS: 0.3194 | LB_LOSS: 1.0839 [2026-04-17 14:32:33] Validation | Batch 470/784 | Loss: 0.3294 | LM_LOSS: 0.3186 | LB_LOSS: 1.0838 [2026-04-17 14:32:34] Validation | Batch 480/784 | Loss: 0.3300 | LM_LOSS: 0.3191 | LB_LOSS: 1.0838 [2026-04-17 14:32:35] Validation | Batch 490/784 | Loss: 0.3293 | LM_LOSS: 0.3185 | LB_LOSS: 1.0837 [2026-04-17 14:32:37] Validation | Batch 500/784 | Loss: 0.3297 | LM_LOSS: 0.3188 | LB_LOSS: 1.0837 [2026-04-17 14:32:38] Validation | Batch 510/784 | Loss: 0.3294 | LM_LOSS: 0.3185 | LB_LOSS: 1.0836 [2026-04-17 14:32:39] Validation | Batch 520/784 | Loss: 0.3296 | LM_LOSS: 0.3188 | LB_LOSS: 1.0836 [2026-04-17 14:32:41] Validation | Batch 530/784 | Loss: 0.3305 | LM_LOSS: 0.3196 | LB_LOSS: 1.0835 [2026-04-17 14:32:42] Validation | Batch 540/784 | Loss: 0.3308 | LM_LOSS: 0.3200 | LB_LOSS: 1.0835 [2026-04-17 14:32:44] Validation | Batch 550/784 | Loss: 0.3321 | LM_LOSS: 0.3213 | LB_LOSS: 1.0835 [2026-04-17 14:32:45] Validation | Batch 560/784 | Loss: 0.3322 | LM_LOSS: 0.3214 | LB_LOSS: 1.0835 [2026-04-17 14:32:46] Validation | Batch 570/784 | Loss: 0.3318 | LM_LOSS: 0.3209 | LB_LOSS: 1.0835 [2026-04-17 14:32:48] Validation | Batch 580/784 | Loss: 0.3312 | LM_LOSS: 0.3204 | LB_LOSS: 1.0835 [2026-04-17 14:32:49] Validation | Batch 590/784 | Loss: 0.3315 | LM_LOSS: 0.3206 | LB_LOSS: 1.0834 [2026-04-17 14:32:50] Validation | Batch 600/784 | Loss: 0.3314 | LM_LOSS: 0.3205 | LB_LOSS: 1.0834 [2026-04-17 14:32:52] Validation | Batch 610/784 | Loss: 0.3315 | LM_LOSS: 0.3207 | LB_LOSS: 1.0834 [2026-04-17 14:32:53] Validation | Batch 620/784 | Loss: 0.3313 | LM_LOSS: 0.3205 | LB_LOSS: 1.0834 [2026-04-17 14:32:54] Validation | Batch 630/784 | Loss: 0.3322 | LM_LOSS: 0.3213 | LB_LOSS: 1.0834 [2026-04-17 14:32:56] Validation | Batch 640/784 | Loss: 0.3322 | LM_LOSS: 0.3214 | LB_LOSS: 1.0834 [2026-04-17 14:32:58] Validation | Batch 650/784 | Loss: 0.3321 | LM_LOSS: 0.3212 | LB_LOSS: 1.0835 [2026-04-17 14:32:59] Validation | Batch 660/784 | Loss: 0.3324 | LM_LOSS: 0.3216 | LB_LOSS: 1.0834 [2026-04-17 14:33:00] Validation | Batch 670/784 | Loss: 0.3329 | LM_LOSS: 0.3220 | LB_LOSS: 1.0835 [2026-04-17 14:33:02] Validation | Batch 680/784 | Loss: 0.3326 | LM_LOSS: 0.3217 | LB_LOSS: 1.0835 [2026-04-17 14:33:03] Validation | Batch 690/784 | Loss: 0.3327 | LM_LOSS: 0.3219 | LB_LOSS: 1.0834 [2026-04-17 14:33:05] Validation | Batch 700/784 | Loss: 0.3328 | LM_LOSS: 0.3219 | LB_LOSS: 1.0834 [2026-04-17 14:33:06] Validation | Batch 710/784 | Loss: 0.3325 | LM_LOSS: 0.3217 | LB_LOSS: 1.0834 [2026-04-17 14:33:07] Validation | Batch 720/784 | Loss: 0.3323 | LM_LOSS: 0.3214 | LB_LOSS: 1.0833 [2026-04-17 14:33:09] Validation | Batch 730/784 | Loss: 0.3317 | LM_LOSS: 0.3209 | LB_LOSS: 1.0832 [2026-04-17 14:33:10] Validation | Batch 740/784 | Loss: 0.3318 | LM_LOSS: 0.3210 | LB_LOSS: 1.0833 [2026-04-17 14:33:11] Validation | Batch 750/784 | Loss: 0.3311 | LM_LOSS: 0.3203 | LB_LOSS: 1.0833 [2026-04-17 14:33:12] Validation | Batch 760/784 | Loss: 0.3313 | LM_LOSS: 0.3204 | LB_LOSS: 1.0833 [2026-04-17 14:33:14] Validation | Batch 770/784 | Loss: 0.3314 | LM_LOSS: 0.3206 | LB_LOSS: 1.0834 [2026-04-17 14:33:15] Validation | Batch 780/784 | Loss: 0.3318 | LM_LOSS: 0.3210 | LB_LOSS: 1.0833 [2026-04-17 14:33:16] Validation | Batch 784/784 | Loss: 0.3320 | LM_LOSS: 0.3212 | LB_LOSS: 1.0833 [2026-04-17 14:33:18] Validation | Loss: 0.3320 | LM_LOSS: 0.3212 | LB_LOSS: 1.0833 | PPL: 1.38 | Time: 106.43s [2026-04-17 14:33:25] Epoch 3 | Step 26010 | Loss: 0.2071 | LM: 0.1961 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 14:33:31] Epoch 3 | Step 26020 | Loss: 0.2071 | LM: 0.1961 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 14:33:38] Epoch 3 | Step 26030 | Loss: 0.2071 | LM: 0.1961 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 14:33:44] Epoch 3 | Step 26040 | Loss: 0.2071 | LM: 0.1961 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 14:33:50] Epoch 3 | Step 26050 | Loss: 0.2071 | LM: 0.1961 | LB: 1.0878 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 14:33:57] Epoch 3 | Step 26060 | Loss: 0.2071 | LM: 0.1962 | LB: 1.0878 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 14:34:03] Epoch 3 | Step 26070 | Loss: 0.2071 | LM: 0.1961 | LB: 1.0878 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 14:34:10] Epoch 3 | Step 26080 | Loss: 0.2071 | LM: 0.1960 | LB: 1.0878 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 14:34:16] Epoch 3 | Step 26090 | Loss: 0.2071 | LM: 0.1959 | LB: 1.0878 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 14:34:22] Epoch 3 | Step 26100 | Loss: 0.2070 | LM: 0.1958 | LB: 1.0878 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 14:34:29] Epoch 3 | Step 26110 | Loss: 0.2070 | LM: 0.1957 | LB: 1.0878 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 14:34:35] Epoch 3 | Step 26120 | Loss: 0.2070 | LM: 0.1957 | LB: 1.0878 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 14:34:42] Epoch 3 | Step 26130 | Loss: 0.2069 | LM: 0.1957 | LB: 1.0878 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 14:34:48] Epoch 3 | Step 26140 | Loss: 0.2069 | LM: 0.1956 | LB: 1.0878 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 14:34:55] Epoch 3 | Step 26150 | Loss: 0.2068 | LM: 0.1954 | LB: 1.0878 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 14:35:00] Epoch 3 | Step 26160 | Loss: 0.2068 | LM: 0.1953 | LB: 1.0878 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 14:35:07] Epoch 3 | Step 26170 | Loss: 0.2068 | LM: 0.1953 | LB: 1.0878 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 14:35:13] Epoch 3 | Step 26180 | Loss: 0.2068 | LM: 0.1954 | LB: 1.0878 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 14:35:19] Epoch 3 | Step 26190 | Loss: 0.2069 | LM: 0.1954 | LB: 1.0878 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 14:35:26] Epoch 3 | Step 26200 | Loss: 0.2069 | LM: 0.1954 | LB: 1.0878 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 14:35:32] Epoch 3 | Step 26210 | Loss: 0.2069 | LM: 0.1955 | LB: 1.0878 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 14:35:39] Epoch 3 | Step 26220 | Loss: 0.2069 | LM: 0.1953 | LB: 1.0878 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 14:35:45] Epoch 3 | Step 26230 | Loss: 0.2068 | LM: 0.1952 | LB: 1.0878 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 14:35:52] Epoch 3 | Step 26240 | Loss: 0.2069 | LM: 0.1954 | LB: 1.0878 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 14:35:58] Epoch 3 | Step 26250 | Loss: 0.2069 | LM: 0.1955 | LB: 1.0878 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 14:36:05] Epoch 3 | Step 26260 | Loss: 0.2070 | LM: 0.1956 | LB: 1.0878 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 14:36:11] Epoch 3 | Step 26270 | Loss: 0.2069 | LM: 0.1955 | LB: 1.0878 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 14:36:18] Epoch 3 | Step 26280 | Loss: 0.2069 | LM: 0.1955 | LB: 1.0878 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 14:36:24] Epoch 3 | Step 26290 | Loss: 0.2069 | LM: 0.1954 | LB: 1.0878 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 14:36:31] Epoch 3 | Step 26300 | Loss: 0.2069 | LM: 0.1954 | LB: 1.0878 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 14:36:37] Epoch 3 | Step 26310 | Loss: 0.2068 | LM: 0.1954 | LB: 1.0878 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 14:36:43] Epoch 3 | Step 26320 | Loss: 0.2068 | LM: 0.1953 | LB: 1.0878 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 14:36:50] Epoch 3 | Step 26330 | Loss: 0.2068 | LM: 0.1953 | LB: 1.0878 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 14:36:56] Epoch 3 | Step 26340 | Loss: 0.2068 | LM: 0.1954 | LB: 1.0878 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 14:37:02] Epoch 3 | Step 26350 | Loss: 0.2068 | LM: 0.1954 | LB: 1.0878 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 14:37:09] Epoch 3 | Step 26360 | Loss: 0.2067 | LM: 0.1954 | LB: 1.0878 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 14:37:16] Epoch 3 | Step 26370 | Loss: 0.2067 | LM: 0.1954 | LB: 1.0878 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 14:37:22] Epoch 3 | Step 26380 | Loss: 0.2067 | LM: 0.1954 | LB: 1.0878 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 14:37:28] Epoch 3 | Step 26390 | Loss: 0.2068 | LM: 0.1954 | LB: 1.0878 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 14:37:35] Epoch 3 | Step 26400 | Loss: 0.2067 | LM: 0.1954 | LB: 1.0878 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 14:37:41] Epoch 3 | Step 26410 | Loss: 0.2067 | LM: 0.1954 | LB: 1.0878 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 14:37:47] Epoch 3 | Step 26420 | Loss: 0.2067 | LM: 0.1955 | LB: 1.0878 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 14:37:54] Epoch 3 | Step 26430 | Loss: 0.2067 | LM: 0.1956 | LB: 1.0879 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 14:38:01] Epoch 3 | Step 26440 | Loss: 0.2068 | LM: 0.1957 | LB: 1.0878 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 14:38:07] Epoch 3 | Step 26450 | Loss: 0.2068 | LM: 0.1956 | LB: 1.0879 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 14:38:13] Epoch 3 | Step 26460 | Loss: 0.2068 | LM: 0.1957 | LB: 1.0879 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 14:38:20] Epoch 3 | Step 26470 | Loss: 0.2068 | LM: 0.1957 | LB: 1.0879 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 14:38:26] Epoch 3 | Step 26480 | Loss: 0.2068 | LM: 0.1956 | LB: 1.0879 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 14:38:32] Epoch 3 | Step 26490 | Loss: 0.2068 | LM: 0.1956 | LB: 1.0879 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 14:38:39] Epoch 3 | Step 26500 | Loss: 0.2068 | LM: 0.1955 | LB: 1.0879 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 14:38:45] Epoch 3 | Step 26510 | Loss: 0.2068 | LM: 0.1955 | LB: 1.0879 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 14:38:52] Epoch 3 | Step 26520 | Loss: 0.2068 | LM: 0.1956 | LB: 1.0879 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 14:38:58] Epoch 3 | Step 26530 | Loss: 0.2068 | LM: 0.1955 | LB: 1.0879 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 14:39:04] Epoch 3 | Step 26540 | Loss: 0.2068 | LM: 0.1955 | LB: 1.0879 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 14:39:10] Epoch 3 | Step 26550 | Loss: 0.2068 | LM: 0.1955 | LB: 1.0879 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 14:39:17] Epoch 3 | Step 26560 | Loss: 0.2069 | LM: 0.1956 | LB: 1.0879 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 14:39:23] Epoch 3 | Step 26570 | Loss: 0.2069 | LM: 0.1957 | LB: 1.0879 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 14:39:29] Epoch 3 | Step 26580 | Loss: 0.2069 | LM: 0.1957 | LB: 1.0879 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 14:39:35] Epoch 3 | Step 26590 | Loss: 0.2069 | LM: 0.1956 | LB: 1.0879 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 14:39:41] Epoch 3 | Step 26600 | Loss: 0.2069 | LM: 0.1957 | LB: 1.0879 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 14:39:48] Epoch 3 | Step 26610 | Loss: 0.2070 | LM: 0.1958 | LB: 1.0879 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 14:39:54] Epoch 3 | Step 26620 | Loss: 0.2070 | LM: 0.1958 | LB: 1.0879 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 14:40:00] Epoch 3 | Step 26630 | Loss: 0.2070 | LM: 0.1957 | LB: 1.0879 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 14:40:07] Epoch 3 | Step 26640 | Loss: 0.2070 | LM: 0.1956 | LB: 1.0880 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 14:40:13] Epoch 3 | Step 26650 | Loss: 0.2070 | LM: 0.1956 | LB: 1.0880 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 14:40:19] Epoch 3 | Step 26660 | Loss: 0.2070 | LM: 0.1957 | LB: 1.0879 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 14:40:25] Epoch 3 | Step 26670 | Loss: 0.2070 | LM: 0.1958 | LB: 1.0880 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 14:40:32] Epoch 3 | Step 26680 | Loss: 0.2070 | LM: 0.1957 | LB: 1.0879 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 14:40:38] Epoch 3 | Step 26690 | Loss: 0.2070 | LM: 0.1955 | LB: 1.0879 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 14:40:45] Epoch 3 | Step 26700 | Loss: 0.2071 | LM: 0.1957 | LB: 1.0879 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 14:40:51] Epoch 3 | Step 26710 | Loss: 0.2071 | LM: 0.1958 | LB: 1.0879 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 14:40:58] Epoch 3 | Step 26720 | Loss: 0.2071 | LM: 0.1958 | LB: 1.0879 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 14:41:04] Epoch 3 | Step 26730 | Loss: 0.2071 | LM: 0.1959 | LB: 1.0879 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 14:41:11] Epoch 3 | Step 26740 | Loss: 0.2072 | LM: 0.1961 | LB: 1.0879 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 14:41:17] Epoch 3 | Step 26750 | Loss: 0.2072 | LM: 0.1961 | LB: 1.0879 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 14:41:23] Epoch 3 | Step 26760 | Loss: 0.2072 | LM: 0.1961 | LB: 1.0879 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 14:41:30] Epoch 3 | Step 26770 | Loss: 0.2071 | LM: 0.1960 | LB: 1.0879 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 14:41:36] Epoch 3 | Step 26780 | Loss: 0.2072 | LM: 0.1961 | LB: 1.0879 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 14:41:43] Epoch 3 | Step 26790 | Loss: 0.2071 | LM: 0.1961 | LB: 1.0879 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 14:41:49] Epoch 3 | Step 26800 | Loss: 0.2071 | LM: 0.1961 | LB: 1.0879 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 14:41:55] Epoch 3 | Step 26810 | Loss: 0.2071 | LM: 0.1960 | LB: 1.0879 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 14:42:02] Epoch 3 | Step 26820 | Loss: 0.2072 | LM: 0.1960 | LB: 1.0879 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 14:42:08] Epoch 3 | Step 26830 | Loss: 0.2072 | LM: 0.1959 | LB: 1.0879 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 14:42:15] Epoch 3 | Step 26840 | Loss: 0.2072 | LM: 0.1958 | LB: 1.0879 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 14:42:21] Epoch 3 | Step 26850 | Loss: 0.2072 | LM: 0.1958 | LB: 1.0879 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 14:42:28] Epoch 3 | Step 26860 | Loss: 0.2071 | LM: 0.1957 | LB: 1.0879 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 14:42:34] Epoch 3 | Step 26870 | Loss: 0.2072 | LM: 0.1956 | LB: 1.0879 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 14:42:40] Epoch 3 | Step 26880 | Loss: 0.2072 | LM: 0.1956 | LB: 1.0879 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 14:42:47] Epoch 3 | Step 26890 | Loss: 0.2072 | LM: 0.1956 | LB: 1.0879 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 14:42:53] Epoch 3 | Step 26900 | Loss: 0.2072 | LM: 0.1956 | LB: 1.0879 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 14:43:00] Epoch 3 | Step 26910 | Loss: 0.2071 | LM: 0.1955 | LB: 1.0879 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 14:43:06] Epoch 3 | Step 26920 | Loss: 0.2072 | LM: 0.1955 | LB: 1.0878 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 14:43:13] Epoch 3 | Step 26930 | Loss: 0.2072 | LM: 0.1956 | LB: 1.0878 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 14:43:19] Epoch 3 | Step 26940 | Loss: 0.2072 | LM: 0.1957 | LB: 1.0878 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 14:43:25] Epoch 3 | Step 26950 | Loss: 0.2072 | LM: 0.1957 | LB: 1.0878 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 14:43:31] Epoch 3 | Step 26960 | Loss: 0.2073 | LM: 0.1957 | LB: 1.0878 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 14:43:38] Epoch 3 | Step 26970 | Loss: 0.2073 | LM: 0.1957 | LB: 1.0878 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 14:43:45] Epoch 3 | Step 26980 | Loss: 0.2073 | LM: 0.1957 | LB: 1.0878 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 14:43:51] Epoch 3 | Step 26990 | Loss: 0.2073 | LM: 0.1958 | LB: 1.0879 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 14:43:57] Epoch 3 | Step 27000 | Loss: 0.2072 | LM: 0.1957 | LB: 1.0879 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 14:44:06] Checkpoint saved: outputs/2026-04-17/08-57-56/checkpoints/checkpoint_step_27000.pt [2026-04-17 14:44:19] Validation | Batch 10/784 | Loss: 0.3352 | LM_LOSS: 0.3244 | LB_LOSS: 1.0847 [2026-04-17 14:44:21] Validation | Batch 20/784 | Loss: 0.3470 | LM_LOSS: 0.3362 | LB_LOSS: 1.0849 [2026-04-17 14:44:22] Validation | Batch 30/784 | Loss: 0.3328 | LM_LOSS: 0.3220 | LB_LOSS: 1.0842 [2026-04-17 14:44:23] Validation | Batch 40/784 | Loss: 0.3355 | LM_LOSS: 0.3246 | LB_LOSS: 1.0841 [2026-04-17 14:44:25] Validation | Batch 50/784 | Loss: 0.3330 | LM_LOSS: 0.3221 | LB_LOSS: 1.0834 [2026-04-17 14:44:26] Validation | Batch 60/784 | Loss: 0.3349 | LM_LOSS: 0.3241 | LB_LOSS: 1.0830 [2026-04-17 14:44:27] Validation | Batch 70/784 | Loss: 0.3322 | LM_LOSS: 0.3214 | LB_LOSS: 1.0823 [2026-04-17 14:44:29] Validation | Batch 80/784 | Loss: 0.3284 | LM_LOSS: 0.3176 | LB_LOSS: 1.0819 [2026-04-17 14:44:30] Validation | Batch 90/784 | Loss: 0.3273 | LM_LOSS: 0.3165 | LB_LOSS: 1.0824 [2026-04-17 14:44:31] Validation | Batch 100/784 | Loss: 0.3293 | LM_LOSS: 0.3184 | LB_LOSS: 1.0829 [2026-04-17 14:44:33] Validation | Batch 110/784 | Loss: 0.3238 | LM_LOSS: 0.3129 | LB_LOSS: 1.0830 [2026-04-17 14:44:34] Validation | Batch 120/784 | Loss: 0.3274 | LM_LOSS: 0.3165 | LB_LOSS: 1.0829 [2026-04-17 14:44:35] Validation | Batch 130/784 | Loss: 0.3305 | LM_LOSS: 0.3196 | LB_LOSS: 1.0829 [2026-04-17 14:44:37] Validation | Batch 140/784 | Loss: 0.3298 | LM_LOSS: 0.3190 | LB_LOSS: 1.0827 [2026-04-17 14:44:38] Validation | Batch 150/784 | Loss: 0.3259 | LM_LOSS: 0.3150 | LB_LOSS: 1.0830 [2026-04-17 14:44:40] Validation | Batch 160/784 | Loss: 0.3267 | LM_LOSS: 0.3159 | LB_LOSS: 1.0827 [2026-04-17 14:44:41] Validation | Batch 170/784 | Loss: 0.3268 | LM_LOSS: 0.3160 | LB_LOSS: 1.0824 [2026-04-17 14:44:43] Validation | Batch 180/784 | Loss: 0.3243 | LM_LOSS: 0.3135 | LB_LOSS: 1.0824 [2026-04-17 14:44:44] Validation | Batch 190/784 | Loss: 0.3266 | LM_LOSS: 0.3158 | LB_LOSS: 1.0829 [2026-04-17 14:44:45] Validation | Batch 200/784 | Loss: 0.3271 | LM_LOSS: 0.3163 | LB_LOSS: 1.0829 [2026-04-17 14:44:47] Validation | Batch 210/784 | Loss: 0.3259 | LM_LOSS: 0.3151 | LB_LOSS: 1.0828 [2026-04-17 14:44:48] Validation | Batch 220/784 | Loss: 0.3268 | LM_LOSS: 0.3160 | LB_LOSS: 1.0829 [2026-04-17 14:44:49] Validation | Batch 230/784 | Loss: 0.3274 | LM_LOSS: 0.3166 | LB_LOSS: 1.0828 [2026-04-17 14:44:51] Validation | Batch 240/784 | Loss: 0.3279 | LM_LOSS: 0.3171 | LB_LOSS: 1.0831 [2026-04-17 14:44:52] Validation | Batch 250/784 | Loss: 0.3278 | LM_LOSS: 0.3169 | LB_LOSS: 1.0830 [2026-04-17 14:44:54] Validation | Batch 260/784 | Loss: 0.3281 | LM_LOSS: 0.3172 | LB_LOSS: 1.0832 [2026-04-17 14:44:55] Validation | Batch 270/784 | Loss: 0.3279 | LM_LOSS: 0.3171 | LB_LOSS: 1.0832 [2026-04-17 14:44:57] Validation | Batch 280/784 | Loss: 0.3284 | LM_LOSS: 0.3175 | LB_LOSS: 1.0834 [2026-04-17 14:44:58] Validation | Batch 290/784 | Loss: 0.3295 | LM_LOSS: 0.3187 | LB_LOSS: 1.0835 [2026-04-17 14:44:59] Validation | Batch 300/784 | Loss: 0.3303 | LM_LOSS: 0.3195 | LB_LOSS: 1.0836 [2026-04-17 14:45:00] Validation | Batch 310/784 | Loss: 0.3297 | LM_LOSS: 0.3189 | LB_LOSS: 1.0835 [2026-04-17 14:45:02] Validation | Batch 320/784 | Loss: 0.3313 | LM_LOSS: 0.3205 | LB_LOSS: 1.0835 [2026-04-17 14:45:03] Validation | Batch 330/784 | Loss: 0.3311 | LM_LOSS: 0.3203 | LB_LOSS: 1.0835 [2026-04-17 14:45:05] Validation | Batch 340/784 | Loss: 0.3299 | LM_LOSS: 0.3191 | LB_LOSS: 1.0836 [2026-04-17 14:45:06] Validation | Batch 350/784 | Loss: 0.3301 | LM_LOSS: 0.3193 | LB_LOSS: 1.0838 [2026-04-17 14:45:07] Validation | Batch 360/784 | Loss: 0.3299 | LM_LOSS: 0.3191 | LB_LOSS: 1.0838 [2026-04-17 14:45:09] Validation | Batch 370/784 | Loss: 0.3305 | LM_LOSS: 0.3196 | LB_LOSS: 1.0837 [2026-04-17 14:45:10] Validation | Batch 380/784 | Loss: 0.3303 | LM_LOSS: 0.3194 | LB_LOSS: 1.0837 [2026-04-17 14:45:11] Validation | Batch 390/784 | Loss: 0.3302 | LM_LOSS: 0.3194 | LB_LOSS: 1.0838 [2026-04-17 14:45:12] Validation | Batch 400/784 | Loss: 0.3305 | LM_LOSS: 0.3197 | LB_LOSS: 1.0838 [2026-04-17 14:45:14] Validation | Batch 410/784 | Loss: 0.3308 | LM_LOSS: 0.3200 | LB_LOSS: 1.0838 [2026-04-17 14:45:15] Validation | Batch 420/784 | Loss: 0.3311 | LM_LOSS: 0.3202 | LB_LOSS: 1.0839 [2026-04-17 14:45:16] Validation | Batch 430/784 | Loss: 0.3312 | LM_LOSS: 0.3204 | LB_LOSS: 1.0838 [2026-04-17 14:45:17] Validation | Batch 440/784 | Loss: 0.3309 | LM_LOSS: 0.3200 | LB_LOSS: 1.0838 [2026-04-17 14:45:19] Validation | Batch 450/784 | Loss: 0.3301 | LM_LOSS: 0.3193 | LB_LOSS: 1.0838 [2026-04-17 14:45:20] Validation | Batch 460/784 | Loss: 0.3306 | LM_LOSS: 0.3198 | LB_LOSS: 1.0839 [2026-04-17 14:45:22] Validation | Batch 470/784 | Loss: 0.3298 | LM_LOSS: 0.3189 | LB_LOSS: 1.0838 [2026-04-17 14:45:23] Validation | Batch 480/784 | Loss: 0.3303 | LM_LOSS: 0.3195 | LB_LOSS: 1.0838 [2026-04-17 14:45:24] Validation | Batch 490/784 | Loss: 0.3296 | LM_LOSS: 0.3188 | LB_LOSS: 1.0837 [2026-04-17 14:45:25] Validation | Batch 500/784 | Loss: 0.3300 | LM_LOSS: 0.3191 | LB_LOSS: 1.0837 [2026-04-17 14:45:27] Validation | Batch 510/784 | Loss: 0.3297 | LM_LOSS: 0.3189 | LB_LOSS: 1.0836 [2026-04-17 14:45:28] Validation | Batch 520/784 | Loss: 0.3299 | LM_LOSS: 0.3191 | LB_LOSS: 1.0836 [2026-04-17 14:45:30] Validation | Batch 530/784 | Loss: 0.3308 | LM_LOSS: 0.3200 | LB_LOSS: 1.0835 [2026-04-17 14:45:31] Validation | Batch 540/784 | Loss: 0.3311 | LM_LOSS: 0.3203 | LB_LOSS: 1.0835 [2026-04-17 14:45:32] Validation | Batch 550/784 | Loss: 0.3324 | LM_LOSS: 0.3216 | LB_LOSS: 1.0835 [2026-04-17 14:45:34] Validation | Batch 560/784 | Loss: 0.3326 | LM_LOSS: 0.3217 | LB_LOSS: 1.0835 [2026-04-17 14:45:35] Validation | Batch 570/784 | Loss: 0.3321 | LM_LOSS: 0.3213 | LB_LOSS: 1.0835 [2026-04-17 14:45:37] Validation | Batch 580/784 | Loss: 0.3316 | LM_LOSS: 0.3207 | LB_LOSS: 1.0835 [2026-04-17 14:45:38] Validation | Batch 590/784 | Loss: 0.3318 | LM_LOSS: 0.3210 | LB_LOSS: 1.0834 [2026-04-17 14:45:39] Validation | Batch 600/784 | Loss: 0.3317 | LM_LOSS: 0.3208 | LB_LOSS: 1.0834 [2026-04-17 14:45:41] Validation | Batch 610/784 | Loss: 0.3318 | LM_LOSS: 0.3210 | LB_LOSS: 1.0834 [2026-04-17 14:45:42] Validation | Batch 620/784 | Loss: 0.3317 | LM_LOSS: 0.3208 | LB_LOSS: 1.0834 [2026-04-17 14:45:44] Validation | Batch 630/784 | Loss: 0.3325 | LM_LOSS: 0.3217 | LB_LOSS: 1.0834 [2026-04-17 14:45:45] Validation | Batch 640/784 | Loss: 0.3325 | LM_LOSS: 0.3217 | LB_LOSS: 1.0834 [2026-04-17 14:45:47] Validation | Batch 650/784 | Loss: 0.3324 | LM_LOSS: 0.3216 | LB_LOSS: 1.0835 [2026-04-17 14:45:48] Validation | Batch 660/784 | Loss: 0.3328 | LM_LOSS: 0.3220 | LB_LOSS: 1.0834 [2026-04-17 14:45:50] Validation | Batch 670/784 | Loss: 0.3332 | LM_LOSS: 0.3224 | LB_LOSS: 1.0835 [2026-04-17 14:45:51] Validation | Batch 680/784 | Loss: 0.3329 | LM_LOSS: 0.3221 | LB_LOSS: 1.0835 [2026-04-17 14:45:52] Validation | Batch 690/784 | Loss: 0.3331 | LM_LOSS: 0.3222 | LB_LOSS: 1.0834 [2026-04-17 14:45:54] Validation | Batch 700/784 | Loss: 0.3331 | LM_LOSS: 0.3223 | LB_LOSS: 1.0834 [2026-04-17 14:45:55] Validation | Batch 710/784 | Loss: 0.3329 | LM_LOSS: 0.3220 | LB_LOSS: 1.0833 [2026-04-17 14:45:57] Validation | Batch 720/784 | Loss: 0.3326 | LM_LOSS: 0.3218 | LB_LOSS: 1.0833 [2026-04-17 14:45:58] Validation | Batch 730/784 | Loss: 0.3321 | LM_LOSS: 0.3212 | LB_LOSS: 1.0832 [2026-04-17 14:45:59] Validation | Batch 740/784 | Loss: 0.3321 | LM_LOSS: 0.3213 | LB_LOSS: 1.0833 [2026-04-17 14:46:00] Validation | Batch 750/784 | Loss: 0.3315 | LM_LOSS: 0.3206 | LB_LOSS: 1.0833 [2026-04-17 14:46:02] Validation | Batch 760/784 | Loss: 0.3316 | LM_LOSS: 0.3208 | LB_LOSS: 1.0833 [2026-04-17 14:46:03] Validation | Batch 770/784 | Loss: 0.3318 | LM_LOSS: 0.3210 | LB_LOSS: 1.0833 [2026-04-17 14:46:04] Validation | Batch 780/784 | Loss: 0.3321 | LM_LOSS: 0.3213 | LB_LOSS: 1.0833 [2026-04-17 14:46:05] Validation | Batch 784/784 | Loss: 0.3323 | LM_LOSS: 0.3215 | LB_LOSS: 1.0833 [2026-04-17 14:46:08] Validation | Loss: 0.3323 | LM_LOSS: 0.3215 | LB_LOSS: 1.0833 | PPL: 1.38 | Time: 106.95s [2026-04-17 14:46:13] Epoch 3 | Step 27010 | Loss: 0.2072 | LM: 0.1957 | LB: 1.0879 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 14:46:20] Epoch 3 | Step 27020 | Loss: 0.2072 | LM: 0.1958 | LB: 1.0878 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 14:46:26] Epoch 3 | Step 27030 | Loss: 0.2071 | LM: 0.1957 | LB: 1.0878 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 14:46:33] Epoch 3 | Step 27040 | Loss: 0.2071 | LM: 0.1956 | LB: 1.0879 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 14:46:39] Epoch 3 | Step 27050 | Loss: 0.2072 | LM: 0.1957 | LB: 1.0878 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 14:46:45] Epoch 3 | Step 27060 | Loss: 0.2072 | LM: 0.1956 | LB: 1.0878 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 14:46:52] Epoch 3 | Step 27070 | Loss: 0.2072 | LM: 0.1956 | LB: 1.0878 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 14:46:58] Epoch 3 | Step 27080 | Loss: 0.2072 | LM: 0.1958 | LB: 1.0878 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 14:47:04] Epoch 3 | Step 27090 | Loss: 0.2072 | LM: 0.1959 | LB: 1.0878 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 14:47:11] Epoch 3 | Step 27100 | Loss: 0.2072 | LM: 0.1959 | LB: 1.0878 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 14:47:17] Epoch 3 | Step 27110 | Loss: 0.2072 | LM: 0.1960 | LB: 1.0878 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 14:47:24] Epoch 3 | Step 27120 | Loss: 0.2072 | LM: 0.1960 | LB: 1.0878 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 14:47:30] Epoch 3 | Step 27130 | Loss: 0.2072 | LM: 0.1960 | LB: 1.0878 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 14:47:36] Epoch 3 | Step 27140 | Loss: 0.2072 | LM: 0.1959 | LB: 1.0878 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 14:47:43] Epoch 3 | Step 27150 | Loss: 0.2071 | LM: 0.1959 | LB: 1.0878 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 14:47:49] Epoch 3 | Step 27160 | Loss: 0.2071 | LM: 0.1959 | LB: 1.0878 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 14:47:56] Epoch 3 | Step 27170 | Loss: 0.2071 | LM: 0.1958 | LB: 1.0878 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 14:48:02] Epoch 3 | Step 27180 | Loss: 0.2070 | LM: 0.1958 | LB: 1.0878 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 14:48:08] Epoch 3 | Step 27190 | Loss: 0.2070 | LM: 0.1958 | LB: 1.0878 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 14:48:15] Epoch 3 | Step 27200 | Loss: 0.2070 | LM: 0.1958 | LB: 1.0878 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 14:48:21] Epoch 3 | Step 27210 | Loss: 0.2069 | LM: 0.1957 | LB: 1.0878 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 14:48:28] Epoch 3 | Step 27220 | Loss: 0.2069 | LM: 0.1958 | LB: 1.0878 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 14:48:34] Epoch 3 | Step 27230 | Loss: 0.2070 | LM: 0.1959 | LB: 1.0878 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 14:48:40] Epoch 3 | Step 27240 | Loss: 0.2069 | LM: 0.1959 | LB: 1.0878 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 14:48:47] Epoch 3 | Step 27250 | Loss: 0.2070 | LM: 0.1958 | LB: 1.0878 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 14:48:53] Epoch 3 | Step 27260 | Loss: 0.2069 | LM: 0.1957 | LB: 1.0878 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 14:49:00] Epoch 3 | Step 27270 | Loss: 0.2070 | LM: 0.1959 | LB: 1.0878 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 14:49:06] Epoch 3 | Step 27280 | Loss: 0.2070 | LM: 0.1959 | LB: 1.0878 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 14:49:13] Epoch 3 | Step 27290 | Loss: 0.2070 | LM: 0.1959 | LB: 1.0878 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 14:49:19] Epoch 3 | Step 27300 | Loss: 0.2069 | LM: 0.1958 | LB: 1.0878 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 14:49:26] Epoch 3 | Step 27310 | Loss: 0.2069 | LM: 0.1959 | LB: 1.0878 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 14:49:32] Epoch 3 | Step 27320 | Loss: 0.2069 | LM: 0.1959 | LB: 1.0878 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 14:49:38] Epoch 3 | Step 27330 | Loss: 0.2070 | LM: 0.1959 | LB: 1.0878 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 14:49:45] Epoch 3 | Step 27340 | Loss: 0.2070 | LM: 0.1960 | LB: 1.0878 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 14:49:51] Epoch 3 | Step 27350 | Loss: 0.2070 | LM: 0.1960 | LB: 1.0878 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 14:49:57] Epoch 3 | Step 27360 | Loss: 0.2070 | LM: 0.1961 | LB: 1.0878 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 14:50:04] Epoch 3 | Step 27370 | Loss: 0.2071 | LM: 0.1963 | LB: 1.0878 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 14:50:10] Epoch 3 | Step 27380 | Loss: 0.2071 | LM: 0.1963 | LB: 1.0878 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 14:50:17] Epoch 3 | Step 27390 | Loss: 0.2071 | LM: 0.1964 | LB: 1.0879 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 14:50:23] Epoch 3 | Step 27400 | Loss: 0.2071 | LM: 0.1964 | LB: 1.0878 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 14:50:30] Epoch 3 | Step 27410 | Loss: 0.2071 | LM: 0.1965 | LB: 1.0879 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 14:50:36] Epoch 3 | Step 27420 | Loss: 0.2071 | LM: 0.1965 | LB: 1.0878 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 14:50:42] Epoch 3 | Step 27430 | Loss: 0.2072 | LM: 0.1965 | LB: 1.0878 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 14:50:49] Epoch 3 | Step 27440 | Loss: 0.2071 | LM: 0.1964 | LB: 1.0878 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 14:50:55] Epoch 3 | Step 27450 | Loss: 0.2071 | LM: 0.1963 | LB: 1.0878 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 14:51:02] Epoch 3 | Step 27460 | Loss: 0.2071 | LM: 0.1963 | LB: 1.0879 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 14:51:08] Epoch 3 | Step 27470 | Loss: 0.2071 | LM: 0.1962 | LB: 1.0879 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 14:51:14] Epoch 3 | Step 27480 | Loss: 0.2071 | LM: 0.1962 | LB: 1.0878 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 14:51:21] Epoch 3 | Step 27490 | Loss: 0.2071 | LM: 0.1962 | LB: 1.0879 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 14:51:27] Epoch 3 | Step 27500 | Loss: 0.2071 | LM: 0.1962 | LB: 1.0879 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 14:51:33] Epoch 3 | Step 27510 | Loss: 0.2071 | LM: 0.1962 | LB: 1.0878 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 14:51:40] Epoch 3 | Step 27520 | Loss: 0.2071 | LM: 0.1962 | LB: 1.0878 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 14:51:46] Epoch 3 | Step 27530 | Loss: 0.2071 | LM: 0.1961 | LB: 1.0878 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 14:51:53] Epoch 3 | Step 27540 | Loss: 0.2072 | LM: 0.1962 | LB: 1.0878 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 14:51:59] Epoch 3 | Step 27550 | Loss: 0.2071 | LM: 0.1961 | LB: 1.0878 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 14:52:05] Epoch 3 | Step 27560 | Loss: 0.2071 | LM: 0.1960 | LB: 1.0879 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 14:52:12] Epoch 3 | Step 27570 | Loss: 0.2071 | LM: 0.1959 | LB: 1.0878 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 14:52:18] Epoch 3 | Step 27580 | Loss: 0.2071 | LM: 0.1959 | LB: 1.0878 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 14:52:25] Epoch 3 | Step 27590 | Loss: 0.2071 | LM: 0.1959 | LB: 1.0878 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 14:52:31] Epoch 3 | Step 27600 | Loss: 0.2071 | LM: 0.1958 | LB: 1.0878 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 14:52:38] Epoch 3 | Step 27610 | Loss: 0.2071 | LM: 0.1959 | LB: 1.0878 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 14:52:44] Epoch 3 | Step 27620 | Loss: 0.2071 | LM: 0.1959 | LB: 1.0878 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 14:52:50] Epoch 3 | Step 27630 | Loss: 0.2071 | LM: 0.1958 | LB: 1.0878 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 14:52:57] Epoch 3 | Step 27640 | Loss: 0.2071 | LM: 0.1959 | LB: 1.0878 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 14:53:03] Epoch 3 | Step 27650 | Loss: 0.2071 | LM: 0.1959 | LB: 1.0878 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 14:53:10] Epoch 3 | Step 27660 | Loss: 0.2071 | LM: 0.1958 | LB: 1.0878 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 14:53:17] Epoch 3 | Step 27670 | Loss: 0.2070 | LM: 0.1958 | LB: 1.0878 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 14:53:23] Epoch 3 | Step 27680 | Loss: 0.2070 | LM: 0.1958 | LB: 1.0878 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 14:53:29] Epoch 3 | Step 27690 | Loss: 0.2070 | LM: 0.1959 | LB: 1.0878 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 14:53:36] Epoch 3 | Step 27700 | Loss: 0.2070 | LM: 0.1958 | LB: 1.0878 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 14:53:42] Epoch 3 | Step 27710 | Loss: 0.2069 | LM: 0.1958 | LB: 1.0878 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 14:53:49] Epoch 3 | Step 27720 | Loss: 0.2069 | LM: 0.1958 | LB: 1.0878 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 14:53:55] Epoch 3 | Step 27730 | Loss: 0.2069 | LM: 0.1957 | LB: 1.0878 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 14:54:01] Epoch 3 | Step 27740 | Loss: 0.2068 | LM: 0.1958 | LB: 1.0879 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 14:54:08] Epoch 3 | Step 27750 | Loss: 0.2068 | LM: 0.1959 | LB: 1.0878 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 14:54:14] Epoch 3 | Step 27760 | Loss: 0.2068 | LM: 0.1958 | LB: 1.0879 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 14:54:20] Epoch 3 | Step 27770 | Loss: 0.2068 | LM: 0.1958 | LB: 1.0878 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 14:54:27] Epoch 3 | Step 27780 | Loss: 0.2068 | LM: 0.1958 | LB: 1.0878 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 14:54:33] Epoch 3 | Step 27790 | Loss: 0.2068 | LM: 0.1958 | LB: 1.0879 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 14:54:39] Epoch 3 | Step 27800 | Loss: 0.2068 | LM: 0.1958 | LB: 1.0878 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 14:54:46] Epoch 3 | Step 27810 | Loss: 0.2068 | LM: 0.1958 | LB: 1.0878 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 14:54:52] Epoch 3 | Step 27820 | Loss: 0.2068 | LM: 0.1959 | LB: 1.0878 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 14:54:59] Epoch 3 | Step 27830 | Loss: 0.2068 | LM: 0.1958 | LB: 1.0878 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 14:55:05] Epoch 3 | Step 27840 | Loss: 0.2068 | LM: 0.1958 | LB: 1.0879 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 14:55:12] Epoch 3 | Step 27850 | Loss: 0.2068 | LM: 0.1958 | LB: 1.0878 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 14:55:19] Epoch 3 | Step 27860 | Loss: 0.2068 | LM: 0.1958 | LB: 1.0878 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 14:55:25] Epoch 3 | Step 27870 | Loss: 0.2067 | LM: 0.1958 | LB: 1.0878 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 14:55:32] Epoch 3 | Step 27880 | Loss: 0.2067 | LM: 0.1957 | LB: 1.0878 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 14:55:38] Epoch 3 | Step 27890 | Loss: 0.2067 | LM: 0.1956 | LB: 1.0878 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 14:55:45] Epoch 3 | Step 27900 | Loss: 0.2067 | LM: 0.1956 | LB: 1.0878 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 14:55:52] Epoch 3 | Step 27910 | Loss: 0.2068 | LM: 0.1956 | LB: 1.0878 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 14:55:58] Epoch 3 | Step 27920 | Loss: 0.2067 | LM: 0.1956 | LB: 1.0878 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 14:56:04] Epoch 3 | Step 27930 | Loss: 0.2068 | LM: 0.1957 | LB: 1.0878 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 14:56:11] Epoch 3 | Step 27940 | Loss: 0.2067 | LM: 0.1957 | LB: 1.0878 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 14:56:17] Epoch 3 | Step 27950 | Loss: 0.2067 | LM: 0.1958 | LB: 1.0878 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 14:56:24] Epoch 3 | Step 27960 | Loss: 0.2067 | LM: 0.1957 | LB: 1.0878 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 14:56:31] Epoch 3 | Step 27970 | Loss: 0.2067 | LM: 0.1958 | LB: 1.0878 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 14:56:37] Epoch 3 | Step 27980 | Loss: 0.2067 | LM: 0.1957 | LB: 1.0878 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 14:56:44] Epoch 3 | Step 27990 | Loss: 0.2066 | LM: 0.1957 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 14:56:50] Epoch 3 | Step 28000 | Loss: 0.2066 | LM: 0.1957 | LB: 1.0878 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 14:56:51] Validation | Batch 10/784 | Loss: 0.3352 | LM_LOSS: 0.3244 | LB_LOSS: 1.0846 [2026-04-17 14:56:52] Validation | Batch 20/784 | Loss: 0.3469 | LM_LOSS: 0.3361 | LB_LOSS: 1.0848 [2026-04-17 14:56:54] Validation | Batch 30/784 | Loss: 0.3327 | LM_LOSS: 0.3219 | LB_LOSS: 1.0841 [2026-04-17 14:56:55] Validation | Batch 40/784 | Loss: 0.3355 | LM_LOSS: 0.3246 | LB_LOSS: 1.0840 [2026-04-17 14:56:56] Validation | Batch 50/784 | Loss: 0.3331 | LM_LOSS: 0.3222 | LB_LOSS: 1.0834 [2026-04-17 14:56:58] Validation | Batch 60/784 | Loss: 0.3350 | LM_LOSS: 0.3242 | LB_LOSS: 1.0829 [2026-04-17 14:56:59] Validation | Batch 70/784 | Loss: 0.3322 | LM_LOSS: 0.3214 | LB_LOSS: 1.0823 [2026-04-17 14:57:00] Validation | Batch 80/784 | Loss: 0.3286 | LM_LOSS: 0.3177 | LB_LOSS: 1.0818 [2026-04-17 14:57:02] Validation | Batch 90/784 | Loss: 0.3274 | LM_LOSS: 0.3166 | LB_LOSS: 1.0824 [2026-04-17 14:57:03] Validation | Batch 100/784 | Loss: 0.3294 | LM_LOSS: 0.3186 | LB_LOSS: 1.0828 [2026-04-17 14:57:04] Validation | Batch 110/784 | Loss: 0.3238 | LM_LOSS: 0.3130 | LB_LOSS: 1.0830 [2026-04-17 14:57:06] Validation | Batch 120/784 | Loss: 0.3275 | LM_LOSS: 0.3166 | LB_LOSS: 1.0829 [2026-04-17 14:57:07] Validation | Batch 130/784 | Loss: 0.3305 | LM_LOSS: 0.3197 | LB_LOSS: 1.0829 [2026-04-17 14:57:09] Validation | Batch 140/784 | Loss: 0.3299 | LM_LOSS: 0.3191 | LB_LOSS: 1.0827 [2026-04-17 14:57:10] Validation | Batch 150/784 | Loss: 0.3260 | LM_LOSS: 0.3151 | LB_LOSS: 1.0830 [2026-04-17 14:57:11] Validation | Batch 160/784 | Loss: 0.3268 | LM_LOSS: 0.3160 | LB_LOSS: 1.0827 [2026-04-17 14:57:13] Validation | Batch 170/784 | Loss: 0.3269 | LM_LOSS: 0.3161 | LB_LOSS: 1.0824 [2026-04-17 14:57:14] Validation | Batch 180/784 | Loss: 0.3244 | LM_LOSS: 0.3136 | LB_LOSS: 1.0824 [2026-04-17 14:57:16] Validation | Batch 190/784 | Loss: 0.3267 | LM_LOSS: 0.3158 | LB_LOSS: 1.0828 [2026-04-17 14:57:17] Validation | Batch 200/784 | Loss: 0.3271 | LM_LOSS: 0.3163 | LB_LOSS: 1.0829 [2026-04-17 14:57:18] Validation | Batch 210/784 | Loss: 0.3260 | LM_LOSS: 0.3151 | LB_LOSS: 1.0828 [2026-04-17 14:57:20] Validation | Batch 220/784 | Loss: 0.3269 | LM_LOSS: 0.3160 | LB_LOSS: 1.0828 [2026-04-17 14:57:21] Validation | Batch 230/784 | Loss: 0.3275 | LM_LOSS: 0.3166 | LB_LOSS: 1.0827 [2026-04-17 14:57:22] Validation | Batch 240/784 | Loss: 0.3279 | LM_LOSS: 0.3171 | LB_LOSS: 1.0831 [2026-04-17 14:57:23] Validation | Batch 250/784 | Loss: 0.3278 | LM_LOSS: 0.3170 | LB_LOSS: 1.0829 [2026-04-17 14:57:25] Validation | Batch 260/784 | Loss: 0.3281 | LM_LOSS: 0.3173 | LB_LOSS: 1.0831 [2026-04-17 14:57:26] Validation | Batch 270/784 | Loss: 0.3280 | LM_LOSS: 0.3172 | LB_LOSS: 1.0832 [2026-04-17 14:57:28] Validation | Batch 280/784 | Loss: 0.3284 | LM_LOSS: 0.3176 | LB_LOSS: 1.0834 [2026-04-17 14:57:29] Validation | Batch 290/784 | Loss: 0.3296 | LM_LOSS: 0.3187 | LB_LOSS: 1.0835 [2026-04-17 14:57:30] Validation | Batch 300/784 | Loss: 0.3304 | LM_LOSS: 0.3195 | LB_LOSS: 1.0835 [2026-04-17 14:57:32] Validation | Batch 310/784 | Loss: 0.3298 | LM_LOSS: 0.3189 | LB_LOSS: 1.0835 [2026-04-17 14:57:33] Validation | Batch 320/784 | Loss: 0.3314 | LM_LOSS: 0.3206 | LB_LOSS: 1.0835 [2026-04-17 14:57:35] Validation | Batch 330/784 | Loss: 0.3312 | LM_LOSS: 0.3203 | LB_LOSS: 1.0834 [2026-04-17 14:57:36] Validation | Batch 340/784 | Loss: 0.3300 | LM_LOSS: 0.3191 | LB_LOSS: 1.0835 [2026-04-17 14:57:37] Validation | Batch 350/784 | Loss: 0.3302 | LM_LOSS: 0.3194 | LB_LOSS: 1.0837 [2026-04-17 14:57:38] Validation | Batch 360/784 | Loss: 0.3300 | LM_LOSS: 0.3192 | LB_LOSS: 1.0837 [2026-04-17 14:57:40] Validation | Batch 370/784 | Loss: 0.3305 | LM_LOSS: 0.3197 | LB_LOSS: 1.0837 [2026-04-17 14:57:41] Validation | Batch 380/784 | Loss: 0.3303 | LM_LOSS: 0.3195 | LB_LOSS: 1.0837 [2026-04-17 14:57:42] Validation | Batch 390/784 | Loss: 0.3303 | LM_LOSS: 0.3194 | LB_LOSS: 1.0838 [2026-04-17 14:57:43] Validation | Batch 400/784 | Loss: 0.3305 | LM_LOSS: 0.3197 | LB_LOSS: 1.0838 [2026-04-17 14:57:45] Validation | Batch 410/784 | Loss: 0.3309 | LM_LOSS: 0.3201 | LB_LOSS: 1.0838 [2026-04-17 14:57:46] Validation | Batch 420/784 | Loss: 0.3311 | LM_LOSS: 0.3203 | LB_LOSS: 1.0838 [2026-04-17 14:57:47] Validation | Batch 430/784 | Loss: 0.3312 | LM_LOSS: 0.3204 | LB_LOSS: 1.0838 [2026-04-17 14:57:48] Validation | Batch 440/784 | Loss: 0.3309 | LM_LOSS: 0.3201 | LB_LOSS: 1.0838 [2026-04-17 14:57:50] Validation | Batch 450/784 | Loss: 0.3302 | LM_LOSS: 0.3193 | LB_LOSS: 1.0838 [2026-04-17 14:57:51] Validation | Batch 460/784 | Loss: 0.3307 | LM_LOSS: 0.3198 | LB_LOSS: 1.0838 [2026-04-17 14:57:53] Validation | Batch 470/784 | Loss: 0.3298 | LM_LOSS: 0.3190 | LB_LOSS: 1.0838 [2026-04-17 14:57:54] Validation | Batch 480/784 | Loss: 0.3303 | LM_LOSS: 0.3195 | LB_LOSS: 1.0838 [2026-04-17 14:57:55] Validation | Batch 490/784 | Loss: 0.3297 | LM_LOSS: 0.3188 | LB_LOSS: 1.0837 [2026-04-17 14:57:57] Validation | Batch 500/784 | Loss: 0.3300 | LM_LOSS: 0.3192 | LB_LOSS: 1.0836 [2026-04-17 14:57:58] Validation | Batch 510/784 | Loss: 0.3297 | LM_LOSS: 0.3189 | LB_LOSS: 1.0836 [2026-04-17 14:57:59] Validation | Batch 520/784 | Loss: 0.3300 | LM_LOSS: 0.3191 | LB_LOSS: 1.0835 [2026-04-17 14:58:01] Validation | Batch 530/784 | Loss: 0.3308 | LM_LOSS: 0.3200 | LB_LOSS: 1.0835 [2026-04-17 14:58:02] Validation | Batch 540/784 | Loss: 0.3312 | LM_LOSS: 0.3204 | LB_LOSS: 1.0835 [2026-04-17 14:58:04] Validation | Batch 550/784 | Loss: 0.3325 | LM_LOSS: 0.3217 | LB_LOSS: 1.0835 [2026-04-17 14:58:05] Validation | Batch 560/784 | Loss: 0.3326 | LM_LOSS: 0.3218 | LB_LOSS: 1.0835 [2026-04-17 14:58:07] Validation | Batch 570/784 | Loss: 0.3321 | LM_LOSS: 0.3213 | LB_LOSS: 1.0834 [2026-04-17 14:58:08] Validation | Batch 580/784 | Loss: 0.3316 | LM_LOSS: 0.3208 | LB_LOSS: 1.0835 [2026-04-17 14:58:09] Validation | Batch 590/784 | Loss: 0.3319 | LM_LOSS: 0.3210 | LB_LOSS: 1.0834 [2026-04-17 14:58:11] Validation | Batch 600/784 | Loss: 0.3317 | LM_LOSS: 0.3209 | LB_LOSS: 1.0834 [2026-04-17 14:58:12] Validation | Batch 610/784 | Loss: 0.3319 | LM_LOSS: 0.3210 | LB_LOSS: 1.0834 [2026-04-17 14:58:13] Validation | Batch 620/784 | Loss: 0.3317 | LM_LOSS: 0.3209 | LB_LOSS: 1.0834 [2026-04-17 14:58:15] Validation | Batch 630/784 | Loss: 0.3325 | LM_LOSS: 0.3217 | LB_LOSS: 1.0834 [2026-04-17 14:58:17] Validation | Batch 640/784 | Loss: 0.3326 | LM_LOSS: 0.3217 | LB_LOSS: 1.0834 [2026-04-17 14:58:18] Validation | Batch 650/784 | Loss: 0.3324 | LM_LOSS: 0.3216 | LB_LOSS: 1.0834 [2026-04-17 14:58:19] Validation | Batch 660/784 | Loss: 0.3328 | LM_LOSS: 0.3220 | LB_LOSS: 1.0834 [2026-04-17 14:58:21] Validation | Batch 670/784 | Loss: 0.3332 | LM_LOSS: 0.3224 | LB_LOSS: 1.0835 [2026-04-17 14:58:22] Validation | Batch 680/784 | Loss: 0.3329 | LM_LOSS: 0.3221 | LB_LOSS: 1.0835 [2026-04-17 14:58:24] Validation | Batch 690/784 | Loss: 0.3331 | LM_LOSS: 0.3222 | LB_LOSS: 1.0834 [2026-04-17 14:58:25] Validation | Batch 700/784 | Loss: 0.3331 | LM_LOSS: 0.3223 | LB_LOSS: 1.0834 [2026-04-17 14:58:26] Validation | Batch 710/784 | Loss: 0.3329 | LM_LOSS: 0.3220 | LB_LOSS: 1.0833 [2026-04-17 14:58:28] Validation | Batch 720/784 | Loss: 0.3326 | LM_LOSS: 0.3218 | LB_LOSS: 1.0832 [2026-04-17 14:58:29] Validation | Batch 730/784 | Loss: 0.3321 | LM_LOSS: 0.3213 | LB_LOSS: 1.0832 [2026-04-17 14:58:30] Validation | Batch 740/784 | Loss: 0.3321 | LM_LOSS: 0.3213 | LB_LOSS: 1.0833 [2026-04-17 14:58:31] Validation | Batch 750/784 | Loss: 0.3315 | LM_LOSS: 0.3206 | LB_LOSS: 1.0833 [2026-04-17 14:58:33] Validation | Batch 760/784 | Loss: 0.3316 | LM_LOSS: 0.3208 | LB_LOSS: 1.0833 [2026-04-17 14:58:34] Validation | Batch 770/784 | Loss: 0.3318 | LM_LOSS: 0.3210 | LB_LOSS: 1.0833 [2026-04-17 14:58:35] Validation | Batch 780/784 | Loss: 0.3322 | LM_LOSS: 0.3213 | LB_LOSS: 1.0833 [2026-04-17 14:58:36] Validation | Batch 784/784 | Loss: 0.3324 | LM_LOSS: 0.3215 | LB_LOSS: 1.0833 [2026-04-17 14:58:39] Validation | Loss: 0.3324 | LM_LOSS: 0.3215 | LB_LOSS: 1.0833 | PPL: 1.38 | Time: 105.72s [2026-04-17 14:58:45] Epoch 3 | Step 28010 | Loss: 0.2066 | LM: 0.1957 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 14:58:51] Epoch 3 | Step 28020 | Loss: 0.2066 | LM: 0.1958 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 14:58:58] Epoch 3 | Step 28030 | Loss: 0.2066 | LM: 0.1957 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 14:59:04] Epoch 3 | Step 28040 | Loss: 0.2066 | LM: 0.1956 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 14:59:11] Epoch 3 | Step 28050 | Loss: 0.2066 | LM: 0.1955 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 14:59:17] Epoch 3 | Step 28060 | Loss: 0.2065 | LM: 0.1955 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 14:59:23] Epoch 3 | Step 28070 | Loss: 0.2066 | LM: 0.1954 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 14:59:30] Epoch 3 | Step 28080 | Loss: 0.2066 | LM: 0.1955 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 14:59:36] Epoch 3 | Step 28090 | Loss: 0.2066 | LM: 0.1955 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 14:59:42] Epoch 3 | Step 28100 | Loss: 0.2065 | LM: 0.1955 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 14:59:49] Epoch 3 | Step 28110 | Loss: 0.2066 | LM: 0.1955 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 14:59:55] Epoch 3 | Step 28120 | Loss: 0.2065 | LM: 0.1955 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 15:00:01] Epoch 3 | Step 28130 | Loss: 0.2065 | LM: 0.1954 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 15:00:08] Epoch 3 | Step 28140 | Loss: 0.2065 | LM: 0.1954 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 15:00:14] Epoch 3 | Step 28150 | Loss: 0.2065 | LM: 0.1954 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 15:00:21] Epoch 3 | Step 28160 | Loss: 0.2065 | LM: 0.1954 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 15:00:27] Epoch 3 | Step 28170 | Loss: 0.2064 | LM: 0.1953 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 15:00:34] Epoch 3 | Step 28180 | Loss: 0.2065 | LM: 0.1954 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 15:00:40] Epoch 3 | Step 28190 | Loss: 0.2065 | LM: 0.1954 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 15:00:46] Epoch 3 | Step 28200 | Loss: 0.2065 | LM: 0.1953 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 15:00:53] Epoch 3 | Step 28210 | Loss: 0.2065 | LM: 0.1954 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 15:00:59] Epoch 3 | Step 28220 | Loss: 0.2065 | LM: 0.1954 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 15:01:05] Epoch 3 | Step 28230 | Loss: 0.2065 | LM: 0.1954 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 15:01:12] Epoch 3 | Step 28240 | Loss: 0.2066 | LM: 0.1955 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 15:01:18] Epoch 3 | Step 28250 | Loss: 0.2066 | LM: 0.1955 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 15:01:25] Epoch 3 | Step 28260 | Loss: 0.2066 | LM: 0.1956 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 15:01:31] Epoch 3 | Step 28270 | Loss: 0.2066 | LM: 0.1955 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 15:01:38] Epoch 3 | Step 28280 | Loss: 0.2066 | LM: 0.1956 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 15:01:44] Epoch 3 | Step 28290 | Loss: 0.2066 | LM: 0.1955 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 15:01:50] Epoch 3 | Step 28300 | Loss: 0.2066 | LM: 0.1956 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 15:01:56] Epoch 3 | Step 28310 | Loss: 0.2066 | LM: 0.1956 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 15:02:03] Epoch 3 | Step 28320 | Loss: 0.2066 | LM: 0.1956 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 15:02:09] Epoch 3 | Step 28330 | Loss: 0.2066 | LM: 0.1955 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 15:02:16] Epoch 3 | Step 28340 | Loss: 0.2066 | LM: 0.1954 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 15:02:23] Epoch 3 | Step 28350 | Loss: 0.2066 | LM: 0.1954 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 15:02:29] Epoch 3 | Step 28360 | Loss: 0.2066 | LM: 0.1954 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 15:02:35] Epoch 3 | Step 28370 | Loss: 0.2065 | LM: 0.1954 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 15:02:42] Epoch 3 | Step 28380 | Loss: 0.2064 | LM: 0.1953 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 15:02:48] Epoch 3 | Step 28390 | Loss: 0.2065 | LM: 0.1953 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 15:02:54] Epoch 3 | Step 28400 | Loss: 0.2064 | LM: 0.1952 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 15:03:00] Epoch 3 | Step 28410 | Loss: 0.2064 | LM: 0.1952 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 15:03:07] Epoch 3 | Step 28420 | Loss: 0.2064 | LM: 0.1952 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 15:03:13] Epoch 3 | Step 28430 | Loss: 0.2064 | LM: 0.1952 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 15:03:20] Epoch 3 | Step 28440 | Loss: 0.2065 | LM: 0.1952 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 15:03:26] Epoch 3 | Step 28450 | Loss: 0.2064 | LM: 0.1951 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 15:03:33] Epoch 3 | Step 28460 | Loss: 0.2065 | LM: 0.1951 | LB: 1.0878 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 15:03:39] Epoch 3 | Step 28470 | Loss: 0.2064 | LM: 0.1950 | LB: 1.0878 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 15:03:46] Epoch 3 | Step 28480 | Loss: 0.2064 | LM: 0.1950 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 15:03:51] Epoch 3 | Step 28490 | Loss: 0.2065 | LM: 0.1950 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 15:03:58] Epoch 3 | Step 28500 | Loss: 0.2065 | LM: 0.1950 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 15:04:04] Epoch 3 | Step 28510 | Loss: 0.2065 | LM: 0.1951 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 15:04:10] Epoch 3 | Step 28520 | Loss: 0.2065 | LM: 0.1951 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 15:04:17] Epoch 3 | Step 28530 | Loss: 0.2065 | LM: 0.1951 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 15:04:23] Epoch 3 | Step 28540 | Loss: 0.2065 | LM: 0.1951 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 15:04:30] Epoch 3 | Step 28550 | Loss: 0.2065 | LM: 0.1952 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 15:04:36] Epoch 3 | Step 28560 | Loss: 0.2066 | LM: 0.1951 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 15:04:42] Epoch 3 | Step 28570 | Loss: 0.2065 | LM: 0.1950 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 15:04:48] Epoch 3 | Step 28580 | Loss: 0.2065 | LM: 0.1950 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 15:04:55] Epoch 3 | Step 28590 | Loss: 0.2065 | LM: 0.1950 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 15:05:01] Epoch 3 | Step 28600 | Loss: 0.2065 | LM: 0.1950 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 15:05:07] Epoch 3 | Step 28610 | Loss: 0.2065 | LM: 0.1951 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 15:05:13] Epoch 3 | Step 28620 | Loss: 0.2065 | LM: 0.1951 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 15:05:19] Epoch 3 | Step 28630 | Loss: 0.2065 | LM: 0.1950 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 15:05:26] Epoch 3 | Step 28640 | Loss: 0.2065 | LM: 0.1951 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 15:05:33] Epoch 3 | Step 28650 | Loss: 0.2065 | LM: 0.1950 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 15:05:39] Epoch 3 | Step 28660 | Loss: 0.2065 | LM: 0.1950 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 15:05:46] Epoch 3 | Step 28670 | Loss: 0.2064 | LM: 0.1950 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 15:05:52] Epoch 3 | Step 28680 | Loss: 0.2065 | LM: 0.1950 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 15:05:58] Epoch 3 | Step 28690 | Loss: 0.2065 | LM: 0.1950 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 15:06:04] Epoch 3 | Step 28700 | Loss: 0.2064 | LM: 0.1950 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 15:06:10] Epoch 3 | Step 28710 | Loss: 0.2064 | LM: 0.1950 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 15:06:17] Epoch 3 | Step 28720 | Loss: 0.2065 | LM: 0.1950 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 15:06:23] Epoch 3 | Step 28730 | Loss: 0.2065 | LM: 0.1951 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 15:06:29] Epoch 3 | Step 28740 | Loss: 0.2064 | LM: 0.1950 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 15:06:37] Epoch 3 | Step 28750 | Loss: 0.2064 | LM: 0.1951 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 15:06:43] Epoch 3 | Step 28760 | Loss: 0.2064 | LM: 0.1951 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 15:06:49] Epoch 3 | Step 28770 | Loss: 0.2065 | LM: 0.1951 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 15:06:56] Epoch 3 | Step 28780 | Loss: 0.2065 | LM: 0.1952 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 15:07:02] Epoch 3 | Step 28790 | Loss: 0.2065 | LM: 0.1951 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 15:07:08] Epoch 3 | Step 28800 | Loss: 0.2065 | LM: 0.1951 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 15:07:15] Epoch 3 | Step 28810 | Loss: 0.2065 | LM: 0.1951 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 15:07:21] Epoch 3 | Step 28820 | Loss: 0.2065 | LM: 0.1951 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 15:07:27] Epoch 3 | Step 28830 | Loss: 0.2064 | LM: 0.1951 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 15:07:34] Epoch 3 | Step 28840 | Loss: 0.2064 | LM: 0.1951 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 15:07:41] Epoch 3 | Step 28850 | Loss: 0.2064 | LM: 0.1951 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 15:07:47] Epoch 3 | Step 28860 | Loss: 0.2065 | LM: 0.1951 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 15:07:54] Epoch 3 | Step 28870 | Loss: 0.2064 | LM: 0.1951 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 15:08:00] Epoch 3 | Step 28880 | Loss: 0.2064 | LM: 0.1951 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 15:08:07] Epoch 3 | Step 28890 | Loss: 0.2064 | LM: 0.1950 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 15:08:13] Epoch 3 | Step 28900 | Loss: 0.2063 | LM: 0.1950 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 15:08:19] Epoch 3 | Step 28910 | Loss: 0.2063 | LM: 0.1950 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 15:08:26] Epoch 3 | Step 28920 | Loss: 0.2063 | LM: 0.1950 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 15:08:32] Epoch 3 | Step 28930 | Loss: 0.2063 | LM: 0.1949 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 15:08:39] Epoch 3 | Step 28940 | Loss: 0.2063 | LM: 0.1950 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 15:08:46] Epoch 3 | Step 28950 | Loss: 0.2063 | LM: 0.1950 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 15:08:52] Epoch 3 | Step 28960 | Loss: 0.2063 | LM: 0.1950 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 15:08:58] Epoch 3 | Step 28970 | Loss: 0.2062 | LM: 0.1949 | LB: 1.0877 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 15:09:05] Epoch 3 | Step 28980 | Loss: 0.2062 | LM: 0.1950 | LB: 1.0876 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 15:09:11] Epoch 3 | Step 28990 | Loss: 0.2062 | LM: 0.1949 | LB: 1.0876 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 15:09:17] Epoch 3 | Step 29000 | Loss: 0.2062 | LM: 0.1949 | LB: 1.0876 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 15:09:18] Validation | Batch 10/784 | Loss: 0.3360 | LM_LOSS: 0.3251 | LB_LOSS: 1.0846 [2026-04-17 15:09:20] Validation | Batch 20/784 | Loss: 0.3474 | LM_LOSS: 0.3365 | LB_LOSS: 1.0847 [2026-04-17 15:09:21] Validation | Batch 30/784 | Loss: 0.3331 | LM_LOSS: 0.3223 | LB_LOSS: 1.0840 [2026-04-17 15:09:23] Validation | Batch 40/784 | Loss: 0.3360 | LM_LOSS: 0.3252 | LB_LOSS: 1.0839 [2026-04-17 15:09:24] Validation | Batch 50/784 | Loss: 0.3336 | LM_LOSS: 0.3228 | LB_LOSS: 1.0833 [2026-04-17 15:09:25] Validation | Batch 60/784 | Loss: 0.3356 | LM_LOSS: 0.3247 | LB_LOSS: 1.0829 [2026-04-17 15:09:27] Validation | Batch 70/784 | Loss: 0.3329 | LM_LOSS: 0.3221 | LB_LOSS: 1.0822 [2026-04-17 15:09:28] Validation | Batch 80/784 | Loss: 0.3292 | LM_LOSS: 0.3184 | LB_LOSS: 1.0818 [2026-04-17 15:09:29] Validation | Batch 90/784 | Loss: 0.3281 | LM_LOSS: 0.3173 | LB_LOSS: 1.0823 [2026-04-17 15:09:31] Validation | Batch 100/784 | Loss: 0.3301 | LM_LOSS: 0.3193 | LB_LOSS: 1.0828 [2026-04-17 15:09:32] Validation | Batch 110/784 | Loss: 0.3246 | LM_LOSS: 0.3137 | LB_LOSS: 1.0829 [2026-04-17 15:09:33] Validation | Batch 120/784 | Loss: 0.3282 | LM_LOSS: 0.3173 | LB_LOSS: 1.0828 [2026-04-17 15:09:35] Validation | Batch 130/784 | Loss: 0.3312 | LM_LOSS: 0.3204 | LB_LOSS: 1.0828 [2026-04-17 15:09:36] Validation | Batch 140/784 | Loss: 0.3306 | LM_LOSS: 0.3198 | LB_LOSS: 1.0826 [2026-04-17 15:09:38] Validation | Batch 150/784 | Loss: 0.3266 | LM_LOSS: 0.3158 | LB_LOSS: 1.0829 [2026-04-17 15:09:39] Validation | Batch 160/784 | Loss: 0.3274 | LM_LOSS: 0.3166 | LB_LOSS: 1.0826 [2026-04-17 15:09:41] Validation | Batch 170/784 | Loss: 0.3276 | LM_LOSS: 0.3167 | LB_LOSS: 1.0823 [2026-04-17 15:09:42] Validation | Batch 180/784 | Loss: 0.3251 | LM_LOSS: 0.3143 | LB_LOSS: 1.0823 [2026-04-17 15:09:43] Validation | Batch 190/784 | Loss: 0.3273 | LM_LOSS: 0.3165 | LB_LOSS: 1.0828 [2026-04-17 15:09:44] Validation | Batch 200/784 | Loss: 0.3278 | LM_LOSS: 0.3169 | LB_LOSS: 1.0828 [2026-04-17 15:09:46] Validation | Batch 210/784 | Loss: 0.3266 | LM_LOSS: 0.3158 | LB_LOSS: 1.0827 [2026-04-17 15:09:47] Validation | Batch 220/784 | Loss: 0.3275 | LM_LOSS: 0.3167 | LB_LOSS: 1.0828 [2026-04-17 15:09:49] Validation | Batch 230/784 | Loss: 0.3281 | LM_LOSS: 0.3173 | LB_LOSS: 1.0827 [2026-04-17 15:09:50] Validation | Batch 240/784 | Loss: 0.3286 | LM_LOSS: 0.3177 | LB_LOSS: 1.0830 [2026-04-17 15:09:51] Validation | Batch 250/784 | Loss: 0.3284 | LM_LOSS: 0.3176 | LB_LOSS: 1.0829 [2026-04-17 15:09:53] Validation | Batch 260/784 | Loss: 0.3287 | LM_LOSS: 0.3179 | LB_LOSS: 1.0831 [2026-04-17 15:09:54] Validation | Batch 270/784 | Loss: 0.3286 | LM_LOSS: 0.3178 | LB_LOSS: 1.0831 [2026-04-17 15:09:56] Validation | Batch 280/784 | Loss: 0.3290 | LM_LOSS: 0.3182 | LB_LOSS: 1.0833 [2026-04-17 15:09:57] Validation | Batch 290/784 | Loss: 0.3301 | LM_LOSS: 0.3193 | LB_LOSS: 1.0834 [2026-04-17 15:09:58] Validation | Batch 300/784 | Loss: 0.3310 | LM_LOSS: 0.3201 | LB_LOSS: 1.0834 [2026-04-17 15:09:59] Validation | Batch 310/784 | Loss: 0.3304 | LM_LOSS: 0.3195 | LB_LOSS: 1.0834 [2026-04-17 15:10:01] Validation | Batch 320/784 | Loss: 0.3320 | LM_LOSS: 0.3211 | LB_LOSS: 1.0834 [2026-04-17 15:10:02] Validation | Batch 330/784 | Loss: 0.3318 | LM_LOSS: 0.3209 | LB_LOSS: 1.0834 [2026-04-17 15:10:03] Validation | Batch 340/784 | Loss: 0.3305 | LM_LOSS: 0.3197 | LB_LOSS: 1.0835 [2026-04-17 15:10:05] Validation | Batch 350/784 | Loss: 0.3308 | LM_LOSS: 0.3199 | LB_LOSS: 1.0836 [2026-04-17 15:10:06] Validation | Batch 360/784 | Loss: 0.3306 | LM_LOSS: 0.3197 | LB_LOSS: 1.0837 [2026-04-17 15:10:07] Validation | Batch 370/784 | Loss: 0.3311 | LM_LOSS: 0.3203 | LB_LOSS: 1.0836 [2026-04-17 15:10:09] Validation | Batch 380/784 | Loss: 0.3309 | LM_LOSS: 0.3201 | LB_LOSS: 1.0836 [2026-04-17 15:10:10] Validation | Batch 390/784 | Loss: 0.3308 | LM_LOSS: 0.3200 | LB_LOSS: 1.0837 [2026-04-17 15:10:11] Validation | Batch 400/784 | Loss: 0.3311 | LM_LOSS: 0.3203 | LB_LOSS: 1.0837 [2026-04-17 15:10:12] Validation | Batch 410/784 | Loss: 0.3315 | LM_LOSS: 0.3206 | LB_LOSS: 1.0837 [2026-04-17 15:10:14] Validation | Batch 420/784 | Loss: 0.3317 | LM_LOSS: 0.3209 | LB_LOSS: 1.0838 [2026-04-17 15:10:15] Validation | Batch 430/784 | Loss: 0.3318 | LM_LOSS: 0.3210 | LB_LOSS: 1.0837 [2026-04-17 15:10:16] Validation | Batch 440/784 | Loss: 0.3315 | LM_LOSS: 0.3207 | LB_LOSS: 1.0837 [2026-04-17 15:10:18] Validation | Batch 450/784 | Loss: 0.3308 | LM_LOSS: 0.3199 | LB_LOSS: 1.0837 [2026-04-17 15:10:19] Validation | Batch 460/784 | Loss: 0.3312 | LM_LOSS: 0.3204 | LB_LOSS: 1.0838 [2026-04-17 15:10:20] Validation | Batch 470/784 | Loss: 0.3304 | LM_LOSS: 0.3196 | LB_LOSS: 1.0837 [2026-04-17 15:10:22] Validation | Batch 480/784 | Loss: 0.3309 | LM_LOSS: 0.3201 | LB_LOSS: 1.0837 [2026-04-17 15:10:23] Validation | Batch 490/784 | Loss: 0.3303 | LM_LOSS: 0.3194 | LB_LOSS: 1.0836 [2026-04-17 15:10:24] Validation | Batch 500/784 | Loss: 0.3306 | LM_LOSS: 0.3198 | LB_LOSS: 1.0836 [2026-04-17 15:10:26] Validation | Batch 510/784 | Loss: 0.3303 | LM_LOSS: 0.3195 | LB_LOSS: 1.0835 [2026-04-17 15:10:27] Validation | Batch 520/784 | Loss: 0.3306 | LM_LOSS: 0.3198 | LB_LOSS: 1.0835 [2026-04-17 15:10:28] Validation | Batch 530/784 | Loss: 0.3315 | LM_LOSS: 0.3206 | LB_LOSS: 1.0834 [2026-04-17 15:10:30] Validation | Batch 540/784 | Loss: 0.3318 | LM_LOSS: 0.3210 | LB_LOSS: 1.0834 [2026-04-17 15:10:31] Validation | Batch 550/784 | Loss: 0.3331 | LM_LOSS: 0.3223 | LB_LOSS: 1.0834 [2026-04-17 15:10:33] Validation | Batch 560/784 | Loss: 0.3333 | LM_LOSS: 0.3224 | LB_LOSS: 1.0834 [2026-04-17 15:10:34] Validation | Batch 570/784 | Loss: 0.3328 | LM_LOSS: 0.3219 | LB_LOSS: 1.0833 [2026-04-17 15:10:35] Validation | Batch 580/784 | Loss: 0.3322 | LM_LOSS: 0.3214 | LB_LOSS: 1.0834 [2026-04-17 15:10:37] Validation | Batch 590/784 | Loss: 0.3325 | LM_LOSS: 0.3217 | LB_LOSS: 1.0833 [2026-04-17 15:10:38] Validation | Batch 600/784 | Loss: 0.3324 | LM_LOSS: 0.3215 | LB_LOSS: 1.0833 [2026-04-17 15:10:39] Validation | Batch 610/784 | Loss: 0.3325 | LM_LOSS: 0.3217 | LB_LOSS: 1.0833 [2026-04-17 15:10:41] Validation | Batch 620/784 | Loss: 0.3323 | LM_LOSS: 0.3215 | LB_LOSS: 1.0833 [2026-04-17 15:10:42] Validation | Batch 630/784 | Loss: 0.3332 | LM_LOSS: 0.3223 | LB_LOSS: 1.0833 [2026-04-17 15:10:44] Validation | Batch 640/784 | Loss: 0.3332 | LM_LOSS: 0.3224 | LB_LOSS: 1.0833 [2026-04-17 15:10:45] Validation | Batch 650/784 | Loss: 0.3331 | LM_LOSS: 0.3222 | LB_LOSS: 1.0834 [2026-04-17 15:10:47] Validation | Batch 660/784 | Loss: 0.3335 | LM_LOSS: 0.3226 | LB_LOSS: 1.0833 [2026-04-17 15:10:48] Validation | Batch 670/784 | Loss: 0.3339 | LM_LOSS: 0.3230 | LB_LOSS: 1.0834 [2026-04-17 15:10:49] Validation | Batch 680/784 | Loss: 0.3336 | LM_LOSS: 0.3227 | LB_LOSS: 1.0834 [2026-04-17 15:10:51] Validation | Batch 690/784 | Loss: 0.3337 | LM_LOSS: 0.3229 | LB_LOSS: 1.0833 [2026-04-17 15:10:52] Validation | Batch 700/784 | Loss: 0.3338 | LM_LOSS: 0.3229 | LB_LOSS: 1.0833 [2026-04-17 15:10:54] Validation | Batch 710/784 | Loss: 0.3335 | LM_LOSS: 0.3227 | LB_LOSS: 1.0832 [2026-04-17 15:10:55] Validation | Batch 720/784 | Loss: 0.3332 | LM_LOSS: 0.3224 | LB_LOSS: 1.0832 [2026-04-17 15:10:56] Validation | Batch 730/784 | Loss: 0.3327 | LM_LOSS: 0.3219 | LB_LOSS: 1.0831 [2026-04-17 15:10:58] Validation | Batch 740/784 | Loss: 0.3328 | LM_LOSS: 0.3220 | LB_LOSS: 1.0832 [2026-04-17 15:10:59] Validation | Batch 750/784 | Loss: 0.3321 | LM_LOSS: 0.3213 | LB_LOSS: 1.0832 [2026-04-17 15:11:00] Validation | Batch 760/784 | Loss: 0.3322 | LM_LOSS: 0.3214 | LB_LOSS: 1.0832 [2026-04-17 15:11:02] Validation | Batch 770/784 | Loss: 0.3324 | LM_LOSS: 0.3216 | LB_LOSS: 1.0832 [2026-04-17 15:11:03] Validation | Batch 780/784 | Loss: 0.3328 | LM_LOSS: 0.3219 | LB_LOSS: 1.0832 [2026-04-17 15:11:04] Validation | Batch 784/784 | Loss: 0.3330 | LM_LOSS: 0.3222 | LB_LOSS: 1.0832 [2026-04-17 15:11:06] Validation | Loss: 0.3330 | LM_LOSS: 0.3222 | LB_LOSS: 1.0832 | PPL: 1.38 | Time: 106.39s [2026-04-17 15:11:13] Epoch 3 | Step 29010 | Loss: 0.2061 | LM: 0.1948 | LB: 1.0876 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 15:11:19] Epoch 3 | Step 29020 | Loss: 0.2061 | LM: 0.1948 | LB: 1.0876 | CL0: 2.9 | CL1: 2.4 | HR0: 0.348/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 15:11:25] Epoch 3 | Step 29030 | Loss: 0.2061 | LM: 0.1948 | LB: 1.0876 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 15:11:32] Epoch 3 | Step 29040 | Loss: 0.2061 | LM: 0.1948 | LB: 1.0876 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 15:11:38] Epoch 3 | Step 29050 | Loss: 0.2061 | LM: 0.1948 | LB: 1.0876 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 15:11:45] Epoch 3 | Step 29060 | Loss: 0.2061 | LM: 0.1947 | LB: 1.0876 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 15:11:51] Epoch 3 | Step 29070 | Loss: 0.2061 | LM: 0.1948 | LB: 1.0876 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 15:11:57] Epoch 3 | Step 29080 | Loss: 0.2061 | LM: 0.1948 | LB: 1.0876 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 15:12:04] Epoch 3 | Step 29090 | Loss: 0.2061 | LM: 0.1949 | LB: 1.0876 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 15:12:10] Epoch 3 | Step 29100 | Loss: 0.2062 | LM: 0.1949 | LB: 1.0876 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 15:12:17] Epoch 3 | Step 29110 | Loss: 0.2061 | LM: 0.1949 | LB: 1.0876 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 15:12:23] Epoch 3 | Step 29120 | Loss: 0.2061 | LM: 0.1948 | LB: 1.0876 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 15:12:30] Epoch 3 | Step 29130 | Loss: 0.2061 | LM: 0.1947 | LB: 1.0876 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 15:12:36] Epoch 3 | Step 29140 | Loss: 0.2061 | LM: 0.1946 | LB: 1.0876 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 15:12:43] Epoch 3 | Step 29150 | Loss: 0.2061 | LM: 0.1946 | LB: 1.0876 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 15:12:49] Epoch 3 | Step 29160 | Loss: 0.2061 | LM: 0.1945 | LB: 1.0876 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 15:12:55] Epoch 3 | Step 29170 | Loss: 0.2061 | LM: 0.1945 | LB: 1.0876 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 15:13:02] Epoch 3 | Step 29180 | Loss: 0.2060 | LM: 0.1945 | LB: 1.0876 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 15:13:08] Epoch 3 | Step 29190 | Loss: 0.2061 | LM: 0.1945 | LB: 1.0876 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 15:13:14] Epoch 3 | Step 29200 | Loss: 0.2060 | LM: 0.1946 | LB: 1.0876 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 15:13:21] Epoch 3 | Step 29210 | Loss: 0.2061 | LM: 0.1946 | LB: 1.0876 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 15:13:27] Epoch 3 | Step 29220 | Loss: 0.2061 | LM: 0.1946 | LB: 1.0876 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 15:13:34] Epoch 3 | Step 29230 | Loss: 0.2061 | LM: 0.1947 | LB: 1.0876 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 15:13:40] Epoch 3 | Step 29240 | Loss: 0.2061 | LM: 0.1947 | LB: 1.0876 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 15:13:46] Epoch 3 | Step 29250 | Loss: 0.2061 | LM: 0.1948 | LB: 1.0876 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 15:13:53] Epoch 3 | Step 29260 | Loss: 0.2060 | LM: 0.1947 | LB: 1.0876 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 15:13:59] Epoch 3 | Step 29270 | Loss: 0.2060 | LM: 0.1947 | LB: 1.0876 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 15:14:06] Epoch 3 | Step 29280 | Loss: 0.2060 | LM: 0.1947 | LB: 1.0876 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 15:14:12] Epoch 3 | Step 29290 | Loss: 0.2060 | LM: 0.1946 | LB: 1.0876 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 15:14:18] Epoch 3 | Step 29300 | Loss: 0.2060 | LM: 0.1946 | LB: 1.0876 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 15:14:25] Epoch 3 | Step 29310 | Loss: 0.2060 | LM: 0.1947 | LB: 1.0876 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 15:14:31] Epoch 3 | Step 29320 | Loss: 0.2060 | LM: 0.1946 | LB: 1.0876 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 15:14:38] Epoch 3 | Step 29330 | Loss: 0.2060 | LM: 0.1946 | LB: 1.0876 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 15:14:44] Epoch 3 | Step 29340 | Loss: 0.2060 | LM: 0.1945 | LB: 1.0876 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 15:14:50] Epoch 3 | Step 29350 | Loss: 0.2060 | LM: 0.1945 | LB: 1.0876 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 15:14:57] Epoch 3 | Step 29360 | Loss: 0.2059 | LM: 0.1945 | LB: 1.0876 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.385 | LR: 1.00e-05 [2026-04-17 15:15:03] Epoch 3 | Step 29370 | Loss: 0.2059 | LM: 0.1945 | LB: 1.0876 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 15:15:10] Epoch 3 | Step 29380 | Loss: 0.2059 | LM: 0.1945 | LB: 1.0876 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 15:15:16] Epoch 3 | Step 29390 | Loss: 0.2060 | LM: 0.1945 | LB: 1.0876 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 15:15:23] Epoch 3 | Step 29400 | Loss: 0.2060 | LM: 0.1945 | LB: 1.0876 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 15:15:29] Epoch 3 | Step 29410 | Loss: 0.2060 | LM: 0.1945 | LB: 1.0876 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 15:15:35] Epoch 3 | Step 29420 | Loss: 0.2060 | LM: 0.1944 | LB: 1.0876 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 15:15:42] Epoch 3 | Step 29430 | Loss: 0.2060 | LM: 0.1945 | LB: 1.0876 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 15:15:48] Epoch 3 | Step 29440 | Loss: 0.2060 | LM: 0.1945 | LB: 1.0876 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 15:15:54] Epoch 3 | Step 29450 | Loss: 0.2060 | LM: 0.1945 | LB: 1.0876 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 15:16:01] Epoch 3 | Step 29460 | Loss: 0.2060 | LM: 0.1945 | LB: 1.0876 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 15:16:07] Epoch 3 | Step 29470 | Loss: 0.2059 | LM: 0.1944 | LB: 1.0876 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 15:16:13] Epoch 3 | Step 29480 | Loss: 0.2060 | LM: 0.1944 | LB: 1.0876 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 15:16:20] Epoch 3 | Step 29490 | Loss: 0.2060 | LM: 0.1945 | LB: 1.0876 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 15:16:26] Epoch 3 | Step 29500 | Loss: 0.2060 | LM: 0.1944 | LB: 1.0876 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 15:16:32] Epoch 3 | Step 29510 | Loss: 0.2060 | LM: 0.1944 | LB: 1.0876 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 15:16:39] Epoch 3 | Step 29520 | Loss: 0.2060 | LM: 0.1945 | LB: 1.0876 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 15:16:45] Epoch 3 | Step 29530 | Loss: 0.2060 | LM: 0.1945 | LB: 1.0876 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 15:16:52] Epoch 3 | Step 29540 | Loss: 0.2060 | LM: 0.1945 | LB: 1.0876 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 15:16:58] Epoch 3 | Step 29550 | Loss: 0.2060 | LM: 0.1945 | LB: 1.0876 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 15:17:05] Epoch 3 | Step 29560 | Loss: 0.2060 | LM: 0.1945 | LB: 1.0875 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 15:17:11] Epoch 3 | Step 29570 | Loss: 0.2060 | LM: 0.1945 | LB: 1.0876 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 15:17:17] Epoch 3 | Step 29580 | Loss: 0.2060 | LM: 0.1944 | LB: 1.0876 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 15:17:24] Epoch 3 | Step 29590 | Loss: 0.2059 | LM: 0.1944 | LB: 1.0875 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 15:17:30] Epoch 3 | Step 29600 | Loss: 0.2059 | LM: 0.1944 | LB: 1.0875 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 15:17:37] Epoch 3 | Step 29610 | Loss: 0.2059 | LM: 0.1943 | LB: 1.0875 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 15:17:43] Epoch 3 | Step 29620 | Loss: 0.2059 | LM: 0.1943 | LB: 1.0875 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 15:17:49] Epoch 3 | Step 29630 | Loss: 0.2059 | LM: 0.1943 | LB: 1.0875 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 15:17:56] Epoch 3 | Step 29640 | Loss: 0.2059 | LM: 0.1943 | LB: 1.0875 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 15:18:02] Epoch 3 | Step 29650 | Loss: 0.2059 | LM: 0.1942 | LB: 1.0875 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 15:18:09] Epoch 3 | Step 29660 | Loss: 0.2059 | LM: 0.1942 | LB: 1.0876 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 15:18:15] Epoch 3 | Step 29670 | Loss: 0.2059 | LM: 0.1942 | LB: 1.0876 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 15:18:22] Epoch 3 | Step 29680 | Loss: 0.2059 | LM: 0.1941 | LB: 1.0875 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 15:18:28] Epoch 3 | Step 29690 | Loss: 0.2059 | LM: 0.1941 | LB: 1.0875 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 15:18:35] Epoch 3 | Step 29700 | Loss: 0.2059 | LM: 0.1941 | LB: 1.0875 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 15:18:41] Epoch 3 | Step 29710 | Loss: 0.2059 | LM: 0.1942 | LB: 1.0875 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 15:18:47] Epoch 3 | Step 29720 | Loss: 0.2059 | LM: 0.1941 | LB: 1.0875 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 15:18:54] Epoch 3 | Step 29730 | Loss: 0.2059 | LM: 0.1941 | LB: 1.0875 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 15:19:00] Epoch 3 | Step 29740 | Loss: 0.2059 | LM: 0.1941 | LB: 1.0875 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 15:19:07] Epoch 3 | Step 29750 | Loss: 0.2059 | LM: 0.1941 | LB: 1.0875 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 15:19:13] Epoch 3 | Step 29760 | Loss: 0.2059 | LM: 0.1941 | LB: 1.0875 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 15:19:19] Epoch 3 | Step 29770 | Loss: 0.2059 | LM: 0.1940 | LB: 1.0875 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 15:19:26] Epoch 3 | Step 29780 | Loss: 0.2059 | LM: 0.1940 | LB: 1.0875 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 15:19:32] Epoch 3 | Step 29790 | Loss: 0.2059 | LM: 0.1940 | LB: 1.0875 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 15:19:38] Epoch 3 | Step 29800 | Loss: 0.2059 | LM: 0.1940 | LB: 1.0875 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 15:19:45] Epoch 3 | Step 29810 | Loss: 0.2059 | LM: 0.1941 | LB: 1.0875 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 15:19:51] Epoch 3 | Step 29820 | Loss: 0.2059 | LM: 0.1940 | LB: 1.0875 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 15:19:58] Epoch 3 | Step 29830 | Loss: 0.2059 | LM: 0.1941 | LB: 1.0875 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 15:20:04] Epoch 3 | Step 29840 | Loss: 0.2059 | LM: 0.1941 | LB: 1.0875 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 15:20:10] Epoch 3 | Step 29850 | Loss: 0.2058 | LM: 0.1941 | LB: 1.0875 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 15:20:17] Epoch 3 | Step 29860 | Loss: 0.2058 | LM: 0.1940 | LB: 1.0875 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 15:20:23] Epoch 3 | Step 29870 | Loss: 0.2058 | LM: 0.1940 | LB: 1.0875 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 15:20:30] Epoch 3 | Step 29880 | Loss: 0.2058 | LM: 0.1940 | LB: 1.0875 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 15:20:36] Epoch 3 | Step 29890 | Loss: 0.2058 | LM: 0.1942 | LB: 1.0875 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 15:20:42] Epoch 3 | Step 29900 | Loss: 0.2058 | LM: 0.1942 | LB: 1.0875 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 15:20:49] Epoch 3 | Step 29910 | Loss: 0.2058 | LM: 0.1941 | LB: 1.0875 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 15:20:55] Epoch 3 | Step 29920 | Loss: 0.2059 | LM: 0.1941 | LB: 1.0875 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 15:21:01] Epoch 3 | Step 29930 | Loss: 0.2059 | LM: 0.1941 | LB: 1.0875 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 15:21:08] Epoch 3 | Step 29940 | Loss: 0.2059 | LM: 0.1941 | LB: 1.0875 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 15:21:14] Epoch 3 | Step 29950 | Loss: 0.2059 | LM: 0.1942 | LB: 1.0875 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 15:21:21] Epoch 3 | Step 29960 | Loss: 0.2059 | LM: 0.1942 | LB: 1.0875 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 15:21:27] Epoch 3 | Step 29970 | Loss: 0.2059 | LM: 0.1941 | LB: 1.0875 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 15:21:33] Epoch 3 | Step 29980 | Loss: 0.2059 | LM: 0.1941 | LB: 1.0875 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 15:21:40] Epoch 3 | Step 29990 | Loss: 0.2059 | LM: 0.1941 | LB: 1.0875 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 15:21:46] Epoch 3 | Step 30000 | Loss: 0.2059 | LM: 0.1940 | LB: 1.0875 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 15:21:55] Checkpoint saved: outputs/2026-04-17/08-57-56/checkpoints/checkpoint_step_30000.pt [2026-04-17 15:22:11] Validation | Batch 10/784 | Loss: 0.3357 | LM_LOSS: 0.3248 | LB_LOSS: 1.0845 [2026-04-17 15:22:12] Validation | Batch 20/784 | Loss: 0.3471 | LM_LOSS: 0.3363 | LB_LOSS: 1.0847 [2026-04-17 15:22:13] Validation | Batch 30/784 | Loss: 0.3328 | LM_LOSS: 0.3220 | LB_LOSS: 1.0840 [2026-04-17 15:22:15] Validation | Batch 40/784 | Loss: 0.3355 | LM_LOSS: 0.3247 | LB_LOSS: 1.0839 [2026-04-17 15:22:16] Validation | Batch 50/784 | Loss: 0.3330 | LM_LOSS: 0.3222 | LB_LOSS: 1.0833 [2026-04-17 15:22:18] Validation | Batch 60/784 | Loss: 0.3350 | LM_LOSS: 0.3242 | LB_LOSS: 1.0828 [2026-04-17 15:22:19] Validation | Batch 70/784 | Loss: 0.3323 | LM_LOSS: 0.3215 | LB_LOSS: 1.0822 [2026-04-17 15:22:20] Validation | Batch 80/784 | Loss: 0.3287 | LM_LOSS: 0.3179 | LB_LOSS: 1.0817 [2026-04-17 15:22:21] Validation | Batch 90/784 | Loss: 0.3276 | LM_LOSS: 0.3167 | LB_LOSS: 1.0823 [2026-04-17 15:22:23] Validation | Batch 100/784 | Loss: 0.3295 | LM_LOSS: 0.3187 | LB_LOSS: 1.0828 [2026-04-17 15:22:24] Validation | Batch 110/784 | Loss: 0.3240 | LM_LOSS: 0.3132 | LB_LOSS: 1.0829 [2026-04-17 15:22:26] Validation | Batch 120/784 | Loss: 0.3276 | LM_LOSS: 0.3168 | LB_LOSS: 1.0828 [2026-04-17 15:22:27] Validation | Batch 130/784 | Loss: 0.3307 | LM_LOSS: 0.3199 | LB_LOSS: 1.0828 [2026-04-17 15:22:28] Validation | Batch 140/784 | Loss: 0.3301 | LM_LOSS: 0.3192 | LB_LOSS: 1.0826 [2026-04-17 15:22:30] Validation | Batch 150/784 | Loss: 0.3261 | LM_LOSS: 0.3153 | LB_LOSS: 1.0829 [2026-04-17 15:22:31] Validation | Batch 160/784 | Loss: 0.3269 | LM_LOSS: 0.3161 | LB_LOSS: 1.0826 [2026-04-17 15:22:33] Validation | Batch 170/784 | Loss: 0.3270 | LM_LOSS: 0.3162 | LB_LOSS: 1.0823 [2026-04-17 15:22:34] Validation | Batch 180/784 | Loss: 0.3245 | LM_LOSS: 0.3137 | LB_LOSS: 1.0823 [2026-04-17 15:22:35] Validation | Batch 190/784 | Loss: 0.3268 | LM_LOSS: 0.3159 | LB_LOSS: 1.0827 [2026-04-17 15:22:37] Validation | Batch 200/784 | Loss: 0.3272 | LM_LOSS: 0.3164 | LB_LOSS: 1.0828 [2026-04-17 15:22:38] Validation | Batch 210/784 | Loss: 0.3260 | LM_LOSS: 0.3152 | LB_LOSS: 1.0827 [2026-04-17 15:22:39] Validation | Batch 220/784 | Loss: 0.3269 | LM_LOSS: 0.3161 | LB_LOSS: 1.0827 [2026-04-17 15:22:41] Validation | Batch 230/784 | Loss: 0.3275 | LM_LOSS: 0.3167 | LB_LOSS: 1.0827 [2026-04-17 15:22:42] Validation | Batch 240/784 | Loss: 0.3280 | LM_LOSS: 0.3172 | LB_LOSS: 1.0830 [2026-04-17 15:22:44] Validation | Batch 250/784 | Loss: 0.3279 | LM_LOSS: 0.3171 | LB_LOSS: 1.0828 [2026-04-17 15:22:45] Validation | Batch 260/784 | Loss: 0.3282 | LM_LOSS: 0.3174 | LB_LOSS: 1.0831 [2026-04-17 15:22:47] Validation | Batch 270/784 | Loss: 0.3280 | LM_LOSS: 0.3172 | LB_LOSS: 1.0831 [2026-04-17 15:22:48] Validation | Batch 280/784 | Loss: 0.3285 | LM_LOSS: 0.3176 | LB_LOSS: 1.0833 [2026-04-17 15:22:49] Validation | Batch 290/784 | Loss: 0.3296 | LM_LOSS: 0.3188 | LB_LOSS: 1.0834 [2026-04-17 15:22:51] Validation | Batch 300/784 | Loss: 0.3304 | LM_LOSS: 0.3196 | LB_LOSS: 1.0834 [2026-04-17 15:22:52] Validation | Batch 310/784 | Loss: 0.3298 | LM_LOSS: 0.3190 | LB_LOSS: 1.0834 [2026-04-17 15:22:53] Validation | Batch 320/784 | Loss: 0.3314 | LM_LOSS: 0.3206 | LB_LOSS: 1.0834 [2026-04-17 15:22:55] Validation | Batch 330/784 | Loss: 0.3312 | LM_LOSS: 0.3204 | LB_LOSS: 1.0833 [2026-04-17 15:22:56] Validation | Batch 340/784 | Loss: 0.3300 | LM_LOSS: 0.3192 | LB_LOSS: 1.0834 [2026-04-17 15:22:57] Validation | Batch 350/784 | Loss: 0.3303 | LM_LOSS: 0.3194 | LB_LOSS: 1.0836 [2026-04-17 15:22:58] Validation | Batch 360/784 | Loss: 0.3301 | LM_LOSS: 0.3192 | LB_LOSS: 1.0836 [2026-04-17 15:23:00] Validation | Batch 370/784 | Loss: 0.3306 | LM_LOSS: 0.3197 | LB_LOSS: 1.0836 [2026-04-17 15:23:01] Validation | Batch 380/784 | Loss: 0.3304 | LM_LOSS: 0.3196 | LB_LOSS: 1.0836 [2026-04-17 15:23:03] Validation | Batch 390/784 | Loss: 0.3303 | LM_LOSS: 0.3195 | LB_LOSS: 1.0837 [2026-04-17 15:23:04] Validation | Batch 400/784 | Loss: 0.3306 | LM_LOSS: 0.3198 | LB_LOSS: 1.0837 [2026-04-17 15:23:05] Validation | Batch 410/784 | Loss: 0.3310 | LM_LOSS: 0.3201 | LB_LOSS: 1.0837 [2026-04-17 15:23:06] Validation | Batch 420/784 | Loss: 0.3312 | LM_LOSS: 0.3204 | LB_LOSS: 1.0838 [2026-04-17 15:23:07] Validation | Batch 430/784 | Loss: 0.3313 | LM_LOSS: 0.3205 | LB_LOSS: 1.0837 [2026-04-17 15:23:09] Validation | Batch 440/784 | Loss: 0.3310 | LM_LOSS: 0.3202 | LB_LOSS: 1.0837 [2026-04-17 15:23:10] Validation | Batch 450/784 | Loss: 0.3302 | LM_LOSS: 0.3194 | LB_LOSS: 1.0837 [2026-04-17 15:23:11] Validation | Batch 460/784 | Loss: 0.3307 | LM_LOSS: 0.3199 | LB_LOSS: 1.0837 [2026-04-17 15:23:13] Validation | Batch 470/784 | Loss: 0.3299 | LM_LOSS: 0.3191 | LB_LOSS: 1.0837 [2026-04-17 15:23:14] Validation | Batch 480/784 | Loss: 0.3304 | LM_LOSS: 0.3196 | LB_LOSS: 1.0837 [2026-04-17 15:23:16] Validation | Batch 490/784 | Loss: 0.3297 | LM_LOSS: 0.3189 | LB_LOSS: 1.0836 [2026-04-17 15:23:17] Validation | Batch 500/784 | Loss: 0.3301 | LM_LOSS: 0.3193 | LB_LOSS: 1.0835 [2026-04-17 15:23:18] Validation | Batch 510/784 | Loss: 0.3298 | LM_LOSS: 0.3190 | LB_LOSS: 1.0835 [2026-04-17 15:23:20] Validation | Batch 520/784 | Loss: 0.3301 | LM_LOSS: 0.3192 | LB_LOSS: 1.0834 [2026-04-17 15:23:21] Validation | Batch 530/784 | Loss: 0.3309 | LM_LOSS: 0.3201 | LB_LOSS: 1.0834 [2026-04-17 15:23:22] Validation | Batch 540/784 | Loss: 0.3313 | LM_LOSS: 0.3205 | LB_LOSS: 1.0834 [2026-04-17 15:23:24] Validation | Batch 550/784 | Loss: 0.3326 | LM_LOSS: 0.3218 | LB_LOSS: 1.0834 [2026-04-17 15:23:25] Validation | Batch 560/784 | Loss: 0.3327 | LM_LOSS: 0.3219 | LB_LOSS: 1.0834 [2026-04-17 15:23:27] Validation | Batch 570/784 | Loss: 0.3322 | LM_LOSS: 0.3214 | LB_LOSS: 1.0833 [2026-04-17 15:23:28] Validation | Batch 580/784 | Loss: 0.3317 | LM_LOSS: 0.3209 | LB_LOSS: 1.0834 [2026-04-17 15:23:29] Validation | Batch 590/784 | Loss: 0.3320 | LM_LOSS: 0.3211 | LB_LOSS: 1.0833 [2026-04-17 15:23:31] Validation | Batch 600/784 | Loss: 0.3318 | LM_LOSS: 0.3210 | LB_LOSS: 1.0833 [2026-04-17 15:23:32] Validation | Batch 610/784 | Loss: 0.3320 | LM_LOSS: 0.3212 | LB_LOSS: 1.0833 [2026-04-17 15:23:33] Validation | Batch 620/784 | Loss: 0.3318 | LM_LOSS: 0.3210 | LB_LOSS: 1.0833 [2026-04-17 15:23:35] Validation | Batch 630/784 | Loss: 0.3326 | LM_LOSS: 0.3218 | LB_LOSS: 1.0833 [2026-04-17 15:23:37] Validation | Batch 640/784 | Loss: 0.3327 | LM_LOSS: 0.3218 | LB_LOSS: 1.0833 [2026-04-17 15:23:38] Validation | Batch 650/784 | Loss: 0.3325 | LM_LOSS: 0.3217 | LB_LOSS: 1.0833 [2026-04-17 15:23:39] Validation | Batch 660/784 | Loss: 0.3329 | LM_LOSS: 0.3221 | LB_LOSS: 1.0833 [2026-04-17 15:23:41] Validation | Batch 670/784 | Loss: 0.3334 | LM_LOSS: 0.3225 | LB_LOSS: 1.0834 [2026-04-17 15:23:42] Validation | Batch 680/784 | Loss: 0.3331 | LM_LOSS: 0.3222 | LB_LOSS: 1.0834 [2026-04-17 15:23:44] Validation | Batch 690/784 | Loss: 0.3332 | LM_LOSS: 0.3224 | LB_LOSS: 1.0833 [2026-04-17 15:23:45] Validation | Batch 700/784 | Loss: 0.3333 | LM_LOSS: 0.3224 | LB_LOSS: 1.0833 [2026-04-17 15:23:46] Validation | Batch 710/784 | Loss: 0.3330 | LM_LOSS: 0.3222 | LB_LOSS: 1.0832 [2026-04-17 15:23:48] Validation | Batch 720/784 | Loss: 0.3327 | LM_LOSS: 0.3219 | LB_LOSS: 1.0831 [2026-04-17 15:23:49] Validation | Batch 730/784 | Loss: 0.3322 | LM_LOSS: 0.3214 | LB_LOSS: 1.0831 [2026-04-17 15:23:50] Validation | Batch 740/784 | Loss: 0.3323 | LM_LOSS: 0.3215 | LB_LOSS: 1.0832 [2026-04-17 15:23:51] Validation | Batch 750/784 | Loss: 0.3316 | LM_LOSS: 0.3208 | LB_LOSS: 1.0832 [2026-04-17 15:23:53] Validation | Batch 760/784 | Loss: 0.3317 | LM_LOSS: 0.3209 | LB_LOSS: 1.0832 [2026-04-17 15:23:54] Validation | Batch 770/784 | Loss: 0.3319 | LM_LOSS: 0.3211 | LB_LOSS: 1.0832 [2026-04-17 15:23:56] Validation | Batch 780/784 | Loss: 0.3323 | LM_LOSS: 0.3215 | LB_LOSS: 1.0832 [2026-04-17 15:23:56] Validation | Batch 784/784 | Loss: 0.3325 | LM_LOSS: 0.3217 | LB_LOSS: 1.0832 [2026-04-17 15:23:59] Validation | Loss: 0.3325 | LM_LOSS: 0.3217 | LB_LOSS: 1.0832 | PPL: 1.38 | Time: 106.73s [2026-04-17 15:24:05] Epoch 3 | Step 30010 | Loss: 0.2059 | LM: 0.1940 | LB: 1.0875 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 15:24:12] Epoch 3 | Step 30020 | Loss: 0.2059 | LM: 0.1939 | LB: 1.0875 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 15:24:18] Epoch 3 | Step 30030 | Loss: 0.2059 | LM: 0.1940 | LB: 1.0875 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 15:24:25] Epoch 3 | Step 30040 | Loss: 0.2059 | LM: 0.1941 | LB: 1.0875 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 15:24:31] Epoch 3 | Step 30050 | Loss: 0.2059 | LM: 0.1941 | LB: 1.0875 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 15:24:37] Epoch 3 | Step 30060 | Loss: 0.2059 | LM: 0.1941 | LB: 1.0875 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 15:24:43] Epoch 3 | Step 30070 | Loss: 0.2059 | LM: 0.1941 | LB: 1.0875 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 15:24:50] Epoch 3 | Step 30080 | Loss: 0.2059 | LM: 0.1941 | LB: 1.0875 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 15:24:56] Epoch 3 | Step 30090 | Loss: 0.2058 | LM: 0.1941 | LB: 1.0875 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 15:25:03] Epoch 3 | Step 30100 | Loss: 0.2058 | LM: 0.1941 | LB: 1.0875 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 15:25:09] Epoch 3 | Step 30110 | Loss: 0.2059 | LM: 0.1941 | LB: 1.0875 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 15:25:16] Epoch 3 | Step 30120 | Loss: 0.2058 | LM: 0.1941 | LB: 1.0875 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 15:25:22] Epoch 3 | Step 30130 | Loss: 0.2059 | LM: 0.1941 | LB: 1.0875 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 15:25:28] Epoch 3 | Step 30140 | Loss: 0.2058 | LM: 0.1941 | LB: 1.0874 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 15:25:34] Epoch 3 | Step 30150 | Loss: 0.2058 | LM: 0.1940 | LB: 1.0875 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 15:25:41] Epoch 3 | Step 30160 | Loss: 0.2057 | LM: 0.1940 | LB: 1.0875 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 15:25:47] Epoch 3 | Step 30170 | Loss: 0.2058 | LM: 0.1940 | LB: 1.0875 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 15:25:54] Epoch 3 | Step 30180 | Loss: 0.2057 | LM: 0.1939 | LB: 1.0875 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 15:26:00] Epoch 3 | Step 30190 | Loss: 0.2057 | LM: 0.1938 | LB: 1.0875 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 15:26:07] Epoch 3 | Step 30200 | Loss: 0.2057 | LM: 0.1939 | LB: 1.0875 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 15:26:13] Epoch 3 | Step 30210 | Loss: 0.2057 | LM: 0.1939 | LB: 1.0875 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 15:26:20] Epoch 3 | Step 30220 | Loss: 0.2057 | LM: 0.1939 | LB: 1.0875 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 15:26:27] Epoch 3 | Step 30230 | Loss: 0.2058 | LM: 0.1940 | LB: 1.0875 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 15:26:33] Epoch 3 | Step 30240 | Loss: 0.2058 | LM: 0.1941 | LB: 1.0875 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 15:26:40] Epoch 3 | Step 30250 | Loss: 0.2058 | LM: 0.1941 | LB: 1.0875 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 15:26:47] Epoch 3 | Step 30260 | Loss: 0.2058 | LM: 0.1942 | LB: 1.0875 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 15:26:53] Epoch 3 | Step 30270 | Loss: 0.2058 | LM: 0.1942 | LB: 1.0875 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 15:26:59] Epoch 3 | Step 30280 | Loss: 0.2058 | LM: 0.1943 | LB: 1.0875 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 15:27:06] Epoch 3 | Step 30290 | Loss: 0.2058 | LM: 0.1943 | LB: 1.0875 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 15:27:12] Epoch 3 | Step 30300 | Loss: 0.2058 | LM: 0.1943 | LB: 1.0875 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 15:27:19] Epoch 3 | Step 30310 | Loss: 0.2058 | LM: 0.1943 | LB: 1.0875 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 15:27:25] Epoch 3 | Step 30320 | Loss: 0.2058 | LM: 0.1943 | LB: 1.0875 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 15:27:32] Epoch 3 | Step 30330 | Loss: 0.2058 | LM: 0.1943 | LB: 1.0875 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 15:27:38] Epoch 3 | Step 30340 | Loss: 0.2058 | LM: 0.1944 | LB: 1.0875 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 15:27:45] Epoch 3 | Step 30350 | Loss: 0.2058 | LM: 0.1943 | LB: 1.0875 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 15:27:52] Epoch 3 | Step 30360 | Loss: 0.2058 | LM: 0.1943 | LB: 1.0875 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 15:27:58] Epoch 3 | Step 30370 | Loss: 0.2058 | LM: 0.1942 | LB: 1.0875 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 15:28:04] Epoch 3 | Step 30380 | Loss: 0.2058 | LM: 0.1942 | LB: 1.0875 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 15:28:11] Epoch 3 | Step 30390 | Loss: 0.2058 | LM: 0.1942 | LB: 1.0875 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 15:28:17] Epoch 3 | Step 30400 | Loss: 0.2058 | LM: 0.1943 | LB: 1.0875 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 15:28:24] Epoch 3 | Step 30410 | Loss: 0.2058 | LM: 0.1942 | LB: 1.0875 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 15:28:30] Epoch 3 | Step 30420 | Loss: 0.2058 | LM: 0.1942 | LB: 1.0875 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 15:28:37] Epoch 3 | Step 30430 | Loss: 0.2058 | LM: 0.1942 | LB: 1.0875 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 15:28:43] Epoch 3 | Step 30440 | Loss: 0.2058 | LM: 0.1942 | LB: 1.0875 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 15:28:49] Epoch 3 | Step 30450 | Loss: 0.2058 | LM: 0.1942 | LB: 1.0875 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 15:28:56] Epoch 3 | Step 30460 | Loss: 0.2058 | LM: 0.1942 | LB: 1.0875 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 15:29:02] Epoch 3 | Step 30470 | Loss: 0.2058 | LM: 0.1941 | LB: 1.0875 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 15:29:08] Epoch 3 | Step 30480 | Loss: 0.2058 | LM: 0.1942 | LB: 1.0875 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 15:29:15] Epoch 3 | Step 30490 | Loss: 0.2059 | LM: 0.1942 | LB: 1.0875 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 15:29:21] Epoch 3 | Step 30500 | Loss: 0.2059 | LM: 0.1942 | LB: 1.0875 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 15:29:27] Epoch 3 | Step 30510 | Loss: 0.2059 | LM: 0.1941 | LB: 1.0875 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 15:29:33] Epoch 3 | Step 30520 | Loss: 0.2059 | LM: 0.1941 | LB: 1.0875 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 15:29:40] Epoch 3 | Step 30530 | Loss: 0.2059 | LM: 0.1941 | LB: 1.0875 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 15:29:46] Epoch 3 | Step 30540 | Loss: 0.2058 | LM: 0.1940 | LB: 1.0875 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 15:29:52] Epoch 3 | Step 30550 | Loss: 0.2058 | LM: 0.1940 | LB: 1.0875 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 15:29:59] Epoch 3 | Step 30560 | Loss: 0.2058 | LM: 0.1940 | LB: 1.0875 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 15:30:06] Epoch 3 | Step 30570 | Loss: 0.2059 | LM: 0.1940 | LB: 1.0875 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 15:30:12] Epoch 3 | Step 30580 | Loss: 0.2059 | LM: 0.1940 | LB: 1.0875 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 15:30:18] Epoch 3 | Step 30590 | Loss: 0.2058 | LM: 0.1940 | LB: 1.0875 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 15:30:25] Epoch 3 | Step 30600 | Loss: 0.2058 | LM: 0.1940 | LB: 1.0875 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 15:30:31] Epoch 3 | Step 30610 | Loss: 0.2058 | LM: 0.1940 | LB: 1.0875 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 15:30:37] Epoch 3 | Step 30620 | Loss: 0.2059 | LM: 0.1940 | LB: 1.0875 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 15:30:44] Epoch 3 | Step 30630 | Loss: 0.2059 | LM: 0.1940 | LB: 1.0875 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 15:30:50] Epoch 3 | Step 30640 | Loss: 0.2059 | LM: 0.1941 | LB: 1.0875 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 15:30:57] Epoch 3 | Step 30650 | Loss: 0.2059 | LM: 0.1941 | LB: 1.0875 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 15:31:03] Epoch 3 | Step 30660 | Loss: 0.2060 | LM: 0.1941 | LB: 1.0875 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 15:31:10] Epoch 3 | Step 30670 | Loss: 0.2060 | LM: 0.1942 | LB: 1.0875 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 15:31:15] Epoch 3 | Step 30680 | Loss: 0.2060 | LM: 0.1941 | LB: 1.0875 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 15:31:22] Epoch 3 | Step 30690 | Loss: 0.2060 | LM: 0.1941 | LB: 1.0875 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 15:31:28] Epoch 3 | Step 30700 | Loss: 0.2060 | LM: 0.1941 | LB: 1.0875 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 15:31:35] Epoch 3 | Step 30710 | Loss: 0.2060 | LM: 0.1942 | LB: 1.0875 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 15:31:41] Epoch 3 | Step 30720 | Loss: 0.2060 | LM: 0.1942 | LB: 1.0875 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 15:31:48] Epoch 3 | Step 30730 | Loss: 0.2060 | LM: 0.1942 | LB: 1.0875 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 15:31:55] Epoch 3 | Step 30740 | Loss: 0.2059 | LM: 0.1942 | LB: 1.0875 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 15:32:01] Epoch 3 | Step 30750 | Loss: 0.2060 | LM: 0.1943 | LB: 1.0875 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 15:32:07] Epoch 3 | Step 30760 | Loss: 0.2059 | LM: 0.1942 | LB: 1.0875 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 15:32:14] Epoch 3 | Step 30770 | Loss: 0.2059 | LM: 0.1942 | LB: 1.0875 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 15:32:20] Epoch 3 | Step 30780 | Loss: 0.2059 | LM: 0.1941 | LB: 1.0875 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 15:32:27] Epoch 3 | Step 30790 | Loss: 0.2059 | LM: 0.1942 | LB: 1.0875 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 15:32:33] Epoch 3 | Step 30800 | Loss: 0.2059 | LM: 0.1941 | LB: 1.0875 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 15:32:40] Epoch 3 | Step 30810 | Loss: 0.2059 | LM: 0.1940 | LB: 1.0875 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 15:32:46] Epoch 3 | Step 30820 | Loss: 0.2059 | LM: 0.1940 | LB: 1.0875 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 15:32:52] Epoch 3 | Step 30830 | Loss: 0.2059 | LM: 0.1939 | LB: 1.0875 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 15:32:59] Epoch 3 | Step 30840 | Loss: 0.2059 | LM: 0.1940 | LB: 1.0875 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 15:33:05] Epoch 3 | Step 30850 | Loss: 0.2059 | LM: 0.1940 | LB: 1.0875 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 15:33:12] Epoch 3 | Step 30860 | Loss: 0.2059 | LM: 0.1939 | LB: 1.0875 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 15:33:18] Epoch 3 | Step 30870 | Loss: 0.2059 | LM: 0.1940 | LB: 1.0875 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 15:33:25] Epoch 3 | Step 30880 | Loss: 0.2059 | LM: 0.1939 | LB: 1.0875 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 15:33:31] Epoch 3 | Step 30890 | Loss: 0.2059 | LM: 0.1939 | LB: 1.0875 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 15:33:38] Epoch 3 | Step 30900 | Loss: 0.2059 | LM: 0.1939 | LB: 1.0875 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 15:33:44] Epoch 3 | Step 30910 | Loss: 0.2059 | LM: 0.1940 | LB: 1.0875 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 15:33:51] Epoch 3 | Step 30920 | Loss: 0.2059 | LM: 0.1939 | LB: 1.0875 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 15:33:57] Epoch 3 | Step 30930 | Loss: 0.2059 | LM: 0.1939 | LB: 1.0875 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 15:34:04] Epoch 3 | Step 30940 | Loss: 0.2059 | LM: 0.1940 | LB: 1.0875 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 15:34:10] Epoch 3 | Step 30950 | Loss: 0.2059 | LM: 0.1940 | LB: 1.0875 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 15:34:16] Epoch 3 | Step 30960 | Loss: 0.2059 | LM: 0.1941 | LB: 1.0875 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 15:34:23] Epoch 3 | Step 30970 | Loss: 0.2059 | LM: 0.1941 | LB: 1.0875 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 15:34:29] Epoch 3 | Step 30980 | Loss: 0.2060 | LM: 0.1942 | LB: 1.0875 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 15:34:35] Epoch 3 | Step 30990 | Loss: 0.2060 | LM: 0.1941 | LB: 1.0875 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 15:34:42] Epoch 3 | Step 31000 | Loss: 0.2060 | LM: 0.1941 | LB: 1.0875 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 15:34:43] Validation | Batch 10/784 | Loss: 0.3358 | LM_LOSS: 0.3250 | LB_LOSS: 1.0845 [2026-04-17 15:34:44] Validation | Batch 20/784 | Loss: 0.3475 | LM_LOSS: 0.3367 | LB_LOSS: 1.0847 [2026-04-17 15:34:46] Validation | Batch 30/784 | Loss: 0.3332 | LM_LOSS: 0.3223 | LB_LOSS: 1.0839 [2026-04-17 15:34:47] Validation | Batch 40/784 | Loss: 0.3360 | LM_LOSS: 0.3251 | LB_LOSS: 1.0839 [2026-04-17 15:34:49] Validation | Batch 50/784 | Loss: 0.3335 | LM_LOSS: 0.3227 | LB_LOSS: 1.0832 [2026-04-17 15:34:50] Validation | Batch 60/784 | Loss: 0.3354 | LM_LOSS: 0.3246 | LB_LOSS: 1.0828 [2026-04-17 15:34:51] Validation | Batch 70/784 | Loss: 0.3327 | LM_LOSS: 0.3219 | LB_LOSS: 1.0822 [2026-04-17 15:34:52] Validation | Batch 80/784 | Loss: 0.3290 | LM_LOSS: 0.3182 | LB_LOSS: 1.0817 [2026-04-17 15:34:54] Validation | Batch 90/784 | Loss: 0.3279 | LM_LOSS: 0.3170 | LB_LOSS: 1.0823 [2026-04-17 15:34:55] Validation | Batch 100/784 | Loss: 0.3298 | LM_LOSS: 0.3190 | LB_LOSS: 1.0827 [2026-04-17 15:34:57] Validation | Batch 110/784 | Loss: 0.3243 | LM_LOSS: 0.3135 | LB_LOSS: 1.0829 [2026-04-17 15:34:58] Validation | Batch 120/784 | Loss: 0.3279 | LM_LOSS: 0.3171 | LB_LOSS: 1.0828 [2026-04-17 15:34:59] Validation | Batch 130/784 | Loss: 0.3310 | LM_LOSS: 0.3201 | LB_LOSS: 1.0827 [2026-04-17 15:35:01] Validation | Batch 140/784 | Loss: 0.3304 | LM_LOSS: 0.3195 | LB_LOSS: 1.0825 [2026-04-17 15:35:02] Validation | Batch 150/784 | Loss: 0.3264 | LM_LOSS: 0.3155 | LB_LOSS: 1.0829 [2026-04-17 15:35:04] Validation | Batch 160/784 | Loss: 0.3271 | LM_LOSS: 0.3163 | LB_LOSS: 1.0825 [2026-04-17 15:35:05] Validation | Batch 170/784 | Loss: 0.3273 | LM_LOSS: 0.3164 | LB_LOSS: 1.0823 [2026-04-17 15:35:07] Validation | Batch 180/784 | Loss: 0.3248 | LM_LOSS: 0.3140 | LB_LOSS: 1.0823 [2026-04-17 15:35:08] Validation | Batch 190/784 | Loss: 0.3270 | LM_LOSS: 0.3162 | LB_LOSS: 1.0827 [2026-04-17 15:35:09] Validation | Batch 200/784 | Loss: 0.3275 | LM_LOSS: 0.3167 | LB_LOSS: 1.0828 [2026-04-17 15:35:11] Validation | Batch 210/784 | Loss: 0.3263 | LM_LOSS: 0.3155 | LB_LOSS: 1.0827 [2026-04-17 15:35:12] Validation | Batch 220/784 | Loss: 0.3273 | LM_LOSS: 0.3164 | LB_LOSS: 1.0827 [2026-04-17 15:35:13] Validation | Batch 230/784 | Loss: 0.3279 | LM_LOSS: 0.3170 | LB_LOSS: 1.0826 [2026-04-17 15:35:15] Validation | Batch 240/784 | Loss: 0.3283 | LM_LOSS: 0.3175 | LB_LOSS: 1.0830 [2026-04-17 15:35:16] Validation | Batch 250/784 | Loss: 0.3282 | LM_LOSS: 0.3174 | LB_LOSS: 1.0828 [2026-04-17 15:35:18] Validation | Batch 260/784 | Loss: 0.3285 | LM_LOSS: 0.3177 | LB_LOSS: 1.0830 [2026-04-17 15:35:19] Validation | Batch 270/784 | Loss: 0.3284 | LM_LOSS: 0.3175 | LB_LOSS: 1.0831 [2026-04-17 15:35:21] Validation | Batch 280/784 | Loss: 0.3288 | LM_LOSS: 0.3180 | LB_LOSS: 1.0832 [2026-04-17 15:35:22] Validation | Batch 290/784 | Loss: 0.3299 | LM_LOSS: 0.3191 | LB_LOSS: 1.0834 [2026-04-17 15:35:23] Validation | Batch 300/784 | Loss: 0.3307 | LM_LOSS: 0.3199 | LB_LOSS: 1.0834 [2026-04-17 15:35:24] Validation | Batch 310/784 | Loss: 0.3301 | LM_LOSS: 0.3193 | LB_LOSS: 1.0833 [2026-04-17 15:35:26] Validation | Batch 320/784 | Loss: 0.3318 | LM_LOSS: 0.3209 | LB_LOSS: 1.0833 [2026-04-17 15:35:28] Validation | Batch 330/784 | Loss: 0.3315 | LM_LOSS: 0.3207 | LB_LOSS: 1.0833 [2026-04-17 15:35:29] Validation | Batch 340/784 | Loss: 0.3303 | LM_LOSS: 0.3195 | LB_LOSS: 1.0834 [2026-04-17 15:35:30] Validation | Batch 350/784 | Loss: 0.3306 | LM_LOSS: 0.3197 | LB_LOSS: 1.0836 [2026-04-17 15:35:31] Validation | Batch 360/784 | Loss: 0.3303 | LM_LOSS: 0.3195 | LB_LOSS: 1.0836 [2026-04-17 15:35:32] Validation | Batch 370/784 | Loss: 0.3309 | LM_LOSS: 0.3200 | LB_LOSS: 1.0835 [2026-04-17 15:35:34] Validation | Batch 380/784 | Loss: 0.3307 | LM_LOSS: 0.3198 | LB_LOSS: 1.0836 [2026-04-17 15:35:35] Validation | Batch 390/784 | Loss: 0.3306 | LM_LOSS: 0.3198 | LB_LOSS: 1.0837 [2026-04-17 15:35:36] Validation | Batch 400/784 | Loss: 0.3309 | LM_LOSS: 0.3201 | LB_LOSS: 1.0836 [2026-04-17 15:35:37] Validation | Batch 410/784 | Loss: 0.3312 | LM_LOSS: 0.3204 | LB_LOSS: 1.0837 [2026-04-17 15:35:38] Validation | Batch 420/784 | Loss: 0.3315 | LM_LOSS: 0.3207 | LB_LOSS: 1.0837 [2026-04-17 15:35:40] Validation | Batch 430/784 | Loss: 0.3316 | LM_LOSS: 0.3208 | LB_LOSS: 1.0836 [2026-04-17 15:35:41] Validation | Batch 440/784 | Loss: 0.3313 | LM_LOSS: 0.3204 | LB_LOSS: 1.0837 [2026-04-17 15:35:42] Validation | Batch 450/784 | Loss: 0.3305 | LM_LOSS: 0.3197 | LB_LOSS: 1.0836 [2026-04-17 15:35:44] Validation | Batch 460/784 | Loss: 0.3310 | LM_LOSS: 0.3202 | LB_LOSS: 1.0837 [2026-04-17 15:35:45] Validation | Batch 470/784 | Loss: 0.3302 | LM_LOSS: 0.3194 | LB_LOSS: 1.0837 [2026-04-17 15:35:47] Validation | Batch 480/784 | Loss: 0.3307 | LM_LOSS: 0.3199 | LB_LOSS: 1.0836 [2026-04-17 15:35:48] Validation | Batch 490/784 | Loss: 0.3300 | LM_LOSS: 0.3192 | LB_LOSS: 1.0836 [2026-04-17 15:35:49] Validation | Batch 500/784 | Loss: 0.3304 | LM_LOSS: 0.3196 | LB_LOSS: 1.0835 [2026-04-17 15:35:51] Validation | Batch 510/784 | Loss: 0.3301 | LM_LOSS: 0.3193 | LB_LOSS: 1.0835 [2026-04-17 15:35:52] Validation | Batch 520/784 | Loss: 0.3304 | LM_LOSS: 0.3195 | LB_LOSS: 1.0834 [2026-04-17 15:35:53] Validation | Batch 530/784 | Loss: 0.3312 | LM_LOSS: 0.3204 | LB_LOSS: 1.0834 [2026-04-17 15:35:55] Validation | Batch 540/784 | Loss: 0.3316 | LM_LOSS: 0.3208 | LB_LOSS: 1.0834 [2026-04-17 15:35:56] Validation | Batch 550/784 | Loss: 0.3329 | LM_LOSS: 0.3221 | LB_LOSS: 1.0833 [2026-04-17 15:35:58] Validation | Batch 560/784 | Loss: 0.3331 | LM_LOSS: 0.3222 | LB_LOSS: 1.0834 [2026-04-17 15:35:59] Validation | Batch 570/784 | Loss: 0.3326 | LM_LOSS: 0.3217 | LB_LOSS: 1.0833 [2026-04-17 15:36:00] Validation | Batch 580/784 | Loss: 0.3320 | LM_LOSS: 0.3212 | LB_LOSS: 1.0833 [2026-04-17 15:36:02] Validation | Batch 590/784 | Loss: 0.3323 | LM_LOSS: 0.3214 | LB_LOSS: 1.0833 [2026-04-17 15:36:03] Validation | Batch 600/784 | Loss: 0.3321 | LM_LOSS: 0.3213 | LB_LOSS: 1.0832 [2026-04-17 15:36:05] Validation | Batch 610/784 | Loss: 0.3323 | LM_LOSS: 0.3215 | LB_LOSS: 1.0832 [2026-04-17 15:36:06] Validation | Batch 620/784 | Loss: 0.3321 | LM_LOSS: 0.3213 | LB_LOSS: 1.0832 [2026-04-17 15:36:07] Validation | Batch 630/784 | Loss: 0.3329 | LM_LOSS: 0.3221 | LB_LOSS: 1.0833 [2026-04-17 15:36:09] Validation | Batch 640/784 | Loss: 0.3330 | LM_LOSS: 0.3222 | LB_LOSS: 1.0832 [2026-04-17 15:36:11] Validation | Batch 650/784 | Loss: 0.3328 | LM_LOSS: 0.3220 | LB_LOSS: 1.0833 [2026-04-17 15:36:12] Validation | Batch 660/784 | Loss: 0.3332 | LM_LOSS: 0.3224 | LB_LOSS: 1.0833 [2026-04-17 15:36:14] Validation | Batch 670/784 | Loss: 0.3337 | LM_LOSS: 0.3229 | LB_LOSS: 1.0833 [2026-04-17 15:36:15] Validation | Batch 680/784 | Loss: 0.3334 | LM_LOSS: 0.3225 | LB_LOSS: 1.0833 [2026-04-17 15:36:16] Validation | Batch 690/784 | Loss: 0.3335 | LM_LOSS: 0.3227 | LB_LOSS: 1.0833 [2026-04-17 15:36:18] Validation | Batch 700/784 | Loss: 0.3336 | LM_LOSS: 0.3227 | LB_LOSS: 1.0832 [2026-04-17 15:36:19] Validation | Batch 710/784 | Loss: 0.3333 | LM_LOSS: 0.3225 | LB_LOSS: 1.0832 [2026-04-17 15:36:21] Validation | Batch 720/784 | Loss: 0.3330 | LM_LOSS: 0.3222 | LB_LOSS: 1.0831 [2026-04-17 15:36:22] Validation | Batch 730/784 | Loss: 0.3325 | LM_LOSS: 0.3217 | LB_LOSS: 1.0831 [2026-04-17 15:36:23] Validation | Batch 740/784 | Loss: 0.3326 | LM_LOSS: 0.3218 | LB_LOSS: 1.0832 [2026-04-17 15:36:24] Validation | Batch 750/784 | Loss: 0.3319 | LM_LOSS: 0.3211 | LB_LOSS: 1.0831 [2026-04-17 15:36:26] Validation | Batch 760/784 | Loss: 0.3320 | LM_LOSS: 0.3212 | LB_LOSS: 1.0831 [2026-04-17 15:36:27] Validation | Batch 770/784 | Loss: 0.3322 | LM_LOSS: 0.3214 | LB_LOSS: 1.0832 [2026-04-17 15:36:28] Validation | Batch 780/784 | Loss: 0.3326 | LM_LOSS: 0.3217 | LB_LOSS: 1.0832 [2026-04-17 15:36:29] Validation | Batch 784/784 | Loss: 0.3328 | LM_LOSS: 0.3219 | LB_LOSS: 1.0832 [2026-04-17 15:36:31] Validation | Loss: 0.3328 | LM_LOSS: 0.3219 | LB_LOSS: 1.0832 | PPL: 1.38 | Time: 107.17s [2026-04-17 15:36:37] Epoch 3 | Step 31010 | Loss: 0.2060 | LM: 0.1941 | LB: 1.0875 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 15:36:44] Epoch 3 | Step 31020 | Loss: 0.2060 | LM: 0.1941 | LB: 1.0875 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 15:36:50] Epoch 3 | Step 31030 | Loss: 0.2059 | LM: 0.1941 | LB: 1.0875 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 15:36:56] Epoch 3 | Step 31040 | Loss: 0.2059 | LM: 0.1941 | LB: 1.0875 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 15:37:03] Epoch 3 | Step 31050 | Loss: 0.2059 | LM: 0.1940 | LB: 1.0875 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 15:37:09] Epoch 3 | Step 31060 | Loss: 0.2059 | LM: 0.1940 | LB: 1.0875 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 15:37:16] Epoch 3 | Step 31070 | Loss: 0.2059 | LM: 0.1940 | LB: 1.0875 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 15:37:22] Epoch 3 | Step 31080 | Loss: 0.2059 | LM: 0.1940 | LB: 1.0875 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 15:37:28] Epoch 3 | Step 31090 | Loss: 0.2059 | LM: 0.1940 | LB: 1.0875 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 15:37:35] Epoch 3 | Step 31100 | Loss: 0.2059 | LM: 0.1941 | LB: 1.0875 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 15:37:41] Epoch 3 | Step 31110 | Loss: 0.2059 | LM: 0.1941 | LB: 1.0875 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 15:37:47] Epoch 3 | Step 31120 | Loss: 0.2059 | LM: 0.1941 | LB: 1.0875 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 15:37:53] Epoch 3 | Step 31130 | Loss: 0.2059 | LM: 0.1941 | LB: 1.0875 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 15:38:00] Epoch 3 | Step 31140 | Loss: 0.2059 | LM: 0.1941 | LB: 1.0875 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 15:38:06] Epoch 3 | Step 31150 | Loss: 0.2059 | LM: 0.1941 | LB: 1.0875 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 15:38:13] Epoch 3 | Step 31160 | Loss: 0.2059 | LM: 0.1941 | LB: 1.0875 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 15:38:19] Epoch 3 | Step 31170 | Loss: 0.2059 | LM: 0.1941 | LB: 1.0875 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 15:38:25] Epoch 3 | Step 31180 | Loss: 0.2059 | LM: 0.1940 | LB: 1.0875 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 15:38:32] Epoch 3 | Step 31190 | Loss: 0.2059 | LM: 0.1940 | LB: 1.0875 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 15:38:38] Epoch 3 | Step 31200 | Loss: 0.2059 | LM: 0.1940 | LB: 1.0875 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 15:38:44] Epoch 3 | Step 31210 | Loss: 0.2059 | LM: 0.1940 | LB: 1.0875 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 15:38:50] Epoch 3 | Step 31220 | Loss: 0.2059 | LM: 0.1939 | LB: 1.0875 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 15:38:56] Epoch 3 | Step 31230 | Loss: 0.2059 | LM: 0.1939 | LB: 1.0875 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 15:39:03] Epoch 3 | Step 31240 | Loss: 0.2059 | LM: 0.1939 | LB: 1.0875 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 15:39:09] Epoch 3 | Step 31250 | Loss: 0.2059 | LM: 0.1939 | LB: 1.0875 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 15:39:15] Epoch 3 | Step 31260 | Loss: 0.2059 | LM: 0.1939 | LB: 1.0875 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 15:39:22] Epoch 3 | Step 31270 | Loss: 0.2059 | LM: 0.1940 | LB: 1.0875 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 15:39:28] Epoch 3 | Step 31280 | Loss: 0.2059 | LM: 0.1940 | LB: 1.0875 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 15:39:34] Epoch 3 | Step 31290 | Loss: 0.2059 | LM: 0.1940 | LB: 1.0875 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 15:39:41] Epoch 3 | Step 31300 | Loss: 0.2058 | LM: 0.1940 | LB: 1.0875 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 15:39:47] Epoch 3 | Step 31310 | Loss: 0.2058 | LM: 0.1940 | LB: 1.0875 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 15:39:53] Epoch 3 | Step 31320 | Loss: 0.2059 | LM: 0.1941 | LB: 1.0875 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 15:39:59] Epoch 3 | Step 31330 | Loss: 0.2059 | LM: 0.1941 | LB: 1.0875 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 15:40:05] Epoch 3 | Step 31340 | Loss: 0.2059 | LM: 0.1941 | LB: 1.0875 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 15:40:12] Epoch 3 | Step 31350 | Loss: 0.2059 | LM: 0.1942 | LB: 1.0875 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 15:40:18] Epoch 3 | Step 31360 | Loss: 0.2059 | LM: 0.1941 | LB: 1.0875 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 15:40:25] Epoch 3 | Step 31370 | Loss: 0.2059 | LM: 0.1941 | LB: 1.0875 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 15:40:31] Epoch 3 | Step 31380 | Loss: 0.2058 | LM: 0.1940 | LB: 1.0875 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 15:40:37] Epoch 3 | Step 31390 | Loss: 0.2059 | LM: 0.1940 | LB: 1.0875 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 15:40:44] Epoch 3 | Step 31400 | Loss: 0.2059 | LM: 0.1941 | LB: 1.0875 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 15:40:50] Epoch 3 | Step 31410 | Loss: 0.2058 | LM: 0.1940 | LB: 1.0875 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 15:40:56] Epoch 3 | Step 31420 | Loss: 0.2058 | LM: 0.1941 | LB: 1.0875 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 15:41:02] Epoch 3 | Step 31430 | Loss: 0.2058 | LM: 0.1941 | LB: 1.0875 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 15:41:08] Epoch 3 | Step 31440 | Loss: 0.2058 | LM: 0.1941 | LB: 1.0875 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 15:41:15] Epoch 3 | Step 31450 | Loss: 0.2058 | LM: 0.1941 | LB: 1.0875 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 15:41:21] Epoch 3 | Step 31460 | Loss: 0.2058 | LM: 0.1941 | LB: 1.0875 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 15:41:28] Epoch 3 | Step 31470 | Loss: 0.2057 | LM: 0.1941 | LB: 1.0875 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 15:41:34] Epoch 3 | Step 31480 | Loss: 0.2057 | LM: 0.1941 | LB: 1.0875 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 15:41:41] Epoch 3 | Step 31490 | Loss: 0.2057 | LM: 0.1941 | LB: 1.0875 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 15:41:47] Epoch 3 | Step 31500 | Loss: 0.2057 | LM: 0.1941 | LB: 1.0875 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 15:41:53] Epoch 3 | Step 31510 | Loss: 0.2057 | LM: 0.1941 | LB: 1.0875 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 15:41:59] Epoch 3 | Step 31520 | Loss: 0.2057 | LM: 0.1941 | LB: 1.0875 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 15:42:05] Epoch 3 | Step 31530 | Loss: 0.2057 | LM: 0.1941 | LB: 1.0875 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 15:42:12] Epoch 3 | Step 31540 | Loss: 0.2057 | LM: 0.1941 | LB: 1.0875 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 15:42:18] Epoch 3 | Step 31550 | Loss: 0.2057 | LM: 0.1941 | LB: 1.0875 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 15:42:24] Epoch 3 | Step 31560 | Loss: 0.2057 | LM: 0.1941 | LB: 1.0875 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 15:42:30] Epoch 3 | Step 31570 | Loss: 0.2057 | LM: 0.1942 | LB: 1.0875 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 15:42:36] Epoch 3 | Step 31580 | Loss: 0.2057 | LM: 0.1941 | LB: 1.0875 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 15:42:42] Epoch 3 | Step 31590 | Loss: 0.2057 | LM: 0.1941 | LB: 1.0875 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 15:42:49] Epoch 3 | Step 31600 | Loss: 0.2057 | LM: 0.1943 | LB: 1.0875 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 15:42:55] Epoch 3 | Step 31610 | Loss: 0.2057 | LM: 0.1943 | LB: 1.0875 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 15:43:01] Epoch 3 | Step 31620 | Loss: 0.2058 | LM: 0.1944 | LB: 1.0875 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 15:43:07] Epoch 3 | Step 31630 | Loss: 0.2058 | LM: 0.1944 | LB: 1.0875 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 15:43:13] Epoch 3 | Step 31640 | Loss: 0.2058 | LM: 0.1944 | LB: 1.0875 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 15:43:20] Epoch 3 | Step 31650 | Loss: 0.2058 | LM: 0.1945 | LB: 1.0875 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 15:43:26] Epoch 3 | Step 31660 | Loss: 0.2058 | LM: 0.1944 | LB: 1.0874 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 15:43:33] Epoch 3 | Step 31670 | Loss: 0.2058 | LM: 0.1944 | LB: 1.0875 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 15:43:39] Epoch 3 | Step 31680 | Loss: 0.2058 | LM: 0.1944 | LB: 1.0875 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 15:43:45] Epoch 3 | Step 31690 | Loss: 0.2058 | LM: 0.1944 | LB: 1.0874 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 15:43:52] Epoch 3 | Step 31700 | Loss: 0.2058 | LM: 0.1944 | LB: 1.0874 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 15:43:58] Epoch 3 | Step 31710 | Loss: 0.2058 | LM: 0.1944 | LB: 1.0875 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 15:44:04] Epoch 3 | Step 31720 | Loss: 0.2058 | LM: 0.1944 | LB: 1.0874 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 15:44:10] Epoch 3 | Step 31730 | Loss: 0.2058 | LM: 0.1944 | LB: 1.0874 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 15:44:17] Epoch 3 | Step 31740 | Loss: 0.2058 | LM: 0.1944 | LB: 1.0875 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 15:44:23] Epoch 3 | Step 31750 | Loss: 0.2058 | LM: 0.1944 | LB: 1.0875 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 15:44:30] Epoch 3 | Step 31760 | Loss: 0.2058 | LM: 0.1945 | LB: 1.0875 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 15:44:36] Epoch 3 | Step 31770 | Loss: 0.2058 | LM: 0.1945 | LB: 1.0875 | CL0: 2.9 | CL1: 2.4 | HR0: 0.347/SR0: 0.347 | HR1: 0.416/SR1: 0.384 | LR: 1.00e-05 [2026-04-17 15:44:39] Epoch 3 completed in 7926.80s | Loss: 0.2058 | CL0: 2.9 | CL1: 2.4 [2026-04-17 15:44:47] Checkpoint saved: outputs/2026-04-17/08-57-56/checkpoints/checkpoint_step_31773.pt [2026-04-17 15:45:02] Training completed! [2026-04-17 15:45:05] Final model: outputs/2026-04-17/08-57-56/model_final.pt