[2026-05-23 17:26:56] CUDA_VISIBLE_DEVICES: 0 [2026-05-23 17:26:56] Number of processes: 1 [2026-05-23 17:26:56] Mixed precision: bf16 [2026-05-23 17:26:56] ============================================================ [2026-05-23 17:26:56] Routing Evolution Experiment | N=2.5 [2026-05-23 17:26:56] ============================================================ [2026-05-23 17:26:56] 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 max_steps: null batch_size: 8 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: 0.5 load_balancing_N: 2.5 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_V4_full max_context_len: 4096 max_target_len: 256 num_workers: 0 pin_memory: true max_train_samples: 150000 max_val_samples: null logging: log_interval: 10 save_interval: 3000 eval_interval: 1000 save_every_epoch: false model_only_checkpoints: true tracking: enabled: true backend: wandb project: routing-evolution run_name: routing_N2.5 entity: null base_url: https://wandb.platun0v.ru local_dir: outputs/2026-05-23/N_2.5 paths: output_dir: outputs/2026-05-23/N_2.5 seed: 42 device: cuda [2026-05-23 17:26:59] Loading model... [2026-05-23 17:27:05] Loaded pretrained: /workspace/byte-llms-code/hnet_project/checkpoints/hnet_2stage_XL_code.pt [2026-05-23 17:27:05] Applied LR multipliers: [2.0, 1.5, 1.0] [2026-05-23 17:27:05] Warming up model... [2026-05-23 17:28:18] CUDA_VISIBLE_DEVICES: 0 [2026-05-23 17:28:18] Number of processes: 1 [2026-05-23 17:28:18] Mixed precision: bf16 [2026-05-23 17:28:18] ============================================================ [2026-05-23 17:28:18] Routing Evolution Experiment | N=2.5 [2026-05-23 17:28:18] ============================================================ [2026-05-23 17:28:18] 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 max_steps: null batch_size: 8 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: 0.5 load_balancing_N: 2.5 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_V4_full max_context_len: 4096 max_target_len: 256 num_workers: 0 pin_memory: true max_train_samples: 150000 max_val_samples: null logging: log_interval: 10 save_interval: 3000 eval_interval: 1000 save_every_epoch: false model_only_checkpoints: true tracking: enabled: true backend: wandb project: routing-evolution run_name: routing_N2.5 entity: null base_url: https://wandb.platun0v.ru local_dir: outputs/2026-05-23/N_2.5 paths: output_dir: outputs/2026-05-23/N_2.5 seed: 42 device: cuda [2026-05-23 17:28:20] Loading model... [2026-05-23 17:28:26] Loaded pretrained: /workspace/byte-llms-code/hnet_project/checkpoints/hnet_2stage_XL_code.pt [2026-05-23 17:28:26] Applied LR multipliers: [2.0, 1.5, 1.0] [2026-05-23 17:28:26] Warming up model... [2026-05-23 17:32:52] Total params: 1,654,090,112 [2026-05-23 17:32:52] Trainable params: 1,654,090,112 [2026-05-23 17:32:52] Creating dataloaders... [2026-05-23 17:32:52] Train dataset size: 150000 (max_train_samples=150000) | Epochs: 3 [2026-05-23 17:32:52] Max steps: 14062, Steps per epoch: 18750 [2026-05-23 17:32:52] Starting training... [2026-05-23 17:32:52] ============================================================ [2026-05-23 17:32:52] EPOCH 1/3 (step 0) [2026-05-23 17:32:52] ============================================================ [2026-05-23 17:36:49] Epoch 1 | Step 10 | Loss: 1.1261 | LM: 0.6789 | LB: 1.0137 | CL0: 2.8 | CL1: 2.1 | HR0: 0.356/SR0: 0.354 | HR1: 0.471/SR1: 0.448 | LR: 1.06e-05 [2026-05-23 17:37:06] Epoch 1 | Step 20 | Loss: 1.1214 | LM: 0.6306 | LB: 1.0141 | CL0: 2.8 | CL1: 2.1 | HR0: 0.355/SR0: 0.354 | HR1: 0.474/SR1: 0.452 | LR: 1.13e-05 [2026-05-23 17:37:20] Epoch 1 | Step 30 | Loss: 1.1086 | LM: 0.6218 | LB: 1.0152 | CL0: 2.9 | CL1: 2.1 | HR0: 0.350/SR0: 0.350 | HR1: 0.474/SR1: 0.451 | LR: 1.19e-05 [2026-05-23 17:37:32] Epoch 1 | Step 40 | Loss: 1.1008 | LM: 0.6120 | LB: 1.0160 | CL0: 2.9 | CL1: 2.1 | HR0: 0.352/SR0: 0.351 | HR1: 0.476/SR1: 0.454 | LR: 1.26e-05 [2026-05-23 17:37:45] Epoch 1 | Step 50 | Loss: 1.0841 | LM: 0.5950 | LB: 1.0157 | CL0: 2.9 | CL1: 2.1 | HR0: 0.352/SR0: 0.351 | HR1: 0.475/SR1: 0.452 | LR: 1.32e-05 [2026-05-23 17:37:56] Epoch 1 | Step 60 | Loss: 1.0621 | LM: 0.5725 | LB: 1.0160 | CL0: 2.9 | CL1: 2.1 | HR0: 0.352/SR0: 0.351 | HR1: 0.477/SR1: 0.454 | LR: 1.38e-05 [2026-05-23 17:38:08] Epoch 1 | Step 70 | Loss: 1.0499 | LM: 0.5534 | LB: 1.0156 | CL0: 2.9 | CL1: 2.1 | HR0: 0.352/SR0: 0.351 | HR1: 0.476/SR1: 0.454 | LR: 1.45e-05 [2026-05-23 17:38:20] Epoch 1 | Step 80 | Loss: 1.0323 | LM: 0.5292 | LB: 1.0155 | CL0: 2.8 | CL1: 2.1 | HR0: 0.353/SR0: 0.352 | HR1: 0.476/SR1: 0.454 | LR: 1.51e-05 [2026-05-23 17:38:32] Epoch 1 | Step 90 | Loss: 1.0236 | LM: 0.5216 | LB: 1.0156 | CL0: 2.8 | CL1: 2.1 | HR0: 0.353/SR0: 0.352 | HR1: 0.477/SR1: 0.454 | LR: 1.58e-05 [2026-05-23 17:38:44] Epoch 1 | Step 100 | Loss: 1.0103 | LM: 0.5088 | LB: 1.0157 | CL0: 2.8 | CL1: 2.1 | HR0: 0.353/SR0: 0.352 | HR1: 0.478/SR1: 0.455 | LR: 1.64e-05 [2026-05-23 17:38:56] Epoch 1 | Step 110 | Loss: 1.0008 | LM: 0.4941 | LB: 1.0157 | CL0: 2.8 | CL1: 2.1 | HR0: 0.354/SR0: 0.353 | HR1: 0.478/SR1: 0.455 | LR: 1.70e-05 [2026-05-23 17:39:08] Epoch 1 | Step 120 | Loss: 0.9916 | LM: 0.4873 | LB: 1.0157 | CL0: 2.8 | CL1: 2.1 | HR0: 0.354/SR0: 0.353 | HR1: 0.479/SR1: 0.455 | LR: 1.77e-05 [2026-05-23 17:39:19] Epoch 1 | Step 130 | Loss: 0.9802 | LM: 0.4751 | LB: 1.0154 | CL0: 2.8 | CL1: 2.1 | HR0: 0.354/SR0: 0.353 | HR1: 0.477/SR1: 0.455 | LR: 1.83e-05 [2026-05-23 17:39:30] Epoch 1 | Step 140 | Loss: 0.9714 | LM: 0.4614 | LB: 1.0150 | CL0: 2.8 | CL1: 2.1 | HR0: 0.355/SR0: 0.354 | HR1: 0.477/SR1: 0.454 | LR: 1.90e-05 [2026-05-23 17:39:42] Epoch 1 | Step 150 | Loss: 0.9627 | LM: 0.4532 | LB: 1.0149 | CL0: 2.8 | CL1: 2.1 | HR0: 0.356/SR0: 0.355 | HR1: 0.477/SR1: 0.454 | LR: 1.96e-05 [2026-05-23 17:39:55] Epoch 1 | Step 160 | Loss: 0.9574 | LM: 0.4514 | LB: 1.0148 | CL0: 2.8 | CL1: 2.1 | HR0: 0.356/SR0: 0.354 | HR1: 0.476/SR1: 0.454 | LR: 2.02e-05 [2026-05-23 17:40:07] Epoch 1 | Step 170 | Loss: 0.9508 | LM: 0.4451 | LB: 1.0151 | CL0: 2.8 | CL1: 2.1 | HR0: 0.355/SR0: 0.354 | HR1: 0.476/SR1: 0.454 | LR: 2.09e-05 [2026-05-23 17:40:18] Epoch 1 | Step 180 | Loss: 0.9432 | LM: 0.4388 | LB: 1.0147 | CL0: 2.8 | CL1: 2.1 | HR0: 0.355/SR0: 0.354 | HR1: 0.475/SR1: 0.453 | LR: 2.15e-05 [2026-05-23 17:40:29] Epoch 1 | Step 190 | Loss: 0.9382 | LM: 0.4338 | LB: 1.0148 | CL0: 2.8 | CL1: 2.1 | HR0: 0.355/SR0: 0.354 | HR1: 0.475/SR1: 0.453 | LR: 2.22e-05 [2026-05-23 17:40:41] Epoch 1 | Step 200 | Loss: 0.9343 | LM: 0.4305 | LB: 1.0148 | CL0: 2.8 | CL1: 2.1 | HR0: 0.355/SR0: 0.353 | HR1: 0.475/SR1: 0.453 | LR: 2.28e-05 [2026-05-23 17:40:53] Epoch 1 | Step 210 | Loss: 0.9296 | LM: 0.4239 | LB: 1.0147 | CL0: 2.8 | CL1: 2.1 | HR0: 0.354/SR0: 0.353 | HR1: 0.475/SR1: 0.453 | LR: 2.34e-05 [2026-05-23 17:41:04] Epoch 1 | Step 220 | Loss: 0.9265 | LM: 0.4222 | LB: 1.0145 | CL0: 2.8 | CL1: 2.1 | HR0: 0.355/SR0: 0.354 | HR1: 0.474/SR1: 0.452 | LR: 2.41e-05 [2026-05-23 17:41:16] Epoch 1 | Step 230 | Loss: 0.9213 | LM: 0.4181 | LB: 1.0144 | CL0: 2.8 | CL1: 2.1 | HR0: 0.355/SR0: 0.354 | HR1: 0.473/SR1: 0.452 | LR: 2.47e-05 [2026-05-23 17:41:27] Epoch 1 | Step 240 | Loss: 0.9183 | LM: 0.4173 | LB: 1.0144 | CL0: 2.8 | CL1: 2.1 | HR0: 0.355/SR0: 0.354 | HR1: 0.473/SR1: 0.452 | LR: 2.54e-05 [2026-05-23 17:41:38] Epoch 1 | Step 250 | Loss: 0.9138 | LM: 0.4134 | LB: 1.0143 | CL0: 2.8 | CL1: 2.1 | HR0: 0.355/SR0: 0.354 | HR1: 0.473/SR1: 0.452 | LR: 2.60e-05 [2026-05-23 17:41:50] Epoch 1 | Step 260 | Loss: 0.9103 | LM: 0.4090 | LB: 1.0143 | CL0: 2.8 | CL1: 2.1 | HR0: 0.355/SR0: 0.353 | HR1: 0.473/SR1: 0.452 | LR: 2.66e-05 [2026-05-23 17:42:01] Epoch 1 | Step 270 | Loss: 0.9073 | LM: 0.4060 | LB: 1.0142 | CL0: 2.8 | CL1: 2.1 | HR0: 0.355/SR0: 0.353 | HR1: 0.472/SR1: 0.451 | LR: 2.73e-05 [2026-05-23 17:42:13] Epoch 1 | Step 280 | Loss: 0.9040 | LM: 0.4011 | LB: 1.0140 | CL0: 2.8 | CL1: 2.1 | HR0: 0.355/SR0: 0.354 | HR1: 0.472/SR1: 0.451 | LR: 2.79e-05 [2026-05-23 17:42:26] Epoch 1 | Step 290 | Loss: 0.9020 | LM: 0.3997 | LB: 1.0138 | CL0: 2.8 | CL1: 2.1 | HR0: 0.355/SR0: 0.354 | HR1: 0.471/SR1: 0.450 | LR: 2.86e-05 [2026-05-23 17:42:38] Epoch 1 | Step 300 | Loss: 0.9006 | LM: 0.3988 | LB: 1.0138 | CL0: 2.8 | CL1: 2.1 | HR0: 0.355/SR0: 0.354 | HR1: 0.471/SR1: 0.450 | LR: 2.92e-05 [2026-05-23 17:42:50] Epoch 1 | Step 310 | Loss: 0.8979 | LM: 0.3960 | LB: 1.0138 | CL0: 2.8 | CL1: 2.1 | HR0: 0.355/SR0: 0.353 | HR1: 0.471/SR1: 0.449 | LR: 2.98e-05 [2026-05-23 17:43:02] Epoch 1 | Step 320 | Loss: 0.8959 | LM: 0.3942 | LB: 1.0137 | CL0: 2.8 | CL1: 2.1 | HR0: 0.355/SR0: 0.354 | HR1: 0.470/SR1: 0.449 | LR: 3.05e-05 [2026-05-23 17:43:13] Epoch 1 | Step 330 | Loss: 0.8943 | LM: 0.3937 | LB: 1.0136 | CL0: 2.8 | CL1: 2.1 | HR0: 0.355/SR0: 0.354 | HR1: 0.470/SR1: 0.449 | LR: 3.11e-05 [2026-05-23 17:43:24] Epoch 1 | Step 340 | Loss: 0.8912 | LM: 0.3902 | LB: 1.0135 | CL0: 2.8 | CL1: 2.1 | HR0: 0.355/SR0: 0.354 | HR1: 0.470/SR1: 0.449 | LR: 3.18e-05 [2026-05-23 17:43:35] Epoch 1 | Step 350 | Loss: 0.8895 | LM: 0.3897 | LB: 1.0133 | CL0: 2.8 | CL1: 2.1 | HR0: 0.355/SR0: 0.354 | HR1: 0.469/SR1: 0.448 | LR: 3.24e-05 [2026-05-23 17:43:47] Epoch 1 | Step 360 | Loss: 0.8879 | LM: 0.3880 | LB: 1.0133 | CL0: 2.8 | CL1: 2.1 | HR0: 0.355/SR0: 0.354 | HR1: 0.469/SR1: 0.448 | LR: 3.30e-05 [2026-05-23 17:43:59] Epoch 1 | Step 370 | Loss: 0.8860 | LM: 0.3858 | LB: 1.0131 | CL0: 2.8 | CL1: 2.1 | HR0: 0.355/SR0: 0.354 | HR1: 0.469/SR1: 0.447 | LR: 3.37e-05 [2026-05-23 17:44:10] Epoch 1 | Step 380 | Loss: 0.8838 | LM: 0.3833 | LB: 1.0130 | CL0: 2.8 | CL1: 2.1 | HR0: 0.355/SR0: 0.354 | HR1: 0.468/SR1: 0.447 | LR: 3.43e-05 [2026-05-23 17:44:22] Epoch 1 | Step 390 | Loss: 0.8817 | LM: 0.3804 | LB: 1.0129 | CL0: 2.8 | CL1: 2.1 | HR0: 0.356/SR0: 0.354 | HR1: 0.468/SR1: 0.446 | LR: 3.50e-05 [2026-05-23 17:44:33] Epoch 1 | Step 400 | Loss: 0.8806 | LM: 0.3796 | LB: 1.0128 | CL0: 2.8 | CL1: 2.1 | HR0: 0.355/SR0: 0.354 | HR1: 0.467/SR1: 0.446 | LR: 3.56e-05 [2026-05-23 17:44:44] Epoch 1 | Step 410 | Loss: 0.8795 | LM: 0.3795 | LB: 1.0127 | CL0: 2.8 | CL1: 2.1 | HR0: 0.356/SR0: 0.354 | HR1: 0.467/SR1: 0.446 | LR: 3.62e-05 [2026-05-23 17:44:57] Epoch 1 | Step 420 | Loss: 0.8786 | LM: 0.3783 | LB: 1.0126 | CL0: 2.8 | CL1: 2.1 | HR0: 0.356/SR0: 0.354 | HR1: 0.467/SR1: 0.445 | LR: 3.69e-05 [2026-05-23 17:45:08] Epoch 1 | Step 430 | Loss: 0.8781 | LM: 0.3787 | LB: 1.0125 | CL0: 2.8 | CL1: 2.2 | HR0: 0.356/SR0: 0.354 | HR1: 0.466/SR1: 0.445 | LR: 3.75e-05 [2026-05-23 17:45:20] Epoch 1 | Step 440 | Loss: 0.8763 | LM: 0.3757 | LB: 1.0125 | CL0: 2.8 | CL1: 2.2 | HR0: 0.356/SR0: 0.354 | HR1: 0.466/SR1: 0.445 | LR: 3.82e-05 [2026-05-23 17:45:32] Epoch 1 | Step 450 | Loss: 0.8746 | LM: 0.3743 | LB: 1.0124 | CL0: 2.8 | CL1: 2.2 | HR0: 0.356/SR0: 0.354 | HR1: 0.466/SR1: 0.445 | LR: 3.88e-05 [2026-05-23 17:45:43] Epoch 1 | Step 460 | Loss: 0.8736 | LM: 0.3729 | LB: 1.0123 | CL0: 2.8 | CL1: 2.2 | HR0: 0.356/SR0: 0.355 | HR1: 0.466/SR1: 0.444 | LR: 3.94e-05 [2026-05-23 17:45:55] Epoch 1 | Step 470 | Loss: 0.8725 | LM: 0.3715 | LB: 1.0122 | CL0: 2.8 | CL1: 2.2 | HR0: 0.356/SR0: 0.355 | HR1: 0.465/SR1: 0.444 | LR: 4.01e-05 [2026-05-23 17:46:07] Epoch 1 | Step 480 | Loss: 0.8706 | LM: 0.3693 | LB: 1.0121 | CL0: 2.8 | CL1: 2.2 | HR0: 0.356/SR0: 0.355 | HR1: 0.465/SR1: 0.443 | LR: 4.07e-05 [2026-05-23 17:46:19] Epoch 1 | Step 490 | Loss: 0.8688 | LM: 0.3687 | LB: 1.0120 | CL0: 2.8 | CL1: 2.2 | HR0: 0.356/SR0: 0.355 | HR1: 0.464/SR1: 0.443 | LR: 4.14e-05 [2026-05-23 17:46:31] Epoch 1 | Step 500 | Loss: 0.8681 | LM: 0.3676 | LB: 1.0119 | CL0: 2.8 | CL1: 2.2 | HR0: 0.357/SR0: 0.355 | HR1: 0.464/SR1: 0.442 | LR: 4.20e-05 [2026-05-23 17:46:42] Epoch 1 | Step 510 | Loss: 0.8673 | LM: 0.3664 | LB: 1.0118 | CL0: 2.8 | CL1: 2.2 | HR0: 0.357/SR0: 0.356 | HR1: 0.464/SR1: 0.442 | LR: 4.26e-05 [2026-05-23 17:46:54] Epoch 1 | Step 520 | Loss: 0.8660 | LM: 0.3643 | LB: 1.0117 | CL0: 2.8 | CL1: 2.2 | HR0: 0.357/SR0: 0.356 | HR1: 0.464/SR1: 0.442 | LR: 4.33e-05 [2026-05-23 17:47:06] Epoch 1 | Step 530 | Loss: 0.8649 | LM: 0.3630 | LB: 1.0116 | CL0: 2.8 | CL1: 2.2 | HR0: 0.357/SR0: 0.356 | HR1: 0.463/SR1: 0.442 | LR: 4.39e-05 [2026-05-23 17:47:19] Epoch 1 | Step 540 | Loss: 0.8629 | LM: 0.3611 | LB: 1.0115 | CL0: 2.8 | CL1: 2.2 | HR0: 0.357/SR0: 0.356 | HR1: 0.463/SR1: 0.441 | LR: 4.46e-05 [2026-05-23 17:47:32] Epoch 1 | Step 550 | Loss: 0.8615 | LM: 0.3593 | LB: 1.0115 | CL0: 2.8 | CL1: 2.2 | HR0: 0.357/SR0: 0.356 | HR1: 0.462/SR1: 0.441 | LR: 4.52e-05 [2026-05-23 17:47:44] Epoch 1 | Step 560 | Loss: 0.8602 | LM: 0.3577 | LB: 1.0114 | CL0: 2.8 | CL1: 2.2 | HR0: 0.357/SR0: 0.355 | HR1: 0.462/SR1: 0.440 | LR: 4.58e-05 [2026-05-23 17:47:56] Epoch 1 | Step 570 | Loss: 0.8596 | LM: 0.3568 | LB: 1.0114 | CL0: 2.8 | CL1: 2.2 | HR0: 0.357/SR0: 0.356 | HR1: 0.462/SR1: 0.440 | LR: 4.65e-05 [2026-05-23 17:48:09] Epoch 1 | Step 580 | Loss: 0.8592 | LM: 0.3571 | LB: 1.0113 | CL0: 2.8 | CL1: 2.2 | HR0: 0.357/SR0: 0.356 | HR1: 0.462/SR1: 0.440 | LR: 4.71e-05 [2026-05-23 17:48:22] Epoch 1 | Step 590 | Loss: 0.8585 | LM: 0.3556 | LB: 1.0112 | CL0: 2.8 | CL1: 2.2 | HR0: 0.357/SR0: 0.356 | HR1: 0.461/SR1: 0.439 | LR: 4.78e-05 [2026-05-23 17:48:35] Epoch 1 | Step 600 | Loss: 0.8581 | LM: 0.3555 | LB: 1.0112 | CL0: 2.8 | CL1: 2.2 | HR0: 0.356/SR0: 0.356 | HR1: 0.461/SR1: 0.439 | LR: 4.84e-05 [2026-05-23 17:48:46] Epoch 1 | Step 610 | Loss: 0.8568 | LM: 0.3541 | LB: 1.0111 | CL0: 2.8 | CL1: 2.2 | HR0: 0.357/SR0: 0.356 | HR1: 0.461/SR1: 0.439 | LR: 4.90e-05 [2026-05-23 17:48:59] Epoch 1 | Step 620 | Loss: 0.8556 | LM: 0.3525 | LB: 1.0110 | CL0: 2.8 | CL1: 2.2 | HR0: 0.357/SR0: 0.356 | HR1: 0.460/SR1: 0.438 | LR: 4.97e-05 [2026-05-23 17:49:11] Epoch 1 | Step 630 | Loss: 0.8544 | LM: 0.3517 | LB: 1.0109 | CL0: 2.8 | CL1: 2.2 | HR0: 0.357/SR0: 0.356 | HR1: 0.460/SR1: 0.438 | LR: 5.03e-05 [2026-05-23 17:49:23] Epoch 1 | Step 640 | Loss: 0.8535 | LM: 0.3513 | LB: 1.0108 | CL0: 2.8 | CL1: 2.2 | HR0: 0.357/SR0: 0.356 | HR1: 0.460/SR1: 0.437 | LR: 5.10e-05 [2026-05-23 17:49:35] Epoch 1 | Step 650 | Loss: 0.8524 | LM: 0.3497 | LB: 1.0107 | CL0: 2.8 | CL1: 2.2 | HR0: 0.357/SR0: 0.356 | HR1: 0.459/SR1: 0.437 | LR: 5.16e-05 [2026-05-23 17:49:48] Epoch 1 | Step 660 | Loss: 0.8517 | LM: 0.3497 | LB: 1.0107 | CL0: 2.8 | CL1: 2.2 | HR0: 0.357/SR0: 0.356 | HR1: 0.459/SR1: 0.437 | LR: 5.22e-05 [2026-05-23 17:49:59] Epoch 1 | Step 670 | Loss: 0.8509 | LM: 0.3489 | LB: 1.0106 | CL0: 2.8 | CL1: 2.2 | HR0: 0.357/SR0: 0.356 | HR1: 0.459/SR1: 0.436 | LR: 5.29e-05 [2026-05-23 17:50:12] Epoch 1 | Step 680 | Loss: 0.8501 | LM: 0.3487 | LB: 1.0105 | CL0: 2.8 | CL1: 2.2 | HR0: 0.357/SR0: 0.356 | HR1: 0.458/SR1: 0.436 | LR: 5.35e-05 [2026-05-23 17:50:25] Epoch 1 | Step 690 | Loss: 0.8496 | LM: 0.3476 | LB: 1.0105 | CL0: 2.8 | CL1: 2.2 | HR0: 0.357/SR0: 0.356 | HR1: 0.458/SR1: 0.436 | LR: 5.42e-05 [2026-05-23 17:50:36] Epoch 1 | Step 700 | Loss: 0.8484 | LM: 0.3462 | LB: 1.0104 | CL0: 2.8 | CL1: 2.2 | HR0: 0.357/SR0: 0.356 | HR1: 0.458/SR1: 0.435 | LR: 5.48e-05 [2026-05-23 17:50:48] Epoch 1 | Step 710 | Loss: 0.8479 | LM: 0.3455 | LB: 1.0104 | CL0: 2.8 | CL1: 2.2 | HR0: 0.357/SR0: 0.356 | HR1: 0.457/SR1: 0.435 | LR: 5.54e-05 [2026-05-23 17:51:00] Epoch 1 | Step 720 | Loss: 0.8472 | LM: 0.3448 | LB: 1.0103 | CL0: 2.8 | CL1: 2.2 | HR0: 0.357/SR0: 0.356 | HR1: 0.457/SR1: 0.434 | LR: 5.61e-05 [2026-05-23 17:51:12] Epoch 1 | Step 730 | Loss: 0.8469 | LM: 0.3445 | LB: 1.0102 | CL0: 2.8 | CL1: 2.2 | HR0: 0.357/SR0: 0.356 | HR1: 0.457/SR1: 0.434 | LR: 5.67e-05 [2026-05-23 17:51:24] Epoch 1 | Step 740 | Loss: 0.8460 | LM: 0.3433 | LB: 1.0102 | CL0: 2.8 | CL1: 2.2 | HR0: 0.357/SR0: 0.357 | HR1: 0.456/SR1: 0.434 | LR: 5.74e-05 [2026-05-23 17:51:35] Epoch 1 | Step 750 | Loss: 0.8457 | LM: 0.3427 | LB: 1.0101 | CL0: 2.8 | CL1: 2.2 | HR0: 0.357/SR0: 0.356 | HR1: 0.456/SR1: 0.433 | LR: 5.80e-05 [2026-05-23 17:51:47] Epoch 1 | Step 760 | Loss: 0.8453 | LM: 0.3424 | LB: 1.0101 | CL0: 2.8 | CL1: 2.2 | HR0: 0.357/SR0: 0.356 | HR1: 0.455/SR1: 0.433 | LR: 5.86e-05 [2026-05-23 17:51:59] Epoch 1 | Step 770 | Loss: 0.8439 | LM: 0.3412 | LB: 1.0100 | CL0: 2.8 | CL1: 2.2 | HR0: 0.357/SR0: 0.356 | HR1: 0.455/SR1: 0.432 | LR: 5.93e-05 [2026-05-23 17:52:11] Epoch 1 | Step 780 | Loss: 0.8433 | LM: 0.3403 | LB: 1.0100 | CL0: 2.8 | CL1: 2.2 | HR0: 0.357/SR0: 0.356 | HR1: 0.455/SR1: 0.432 | LR: 5.99e-05 [2026-05-23 17:52:22] Epoch 1 | Step 790 | Loss: 0.8427 | LM: 0.3398 | LB: 1.0099 | CL0: 2.8 | CL1: 2.2 | HR0: 0.357/SR0: 0.357 | HR1: 0.454/SR1: 0.432 | LR: 6.06e-05 [2026-05-23 17:52:35] Epoch 1 | Step 800 | Loss: 0.8424 | LM: 0.3399 | LB: 1.0099 | CL0: 2.8 | CL1: 2.2 | HR0: 0.357/SR0: 0.357 | HR1: 0.454/SR1: 0.431 | LR: 6.12e-05 [2026-05-23 17:52:47] Epoch 1 | Step 810 | Loss: 0.8421 | LM: 0.3400 | LB: 1.0098 | CL0: 2.8 | CL1: 2.2 | HR0: 0.357/SR0: 0.357 | HR1: 0.454/SR1: 0.431 | LR: 6.18e-05 [2026-05-23 17:52:59] Epoch 1 | Step 820 | Loss: 0.8419 | LM: 0.3399 | LB: 1.0097 | CL0: 2.8 | CL1: 2.2 | HR0: 0.357/SR0: 0.357 | HR1: 0.453/SR1: 0.430 | LR: 6.25e-05 [2026-05-23 17:53:10] Epoch 1 | Step 830 | Loss: 0.8412 | LM: 0.3392 | LB: 1.0097 | CL0: 2.8 | CL1: 2.2 | HR0: 0.357/SR0: 0.357 | HR1: 0.453/SR1: 0.430 | LR: 6.31e-05 [2026-05-23 17:53:22] Epoch 1 | Step 840 | Loss: 0.8408 | LM: 0.3391 | LB: 1.0096 | CL0: 2.8 | CL1: 2.2 | HR0: 0.357/SR0: 0.357 | HR1: 0.453/SR1: 0.430 | LR: 6.38e-05 [2026-05-23 17:53:34] Epoch 1 | Step 850 | Loss: 0.8402 | LM: 0.3383 | LB: 1.0096 | CL0: 2.8 | CL1: 2.2 | HR0: 0.357/SR0: 0.357 | HR1: 0.452/SR1: 0.429 | LR: 6.44e-05 [2026-05-23 17:53:46] Epoch 1 | Step 860 | Loss: 0.8396 | LM: 0.3381 | LB: 1.0096 | CL0: 2.8 | CL1: 2.2 | HR0: 0.357/SR0: 0.357 | HR1: 0.452/SR1: 0.429 | LR: 6.50e-05 [2026-05-23 17:53:57] Epoch 1 | Step 870 | Loss: 0.8388 | LM: 0.3368 | LB: 1.0095 | CL0: 2.8 | CL1: 2.2 | HR0: 0.357/SR0: 0.357 | HR1: 0.452/SR1: 0.429 | LR: 6.57e-05 [2026-05-23 17:54:09] Epoch 1 | Step 880 | Loss: 0.8381 | LM: 0.3361 | LB: 1.0095 | CL0: 2.8 | CL1: 2.2 | HR0: 0.357/SR0: 0.357 | HR1: 0.451/SR1: 0.428 | LR: 6.63e-05 [2026-05-23 17:54:21] Epoch 1 | Step 890 | Loss: 0.8380 | LM: 0.3357 | LB: 1.0094 | CL0: 2.8 | CL1: 2.2 | HR0: 0.357/SR0: 0.357 | HR1: 0.451/SR1: 0.428 | LR: 6.70e-05 [2026-05-23 17:54:32] Epoch 1 | Step 900 | Loss: 0.8373 | LM: 0.3346 | LB: 1.0093 | CL0: 2.8 | CL1: 2.2 | HR0: 0.357/SR0: 0.357 | HR1: 0.451/SR1: 0.428 | LR: 6.76e-05 [2026-05-23 17:54:44] Epoch 1 | Step 910 | Loss: 0.8367 | LM: 0.3341 | LB: 1.0093 | CL0: 2.8 | CL1: 2.2 | HR0: 0.357/SR0: 0.357 | HR1: 0.450/SR1: 0.427 | LR: 6.83e-05 [2026-05-23 17:54:55] Epoch 1 | Step 920 | Loss: 0.8360 | LM: 0.3336 | LB: 1.0092 | CL0: 2.8 | CL1: 2.2 | HR0: 0.357/SR0: 0.357 | HR1: 0.450/SR1: 0.427 | LR: 6.89e-05 [2026-05-23 17:55:06] Epoch 1 | Step 930 | Loss: 0.8357 | LM: 0.3328 | LB: 1.0092 | CL0: 2.8 | CL1: 2.2 | HR0: 0.357/SR0: 0.357 | HR1: 0.450/SR1: 0.426 | LR: 6.95e-05 [2026-05-23 17:55:18] Epoch 1 | Step 940 | Loss: 0.8349 | LM: 0.3318 | LB: 1.0091 | CL0: 2.8 | CL1: 2.2 | HR0: 0.357/SR0: 0.357 | HR1: 0.449/SR1: 0.426 | LR: 7.02e-05 [2026-05-23 17:55:30] Epoch 1 | Step 950 | Loss: 0.8346 | LM: 0.3317 | LB: 1.0091 | CL0: 2.8 | CL1: 2.2 | HR0: 0.357/SR0: 0.357 | HR1: 0.449/SR1: 0.426 | LR: 7.08e-05 [2026-05-23 17:55:41] Epoch 1 | Step 960 | Loss: 0.8345 | LM: 0.3312 | LB: 1.0091 | CL0: 2.8 | CL1: 2.2 | HR0: 0.357/SR0: 0.357 | HR1: 0.449/SR1: 0.425 | LR: 7.15e-05 [2026-05-23 17:55:53] Epoch 1 | Step 970 | Loss: 0.8341 | LM: 0.3304 | LB: 1.0090 | CL0: 2.8 | CL1: 2.2 | HR0: 0.357/SR0: 0.358 | HR1: 0.448/SR1: 0.425 | LR: 7.21e-05 [2026-05-23 17:56:05] Epoch 1 | Step 980 | Loss: 0.8338 | LM: 0.3301 | LB: 1.0090 | CL0: 2.8 | CL1: 2.2 | HR0: 0.357/SR0: 0.358 | HR1: 0.448/SR1: 0.425 | LR: 7.27e-05 [2026-05-23 17:56:16] Epoch 1 | Step 990 | Loss: 0.8333 | LM: 0.3294 | LB: 1.0089 | CL0: 2.8 | CL1: 2.2 | HR0: 0.357/SR0: 0.358 | HR1: 0.448/SR1: 0.425 | LR: 7.34e-05 [2026-05-23 17:56:28] Epoch 1 | Step 1000 | Loss: 0.8327 | LM: 0.3295 | LB: 1.0089 | CL0: 2.8 | CL1: 2.2 | HR0: 0.357/SR0: 0.358 | HR1: 0.447/SR1: 0.424 | LR: 7.40e-05 [2026-05-23 17:56:30] Validation | Batch 10/1463 | Loss: 0.7602 | LM_LOSS: 0.2579 | LB_LOSS: 1.0048 [2026-05-23 17:56:32] Validation | Batch 20/1463 | Loss: 0.7707 | LM_LOSS: 0.2689 | LB_LOSS: 1.0036 [2026-05-23 17:56:35] Validation | Batch 30/1463 | Loss: 0.7776 | LM_LOSS: 0.2759 | LB_LOSS: 1.0036 [2026-05-23 17:56:38] Validation | Batch 40/1463 | Loss: 0.7837 | LM_LOSS: 0.2820 | LB_LOSS: 1.0032 [2026-05-23 17:56:40] Validation | Batch 50/1463 | Loss: 0.7915 | LM_LOSS: 0.2898 | LB_LOSS: 1.0035 [2026-05-23 17:56:44] Validation | Batch 60/1463 | Loss: 0.7915 | LM_LOSS: 0.2897 | LB_LOSS: 1.0035 [2026-05-23 17:56:47] Validation | Batch 70/1463 | Loss: 0.7917 | LM_LOSS: 0.2900 | LB_LOSS: 1.0035 [2026-05-23 17:56:50] Validation | Batch 80/1463 | Loss: 0.7935 | LM_LOSS: 0.2918 | LB_LOSS: 1.0035 [2026-05-23 17:56:53] Validation | Batch 90/1463 | Loss: 0.7933 | LM_LOSS: 0.2916 | LB_LOSS: 1.0034 [2026-05-23 17:56:55] Validation | Batch 100/1463 | Loss: 0.7934 | LM_LOSS: 0.2917 | LB_LOSS: 1.0034 [2026-05-23 17:56:58] Validation | Batch 110/1463 | Loss: 0.7923 | LM_LOSS: 0.2907 | LB_LOSS: 1.0033 [2026-05-23 17:57:00] Validation | Batch 120/1463 | Loss: 0.7938 | LM_LOSS: 0.2922 | LB_LOSS: 1.0033 [2026-05-23 17:57:02] Validation | Batch 130/1463 | Loss: 0.7953 | LM_LOSS: 0.2937 | LB_LOSS: 1.0034 [2026-05-23 17:57:06] Validation | Batch 140/1463 | Loss: 0.7962 | LM_LOSS: 0.2945 | LB_LOSS: 1.0034 [2026-05-23 17:57:08] Validation | Batch 150/1463 | Loss: 0.7941 | LM_LOSS: 0.2924 | LB_LOSS: 1.0034 [2026-05-23 17:57:11] Validation | Batch 160/1463 | Loss: 0.7942 | LM_LOSS: 0.2925 | LB_LOSS: 1.0034 [2026-05-23 17:57:14] Validation | Batch 170/1463 | Loss: 0.7936 | LM_LOSS: 0.2919 | LB_LOSS: 1.0034 [2026-05-23 17:57:17] Validation | Batch 180/1463 | Loss: 0.7945 | LM_LOSS: 0.2928 | LB_LOSS: 1.0034 [2026-05-23 17:57:19] Validation | Batch 190/1463 | Loss: 0.7952 | LM_LOSS: 0.2935 | LB_LOSS: 1.0034 [2026-05-23 17:57:22] Validation | Batch 200/1463 | Loss: 0.7953 | LM_LOSS: 0.2936 | LB_LOSS: 1.0034 [2026-05-23 17:57:24] Validation | Batch 210/1463 | Loss: 0.7960 | LM_LOSS: 0.2943 | LB_LOSS: 1.0034 [2026-05-23 17:57:28] Validation | Batch 220/1463 | Loss: 0.7951 | LM_LOSS: 0.2934 | LB_LOSS: 1.0034 [2026-05-23 17:57:30] Validation | Batch 230/1463 | Loss: 0.7968 | LM_LOSS: 0.2951 | LB_LOSS: 1.0034 [2026-05-23 17:57:33] Validation | Batch 240/1463 | Loss: 0.7982 | LM_LOSS: 0.2965 | LB_LOSS: 1.0034 [2026-05-23 17:57:36] Validation | Batch 250/1463 | Loss: 0.7974 | LM_LOSS: 0.2957 | LB_LOSS: 1.0033 [2026-05-23 17:57:39] Validation | Batch 260/1463 | Loss: 0.7988 | LM_LOSS: 0.2971 | LB_LOSS: 1.0033 [2026-05-23 17:57:41] Validation | Batch 270/1463 | Loss: 0.7972 | LM_LOSS: 0.2956 | LB_LOSS: 1.0033 [2026-05-23 17:57:44] Validation | Batch 280/1463 | Loss: 0.7978 | LM_LOSS: 0.2961 | LB_LOSS: 1.0033 [2026-05-23 17:57:47] Validation | Batch 290/1463 | Loss: 0.7970 | LM_LOSS: 0.2954 | LB_LOSS: 1.0033 [2026-05-23 17:57:49] Validation | Batch 300/1463 | Loss: 0.7969 | LM_LOSS: 0.2953 | LB_LOSS: 1.0033 [2026-05-23 17:57:52] Validation | Batch 310/1463 | Loss: 0.7981 | LM_LOSS: 0.2965 | LB_LOSS: 1.0033 [2026-05-23 17:57:54] Validation | Batch 320/1463 | Loss: 0.7967 | LM_LOSS: 0.2950 | LB_LOSS: 1.0033 [2026-05-23 17:57:57] Validation | Batch 330/1463 | Loss: 0.7967 | LM_LOSS: 0.2950 | LB_LOSS: 1.0033 [2026-05-23 17:57:59] Validation | Batch 340/1463 | Loss: 0.7965 | LM_LOSS: 0.2948 | LB_LOSS: 1.0033 [2026-05-23 17:58:02] Validation | Batch 350/1463 | Loss: 0.7972 | LM_LOSS: 0.2955 | LB_LOSS: 1.0033 [2026-05-23 17:58:05] Validation | Batch 360/1463 | Loss: 0.7967 | LM_LOSS: 0.2951 | LB_LOSS: 1.0033 [2026-05-23 17:58:07] Validation | Batch 370/1463 | Loss: 0.7964 | LM_LOSS: 0.2948 | LB_LOSS: 1.0033 [2026-05-23 17:58:10] Validation | Batch 380/1463 | Loss: 0.7964 | LM_LOSS: 0.2948 | LB_LOSS: 1.0033 [2026-05-23 17:58:12] Validation | Batch 390/1463 | Loss: 0.7970 | LM_LOSS: 0.2954 | LB_LOSS: 1.0033 [2026-05-23 17:58:15] Validation | Batch 400/1463 | Loss: 0.7968 | LM_LOSS: 0.2952 | LB_LOSS: 1.0033 [2026-05-23 17:58:17] Validation | Batch 410/1463 | Loss: 0.7965 | LM_LOSS: 0.2948 | LB_LOSS: 1.0033 [2026-05-23 17:58:20] Validation | Batch 420/1463 | Loss: 0.7965 | LM_LOSS: 0.2949 | LB_LOSS: 1.0032 [2026-05-23 17:58:24] Validation | Batch 430/1463 | Loss: 0.7960 | LM_LOSS: 0.2944 | LB_LOSS: 1.0032 [2026-05-23 17:58:26] Validation | Batch 440/1463 | Loss: 0.7961 | LM_LOSS: 0.2945 | LB_LOSS: 1.0032 [2026-05-23 17:58:28] Validation | Batch 450/1463 | Loss: 0.7959 | LM_LOSS: 0.2943 | LB_LOSS: 1.0032 [2026-05-23 17:58:31] Validation | Batch 460/1463 | Loss: 0.7960 | LM_LOSS: 0.2943 | LB_LOSS: 1.0032 [2026-05-23 17:58:34] Validation | Batch 470/1463 | Loss: 0.7966 | LM_LOSS: 0.2950 | LB_LOSS: 1.0033 [2026-05-23 17:58:37] Validation | Batch 480/1463 | Loss: 0.7964 | LM_LOSS: 0.2948 | LB_LOSS: 1.0033 [2026-05-23 17:58:40] Validation | Batch 490/1463 | Loss: 0.7963 | LM_LOSS: 0.2947 | LB_LOSS: 1.0033 [2026-05-23 17:58:43] Validation | Batch 500/1463 | Loss: 0.7962 | LM_LOSS: 0.2946 | LB_LOSS: 1.0033 [2026-05-23 17:58:46] Validation | Batch 510/1463 | Loss: 0.7957 | LM_LOSS: 0.2940 | LB_LOSS: 1.0033 [2026-05-23 17:58:48] Validation | Batch 520/1463 | Loss: 0.7953 | LM_LOSS: 0.2937 | LB_LOSS: 1.0033 [2026-05-23 17:58:50] Validation | Batch 530/1463 | Loss: 0.7956 | LM_LOSS: 0.2939 | LB_LOSS: 1.0033 [2026-05-23 17:58:53] Validation | Batch 540/1463 | Loss: 0.7958 | LM_LOSS: 0.2942 | LB_LOSS: 1.0033 [2026-05-23 17:58:55] Validation | Batch 550/1463 | Loss: 0.7955 | LM_LOSS: 0.2939 | LB_LOSS: 1.0032 [2026-05-23 17:58:58] Validation | Batch 560/1463 | Loss: 0.7957 | LM_LOSS: 0.2941 | LB_LOSS: 1.0032 [2026-05-23 17:59:01] Validation | Batch 570/1463 | Loss: 0.7957 | LM_LOSS: 0.2941 | LB_LOSS: 1.0032 [2026-05-23 17:59:03] Validation | Batch 580/1463 | Loss: 0.7956 | LM_LOSS: 0.2939 | LB_LOSS: 1.0032 [2026-05-23 17:59:06] Validation | Batch 590/1463 | Loss: 0.7953 | LM_LOSS: 0.2937 | LB_LOSS: 1.0032 [2026-05-23 17:59:09] Validation | Batch 600/1463 | Loss: 0.7957 | LM_LOSS: 0.2941 | LB_LOSS: 1.0032 [2026-05-23 17:59:12] Validation | Batch 610/1463 | Loss: 0.7958 | LM_LOSS: 0.2942 | LB_LOSS: 1.0032 [2026-05-23 17:59:15] Validation | Batch 620/1463 | Loss: 0.7954 | LM_LOSS: 0.2938 | LB_LOSS: 1.0032 [2026-05-23 17:59:18] Validation | Batch 630/1463 | Loss: 0.7948 | LM_LOSS: 0.2932 | LB_LOSS: 1.0032 [2026-05-23 17:59:20] Validation | Batch 640/1463 | Loss: 0.7945 | LM_LOSS: 0.2930 | LB_LOSS: 1.0032 [2026-05-23 17:59:23] Validation | Batch 650/1463 | Loss: 0.7941 | LM_LOSS: 0.2925 | LB_LOSS: 1.0032 [2026-05-23 17:59:25] Validation | Batch 660/1463 | Loss: 0.7937 | LM_LOSS: 0.2921 | LB_LOSS: 1.0032 [2026-05-23 17:59:28] Validation | Batch 670/1463 | Loss: 0.7935 | LM_LOSS: 0.2918 | LB_LOSS: 1.0032 [2026-05-23 17:59:30] Validation | Batch 680/1463 | Loss: 0.7934 | LM_LOSS: 0.2918 | LB_LOSS: 1.0032 [2026-05-23 17:59:33] Validation | Batch 690/1463 | Loss: 0.7937 | LM_LOSS: 0.2921 | LB_LOSS: 1.0032 [2026-05-23 17:59:35] Validation | Batch 700/1463 | Loss: 0.7935 | LM_LOSS: 0.2919 | LB_LOSS: 1.0032 [2026-05-23 17:59:38] Validation | Batch 710/1463 | Loss: 0.7939 | LM_LOSS: 0.2923 | LB_LOSS: 1.0032 [2026-05-23 17:59:41] Validation | Batch 720/1463 | Loss: 0.7947 | LM_LOSS: 0.2931 | LB_LOSS: 1.0032 [2026-05-23 17:59:43] Validation | Batch 730/1463 | Loss: 0.7945 | LM_LOSS: 0.2928 | LB_LOSS: 1.0032 [2026-05-23 17:59:45] Validation | Batch 740/1463 | Loss: 0.7941 | LM_LOSS: 0.2925 | LB_LOSS: 1.0032 [2026-05-23 17:59:47] Validation | Batch 750/1463 | Loss: 0.7941 | LM_LOSS: 0.2924 | LB_LOSS: 1.0032 [2026-05-23 17:59:50] Validation | Batch 760/1463 | Loss: 0.7939 | LM_LOSS: 0.2923 | LB_LOSS: 1.0032 [2026-05-23 17:59:52] Validation | Batch 770/1463 | Loss: 0.7938 | LM_LOSS: 0.2922 | LB_LOSS: 1.0032 [2026-05-23 17:59:54] Validation | Batch 780/1463 | Loss: 0.7936 | LM_LOSS: 0.2920 | LB_LOSS: 1.0032 [2026-05-23 17:59:56] Validation | Batch 790/1463 | Loss: 0.7938 | LM_LOSS: 0.2922 | LB_LOSS: 1.0032 [2026-05-23 17:59:59] Validation | Batch 800/1463 | Loss: 0.7937 | LM_LOSS: 0.2921 | LB_LOSS: 1.0032 [2026-05-23 18:00:01] Validation | Batch 810/1463 | Loss: 0.7934 | LM_LOSS: 0.2917 | LB_LOSS: 1.0032 [2026-05-23 18:00:04] Validation | Batch 820/1463 | Loss: 0.7936 | LM_LOSS: 0.2920 | LB_LOSS: 1.0032 [2026-05-23 18:00:07] Validation | Batch 830/1463 | Loss: 0.7939 | LM_LOSS: 0.2923 | LB_LOSS: 1.0032 [2026-05-23 18:00:10] Validation | Batch 840/1463 | Loss: 0.7940 | LM_LOSS: 0.2923 | LB_LOSS: 1.0032 [2026-05-23 18:00:13] Validation | Batch 850/1463 | Loss: 0.7943 | LM_LOSS: 0.2926 | LB_LOSS: 1.0032 [2026-05-23 18:00:15] Validation | Batch 860/1463 | Loss: 0.7943 | LM_LOSS: 0.2927 | LB_LOSS: 1.0032 [2026-05-23 18:00:18] Validation | Batch 870/1463 | Loss: 0.7943 | LM_LOSS: 0.2927 | LB_LOSS: 1.0032 [2026-05-23 18:00:20] Validation | Batch 880/1463 | Loss: 0.7935 | LM_LOSS: 0.2919 | LB_LOSS: 1.0032 [2026-05-23 18:00:23] Validation | Batch 890/1463 | Loss: 0.7935 | LM_LOSS: 0.2919 | LB_LOSS: 1.0032 [2026-05-23 18:00:26] Validation | Batch 900/1463 | Loss: 0.7932 | LM_LOSS: 0.2916 | LB_LOSS: 1.0032 [2026-05-23 18:00:29] Validation | Batch 910/1463 | Loss: 0.7932 | LM_LOSS: 0.2915 | LB_LOSS: 1.0032 [2026-05-23 18:00:32] Validation | Batch 920/1463 | Loss: 0.7933 | LM_LOSS: 0.2917 | LB_LOSS: 1.0032 [2026-05-23 18:00:34] Validation | Batch 930/1463 | Loss: 0.7931 | LM_LOSS: 0.2915 | LB_LOSS: 1.0033 [2026-05-23 18:00:37] Validation | Batch 940/1463 | Loss: 0.7929 | LM_LOSS: 0.2913 | LB_LOSS: 1.0033 [2026-05-23 18:00:40] Validation | Batch 950/1463 | Loss: 0.7932 | LM_LOSS: 0.2915 | LB_LOSS: 1.0033 [2026-05-23 18:00:42] Validation | Batch 960/1463 | Loss: 0.7932 | LM_LOSS: 0.2916 | LB_LOSS: 1.0033 [2026-05-23 18:00:45] Validation | Batch 970/1463 | Loss: 0.7933 | LM_LOSS: 0.2917 | LB_LOSS: 1.0033 [2026-05-23 18:00:48] Validation | Batch 980/1463 | Loss: 0.7936 | LM_LOSS: 0.2920 | LB_LOSS: 1.0033 [2026-05-23 18:00:50] Validation | Batch 990/1463 | Loss: 0.7939 | LM_LOSS: 0.2922 | LB_LOSS: 1.0033 [2026-05-23 18:00:53] Validation | Batch 1000/1463 | Loss: 0.7945 | LM_LOSS: 0.2928 | LB_LOSS: 1.0033 [2026-05-23 18:00:56] Validation | Batch 1010/1463 | Loss: 0.7945 | LM_LOSS: 0.2928 | LB_LOSS: 1.0033 [2026-05-23 18:00:59] Validation | Batch 1020/1463 | Loss: 0.7945 | LM_LOSS: 0.2929 | LB_LOSS: 1.0033 [2026-05-23 18:01:02] Validation | Batch 1030/1463 | Loss: 0.7949 | LM_LOSS: 0.2933 | LB_LOSS: 1.0033 [2026-05-23 18:01:04] Validation | Batch 1040/1463 | Loss: 0.7945 | LM_LOSS: 0.2929 | LB_LOSS: 1.0033 [2026-05-23 18:01:07] Validation | Batch 1050/1463 | Loss: 0.7944 | LM_LOSS: 0.2928 | LB_LOSS: 1.0033 [2026-05-23 18:01:10] Validation | Batch 1060/1463 | Loss: 0.7944 | LM_LOSS: 0.2928 | LB_LOSS: 1.0033 [2026-05-23 18:01:13] Validation | Batch 1070/1463 | Loss: 0.7944 | LM_LOSS: 0.2927 | LB_LOSS: 1.0033 [2026-05-23 18:01:15] Validation | Batch 1080/1463 | Loss: 0.7941 | LM_LOSS: 0.2925 | LB_LOSS: 1.0033 [2026-05-23 18:01:17] Validation | Batch 1090/1463 | Loss: 0.7943 | LM_LOSS: 0.2926 | LB_LOSS: 1.0033 [2026-05-23 18:01:20] Validation | Batch 1100/1463 | Loss: 0.7941 | LM_LOSS: 0.2925 | LB_LOSS: 1.0033 [2026-05-23 18:01:22] Validation | Batch 1110/1463 | Loss: 0.7941 | LM_LOSS: 0.2924 | LB_LOSS: 1.0033 [2026-05-23 18:01:25] Validation | Batch 1120/1463 | Loss: 0.7939 | LM_LOSS: 0.2923 | LB_LOSS: 1.0033 [2026-05-23 18:01:28] Validation | Batch 1130/1463 | Loss: 0.7937 | LM_LOSS: 0.2921 | LB_LOSS: 1.0033 [2026-05-23 18:01:31] Validation | Batch 1140/1463 | Loss: 0.7939 | LM_LOSS: 0.2923 | LB_LOSS: 1.0033 [2026-05-23 18:01:34] Validation | Batch 1150/1463 | Loss: 0.7938 | LM_LOSS: 0.2922 | LB_LOSS: 1.0033 [2026-05-23 18:01:36] Validation | Batch 1160/1463 | Loss: 0.7938 | LM_LOSS: 0.2922 | LB_LOSS: 1.0033 [2026-05-23 18:01:39] Validation | Batch 1170/1463 | Loss: 0.7938 | LM_LOSS: 0.2922 | LB_LOSS: 1.0033 [2026-05-23 18:01:42] Validation | Batch 1180/1463 | Loss: 0.7938 | LM_LOSS: 0.2922 | LB_LOSS: 1.0033 [2026-05-23 18:01:44] Validation | Batch 1190/1463 | Loss: 0.7940 | LM_LOSS: 0.2923 | LB_LOSS: 1.0033 [2026-05-23 18:01:48] Validation | Batch 1200/1463 | Loss: 0.7940 | LM_LOSS: 0.2923 | LB_LOSS: 1.0033 [2026-05-23 18:01:50] Validation | Batch 1210/1463 | Loss: 0.7942 | LM_LOSS: 0.2925 | LB_LOSS: 1.0033 [2026-05-23 18:01:53] Validation | Batch 1220/1463 | Loss: 0.7945 | LM_LOSS: 0.2929 | LB_LOSS: 1.0033 [2026-05-23 18:01:55] Validation | Batch 1230/1463 | Loss: 0.7945 | LM_LOSS: 0.2929 | LB_LOSS: 1.0033 [2026-05-23 18:01:58] Validation | Batch 1240/1463 | Loss: 0.7948 | LM_LOSS: 0.2932 | LB_LOSS: 1.0033 [2026-05-23 18:02:01] Validation | Batch 1250/1463 | Loss: 0.7950 | LM_LOSS: 0.2934 | LB_LOSS: 1.0033 [2026-05-23 18:02:04] Validation | Batch 1260/1463 | Loss: 0.7950 | LM_LOSS: 0.2934 | LB_LOSS: 1.0033 [2026-05-23 18:02:06] Validation | Batch 1270/1463 | Loss: 0.7948 | LM_LOSS: 0.2932 | LB_LOSS: 1.0032 [2026-05-23 18:02:09] Validation | Batch 1280/1463 | Loss: 0.7947 | LM_LOSS: 0.2931 | LB_LOSS: 1.0033 [2026-05-23 18:02:12] Validation | Batch 1290/1463 | Loss: 0.7947 | LM_LOSS: 0.2931 | LB_LOSS: 1.0033 [2026-05-23 18:02:14] Validation | Batch 1300/1463 | Loss: 0.7946 | LM_LOSS: 0.2929 | LB_LOSS: 1.0033 [2026-05-23 18:02:17] Validation | Batch 1310/1463 | Loss: 0.7945 | LM_LOSS: 0.2929 | LB_LOSS: 1.0033 [2026-05-23 18:02:19] Validation | Batch 1320/1463 | Loss: 0.7945 | LM_LOSS: 0.2929 | LB_LOSS: 1.0033 [2026-05-23 18:02:21] Validation | Batch 1330/1463 | Loss: 0.7946 | LM_LOSS: 0.2929 | LB_LOSS: 1.0033 [2026-05-23 18:02:23] Validation | Batch 1340/1463 | Loss: 0.7948 | LM_LOSS: 0.2932 | LB_LOSS: 1.0032 [2026-05-23 18:02:26] Validation | Batch 1350/1463 | Loss: 0.7949 | LM_LOSS: 0.2933 | LB_LOSS: 1.0032 [2026-05-23 18:02:30] Validation | Batch 1360/1463 | Loss: 0.7951 | LM_LOSS: 0.2935 | LB_LOSS: 1.0033 [2026-05-23 18:02:33] Validation | Batch 1370/1463 | Loss: 0.7954 | LM_LOSS: 0.2938 | LB_LOSS: 1.0033 [2026-05-23 18:02:35] Validation | Batch 1380/1463 | Loss: 0.7952 | LM_LOSS: 0.2936 | LB_LOSS: 1.0033 [2026-05-23 18:02:38] Validation | Batch 1390/1463 | Loss: 0.7951 | LM_LOSS: 0.2935 | LB_LOSS: 1.0033 [2026-05-23 18:02:40] Validation | Batch 1400/1463 | Loss: 0.7955 | LM_LOSS: 0.2938 | LB_LOSS: 1.0033 [2026-05-23 18:02:42] Validation | Batch 1410/1463 | Loss: 0.7957 | LM_LOSS: 0.2940 | LB_LOSS: 1.0033 [2026-05-23 18:02:45] Validation | Batch 1420/1463 | Loss: 0.7956 | LM_LOSS: 0.2940 | LB_LOSS: 1.0033 [2026-05-23 18:02:48] Validation | Batch 1430/1463 | Loss: 0.7957 | LM_LOSS: 0.2941 | LB_LOSS: 1.0033 [2026-05-23 18:02:51] Validation | Batch 1440/1463 | Loss: 0.7961 | LM_LOSS: 0.2945 | LB_LOSS: 1.0033 [2026-05-23 18:02:54] Validation | Batch 1450/1463 | Loss: 0.7958 | LM_LOSS: 0.2942 | LB_LOSS: 1.0033 [2026-05-23 18:02:56] Validation | Batch 1460/1463 | Loss: 0.7962 | LM_LOSS: 0.2946 | LB_LOSS: 1.0033 [2026-05-23 18:02:57] Validation | Batch 1463/1463 | Loss: 0.7961 | LM_LOSS: 0.2944 | LB_LOSS: 1.0033 [2026-05-23 18:02:57] Validation | Loss: 0.7961 | LM_LOSS: 0.2944 | LB_LOSS: 1.0033 | PPL: 1.34 | Time: 389.31s [2026-05-23 18:03:01] New best model saved! Val loss: 0.7961 [2026-05-23 18:03:12] Epoch 1 | Step 1010 | Loss: 0.8324 | LM: 0.3294 | LB: 1.0089 | CL0: 2.8 | CL1: 2.2 | HR0: 0.357/SR0: 0.358 | HR1: 0.447/SR1: 0.424 | LR: 7.47e-05 [2026-05-23 18:03:24] Epoch 1 | Step 1020 | Loss: 0.8320 | LM: 0.3288 | LB: 1.0088 | CL0: 2.8 | CL1: 2.2 | HR0: 0.357/SR0: 0.358 | HR1: 0.447/SR1: 0.423 | LR: 7.53e-05 [2026-05-23 18:03:36] Epoch 1 | Step 1030 | Loss: 0.8316 | LM: 0.3292 | LB: 1.0088 | CL0: 2.8 | CL1: 2.2 | HR0: 0.357/SR0: 0.358 | HR1: 0.447/SR1: 0.423 | LR: 7.59e-05 [2026-05-23 18:03:47] Epoch 1 | Step 1040 | Loss: 0.8312 | LM: 0.3293 | LB: 1.0087 | CL0: 2.8 | CL1: 2.3 | HR0: 0.357/SR0: 0.358 | HR1: 0.446/SR1: 0.423 | LR: 7.66e-05 [2026-05-23 18:03:59] Epoch 1 | Step 1050 | Loss: 0.8310 | LM: 0.3290 | LB: 1.0086 | CL0: 2.8 | CL1: 2.3 | HR0: 0.358/SR0: 0.358 | HR1: 0.446/SR1: 0.423 | LR: 7.72e-05 [2026-05-23 18:04:10] Epoch 1 | Step 1060 | Loss: 0.8306 | LM: 0.3288 | LB: 1.0086 | CL0: 2.8 | CL1: 2.3 | HR0: 0.358/SR0: 0.359 | HR1: 0.446/SR1: 0.422 | LR: 7.79e-05 [2026-05-23 18:04:22] Epoch 1 | Step 1070 | Loss: 0.8303 | LM: 0.3280 | LB: 1.0085 | CL0: 2.8 | CL1: 2.3 | HR0: 0.358/SR0: 0.359 | HR1: 0.445/SR1: 0.422 | LR: 7.85e-05 [2026-05-23 18:04:33] Epoch 1 | Step 1080 | Loss: 0.8298 | LM: 0.3272 | LB: 1.0085 | CL0: 2.8 | CL1: 2.3 | HR0: 0.358/SR0: 0.359 | HR1: 0.445/SR1: 0.422 | LR: 7.91e-05 [2026-05-23 18:04:45] Epoch 1 | Step 1090 | Loss: 0.8293 | LM: 0.3266 | LB: 1.0085 | CL0: 2.8 | CL1: 2.3 | HR0: 0.358/SR0: 0.359 | HR1: 0.445/SR1: 0.421 | LR: 7.98e-05 [2026-05-23 18:04:56] Epoch 1 | Step 1100 | Loss: 0.8289 | LM: 0.3262 | LB: 1.0084 | CL0: 2.8 | CL1: 2.3 | HR0: 0.358/SR0: 0.359 | HR1: 0.445/SR1: 0.421 | LR: 8.04e-05 [2026-05-23 18:05:08] Epoch 1 | Step 1110 | Loss: 0.8286 | LM: 0.3258 | LB: 1.0084 | CL0: 2.8 | CL1: 2.3 | HR0: 0.358/SR0: 0.359 | HR1: 0.444/SR1: 0.421 | LR: 8.11e-05 [2026-05-23 18:05:19] Epoch 1 | Step 1120 | Loss: 0.8286 | LM: 0.3258 | LB: 1.0083 | CL0: 2.8 | CL1: 2.3 | HR0: 0.358/SR0: 0.359 | HR1: 0.444/SR1: 0.420 | LR: 8.17e-05 [2026-05-23 18:05:30] Epoch 1 | Step 1130 | Loss: 0.8282 | LM: 0.3252 | LB: 1.0083 | CL0: 2.8 | CL1: 2.3 | HR0: 0.358/SR0: 0.359 | HR1: 0.444/SR1: 0.420 | LR: 8.23e-05 [2026-05-23 18:05:43] Epoch 1 | Step 1140 | Loss: 0.8280 | LM: 0.3251 | LB: 1.0083 | CL0: 2.8 | CL1: 2.3 | HR0: 0.358/SR0: 0.360 | HR1: 0.443/SR1: 0.420 | LR: 8.30e-05 [2026-05-23 18:05:55] Epoch 1 | Step 1150 | Loss: 0.8279 | LM: 0.3249 | LB: 1.0082 | CL0: 2.8 | CL1: 2.3 | HR0: 0.358/SR0: 0.360 | HR1: 0.443/SR1: 0.419 | LR: 8.36e-05 [2026-05-23 18:06:06] Epoch 1 | Step 1160 | Loss: 0.8275 | LM: 0.3244 | LB: 1.0082 | CL0: 2.8 | CL1: 2.3 | HR0: 0.358/SR0: 0.360 | HR1: 0.443/SR1: 0.419 | LR: 8.43e-05 [2026-05-23 18:06:17] Epoch 1 | Step 1170 | Loss: 0.8273 | LM: 0.3243 | LB: 1.0081 | CL0: 2.8 | CL1: 2.3 | HR0: 0.358/SR0: 0.360 | HR1: 0.443/SR1: 0.419 | LR: 8.49e-05 [2026-05-23 18:06:30] Epoch 1 | Step 1180 | Loss: 0.8272 | LM: 0.3242 | LB: 1.0081 | CL0: 2.8 | CL1: 2.3 | HR0: 0.358/SR0: 0.360 | HR1: 0.442/SR1: 0.418 | LR: 8.55e-05 [2026-05-23 18:06:42] Epoch 1 | Step 1190 | Loss: 0.8270 | LM: 0.3242 | LB: 1.0081 | CL0: 2.8 | CL1: 2.3 | HR0: 0.358/SR0: 0.360 | HR1: 0.442/SR1: 0.418 | LR: 8.62e-05 [2026-05-23 18:06:55] Epoch 1 | Step 1200 | Loss: 0.8267 | LM: 0.3239 | LB: 1.0080 | CL0: 2.8 | CL1: 2.3 | HR0: 0.358/SR0: 0.360 | HR1: 0.442/SR1: 0.418 | LR: 8.68e-05 [2026-05-23 18:07:08] Epoch 1 | Step 1210 | Loss: 0.8266 | LM: 0.3240 | LB: 1.0080 | CL0: 2.8 | CL1: 2.3 | HR0: 0.358/SR0: 0.361 | HR1: 0.441/SR1: 0.418 | LR: 8.75e-05 [2026-05-23 18:07:20] Epoch 1 | Step 1220 | Loss: 0.8263 | LM: 0.3241 | LB: 1.0079 | CL0: 2.8 | CL1: 2.3 | HR0: 0.358/SR0: 0.361 | HR1: 0.441/SR1: 0.417 | LR: 8.81e-05 [2026-05-23 18:07:33] Epoch 1 | Step 1230 | Loss: 0.8260 | LM: 0.3239 | LB: 1.0079 | CL0: 2.8 | CL1: 2.3 | HR0: 0.358/SR0: 0.361 | HR1: 0.441/SR1: 0.417 | LR: 8.87e-05 [2026-05-23 18:07:45] Epoch 1 | Step 1240 | Loss: 0.8260 | LM: 0.3237 | LB: 1.0079 | CL0: 2.8 | CL1: 2.3 | HR0: 0.358/SR0: 0.361 | HR1: 0.441/SR1: 0.417 | LR: 8.94e-05 [2026-05-23 18:07:57] Epoch 1 | Step 1250 | Loss: 0.8258 | LM: 0.3238 | LB: 1.0078 | CL0: 2.8 | CL1: 2.3 | HR0: 0.359/SR0: 0.361 | HR1: 0.440/SR1: 0.416 | LR: 9.00e-05 [2026-05-23 18:08:09] Epoch 1 | Step 1260 | Loss: 0.8253 | LM: 0.3231 | LB: 1.0078 | CL0: 2.8 | CL1: 2.3 | HR0: 0.359/SR0: 0.361 | HR1: 0.440/SR1: 0.416 | LR: 9.07e-05 [2026-05-23 18:08:22] Epoch 1 | Step 1270 | Loss: 0.8249 | LM: 0.3226 | LB: 1.0077 | CL0: 2.8 | CL1: 2.3 | HR0: 0.359/SR0: 0.361 | HR1: 0.440/SR1: 0.416 | LR: 9.13e-05 [2026-05-23 18:08:34] Epoch 1 | Step 1280 | Loss: 0.8246 | LM: 0.3228 | LB: 1.0077 | CL0: 2.8 | CL1: 2.3 | HR0: 0.359/SR0: 0.362 | HR1: 0.440/SR1: 0.416 | LR: 9.19e-05 [2026-05-23 18:08:47] Epoch 1 | Step 1290 | Loss: 0.8244 | LM: 0.3227 | LB: 1.0076 | CL0: 2.8 | CL1: 2.3 | HR0: 0.359/SR0: 0.362 | HR1: 0.439/SR1: 0.415 | LR: 9.26e-05 [2026-05-23 18:08:59] Epoch 1 | Step 1300 | Loss: 0.8244 | LM: 0.3229 | LB: 1.0076 | CL0: 2.8 | CL1: 2.3 | HR0: 0.359/SR0: 0.362 | HR1: 0.439/SR1: 0.415 | LR: 9.32e-05 [2026-05-23 18:09:12] Epoch 1 | Step 1310 | Loss: 0.8243 | LM: 0.3238 | LB: 1.0076 | CL0: 2.8 | CL1: 2.3 | HR0: 0.359/SR0: 0.362 | HR1: 0.439/SR1: 0.415 | LR: 9.39e-05 [2026-05-23 18:09:24] Epoch 1 | Step 1320 | Loss: 0.8240 | LM: 0.3236 | LB: 1.0075 | CL0: 2.8 | CL1: 2.3 | HR0: 0.359/SR0: 0.362 | HR1: 0.439/SR1: 0.415 | LR: 9.45e-05 [2026-05-23 18:09:37] Epoch 1 | Step 1330 | Loss: 0.8239 | LM: 0.3232 | LB: 1.0075 | CL0: 2.8 | CL1: 2.3 | HR0: 0.359/SR0: 0.362 | HR1: 0.439/SR1: 0.414 | LR: 9.51e-05 [2026-05-23 18:09:49] Epoch 1 | Step 1340 | Loss: 0.8237 | LM: 0.3229 | LB: 1.0075 | CL0: 2.8 | CL1: 2.3 | HR0: 0.359/SR0: 0.362 | HR1: 0.438/SR1: 0.414 | LR: 9.58e-05 [2026-05-23 18:10:01] Epoch 1 | Step 1350 | Loss: 0.8234 | LM: 0.3225 | LB: 1.0074 | CL0: 2.8 | CL1: 2.3 | HR0: 0.359/SR0: 0.363 | HR1: 0.438/SR1: 0.414 | LR: 9.64e-05 [2026-05-23 18:10:13] Epoch 1 | Step 1360 | Loss: 0.8233 | LM: 0.3224 | LB: 1.0074 | CL0: 2.8 | CL1: 2.3 | HR0: 0.359/SR0: 0.363 | HR1: 0.438/SR1: 0.414 | LR: 9.71e-05 [2026-05-23 18:10:25] Epoch 1 | Step 1370 | Loss: 0.8231 | LM: 0.3221 | LB: 1.0073 | CL0: 2.8 | CL1: 2.3 | HR0: 0.359/SR0: 0.363 | HR1: 0.438/SR1: 0.413 | LR: 9.77e-05 [2026-05-23 18:10:37] Epoch 1 | Step 1380 | Loss: 0.8227 | LM: 0.3219 | LB: 1.0073 | CL0: 2.8 | CL1: 2.3 | HR0: 0.359/SR0: 0.363 | HR1: 0.437/SR1: 0.413 | LR: 9.83e-05 [2026-05-23 18:10:49] Epoch 1 | Step 1390 | Loss: 0.8224 | LM: 0.3216 | LB: 1.0073 | CL0: 2.8 | CL1: 2.3 | HR0: 0.359/SR0: 0.363 | HR1: 0.437/SR1: 0.413 | LR: 9.90e-05 [2026-05-23 18:11:02] Epoch 1 | Step 1400 | Loss: 0.8223 | LM: 0.3219 | LB: 1.0073 | CL0: 2.8 | CL1: 2.3 | HR0: 0.359/SR0: 0.363 | HR1: 0.437/SR1: 0.413 | LR: 9.96e-05 [2026-05-23 18:11:14] Epoch 1 | Step 1410 | Loss: 0.8221 | LM: 0.3223 | LB: 1.0072 | CL0: 2.8 | CL1: 2.3 | HR0: 0.359/SR0: 0.363 | HR1: 0.437/SR1: 0.412 | LR: 1.00e-04 [2026-05-23 18:11:26] Epoch 1 | Step 1420 | Loss: 0.8220 | LM: 0.3220 | LB: 1.0072 | CL0: 2.8 | CL1: 2.3 | HR0: 0.359/SR0: 0.363 | HR1: 0.436/SR1: 0.412 | LR: 1.00e-04 [2026-05-23 18:11:39] Epoch 1 | Step 1430 | Loss: 0.8216 | LM: 0.3215 | LB: 1.0072 | CL0: 2.8 | CL1: 2.3 | HR0: 0.360/SR0: 0.364 | HR1: 0.436/SR1: 0.412 | LR: 1.00e-04 [2026-05-23 18:11:51] Epoch 1 | Step 1440 | Loss: 0.8214 | LM: 0.3212 | LB: 1.0071 | CL0: 2.8 | CL1: 2.3 | HR0: 0.360/SR0: 0.364 | HR1: 0.436/SR1: 0.412 | LR: 1.00e-04 [2026-05-23 18:12:03] Epoch 1 | Step 1450 | Loss: 0.8213 | LM: 0.3213 | LB: 1.0071 | CL0: 2.8 | CL1: 2.3 | HR0: 0.360/SR0: 0.364 | HR1: 0.436/SR1: 0.411 | LR: 1.00e-04 [2026-05-23 18:12:16] Epoch 1 | Step 1460 | Loss: 0.8212 | LM: 0.3214 | LB: 1.0071 | CL0: 2.8 | CL1: 2.3 | HR0: 0.360/SR0: 0.364 | HR1: 0.436/SR1: 0.411 | LR: 1.00e-04 [2026-05-23 18:12:29] Epoch 1 | Step 1470 | Loss: 0.8210 | LM: 0.3210 | LB: 1.0070 | CL0: 2.8 | CL1: 2.3 | HR0: 0.360/SR0: 0.364 | HR1: 0.435/SR1: 0.411 | LR: 1.00e-04 [2026-05-23 18:12:41] Epoch 1 | Step 1480 | Loss: 0.8209 | LM: 0.3210 | LB: 1.0070 | CL0: 2.8 | CL1: 2.3 | HR0: 0.360/SR0: 0.364 | HR1: 0.435/SR1: 0.411 | LR: 1.00e-04 [2026-05-23 18:12:53] Epoch 1 | Step 1490 | Loss: 0.8208 | LM: 0.3211 | LB: 1.0070 | CL0: 2.8 | CL1: 2.3 | HR0: 0.360/SR0: 0.364 | HR1: 0.435/SR1: 0.411 | LR: 1.00e-04 [2026-05-23 18:13:05] Epoch 1 | Step 1500 | Loss: 0.8207 | LM: 0.3208 | LB: 1.0070 | CL0: 2.8 | CL1: 2.3 | HR0: 0.360/SR0: 0.364 | HR1: 0.435/SR1: 0.410 | LR: 1.00e-04 [2026-05-23 18:13:16] Epoch 1 | Step 1510 | Loss: 0.8205 | LM: 0.3206 | LB: 1.0069 | CL0: 2.8 | CL1: 2.3 | HR0: 0.360/SR0: 0.365 | HR1: 0.435/SR1: 0.410 | LR: 1.00e-04 [2026-05-23 18:13:27] Epoch 1 | Step 1520 | Loss: 0.8201 | LM: 0.3201 | LB: 1.0069 | CL0: 2.8 | CL1: 2.3 | HR0: 0.360/SR0: 0.365 | HR1: 0.434/SR1: 0.410 | LR: 1.00e-04 [2026-05-23 18:13:39] Epoch 1 | Step 1530 | Loss: 0.8198 | LM: 0.3196 | LB: 1.0069 | CL0: 2.8 | CL1: 2.3 | HR0: 0.360/SR0: 0.365 | HR1: 0.434/SR1: 0.410 | LR: 1.00e-04 [2026-05-23 18:13:51] Epoch 1 | Step 1540 | Loss: 0.8196 | LM: 0.3192 | LB: 1.0068 | CL0: 2.8 | CL1: 2.3 | HR0: 0.360/SR0: 0.365 | HR1: 0.434/SR1: 0.410 | LR: 1.00e-04 [2026-05-23 18:14:03] Epoch 1 | Step 1550 | Loss: 0.8195 | LM: 0.3191 | LB: 1.0068 | CL0: 2.8 | CL1: 2.3 | HR0: 0.360/SR0: 0.365 | HR1: 0.434/SR1: 0.409 | LR: 1.00e-04 [2026-05-23 18:14:15] Epoch 1 | Step 1560 | Loss: 0.8195 | LM: 0.3190 | LB: 1.0068 | CL0: 2.8 | CL1: 2.3 | HR0: 0.360/SR0: 0.365 | HR1: 0.434/SR1: 0.409 | LR: 1.00e-04 [2026-05-23 18:14:27] Epoch 1 | Step 1570 | Loss: 0.8193 | LM: 0.3184 | LB: 1.0068 | CL0: 2.8 | CL1: 2.3 | HR0: 0.360/SR0: 0.365 | HR1: 0.433/SR1: 0.409 | LR: 1.00e-04 [2026-05-23 18:14:39] Epoch 1 | Step 1580 | Loss: 0.8192 | LM: 0.3185 | LB: 1.0067 | CL0: 2.8 | CL1: 2.3 | HR0: 0.360/SR0: 0.366 | HR1: 0.433/SR1: 0.409 | LR: 1.00e-04 [2026-05-23 18:14:51] Epoch 1 | Step 1590 | Loss: 0.8189 | LM: 0.3182 | LB: 1.0067 | CL0: 2.8 | CL1: 2.3 | HR0: 0.360/SR0: 0.366 | HR1: 0.433/SR1: 0.408 | LR: 1.00e-04 [2026-05-23 18:15:03] Epoch 1 | Step 1600 | Loss: 0.8186 | LM: 0.3183 | LB: 1.0067 | CL0: 2.8 | CL1: 2.3 | HR0: 0.360/SR0: 0.366 | HR1: 0.433/SR1: 0.408 | LR: 1.00e-04 [2026-05-23 18:15:16] Epoch 1 | Step 1610 | Loss: 0.8185 | LM: 0.3183 | LB: 1.0067 | CL0: 2.8 | CL1: 2.3 | HR0: 0.360/SR0: 0.366 | HR1: 0.433/SR1: 0.408 | LR: 1.00e-04 [2026-05-23 18:15:28] Epoch 1 | Step 1620 | Loss: 0.8181 | LM: 0.3184 | LB: 1.0066 | CL0: 2.8 | CL1: 2.3 | HR0: 0.361/SR0: 0.366 | HR1: 0.432/SR1: 0.408 | LR: 1.00e-04 [2026-05-23 18:15:41] Epoch 1 | Step 1630 | Loss: 0.8180 | LM: 0.3181 | LB: 1.0066 | CL0: 2.8 | CL1: 2.3 | HR0: 0.361/SR0: 0.366 | HR1: 0.432/SR1: 0.408 | LR: 1.00e-04 [2026-05-23 18:15:53] Epoch 1 | Step 1640 | Loss: 0.8179 | LM: 0.3179 | LB: 1.0066 | CL0: 2.8 | CL1: 2.3 | HR0: 0.361/SR0: 0.367 | HR1: 0.432/SR1: 0.407 | LR: 1.00e-04 [2026-05-23 18:16:05] Epoch 1 | Step 1650 | Loss: 0.8177 | LM: 0.3174 | LB: 1.0065 | CL0: 2.8 | CL1: 2.3 | HR0: 0.361/SR0: 0.367 | HR1: 0.432/SR1: 0.407 | LR: 1.00e-04 [2026-05-23 18:16:16] Epoch 1 | Step 1660 | Loss: 0.8174 | LM: 0.3170 | LB: 1.0065 | CL0: 2.8 | CL1: 2.3 | HR0: 0.361/SR0: 0.367 | HR1: 0.432/SR1: 0.407 | LR: 1.00e-04 [2026-05-23 18:16:29] Epoch 1 | Step 1670 | Loss: 0.8173 | LM: 0.3168 | LB: 1.0065 | CL0: 2.8 | CL1: 2.3 | HR0: 0.361/SR0: 0.367 | HR1: 0.431/SR1: 0.407 | LR: 1.00e-04 [2026-05-23 18:16:40] Epoch 1 | Step 1680 | Loss: 0.8170 | LM: 0.3165 | LB: 1.0065 | CL0: 2.8 | CL1: 2.3 | HR0: 0.361/SR0: 0.367 | HR1: 0.431/SR1: 0.407 | LR: 1.00e-04 [2026-05-23 18:16:52] Epoch 1 | Step 1690 | Loss: 0.8169 | LM: 0.3165 | LB: 1.0064 | CL0: 2.8 | CL1: 2.3 | HR0: 0.361/SR0: 0.367 | HR1: 0.431/SR1: 0.407 | LR: 1.00e-04 [2026-05-23 18:17:04] Epoch 1 | Step 1700 | Loss: 0.8165 | LM: 0.3160 | LB: 1.0064 | CL0: 2.8 | CL1: 2.3 | HR0: 0.361/SR0: 0.368 | HR1: 0.431/SR1: 0.406 | LR: 1.00e-04 [2026-05-23 18:17:16] Epoch 1 | Step 1710 | Loss: 0.8163 | LM: 0.3158 | LB: 1.0064 | CL0: 2.8 | CL1: 2.3 | HR0: 0.361/SR0: 0.368 | HR1: 0.431/SR1: 0.406 | LR: 1.00e-04 [2026-05-23 18:17:29] Epoch 1 | Step 1720 | Loss: 0.8163 | LM: 0.3157 | LB: 1.0063 | CL0: 2.8 | CL1: 2.3 | HR0: 0.361/SR0: 0.368 | HR1: 0.431/SR1: 0.406 | LR: 1.00e-04 [2026-05-23 18:17:40] Epoch 1 | Step 1730 | Loss: 0.8163 | LM: 0.3155 | LB: 1.0063 | CL0: 2.8 | CL1: 2.3 | HR0: 0.361/SR0: 0.368 | HR1: 0.431/SR1: 0.406 | LR: 1.00e-04 [2026-05-23 18:17:52] Epoch 1 | Step 1740 | Loss: 0.8159 | LM: 0.3153 | LB: 1.0063 | CL0: 2.8 | CL1: 2.3 | HR0: 0.361/SR0: 0.368 | HR1: 0.430/SR1: 0.406 | LR: 1.00e-04 [2026-05-23 18:18:04] Epoch 1 | Step 1750 | Loss: 0.8157 | LM: 0.3153 | LB: 1.0063 | CL0: 2.8 | CL1: 2.3 | HR0: 0.361/SR0: 0.368 | HR1: 0.430/SR1: 0.406 | LR: 1.00e-04 [2026-05-23 18:18:17] Epoch 1 | Step 1760 | Loss: 0.8154 | LM: 0.3151 | LB: 1.0062 | CL0: 2.8 | CL1: 2.3 | HR0: 0.361/SR0: 0.368 | HR1: 0.430/SR1: 0.405 | LR: 1.00e-04 [2026-05-23 18:18:28] Epoch 1 | Step 1770 | Loss: 0.8153 | LM: 0.3153 | LB: 1.0062 | CL0: 2.8 | CL1: 2.3 | HR0: 0.361/SR0: 0.369 | HR1: 0.430/SR1: 0.405 | LR: 1.00e-04 [2026-05-23 18:18:40] Epoch 1 | Step 1780 | Loss: 0.8152 | LM: 0.3153 | LB: 1.0062 | CL0: 2.8 | CL1: 2.3 | HR0: 0.362/SR0: 0.369 | HR1: 0.430/SR1: 0.405 | LR: 1.00e-04 [2026-05-23 18:18:53] Epoch 1 | Step 1790 | Loss: 0.8152 | LM: 0.3153 | LB: 1.0062 | CL0: 2.8 | CL1: 2.3 | HR0: 0.362/SR0: 0.369 | HR1: 0.430/SR1: 0.405 | LR: 1.00e-04 [2026-05-23 18:19:05] Epoch 1 | Step 1800 | Loss: 0.8150 | LM: 0.3154 | LB: 1.0062 | CL0: 2.8 | CL1: 2.3 | HR0: 0.362/SR0: 0.369 | HR1: 0.429/SR1: 0.405 | LR: 1.00e-04 [2026-05-23 18:19:17] Epoch 1 | Step 1810 | Loss: 0.8148 | LM: 0.3148 | LB: 1.0061 | CL0: 2.8 | CL1: 2.3 | HR0: 0.362/SR0: 0.369 | HR1: 0.429/SR1: 0.405 | LR: 1.00e-04 [2026-05-23 18:19:29] Epoch 1 | Step 1820 | Loss: 0.8147 | LM: 0.3149 | LB: 1.0061 | CL0: 2.8 | CL1: 2.3 | HR0: 0.362/SR0: 0.369 | HR1: 0.429/SR1: 0.404 | LR: 1.00e-04 [2026-05-23 18:19:41] Epoch 1 | Step 1830 | Loss: 0.8146 | LM: 0.3149 | LB: 1.0061 | CL0: 2.8 | CL1: 2.3 | HR0: 0.362/SR0: 0.369 | HR1: 0.429/SR1: 0.404 | LR: 1.00e-04 [2026-05-23 18:19:53] Epoch 1 | Step 1840 | Loss: 0.8144 | LM: 0.3149 | LB: 1.0061 | CL0: 2.8 | CL1: 2.3 | HR0: 0.362/SR0: 0.370 | HR1: 0.429/SR1: 0.404 | LR: 1.00e-04 [2026-05-23 18:20:06] Epoch 1 | Step 1850 | Loss: 0.8144 | LM: 0.3148 | LB: 1.0060 | CL0: 2.8 | CL1: 2.3 | HR0: 0.362/SR0: 0.370 | HR1: 0.429/SR1: 0.404 | LR: 1.00e-04 [2026-05-23 18:20:18] Epoch 1 | Step 1860 | Loss: 0.8143 | LM: 0.3144 | LB: 1.0060 | CL0: 2.8 | CL1: 2.3 | HR0: 0.362/SR0: 0.370 | HR1: 0.429/SR1: 0.404 | LR: 1.00e-04 [2026-05-23 18:20:31] Epoch 1 | Step 1870 | Loss: 0.8142 | LM: 0.3143 | LB: 1.0060 | CL0: 2.8 | CL1: 2.3 | HR0: 0.362/SR0: 0.370 | HR1: 0.428/SR1: 0.404 | LR: 1.00e-04 [2026-05-23 18:20:42] Epoch 1 | Step 1880 | Loss: 0.8140 | LM: 0.3141 | LB: 1.0060 | CL0: 2.8 | CL1: 2.3 | HR0: 0.362/SR0: 0.370 | HR1: 0.428/SR1: 0.404 | LR: 1.00e-04 [2026-05-23 18:20:54] Epoch 1 | Step 1890 | Loss: 0.8139 | LM: 0.3141 | LB: 1.0059 | CL0: 2.8 | CL1: 2.3 | HR0: 0.362/SR0: 0.370 | HR1: 0.428/SR1: 0.404 | LR: 1.00e-04 [2026-05-23 18:21:07] Epoch 1 | Step 1900 | Loss: 0.8139 | LM: 0.3142 | LB: 1.0059 | CL0: 2.8 | CL1: 2.4 | HR0: 0.362/SR0: 0.371 | HR1: 0.428/SR1: 0.403 | LR: 1.00e-04 [2026-05-23 18:21:19] Epoch 1 | Step 1910 | Loss: 0.8138 | LM: 0.3145 | LB: 1.0059 | CL0: 2.8 | CL1: 2.4 | HR0: 0.362/SR0: 0.371 | HR1: 0.428/SR1: 0.403 | LR: 1.00e-04 [2026-05-23 18:21:31] Epoch 1 | Step 1920 | Loss: 0.8135 | LM: 0.3148 | LB: 1.0059 | CL0: 2.8 | CL1: 2.4 | HR0: 0.362/SR0: 0.371 | HR1: 0.428/SR1: 0.403 | LR: 1.00e-04 [2026-05-23 18:21:43] Epoch 1 | Step 1930 | Loss: 0.8135 | LM: 0.3147 | LB: 1.0058 | CL0: 2.8 | CL1: 2.4 | HR0: 0.363/SR0: 0.371 | HR1: 0.428/SR1: 0.403 | LR: 1.00e-04 [2026-05-23 18:21:54] Epoch 1 | Step 1940 | Loss: 0.8134 | LM: 0.3148 | LB: 1.0058 | CL0: 2.8 | CL1: 2.4 | HR0: 0.363/SR0: 0.371 | HR1: 0.427/SR1: 0.403 | LR: 1.00e-04 [2026-05-23 18:22:07] Epoch 1 | Step 1950 | Loss: 0.8134 | LM: 0.3147 | LB: 1.0058 | CL0: 2.8 | CL1: 2.4 | HR0: 0.363/SR0: 0.371 | HR1: 0.427/SR1: 0.403 | LR: 1.00e-04 [2026-05-23 18:22:19] Epoch 1 | Step 1960 | Loss: 0.8132 | LM: 0.3145 | LB: 1.0058 | CL0: 2.8 | CL1: 2.4 | HR0: 0.363/SR0: 0.372 | HR1: 0.427/SR1: 0.403 | LR: 1.00e-04 [2026-05-23 18:22:31] Epoch 1 | Step 1970 | Loss: 0.8129 | LM: 0.3144 | LB: 1.0057 | CL0: 2.8 | CL1: 2.4 | HR0: 0.363/SR0: 0.372 | HR1: 0.427/SR1: 0.402 | LR: 1.00e-04 [2026-05-23 18:22:43] Epoch 1 | Step 1980 | Loss: 0.8127 | LM: 0.3142 | LB: 1.0057 | CL0: 2.8 | CL1: 2.4 | HR0: 0.363/SR0: 0.372 | HR1: 0.427/SR1: 0.402 | LR: 1.00e-04 [2026-05-23 18:22:55] Epoch 1 | Step 1990 | Loss: 0.8126 | LM: 0.3143 | LB: 1.0057 | CL0: 2.8 | CL1: 2.4 | HR0: 0.363/SR0: 0.372 | HR1: 0.427/SR1: 0.402 | LR: 1.00e-04 [2026-05-23 18:23:06] Epoch 1 | Step 2000 | Loss: 0.8124 | LM: 0.3140 | LB: 1.0057 | CL0: 2.8 | CL1: 2.4 | HR0: 0.363/SR0: 0.372 | HR1: 0.427/SR1: 0.402 | LR: 1.00e-04 [2026-05-23 18:23:09] Validation | Batch 10/1463 | Loss: 0.7572 | LM_LOSS: 0.2567 | LB_LOSS: 1.0009 [2026-05-23 18:23:11] Validation | Batch 20/1463 | Loss: 0.7655 | LM_LOSS: 0.2653 | LB_LOSS: 1.0005 [2026-05-23 18:23:14] Validation | Batch 30/1463 | Loss: 0.7717 | LM_LOSS: 0.2715 | LB_LOSS: 1.0005 [2026-05-23 18:23:17] Validation | Batch 40/1463 | Loss: 0.7787 | LM_LOSS: 0.2785 | LB_LOSS: 1.0004 [2026-05-23 18:23:19] Validation | Batch 50/1463 | Loss: 0.7864 | LM_LOSS: 0.2862 | LB_LOSS: 1.0004 [2026-05-23 18:23:23] Validation | Batch 60/1463 | Loss: 0.7861 | LM_LOSS: 0.2858 | LB_LOSS: 1.0005 [2026-05-23 18:23:26] Validation | Batch 70/1463 | Loss: 0.7861 | LM_LOSS: 0.2858 | LB_LOSS: 1.0004 [2026-05-23 18:23:29] Validation | Batch 80/1463 | Loss: 0.7878 | LM_LOSS: 0.2876 | LB_LOSS: 1.0005 [2026-05-23 18:23:32] Validation | Batch 90/1463 | Loss: 0.7870 | LM_LOSS: 0.2868 | LB_LOSS: 1.0004 [2026-05-23 18:23:34] Validation | Batch 100/1463 | Loss: 0.7868 | LM_LOSS: 0.2866 | LB_LOSS: 1.0004 [2026-05-23 18:23:38] Validation | Batch 110/1463 | Loss: 0.7856 | LM_LOSS: 0.2854 | LB_LOSS: 1.0004 [2026-05-23 18:23:40] Validation | Batch 120/1463 | Loss: 0.7869 | LM_LOSS: 0.2867 | LB_LOSS: 1.0004 [2026-05-23 18:23:42] Validation | Batch 130/1463 | Loss: 0.7884 | LM_LOSS: 0.2882 | LB_LOSS: 1.0004 [2026-05-23 18:23:45] Validation | Batch 140/1463 | Loss: 0.7892 | LM_LOSS: 0.2890 | LB_LOSS: 1.0004 [2026-05-23 18:23:48] Validation | Batch 150/1463 | Loss: 0.7872 | LM_LOSS: 0.2870 | LB_LOSS: 1.0005 [2026-05-23 18:23:51] Validation | Batch 160/1463 | Loss: 0.7874 | LM_LOSS: 0.2872 | LB_LOSS: 1.0005 [2026-05-23 18:23:54] Validation | Batch 170/1463 | Loss: 0.7869 | LM_LOSS: 0.2867 | LB_LOSS: 1.0005 [2026-05-23 18:23:57] Validation | Batch 180/1463 | Loss: 0.7878 | LM_LOSS: 0.2875 | LB_LOSS: 1.0005 [2026-05-23 18:24:00] Validation | Batch 190/1463 | Loss: 0.7887 | LM_LOSS: 0.2884 | LB_LOSS: 1.0005 [2026-05-23 18:24:02] Validation | Batch 200/1463 | Loss: 0.7888 | LM_LOSS: 0.2885 | LB_LOSS: 1.0005 [2026-05-23 18:24:05] Validation | Batch 210/1463 | Loss: 0.7896 | LM_LOSS: 0.2893 | LB_LOSS: 1.0005 [2026-05-23 18:24:08] Validation | Batch 220/1463 | Loss: 0.7887 | LM_LOSS: 0.2885 | LB_LOSS: 1.0005 [2026-05-23 18:24:11] Validation | Batch 230/1463 | Loss: 0.7903 | LM_LOSS: 0.2900 | LB_LOSS: 1.0005 [2026-05-23 18:24:14] Validation | Batch 240/1463 | Loss: 0.7917 | LM_LOSS: 0.2915 | LB_LOSS: 1.0005 [2026-05-23 18:24:17] Validation | Batch 250/1463 | Loss: 0.7908 | LM_LOSS: 0.2906 | LB_LOSS: 1.0005 [2026-05-23 18:24:20] Validation | Batch 260/1463 | Loss: 0.7923 | LM_LOSS: 0.2921 | LB_LOSS: 1.0005 [2026-05-23 18:24:23] Validation | Batch 270/1463 | Loss: 0.7909 | LM_LOSS: 0.2907 | LB_LOSS: 1.0005 [2026-05-23 18:24:25] Validation | Batch 280/1463 | Loss: 0.7915 | LM_LOSS: 0.2913 | LB_LOSS: 1.0005 [2026-05-23 18:24:28] Validation | Batch 290/1463 | Loss: 0.7907 | LM_LOSS: 0.2904 | LB_LOSS: 1.0005 [2026-05-23 18:24:30] Validation | Batch 300/1463 | Loss: 0.7907 | LM_LOSS: 0.2904 | LB_LOSS: 1.0005 [2026-05-23 18:24:34] Validation | Batch 310/1463 | Loss: 0.7919 | LM_LOSS: 0.2916 | LB_LOSS: 1.0005 [2026-05-23 18:24:36] Validation | Batch 320/1463 | Loss: 0.7905 | LM_LOSS: 0.2903 | LB_LOSS: 1.0005 [2026-05-23 18:24:39] Validation | Batch 330/1463 | Loss: 0.7905 | LM_LOSS: 0.2902 | LB_LOSS: 1.0005 [2026-05-23 18:24:41] Validation | Batch 340/1463 | Loss: 0.7902 | LM_LOSS: 0.2899 | LB_LOSS: 1.0005 [2026-05-23 18:24:44] Validation | Batch 350/1463 | Loss: 0.7908 | LM_LOSS: 0.2905 | LB_LOSS: 1.0005 [2026-05-23 18:24:47] Validation | Batch 360/1463 | Loss: 0.7903 | LM_LOSS: 0.2900 | LB_LOSS: 1.0005 [2026-05-23 18:24:49] Validation | Batch 370/1463 | Loss: 0.7901 | LM_LOSS: 0.2898 | LB_LOSS: 1.0005 [2026-05-23 18:24:52] Validation | Batch 380/1463 | Loss: 0.7901 | LM_LOSS: 0.2898 | LB_LOSS: 1.0005 [2026-05-23 18:24:55] Validation | Batch 390/1463 | Loss: 0.7907 | LM_LOSS: 0.2904 | LB_LOSS: 1.0005 [2026-05-23 18:24:57] Validation | Batch 400/1463 | Loss: 0.7904 | LM_LOSS: 0.2902 | LB_LOSS: 1.0005 [2026-05-23 18:25:00] Validation | Batch 410/1463 | Loss: 0.7901 | LM_LOSS: 0.2899 | LB_LOSS: 1.0005 [2026-05-23 18:25:03] Validation | Batch 420/1463 | Loss: 0.7902 | LM_LOSS: 0.2900 | LB_LOSS: 1.0005 [2026-05-23 18:25:06] Validation | Batch 430/1463 | Loss: 0.7897 | LM_LOSS: 0.2895 | LB_LOSS: 1.0005 [2026-05-23 18:25:08] Validation | Batch 440/1463 | Loss: 0.7899 | LM_LOSS: 0.2896 | LB_LOSS: 1.0005 [2026-05-23 18:25:11] Validation | Batch 450/1463 | Loss: 0.7897 | LM_LOSS: 0.2895 | LB_LOSS: 1.0005 [2026-05-23 18:25:14] Validation | Batch 460/1463 | Loss: 0.7897 | LM_LOSS: 0.2894 | LB_LOSS: 1.0005 [2026-05-23 18:25:17] Validation | Batch 470/1463 | Loss: 0.7903 | LM_LOSS: 0.2901 | LB_LOSS: 1.0005 [2026-05-23 18:25:20] Validation | Batch 480/1463 | Loss: 0.7901 | LM_LOSS: 0.2899 | LB_LOSS: 1.0005 [2026-05-23 18:25:23] Validation | Batch 490/1463 | Loss: 0.7900 | LM_LOSS: 0.2898 | LB_LOSS: 1.0005 [2026-05-23 18:25:26] Validation | Batch 500/1463 | Loss: 0.7899 | LM_LOSS: 0.2897 | LB_LOSS: 1.0005 [2026-05-23 18:25:29] Validation | Batch 510/1463 | Loss: 0.7893 | LM_LOSS: 0.2891 | LB_LOSS: 1.0005 [2026-05-23 18:25:31] Validation | Batch 520/1463 | Loss: 0.7891 | LM_LOSS: 0.2888 | LB_LOSS: 1.0005 [2026-05-23 18:25:34] Validation | Batch 530/1463 | Loss: 0.7894 | LM_LOSS: 0.2891 | LB_LOSS: 1.0005 [2026-05-23 18:25:36] Validation | Batch 540/1463 | Loss: 0.7896 | LM_LOSS: 0.2894 | LB_LOSS: 1.0005 [2026-05-23 18:25:38] Validation | Batch 550/1463 | Loss: 0.7893 | LM_LOSS: 0.2891 | LB_LOSS: 1.0004 [2026-05-23 18:25:41] Validation | Batch 560/1463 | Loss: 0.7895 | LM_LOSS: 0.2893 | LB_LOSS: 1.0004 [2026-05-23 18:25:44] Validation | Batch 570/1463 | Loss: 0.7894 | LM_LOSS: 0.2892 | LB_LOSS: 1.0004 [2026-05-23 18:25:47] Validation | Batch 580/1463 | Loss: 0.7892 | LM_LOSS: 0.2889 | LB_LOSS: 1.0004 [2026-05-23 18:25:50] Validation | Batch 590/1463 | Loss: 0.7889 | LM_LOSS: 0.2887 | LB_LOSS: 1.0004 [2026-05-23 18:25:53] Validation | Batch 600/1463 | Loss: 0.7892 | LM_LOSS: 0.2890 | LB_LOSS: 1.0004 [2026-05-23 18:25:57] Validation | Batch 610/1463 | Loss: 0.7894 | LM_LOSS: 0.2891 | LB_LOSS: 1.0004 [2026-05-23 18:25:59] Validation | Batch 620/1463 | Loss: 0.7889 | LM_LOSS: 0.2887 | LB_LOSS: 1.0004 [2026-05-23 18:26:02] Validation | Batch 630/1463 | Loss: 0.7884 | LM_LOSS: 0.2882 | LB_LOSS: 1.0004 [2026-05-23 18:26:05] Validation | Batch 640/1463 | Loss: 0.7881 | LM_LOSS: 0.2879 | LB_LOSS: 1.0004 [2026-05-23 18:26:07] Validation | Batch 650/1463 | Loss: 0.7877 | LM_LOSS: 0.2874 | LB_LOSS: 1.0004 [2026-05-23 18:26:10] Validation | Batch 660/1463 | Loss: 0.7873 | LM_LOSS: 0.2871 | LB_LOSS: 1.0004 [2026-05-23 18:26:12] Validation | Batch 670/1463 | Loss: 0.7870 | LM_LOSS: 0.2868 | LB_LOSS: 1.0004 [2026-05-23 18:26:15] Validation | Batch 680/1463 | Loss: 0.7870 | LM_LOSS: 0.2868 | LB_LOSS: 1.0004 [2026-05-23 18:26:18] Validation | Batch 690/1463 | Loss: 0.7873 | LM_LOSS: 0.2871 | LB_LOSS: 1.0004 [2026-05-23 18:26:20] Validation | Batch 700/1463 | Loss: 0.7872 | LM_LOSS: 0.2869 | LB_LOSS: 1.0004 [2026-05-23 18:26:23] Validation | Batch 710/1463 | Loss: 0.7876 | LM_LOSS: 0.2873 | LB_LOSS: 1.0004 [2026-05-23 18:26:26] Validation | Batch 720/1463 | Loss: 0.7883 | LM_LOSS: 0.2881 | LB_LOSS: 1.0004 [2026-05-23 18:26:28] Validation | Batch 730/1463 | Loss: 0.7881 | LM_LOSS: 0.2879 | LB_LOSS: 1.0004 [2026-05-23 18:26:30] Validation | Batch 740/1463 | Loss: 0.7879 | LM_LOSS: 0.2877 | LB_LOSS: 1.0004 [2026-05-23 18:26:33] Validation | Batch 750/1463 | Loss: 0.7879 | LM_LOSS: 0.2876 | LB_LOSS: 1.0004 [2026-05-23 18:26:35] Validation | Batch 760/1463 | Loss: 0.7877 | LM_LOSS: 0.2875 | LB_LOSS: 1.0005 [2026-05-23 18:26:38] Validation | Batch 770/1463 | Loss: 0.7876 | LM_LOSS: 0.2874 | LB_LOSS: 1.0005 [2026-05-23 18:26:40] Validation | Batch 780/1463 | Loss: 0.7875 | LM_LOSS: 0.2872 | LB_LOSS: 1.0005 [2026-05-23 18:26:42] Validation | Batch 790/1463 | Loss: 0.7877 | LM_LOSS: 0.2875 | LB_LOSS: 1.0005 [2026-05-23 18:26:45] Validation | Batch 800/1463 | Loss: 0.7876 | LM_LOSS: 0.2874 | LB_LOSS: 1.0005 [2026-05-23 18:26:47] Validation | Batch 810/1463 | Loss: 0.7873 | LM_LOSS: 0.2870 | LB_LOSS: 1.0004 [2026-05-23 18:26:50] Validation | Batch 820/1463 | Loss: 0.7875 | LM_LOSS: 0.2873 | LB_LOSS: 1.0005 [2026-05-23 18:26:53] Validation | Batch 830/1463 | Loss: 0.7878 | LM_LOSS: 0.2876 | LB_LOSS: 1.0004 [2026-05-23 18:26:56] Validation | Batch 840/1463 | Loss: 0.7879 | LM_LOSS: 0.2877 | LB_LOSS: 1.0005 [2026-05-23 18:26:59] Validation | Batch 850/1463 | Loss: 0.7881 | LM_LOSS: 0.2879 | LB_LOSS: 1.0005 [2026-05-23 18:27:02] Validation | Batch 860/1463 | Loss: 0.7882 | LM_LOSS: 0.2880 | LB_LOSS: 1.0005 [2026-05-23 18:27:04] Validation | Batch 870/1463 | Loss: 0.7882 | LM_LOSS: 0.2879 | LB_LOSS: 1.0005 [2026-05-23 18:27:07] Validation | Batch 880/1463 | Loss: 0.7874 | LM_LOSS: 0.2872 | LB_LOSS: 1.0005 [2026-05-23 18:27:10] Validation | Batch 890/1463 | Loss: 0.7875 | LM_LOSS: 0.2872 | LB_LOSS: 1.0005 [2026-05-23 18:27:12] Validation | Batch 900/1463 | Loss: 0.7872 | LM_LOSS: 0.2870 | LB_LOSS: 1.0005 [2026-05-23 18:27:15] Validation | Batch 910/1463 | Loss: 0.7871 | LM_LOSS: 0.2869 | LB_LOSS: 1.0004 [2026-05-23 18:27:19] Validation | Batch 920/1463 | Loss: 0.7873 | LM_LOSS: 0.2871 | LB_LOSS: 1.0005 [2026-05-23 18:27:21] Validation | Batch 930/1463 | Loss: 0.7871 | LM_LOSS: 0.2869 | LB_LOSS: 1.0005 [2026-05-23 18:27:24] Validation | Batch 940/1463 | Loss: 0.7868 | LM_LOSS: 0.2866 | LB_LOSS: 1.0005 [2026-05-23 18:27:27] Validation | Batch 950/1463 | Loss: 0.7871 | LM_LOSS: 0.2869 | LB_LOSS: 1.0005 [2026-05-23 18:27:30] Validation | Batch 960/1463 | Loss: 0.7872 | LM_LOSS: 0.2869 | LB_LOSS: 1.0005 [2026-05-23 18:27:33] Validation | Batch 970/1463 | Loss: 0.7872 | LM_LOSS: 0.2869 | LB_LOSS: 1.0005 [2026-05-23 18:27:36] Validation | Batch 980/1463 | Loss: 0.7875 | LM_LOSS: 0.2873 | LB_LOSS: 1.0005 [2026-05-23 18:27:38] Validation | Batch 990/1463 | Loss: 0.7877 | LM_LOSS: 0.2875 | LB_LOSS: 1.0005 [2026-05-23 18:27:41] Validation | Batch 1000/1463 | Loss: 0.7883 | LM_LOSS: 0.2881 | LB_LOSS: 1.0005 [2026-05-23 18:27:44] Validation | Batch 1010/1463 | Loss: 0.7884 | LM_LOSS: 0.2881 | LB_LOSS: 1.0005 [2026-05-23 18:27:47] Validation | Batch 1020/1463 | Loss: 0.7884 | LM_LOSS: 0.2882 | LB_LOSS: 1.0005 [2026-05-23 18:27:50] Validation | Batch 1030/1463 | Loss: 0.7888 | LM_LOSS: 0.2886 | LB_LOSS: 1.0005 [2026-05-23 18:27:52] Validation | Batch 1040/1463 | Loss: 0.7885 | LM_LOSS: 0.2883 | LB_LOSS: 1.0005 [2026-05-23 18:27:55] Validation | Batch 1050/1463 | Loss: 0.7884 | LM_LOSS: 0.2881 | LB_LOSS: 1.0005 [2026-05-23 18:27:58] Validation | Batch 1060/1463 | Loss: 0.7883 | LM_LOSS: 0.2881 | LB_LOSS: 1.0005 [2026-05-23 18:28:01] Validation | Batch 1070/1463 | Loss: 0.7883 | LM_LOSS: 0.2881 | LB_LOSS: 1.0005 [2026-05-23 18:28:04] Validation | Batch 1080/1463 | Loss: 0.7881 | LM_LOSS: 0.2879 | LB_LOSS: 1.0005 [2026-05-23 18:28:06] Validation | Batch 1090/1463 | Loss: 0.7882 | LM_LOSS: 0.2879 | LB_LOSS: 1.0005 [2026-05-23 18:28:08] Validation | Batch 1100/1463 | Loss: 0.7880 | LM_LOSS: 0.2877 | LB_LOSS: 1.0005 [2026-05-23 18:28:11] Validation | Batch 1110/1463 | Loss: 0.7880 | LM_LOSS: 0.2877 | LB_LOSS: 1.0005 [2026-05-23 18:28:14] Validation | Batch 1120/1463 | Loss: 0.7878 | LM_LOSS: 0.2876 | LB_LOSS: 1.0005 [2026-05-23 18:28:17] Validation | Batch 1130/1463 | Loss: 0.7876 | LM_LOSS: 0.2874 | LB_LOSS: 1.0005 [2026-05-23 18:28:20] Validation | Batch 1140/1463 | Loss: 0.7878 | LM_LOSS: 0.2876 | LB_LOSS: 1.0005 [2026-05-23 18:28:23] Validation | Batch 1150/1463 | Loss: 0.7877 | LM_LOSS: 0.2875 | LB_LOSS: 1.0005 [2026-05-23 18:28:25] Validation | Batch 1160/1463 | Loss: 0.7877 | LM_LOSS: 0.2875 | LB_LOSS: 1.0005 [2026-05-23 18:28:28] Validation | Batch 1170/1463 | Loss: 0.7877 | LM_LOSS: 0.2874 | LB_LOSS: 1.0005 [2026-05-23 18:28:31] Validation | Batch 1180/1463 | Loss: 0.7877 | LM_LOSS: 0.2875 | LB_LOSS: 1.0005 [2026-05-23 18:28:34] Validation | Batch 1190/1463 | Loss: 0.7878 | LM_LOSS: 0.2876 | LB_LOSS: 1.0005 [2026-05-23 18:28:37] Validation | Batch 1200/1463 | Loss: 0.7878 | LM_LOSS: 0.2876 | LB_LOSS: 1.0005 [2026-05-23 18:28:40] Validation | Batch 1210/1463 | Loss: 0.7880 | LM_LOSS: 0.2878 | LB_LOSS: 1.0005 [2026-05-23 18:28:42] Validation | Batch 1220/1463 | Loss: 0.7884 | LM_LOSS: 0.2881 | LB_LOSS: 1.0005 [2026-05-23 18:28:45] Validation | Batch 1230/1463 | Loss: 0.7884 | LM_LOSS: 0.2881 | LB_LOSS: 1.0005 [2026-05-23 18:28:48] Validation | Batch 1240/1463 | Loss: 0.7886 | LM_LOSS: 0.2884 | LB_LOSS: 1.0005 [2026-05-23 18:28:50] Validation | Batch 1250/1463 | Loss: 0.7888 | LM_LOSS: 0.2886 | LB_LOSS: 1.0005 [2026-05-23 18:28:54] Validation | Batch 1260/1463 | Loss: 0.7887 | LM_LOSS: 0.2885 | LB_LOSS: 1.0005 [2026-05-23 18:28:56] Validation | Batch 1270/1463 | Loss: 0.7886 | LM_LOSS: 0.2883 | LB_LOSS: 1.0005 [2026-05-23 18:28:59] Validation | Batch 1280/1463 | Loss: 0.7885 | LM_LOSS: 0.2882 | LB_LOSS: 1.0005 [2026-05-23 18:29:02] Validation | Batch 1290/1463 | Loss: 0.7885 | LM_LOSS: 0.2882 | LB_LOSS: 1.0005 [2026-05-23 18:29:05] Validation | Batch 1300/1463 | Loss: 0.7883 | LM_LOSS: 0.2881 | LB_LOSS: 1.0005 [2026-05-23 18:29:07] Validation | Batch 1310/1463 | Loss: 0.7882 | LM_LOSS: 0.2880 | LB_LOSS: 1.0005 [2026-05-23 18:29:09] Validation | Batch 1320/1463 | Loss: 0.7883 | LM_LOSS: 0.2881 | LB_LOSS: 1.0005 [2026-05-23 18:29:12] Validation | Batch 1330/1463 | Loss: 0.7883 | LM_LOSS: 0.2881 | LB_LOSS: 1.0005 [2026-05-23 18:29:14] Validation | Batch 1340/1463 | Loss: 0.7886 | LM_LOSS: 0.2883 | LB_LOSS: 1.0005 [2026-05-23 18:29:17] Validation | Batch 1350/1463 | Loss: 0.7886 | LM_LOSS: 0.2884 | LB_LOSS: 1.0005 [2026-05-23 18:29:21] Validation | Batch 1360/1463 | Loss: 0.7889 | LM_LOSS: 0.2887 | LB_LOSS: 1.0005 [2026-05-23 18:29:24] Validation | Batch 1370/1463 | Loss: 0.7891 | LM_LOSS: 0.2889 | LB_LOSS: 1.0005 [2026-05-23 18:29:26] Validation | Batch 1380/1463 | Loss: 0.7890 | LM_LOSS: 0.2887 | LB_LOSS: 1.0005 [2026-05-23 18:29:29] Validation | Batch 1390/1463 | Loss: 0.7889 | LM_LOSS: 0.2887 | LB_LOSS: 1.0005 [2026-05-23 18:29:32] Validation | Batch 1400/1463 | Loss: 0.7892 | LM_LOSS: 0.2890 | LB_LOSS: 1.0005 [2026-05-23 18:29:34] Validation | Batch 1410/1463 | Loss: 0.7894 | LM_LOSS: 0.2892 | LB_LOSS: 1.0005 [2026-05-23 18:29:37] Validation | Batch 1420/1463 | Loss: 0.7894 | LM_LOSS: 0.2891 | LB_LOSS: 1.0005 [2026-05-23 18:29:40] Validation | Batch 1430/1463 | Loss: 0.7895 | LM_LOSS: 0.2892 | LB_LOSS: 1.0005 [2026-05-23 18:29:43] Validation | Batch 1440/1463 | Loss: 0.7898 | LM_LOSS: 0.2896 | LB_LOSS: 1.0005 [2026-05-23 18:29:45] Validation | Batch 1450/1463 | Loss: 0.7896 | LM_LOSS: 0.2893 | LB_LOSS: 1.0005 [2026-05-23 18:29:48] Validation | Batch 1460/1463 | Loss: 0.7899 | LM_LOSS: 0.2897 | LB_LOSS: 1.0005 [2026-05-23 18:29:49] Validation | Batch 1463/1463 | Loss: 0.7898 | LM_LOSS: 0.2896 | LB_LOSS: 1.0005 [2026-05-23 18:29:49] Validation | Loss: 0.7898 | LM_LOSS: 0.2896 | LB_LOSS: 1.0005 | PPL: 1.34 | Time: 402.81s [2026-05-23 18:29:55] New best model saved! Val loss: 0.7898 [2026-05-23 18:30:08] Epoch 1 | Step 2010 | Loss: 0.8123 | LM: 0.3138 | LB: 1.0056 | CL0: 2.8 | CL1: 2.4 | HR0: 0.363/SR0: 0.372 | HR1: 0.427/SR1: 0.402 | LR: 1.00e-04 [2026-05-23 18:30:20] Epoch 1 | Step 2020 | Loss: 0.8123 | LM: 0.3138 | LB: 1.0056 | CL0: 2.8 | CL1: 2.4 | HR0: 0.363/SR0: 0.373 | HR1: 0.426/SR1: 0.402 | LR: 1.00e-04 [2026-05-23 18:30:32] Epoch 1 | Step 2030 | Loss: 0.8120 | LM: 0.3134 | LB: 1.0056 | CL0: 2.8 | CL1: 2.4 | HR0: 0.363/SR0: 0.373 | HR1: 0.426/SR1: 0.402 | LR: 1.00e-04 [2026-05-23 18:30:45] Epoch 1 | Step 2040 | Loss: 0.8120 | LM: 0.3138 | LB: 1.0056 | CL0: 2.8 | CL1: 2.4 | HR0: 0.363/SR0: 0.373 | HR1: 0.426/SR1: 0.402 | LR: 1.00e-04 [2026-05-23 18:30:57] Epoch 1 | Step 2050 | Loss: 0.8120 | LM: 0.3134 | LB: 1.0056 | CL0: 2.8 | CL1: 2.4 | HR0: 0.363/SR0: 0.373 | HR1: 0.426/SR1: 0.401 | LR: 1.00e-04 [2026-05-23 18:31:09] Epoch 1 | Step 2060 | Loss: 0.8118 | LM: 0.3135 | LB: 1.0055 | CL0: 2.8 | CL1: 2.4 | HR0: 0.363/SR0: 0.373 | HR1: 0.426/SR1: 0.401 | LR: 1.00e-04 [2026-05-23 18:31:20] Epoch 1 | Step 2070 | Loss: 0.8117 | LM: 0.3137 | LB: 1.0055 | CL0: 2.8 | CL1: 2.4 | HR0: 0.363/SR0: 0.373 | HR1: 0.426/SR1: 0.401 | LR: 1.00e-04 [2026-05-23 18:31:33] Epoch 1 | Step 2080 | Loss: 0.8116 | LM: 0.3138 | LB: 1.0055 | CL0: 2.8 | CL1: 2.4 | HR0: 0.364/SR0: 0.373 | HR1: 0.426/SR1: 0.401 | LR: 1.00e-04 [2026-05-23 18:31:46] Epoch 1 | Step 2090 | Loss: 0.8113 | LM: 0.3133 | LB: 1.0055 | CL0: 2.8 | CL1: 2.4 | HR0: 0.364/SR0: 0.374 | HR1: 0.426/SR1: 0.401 | LR: 1.00e-04 [2026-05-23 18:31:58] Epoch 1 | Step 2100 | Loss: 0.8113 | LM: 0.3134 | LB: 1.0055 | CL0: 2.8 | CL1: 2.4 | HR0: 0.364/SR0: 0.374 | HR1: 0.426/SR1: 0.401 | LR: 1.00e-04 [2026-05-23 18:32:10] Epoch 1 | Step 2110 | Loss: 0.8111 | LM: 0.3133 | LB: 1.0054 | CL0: 2.8 | CL1: 2.4 | HR0: 0.364/SR0: 0.374 | HR1: 0.425/SR1: 0.401 | LR: 1.00e-04 [2026-05-23 18:32:21] Epoch 1 | Step 2120 | Loss: 0.8109 | LM: 0.3130 | LB: 1.0054 | CL0: 2.8 | CL1: 2.4 | HR0: 0.364/SR0: 0.374 | HR1: 0.425/SR1: 0.401 | LR: 1.00e-04 [2026-05-23 18:32:34] Epoch 1 | Step 2130 | Loss: 0.8109 | LM: 0.3131 | LB: 1.0054 | CL0: 2.8 | CL1: 2.4 | HR0: 0.364/SR0: 0.374 | HR1: 0.425/SR1: 0.400 | LR: 1.00e-04 [2026-05-23 18:32:46] Epoch 1 | Step 2140 | Loss: 0.8106 | LM: 0.3128 | LB: 1.0054 | CL0: 2.8 | CL1: 2.4 | HR0: 0.364/SR0: 0.374 | HR1: 0.425/SR1: 0.400 | LR: 1.00e-04 [2026-05-23 18:32:58] Epoch 1 | Step 2150 | Loss: 0.8105 | LM: 0.3126 | LB: 1.0054 | CL0: 2.8 | CL1: 2.4 | HR0: 0.364/SR0: 0.375 | HR1: 0.425/SR1: 0.400 | LR: 1.00e-04 [2026-05-23 18:33:10] Epoch 1 | Step 2160 | Loss: 0.8104 | LM: 0.3125 | LB: 1.0053 | CL0: 2.8 | CL1: 2.4 | HR0: 0.364/SR0: 0.375 | HR1: 0.425/SR1: 0.400 | LR: 1.00e-04 [2026-05-23 18:33:23] Epoch 1 | Step 2170 | Loss: 0.8103 | LM: 0.3124 | LB: 1.0053 | CL0: 2.8 | CL1: 2.4 | HR0: 0.364/SR0: 0.375 | HR1: 0.425/SR1: 0.400 | LR: 1.00e-04 [2026-05-23 18:33:35] Epoch 1 | Step 2180 | Loss: 0.8101 | LM: 0.3122 | LB: 1.0053 | CL0: 2.8 | CL1: 2.4 | HR0: 0.365/SR0: 0.375 | HR1: 0.425/SR1: 0.400 | LR: 1.00e-04 [2026-05-23 18:33:48] Epoch 1 | Step 2190 | Loss: 0.8100 | LM: 0.3121 | LB: 1.0053 | CL0: 2.8 | CL1: 2.4 | HR0: 0.365/SR0: 0.375 | HR1: 0.425/SR1: 0.400 | LR: 1.00e-04 [2026-05-23 18:34:00] Epoch 1 | Step 2200 | Loss: 0.8099 | LM: 0.3119 | LB: 1.0053 | CL0: 2.8 | CL1: 2.4 | HR0: 0.365/SR0: 0.375 | HR1: 0.424/SR1: 0.400 | LR: 1.00e-04 [2026-05-23 18:34:12] Epoch 1 | Step 2210 | Loss: 0.8098 | LM: 0.3118 | LB: 1.0052 | CL0: 2.8 | CL1: 2.4 | HR0: 0.365/SR0: 0.376 | HR1: 0.424/SR1: 0.400 | LR: 1.00e-04 [2026-05-23 18:34:24] Epoch 1 | Step 2220 | Loss: 0.8096 | LM: 0.3118 | LB: 1.0052 | CL0: 2.8 | CL1: 2.4 | HR0: 0.365/SR0: 0.376 | HR1: 0.424/SR1: 0.400 | LR: 1.00e-04 [2026-05-23 18:34:37] Epoch 1 | Step 2230 | Loss: 0.8096 | LM: 0.3119 | LB: 1.0052 | CL0: 2.8 | CL1: 2.4 | HR0: 0.365/SR0: 0.376 | HR1: 0.424/SR1: 0.400 | LR: 1.00e-04 [2026-05-23 18:34:49] Epoch 1 | Step 2240 | Loss: 0.8095 | LM: 0.3116 | LB: 1.0052 | CL0: 2.8 | CL1: 2.4 | HR0: 0.365/SR0: 0.376 | HR1: 0.424/SR1: 0.399 | LR: 1.00e-04 [2026-05-23 18:35:02] Epoch 1 | Step 2250 | Loss: 0.8095 | LM: 0.3115 | LB: 1.0052 | CL0: 2.8 | CL1: 2.4 | HR0: 0.365/SR0: 0.376 | HR1: 0.424/SR1: 0.399 | LR: 1.00e-04 [2026-05-23 18:35:15] Epoch 1 | Step 2260 | Loss: 0.8093 | LM: 0.3115 | LB: 1.0051 | CL0: 2.8 | CL1: 2.4 | HR0: 0.365/SR0: 0.377 | HR1: 0.424/SR1: 0.399 | LR: 1.00e-04 [2026-05-23 18:35:27] Epoch 1 | Step 2270 | Loss: 0.8093 | LM: 0.3114 | LB: 1.0051 | CL0: 2.8 | CL1: 2.4 | HR0: 0.365/SR0: 0.377 | HR1: 0.424/SR1: 0.399 | LR: 1.00e-04 [2026-05-23 18:35:38] Epoch 1 | Step 2280 | Loss: 0.8093 | LM: 0.3115 | LB: 1.0051 | CL0: 2.7 | CL1: 2.4 | HR0: 0.365/SR0: 0.377 | HR1: 0.424/SR1: 0.399 | LR: 1.00e-04 [2026-05-23 18:35:50] Epoch 1 | Step 2290 | Loss: 0.8092 | LM: 0.3114 | LB: 1.0051 | CL0: 2.7 | CL1: 2.4 | HR0: 0.365/SR0: 0.377 | HR1: 0.424/SR1: 0.399 | LR: 1.00e-04 [2026-05-23 18:36:03] Epoch 1 | Step 2300 | Loss: 0.8093 | LM: 0.3114 | LB: 1.0051 | CL0: 2.7 | CL1: 2.4 | HR0: 0.366/SR0: 0.377 | HR1: 0.424/SR1: 0.399 | LR: 1.00e-04 [2026-05-23 18:36:15] Epoch 1 | Step 2310 | Loss: 0.8092 | LM: 0.3112 | LB: 1.0050 | CL0: 2.7 | CL1: 2.4 | HR0: 0.366/SR0: 0.377 | HR1: 0.423/SR1: 0.399 | LR: 1.00e-04 [2026-05-23 18:36:26] Epoch 1 | Step 2320 | Loss: 0.8090 | LM: 0.3109 | LB: 1.0050 | CL0: 2.7 | CL1: 2.4 | HR0: 0.366/SR0: 0.378 | HR1: 0.423/SR1: 0.399 | LR: 1.00e-04 [2026-05-23 18:36:38] Epoch 1 | Step 2330 | Loss: 0.8090 | LM: 0.3107 | LB: 1.0050 | CL0: 2.7 | CL1: 2.4 | HR0: 0.366/SR0: 0.378 | HR1: 0.423/SR1: 0.399 | LR: 1.00e-04 [2026-05-23 18:36:50] Epoch 1 | Step 2340 | Loss: 0.8090 | LM: 0.3107 | LB: 1.0050 | CL0: 2.7 | CL1: 2.4 | HR0: 0.366/SR0: 0.378 | HR1: 0.423/SR1: 0.399 | LR: 1.00e-04 [2026-05-23 18:37:02] Epoch 1 | Step 2350 | Loss: 0.8089 | LM: 0.3104 | LB: 1.0050 | CL0: 2.7 | CL1: 2.4 | HR0: 0.366/SR0: 0.378 | HR1: 0.423/SR1: 0.398 | LR: 1.00e-04 [2026-05-23 18:37:14] Epoch 1 | Step 2360 | Loss: 0.8089 | LM: 0.3103 | LB: 1.0050 | CL0: 2.7 | CL1: 2.4 | HR0: 0.366/SR0: 0.378 | HR1: 0.423/SR1: 0.398 | LR: 1.00e-04 [2026-05-23 18:37:26] Epoch 1 | Step 2370 | Loss: 0.8086 | LM: 0.3101 | LB: 1.0049 | CL0: 2.7 | CL1: 2.4 | HR0: 0.366/SR0: 0.378 | HR1: 0.423/SR1: 0.398 | LR: 1.00e-04 [2026-05-23 18:37:39] Epoch 1 | Step 2380 | Loss: 0.8086 | LM: 0.3101 | LB: 1.0049 | CL0: 2.7 | CL1: 2.4 | HR0: 0.366/SR0: 0.379 | HR1: 0.423/SR1: 0.398 | LR: 1.00e-04 [2026-05-23 18:37:51] Epoch 1 | Step 2390 | Loss: 0.8085 | LM: 0.3100 | LB: 1.0049 | CL0: 2.7 | CL1: 2.4 | HR0: 0.366/SR0: 0.379 | HR1: 0.423/SR1: 0.398 | LR: 1.00e-04 [2026-05-23 18:38:04] Epoch 1 | Step 2400 | Loss: 0.8084 | LM: 0.3097 | LB: 1.0049 | CL0: 2.7 | CL1: 2.4 | HR0: 0.366/SR0: 0.379 | HR1: 0.423/SR1: 0.398 | LR: 1.00e-04 [2026-05-23 18:38:17] Epoch 1 | Step 2410 | Loss: 0.8082 | LM: 0.3097 | LB: 1.0049 | CL0: 2.7 | CL1: 2.4 | HR0: 0.366/SR0: 0.379 | HR1: 0.423/SR1: 0.398 | LR: 1.00e-04 [2026-05-23 18:38:29] Epoch 1 | Step 2420 | Loss: 0.8081 | LM: 0.3095 | LB: 1.0049 | CL0: 2.7 | CL1: 2.4 | HR0: 0.367/SR0: 0.379 | HR1: 0.422/SR1: 0.398 | LR: 1.00e-04 [2026-05-23 18:38:42] Epoch 1 | Step 2430 | Loss: 0.8081 | LM: 0.3096 | LB: 1.0048 | CL0: 2.7 | CL1: 2.4 | HR0: 0.367/SR0: 0.379 | HR1: 0.422/SR1: 0.398 | LR: 1.00e-04 [2026-05-23 18:38:54] Epoch 1 | Step 2440 | Loss: 0.8081 | LM: 0.3094 | LB: 1.0048 | CL0: 2.7 | CL1: 2.4 | HR0: 0.367/SR0: 0.380 | HR1: 0.422/SR1: 0.398 | LR: 1.00e-04 [2026-05-23 18:39:07] Epoch 1 | Step 2450 | Loss: 0.8081 | LM: 0.3093 | LB: 1.0048 | CL0: 2.7 | CL1: 2.4 | HR0: 0.367/SR0: 0.380 | HR1: 0.422/SR1: 0.398 | LR: 1.00e-04 [2026-05-23 18:39:19] Epoch 1 | Step 2460 | Loss: 0.8078 | LM: 0.3091 | LB: 1.0048 | CL0: 2.7 | CL1: 2.4 | HR0: 0.367/SR0: 0.380 | HR1: 0.422/SR1: 0.397 | LR: 1.00e-04 [2026-05-23 18:39:31] Epoch 1 | Step 2470 | Loss: 0.8076 | LM: 0.3087 | LB: 1.0048 | CL0: 2.7 | CL1: 2.4 | HR0: 0.367/SR0: 0.380 | HR1: 0.422/SR1: 0.397 | LR: 1.00e-04 [2026-05-23 18:39:44] Epoch 1 | Step 2480 | Loss: 0.8075 | LM: 0.3088 | LB: 1.0048 | CL0: 2.7 | CL1: 2.4 | HR0: 0.367/SR0: 0.380 | HR1: 0.422/SR1: 0.397 | LR: 1.00e-04 [2026-05-23 18:39:57] Epoch 1 | Step 2490 | Loss: 0.8075 | LM: 0.3088 | LB: 1.0047 | CL0: 2.7 | CL1: 2.4 | HR0: 0.367/SR0: 0.381 | HR1: 0.422/SR1: 0.397 | LR: 1.00e-04 [2026-05-23 18:40:09] Epoch 1 | Step 2500 | Loss: 0.8073 | LM: 0.3084 | LB: 1.0047 | CL0: 2.7 | CL1: 2.4 | HR0: 0.367/SR0: 0.381 | HR1: 0.422/SR1: 0.397 | LR: 1.00e-04 [2026-05-23 18:40:20] Epoch 1 | Step 2510 | Loss: 0.8072 | LM: 0.3084 | LB: 1.0047 | CL0: 2.7 | CL1: 2.4 | HR0: 0.367/SR0: 0.381 | HR1: 0.422/SR1: 0.397 | LR: 1.00e-04 [2026-05-23 18:40:33] Epoch 1 | Step 2520 | Loss: 0.8073 | LM: 0.3083 | LB: 1.0047 | CL0: 2.7 | CL1: 2.4 | HR0: 0.367/SR0: 0.381 | HR1: 0.422/SR1: 0.397 | LR: 1.00e-04 [2026-05-23 18:40:45] Epoch 1 | Step 2530 | Loss: 0.8071 | LM: 0.3082 | LB: 1.0047 | CL0: 2.7 | CL1: 2.4 | HR0: 0.368/SR0: 0.381 | HR1: 0.422/SR1: 0.397 | LR: 1.00e-04 [2026-05-23 18:40:57] Epoch 1 | Step 2540 | Loss: 0.8071 | LM: 0.3082 | LB: 1.0047 | CL0: 2.7 | CL1: 2.4 | HR0: 0.368/SR0: 0.381 | HR1: 0.421/SR1: 0.397 | LR: 1.00e-04 [2026-05-23 18:41:09] Epoch 1 | Step 2550 | Loss: 0.8070 | LM: 0.3082 | LB: 1.0046 | CL0: 2.7 | CL1: 2.4 | HR0: 0.368/SR0: 0.381 | HR1: 0.421/SR1: 0.397 | LR: 1.00e-04 [2026-05-23 18:41:21] Epoch 1 | Step 2560 | Loss: 0.8070 | LM: 0.3081 | LB: 1.0046 | CL0: 2.7 | CL1: 2.4 | HR0: 0.368/SR0: 0.382 | HR1: 0.421/SR1: 0.397 | LR: 1.00e-04 [2026-05-23 18:41:33] Epoch 1 | Step 2570 | Loss: 0.8070 | LM: 0.3080 | LB: 1.0046 | CL0: 2.7 | CL1: 2.4 | HR0: 0.368/SR0: 0.382 | HR1: 0.421/SR1: 0.397 | LR: 1.00e-04 [2026-05-23 18:41:46] Epoch 1 | Step 2580 | Loss: 0.8070 | LM: 0.3080 | LB: 1.0046 | CL0: 2.7 | CL1: 2.4 | HR0: 0.368/SR0: 0.382 | HR1: 0.421/SR1: 0.397 | LR: 1.00e-04 [2026-05-23 18:41:58] Epoch 1 | Step 2590 | Loss: 0.8070 | LM: 0.3081 | LB: 1.0046 | CL0: 2.7 | CL1: 2.4 | HR0: 0.368/SR0: 0.382 | HR1: 0.421/SR1: 0.397 | LR: 1.00e-04 [2026-05-23 18:42:10] Epoch 1 | Step 2600 | Loss: 0.8068 | LM: 0.3079 | LB: 1.0046 | CL0: 2.7 | CL1: 2.4 | HR0: 0.368/SR0: 0.382 | HR1: 0.421/SR1: 0.397 | LR: 1.00e-04 [2026-05-23 18:42:22] Epoch 1 | Step 2610 | Loss: 0.8066 | LM: 0.3076 | LB: 1.0046 | CL0: 2.7 | CL1: 2.4 | HR0: 0.368/SR0: 0.382 | HR1: 0.421/SR1: 0.397 | LR: 1.00e-04 [2026-05-23 18:42:34] Epoch 1 | Step 2620 | Loss: 0.8064 | LM: 0.3072 | LB: 1.0045 | CL0: 2.7 | CL1: 2.4 | HR0: 0.368/SR0: 0.383 | HR1: 0.421/SR1: 0.396 | LR: 1.00e-04 [2026-05-23 18:42:46] Epoch 1 | Step 2630 | Loss: 0.8064 | LM: 0.3073 | LB: 1.0045 | CL0: 2.7 | CL1: 2.4 | HR0: 0.368/SR0: 0.383 | HR1: 0.421/SR1: 0.396 | LR: 1.00e-04 [2026-05-23 18:42:59] Epoch 1 | Step 2640 | Loss: 0.8063 | LM: 0.3071 | LB: 1.0045 | CL0: 2.7 | CL1: 2.4 | HR0: 0.368/SR0: 0.383 | HR1: 0.421/SR1: 0.396 | LR: 1.00e-04 [2026-05-23 18:43:11] Epoch 1 | Step 2650 | Loss: 0.8063 | LM: 0.3072 | LB: 1.0045 | CL0: 2.7 | CL1: 2.4 | HR0: 0.368/SR0: 0.383 | HR1: 0.421/SR1: 0.396 | LR: 1.00e-04 [2026-05-23 18:43:25] Epoch 1 | Step 2660 | Loss: 0.8063 | LM: 0.3071 | LB: 1.0045 | CL0: 2.7 | CL1: 2.4 | HR0: 0.369/SR0: 0.383 | HR1: 0.420/SR1: 0.396 | LR: 1.00e-04 [2026-05-23 18:43:38] Epoch 1 | Step 2670 | Loss: 0.8063 | LM: 0.3072 | LB: 1.0045 | CL0: 2.7 | CL1: 2.4 | HR0: 0.369/SR0: 0.383 | HR1: 0.420/SR1: 0.396 | LR: 1.00e-04 [2026-05-23 18:43:50] Epoch 1 | Step 2680 | Loss: 0.8062 | LM: 0.3072 | LB: 1.0045 | CL0: 2.7 | CL1: 2.4 | HR0: 0.369/SR0: 0.384 | HR1: 0.420/SR1: 0.396 | LR: 1.00e-04 [2026-05-23 18:44:03] Epoch 1 | Step 2690 | Loss: 0.8062 | LM: 0.3072 | LB: 1.0044 | CL0: 2.7 | CL1: 2.4 | HR0: 0.369/SR0: 0.384 | HR1: 0.420/SR1: 0.396 | LR: 1.00e-04 [2026-05-23 18:44:14] Epoch 1 | Step 2700 | Loss: 0.8060 | LM: 0.3070 | LB: 1.0044 | CL0: 2.7 | CL1: 2.4 | HR0: 0.369/SR0: 0.384 | HR1: 0.420/SR1: 0.396 | LR: 1.00e-04 [2026-05-23 18:44:26] Epoch 1 | Step 2710 | Loss: 0.8058 | LM: 0.3069 | LB: 1.0044 | CL0: 2.7 | CL1: 2.4 | HR0: 0.369/SR0: 0.384 | HR1: 0.420/SR1: 0.396 | LR: 1.00e-04 [2026-05-23 18:44:38] Epoch 1 | Step 2720 | Loss: 0.8057 | LM: 0.3067 | LB: 1.0044 | CL0: 2.7 | CL1: 2.4 | HR0: 0.369/SR0: 0.384 | HR1: 0.420/SR1: 0.396 | LR: 1.00e-04 [2026-05-23 18:44:49] Epoch 1 | Step 2730 | Loss: 0.8056 | LM: 0.3067 | LB: 1.0044 | CL0: 2.7 | CL1: 2.4 | HR0: 0.369/SR0: 0.384 | HR1: 0.420/SR1: 0.396 | LR: 1.00e-04 [2026-05-23 18:45:01] Epoch 1 | Step 2740 | Loss: 0.8055 | LM: 0.3067 | LB: 1.0044 | CL0: 2.7 | CL1: 2.4 | HR0: 0.369/SR0: 0.384 | HR1: 0.420/SR1: 0.396 | LR: 1.00e-04 [2026-05-23 18:45:13] Epoch 1 | Step 2750 | Loss: 0.8054 | LM: 0.3064 | LB: 1.0044 | CL0: 2.7 | CL1: 2.4 | HR0: 0.369/SR0: 0.385 | HR1: 0.420/SR1: 0.396 | LR: 1.00e-04 [2026-05-23 18:45:26] Epoch 1 | Step 2760 | Loss: 0.8053 | LM: 0.3062 | LB: 1.0043 | CL0: 2.7 | CL1: 2.4 | HR0: 0.369/SR0: 0.385 | HR1: 0.420/SR1: 0.396 | LR: 1.00e-04 [2026-05-23 18:45:39] Epoch 1 | Step 2770 | Loss: 0.8052 | LM: 0.3060 | LB: 1.0043 | CL0: 2.7 | CL1: 2.4 | HR0: 0.369/SR0: 0.385 | HR1: 0.420/SR1: 0.396 | LR: 1.00e-04 [2026-05-23 18:45:50] Epoch 1 | Step 2780 | Loss: 0.8052 | LM: 0.3061 | LB: 1.0043 | CL0: 2.7 | CL1: 2.4 | HR0: 0.370/SR0: 0.385 | HR1: 0.420/SR1: 0.396 | LR: 1.00e-04 [2026-05-23 18:46:02] Epoch 1 | Step 2790 | Loss: 0.8052 | LM: 0.3061 | LB: 1.0043 | CL0: 2.7 | CL1: 2.4 | HR0: 0.370/SR0: 0.385 | HR1: 0.420/SR1: 0.396 | LR: 1.00e-04 [2026-05-23 18:46:15] Epoch 1 | Step 2800 | Loss: 0.8052 | LM: 0.3062 | LB: 1.0043 | CL0: 2.7 | CL1: 2.4 | HR0: 0.370/SR0: 0.385 | HR1: 0.419/SR1: 0.396 | LR: 1.00e-04 [2026-05-23 18:46:26] Epoch 1 | Step 2810 | Loss: 0.8051 | LM: 0.3060 | LB: 1.0043 | CL0: 2.7 | CL1: 2.4 | HR0: 0.370/SR0: 0.386 | HR1: 0.419/SR1: 0.396 | LR: 1.00e-04 [2026-05-23 18:46:37] Epoch 1 | Step 2820 | Loss: 0.8051 | LM: 0.3059 | LB: 1.0043 | CL0: 2.7 | CL1: 2.4 | HR0: 0.370/SR0: 0.386 | HR1: 0.419/SR1: 0.395 | LR: 1.00e-04 [2026-05-23 18:46:49] Epoch 1 | Step 2830 | Loss: 0.8051 | LM: 0.3059 | LB: 1.0043 | CL0: 2.7 | CL1: 2.4 | HR0: 0.370/SR0: 0.386 | HR1: 0.419/SR1: 0.395 | LR: 1.00e-04 [2026-05-23 18:47:02] Epoch 1 | Step 2840 | Loss: 0.8049 | LM: 0.3057 | LB: 1.0043 | CL0: 2.7 | CL1: 2.4 | HR0: 0.370/SR0: 0.386 | HR1: 0.419/SR1: 0.395 | LR: 1.00e-04 [2026-05-23 18:47:14] Epoch 1 | Step 2850 | Loss: 0.8049 | LM: 0.3058 | LB: 1.0042 | CL0: 2.7 | CL1: 2.4 | HR0: 0.370/SR0: 0.386 | HR1: 0.419/SR1: 0.395 | LR: 1.00e-04 [2026-05-23 18:47:28] Epoch 1 | Step 2860 | Loss: 0.8048 | LM: 0.3057 | LB: 1.0042 | CL0: 2.7 | CL1: 2.4 | HR0: 0.370/SR0: 0.386 | HR1: 0.419/SR1: 0.395 | LR: 1.00e-04 [2026-05-23 18:47:41] Epoch 1 | Step 2870 | Loss: 0.8047 | LM: 0.3055 | LB: 1.0042 | CL0: 2.7 | CL1: 2.4 | HR0: 0.370/SR0: 0.387 | HR1: 0.419/SR1: 0.395 | LR: 1.00e-04 [2026-05-23 18:47:53] Epoch 1 | Step 2880 | Loss: 0.8047 | LM: 0.3057 | LB: 1.0042 | CL0: 2.7 | CL1: 2.4 | HR0: 0.370/SR0: 0.387 | HR1: 0.419/SR1: 0.395 | LR: 1.00e-04 [2026-05-23 18:48:06] Epoch 1 | Step 2890 | Loss: 0.8046 | LM: 0.3056 | LB: 1.0042 | CL0: 2.7 | CL1: 2.4 | HR0: 0.371/SR0: 0.387 | HR1: 0.419/SR1: 0.395 | LR: 1.00e-04 [2026-05-23 18:48:19] Epoch 1 | Step 2900 | Loss: 0.8045 | LM: 0.3054 | LB: 1.0042 | CL0: 2.7 | CL1: 2.4 | HR0: 0.371/SR0: 0.387 | HR1: 0.419/SR1: 0.395 | LR: 1.00e-04 [2026-05-23 18:48:30] Epoch 1 | Step 2910 | Loss: 0.8044 | LM: 0.3052 | LB: 1.0042 | CL0: 2.7 | CL1: 2.4 | HR0: 0.371/SR0: 0.387 | HR1: 0.419/SR1: 0.395 | LR: 1.00e-04 [2026-05-23 18:48:42] Epoch 1 | Step 2920 | Loss: 0.8044 | LM: 0.3052 | LB: 1.0041 | CL0: 2.7 | CL1: 2.4 | HR0: 0.371/SR0: 0.387 | HR1: 0.419/SR1: 0.395 | LR: 1.00e-04 [2026-05-23 18:48:54] Epoch 1 | Step 2930 | Loss: 0.8043 | LM: 0.3050 | LB: 1.0041 | CL0: 2.7 | CL1: 2.4 | HR0: 0.371/SR0: 0.387 | HR1: 0.419/SR1: 0.395 | LR: 1.00e-04 [2026-05-23 18:49:06] Epoch 1 | Step 2940 | Loss: 0.8042 | LM: 0.3049 | LB: 1.0041 | CL0: 2.7 | CL1: 2.4 | HR0: 0.371/SR0: 0.388 | HR1: 0.419/SR1: 0.395 | LR: 1.00e-04 [2026-05-23 18:49:18] Epoch 1 | Step 2950 | Loss: 0.8042 | LM: 0.3048 | LB: 1.0041 | CL0: 2.7 | CL1: 2.4 | HR0: 0.371/SR0: 0.388 | HR1: 0.418/SR1: 0.395 | LR: 1.00e-04 [2026-05-23 18:49:31] Epoch 1 | Step 2960 | Loss: 0.8041 | LM: 0.3046 | LB: 1.0041 | CL0: 2.7 | CL1: 2.4 | HR0: 0.371/SR0: 0.388 | HR1: 0.418/SR1: 0.395 | LR: 1.00e-04 [2026-05-23 18:49:43] Epoch 1 | Step 2970 | Loss: 0.8039 | LM: 0.3044 | LB: 1.0041 | CL0: 2.7 | CL1: 2.4 | HR0: 0.371/SR0: 0.388 | HR1: 0.418/SR1: 0.395 | LR: 1.00e-04 [2026-05-23 18:49:56] Epoch 1 | Step 2980 | Loss: 0.8038 | LM: 0.3044 | LB: 1.0041 | CL0: 2.7 | CL1: 2.4 | HR0: 0.371/SR0: 0.388 | HR1: 0.418/SR1: 0.395 | LR: 1.00e-04 [2026-05-23 18:50:08] Epoch 1 | Step 2990 | Loss: 0.8037 | LM: 0.3043 | LB: 1.0041 | CL0: 2.7 | CL1: 2.4 | HR0: 0.371/SR0: 0.388 | HR1: 0.418/SR1: 0.395 | LR: 1.00e-04 [2026-05-23 18:50:20] Epoch 1 | Step 3000 | Loss: 0.8037 | LM: 0.3043 | LB: 1.0041 | CL0: 2.7 | CL1: 2.4 | HR0: 0.371/SR0: 0.389 | HR1: 0.418/SR1: 0.395 | LR: 1.00e-04 [2026-05-23 18:50:24] Checkpoint saved: outputs/2026-05-23/N_2.5/checkpoints/checkpoint_step_3000.pt [2026-05-23 18:50:30] Validation | Batch 10/1463 | Loss: 0.7523 | LM_LOSS: 0.2525 | LB_LOSS: 0.9996 [2026-05-23 18:50:32] Validation | Batch 20/1463 | Loss: 0.7611 | LM_LOSS: 0.2611 | LB_LOSS: 0.9999 [2026-05-23 18:50:36] Validation | Batch 30/1463 | Loss: 0.7670 | LM_LOSS: 0.2669 | LB_LOSS: 1.0001 [2026-05-23 18:50:39] Validation | Batch 40/1463 | Loss: 0.7748 | LM_LOSS: 0.2747 | LB_LOSS: 1.0001 [2026-05-23 18:50:42] Validation | Batch 50/1463 | Loss: 0.7823 | LM_LOSS: 0.2823 | LB_LOSS: 1.0001 [2026-05-23 18:50:45] Validation | Batch 60/1463 | Loss: 0.7826 | LM_LOSS: 0.2825 | LB_LOSS: 1.0001 [2026-05-23 18:50:49] Validation | Batch 70/1463 | Loss: 0.7826 | LM_LOSS: 0.2825 | LB_LOSS: 1.0001 [2026-05-23 18:50:53] Validation | Batch 80/1463 | Loss: 0.7841 | LM_LOSS: 0.2840 | LB_LOSS: 1.0002 [2026-05-23 18:50:55] Validation | Batch 90/1463 | Loss: 0.7835 | LM_LOSS: 0.2834 | LB_LOSS: 1.0002 [2026-05-23 18:50:58] Validation | Batch 100/1463 | Loss: 0.7836 | LM_LOSS: 0.2835 | LB_LOSS: 1.0001 [2026-05-23 18:51:01] Validation | Batch 110/1463 | Loss: 0.7823 | LM_LOSS: 0.2822 | LB_LOSS: 1.0001 [2026-05-23 18:51:03] Validation | Batch 120/1463 | Loss: 0.7837 | LM_LOSS: 0.2837 | LB_LOSS: 1.0001 [2026-05-23 18:51:06] Validation | Batch 130/1463 | Loss: 0.7853 | LM_LOSS: 0.2852 | LB_LOSS: 1.0002 [2026-05-23 18:51:10] Validation | Batch 140/1463 | Loss: 0.7863 | LM_LOSS: 0.2863 | LB_LOSS: 1.0002 [2026-05-23 18:51:13] Validation | Batch 150/1463 | Loss: 0.7844 | LM_LOSS: 0.2842 | LB_LOSS: 1.0003 [2026-05-23 18:51:16] Validation | Batch 160/1463 | Loss: 0.7847 | LM_LOSS: 0.2846 | LB_LOSS: 1.0003 [2026-05-23 18:51:19] Validation | Batch 170/1463 | Loss: 0.7842 | LM_LOSS: 0.2840 | LB_LOSS: 1.0003 [2026-05-23 18:51:22] Validation | Batch 180/1463 | Loss: 0.7850 | LM_LOSS: 0.2849 | LB_LOSS: 1.0003 [2026-05-23 18:51:25] Validation | Batch 190/1463 | Loss: 0.7858 | LM_LOSS: 0.2857 | LB_LOSS: 1.0002 [2026-05-23 18:51:28] Validation | Batch 200/1463 | Loss: 0.7859 | LM_LOSS: 0.2858 | LB_LOSS: 1.0002 [2026-05-23 18:51:31] Validation | Batch 210/1463 | Loss: 0.7866 | LM_LOSS: 0.2865 | LB_LOSS: 1.0003 [2026-05-23 18:51:34] Validation | Batch 220/1463 | Loss: 0.7857 | LM_LOSS: 0.2856 | LB_LOSS: 1.0003 [2026-05-23 18:51:38] Validation | Batch 230/1463 | Loss: 0.7870 | LM_LOSS: 0.2869 | LB_LOSS: 1.0002 [2026-05-23 18:51:41] Validation | Batch 240/1463 | Loss: 0.7885 | LM_LOSS: 0.2884 | LB_LOSS: 1.0002 [2026-05-23 18:51:44] Validation | Batch 250/1463 | Loss: 0.7875 | LM_LOSS: 0.2874 | LB_LOSS: 1.0002 [2026-05-23 18:51:47] Validation | Batch 260/1463 | Loss: 0.7889 | LM_LOSS: 0.2888 | LB_LOSS: 1.0003 [2026-05-23 18:51:50] Validation | Batch 270/1463 | Loss: 0.7875 | LM_LOSS: 0.2874 | LB_LOSS: 1.0003 [2026-05-23 18:51:53] Validation | Batch 280/1463 | Loss: 0.7881 | LM_LOSS: 0.2879 | LB_LOSS: 1.0003 [2026-05-23 18:51:56] Validation | Batch 290/1463 | Loss: 0.7873 | LM_LOSS: 0.2871 | LB_LOSS: 1.0003 [2026-05-23 18:51:58] Validation | Batch 300/1463 | Loss: 0.7874 | LM_LOSS: 0.2872 | LB_LOSS: 1.0003 [2026-05-23 18:52:02] Validation | Batch 310/1463 | Loss: 0.7885 | LM_LOSS: 0.2884 | LB_LOSS: 1.0003 [2026-05-23 18:52:04] Validation | Batch 320/1463 | Loss: 0.7872 | LM_LOSS: 0.2871 | LB_LOSS: 1.0002 [2026-05-23 18:52:07] Validation | Batch 330/1463 | Loss: 0.7870 | LM_LOSS: 0.2869 | LB_LOSS: 1.0003 [2026-05-23 18:52:10] Validation | Batch 340/1463 | Loss: 0.7868 | LM_LOSS: 0.2867 | LB_LOSS: 1.0003 [2026-05-23 18:52:13] Validation | Batch 350/1463 | Loss: 0.7873 | LM_LOSS: 0.2872 | LB_LOSS: 1.0003 [2026-05-23 18:52:16] Validation | Batch 360/1463 | Loss: 0.7870 | LM_LOSS: 0.2868 | LB_LOSS: 1.0003 [2026-05-23 18:52:19] Validation | Batch 370/1463 | Loss: 0.7867 | LM_LOSS: 0.2866 | LB_LOSS: 1.0003 [2026-05-23 18:52:21] Validation | Batch 380/1463 | Loss: 0.7866 | LM_LOSS: 0.2865 | LB_LOSS: 1.0002 [2026-05-23 18:52:24] Validation | Batch 390/1463 | Loss: 0.7872 | LM_LOSS: 0.2871 | LB_LOSS: 1.0002 [2026-05-23 18:52:27] Validation | Batch 400/1463 | Loss: 0.7870 | LM_LOSS: 0.2869 | LB_LOSS: 1.0002 [2026-05-23 18:52:30] Validation | Batch 410/1463 | Loss: 0.7867 | LM_LOSS: 0.2866 | LB_LOSS: 1.0002 [2026-05-23 18:52:33] Validation | Batch 420/1463 | Loss: 0.7868 | LM_LOSS: 0.2866 | LB_LOSS: 1.0002 [2026-05-23 18:52:37] Validation | Batch 430/1463 | Loss: 0.7863 | LM_LOSS: 0.2861 | LB_LOSS: 1.0002 [2026-05-23 18:52:39] Validation | Batch 440/1463 | Loss: 0.7865 | LM_LOSS: 0.2863 | LB_LOSS: 1.0002 [2026-05-23 18:52:42] Validation | Batch 450/1463 | Loss: 0.7863 | LM_LOSS: 0.2862 | LB_LOSS: 1.0003 [2026-05-23 18:52:45] Validation | Batch 460/1463 | Loss: 0.7863 | LM_LOSS: 0.2861 | LB_LOSS: 1.0003 [2026-05-23 18:52:48] Validation | Batch 470/1463 | Loss: 0.7868 | LM_LOSS: 0.2867 | LB_LOSS: 1.0002 [2026-05-23 18:52:52] Validation | Batch 480/1463 | Loss: 0.7866 | LM_LOSS: 0.2865 | LB_LOSS: 1.0002 [2026-05-23 18:52:55] Validation | Batch 490/1463 | Loss: 0.7865 | LM_LOSS: 0.2864 | LB_LOSS: 1.0002 [2026-05-23 18:52:58] Validation | Batch 500/1463 | Loss: 0.7864 | LM_LOSS: 0.2863 | LB_LOSS: 1.0002 [2026-05-23 18:53:02] Validation | Batch 510/1463 | Loss: 0.7859 | LM_LOSS: 0.2857 | LB_LOSS: 1.0002 [2026-05-23 18:53:04] Validation | Batch 520/1463 | Loss: 0.7856 | LM_LOSS: 0.2855 | LB_LOSS: 1.0002 [2026-05-23 18:53:07] Validation | Batch 530/1463 | Loss: 0.7859 | LM_LOSS: 0.2858 | LB_LOSS: 1.0002 [2026-05-23 18:53:09] Validation | Batch 540/1463 | Loss: 0.7862 | LM_LOSS: 0.2860 | LB_LOSS: 1.0002 [2026-05-23 18:53:11] Validation | Batch 550/1463 | Loss: 0.7858 | LM_LOSS: 0.2857 | LB_LOSS: 1.0002 [2026-05-23 18:53:14] Validation | Batch 560/1463 | Loss: 0.7859 | LM_LOSS: 0.2858 | LB_LOSS: 1.0002 [2026-05-23 18:53:18] Validation | Batch 570/1463 | Loss: 0.7859 | LM_LOSS: 0.2858 | LB_LOSS: 1.0002 [2026-05-23 18:53:21] Validation | Batch 580/1463 | Loss: 0.7855 | LM_LOSS: 0.2854 | LB_LOSS: 1.0002 [2026-05-23 18:53:25] Validation | Batch 590/1463 | Loss: 0.7853 | LM_LOSS: 0.2851 | LB_LOSS: 1.0002 [2026-05-23 18:53:28] Validation | Batch 600/1463 | Loss: 0.7856 | LM_LOSS: 0.2855 | LB_LOSS: 1.0002 [2026-05-23 18:53:31] Validation | Batch 610/1463 | Loss: 0.7858 | LM_LOSS: 0.2857 | LB_LOSS: 1.0002 [2026-05-23 18:53:34] Validation | Batch 620/1463 | Loss: 0.7854 | LM_LOSS: 0.2853 | LB_LOSS: 1.0002 [2026-05-23 18:53:37] Validation | Batch 630/1463 | Loss: 0.7848 | LM_LOSS: 0.2847 | LB_LOSS: 1.0002 [2026-05-23 18:53:40] Validation | Batch 640/1463 | Loss: 0.7846 | LM_LOSS: 0.2845 | LB_LOSS: 1.0002 [2026-05-23 18:53:42] Validation | Batch 650/1463 | Loss: 0.7842 | LM_LOSS: 0.2840 | LB_LOSS: 1.0002 [2026-05-23 18:53:45] Validation | Batch 660/1463 | Loss: 0.7837 | LM_LOSS: 0.2836 | LB_LOSS: 1.0002 [2026-05-23 18:53:48] Validation | Batch 670/1463 | Loss: 0.7835 | LM_LOSS: 0.2834 | LB_LOSS: 1.0002 [2026-05-23 18:53:50] Validation | Batch 680/1463 | Loss: 0.7835 | LM_LOSS: 0.2834 | LB_LOSS: 1.0002 [2026-05-23 18:53:54] Validation | Batch 690/1463 | Loss: 0.7838 | LM_LOSS: 0.2837 | LB_LOSS: 1.0002 [2026-05-23 18:53:57] Validation | Batch 700/1463 | Loss: 0.7837 | LM_LOSS: 0.2835 | LB_LOSS: 1.0002 [2026-05-23 18:53:59] Validation | Batch 710/1463 | Loss: 0.7839 | LM_LOSS: 0.2838 | LB_LOSS: 1.0002 [2026-05-23 18:54:03] Validation | Batch 720/1463 | Loss: 0.7847 | LM_LOSS: 0.2846 | LB_LOSS: 1.0002 [2026-05-23 18:54:05] Validation | Batch 730/1463 | Loss: 0.7845 | LM_LOSS: 0.2844 | LB_LOSS: 1.0002 [2026-05-23 18:54:07] Validation | Batch 740/1463 | Loss: 0.7844 | LM_LOSS: 0.2842 | LB_LOSS: 1.0002 [2026-05-23 18:54:10] Validation | Batch 750/1463 | Loss: 0.7844 | LM_LOSS: 0.2842 | LB_LOSS: 1.0002 [2026-05-23 18:54:13] Validation | Batch 760/1463 | Loss: 0.7842 | LM_LOSS: 0.2841 | LB_LOSS: 1.0002 [2026-05-23 18:54:15] Validation | Batch 770/1463 | Loss: 0.7841 | LM_LOSS: 0.2840 | LB_LOSS: 1.0002 [2026-05-23 18:54:18] Validation | Batch 780/1463 | Loss: 0.7840 | LM_LOSS: 0.2839 | LB_LOSS: 1.0002 [2026-05-23 18:54:20] Validation | Batch 790/1463 | Loss: 0.7843 | LM_LOSS: 0.2842 | LB_LOSS: 1.0002 [2026-05-23 18:54:23] Validation | Batch 800/1463 | Loss: 0.7842 | LM_LOSS: 0.2841 | LB_LOSS: 1.0002 [2026-05-23 18:54:25] Validation | Batch 810/1463 | Loss: 0.7839 | LM_LOSS: 0.2838 | LB_LOSS: 1.0002 [2026-05-23 18:54:28] Validation | Batch 820/1463 | Loss: 0.7841 | LM_LOSS: 0.2840 | LB_LOSS: 1.0002 [2026-05-23 18:54:32] Validation | Batch 830/1463 | Loss: 0.7843 | LM_LOSS: 0.2842 | LB_LOSS: 1.0002 [2026-05-23 18:54:35] Validation | Batch 840/1463 | Loss: 0.7845 | LM_LOSS: 0.2844 | LB_LOSS: 1.0002 [2026-05-23 18:54:38] Validation | Batch 850/1463 | Loss: 0.7847 | LM_LOSS: 0.2846 | LB_LOSS: 1.0002 [2026-05-23 18:54:41] Validation | Batch 860/1463 | Loss: 0.7848 | LM_LOSS: 0.2847 | LB_LOSS: 1.0002 [2026-05-23 18:54:44] Validation | Batch 870/1463 | Loss: 0.7849 | LM_LOSS: 0.2847 | LB_LOSS: 1.0002 [2026-05-23 18:54:47] Validation | Batch 880/1463 | Loss: 0.7841 | LM_LOSS: 0.2840 | LB_LOSS: 1.0002 [2026-05-23 18:54:50] Validation | Batch 890/1463 | Loss: 0.7841 | LM_LOSS: 0.2840 | LB_LOSS: 1.0002 [2026-05-23 18:54:53] Validation | Batch 900/1463 | Loss: 0.7839 | LM_LOSS: 0.2837 | LB_LOSS: 1.0002 [2026-05-23 18:54:56] Validation | Batch 910/1463 | Loss: 0.7838 | LM_LOSS: 0.2837 | LB_LOSS: 1.0002 [2026-05-23 18:54:59] Validation | Batch 920/1463 | Loss: 0.7840 | LM_LOSS: 0.2839 | LB_LOSS: 1.0002 [2026-05-23 18:55:02] Validation | Batch 930/1463 | Loss: 0.7838 | LM_LOSS: 0.2837 | LB_LOSS: 1.0002 [2026-05-23 18:55:05] Validation | Batch 940/1463 | Loss: 0.7835 | LM_LOSS: 0.2834 | LB_LOSS: 1.0002 [2026-05-23 18:55:09] Validation | Batch 950/1463 | Loss: 0.7838 | LM_LOSS: 0.2837 | LB_LOSS: 1.0002 [2026-05-23 18:55:11] Validation | Batch 960/1463 | Loss: 0.7839 | LM_LOSS: 0.2838 | LB_LOSS: 1.0002 [2026-05-23 18:55:15] Validation | Batch 970/1463 | Loss: 0.7839 | LM_LOSS: 0.2838 | LB_LOSS: 1.0002 [2026-05-23 18:55:18] Validation | Batch 980/1463 | Loss: 0.7842 | LM_LOSS: 0.2841 | LB_LOSS: 1.0002 [2026-05-23 18:55:21] Validation | Batch 990/1463 | Loss: 0.7845 | LM_LOSS: 0.2843 | LB_LOSS: 1.0002 [2026-05-23 18:55:24] Validation | Batch 1000/1463 | Loss: 0.7851 | LM_LOSS: 0.2849 | LB_LOSS: 1.0002 [2026-05-23 18:55:27] Validation | Batch 1010/1463 | Loss: 0.7851 | LM_LOSS: 0.2850 | LB_LOSS: 1.0002 [2026-05-23 18:55:30] Validation | Batch 1020/1463 | Loss: 0.7852 | LM_LOSS: 0.2850 | LB_LOSS: 1.0002 [2026-05-23 18:55:33] Validation | Batch 1030/1463 | Loss: 0.7856 | LM_LOSS: 0.2855 | LB_LOSS: 1.0002 [2026-05-23 18:55:35] Validation | Batch 1040/1463 | Loss: 0.7852 | LM_LOSS: 0.2851 | LB_LOSS: 1.0002 [2026-05-23 18:55:39] Validation | Batch 1050/1463 | Loss: 0.7851 | LM_LOSS: 0.2850 | LB_LOSS: 1.0002 [2026-05-23 18:55:42] Validation | Batch 1060/1463 | Loss: 0.7851 | LM_LOSS: 0.2850 | LB_LOSS: 1.0002 [2026-05-23 18:55:45] Validation | Batch 1070/1463 | Loss: 0.7850 | LM_LOSS: 0.2849 | LB_LOSS: 1.0002 [2026-05-23 18:55:48] Validation | Batch 1080/1463 | Loss: 0.7849 | LM_LOSS: 0.2847 | LB_LOSS: 1.0002 [2026-05-23 18:55:50] Validation | Batch 1090/1463 | Loss: 0.7849 | LM_LOSS: 0.2848 | LB_LOSS: 1.0002 [2026-05-23 18:55:53] Validation | Batch 1100/1463 | Loss: 0.7847 | LM_LOSS: 0.2846 | LB_LOSS: 1.0002 [2026-05-23 18:55:56] Validation | Batch 1110/1463 | Loss: 0.7848 | LM_LOSS: 0.2846 | LB_LOSS: 1.0002 [2026-05-23 18:55:59] Validation | Batch 1120/1463 | Loss: 0.7846 | LM_LOSS: 0.2844 | LB_LOSS: 1.0002 [2026-05-23 18:56:02] Validation | Batch 1130/1463 | Loss: 0.7844 | LM_LOSS: 0.2842 | LB_LOSS: 1.0002 [2026-05-23 18:56:05] Validation | Batch 1140/1463 | Loss: 0.7846 | LM_LOSS: 0.2844 | LB_LOSS: 1.0002 [2026-05-23 18:56:09] Validation | Batch 1150/1463 | Loss: 0.7845 | LM_LOSS: 0.2844 | LB_LOSS: 1.0002 [2026-05-23 18:56:11] Validation | Batch 1160/1463 | Loss: 0.7844 | LM_LOSS: 0.2843 | LB_LOSS: 1.0002 [2026-05-23 18:56:15] Validation | Batch 1170/1463 | Loss: 0.7844 | LM_LOSS: 0.2843 | LB_LOSS: 1.0002 [2026-05-23 18:56:17] Validation | Batch 1180/1463 | Loss: 0.7845 | LM_LOSS: 0.2844 | LB_LOSS: 1.0002 [2026-05-23 18:56:21] Validation | Batch 1190/1463 | Loss: 0.7846 | LM_LOSS: 0.2845 | LB_LOSS: 1.0002 [2026-05-23 18:56:24] Validation | Batch 1200/1463 | Loss: 0.7846 | LM_LOSS: 0.2845 | LB_LOSS: 1.0002 [2026-05-23 18:56:27] Validation | Batch 1210/1463 | Loss: 0.7848 | LM_LOSS: 0.2847 | LB_LOSS: 1.0002 [2026-05-23 18:56:30] Validation | Batch 1220/1463 | Loss: 0.7851 | LM_LOSS: 0.2850 | LB_LOSS: 1.0002 [2026-05-23 18:56:32] Validation | Batch 1230/1463 | Loss: 0.7851 | LM_LOSS: 0.2850 | LB_LOSS: 1.0002 [2026-05-23 18:56:35] Validation | Batch 1240/1463 | Loss: 0.7854 | LM_LOSS: 0.2853 | LB_LOSS: 1.0002 [2026-05-23 18:56:39] Validation | Batch 1250/1463 | Loss: 0.7855 | LM_LOSS: 0.2854 | LB_LOSS: 1.0002 [2026-05-23 18:56:42] Validation | Batch 1260/1463 | Loss: 0.7854 | LM_LOSS: 0.2853 | LB_LOSS: 1.0002 [2026-05-23 18:56:44] Validation | Batch 1270/1463 | Loss: 0.7853 | LM_LOSS: 0.2852 | LB_LOSS: 1.0002 [2026-05-23 18:56:48] Validation | Batch 1280/1463 | Loss: 0.7852 | LM_LOSS: 0.2851 | LB_LOSS: 1.0002 [2026-05-23 18:56:51] Validation | Batch 1290/1463 | Loss: 0.7852 | LM_LOSS: 0.2851 | LB_LOSS: 1.0002 [2026-05-23 18:56:54] Validation | Batch 1300/1463 | Loss: 0.7850 | LM_LOSS: 0.2849 | LB_LOSS: 1.0002 [2026-05-23 18:56:57] Validation | Batch 1310/1463 | Loss: 0.7849 | LM_LOSS: 0.2848 | LB_LOSS: 1.0002 [2026-05-23 18:56:59] Validation | Batch 1320/1463 | Loss: 0.7850 | LM_LOSS: 0.2849 | LB_LOSS: 1.0002 [2026-05-23 18:57:01] Validation | Batch 1330/1463 | Loss: 0.7850 | LM_LOSS: 0.2849 | LB_LOSS: 1.0002 [2026-05-23 18:57:04] Validation | Batch 1340/1463 | Loss: 0.7852 | LM_LOSS: 0.2851 | LB_LOSS: 1.0002 [2026-05-23 18:57:07] Validation | Batch 1350/1463 | Loss: 0.7852 | LM_LOSS: 0.2851 | LB_LOSS: 1.0002 [2026-05-23 18:57:11] Validation | Batch 1360/1463 | Loss: 0.7855 | LM_LOSS: 0.2854 | LB_LOSS: 1.0002 [2026-05-23 18:57:14] Validation | Batch 1370/1463 | Loss: 0.7858 | LM_LOSS: 0.2857 | LB_LOSS: 1.0002 [2026-05-23 18:57:17] Validation | Batch 1380/1463 | Loss: 0.7856 | LM_LOSS: 0.2855 | LB_LOSS: 1.0002 [2026-05-23 18:57:20] Validation | Batch 1390/1463 | Loss: 0.7855 | LM_LOSS: 0.2854 | LB_LOSS: 1.0002 [2026-05-23 18:57:23] Validation | Batch 1400/1463 | Loss: 0.7858 | LM_LOSS: 0.2857 | LB_LOSS: 1.0002 [2026-05-23 18:57:25] Validation | Batch 1410/1463 | Loss: 0.7860 | LM_LOSS: 0.2859 | LB_LOSS: 1.0002 [2026-05-23 18:57:28] Validation | Batch 1420/1463 | Loss: 0.7860 | LM_LOSS: 0.2859 | LB_LOSS: 1.0002 [2026-05-23 18:57:32] Validation | Batch 1430/1463 | Loss: 0.7861 | LM_LOSS: 0.2860 | LB_LOSS: 1.0002 [2026-05-23 18:57:35] Validation | Batch 1440/1463 | Loss: 0.7864 | LM_LOSS: 0.2863 | LB_LOSS: 1.0002 [2026-05-23 18:57:38] Validation | Batch 1450/1463 | Loss: 0.7862 | LM_LOSS: 0.2861 | LB_LOSS: 1.0002 [2026-05-23 18:57:41] Validation | Batch 1460/1463 | Loss: 0.7865 | LM_LOSS: 0.2863 | LB_LOSS: 1.0002 [2026-05-23 18:57:42] Validation | Batch 1463/1463 | Loss: 0.7863 | LM_LOSS: 0.2862 | LB_LOSS: 1.0002 [2026-05-23 18:57:42] Validation | Loss: 0.7863 | LM_LOSS: 0.2862 | LB_LOSS: 1.0002 | PPL: 1.33 | Time: 434.10s [2026-05-23 18:57:48] New best model saved! Val loss: 0.7863 [2026-05-23 18:58:00] Epoch 1 | Step 3010 | Loss: 0.8037 | LM: 0.3042 | LB: 1.0041 | CL0: 2.7 | CL1: 2.4 | HR0: 0.371/SR0: 0.389 | HR1: 0.418/SR1: 0.395 | LR: 1.00e-04 [2026-05-23 18:58:12] Epoch 1 | Step 3020 | Loss: 0.8036 | LM: 0.3041 | LB: 1.0041 | CL0: 2.7 | CL1: 2.4 | HR0: 0.372/SR0: 0.389 | HR1: 0.418/SR1: 0.395 | LR: 1.00e-04 [2026-05-23 18:58:24] Epoch 1 | Step 3030 | Loss: 0.8033 | LM: 0.3041 | LB: 1.0040 | CL0: 2.7 | CL1: 2.4 | HR0: 0.372/SR0: 0.389 | HR1: 0.418/SR1: 0.395 | LR: 1.00e-04 [2026-05-23 18:58:37] Epoch 1 | Step 3040 | Loss: 0.8033 | LM: 0.3039 | LB: 1.0040 | CL0: 2.7 | CL1: 2.4 | HR0: 0.372/SR0: 0.389 | HR1: 0.418/SR1: 0.395 | LR: 1.00e-04 [2026-05-23 18:58:51] Epoch 1 | Step 3050 | Loss: 0.8032 | LM: 0.3039 | LB: 1.0040 | CL0: 2.7 | CL1: 2.4 | HR0: 0.372/SR0: 0.389 | HR1: 0.418/SR1: 0.395 | LR: 1.00e-04 [2026-05-23 18:59:04] Epoch 1 | Step 3060 | Loss: 0.8031 | LM: 0.3040 | LB: 1.0040 | CL0: 2.7 | CL1: 2.4 | HR0: 0.372/SR0: 0.389 | HR1: 0.418/SR1: 0.395 | LR: 1.00e-04 [2026-05-23 18:59:16] Epoch 1 | Step 3070 | Loss: 0.8029 | LM: 0.3038 | LB: 1.0040 | CL0: 2.7 | CL1: 2.4 | HR0: 0.372/SR0: 0.390 | HR1: 0.418/SR1: 0.394 | LR: 1.00e-04 [2026-05-23 18:59:28] Epoch 1 | Step 3080 | Loss: 0.8029 | LM: 0.3036 | LB: 1.0040 | CL0: 2.7 | CL1: 2.4 | HR0: 0.372/SR0: 0.390 | HR1: 0.418/SR1: 0.394 | LR: 1.00e-04 [2026-05-23 18:59:42] Epoch 1 | Step 3090 | Loss: 0.8029 | LM: 0.3035 | LB: 1.0040 | CL0: 2.7 | CL1: 2.4 | HR0: 0.372/SR0: 0.390 | HR1: 0.418/SR1: 0.394 | LR: 1.00e-04 [2026-05-23 18:59:55] Epoch 1 | Step 3100 | Loss: 0.8028 | LM: 0.3033 | LB: 1.0040 | CL0: 2.7 | CL1: 2.4 | HR0: 0.372/SR0: 0.390 | HR1: 0.418/SR1: 0.394 | LR: 1.00e-04 [2026-05-23 19:00:09] Epoch 1 | Step 3110 | Loss: 0.8027 | LM: 0.3031 | LB: 1.0040 | CL0: 2.7 | CL1: 2.4 | HR0: 0.372/SR0: 0.390 | HR1: 0.418/SR1: 0.394 | LR: 1.00e-04 [2026-05-23 19:00:22] Epoch 1 | Step 3120 | Loss: 0.8028 | LM: 0.3031 | LB: 1.0040 | CL0: 2.7 | CL1: 2.4 | HR0: 0.372/SR0: 0.390 | HR1: 0.418/SR1: 0.394 | LR: 1.00e-04 [2026-05-23 19:00:34] Epoch 1 | Step 3130 | Loss: 0.8027 | LM: 0.3030 | LB: 1.0040 | CL0: 2.7 | CL1: 2.4 | HR0: 0.372/SR0: 0.390 | HR1: 0.417/SR1: 0.394 | LR: 1.00e-04 [2026-05-23 19:00:46] Epoch 1 | Step 3140 | Loss: 0.8026 | LM: 0.3027 | LB: 1.0039 | CL0: 2.7 | CL1: 2.4 | HR0: 0.372/SR0: 0.390 | HR1: 0.417/SR1: 0.394 | LR: 1.00e-04 [2026-05-23 19:01:00] Epoch 1 | Step 3150 | Loss: 0.8025 | LM: 0.3027 | LB: 1.0039 | CL0: 2.7 | CL1: 2.4 | HR0: 0.372/SR0: 0.391 | HR1: 0.417/SR1: 0.394 | LR: 1.00e-04 [2026-05-23 19:01:11] Epoch 1 | Step 3160 | Loss: 0.8025 | LM: 0.3026 | LB: 1.0039 | CL0: 2.7 | CL1: 2.4 | HR0: 0.373/SR0: 0.391 | HR1: 0.417/SR1: 0.394 | LR: 1.00e-04 [2026-05-23 19:01:24] Epoch 1 | Step 3170 | Loss: 0.8026 | LM: 0.3028 | LB: 1.0039 | CL0: 2.7 | CL1: 2.4 | HR0: 0.373/SR0: 0.391 | HR1: 0.417/SR1: 0.394 | LR: 1.00e-04 [2026-05-23 19:01:38] Epoch 1 | Step 3180 | Loss: 0.8026 | LM: 0.3027 | LB: 1.0039 | CL0: 2.7 | CL1: 2.4 | HR0: 0.373/SR0: 0.391 | HR1: 0.417/SR1: 0.394 | LR: 1.00e-04 [2026-05-23 19:01:51] Epoch 1 | Step 3190 | Loss: 0.8025 | LM: 0.3026 | LB: 1.0039 | CL0: 2.7 | CL1: 2.4 | HR0: 0.373/SR0: 0.391 | HR1: 0.417/SR1: 0.394 | LR: 1.00e-04 [2026-05-23 19:02:03] Epoch 1 | Step 3200 | Loss: 0.8023 | LM: 0.3023 | LB: 1.0039 | CL0: 2.7 | CL1: 2.4 | HR0: 0.373/SR0: 0.391 | HR1: 0.417/SR1: 0.394 | LR: 1.00e-04 [2026-05-23 19:02:15] Epoch 1 | Step 3210 | Loss: 0.8023 | LM: 0.3022 | LB: 1.0039 | CL0: 2.7 | CL1: 2.4 | HR0: 0.373/SR0: 0.391 | HR1: 0.417/SR1: 0.394 | LR: 1.00e-04 [2026-05-23 19:02:27] Epoch 1 | Step 3220 | Loss: 0.8023 | LM: 0.3025 | LB: 1.0039 | CL0: 2.7 | CL1: 2.4 | HR0: 0.373/SR0: 0.392 | HR1: 0.417/SR1: 0.394 | LR: 1.00e-04 [2026-05-23 19:02:40] Epoch 1 | Step 3230 | Loss: 0.8022 | LM: 0.3023 | LB: 1.0039 | CL0: 2.7 | CL1: 2.4 | HR0: 0.373/SR0: 0.392 | HR1: 0.417/SR1: 0.394 | LR: 1.00e-04 [2026-05-23 19:02:53] Epoch 1 | Step 3240 | Loss: 0.8022 | LM: 0.3023 | LB: 1.0038 | CL0: 2.7 | CL1: 2.4 | HR0: 0.373/SR0: 0.392 | HR1: 0.417/SR1: 0.394 | LR: 1.00e-04 [2026-05-23 19:03:05] Epoch 1 | Step 3250 | Loss: 0.8021 | LM: 0.3021 | LB: 1.0038 | CL0: 2.7 | CL1: 2.4 | HR0: 0.373/SR0: 0.392 | HR1: 0.417/SR1: 0.394 | LR: 1.00e-04 [2026-05-23 19:03:17] Epoch 1 | Step 3260 | Loss: 0.8020 | LM: 0.3022 | LB: 1.0038 | CL0: 2.7 | CL1: 2.4 | HR0: 0.373/SR0: 0.392 | HR1: 0.417/SR1: 0.394 | LR: 1.00e-04 [2026-05-23 19:03:29] Epoch 1 | Step 3270 | Loss: 0.8019 | LM: 0.3021 | LB: 1.0038 | CL0: 2.7 | CL1: 2.4 | HR0: 0.373/SR0: 0.392 | HR1: 0.417/SR1: 0.394 | LR: 1.00e-04 [2026-05-23 19:03:41] Epoch 1 | Step 3280 | Loss: 0.8019 | LM: 0.3020 | LB: 1.0038 | CL0: 2.7 | CL1: 2.4 | HR0: 0.373/SR0: 0.392 | HR1: 0.417/SR1: 0.394 | LR: 1.00e-04 [2026-05-23 19:03:54] Epoch 1 | Step 3290 | Loss: 0.8018 | LM: 0.3017 | LB: 1.0038 | CL0: 2.7 | CL1: 2.4 | HR0: 0.373/SR0: 0.392 | HR1: 0.417/SR1: 0.394 | LR: 1.00e-04 [2026-05-23 19:04:06] Epoch 1 | Step 3300 | Loss: 0.8017 | LM: 0.3017 | LB: 1.0038 | CL0: 2.7 | CL1: 2.4 | HR0: 0.374/SR0: 0.393 | HR1: 0.417/SR1: 0.394 | LR: 1.00e-04 [2026-05-23 19:04:18] Epoch 1 | Step 3310 | Loss: 0.8017 | LM: 0.3017 | LB: 1.0038 | CL0: 2.7 | CL1: 2.4 | HR0: 0.374/SR0: 0.393 | HR1: 0.417/SR1: 0.394 | LR: 1.00e-04 [2026-05-23 19:04:30] Epoch 1 | Step 3320 | Loss: 0.8016 | LM: 0.3018 | LB: 1.0038 | CL0: 2.7 | CL1: 2.4 | HR0: 0.374/SR0: 0.393 | HR1: 0.417/SR1: 0.394 | LR: 1.00e-04 [2026-05-23 19:04:43] Epoch 1 | Step 3330 | Loss: 0.8016 | LM: 0.3017 | LB: 1.0038 | CL0: 2.7 | CL1: 2.4 | HR0: 0.374/SR0: 0.393 | HR1: 0.417/SR1: 0.394 | LR: 1.00e-04 [2026-05-23 19:04:55] Epoch 1 | Step 3340 | Loss: 0.8015 | LM: 0.3017 | LB: 1.0038 | CL0: 2.7 | CL1: 2.4 | HR0: 0.374/SR0: 0.393 | HR1: 0.417/SR1: 0.394 | LR: 1.00e-04 [2026-05-23 19:05:08] Epoch 1 | Step 3350 | Loss: 0.8015 | LM: 0.3018 | LB: 1.0038 | CL0: 2.7 | CL1: 2.4 | HR0: 0.374/SR0: 0.393 | HR1: 0.417/SR1: 0.394 | LR: 1.00e-04 [2026-05-23 19:05:19] Epoch 1 | Step 3360 | Loss: 0.8015 | LM: 0.3017 | LB: 1.0037 | CL0: 2.7 | CL1: 2.4 | HR0: 0.374/SR0: 0.393 | HR1: 0.417/SR1: 0.394 | LR: 1.00e-04 [2026-05-23 19:05:31] Epoch 1 | Step 3370 | Loss: 0.8014 | LM: 0.3016 | LB: 1.0037 | CL0: 2.7 | CL1: 2.4 | HR0: 0.374/SR0: 0.394 | HR1: 0.417/SR1: 0.394 | LR: 1.00e-04 [2026-05-23 19:05:44] Epoch 1 | Step 3380 | Loss: 0.8014 | LM: 0.3014 | LB: 1.0037 | CL0: 2.7 | CL1: 2.4 | HR0: 0.374/SR0: 0.394 | HR1: 0.416/SR1: 0.394 | LR: 1.00e-04 [2026-05-23 19:05:56] Epoch 1 | Step 3390 | Loss: 0.8013 | LM: 0.3012 | LB: 1.0037 | CL0: 2.7 | CL1: 2.4 | HR0: 0.374/SR0: 0.394 | HR1: 0.416/SR1: 0.394 | LR: 1.00e-04 [2026-05-23 19:06:09] Epoch 1 | Step 3400 | Loss: 0.8011 | LM: 0.3012 | LB: 1.0037 | CL0: 2.7 | CL1: 2.4 | HR0: 0.374/SR0: 0.394 | HR1: 0.416/SR1: 0.394 | LR: 1.00e-04 [2026-05-23 19:06:21] Epoch 1 | Step 3410 | Loss: 0.8012 | LM: 0.3012 | LB: 1.0037 | CL0: 2.7 | CL1: 2.4 | HR0: 0.374/SR0: 0.394 | HR1: 0.416/SR1: 0.394 | LR: 1.00e-04 [2026-05-23 19:06:33] Epoch 1 | Step 3420 | Loss: 0.8010 | LM: 0.3012 | LB: 1.0037 | CL0: 2.7 | CL1: 2.4 | HR0: 0.374/SR0: 0.394 | HR1: 0.416/SR1: 0.394 | LR: 1.00e-04 [2026-05-23 19:06:45] Epoch 1 | Step 3430 | Loss: 0.8009 | LM: 0.3009 | LB: 1.0037 | CL0: 2.7 | CL1: 2.4 | HR0: 0.375/SR0: 0.394 | HR1: 0.416/SR1: 0.394 | LR: 1.00e-04 [2026-05-23 19:06:57] Epoch 1 | Step 3440 | Loss: 0.8009 | LM: 0.3008 | LB: 1.0037 | CL0: 2.7 | CL1: 2.4 | HR0: 0.375/SR0: 0.394 | HR1: 0.416/SR1: 0.394 | LR: 1.00e-04 [2026-05-23 19:07:09] Epoch 1 | Step 3450 | Loss: 0.8008 | LM: 0.3006 | LB: 1.0037 | CL0: 2.7 | CL1: 2.4 | HR0: 0.375/SR0: 0.394 | HR1: 0.416/SR1: 0.394 | LR: 1.00e-04 [2026-05-23 19:07:21] Epoch 1 | Step 3460 | Loss: 0.8008 | LM: 0.3006 | LB: 1.0037 | CL0: 2.7 | CL1: 2.4 | HR0: 0.375/SR0: 0.395 | HR1: 0.416/SR1: 0.394 | LR: 1.00e-04 [2026-05-23 19:07:33] Epoch 1 | Step 3470 | Loss: 0.8007 | LM: 0.3006 | LB: 1.0037 | CL0: 2.7 | CL1: 2.4 | HR0: 0.375/SR0: 0.395 | HR1: 0.416/SR1: 0.394 | LR: 1.00e-04 [2026-05-23 19:07:45] Epoch 1 | Step 3480 | Loss: 0.8007 | LM: 0.3007 | LB: 1.0036 | CL0: 2.7 | CL1: 2.4 | HR0: 0.375/SR0: 0.395 | HR1: 0.416/SR1: 0.394 | LR: 1.00e-04 [2026-05-23 19:07:58] Epoch 1 | Step 3490 | Loss: 0.8008 | LM: 0.3007 | LB: 1.0036 | CL0: 2.7 | CL1: 2.4 | HR0: 0.375/SR0: 0.395 | HR1: 0.416/SR1: 0.394 | LR: 1.00e-04 [2026-05-23 19:08:09] Epoch 1 | Step 3500 | Loss: 0.8008 | LM: 0.3009 | LB: 1.0036 | CL0: 2.7 | CL1: 2.4 | HR0: 0.375/SR0: 0.395 | HR1: 0.416/SR1: 0.394 | LR: 1.00e-04 [2026-05-23 19:08:20] Epoch 1 | Step 3510 | Loss: 0.8008 | LM: 0.3009 | LB: 1.0036 | CL0: 2.7 | CL1: 2.4 | HR0: 0.375/SR0: 0.395 | HR1: 0.416/SR1: 0.394 | LR: 1.00e-04 [2026-05-23 19:08:33] Epoch 1 | Step 3520 | Loss: 0.8008 | LM: 0.3009 | LB: 1.0036 | CL0: 2.7 | CL1: 2.4 | HR0: 0.375/SR0: 0.395 | HR1: 0.416/SR1: 0.394 | LR: 1.00e-04 [2026-05-23 19:08:45] Epoch 1 | Step 3530 | Loss: 0.8008 | LM: 0.3008 | LB: 1.0036 | CL0: 2.7 | CL1: 2.4 | HR0: 0.375/SR0: 0.395 | HR1: 0.416/SR1: 0.394 | LR: 1.00e-04 [2026-05-23 19:08:58] Epoch 1 | Step 3540 | Loss: 0.8007 | LM: 0.3007 | LB: 1.0036 | CL0: 2.7 | CL1: 2.4 | HR0: 0.375/SR0: 0.396 | HR1: 0.416/SR1: 0.394 | LR: 1.00e-04 [2026-05-23 19:09:10] Epoch 1 | Step 3550 | Loss: 0.8007 | LM: 0.3006 | LB: 1.0036 | CL0: 2.7 | CL1: 2.4 | HR0: 0.375/SR0: 0.396 | HR1: 0.416/SR1: 0.394 | LR: 1.00e-04 [2026-05-23 19:09:22] Epoch 1 | Step 3560 | Loss: 0.8006 | LM: 0.3007 | LB: 1.0036 | CL0: 2.7 | CL1: 2.4 | HR0: 0.375/SR0: 0.396 | HR1: 0.416/SR1: 0.394 | LR: 1.00e-04 [2026-05-23 19:09:35] Epoch 1 | Step 3570 | Loss: 0.8006 | LM: 0.3007 | LB: 1.0036 | CL0: 2.7 | CL1: 2.4 | HR0: 0.375/SR0: 0.396 | HR1: 0.416/SR1: 0.394 | LR: 1.00e-04 [2026-05-23 19:09:47] Epoch 1 | Step 3580 | Loss: 0.8005 | LM: 0.3005 | LB: 1.0036 | CL0: 2.7 | CL1: 2.4 | HR0: 0.376/SR0: 0.396 | HR1: 0.415/SR1: 0.393 | LR: 1.00e-04 [2026-05-23 19:10:01] Epoch 1 | Step 3590 | Loss: 0.8004 | LM: 0.3007 | LB: 1.0036 | CL0: 2.7 | CL1: 2.4 | HR0: 0.376/SR0: 0.396 | HR1: 0.415/SR1: 0.393 | LR: 1.00e-04 [2026-05-23 19:10:12] Epoch 1 | Step 3600 | Loss: 0.8003 | LM: 0.3005 | LB: 1.0036 | CL0: 2.7 | CL1: 2.4 | HR0: 0.376/SR0: 0.396 | HR1: 0.415/SR1: 0.393 | LR: 1.00e-04 [2026-05-23 19:10:24] Epoch 1 | Step 3610 | Loss: 0.8002 | LM: 0.3004 | LB: 1.0036 | CL0: 2.7 | CL1: 2.4 | HR0: 0.376/SR0: 0.396 | HR1: 0.415/SR1: 0.393 | LR: 1.00e-04 [2026-05-23 19:10:37] Epoch 1 | Step 3620 | Loss: 0.8002 | LM: 0.3004 | LB: 1.0036 | CL0: 2.7 | CL1: 2.4 | HR0: 0.376/SR0: 0.396 | HR1: 0.415/SR1: 0.393 | LR: 1.00e-04 [2026-05-23 19:10:49] Epoch 1 | Step 3630 | Loss: 0.8001 | LM: 0.3004 | LB: 1.0035 | CL0: 2.7 | CL1: 2.4 | HR0: 0.376/SR0: 0.396 | HR1: 0.415/SR1: 0.393 | LR: 1.00e-04 [2026-05-23 19:11:02] Epoch 1 | Step 3640 | Loss: 0.8001 | LM: 0.3005 | LB: 1.0035 | CL0: 2.7 | CL1: 2.4 | HR0: 0.376/SR0: 0.397 | HR1: 0.415/SR1: 0.393 | LR: 1.00e-04 [2026-05-23 19:11:14] Epoch 1 | Step 3650 | Loss: 0.8001 | LM: 0.3006 | LB: 1.0035 | CL0: 2.7 | CL1: 2.4 | HR0: 0.376/SR0: 0.397 | HR1: 0.415/SR1: 0.393 | LR: 1.00e-04 [2026-05-23 19:11:27] Epoch 1 | Step 3660 | Loss: 0.8001 | LM: 0.3007 | LB: 1.0035 | CL0: 2.7 | CL1: 2.4 | HR0: 0.376/SR0: 0.397 | HR1: 0.415/SR1: 0.393 | LR: 1.00e-04 [2026-05-23 19:11:39] Epoch 1 | Step 3670 | Loss: 0.8001 | LM: 0.3006 | LB: 1.0035 | CL0: 2.7 | CL1: 2.4 | HR0: 0.376/SR0: 0.397 | HR1: 0.415/SR1: 0.393 | LR: 1.00e-04 [2026-05-23 19:11:52] Epoch 1 | Step 3680 | Loss: 0.8000 | LM: 0.3006 | LB: 1.0035 | CL0: 2.7 | CL1: 2.4 | HR0: 0.376/SR0: 0.397 | HR1: 0.415/SR1: 0.393 | LR: 1.00e-04 [2026-05-23 19:12:06] Epoch 1 | Step 3690 | Loss: 0.7999 | LM: 0.3006 | LB: 1.0035 | CL0: 2.7 | CL1: 2.4 | HR0: 0.376/SR0: 0.397 | HR1: 0.415/SR1: 0.393 | LR: 1.00e-04 [2026-05-23 19:12:18] Epoch 1 | Step 3700 | Loss: 0.7998 | LM: 0.3007 | LB: 1.0035 | CL0: 2.7 | CL1: 2.4 | HR0: 0.376/SR0: 0.397 | HR1: 0.415/SR1: 0.393 | LR: 1.00e-04 [2026-05-23 19:12:31] Epoch 1 | Step 3710 | Loss: 0.7998 | LM: 0.3006 | LB: 1.0035 | CL0: 2.7 | CL1: 2.4 | HR0: 0.376/SR0: 0.397 | HR1: 0.415/SR1: 0.393 | LR: 1.00e-04 [2026-05-23 19:12:44] Epoch 1 | Step 3720 | Loss: 0.7999 | LM: 0.3006 | LB: 1.0035 | CL0: 2.7 | CL1: 2.4 | HR0: 0.376/SR0: 0.397 | HR1: 0.415/SR1: 0.393 | LR: 1.00e-04 [2026-05-23 19:12:58] Epoch 1 | Step 3730 | Loss: 0.7998 | LM: 0.3004 | LB: 1.0035 | CL0: 2.7 | CL1: 2.4 | HR0: 0.376/SR0: 0.398 | HR1: 0.415/SR1: 0.393 | LR: 1.00e-04 [2026-05-23 19:13:10] Epoch 1 | Step 3740 | Loss: 0.7998 | LM: 0.3003 | LB: 1.0035 | CL0: 2.7 | CL1: 2.4 | HR0: 0.376/SR0: 0.398 | HR1: 0.415/SR1: 0.393 | LR: 1.00e-04 [2026-05-23 19:13:23] Epoch 1 | Step 3750 | Loss: 0.7997 | LM: 0.3003 | LB: 1.0035 | CL0: 2.7 | CL1: 2.4 | HR0: 0.377/SR0: 0.398 | HR1: 0.415/SR1: 0.393 | LR: 1.00e-04 [2026-05-23 19:13:36] Epoch 1 | Step 3760 | Loss: 0.7997 | LM: 0.3003 | LB: 1.0035 | CL0: 2.7 | CL1: 2.4 | HR0: 0.377/SR0: 0.398 | HR1: 0.415/SR1: 0.393 | LR: 1.00e-04 [2026-05-23 19:13:50] Epoch 1 | Step 3770 | Loss: 0.7996 | LM: 0.3003 | LB: 1.0034 | CL0: 2.7 | CL1: 2.4 | HR0: 0.377/SR0: 0.398 | HR1: 0.415/SR1: 0.393 | LR: 1.00e-04 [2026-05-23 19:14:02] Epoch 1 | Step 3780 | Loss: 0.7995 | LM: 0.3003 | LB: 1.0034 | CL0: 2.7 | CL1: 2.4 | HR0: 0.377/SR0: 0.398 | HR1: 0.415/SR1: 0.393 | LR: 1.00e-04 [2026-05-23 19:14:16] Epoch 1 | Step 3790 | Loss: 0.7996 | LM: 0.3003 | LB: 1.0034 | CL0: 2.7 | CL1: 2.4 | HR0: 0.377/SR0: 0.398 | HR1: 0.415/SR1: 0.393 | LR: 1.00e-04 [2026-05-23 19:14:28] Epoch 1 | Step 3800 | Loss: 0.7995 | LM: 0.3002 | LB: 1.0034 | CL0: 2.7 | CL1: 2.4 | HR0: 0.377/SR0: 0.398 | HR1: 0.415/SR1: 0.393 | LR: 1.00e-04 [2026-05-23 19:14:40] Epoch 1 | Step 3810 | Loss: 0.7995 | LM: 0.3001 | LB: 1.0034 | CL0: 2.7 | CL1: 2.4 | HR0: 0.377/SR0: 0.398 | HR1: 0.415/SR1: 0.393 | LR: 1.00e-04 [2026-05-23 19:14:53] Epoch 1 | Step 3820 | Loss: 0.7994 | LM: 0.3000 | LB: 1.0034 | CL0: 2.7 | CL1: 2.4 | HR0: 0.377/SR0: 0.398 | HR1: 0.415/SR1: 0.393 | LR: 1.00e-04 [2026-05-23 19:15:05] Epoch 1 | Step 3830 | Loss: 0.7993 | LM: 0.2999 | LB: 1.0034 | CL0: 2.7 | CL1: 2.4 | HR0: 0.377/SR0: 0.399 | HR1: 0.415/SR1: 0.393 | LR: 1.00e-04 [2026-05-23 19:15:18] Epoch 1 | Step 3840 | Loss: 0.7992 | LM: 0.2999 | LB: 1.0034 | CL0: 2.7 | CL1: 2.4 | HR0: 0.377/SR0: 0.399 | HR1: 0.415/SR1: 0.393 | LR: 1.00e-04 [2026-05-23 19:15:31] Epoch 1 | Step 3850 | Loss: 0.7992 | LM: 0.2998 | LB: 1.0034 | CL0: 2.7 | CL1: 2.4 | HR0: 0.377/SR0: 0.399 | HR1: 0.415/SR1: 0.393 | LR: 1.00e-04 [2026-05-23 19:15:44] Epoch 1 | Step 3860 | Loss: 0.7992 | LM: 0.2999 | LB: 1.0034 | CL0: 2.7 | CL1: 2.4 | HR0: 0.377/SR0: 0.399 | HR1: 0.415/SR1: 0.393 | LR: 1.00e-04 [2026-05-23 19:15:57] Epoch 1 | Step 3870 | Loss: 0.7992 | LM: 0.2997 | LB: 1.0034 | CL0: 2.7 | CL1: 2.4 | HR0: 0.377/SR0: 0.399 | HR1: 0.415/SR1: 0.393 | LR: 1.00e-04 [2026-05-23 19:16:11] Epoch 1 | Step 3880 | Loss: 0.7991 | LM: 0.2997 | LB: 1.0034 | CL0: 2.7 | CL1: 2.4 | HR0: 0.377/SR0: 0.399 | HR1: 0.414/SR1: 0.393 | LR: 1.00e-04 [2026-05-23 19:16:23] Epoch 1 | Step 3890 | Loss: 0.7992 | LM: 0.2997 | LB: 1.0034 | CL0: 2.7 | CL1: 2.4 | HR0: 0.377/SR0: 0.399 | HR1: 0.414/SR1: 0.393 | LR: 1.00e-04 [2026-05-23 19:16:36] Epoch 1 | Step 3900 | Loss: 0.7991 | LM: 0.2996 | LB: 1.0034 | CL0: 2.7 | CL1: 2.4 | HR0: 0.377/SR0: 0.399 | HR1: 0.414/SR1: 0.393 | LR: 1.00e-04 [2026-05-23 19:16:50] Epoch 1 | Step 3910 | Loss: 0.7990 | LM: 0.2993 | LB: 1.0034 | CL0: 2.7 | CL1: 2.4 | HR0: 0.378/SR0: 0.399 | HR1: 0.414/SR1: 0.393 | LR: 1.00e-04 [2026-05-23 19:17:02] Epoch 1 | Step 3920 | Loss: 0.7990 | LM: 0.2992 | LB: 1.0034 | CL0: 2.7 | CL1: 2.4 | HR0: 0.378/SR0: 0.399 | HR1: 0.414/SR1: 0.393 | LR: 1.00e-04 [2026-05-23 19:17:15] Epoch 1 | Step 3930 | Loss: 0.7990 | LM: 0.2992 | LB: 1.0033 | CL0: 2.7 | CL1: 2.4 | HR0: 0.378/SR0: 0.400 | HR1: 0.414/SR1: 0.393 | LR: 1.00e-04 [2026-05-23 19:17:27] Epoch 1 | Step 3940 | Loss: 0.7989 | LM: 0.2990 | LB: 1.0033 | CL0: 2.7 | CL1: 2.4 | HR0: 0.378/SR0: 0.400 | HR1: 0.414/SR1: 0.393 | LR: 1.00e-04 [2026-05-23 19:17:39] Epoch 1 | Step 3950 | Loss: 0.7988 | LM: 0.2990 | LB: 1.0033 | CL0: 2.7 | CL1: 2.4 | HR0: 0.378/SR0: 0.400 | HR1: 0.414/SR1: 0.393 | LR: 1.00e-04 [2026-05-23 19:17:52] Epoch 1 | Step 3960 | Loss: 0.7989 | LM: 0.2990 | LB: 1.0033 | CL0: 2.7 | CL1: 2.4 | HR0: 0.378/SR0: 0.400 | HR1: 0.414/SR1: 0.393 | LR: 1.00e-04 [2026-05-23 19:18:05] Epoch 1 | Step 3970 | Loss: 0.7989 | LM: 0.2989 | LB: 1.0033 | CL0: 2.7 | CL1: 2.4 | HR0: 0.378/SR0: 0.400 | HR1: 0.414/SR1: 0.393 | LR: 1.00e-04 [2026-05-23 19:18:18] Epoch 1 | Step 3980 | Loss: 0.7988 | LM: 0.2989 | LB: 1.0033 | CL0: 2.7 | CL1: 2.4 | HR0: 0.378/SR0: 0.400 | HR1: 0.414/SR1: 0.393 | LR: 1.00e-04 [2026-05-23 19:18:31] Epoch 1 | Step 3990 | Loss: 0.7988 | LM: 0.2990 | LB: 1.0033 | CL0: 2.7 | CL1: 2.4 | HR0: 0.378/SR0: 0.400 | HR1: 0.414/SR1: 0.393 | LR: 1.00e-04 [2026-05-23 19:18:44] Epoch 1 | Step 4000 | Loss: 0.7988 | LM: 0.2989 | LB: 1.0033 | CL0: 2.7 | CL1: 2.4 | HR0: 0.378/SR0: 0.400 | HR1: 0.414/SR1: 0.393 | LR: 1.00e-04 [2026-05-23 19:18:46] Validation | Batch 10/1463 | Loss: 0.7529 | LM_LOSS: 0.2530 | LB_LOSS: 0.9997 [2026-05-23 19:18:49] Validation | Batch 20/1463 | Loss: 0.7615 | LM_LOSS: 0.2614 | LB_LOSS: 1.0002 [2026-05-23 19:18:52] Validation | Batch 30/1463 | Loss: 0.7662 | LM_LOSS: 0.2660 | LB_LOSS: 1.0004 [2026-05-23 19:18:55] Validation | Batch 40/1463 | Loss: 0.7732 | LM_LOSS: 0.2730 | LB_LOSS: 1.0004 [2026-05-23 19:18:58] Validation | Batch 50/1463 | Loss: 0.7803 | LM_LOSS: 0.2801 | LB_LOSS: 1.0003 [2026-05-23 19:19:02] Validation | Batch 60/1463 | Loss: 0.7806 | LM_LOSS: 0.2804 | LB_LOSS: 1.0004 [2026-05-23 19:19:06] Validation | Batch 70/1463 | Loss: 0.7805 | LM_LOSS: 0.2803 | LB_LOSS: 1.0004 [2026-05-23 19:19:09] Validation | Batch 80/1463 | Loss: 0.7822 | LM_LOSS: 0.2819 | LB_LOSS: 1.0004 [2026-05-23 19:19:12] Validation | Batch 90/1463 | Loss: 0.7816 | LM_LOSS: 0.2814 | LB_LOSS: 1.0004 [2026-05-23 19:19:15] Validation | Batch 100/1463 | Loss: 0.7816 | LM_LOSS: 0.2814 | LB_LOSS: 1.0004 [2026-05-23 19:19:18] Validation | Batch 110/1463 | Loss: 0.7799 | LM_LOSS: 0.2797 | LB_LOSS: 1.0004 [2026-05-23 19:19:20] Validation | Batch 120/1463 | Loss: 0.7814 | LM_LOSS: 0.2812 | LB_LOSS: 1.0004 [2026-05-23 19:19:23] Validation | Batch 130/1463 | Loss: 0.7831 | LM_LOSS: 0.2828 | LB_LOSS: 1.0004 [2026-05-23 19:19:27] Validation | Batch 140/1463 | Loss: 0.7842 | LM_LOSS: 0.2839 | LB_LOSS: 1.0004 [2026-05-23 19:19:30] Validation | Batch 150/1463 | Loss: 0.7823 | LM_LOSS: 0.2820 | LB_LOSS: 1.0005 [2026-05-23 19:19:33] Validation | Batch 160/1463 | Loss: 0.7827 | LM_LOSS: 0.2824 | LB_LOSS: 1.0005 [2026-05-23 19:19:36] Validation | Batch 170/1463 | Loss: 0.7823 | LM_LOSS: 0.2820 | LB_LOSS: 1.0005 [2026-05-23 19:19:39] Validation | Batch 180/1463 | Loss: 0.7829 | LM_LOSS: 0.2827 | LB_LOSS: 1.0005 [2026-05-23 19:19:42] Validation | Batch 190/1463 | Loss: 0.7838 | LM_LOSS: 0.2836 | LB_LOSS: 1.0005 [2026-05-23 19:19:45] Validation | Batch 200/1463 | Loss: 0.7840 | LM_LOSS: 0.2837 | LB_LOSS: 1.0005 [2026-05-23 19:19:48] Validation | Batch 210/1463 | Loss: 0.7848 | LM_LOSS: 0.2846 | LB_LOSS: 1.0005 [2026-05-23 19:19:52] Validation | Batch 220/1463 | Loss: 0.7840 | LM_LOSS: 0.2838 | LB_LOSS: 1.0005 [2026-05-23 19:19:55] Validation | Batch 230/1463 | Loss: 0.7855 | LM_LOSS: 0.2852 | LB_LOSS: 1.0005 [2026-05-23 19:19:58] Validation | Batch 240/1463 | Loss: 0.7870 | LM_LOSS: 0.2868 | LB_LOSS: 1.0005 [2026-05-23 19:20:02] Validation | Batch 250/1463 | Loss: 0.7861 | LM_LOSS: 0.2858 | LB_LOSS: 1.0005 [2026-05-23 19:20:04] Validation | Batch 260/1463 | Loss: 0.7876 | LM_LOSS: 0.2873 | LB_LOSS: 1.0005 [2026-05-23 19:20:07] Validation | Batch 270/1463 | Loss: 0.7862 | LM_LOSS: 0.2859 | LB_LOSS: 1.0006 [2026-05-23 19:20:10] Validation | Batch 280/1463 | Loss: 0.7867 | LM_LOSS: 0.2864 | LB_LOSS: 1.0005 [2026-05-23 19:20:14] Validation | Batch 290/1463 | Loss: 0.7858 | LM_LOSS: 0.2856 | LB_LOSS: 1.0005 [2026-05-23 19:20:16] Validation | Batch 300/1463 | Loss: 0.7860 | LM_LOSS: 0.2858 | LB_LOSS: 1.0006 [2026-05-23 19:20:19] Validation | Batch 310/1463 | Loss: 0.7871 | LM_LOSS: 0.2868 | LB_LOSS: 1.0006 [2026-05-23 19:20:22] Validation | Batch 320/1463 | Loss: 0.7858 | LM_LOSS: 0.2856 | LB_LOSS: 1.0005 [2026-05-23 19:20:25] Validation | Batch 330/1463 | Loss: 0.7857 | LM_LOSS: 0.2854 | LB_LOSS: 1.0006 [2026-05-23 19:20:28] Validation | Batch 340/1463 | Loss: 0.7855 | LM_LOSS: 0.2852 | LB_LOSS: 1.0006 [2026-05-23 19:20:31] Validation | Batch 350/1463 | Loss: 0.7860 | LM_LOSS: 0.2857 | LB_LOSS: 1.0006 [2026-05-23 19:20:34] Validation | Batch 360/1463 | Loss: 0.7857 | LM_LOSS: 0.2854 | LB_LOSS: 1.0006 [2026-05-23 19:20:37] Validation | Batch 370/1463 | Loss: 0.7855 | LM_LOSS: 0.2852 | LB_LOSS: 1.0006 [2026-05-23 19:20:40] Validation | Batch 380/1463 | Loss: 0.7855 | LM_LOSS: 0.2852 | LB_LOSS: 1.0005 [2026-05-23 19:20:43] Validation | Batch 390/1463 | Loss: 0.7860 | LM_LOSS: 0.2857 | LB_LOSS: 1.0005 [2026-05-23 19:20:46] Validation | Batch 400/1463 | Loss: 0.7858 | LM_LOSS: 0.2855 | LB_LOSS: 1.0005 [2026-05-23 19:20:48] Validation | Batch 410/1463 | Loss: 0.7856 | LM_LOSS: 0.2853 | LB_LOSS: 1.0005 [2026-05-23 19:20:52] Validation | Batch 420/1463 | Loss: 0.7856 | LM_LOSS: 0.2853 | LB_LOSS: 1.0005 [2026-05-23 19:20:55] Validation | Batch 430/1463 | Loss: 0.7851 | LM_LOSS: 0.2848 | LB_LOSS: 1.0005 [2026-05-23 19:20:58] Validation | Batch 440/1463 | Loss: 0.7853 | LM_LOSS: 0.2850 | LB_LOSS: 1.0005 [2026-05-23 19:21:00] Validation | Batch 450/1463 | Loss: 0.7852 | LM_LOSS: 0.2849 | LB_LOSS: 1.0006 [2026-05-23 19:21:04] Validation | Batch 460/1463 | Loss: 0.7851 | LM_LOSS: 0.2848 | LB_LOSS: 1.0005 [2026-05-23 19:21:07] Validation | Batch 470/1463 | Loss: 0.7856 | LM_LOSS: 0.2854 | LB_LOSS: 1.0005 [2026-05-23 19:21:11] Validation | Batch 480/1463 | Loss: 0.7854 | LM_LOSS: 0.2851 | LB_LOSS: 1.0005 [2026-05-23 19:21:13] Validation | Batch 490/1463 | Loss: 0.7853 | LM_LOSS: 0.2850 | LB_LOSS: 1.0005 [2026-05-23 19:21:17] Validation | Batch 500/1463 | Loss: 0.7852 | LM_LOSS: 0.2849 | LB_LOSS: 1.0005 [2026-05-23 19:21:20] Validation | Batch 510/1463 | Loss: 0.7846 | LM_LOSS: 0.2843 | LB_LOSS: 1.0005 [2026-05-23 19:21:23] Validation | Batch 520/1463 | Loss: 0.7843 | LM_LOSS: 0.2840 | LB_LOSS: 1.0005 [2026-05-23 19:21:26] Validation | Batch 530/1463 | Loss: 0.7847 | LM_LOSS: 0.2844 | LB_LOSS: 1.0005 [2026-05-23 19:21:28] Validation | Batch 540/1463 | Loss: 0.7849 | LM_LOSS: 0.2846 | LB_LOSS: 1.0005 [2026-05-23 19:21:30] Validation | Batch 550/1463 | Loss: 0.7845 | LM_LOSS: 0.2842 | LB_LOSS: 1.0005 [2026-05-23 19:21:33] Validation | Batch 560/1463 | Loss: 0.7845 | LM_LOSS: 0.2843 | LB_LOSS: 1.0005 [2026-05-23 19:21:37] Validation | Batch 570/1463 | Loss: 0.7844 | LM_LOSS: 0.2842 | LB_LOSS: 1.0005 [2026-05-23 19:21:40] Validation | Batch 580/1463 | Loss: 0.7840 | LM_LOSS: 0.2838 | LB_LOSS: 1.0005 [2026-05-23 19:21:44] Validation | Batch 590/1463 | Loss: 0.7839 | LM_LOSS: 0.2836 | LB_LOSS: 1.0005 [2026-05-23 19:21:47] Validation | Batch 600/1463 | Loss: 0.7842 | LM_LOSS: 0.2839 | LB_LOSS: 1.0005 [2026-05-23 19:21:50] Validation | Batch 610/1463 | Loss: 0.7844 | LM_LOSS: 0.2841 | LB_LOSS: 1.0005 [2026-05-23 19:21:54] Validation | Batch 620/1463 | Loss: 0.7839 | LM_LOSS: 0.2837 | LB_LOSS: 1.0005 [2026-05-23 19:21:56] Validation | Batch 630/1463 | Loss: 0.7834 | LM_LOSS: 0.2831 | LB_LOSS: 1.0005 [2026-05-23 19:21:59] Validation | Batch 640/1463 | Loss: 0.7832 | LM_LOSS: 0.2829 | LB_LOSS: 1.0005 [2026-05-23 19:22:02] Validation | Batch 650/1463 | Loss: 0.7828 | LM_LOSS: 0.2825 | LB_LOSS: 1.0005 [2026-05-23 19:22:04] Validation | Batch 660/1463 | Loss: 0.7824 | LM_LOSS: 0.2821 | LB_LOSS: 1.0005 [2026-05-23 19:22:07] Validation | Batch 670/1463 | Loss: 0.7821 | LM_LOSS: 0.2819 | LB_LOSS: 1.0005 [2026-05-23 19:22:10] Validation | Batch 680/1463 | Loss: 0.7821 | LM_LOSS: 0.2818 | LB_LOSS: 1.0005 [2026-05-23 19:22:13] Validation | Batch 690/1463 | Loss: 0.7824 | LM_LOSS: 0.2821 | LB_LOSS: 1.0005 [2026-05-23 19:22:16] Validation | Batch 700/1463 | Loss: 0.7822 | LM_LOSS: 0.2820 | LB_LOSS: 1.0005 [2026-05-23 19:22:19] Validation | Batch 710/1463 | Loss: 0.7825 | LM_LOSS: 0.2822 | LB_LOSS: 1.0005 [2026-05-23 19:22:22] Validation | Batch 720/1463 | Loss: 0.7833 | LM_LOSS: 0.2831 | LB_LOSS: 1.0005 [2026-05-23 19:22:24] Validation | Batch 730/1463 | Loss: 0.7831 | LM_LOSS: 0.2829 | LB_LOSS: 1.0005 [2026-05-23 19:22:27] Validation | Batch 740/1463 | Loss: 0.7829 | LM_LOSS: 0.2826 | LB_LOSS: 1.0005 [2026-05-23 19:22:30] Validation | Batch 750/1463 | Loss: 0.7829 | LM_LOSS: 0.2826 | LB_LOSS: 1.0005 [2026-05-23 19:22:32] Validation | Batch 760/1463 | Loss: 0.7828 | LM_LOSS: 0.2825 | LB_LOSS: 1.0005 [2026-05-23 19:22:35] Validation | Batch 770/1463 | Loss: 0.7827 | LM_LOSS: 0.2824 | LB_LOSS: 1.0005 [2026-05-23 19:22:37] Validation | Batch 780/1463 | Loss: 0.7826 | LM_LOSS: 0.2823 | LB_LOSS: 1.0005 [2026-05-23 19:22:40] Validation | Batch 790/1463 | Loss: 0.7828 | LM_LOSS: 0.2825 | LB_LOSS: 1.0005 [2026-05-23 19:22:43] Validation | Batch 800/1463 | Loss: 0.7827 | LM_LOSS: 0.2824 | LB_LOSS: 1.0005 [2026-05-23 19:22:45] Validation | Batch 810/1463 | Loss: 0.7824 | LM_LOSS: 0.2821 | LB_LOSS: 1.0005 [2026-05-23 19:22:48] Validation | Batch 820/1463 | Loss: 0.7825 | LM_LOSS: 0.2822 | LB_LOSS: 1.0005 [2026-05-23 19:22:52] Validation | Batch 830/1463 | Loss: 0.7827 | LM_LOSS: 0.2825 | LB_LOSS: 1.0005 [2026-05-23 19:22:55] Validation | Batch 840/1463 | Loss: 0.7828 | LM_LOSS: 0.2826 | LB_LOSS: 1.0005 [2026-05-23 19:22:58] Validation | Batch 850/1463 | Loss: 0.7831 | LM_LOSS: 0.2828 | LB_LOSS: 1.0005 [2026-05-23 19:23:01] Validation | Batch 860/1463 | Loss: 0.7832 | LM_LOSS: 0.2830 | LB_LOSS: 1.0005 [2026-05-23 19:23:04] Validation | Batch 870/1463 | Loss: 0.7833 | LM_LOSS: 0.2830 | LB_LOSS: 1.0005 [2026-05-23 19:23:07] Validation | Batch 880/1463 | Loss: 0.7826 | LM_LOSS: 0.2823 | LB_LOSS: 1.0005 [2026-05-23 19:23:10] Validation | Batch 890/1463 | Loss: 0.7826 | LM_LOSS: 0.2823 | LB_LOSS: 1.0005 [2026-05-23 19:23:13] Validation | Batch 900/1463 | Loss: 0.7824 | LM_LOSS: 0.2821 | LB_LOSS: 1.0005 [2026-05-23 19:23:16] Validation | Batch 910/1463 | Loss: 0.7823 | LM_LOSS: 0.2821 | LB_LOSS: 1.0005 [2026-05-23 19:23:20] Validation | Batch 920/1463 | Loss: 0.7825 | LM_LOSS: 0.2822 | LB_LOSS: 1.0005 [2026-05-23 19:23:23] Validation | Batch 930/1463 | Loss: 0.7823 | LM_LOSS: 0.2820 | LB_LOSS: 1.0005 [2026-05-23 19:23:26] Validation | Batch 940/1463 | Loss: 0.7820 | LM_LOSS: 0.2818 | LB_LOSS: 1.0005 [2026-05-23 19:23:29] Validation | Batch 950/1463 | Loss: 0.7822 | LM_LOSS: 0.2820 | LB_LOSS: 1.0005 [2026-05-23 19:23:32] Validation | Batch 960/1463 | Loss: 0.7823 | LM_LOSS: 0.2820 | LB_LOSS: 1.0005 [2026-05-23 19:23:36] Validation | Batch 970/1463 | Loss: 0.7823 | LM_LOSS: 0.2821 | LB_LOSS: 1.0005 [2026-05-23 19:23:39] Validation | Batch 980/1463 | Loss: 0.7826 | LM_LOSS: 0.2824 | LB_LOSS: 1.0005 [2026-05-23 19:23:42] Validation | Batch 990/1463 | Loss: 0.7829 | LM_LOSS: 0.2826 | LB_LOSS: 1.0005 [2026-05-23 19:23:45] Validation | Batch 1000/1463 | Loss: 0.7834 | LM_LOSS: 0.2831 | LB_LOSS: 1.0005 [2026-05-23 19:23:48] Validation | Batch 1010/1463 | Loss: 0.7834 | LM_LOSS: 0.2832 | LB_LOSS: 1.0005 [2026-05-23 19:23:51] Validation | Batch 1020/1463 | Loss: 0.7835 | LM_LOSS: 0.2832 | LB_LOSS: 1.0005 [2026-05-23 19:23:54] Validation | Batch 1030/1463 | Loss: 0.7839 | LM_LOSS: 0.2836 | LB_LOSS: 1.0005 [2026-05-23 19:23:57] Validation | Batch 1040/1463 | Loss: 0.7836 | LM_LOSS: 0.2833 | LB_LOSS: 1.0005 [2026-05-23 19:24:00] Validation | Batch 1050/1463 | Loss: 0.7835 | LM_LOSS: 0.2832 | LB_LOSS: 1.0005 [2026-05-23 19:24:03] Validation | Batch 1060/1463 | Loss: 0.7834 | LM_LOSS: 0.2832 | LB_LOSS: 1.0005 [2026-05-23 19:24:06] Validation | Batch 1070/1463 | Loss: 0.7834 | LM_LOSS: 0.2832 | LB_LOSS: 1.0005 [2026-05-23 19:24:09] Validation | Batch 1080/1463 | Loss: 0.7833 | LM_LOSS: 0.2830 | LB_LOSS: 1.0005 [2026-05-23 19:24:12] Validation | Batch 1090/1463 | Loss: 0.7833 | LM_LOSS: 0.2831 | LB_LOSS: 1.0005 [2026-05-23 19:24:15] Validation | Batch 1100/1463 | Loss: 0.7831 | LM_LOSS: 0.2829 | LB_LOSS: 1.0005 [2026-05-23 19:24:17] Validation | Batch 1110/1463 | Loss: 0.7831 | LM_LOSS: 0.2829 | LB_LOSS: 1.0005 [2026-05-23 19:24:20] Validation | Batch 1120/1463 | Loss: 0.7829 | LM_LOSS: 0.2827 | LB_LOSS: 1.0005 [2026-05-23 19:24:24] Validation | Batch 1130/1463 | Loss: 0.7827 | LM_LOSS: 0.2825 | LB_LOSS: 1.0005 [2026-05-23 19:24:27] Validation | Batch 1140/1463 | Loss: 0.7829 | LM_LOSS: 0.2827 | LB_LOSS: 1.0005 [2026-05-23 19:24:30] Validation | Batch 1150/1463 | Loss: 0.7829 | LM_LOSS: 0.2826 | LB_LOSS: 1.0005 [2026-05-23 19:24:33] Validation | Batch 1160/1463 | Loss: 0.7828 | LM_LOSS: 0.2825 | LB_LOSS: 1.0005 [2026-05-23 19:24:36] Validation | Batch 1170/1463 | Loss: 0.7828 | LM_LOSS: 0.2825 | LB_LOSS: 1.0005 [2026-05-23 19:24:39] Validation | Batch 1180/1463 | Loss: 0.7828 | LM_LOSS: 0.2826 | LB_LOSS: 1.0005 [2026-05-23 19:24:43] Validation | Batch 1190/1463 | Loss: 0.7830 | LM_LOSS: 0.2827 | LB_LOSS: 1.0005 [2026-05-23 19:24:46] Validation | Batch 1200/1463 | Loss: 0.7830 | LM_LOSS: 0.2827 | LB_LOSS: 1.0005 [2026-05-23 19:24:49] Validation | Batch 1210/1463 | Loss: 0.7832 | LM_LOSS: 0.2829 | LB_LOSS: 1.0005 [2026-05-23 19:24:51] Validation | Batch 1220/1463 | Loss: 0.7835 | LM_LOSS: 0.2833 | LB_LOSS: 1.0005 [2026-05-23 19:24:54] Validation | Batch 1230/1463 | Loss: 0.7836 | LM_LOSS: 0.2833 | LB_LOSS: 1.0005 [2026-05-23 19:24:57] Validation | Batch 1240/1463 | Loss: 0.7838 | LM_LOSS: 0.2835 | LB_LOSS: 1.0005 [2026-05-23 19:25:01] Validation | Batch 1250/1463 | Loss: 0.7839 | LM_LOSS: 0.2837 | LB_LOSS: 1.0005 [2026-05-23 19:25:04] Validation | Batch 1260/1463 | Loss: 0.7839 | LM_LOSS: 0.2836 | LB_LOSS: 1.0005 [2026-05-23 19:25:07] Validation | Batch 1270/1463 | Loss: 0.7837 | LM_LOSS: 0.2834 | LB_LOSS: 1.0005 [2026-05-23 19:25:10] Validation | Batch 1280/1463 | Loss: 0.7836 | LM_LOSS: 0.2833 | LB_LOSS: 1.0005 [2026-05-23 19:25:13] Validation | Batch 1290/1463 | Loss: 0.7836 | LM_LOSS: 0.2833 | LB_LOSS: 1.0005 [2026-05-23 19:25:16] Validation | Batch 1300/1463 | Loss: 0.7834 | LM_LOSS: 0.2831 | LB_LOSS: 1.0005 [2026-05-23 19:25:19] Validation | Batch 1310/1463 | Loss: 0.7833 | LM_LOSS: 0.2830 | LB_LOSS: 1.0005 [2026-05-23 19:25:21] Validation | Batch 1320/1463 | Loss: 0.7834 | LM_LOSS: 0.2831 | LB_LOSS: 1.0005 [2026-05-23 19:25:24] Validation | Batch 1330/1463 | Loss: 0.7834 | LM_LOSS: 0.2831 | LB_LOSS: 1.0005 [2026-05-23 19:25:27] Validation | Batch 1340/1463 | Loss: 0.7836 | LM_LOSS: 0.2834 | LB_LOSS: 1.0005 [2026-05-23 19:25:30] Validation | Batch 1350/1463 | Loss: 0.7836 | LM_LOSS: 0.2834 | LB_LOSS: 1.0005 [2026-05-23 19:25:34] Validation | Batch 1360/1463 | Loss: 0.7839 | LM_LOSS: 0.2836 | LB_LOSS: 1.0005 [2026-05-23 19:25:37] Validation | Batch 1370/1463 | Loss: 0.7842 | LM_LOSS: 0.2839 | LB_LOSS: 1.0005 [2026-05-23 19:25:40] Validation | Batch 1380/1463 | Loss: 0.7840 | LM_LOSS: 0.2837 | LB_LOSS: 1.0005 [2026-05-23 19:25:42] Validation | Batch 1390/1463 | Loss: 0.7839 | LM_LOSS: 0.2837 | LB_LOSS: 1.0005 [2026-05-23 19:25:46] Validation | Batch 1400/1463 | Loss: 0.7842 | LM_LOSS: 0.2839 | LB_LOSS: 1.0005 [2026-05-23 19:25:48] Validation | Batch 1410/1463 | Loss: 0.7845 | LM_LOSS: 0.2842 | LB_LOSS: 1.0005 [2026-05-23 19:25:51] Validation | Batch 1420/1463 | Loss: 0.7844 | LM_LOSS: 0.2842 | LB_LOSS: 1.0005 [2026-05-23 19:25:54] Validation | Batch 1430/1463 | Loss: 0.7845 | LM_LOSS: 0.2842 | LB_LOSS: 1.0005 [2026-05-23 19:25:58] Validation | Batch 1440/1463 | Loss: 0.7849 | LM_LOSS: 0.2846 | LB_LOSS: 1.0005 [2026-05-23 19:26:01] Validation | Batch 1450/1463 | Loss: 0.7846 | LM_LOSS: 0.2843 | LB_LOSS: 1.0005 [2026-05-23 19:26:04] Validation | Batch 1460/1463 | Loss: 0.7849 | LM_LOSS: 0.2846 | LB_LOSS: 1.0005 [2026-05-23 19:26:05] Validation | Batch 1463/1463 | Loss: 0.7848 | LM_LOSS: 0.2845 | LB_LOSS: 1.0005 [2026-05-23 19:26:05] Validation | Loss: 0.7848 | LM_LOSS: 0.2845 | LB_LOSS: 1.0005 | PPL: 1.33 | Time: 441.16s [2026-05-23 19:26:12] New best model saved! Val loss: 0.7848 [2026-05-23 19:26:25] Epoch 1 | Step 4010 | Loss: 0.7986 | LM: 0.2987 | LB: 1.0033 | CL0: 2.7 | CL1: 2.4 | HR0: 0.378/SR0: 0.400 | HR1: 0.414/SR1: 0.393 | LR: 1.00e-04 [2026-05-23 19:26:38] Epoch 1 | Step 4020 | Loss: 0.7986 | LM: 0.2987 | LB: 1.0033 | CL0: 2.7 | CL1: 2.4 | HR0: 0.378/SR0: 0.400 | HR1: 0.414/SR1: 0.393 | LR: 1.00e-04 [2026-05-23 19:26:50] Epoch 1 | Step 4030 | Loss: 0.7987 | LM: 0.2986 | LB: 1.0033 | CL0: 2.7 | CL1: 2.4 | HR0: 0.378/SR0: 0.400 | HR1: 0.414/SR1: 0.393 | LR: 1.00e-04 [2026-05-23 19:27:03] Epoch 1 | Step 4040 | Loss: 0.7987 | LM: 0.2985 | LB: 1.0033 | CL0: 2.7 | CL1: 2.4 | HR0: 0.378/SR0: 0.401 | HR1: 0.414/SR1: 0.393 | LR: 1.00e-04 [2026-05-23 19:27:15] Epoch 1 | Step 4050 | Loss: 0.7986 | LM: 0.2984 | LB: 1.0033 | CL0: 2.7 | CL1: 2.4 | HR0: 0.378/SR0: 0.401 | HR1: 0.414/SR1: 0.393 | LR: 1.00e-04 [2026-05-23 19:27:28] Epoch 1 | Step 4060 | Loss: 0.7986 | LM: 0.2985 | LB: 1.0033 | CL0: 2.7 | CL1: 2.4 | HR0: 0.378/SR0: 0.401 | HR1: 0.414/SR1: 0.393 | LR: 1.00e-04 [2026-05-23 19:27:40] Epoch 1 | Step 4070 | Loss: 0.7986 | LM: 0.2984 | LB: 1.0033 | CL0: 2.7 | CL1: 2.4 | HR0: 0.378/SR0: 0.401 | HR1: 0.414/SR1: 0.393 | LR: 1.00e-04 [2026-05-23 19:27:52] Epoch 1 | Step 4080 | Loss: 0.7986 | LM: 0.2985 | LB: 1.0033 | CL0: 2.7 | CL1: 2.4 | HR0: 0.378/SR0: 0.401 | HR1: 0.414/SR1: 0.393 | LR: 1.00e-04 [2026-05-23 19:28:04] Epoch 1 | Step 4090 | Loss: 0.7985 | LM: 0.2985 | LB: 1.0033 | CL0: 2.7 | CL1: 2.4 | HR0: 0.378/SR0: 0.401 | HR1: 0.414/SR1: 0.393 | LR: 1.00e-04 [2026-05-23 19:28:16] Epoch 1 | Step 4100 | Loss: 0.7985 | LM: 0.2984 | LB: 1.0033 | CL0: 2.7 | CL1: 2.4 | HR0: 0.378/SR0: 0.401 | HR1: 0.414/SR1: 0.393 | LR: 1.00e-04 [2026-05-23 19:28:29] Epoch 1 | Step 4110 | Loss: 0.7985 | LM: 0.2986 | LB: 1.0032 | CL0: 2.7 | CL1: 2.4 | HR0: 0.379/SR0: 0.401 | HR1: 0.414/SR1: 0.393 | LR: 1.00e-04 [2026-05-23 19:28:42] Epoch 1 | Step 4120 | Loss: 0.7985 | LM: 0.2985 | LB: 1.0032 | CL0: 2.7 | CL1: 2.4 | HR0: 0.379/SR0: 0.401 | HR1: 0.414/SR1: 0.393 | LR: 1.00e-04 [2026-05-23 19:28:54] Epoch 1 | Step 4130 | Loss: 0.7984 | LM: 0.2985 | LB: 1.0032 | CL0: 2.7 | CL1: 2.4 | HR0: 0.379/SR0: 0.401 | HR1: 0.414/SR1: 0.393 | LR: 1.00e-04 [2026-05-23 19:29:06] Epoch 1 | Step 4140 | Loss: 0.7983 | LM: 0.2984 | LB: 1.0032 | CL0: 2.7 | CL1: 2.4 | HR0: 0.379/SR0: 0.401 | HR1: 0.414/SR1: 0.393 | LR: 1.00e-04 [2026-05-23 19:29:19] Epoch 1 | Step 4150 | Loss: 0.7983 | LM: 0.2984 | LB: 1.0032 | CL0: 2.7 | CL1: 2.4 | HR0: 0.379/SR0: 0.401 | HR1: 0.414/SR1: 0.393 | LR: 1.00e-04 [2026-05-23 19:29:32] Epoch 1 | Step 4160 | Loss: 0.7982 | LM: 0.2984 | LB: 1.0032 | CL0: 2.7 | CL1: 2.4 | HR0: 0.379/SR0: 0.402 | HR1: 0.414/SR1: 0.393 | LR: 1.00e-04 [2026-05-23 19:29:44] Epoch 1 | Step 4170 | Loss: 0.7982 | LM: 0.2984 | LB: 1.0032 | CL0: 2.7 | CL1: 2.4 | HR0: 0.379/SR0: 0.402 | HR1: 0.414/SR1: 0.393 | LR: 1.00e-04 [2026-05-23 19:29:56] Epoch 1 | Step 4180 | Loss: 0.7982 | LM: 0.2984 | LB: 1.0032 | CL0: 2.7 | CL1: 2.4 | HR0: 0.379/SR0: 0.402 | HR1: 0.414/SR1: 0.393 | LR: 1.00e-04 [2026-05-23 19:30:09] Epoch 1 | Step 4190 | Loss: 0.7981 | LM: 0.2984 | LB: 1.0032 | CL0: 2.7 | CL1: 2.4 | HR0: 0.379/SR0: 0.402 | HR1: 0.414/SR1: 0.393 | LR: 1.00e-04 [2026-05-23 19:30:21] Epoch 1 | Step 4200 | Loss: 0.7981 | LM: 0.2984 | LB: 1.0032 | CL0: 2.7 | CL1: 2.4 | HR0: 0.379/SR0: 0.402 | HR1: 0.414/SR1: 0.393 | LR: 1.00e-04 [2026-05-23 19:30:34] Epoch 1 | Step 4210 | Loss: 0.7981 | LM: 0.2985 | LB: 1.0032 | CL0: 2.7 | CL1: 2.4 | HR0: 0.379/SR0: 0.402 | HR1: 0.414/SR1: 0.393 | LR: 1.00e-04 [2026-05-23 19:30:47] Epoch 1 | Step 4220 | Loss: 0.7980 | LM: 0.2983 | LB: 1.0032 | CL0: 2.7 | CL1: 2.4 | HR0: 0.379/SR0: 0.402 | HR1: 0.413/SR1: 0.393 | LR: 1.00e-04 [2026-05-23 19:31:00] Epoch 1 | Step 4230 | Loss: 0.7980 | LM: 0.2981 | LB: 1.0032 | CL0: 2.7 | CL1: 2.4 | HR0: 0.379/SR0: 0.402 | HR1: 0.413/SR1: 0.393 | LR: 1.00e-04 [2026-05-23 19:31:12] Epoch 1 | Step 4240 | Loss: 0.7980 | LM: 0.2981 | LB: 1.0032 | CL0: 2.7 | CL1: 2.4 | HR0: 0.379/SR0: 0.402 | HR1: 0.413/SR1: 0.393 | LR: 1.00e-04 [2026-05-23 19:31:24] Epoch 1 | Step 4250 | Loss: 0.7979 | LM: 0.2980 | LB: 1.0032 | CL0: 2.7 | CL1: 2.4 | HR0: 0.379/SR0: 0.402 | HR1: 0.413/SR1: 0.393 | LR: 1.00e-04 [2026-05-23 19:31:36] Epoch 1 | Step 4260 | Loss: 0.7978 | LM: 0.2978 | LB: 1.0032 | CL0: 2.7 | CL1: 2.4 | HR0: 0.379/SR0: 0.402 | HR1: 0.413/SR1: 0.393 | LR: 1.00e-04 [2026-05-23 19:31:48] Epoch 1 | Step 4270 | Loss: 0.7978 | LM: 0.2976 | LB: 1.0032 | CL0: 2.7 | CL1: 2.4 | HR0: 0.379/SR0: 0.402 | HR1: 0.413/SR1: 0.393 | LR: 1.00e-04 [2026-05-23 19:32:00] Epoch 1 | Step 4280 | Loss: 0.7977 | LM: 0.2975 | LB: 1.0032 | CL0: 2.7 | CL1: 2.4 | HR0: 0.379/SR0: 0.403 | HR1: 0.413/SR1: 0.393 | LR: 1.00e-04 [2026-05-23 19:32:13] Epoch 1 | Step 4290 | Loss: 0.7977 | LM: 0.2975 | LB: 1.0031 | CL0: 2.6 | CL1: 2.4 | HR0: 0.379/SR0: 0.403 | HR1: 0.413/SR1: 0.393 | LR: 1.00e-04 [2026-05-23 19:32:27] Epoch 1 | Step 4300 | Loss: 0.7977 | LM: 0.2976 | LB: 1.0031 | CL0: 2.6 | CL1: 2.4 | HR0: 0.379/SR0: 0.403 | HR1: 0.413/SR1: 0.393 | LR: 1.00e-04 [2026-05-23 19:32:39] Epoch 1 | Step 4310 | Loss: 0.7977 | LM: 0.2976 | LB: 1.0031 | CL0: 2.6 | CL1: 2.4 | HR0: 0.380/SR0: 0.403 | HR1: 0.413/SR1: 0.393 | LR: 1.00e-04 [2026-05-23 19:32:53] Epoch 1 | Step 4320 | Loss: 0.7977 | LM: 0.2976 | LB: 1.0031 | CL0: 2.6 | CL1: 2.4 | HR0: 0.380/SR0: 0.403 | HR1: 0.413/SR1: 0.393 | LR: 1.00e-04 [2026-05-23 19:33:06] Epoch 1 | Step 4330 | Loss: 0.7976 | LM: 0.2976 | LB: 1.0031 | CL0: 2.6 | CL1: 2.4 | HR0: 0.380/SR0: 0.403 | HR1: 0.413/SR1: 0.393 | LR: 1.00e-04 [2026-05-23 19:33:18] Epoch 1 | Step 4340 | Loss: 0.7976 | LM: 0.2976 | LB: 1.0031 | CL0: 2.6 | CL1: 2.4 | HR0: 0.380/SR0: 0.403 | HR1: 0.413/SR1: 0.393 | LR: 1.00e-04 [2026-05-23 19:33:32] Epoch 1 | Step 4350 | Loss: 0.7976 | LM: 0.2977 | LB: 1.0031 | CL0: 2.6 | CL1: 2.4 | HR0: 0.380/SR0: 0.403 | HR1: 0.413/SR1: 0.393 | LR: 1.00e-04 [2026-05-23 19:33:44] Epoch 1 | Step 4360 | Loss: 0.7975 | LM: 0.2975 | LB: 1.0031 | CL0: 2.6 | CL1: 2.4 | HR0: 0.380/SR0: 0.403 | HR1: 0.413/SR1: 0.393 | LR: 1.00e-04 [2026-05-23 19:33:56] Epoch 1 | Step 4370 | Loss: 0.7975 | LM: 0.2974 | LB: 1.0031 | CL0: 2.6 | CL1: 2.4 | HR0: 0.380/SR0: 0.403 | HR1: 0.413/SR1: 0.393 | LR: 1.00e-04 [2026-05-23 19:34:09] Epoch 1 | Step 4380 | Loss: 0.7974 | LM: 0.2973 | LB: 1.0031 | CL0: 2.6 | CL1: 2.4 | HR0: 0.380/SR0: 0.403 | HR1: 0.413/SR1: 0.393 | LR: 1.00e-04 [2026-05-23 19:34:22] Epoch 1 | Step 4390 | Loss: 0.7973 | LM: 0.2971 | LB: 1.0031 | CL0: 2.6 | CL1: 2.4 | HR0: 0.380/SR0: 0.403 | HR1: 0.413/SR1: 0.393 | LR: 1.00e-04 [2026-05-23 19:34:35] Epoch 1 | Step 4400 | Loss: 0.7973 | LM: 0.2971 | LB: 1.0031 | CL0: 2.6 | CL1: 2.4 | HR0: 0.380/SR0: 0.404 | HR1: 0.413/SR1: 0.393 | LR: 1.00e-04 [2026-05-23 19:34:48] Epoch 1 | Step 4410 | Loss: 0.7972 | LM: 0.2969 | LB: 1.0031 | CL0: 2.6 | CL1: 2.4 | HR0: 0.380/SR0: 0.404 | HR1: 0.413/SR1: 0.393 | LR: 1.00e-04 [2026-05-23 19:35:01] Epoch 1 | Step 4420 | Loss: 0.7972 | LM: 0.2969 | LB: 1.0031 | CL0: 2.6 | CL1: 2.4 | HR0: 0.380/SR0: 0.404 | HR1: 0.413/SR1: 0.392 | LR: 1.00e-04 [2026-05-23 19:35:13] Epoch 1 | Step 4430 | Loss: 0.7971 | LM: 0.2967 | LB: 1.0031 | CL0: 2.6 | CL1: 2.4 | HR0: 0.380/SR0: 0.404 | HR1: 0.413/SR1: 0.392 | LR: 1.00e-04 [2026-05-23 19:35:25] Epoch 1 | Step 4440 | Loss: 0.7971 | LM: 0.2967 | LB: 1.0031 | CL0: 2.6 | CL1: 2.4 | HR0: 0.380/SR0: 0.404 | HR1: 0.413/SR1: 0.392 | LR: 1.00e-04 [2026-05-23 19:35:38] Epoch 1 | Step 4450 | Loss: 0.7971 | LM: 0.2966 | LB: 1.0031 | CL0: 2.6 | CL1: 2.4 | HR0: 0.380/SR0: 0.404 | HR1: 0.413/SR1: 0.392 | LR: 1.00e-04 [2026-05-23 19:35:50] Epoch 1 | Step 4460 | Loss: 0.7969 | LM: 0.2965 | LB: 1.0031 | CL0: 2.6 | CL1: 2.4 | HR0: 0.380/SR0: 0.404 | HR1: 0.413/SR1: 0.392 | LR: 1.00e-04 [2026-05-23 19:36:03] Epoch 1 | Step 4470 | Loss: 0.7969 | LM: 0.2965 | LB: 1.0031 | CL0: 2.6 | CL1: 2.4 | HR0: 0.380/SR0: 0.404 | HR1: 0.413/SR1: 0.392 | LR: 1.00e-04 [2026-05-23 19:36:16] Epoch 1 | Step 4480 | Loss: 0.7969 | LM: 0.2966 | LB: 1.0031 | CL0: 2.6 | CL1: 2.4 | HR0: 0.380/SR0: 0.404 | HR1: 0.413/SR1: 0.392 | LR: 1.00e-04 [2026-05-23 19:36:28] Epoch 1 | Step 4490 | Loss: 0.7968 | LM: 0.2965 | LB: 1.0031 | CL0: 2.6 | CL1: 2.4 | HR0: 0.380/SR0: 0.404 | HR1: 0.413/SR1: 0.392 | LR: 1.00e-04 [2026-05-23 19:36:41] Epoch 1 | Step 4500 | Loss: 0.7968 | LM: 0.2964 | LB: 1.0031 | CL0: 2.6 | CL1: 2.4 | HR0: 0.380/SR0: 0.404 | HR1: 0.413/SR1: 0.392 | LR: 1.00e-04 [2026-05-23 19:36:53] Epoch 1 | Step 4510 | Loss: 0.7968 | LM: 0.2963 | LB: 1.0031 | CL0: 2.6 | CL1: 2.4 | HR0: 0.380/SR0: 0.404 | HR1: 0.413/SR1: 0.392 | LR: 1.00e-04 [2026-05-23 19:37:06] Epoch 1 | Step 4520 | Loss: 0.7968 | LM: 0.2964 | LB: 1.0030 | CL0: 2.6 | CL1: 2.4 | HR0: 0.380/SR0: 0.404 | HR1: 0.413/SR1: 0.392 | LR: 1.00e-04 [2026-05-23 19:37:18] Epoch 1 | Step 4530 | Loss: 0.7967 | LM: 0.2963 | LB: 1.0030 | CL0: 2.6 | CL1: 2.4 | HR0: 0.380/SR0: 0.404 | HR1: 0.413/SR1: 0.392 | LR: 1.00e-04 [2026-05-23 19:37:31] Epoch 1 | Step 4540 | Loss: 0.7967 | LM: 0.2963 | LB: 1.0030 | CL0: 2.6 | CL1: 2.4 | HR0: 0.380/SR0: 0.404 | HR1: 0.413/SR1: 0.392 | LR: 1.00e-04 [2026-05-23 19:37:43] Epoch 1 | Step 4550 | Loss: 0.7966 | LM: 0.2963 | LB: 1.0030 | CL0: 2.6 | CL1: 2.4 | HR0: 0.380/SR0: 0.405 | HR1: 0.413/SR1: 0.392 | LR: 1.00e-04 [2026-05-23 19:37:56] Epoch 1 | Step 4560 | Loss: 0.7965 | LM: 0.2961 | LB: 1.0030 | CL0: 2.6 | CL1: 2.4 | HR0: 0.380/SR0: 0.405 | HR1: 0.413/SR1: 0.392 | LR: 1.00e-04 [2026-05-23 19:38:09] Epoch 1 | Step 4570 | Loss: 0.7965 | LM: 0.2962 | LB: 1.0030 | CL0: 2.6 | CL1: 2.4 | HR0: 0.381/SR0: 0.405 | HR1: 0.413/SR1: 0.392 | LR: 1.00e-04 [2026-05-23 19:38:22] Epoch 1 | Step 4580 | Loss: 0.7965 | LM: 0.2962 | LB: 1.0030 | CL0: 2.6 | CL1: 2.4 | HR0: 0.381/SR0: 0.405 | HR1: 0.413/SR1: 0.392 | LR: 1.00e-04 [2026-05-23 19:38:35] Epoch 1 | Step 4590 | Loss: 0.7965 | LM: 0.2961 | LB: 1.0030 | CL0: 2.6 | CL1: 2.4 | HR0: 0.381/SR0: 0.405 | HR1: 0.413/SR1: 0.392 | LR: 1.00e-04 [2026-05-23 19:38:46] Epoch 1 | Step 4600 | Loss: 0.7965 | LM: 0.2960 | LB: 1.0030 | CL0: 2.6 | CL1: 2.4 | HR0: 0.381/SR0: 0.405 | HR1: 0.412/SR1: 0.392 | LR: 1.00e-04 [2026-05-23 19:38:59] Epoch 1 | Step 4610 | Loss: 0.7964 | LM: 0.2959 | LB: 1.0030 | CL0: 2.6 | CL1: 2.4 | HR0: 0.381/SR0: 0.405 | HR1: 0.412/SR1: 0.392 | LR: 1.00e-04 [2026-05-23 19:39:12] Epoch 1 | Step 4620 | Loss: 0.7964 | LM: 0.2960 | LB: 1.0030 | CL0: 2.6 | CL1: 2.4 | HR0: 0.381/SR0: 0.405 | HR1: 0.412/SR1: 0.392 | LR: 1.00e-04 [2026-05-23 19:39:24] Epoch 1 | Step 4630 | Loss: 0.7963 | LM: 0.2959 | LB: 1.0030 | CL0: 2.6 | CL1: 2.4 | HR0: 0.381/SR0: 0.405 | HR1: 0.412/SR1: 0.392 | LR: 1.00e-04 [2026-05-23 19:39:36] Epoch 1 | Step 4640 | Loss: 0.7963 | LM: 0.2958 | LB: 1.0030 | CL0: 2.6 | CL1: 2.4 | HR0: 0.381/SR0: 0.405 | HR1: 0.412/SR1: 0.392 | LR: 1.00e-04 [2026-05-23 19:39:49] Epoch 1 | Step 4650 | Loss: 0.7963 | LM: 0.2957 | LB: 1.0030 | CL0: 2.6 | CL1: 2.4 | HR0: 0.381/SR0: 0.405 | HR1: 0.412/SR1: 0.392 | LR: 1.00e-04 [2026-05-23 19:40:02] Epoch 1 | Step 4660 | Loss: 0.7962 | LM: 0.2956 | LB: 1.0030 | CL0: 2.6 | CL1: 2.4 | HR0: 0.381/SR0: 0.405 | HR1: 0.412/SR1: 0.392 | LR: 1.00e-04 [2026-05-23 19:40:14] Epoch 1 | Step 4670 | Loss: 0.7961 | LM: 0.2955 | LB: 1.0030 | CL0: 2.6 | CL1: 2.4 | HR0: 0.381/SR0: 0.405 | HR1: 0.412/SR1: 0.392 | LR: 1.00e-04 [2026-05-23 19:40:26] Epoch 1 | Step 4680 | Loss: 0.7961 | LM: 0.2955 | LB: 1.0030 | CL0: 2.6 | CL1: 2.4 | HR0: 0.381/SR0: 0.405 | HR1: 0.412/SR1: 0.392 | LR: 1.00e-04 [2026-05-23 19:40:37] Epoch 1 completed in 7665.28s | Loss: 0.7961 | CL0: 2.6 | CL1: 2.4 [2026-05-23 19:40:37] ============================================================ [2026-05-23 19:40:37] EPOCH 2/3 (step 4687) [2026-05-23 19:40:37] ============================================================ [2026-05-23 19:40:40] Epoch 2 | Step 4690 | Loss: 0.7223 | LM: 0.1851 | LB: 1.0008 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.395/SR1: 0.387 | LR: 1.00e-04 [2026-05-23 19:40:53] Epoch 2 | Step 4700 | Loss: 0.7326 | LM: 0.2470 | LB: 1.0006 | CL0: 2.5 | CL1: 2.5 | HR0: 0.399/SR0: 0.437 | HR1: 0.397/SR1: 0.388 | LR: 1.00e-04 [2026-05-23 19:41:05] Epoch 2 | Step 4710 | Loss: 0.7301 | LM: 0.2520 | LB: 1.0019 | CL0: 2.5 | CL1: 2.5 | HR0: 0.407/SR0: 0.442 | HR1: 0.403/SR1: 0.391 | LR: 1.00e-04 [2026-05-23 19:41:18] Epoch 2 | Step 4720 | Loss: 0.7258 | LM: 0.2463 | LB: 1.0017 | CL0: 2.5 | CL1: 2.5 | HR0: 0.405/SR0: 0.441 | HR1: 0.406/SR1: 0.391 | LR: 1.00e-04 [2026-05-23 19:41:31] Epoch 2 | Step 4730 | Loss: 0.7248 | LM: 0.2466 | LB: 1.0018 | CL0: 2.5 | CL1: 2.5 | HR0: 0.405/SR0: 0.442 | HR1: 0.408/SR1: 0.393 | LR: 1.00e-04 [2026-05-23 19:41:44] Epoch 2 | Step 4740 | Loss: 0.7215 | LM: 0.2362 | LB: 1.0020 | CL0: 2.5 | CL1: 2.5 | HR0: 0.407/SR0: 0.443 | HR1: 0.407/SR1: 0.392 | LR: 1.00e-04 [2026-05-23 19:41:58] Epoch 2 | Step 4750 | Loss: 0.7182 | LM: 0.2345 | LB: 1.0020 | CL0: 2.5 | CL1: 2.5 | HR0: 0.408/SR0: 0.443 | HR1: 0.407/SR1: 0.391 | LR: 1.00e-04 [2026-05-23 19:42:11] Epoch 2 | Step 4760 | Loss: 0.7188 | LM: 0.2352 | LB: 1.0018 | CL0: 2.5 | CL1: 2.5 | HR0: 0.406/SR0: 0.442 | HR1: 0.406/SR1: 0.391 | LR: 1.00e-04 [2026-05-23 19:42:23] Epoch 2 | Step 4770 | Loss: 0.7186 | LM: 0.2379 | LB: 1.0017 | CL0: 2.5 | CL1: 2.5 | HR0: 0.405/SR0: 0.442 | HR1: 0.406/SR1: 0.391 | LR: 1.00e-04 [2026-05-23 19:42:36] Epoch 2 | Step 4780 | Loss: 0.7170 | LM: 0.2385 | LB: 1.0017 | CL0: 2.5 | CL1: 2.5 | HR0: 0.404/SR0: 0.441 | HR1: 0.406/SR1: 0.391 | LR: 1.00e-04 [2026-05-23 19:42:49] Epoch 2 | Step 4790 | Loss: 0.7173 | LM: 0.2360 | LB: 1.0017 | CL0: 2.5 | CL1: 2.5 | HR0: 0.404/SR0: 0.441 | HR1: 0.405/SR1: 0.390 | LR: 1.00e-04 [2026-05-23 19:43:01] Epoch 2 | Step 4800 | Loss: 0.7172 | LM: 0.2374 | LB: 1.0016 | CL0: 2.5 | CL1: 2.5 | HR0: 0.403/SR0: 0.440 | HR1: 0.404/SR1: 0.389 | LR: 1.00e-04 [2026-05-23 19:43:14] Epoch 2 | Step 4810 | Loss: 0.7188 | LM: 0.2379 | LB: 1.0018 | CL0: 2.5 | CL1: 2.5 | HR0: 0.404/SR0: 0.441 | HR1: 0.403/SR1: 0.389 | LR: 1.00e-04 [2026-05-23 19:43:27] Epoch 2 | Step 4820 | Loss: 0.7174 | LM: 0.2343 | LB: 1.0017 | CL0: 2.5 | CL1: 2.5 | HR0: 0.404/SR0: 0.441 | HR1: 0.403/SR1: 0.388 | LR: 1.00e-04 [2026-05-23 19:43:40] Epoch 2 | Step 4830 | Loss: 0.7184 | LM: 0.2332 | LB: 1.0017 | CL0: 2.5 | CL1: 2.5 | HR0: 0.403/SR0: 0.440 | HR1: 0.402/SR1: 0.388 | LR: 1.00e-04 [2026-05-23 19:43:53] Epoch 2 | Step 4840 | Loss: 0.7186 | LM: 0.2339 | LB: 1.0017 | CL0: 2.5 | CL1: 2.5 | HR0: 0.404/SR0: 0.441 | HR1: 0.403/SR1: 0.388 | LR: 1.00e-04 [2026-05-23 19:44:06] Epoch 2 | Step 4850 | Loss: 0.7181 | LM: 0.2324 | LB: 1.0017 | CL0: 2.5 | CL1: 2.5 | HR0: 0.404/SR0: 0.441 | HR1: 0.403/SR1: 0.388 | LR: 1.00e-04 [2026-05-23 19:44:19] Epoch 2 | Step 4860 | Loss: 0.7192 | LM: 0.2311 | LB: 1.0016 | CL0: 2.5 | CL1: 2.5 | HR0: 0.403/SR0: 0.440 | HR1: 0.403/SR1: 0.388 | LR: 1.00e-04 [2026-05-23 19:44:32] Epoch 2 | Step 4870 | Loss: 0.7204 | LM: 0.2305 | LB: 1.0016 | CL0: 2.5 | CL1: 2.5 | HR0: 0.403/SR0: 0.440 | HR1: 0.402/SR1: 0.388 | LR: 1.00e-04 [2026-05-23 19:44:45] Epoch 2 | Step 4880 | Loss: 0.7214 | LM: 0.2304 | LB: 1.0015 | CL0: 2.5 | CL1: 2.5 | HR0: 0.403/SR0: 0.440 | HR1: 0.402/SR1: 0.388 | LR: 1.00e-04 [2026-05-23 19:44:57] Epoch 2 | Step 4890 | Loss: 0.7206 | LM: 0.2298 | LB: 1.0015 | CL0: 2.5 | CL1: 2.5 | HR0: 0.403/SR0: 0.440 | HR1: 0.402/SR1: 0.388 | LR: 1.00e-04 [2026-05-23 19:45:10] Epoch 2 | Step 4900 | Loss: 0.7204 | LM: 0.2296 | LB: 1.0015 | CL0: 2.5 | CL1: 2.5 | HR0: 0.403/SR0: 0.440 | HR1: 0.402/SR1: 0.388 | LR: 1.00e-04 [2026-05-23 19:45:22] Epoch 2 | Step 4910 | Loss: 0.7201 | LM: 0.2283 | LB: 1.0014 | CL0: 2.5 | CL1: 2.5 | HR0: 0.402/SR0: 0.440 | HR1: 0.402/SR1: 0.388 | LR: 1.00e-04 [2026-05-23 19:45:34] Epoch 2 | Step 4920 | Loss: 0.7195 | LM: 0.2277 | LB: 1.0014 | CL0: 2.5 | CL1: 2.5 | HR0: 0.402/SR0: 0.439 | HR1: 0.403/SR1: 0.388 | LR: 1.00e-04 [2026-05-23 19:45:46] Epoch 2 | Step 4930 | Loss: 0.7191 | LM: 0.2269 | LB: 1.0014 | CL0: 2.5 | CL1: 2.5 | HR0: 0.402/SR0: 0.439 | HR1: 0.402/SR1: 0.388 | LR: 1.00e-04 [2026-05-23 19:45:59] Epoch 2 | Step 4940 | Loss: 0.7197 | LM: 0.2263 | LB: 1.0014 | CL0: 2.5 | CL1: 2.5 | HR0: 0.402/SR0: 0.439 | HR1: 0.402/SR1: 0.388 | LR: 1.00e-04 [2026-05-23 19:46:12] Epoch 2 | Step 4950 | Loss: 0.7186 | LM: 0.2254 | LB: 1.0013 | CL0: 2.5 | CL1: 2.5 | HR0: 0.402/SR0: 0.439 | HR1: 0.402/SR1: 0.388 | LR: 1.00e-04 [2026-05-23 19:46:23] Epoch 2 | Step 4960 | Loss: 0.7191 | LM: 0.2273 | LB: 1.0013 | CL0: 2.5 | CL1: 2.5 | HR0: 0.402/SR0: 0.439 | HR1: 0.402/SR1: 0.388 | LR: 1.00e-04 [2026-05-23 19:46:35] Epoch 2 | Step 4970 | Loss: 0.7190 | LM: 0.2260 | LB: 1.0013 | CL0: 2.5 | CL1: 2.5 | HR0: 0.402/SR0: 0.439 | HR1: 0.401/SR1: 0.388 | LR: 1.00e-04 [2026-05-23 19:46:48] Epoch 2 | Step 4980 | Loss: 0.7185 | LM: 0.2248 | LB: 1.0013 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.439 | HR1: 0.401/SR1: 0.388 | LR: 1.00e-04 [2026-05-23 19:47:00] Epoch 2 | Step 4990 | Loss: 0.7183 | LM: 0.2244 | LB: 1.0013 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.439 | HR1: 0.401/SR1: 0.388 | LR: 1.00e-04 [2026-05-23 19:47:13] Epoch 2 | Step 5000 | Loss: 0.7182 | LM: 0.2243 | LB: 1.0013 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.439 | HR1: 0.401/SR1: 0.388 | LR: 1.00e-04 [2026-05-23 19:47:16] Validation | Batch 10/1463 | Loss: 0.7661 | LM_LOSS: 0.2662 | LB_LOSS: 0.9997 [2026-05-23 19:47:18] Validation | Batch 20/1463 | Loss: 0.7767 | LM_LOSS: 0.2766 | LB_LOSS: 1.0001 [2026-05-23 19:47:22] Validation | Batch 30/1463 | Loss: 0.7797 | LM_LOSS: 0.2795 | LB_LOSS: 1.0003 [2026-05-23 19:47:25] Validation | Batch 40/1463 | Loss: 0.7866 | LM_LOSS: 0.2864 | LB_LOSS: 1.0003 [2026-05-23 19:47:27] Validation | Batch 50/1463 | Loss: 0.7946 | LM_LOSS: 0.2945 | LB_LOSS: 1.0003 [2026-05-23 19:47:32] Validation | Batch 60/1463 | Loss: 0.7942 | LM_LOSS: 0.2940 | LB_LOSS: 1.0004 [2026-05-23 19:47:35] Validation | Batch 70/1463 | Loss: 0.7941 | LM_LOSS: 0.2939 | LB_LOSS: 1.0004 [2026-05-23 19:47:39] Validation | Batch 80/1463 | Loss: 0.7960 | LM_LOSS: 0.2958 | LB_LOSS: 1.0004 [2026-05-23 19:47:42] Validation | Batch 90/1463 | Loss: 0.7952 | LM_LOSS: 0.2949 | LB_LOSS: 1.0004 [2026-05-23 19:47:44] Validation | Batch 100/1463 | Loss: 0.7949 | LM_LOSS: 0.2947 | LB_LOSS: 1.0004 [2026-05-23 19:47:48] Validation | Batch 110/1463 | Loss: 0.7927 | LM_LOSS: 0.2925 | LB_LOSS: 1.0004 [2026-05-23 19:47:50] Validation | Batch 120/1463 | Loss: 0.7944 | LM_LOSS: 0.2942 | LB_LOSS: 1.0004 [2026-05-23 19:47:52] Validation | Batch 130/1463 | Loss: 0.7960 | LM_LOSS: 0.2958 | LB_LOSS: 1.0004 [2026-05-23 19:47:56] Validation | Batch 140/1463 | Loss: 0.7973 | LM_LOSS: 0.2971 | LB_LOSS: 1.0004 [2026-05-23 19:48:00] Validation | Batch 150/1463 | Loss: 0.7951 | LM_LOSS: 0.2948 | LB_LOSS: 1.0005 [2026-05-23 19:48:03] Validation | Batch 160/1463 | Loss: 0.7954 | LM_LOSS: 0.2951 | LB_LOSS: 1.0005 [2026-05-23 19:48:06] Validation | Batch 170/1463 | Loss: 0.7946 | LM_LOSS: 0.2944 | LB_LOSS: 1.0005 [2026-05-23 19:48:09] Validation | Batch 180/1463 | Loss: 0.7954 | LM_LOSS: 0.2951 | LB_LOSS: 1.0005 [2026-05-23 19:48:12] Validation | Batch 190/1463 | Loss: 0.7964 | LM_LOSS: 0.2961 | LB_LOSS: 1.0005 [2026-05-23 19:48:15] Validation | Batch 200/1463 | Loss: 0.7965 | LM_LOSS: 0.2962 | LB_LOSS: 1.0005 [2026-05-23 19:48:18] Validation | Batch 210/1463 | Loss: 0.7973 | LM_LOSS: 0.2970 | LB_LOSS: 1.0005 [2026-05-23 19:48:22] Validation | Batch 220/1463 | Loss: 0.7967 | LM_LOSS: 0.2964 | LB_LOSS: 1.0005 [2026-05-23 19:48:25] Validation | Batch 230/1463 | Loss: 0.7981 | LM_LOSS: 0.2978 | LB_LOSS: 1.0005 [2026-05-23 19:48:28] Validation | Batch 240/1463 | Loss: 0.7997 | LM_LOSS: 0.2994 | LB_LOSS: 1.0005 [2026-05-23 19:48:32] Validation | Batch 250/1463 | Loss: 0.7987 | LM_LOSS: 0.2984 | LB_LOSS: 1.0005 [2026-05-23 19:48:34] Validation | Batch 260/1463 | Loss: 0.8002 | LM_LOSS: 0.3000 | LB_LOSS: 1.0005 [2026-05-23 19:48:37] Validation | Batch 270/1463 | Loss: 0.7988 | LM_LOSS: 0.2985 | LB_LOSS: 1.0005 [2026-05-23 19:48:40] Validation | Batch 280/1463 | Loss: 0.7991 | LM_LOSS: 0.2989 | LB_LOSS: 1.0005 [2026-05-23 19:48:44] Validation | Batch 290/1463 | Loss: 0.7981 | LM_LOSS: 0.2979 | LB_LOSS: 1.0005 [2026-05-23 19:48:46] Validation | Batch 300/1463 | Loss: 0.7984 | LM_LOSS: 0.2981 | LB_LOSS: 1.0005 [2026-05-23 19:48:49] Validation | Batch 310/1463 | Loss: 0.7993 | LM_LOSS: 0.2991 | LB_LOSS: 1.0005 [2026-05-23 19:48:52] Validation | Batch 320/1463 | Loss: 0.7981 | LM_LOSS: 0.2979 | LB_LOSS: 1.0005 [2026-05-23 19:48:55] Validation | Batch 330/1463 | Loss: 0.7979 | LM_LOSS: 0.2976 | LB_LOSS: 1.0005 [2026-05-23 19:48:58] Validation | Batch 340/1463 | Loss: 0.7975 | LM_LOSS: 0.2973 | LB_LOSS: 1.0005 [2026-05-23 19:49:02] Validation | Batch 350/1463 | Loss: 0.7982 | LM_LOSS: 0.2979 | LB_LOSS: 1.0005 [2026-05-23 19:49:04] Validation | Batch 360/1463 | Loss: 0.7979 | LM_LOSS: 0.2977 | LB_LOSS: 1.0005 [2026-05-23 19:49:07] Validation | Batch 370/1463 | Loss: 0.7978 | LM_LOSS: 0.2975 | LB_LOSS: 1.0005 [2026-05-23 19:49:10] Validation | Batch 380/1463 | Loss: 0.7977 | LM_LOSS: 0.2974 | LB_LOSS: 1.0005 [2026-05-23 19:49:13] Validation | Batch 390/1463 | Loss: 0.7982 | LM_LOSS: 0.2980 | LB_LOSS: 1.0005 [2026-05-23 19:49:16] Validation | Batch 400/1463 | Loss: 0.7981 | LM_LOSS: 0.2979 | LB_LOSS: 1.0005 [2026-05-23 19:49:19] Validation | Batch 410/1463 | Loss: 0.7978 | LM_LOSS: 0.2976 | LB_LOSS: 1.0005 [2026-05-23 19:49:22] Validation | Batch 420/1463 | Loss: 0.7980 | LM_LOSS: 0.2977 | LB_LOSS: 1.0005 [2026-05-23 19:49:25] Validation | Batch 430/1463 | Loss: 0.7975 | LM_LOSS: 0.2972 | LB_LOSS: 1.0005 [2026-05-23 19:49:28] Validation | Batch 440/1463 | Loss: 0.7976 | LM_LOSS: 0.2974 | LB_LOSS: 1.0005 [2026-05-23 19:49:31] Validation | Batch 450/1463 | Loss: 0.7973 | LM_LOSS: 0.2971 | LB_LOSS: 1.0005 [2026-05-23 19:49:34] Validation | Batch 460/1463 | Loss: 0.7972 | LM_LOSS: 0.2970 | LB_LOSS: 1.0005 [2026-05-23 19:49:37] Validation | Batch 470/1463 | Loss: 0.7979 | LM_LOSS: 0.2976 | LB_LOSS: 1.0005 [2026-05-23 19:49:41] Validation | Batch 480/1463 | Loss: 0.7976 | LM_LOSS: 0.2973 | LB_LOSS: 1.0005 [2026-05-23 19:49:44] Validation | Batch 490/1463 | Loss: 0.7975 | LM_LOSS: 0.2972 | LB_LOSS: 1.0005 [2026-05-23 19:49:47] Validation | Batch 500/1463 | Loss: 0.7973 | LM_LOSS: 0.2971 | LB_LOSS: 1.0005 [2026-05-23 19:49:51] Validation | Batch 510/1463 | Loss: 0.7966 | LM_LOSS: 0.2964 | LB_LOSS: 1.0005 [2026-05-23 19:49:53] Validation | Batch 520/1463 | Loss: 0.7963 | LM_LOSS: 0.2961 | LB_LOSS: 1.0005 [2026-05-23 19:49:56] Validation | Batch 530/1463 | Loss: 0.7968 | LM_LOSS: 0.2966 | LB_LOSS: 1.0005 [2026-05-23 19:49:58] Validation | Batch 540/1463 | Loss: 0.7970 | LM_LOSS: 0.2967 | LB_LOSS: 1.0005 [2026-05-23 19:50:01] Validation | Batch 550/1463 | Loss: 0.7965 | LM_LOSS: 0.2963 | LB_LOSS: 1.0005 [2026-05-23 19:50:04] Validation | Batch 560/1463 | Loss: 0.7966 | LM_LOSS: 0.2964 | LB_LOSS: 1.0005 [2026-05-23 19:50:07] Validation | Batch 570/1463 | Loss: 0.7965 | LM_LOSS: 0.2963 | LB_LOSS: 1.0005 [2026-05-23 19:50:11] Validation | Batch 580/1463 | Loss: 0.7960 | LM_LOSS: 0.2958 | LB_LOSS: 1.0005 [2026-05-23 19:50:14] Validation | Batch 590/1463 | Loss: 0.7957 | LM_LOSS: 0.2955 | LB_LOSS: 1.0005 [2026-05-23 19:50:17] Validation | Batch 600/1463 | Loss: 0.7960 | LM_LOSS: 0.2957 | LB_LOSS: 1.0005 [2026-05-23 19:50:21] Validation | Batch 610/1463 | Loss: 0.7962 | LM_LOSS: 0.2960 | LB_LOSS: 1.0005 [2026-05-23 19:50:24] Validation | Batch 620/1463 | Loss: 0.7957 | LM_LOSS: 0.2955 | LB_LOSS: 1.0005 [2026-05-23 19:50:27] Validation | Batch 630/1463 | Loss: 0.7951 | LM_LOSS: 0.2949 | LB_LOSS: 1.0005 [2026-05-23 19:50:29] Validation | Batch 640/1463 | Loss: 0.7949 | LM_LOSS: 0.2947 | LB_LOSS: 1.0005 [2026-05-23 19:50:32] Validation | Batch 650/1463 | Loss: 0.7945 | LM_LOSS: 0.2943 | LB_LOSS: 1.0005 [2026-05-23 19:50:35] Validation | Batch 660/1463 | Loss: 0.7940 | LM_LOSS: 0.2938 | LB_LOSS: 1.0005 [2026-05-23 19:50:38] Validation | Batch 670/1463 | Loss: 0.7937 | LM_LOSS: 0.2934 | LB_LOSS: 1.0005 [2026-05-23 19:50:40] Validation | Batch 680/1463 | Loss: 0.7937 | LM_LOSS: 0.2934 | LB_LOSS: 1.0005 [2026-05-23 19:50:44] Validation | Batch 690/1463 | Loss: 0.7939 | LM_LOSS: 0.2936 | LB_LOSS: 1.0005 [2026-05-23 19:50:46] Validation | Batch 700/1463 | Loss: 0.7937 | LM_LOSS: 0.2934 | LB_LOSS: 1.0005 [2026-05-23 19:50:49] Validation | Batch 710/1463 | Loss: 0.7940 | LM_LOSS: 0.2937 | LB_LOSS: 1.0005 [2026-05-23 19:50:52] Validation | Batch 720/1463 | Loss: 0.7948 | LM_LOSS: 0.2945 | LB_LOSS: 1.0005 [2026-05-23 19:50:55] Validation | Batch 730/1463 | Loss: 0.7947 | LM_LOSS: 0.2944 | LB_LOSS: 1.0005 [2026-05-23 19:50:57] Validation | Batch 740/1463 | Loss: 0.7944 | LM_LOSS: 0.2942 | LB_LOSS: 1.0005 [2026-05-23 19:51:00] Validation | Batch 750/1463 | Loss: 0.7944 | LM_LOSS: 0.2942 | LB_LOSS: 1.0005 [2026-05-23 19:51:03] Validation | Batch 760/1463 | Loss: 0.7943 | LM_LOSS: 0.2941 | LB_LOSS: 1.0005 [2026-05-23 19:51:05] Validation | Batch 770/1463 | Loss: 0.7943 | LM_LOSS: 0.2940 | LB_LOSS: 1.0005 [2026-05-23 19:51:08] Validation | Batch 780/1463 | Loss: 0.7942 | LM_LOSS: 0.2940 | LB_LOSS: 1.0005 [2026-05-23 19:51:10] Validation | Batch 790/1463 | Loss: 0.7945 | LM_LOSS: 0.2942 | LB_LOSS: 1.0005 [2026-05-23 19:51:13] Validation | Batch 800/1463 | Loss: 0.7943 | LM_LOSS: 0.2941 | LB_LOSS: 1.0005 [2026-05-23 19:51:15] Validation | Batch 810/1463 | Loss: 0.7940 | LM_LOSS: 0.2938 | LB_LOSS: 1.0005 [2026-05-23 19:51:19] Validation | Batch 820/1463 | Loss: 0.7942 | LM_LOSS: 0.2940 | LB_LOSS: 1.0005 [2026-05-23 19:51:22] Validation | Batch 830/1463 | Loss: 0.7944 | LM_LOSS: 0.2942 | LB_LOSS: 1.0005 [2026-05-23 19:51:25] Validation | Batch 840/1463 | Loss: 0.7945 | LM_LOSS: 0.2942 | LB_LOSS: 1.0005 [2026-05-23 19:51:28] Validation | Batch 850/1463 | Loss: 0.7947 | LM_LOSS: 0.2945 | LB_LOSS: 1.0005 [2026-05-23 19:51:32] Validation | Batch 860/1463 | Loss: 0.7949 | LM_LOSS: 0.2946 | LB_LOSS: 1.0005 [2026-05-23 19:51:35] Validation | Batch 870/1463 | Loss: 0.7949 | LM_LOSS: 0.2947 | LB_LOSS: 1.0005 [2026-05-23 19:51:37] Validation | Batch 880/1463 | Loss: 0.7941 | LM_LOSS: 0.2939 | LB_LOSS: 1.0005 [2026-05-23 19:51:40] Validation | Batch 890/1463 | Loss: 0.7941 | LM_LOSS: 0.2939 | LB_LOSS: 1.0005 [2026-05-23 19:51:43] Validation | Batch 900/1463 | Loss: 0.7938 | LM_LOSS: 0.2936 | LB_LOSS: 1.0005 [2026-05-23 19:51:47] Validation | Batch 910/1463 | Loss: 0.7938 | LM_LOSS: 0.2936 | LB_LOSS: 1.0005 [2026-05-23 19:51:50] Validation | Batch 920/1463 | Loss: 0.7940 | LM_LOSS: 0.2937 | LB_LOSS: 1.0005 [2026-05-23 19:51:53] Validation | Batch 930/1463 | Loss: 0.7938 | LM_LOSS: 0.2935 | LB_LOSS: 1.0005 [2026-05-23 19:51:56] Validation | Batch 940/1463 | Loss: 0.7935 | LM_LOSS: 0.2933 | LB_LOSS: 1.0005 [2026-05-23 19:52:00] Validation | Batch 950/1463 | Loss: 0.7937 | LM_LOSS: 0.2935 | LB_LOSS: 1.0005 [2026-05-23 19:52:02] Validation | Batch 960/1463 | Loss: 0.7938 | LM_LOSS: 0.2935 | LB_LOSS: 1.0005 [2026-05-23 19:52:06] Validation | Batch 970/1463 | Loss: 0.7938 | LM_LOSS: 0.2935 | LB_LOSS: 1.0005 [2026-05-23 19:52:09] Validation | Batch 980/1463 | Loss: 0.7941 | LM_LOSS: 0.2939 | LB_LOSS: 1.0005 [2026-05-23 19:52:12] Validation | Batch 990/1463 | Loss: 0.7944 | LM_LOSS: 0.2941 | LB_LOSS: 1.0005 [2026-05-23 19:52:15] Validation | Batch 1000/1463 | Loss: 0.7949 | LM_LOSS: 0.2947 | LB_LOSS: 1.0005 [2026-05-23 19:52:18] Validation | Batch 1010/1463 | Loss: 0.7949 | LM_LOSS: 0.2947 | LB_LOSS: 1.0005 [2026-05-23 19:52:21] Validation | Batch 1020/1463 | Loss: 0.7950 | LM_LOSS: 0.2948 | LB_LOSS: 1.0005 [2026-05-23 19:52:25] Validation | Batch 1030/1463 | Loss: 0.7955 | LM_LOSS: 0.2952 | LB_LOSS: 1.0005 [2026-05-23 19:52:27] Validation | Batch 1040/1463 | Loss: 0.7951 | LM_LOSS: 0.2949 | LB_LOSS: 1.0005 [2026-05-23 19:52:30] Validation | Batch 1050/1463 | Loss: 0.7951 | LM_LOSS: 0.2948 | LB_LOSS: 1.0005 [2026-05-23 19:52:34] Validation | Batch 1060/1463 | Loss: 0.7950 | LM_LOSS: 0.2948 | LB_LOSS: 1.0005 [2026-05-23 19:52:37] Validation | Batch 1070/1463 | Loss: 0.7950 | LM_LOSS: 0.2948 | LB_LOSS: 1.0005 [2026-05-23 19:52:39] Validation | Batch 1080/1463 | Loss: 0.7948 | LM_LOSS: 0.2946 | LB_LOSS: 1.0005 [2026-05-23 19:52:42] Validation | Batch 1090/1463 | Loss: 0.7949 | LM_LOSS: 0.2946 | LB_LOSS: 1.0005 [2026-05-23 19:52:45] Validation | Batch 1100/1463 | Loss: 0.7947 | LM_LOSS: 0.2945 | LB_LOSS: 1.0005 [2026-05-23 19:52:47] Validation | Batch 1110/1463 | Loss: 0.7948 | LM_LOSS: 0.2945 | LB_LOSS: 1.0005 [2026-05-23 19:52:50] Validation | Batch 1120/1463 | Loss: 0.7945 | LM_LOSS: 0.2943 | LB_LOSS: 1.0005 [2026-05-23 19:52:54] Validation | Batch 1130/1463 | Loss: 0.7943 | LM_LOSS: 0.2940 | LB_LOSS: 1.0005 [2026-05-23 19:52:57] Validation | Batch 1140/1463 | Loss: 0.7945 | LM_LOSS: 0.2943 | LB_LOSS: 1.0005 [2026-05-23 19:53:01] Validation | Batch 1150/1463 | Loss: 0.7944 | LM_LOSS: 0.2942 | LB_LOSS: 1.0005 [2026-05-23 19:53:03] Validation | Batch 1160/1463 | Loss: 0.7943 | LM_LOSS: 0.2940 | LB_LOSS: 1.0005 [2026-05-23 19:53:07] Validation | Batch 1170/1463 | Loss: 0.7943 | LM_LOSS: 0.2941 | LB_LOSS: 1.0005 [2026-05-23 19:53:09] Validation | Batch 1180/1463 | Loss: 0.7943 | LM_LOSS: 0.2941 | LB_LOSS: 1.0005 [2026-05-23 19:53:13] Validation | Batch 1190/1463 | Loss: 0.7945 | LM_LOSS: 0.2943 | LB_LOSS: 1.0005 [2026-05-23 19:53:16] Validation | Batch 1200/1463 | Loss: 0.7945 | LM_LOSS: 0.2942 | LB_LOSS: 1.0005 [2026-05-23 19:53:19] Validation | Batch 1210/1463 | Loss: 0.7947 | LM_LOSS: 0.2944 | LB_LOSS: 1.0005 [2026-05-23 19:53:22] Validation | Batch 1220/1463 | Loss: 0.7950 | LM_LOSS: 0.2947 | LB_LOSS: 1.0005 [2026-05-23 19:53:24] Validation | Batch 1230/1463 | Loss: 0.7950 | LM_LOSS: 0.2948 | LB_LOSS: 1.0005 [2026-05-23 19:53:28] Validation | Batch 1240/1463 | Loss: 0.7953 | LM_LOSS: 0.2950 | LB_LOSS: 1.0005 [2026-05-23 19:53:31] Validation | Batch 1250/1463 | Loss: 0.7954 | LM_LOSS: 0.2951 | LB_LOSS: 1.0005 [2026-05-23 19:53:34] Validation | Batch 1260/1463 | Loss: 0.7953 | LM_LOSS: 0.2950 | LB_LOSS: 1.0005 [2026-05-23 19:53:37] Validation | Batch 1270/1463 | Loss: 0.7951 | LM_LOSS: 0.2948 | LB_LOSS: 1.0005 [2026-05-23 19:53:40] Validation | Batch 1280/1463 | Loss: 0.7950 | LM_LOSS: 0.2947 | LB_LOSS: 1.0005 [2026-05-23 19:53:43] Validation | Batch 1290/1463 | Loss: 0.7950 | LM_LOSS: 0.2947 | LB_LOSS: 1.0005 [2026-05-23 19:53:46] Validation | Batch 1300/1463 | Loss: 0.7948 | LM_LOSS: 0.2945 | LB_LOSS: 1.0005 [2026-05-23 19:53:49] Validation | Batch 1310/1463 | Loss: 0.7947 | LM_LOSS: 0.2944 | LB_LOSS: 1.0005 [2026-05-23 19:53:52] Validation | Batch 1320/1463 | Loss: 0.7947 | LM_LOSS: 0.2945 | LB_LOSS: 1.0005 [2026-05-23 19:53:54] Validation | Batch 1330/1463 | Loss: 0.7948 | LM_LOSS: 0.2945 | LB_LOSS: 1.0005 [2026-05-23 19:53:57] Validation | Batch 1340/1463 | Loss: 0.7951 | LM_LOSS: 0.2948 | LB_LOSS: 1.0005 [2026-05-23 19:54:00] Validation | Batch 1350/1463 | Loss: 0.7951 | LM_LOSS: 0.2948 | LB_LOSS: 1.0005 [2026-05-23 19:54:04] Validation | Batch 1360/1463 | Loss: 0.7953 | LM_LOSS: 0.2951 | LB_LOSS: 1.0005 [2026-05-23 19:54:07] Validation | Batch 1370/1463 | Loss: 0.7956 | LM_LOSS: 0.2954 | LB_LOSS: 1.0005 [2026-05-23 19:54:10] Validation | Batch 1380/1463 | Loss: 0.7954 | LM_LOSS: 0.2952 | LB_LOSS: 1.0005 [2026-05-23 19:54:13] Validation | Batch 1390/1463 | Loss: 0.7954 | LM_LOSS: 0.2951 | LB_LOSS: 1.0005 [2026-05-23 19:54:16] Validation | Batch 1400/1463 | Loss: 0.7956 | LM_LOSS: 0.2954 | LB_LOSS: 1.0005 [2026-05-23 19:54:18] Validation | Batch 1410/1463 | Loss: 0.7959 | LM_LOSS: 0.2957 | LB_LOSS: 1.0005 [2026-05-23 19:54:21] Validation | Batch 1420/1463 | Loss: 0.7958 | LM_LOSS: 0.2956 | LB_LOSS: 1.0005 [2026-05-23 19:54:25] Validation | Batch 1430/1463 | Loss: 0.7959 | LM_LOSS: 0.2957 | LB_LOSS: 1.0005 [2026-05-23 19:54:28] Validation | Batch 1440/1463 | Loss: 0.7963 | LM_LOSS: 0.2961 | LB_LOSS: 1.0005 [2026-05-23 19:54:31] Validation | Batch 1450/1463 | Loss: 0.7961 | LM_LOSS: 0.2958 | LB_LOSS: 1.0005 [2026-05-23 19:54:34] Validation | Batch 1460/1463 | Loss: 0.7964 | LM_LOSS: 0.2961 | LB_LOSS: 1.0005 [2026-05-23 19:54:35] Validation | Batch 1463/1463 | Loss: 0.7963 | LM_LOSS: 0.2960 | LB_LOSS: 1.0005 [2026-05-23 19:54:35] Validation | Loss: 0.7963 | LM_LOSS: 0.2960 | LB_LOSS: 1.0005 | PPL: 1.35 | Time: 441.84s [2026-05-23 19:54:48] Epoch 2 | Step 5010 | Loss: 0.7180 | LM: 0.2237 | LB: 1.0013 | CL0: 2.5 | CL1: 2.5 | HR0: 0.402/SR0: 0.439 | HR1: 0.401/SR1: 0.388 | LR: 1.00e-04 [2026-05-23 19:55:00] Epoch 2 | Step 5020 | Loss: 0.7184 | LM: 0.2238 | LB: 1.0013 | CL0: 2.5 | CL1: 2.5 | HR0: 0.402/SR0: 0.439 | HR1: 0.401/SR1: 0.388 | LR: 1.00e-04 [2026-05-23 19:55:12] Epoch 2 | Step 5030 | Loss: 0.7184 | LM: 0.2236 | LB: 1.0013 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.439 | HR1: 0.401/SR1: 0.388 | LR: 1.00e-04 [2026-05-23 19:55:25] Epoch 2 | Step 5040 | Loss: 0.7178 | LM: 0.2226 | LB: 1.0014 | CL0: 2.5 | CL1: 2.5 | HR0: 0.402/SR0: 0.439 | HR1: 0.401/SR1: 0.388 | LR: 1.00e-04 [2026-05-23 19:55:36] Epoch 2 | Step 5050 | Loss: 0.7170 | LM: 0.2217 | LB: 1.0013 | CL0: 2.5 | CL1: 2.5 | HR0: 0.402/SR0: 0.439 | HR1: 0.401/SR1: 0.388 | LR: 1.00e-04 [2026-05-23 19:55:48] Epoch 2 | Step 5060 | Loss: 0.7167 | LM: 0.2223 | LB: 1.0013 | CL0: 2.5 | CL1: 2.5 | HR0: 0.402/SR0: 0.439 | HR1: 0.401/SR1: 0.388 | LR: 1.00e-04 [2026-05-23 19:56:01] Epoch 2 | Step 5070 | Loss: 0.7161 | LM: 0.2221 | LB: 1.0013 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.439 | HR1: 0.401/SR1: 0.388 | LR: 1.00e-04 [2026-05-23 19:56:13] Epoch 2 | Step 5080 | Loss: 0.7159 | LM: 0.2231 | LB: 1.0014 | CL0: 2.5 | CL1: 2.5 | HR0: 0.402/SR0: 0.439 | HR1: 0.401/SR1: 0.388 | LR: 1.00e-04 [2026-05-23 19:56:26] Epoch 2 | Step 5090 | Loss: 0.7161 | LM: 0.2235 | LB: 1.0014 | CL0: 2.5 | CL1: 2.5 | HR0: 0.402/SR0: 0.439 | HR1: 0.401/SR1: 0.389 | LR: 1.00e-04 [2026-05-23 19:56:39] Epoch 2 | Step 5100 | Loss: 0.7163 | LM: 0.2239 | LB: 1.0014 | CL0: 2.5 | CL1: 2.5 | HR0: 0.402/SR0: 0.439 | HR1: 0.402/SR1: 0.389 | LR: 1.00e-04 [2026-05-23 19:56:50] Epoch 2 | Step 5110 | Loss: 0.7165 | LM: 0.2223 | LB: 1.0014 | CL0: 2.5 | CL1: 2.5 | HR0: 0.402/SR0: 0.439 | HR1: 0.401/SR1: 0.389 | LR: 1.00e-04 [2026-05-23 19:57:02] Epoch 2 | Step 5120 | Loss: 0.7167 | LM: 0.2241 | LB: 1.0014 | CL0: 2.5 | CL1: 2.5 | HR0: 0.402/SR0: 0.439 | HR1: 0.401/SR1: 0.389 | LR: 1.00e-04 [2026-05-23 19:57:15] Epoch 2 | Step 5130 | Loss: 0.7168 | LM: 0.2252 | LB: 1.0014 | CL0: 2.5 | CL1: 2.5 | HR0: 0.402/SR0: 0.439 | HR1: 0.401/SR1: 0.389 | LR: 1.00e-04 [2026-05-23 19:57:27] Epoch 2 | Step 5140 | Loss: 0.7168 | LM: 0.2255 | LB: 1.0014 | CL0: 2.5 | CL1: 2.5 | HR0: 0.402/SR0: 0.439 | HR1: 0.401/SR1: 0.389 | LR: 1.00e-04 [2026-05-23 19:57:38] Epoch 2 | Step 5150 | Loss: 0.7163 | LM: 0.2249 | LB: 1.0014 | CL0: 2.5 | CL1: 2.5 | HR0: 0.402/SR0: 0.439 | HR1: 0.401/SR1: 0.389 | LR: 1.00e-04 [2026-05-23 19:57:50] Epoch 2 | Step 5160 | Loss: 0.7160 | LM: 0.2250 | LB: 1.0014 | CL0: 2.5 | CL1: 2.5 | HR0: 0.402/SR0: 0.439 | HR1: 0.401/SR1: 0.389 | LR: 1.00e-04 [2026-05-23 19:58:02] Epoch 2 | Step 5170 | Loss: 0.7158 | LM: 0.2238 | LB: 1.0014 | CL0: 2.5 | CL1: 2.5 | HR0: 0.402/SR0: 0.439 | HR1: 0.401/SR1: 0.389 | LR: 1.00e-04 [2026-05-23 19:58:15] Epoch 2 | Step 5180 | Loss: 0.7161 | LM: 0.2235 | LB: 1.0014 | CL0: 2.5 | CL1: 2.5 | HR0: 0.402/SR0: 0.438 | HR1: 0.401/SR1: 0.389 | LR: 1.00e-04 [2026-05-23 19:58:26] Epoch 2 | Step 5190 | Loss: 0.7160 | LM: 0.2234 | LB: 1.0014 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.438 | HR1: 0.400/SR1: 0.388 | LR: 1.00e-04 [2026-05-23 19:58:37] Epoch 2 | Step 5200 | Loss: 0.7164 | LM: 0.2236 | LB: 1.0014 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.438 | HR1: 0.400/SR1: 0.388 | LR: 1.00e-04 [2026-05-23 19:58:49] Epoch 2 | Step 5210 | Loss: 0.7163 | LM: 0.2228 | LB: 1.0014 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.438 | HR1: 0.400/SR1: 0.388 | LR: 1.00e-04 [2026-05-23 19:59:02] Epoch 2 | Step 5220 | Loss: 0.7167 | LM: 0.2232 | LB: 1.0014 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.438 | HR1: 0.400/SR1: 0.388 | LR: 1.00e-04 [2026-05-23 19:59:14] Epoch 2 | Step 5230 | Loss: 0.7171 | LM: 0.2234 | LB: 1.0013 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.438 | HR1: 0.400/SR1: 0.388 | LR: 1.00e-04 [2026-05-23 19:59:26] Epoch 2 | Step 5240 | Loss: 0.7170 | LM: 0.2237 | LB: 1.0013 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.438 | HR1: 0.400/SR1: 0.388 | LR: 1.00e-04 [2026-05-23 19:59:38] Epoch 2 | Step 5250 | Loss: 0.7171 | LM: 0.2237 | LB: 1.0014 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.438 | HR1: 0.400/SR1: 0.388 | LR: 1.00e-04 [2026-05-23 19:59:50] Epoch 2 | Step 5260 | Loss: 0.7173 | LM: 0.2241 | LB: 1.0014 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.438 | HR1: 0.400/SR1: 0.388 | LR: 1.00e-04 [2026-05-23 20:00:02] Epoch 2 | Step 5270 | Loss: 0.7173 | LM: 0.2241 | LB: 1.0014 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.438 | HR1: 0.400/SR1: 0.388 | LR: 1.00e-04 [2026-05-23 20:00:13] Epoch 2 | Step 5280 | Loss: 0.7176 | LM: 0.2252 | LB: 1.0014 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.438 | HR1: 0.400/SR1: 0.389 | LR: 1.00e-04 [2026-05-23 20:00:27] Epoch 2 | Step 5290 | Loss: 0.7179 | LM: 0.2246 | LB: 1.0014 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.438 | HR1: 0.400/SR1: 0.388 | LR: 1.00e-04 [2026-05-23 20:00:39] Epoch 2 | Step 5300 | Loss: 0.7179 | LM: 0.2249 | LB: 1.0013 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.438 | HR1: 0.400/SR1: 0.388 | LR: 1.00e-04 [2026-05-23 20:00:50] Epoch 2 | Step 5310 | Loss: 0.7182 | LM: 0.2252 | LB: 1.0013 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.438 | HR1: 0.400/SR1: 0.388 | LR: 1.00e-04 [2026-05-23 20:01:02] Epoch 2 | Step 5320 | Loss: 0.7183 | LM: 0.2247 | LB: 1.0013 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.438 | HR1: 0.400/SR1: 0.388 | LR: 1.00e-04 [2026-05-23 20:01:15] Epoch 2 | Step 5330 | Loss: 0.7179 | LM: 0.2246 | LB: 1.0013 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.438 | HR1: 0.400/SR1: 0.388 | LR: 1.00e-04 [2026-05-23 20:01:26] Epoch 2 | Step 5340 | Loss: 0.7178 | LM: 0.2247 | LB: 1.0013 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.438 | HR1: 0.400/SR1: 0.388 | LR: 1.00e-04 [2026-05-23 20:01:39] Epoch 2 | Step 5350 | Loss: 0.7175 | LM: 0.2242 | LB: 1.0013 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.438 | HR1: 0.400/SR1: 0.389 | LR: 1.00e-04 [2026-05-23 20:01:51] Epoch 2 | Step 5360 | Loss: 0.7173 | LM: 0.2234 | LB: 1.0013 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.438 | HR1: 0.400/SR1: 0.389 | LR: 1.00e-04 [2026-05-23 20:02:03] Epoch 2 | Step 5370 | Loss: 0.7177 | LM: 0.2234 | LB: 1.0013 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.438 | HR1: 0.400/SR1: 0.389 | LR: 1.00e-04 [2026-05-23 20:02:14] Epoch 2 | Step 5380 | Loss: 0.7173 | LM: 0.2233 | LB: 1.0013 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.438 | HR1: 0.400/SR1: 0.389 | LR: 1.00e-04 [2026-05-23 20:02:27] Epoch 2 | Step 5390 | Loss: 0.7171 | LM: 0.2233 | LB: 1.0013 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.438 | HR1: 0.400/SR1: 0.389 | LR: 1.00e-04 [2026-05-23 20:02:39] Epoch 2 | Step 5400 | Loss: 0.7172 | LM: 0.2224 | LB: 1.0013 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.389 | LR: 1.00e-04 [2026-05-23 20:02:51] Epoch 2 | Step 5410 | Loss: 0.7173 | LM: 0.2224 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.389 | LR: 1.00e-04 [2026-05-23 20:03:05] Epoch 2 | Step 5420 | Loss: 0.7173 | LM: 0.2225 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.437 | HR1: 0.400/SR1: 0.389 | LR: 1.00e-04 [2026-05-23 20:03:17] Epoch 2 | Step 5430 | Loss: 0.7175 | LM: 0.2225 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.437 | HR1: 0.400/SR1: 0.389 | LR: 1.00e-04 [2026-05-23 20:03:29] Epoch 2 | Step 5440 | Loss: 0.7173 | LM: 0.2221 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.437 | HR1: 0.400/SR1: 0.389 | LR: 1.00e-04 [2026-05-23 20:03:41] Epoch 2 | Step 5450 | Loss: 0.7172 | LM: 0.2221 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.437 | HR1: 0.400/SR1: 0.389 | LR: 1.00e-04 [2026-05-23 20:03:53] Epoch 2 | Step 5460 | Loss: 0.7173 | LM: 0.2220 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.389 | LR: 1.00e-04 [2026-05-23 20:04:05] Epoch 2 | Step 5470 | Loss: 0.7173 | LM: 0.2216 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.389 | LR: 1.00e-04 [2026-05-23 20:04:16] Epoch 2 | Step 5480 | Loss: 0.7172 | LM: 0.2215 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.389 | LR: 1.00e-04 [2026-05-23 20:04:29] Epoch 2 | Step 5490 | Loss: 0.7169 | LM: 0.2217 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.389 | LR: 1.00e-04 [2026-05-23 20:04:41] Epoch 2 | Step 5500 | Loss: 0.7171 | LM: 0.2220 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.389 | LR: 1.00e-04 [2026-05-23 20:04:53] Epoch 2 | Step 5510 | Loss: 0.7169 | LM: 0.2217 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.389 | LR: 1.00e-04 [2026-05-23 20:05:05] Epoch 2 | Step 5520 | Loss: 0.7171 | LM: 0.2215 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.389 | LR: 1.00e-04 [2026-05-23 20:05:18] Epoch 2 | Step 5530 | Loss: 0.7169 | LM: 0.2212 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.389 | LR: 1.00e-04 [2026-05-23 20:05:31] Epoch 2 | Step 5540 | Loss: 0.7166 | LM: 0.2210 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.389 | LR: 1.00e-04 [2026-05-23 20:05:44] Epoch 2 | Step 5550 | Loss: 0.7167 | LM: 0.2217 | LB: 1.0013 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.401/SR1: 0.389 | LR: 1.00e-04 [2026-05-23 20:05:55] Epoch 2 | Step 5560 | Loss: 0.7166 | LM: 0.2212 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.389 | LR: 1.00e-04 [2026-05-23 20:06:08] Epoch 2 | Step 5570 | Loss: 0.7165 | LM: 0.2206 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.389 | LR: 1.00e-04 [2026-05-23 20:06:21] Epoch 2 | Step 5580 | Loss: 0.7165 | LM: 0.2205 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.389 | LR: 1.00e-04 [2026-05-23 20:06:33] Epoch 2 | Step 5590 | Loss: 0.7164 | LM: 0.2206 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.389 | LR: 1.00e-04 [2026-05-23 20:06:47] Epoch 2 | Step 5600 | Loss: 0.7165 | LM: 0.2208 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.389 | LR: 1.00e-04 [2026-05-23 20:06:59] Epoch 2 | Step 5610 | Loss: 0.7167 | LM: 0.2212 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.389 | LR: 1.00e-04 [2026-05-23 20:07:11] Epoch 2 | Step 5620 | Loss: 0.7166 | LM: 0.2212 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.389 | LR: 1.00e-04 [2026-05-23 20:07:24] Epoch 2 | Step 5630 | Loss: 0.7167 | LM: 0.2212 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.437 | HR1: 0.400/SR1: 0.389 | LR: 1.00e-04 [2026-05-23 20:07:37] Epoch 2 | Step 5640 | Loss: 0.7169 | LM: 0.2212 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.389 | LR: 1.00e-04 [2026-05-23 20:07:49] Epoch 2 | Step 5650 | Loss: 0.7168 | LM: 0.2209 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.389 | LR: 1.00e-04 [2026-05-23 20:08:02] Epoch 2 | Step 5660 | Loss: 0.7167 | LM: 0.2202 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.437 | HR1: 0.400/SR1: 0.389 | LR: 1.00e-04 [2026-05-23 20:08:14] Epoch 2 | Step 5670 | Loss: 0.7165 | LM: 0.2198 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.437 | HR1: 0.400/SR1: 0.389 | LR: 1.00e-04 [2026-05-23 20:08:26] Epoch 2 | Step 5680 | Loss: 0.7166 | LM: 0.2199 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.437 | HR1: 0.400/SR1: 0.389 | LR: 1.00e-04 [2026-05-23 20:08:38] Epoch 2 | Step 5690 | Loss: 0.7166 | LM: 0.2195 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.437 | HR1: 0.400/SR1: 0.389 | LR: 1.00e-04 [2026-05-23 20:08:50] Epoch 2 | Step 5700 | Loss: 0.7165 | LM: 0.2194 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.437 | HR1: 0.400/SR1: 0.389 | LR: 1.00e-04 [2026-05-23 20:09:03] Epoch 2 | Step 5710 | Loss: 0.7167 | LM: 0.2195 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.437 | HR1: 0.400/SR1: 0.389 | LR: 1.00e-04 [2026-05-23 20:09:15] Epoch 2 | Step 5720 | Loss: 0.7170 | LM: 0.2202 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.437 | HR1: 0.400/SR1: 0.389 | LR: 1.00e-04 [2026-05-23 20:09:29] Epoch 2 | Step 5730 | Loss: 0.7171 | LM: 0.2203 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.437 | HR1: 0.400/SR1: 0.389 | LR: 1.00e-04 [2026-05-23 20:09:42] Epoch 2 | Step 5740 | Loss: 0.7171 | LM: 0.2201 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.437 | HR1: 0.400/SR1: 0.389 | LR: 1.00e-04 [2026-05-23 20:09:54] Epoch 2 | Step 5750 | Loss: 0.7172 | LM: 0.2198 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.437 | HR1: 0.400/SR1: 0.389 | LR: 1.00e-04 [2026-05-23 20:10:06] Epoch 2 | Step 5760 | Loss: 0.7173 | LM: 0.2200 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.389 | LR: 1.00e-04 [2026-05-23 20:10:18] Epoch 2 | Step 5770 | Loss: 0.7173 | LM: 0.2201 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.389 | LR: 1.00e-04 [2026-05-23 20:10:31] Epoch 2 | Step 5780 | Loss: 0.7174 | LM: 0.2201 | LB: 1.0013 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.389 | LR: 1.00e-04 [2026-05-23 20:10:45] Epoch 2 | Step 5790 | Loss: 0.7177 | LM: 0.2203 | LB: 1.0013 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.389 | LR: 1.00e-04 [2026-05-23 20:10:56] Epoch 2 | Step 5800 | Loss: 0.7177 | LM: 0.2203 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.389 | LR: 1.00e-04 [2026-05-23 20:11:09] Epoch 2 | Step 5810 | Loss: 0.7177 | LM: 0.2203 | LB: 1.0013 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.389 | LR: 1.00e-04 [2026-05-23 20:11:21] Epoch 2 | Step 5820 | Loss: 0.7180 | LM: 0.2205 | LB: 1.0013 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.389 | LR: 1.00e-04 [2026-05-23 20:11:34] Epoch 2 | Step 5830 | Loss: 0.7181 | LM: 0.2202 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.389 | LR: 1.00e-04 [2026-05-23 20:11:46] Epoch 2 | Step 5840 | Loss: 0.7182 | LM: 0.2202 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.389 | LR: 1.00e-04 [2026-05-23 20:11:59] Epoch 2 | Step 5850 | Loss: 0.7184 | LM: 0.2198 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.389 | LR: 1.00e-04 [2026-05-23 20:12:11] Epoch 2 | Step 5860 | Loss: 0.7182 | LM: 0.2195 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.389 | LR: 1.00e-04 [2026-05-23 20:12:24] Epoch 2 | Step 5870 | Loss: 0.7181 | LM: 0.2197 | LB: 1.0013 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.389 | LR: 1.00e-04 [2026-05-23 20:12:36] Epoch 2 | Step 5880 | Loss: 0.7180 | LM: 0.2193 | LB: 1.0013 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.389 | LR: 1.00e-04 [2026-05-23 20:12:48] Epoch 2 | Step 5890 | Loss: 0.7180 | LM: 0.2189 | LB: 1.0013 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.389 | LR: 1.00e-04 [2026-05-23 20:13:01] Epoch 2 | Step 5900 | Loss: 0.7181 | LM: 0.2191 | LB: 1.0013 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.389 | LR: 1.00e-04 [2026-05-23 20:13:13] Epoch 2 | Step 5910 | Loss: 0.7181 | LM: 0.2188 | LB: 1.0013 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.389 | LR: 1.00e-04 [2026-05-23 20:13:26] Epoch 2 | Step 5920 | Loss: 0.7180 | LM: 0.2188 | LB: 1.0013 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.389 | LR: 1.00e-04 [2026-05-23 20:13:38] Epoch 2 | Step 5930 | Loss: 0.7182 | LM: 0.2190 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.389 | LR: 1.00e-04 [2026-05-23 20:13:52] Epoch 2 | Step 5940 | Loss: 0.7184 | LM: 0.2189 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.389 | LR: 1.00e-04 [2026-05-23 20:14:04] Epoch 2 | Step 5950 | Loss: 0.7184 | LM: 0.2190 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.389 | LR: 1.00e-04 [2026-05-23 20:14:17] Epoch 2 | Step 5960 | Loss: 0.7183 | LM: 0.2189 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.389 | LR: 1.00e-04 [2026-05-23 20:14:30] Epoch 2 | Step 5970 | Loss: 0.7182 | LM: 0.2187 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.389 | LR: 1.00e-04 [2026-05-23 20:14:42] Epoch 2 | Step 5980 | Loss: 0.7183 | LM: 0.2187 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.389 | LR: 1.00e-04 [2026-05-23 20:14:54] Epoch 2 | Step 5990 | Loss: 0.7183 | LM: 0.2187 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.389 | LR: 1.00e-04 [2026-05-23 20:15:07] Epoch 2 | Step 6000 | Loss: 0.7183 | LM: 0.2190 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.389 | LR: 1.00e-04 [2026-05-23 20:15:11] Checkpoint saved: outputs/2026-05-23/N_2.5/checkpoints/checkpoint_step_6000.pt [2026-05-23 20:15:17] Validation | Batch 10/1463 | Loss: 0.7650 | LM_LOSS: 0.2651 | LB_LOSS: 0.9997 [2026-05-23 20:15:20] Validation | Batch 20/1463 | Loss: 0.7735 | LM_LOSS: 0.2734 | LB_LOSS: 1.0002 [2026-05-23 20:15:24] Validation | Batch 30/1463 | Loss: 0.7759 | LM_LOSS: 0.2757 | LB_LOSS: 1.0004 [2026-05-23 20:15:26] Validation | Batch 40/1463 | Loss: 0.7824 | LM_LOSS: 0.2822 | LB_LOSS: 1.0004 [2026-05-23 20:15:29] Validation | Batch 50/1463 | Loss: 0.7900 | LM_LOSS: 0.2898 | LB_LOSS: 1.0003 [2026-05-23 20:15:33] Validation | Batch 60/1463 | Loss: 0.7897 | LM_LOSS: 0.2895 | LB_LOSS: 1.0004 [2026-05-23 20:15:37] Validation | Batch 70/1463 | Loss: 0.7899 | LM_LOSS: 0.2897 | LB_LOSS: 1.0005 [2026-05-23 20:15:41] Validation | Batch 80/1463 | Loss: 0.7920 | LM_LOSS: 0.2918 | LB_LOSS: 1.0005 [2026-05-23 20:15:43] Validation | Batch 90/1463 | Loss: 0.7915 | LM_LOSS: 0.2913 | LB_LOSS: 1.0005 [2026-05-23 20:15:46] Validation | Batch 100/1463 | Loss: 0.7911 | LM_LOSS: 0.2909 | LB_LOSS: 1.0004 [2026-05-23 20:15:50] Validation | Batch 110/1463 | Loss: 0.7889 | LM_LOSS: 0.2886 | LB_LOSS: 1.0005 [2026-05-23 20:15:52] Validation | Batch 120/1463 | Loss: 0.7907 | LM_LOSS: 0.2904 | LB_LOSS: 1.0004 [2026-05-23 20:15:54] Validation | Batch 130/1463 | Loss: 0.7920 | LM_LOSS: 0.2918 | LB_LOSS: 1.0005 [2026-05-23 20:15:58] Validation | Batch 140/1463 | Loss: 0.7933 | LM_LOSS: 0.2931 | LB_LOSS: 1.0005 [2026-05-23 20:16:02] Validation | Batch 150/1463 | Loss: 0.7912 | LM_LOSS: 0.2909 | LB_LOSS: 1.0006 [2026-05-23 20:16:05] Validation | Batch 160/1463 | Loss: 0.7917 | LM_LOSS: 0.2915 | LB_LOSS: 1.0006 [2026-05-23 20:16:08] Validation | Batch 170/1463 | Loss: 0.7909 | LM_LOSS: 0.2906 | LB_LOSS: 1.0006 [2026-05-23 20:16:11] Validation | Batch 180/1463 | Loss: 0.7918 | LM_LOSS: 0.2916 | LB_LOSS: 1.0006 [2026-05-23 20:16:14] Validation | Batch 190/1463 | Loss: 0.7929 | LM_LOSS: 0.2927 | LB_LOSS: 1.0006 [2026-05-23 20:16:17] Validation | Batch 200/1463 | Loss: 0.7929 | LM_LOSS: 0.2926 | LB_LOSS: 1.0006 [2026-05-23 20:16:20] Validation | Batch 210/1463 | Loss: 0.7937 | LM_LOSS: 0.2935 | LB_LOSS: 1.0006 [2026-05-23 20:16:24] Validation | Batch 220/1463 | Loss: 0.7932 | LM_LOSS: 0.2929 | LB_LOSS: 1.0006 [2026-05-23 20:16:27] Validation | Batch 230/1463 | Loss: 0.7947 | LM_LOSS: 0.2944 | LB_LOSS: 1.0006 [2026-05-23 20:16:30] Validation | Batch 240/1463 | Loss: 0.7963 | LM_LOSS: 0.2961 | LB_LOSS: 1.0006 [2026-05-23 20:16:34] Validation | Batch 250/1463 | Loss: 0.7954 | LM_LOSS: 0.2951 | LB_LOSS: 1.0006 [2026-05-23 20:16:36] Validation | Batch 260/1463 | Loss: 0.7970 | LM_LOSS: 0.2967 | LB_LOSS: 1.0006 [2026-05-23 20:16:40] Validation | Batch 270/1463 | Loss: 0.7957 | LM_LOSS: 0.2954 | LB_LOSS: 1.0006 [2026-05-23 20:16:43] Validation | Batch 280/1463 | Loss: 0.7960 | LM_LOSS: 0.2957 | LB_LOSS: 1.0006 [2026-05-23 20:16:46] Validation | Batch 290/1463 | Loss: 0.7951 | LM_LOSS: 0.2948 | LB_LOSS: 1.0006 [2026-05-23 20:16:48] Validation | Batch 300/1463 | Loss: 0.7955 | LM_LOSS: 0.2952 | LB_LOSS: 1.0006 [2026-05-23 20:16:52] Validation | Batch 310/1463 | Loss: 0.7965 | LM_LOSS: 0.2962 | LB_LOSS: 1.0006 [2026-05-23 20:16:54] Validation | Batch 320/1463 | Loss: 0.7953 | LM_LOSS: 0.2950 | LB_LOSS: 1.0006 [2026-05-23 20:16:58] Validation | Batch 330/1463 | Loss: 0.7950 | LM_LOSS: 0.2947 | LB_LOSS: 1.0006 [2026-05-23 20:17:00] Validation | Batch 340/1463 | Loss: 0.7946 | LM_LOSS: 0.2943 | LB_LOSS: 1.0006 [2026-05-23 20:17:04] Validation | Batch 350/1463 | Loss: 0.7953 | LM_LOSS: 0.2950 | LB_LOSS: 1.0006 [2026-05-23 20:17:07] Validation | Batch 360/1463 | Loss: 0.7950 | LM_LOSS: 0.2947 | LB_LOSS: 1.0006 [2026-05-23 20:17:09] Validation | Batch 370/1463 | Loss: 0.7949 | LM_LOSS: 0.2946 | LB_LOSS: 1.0006 [2026-05-23 20:17:12] Validation | Batch 380/1463 | Loss: 0.7947 | LM_LOSS: 0.2944 | LB_LOSS: 1.0006 [2026-05-23 20:17:15] Validation | Batch 390/1463 | Loss: 0.7954 | LM_LOSS: 0.2952 | LB_LOSS: 1.0006 [2026-05-23 20:17:18] Validation | Batch 400/1463 | Loss: 0.7953 | LM_LOSS: 0.2950 | LB_LOSS: 1.0006 [2026-05-23 20:17:21] Validation | Batch 410/1463 | Loss: 0.7950 | LM_LOSS: 0.2947 | LB_LOSS: 1.0006 [2026-05-23 20:17:24] Validation | Batch 420/1463 | Loss: 0.7951 | LM_LOSS: 0.2948 | LB_LOSS: 1.0006 [2026-05-23 20:17:28] Validation | Batch 430/1463 | Loss: 0.7947 | LM_LOSS: 0.2944 | LB_LOSS: 1.0006 [2026-05-23 20:17:30] Validation | Batch 440/1463 | Loss: 0.7949 | LM_LOSS: 0.2946 | LB_LOSS: 1.0006 [2026-05-23 20:17:33] Validation | Batch 450/1463 | Loss: 0.7946 | LM_LOSS: 0.2943 | LB_LOSS: 1.0006 [2026-05-23 20:17:36] Validation | Batch 460/1463 | Loss: 0.7946 | LM_LOSS: 0.2943 | LB_LOSS: 1.0006 [2026-05-23 20:17:40] Validation | Batch 470/1463 | Loss: 0.7951 | LM_LOSS: 0.2948 | LB_LOSS: 1.0006 [2026-05-23 20:17:43] Validation | Batch 480/1463 | Loss: 0.7949 | LM_LOSS: 0.2946 | LB_LOSS: 1.0006 [2026-05-23 20:17:46] Validation | Batch 490/1463 | Loss: 0.7948 | LM_LOSS: 0.2945 | LB_LOSS: 1.0006 [2026-05-23 20:17:50] Validation | Batch 500/1463 | Loss: 0.7946 | LM_LOSS: 0.2943 | LB_LOSS: 1.0006 [2026-05-23 20:17:53] Validation | Batch 510/1463 | Loss: 0.7939 | LM_LOSS: 0.2936 | LB_LOSS: 1.0006 [2026-05-23 20:17:56] Validation | Batch 520/1463 | Loss: 0.7937 | LM_LOSS: 0.2934 | LB_LOSS: 1.0006 [2026-05-23 20:17:58] Validation | Batch 530/1463 | Loss: 0.7941 | LM_LOSS: 0.2938 | LB_LOSS: 1.0006 [2026-05-23 20:18:01] Validation | Batch 540/1463 | Loss: 0.7943 | LM_LOSS: 0.2940 | LB_LOSS: 1.0006 [2026-05-23 20:18:03] Validation | Batch 550/1463 | Loss: 0.7938 | LM_LOSS: 0.2935 | LB_LOSS: 1.0006 [2026-05-23 20:18:06] Validation | Batch 560/1463 | Loss: 0.7940 | LM_LOSS: 0.2937 | LB_LOSS: 1.0006 [2026-05-23 20:18:10] Validation | Batch 570/1463 | Loss: 0.7939 | LM_LOSS: 0.2936 | LB_LOSS: 1.0006 [2026-05-23 20:18:13] Validation | Batch 580/1463 | Loss: 0.7935 | LM_LOSS: 0.2932 | LB_LOSS: 1.0006 [2026-05-23 20:18:17] Validation | Batch 590/1463 | Loss: 0.7932 | LM_LOSS: 0.2929 | LB_LOSS: 1.0006 [2026-05-23 20:18:20] Validation | Batch 600/1463 | Loss: 0.7935 | LM_LOSS: 0.2932 | LB_LOSS: 1.0006 [2026-05-23 20:18:23] Validation | Batch 610/1463 | Loss: 0.7938 | LM_LOSS: 0.2935 | LB_LOSS: 1.0006 [2026-05-23 20:18:26] Validation | Batch 620/1463 | Loss: 0.7933 | LM_LOSS: 0.2930 | LB_LOSS: 1.0006 [2026-05-23 20:18:29] Validation | Batch 630/1463 | Loss: 0.7927 | LM_LOSS: 0.2924 | LB_LOSS: 1.0006 [2026-05-23 20:18:32] Validation | Batch 640/1463 | Loss: 0.7926 | LM_LOSS: 0.2923 | LB_LOSS: 1.0006 [2026-05-23 20:18:35] Validation | Batch 650/1463 | Loss: 0.7921 | LM_LOSS: 0.2918 | LB_LOSS: 1.0006 [2026-05-23 20:18:37] Validation | Batch 660/1463 | Loss: 0.7917 | LM_LOSS: 0.2914 | LB_LOSS: 1.0006 [2026-05-23 20:18:40] Validation | Batch 670/1463 | Loss: 0.7913 | LM_LOSS: 0.2910 | LB_LOSS: 1.0006 [2026-05-23 20:18:43] Validation | Batch 680/1463 | Loss: 0.7913 | LM_LOSS: 0.2910 | LB_LOSS: 1.0006 [2026-05-23 20:18:47] Validation | Batch 690/1463 | Loss: 0.7915 | LM_LOSS: 0.2913 | LB_LOSS: 1.0006 [2026-05-23 20:18:49] Validation | Batch 700/1463 | Loss: 0.7914 | LM_LOSS: 0.2911 | LB_LOSS: 1.0006 [2026-05-23 20:18:52] Validation | Batch 710/1463 | Loss: 0.7916 | LM_LOSS: 0.2913 | LB_LOSS: 1.0006 [2026-05-23 20:18:55] Validation | Batch 720/1463 | Loss: 0.7924 | LM_LOSS: 0.2921 | LB_LOSS: 1.0006 [2026-05-23 20:18:58] Validation | Batch 730/1463 | Loss: 0.7922 | LM_LOSS: 0.2919 | LB_LOSS: 1.0006 [2026-05-23 20:19:00] Validation | Batch 740/1463 | Loss: 0.7920 | LM_LOSS: 0.2917 | LB_LOSS: 1.0006 [2026-05-23 20:19:03] Validation | Batch 750/1463 | Loss: 0.7920 | LM_LOSS: 0.2917 | LB_LOSS: 1.0006 [2026-05-23 20:19:06] Validation | Batch 760/1463 | Loss: 0.7919 | LM_LOSS: 0.2916 | LB_LOSS: 1.0006 [2026-05-23 20:19:08] Validation | Batch 770/1463 | Loss: 0.7919 | LM_LOSS: 0.2916 | LB_LOSS: 1.0006 [2026-05-23 20:19:11] Validation | Batch 780/1463 | Loss: 0.7918 | LM_LOSS: 0.2915 | LB_LOSS: 1.0006 [2026-05-23 20:19:13] Validation | Batch 790/1463 | Loss: 0.7920 | LM_LOSS: 0.2918 | LB_LOSS: 1.0006 [2026-05-23 20:19:16] Validation | Batch 800/1463 | Loss: 0.7920 | LM_LOSS: 0.2917 | LB_LOSS: 1.0006 [2026-05-23 20:19:19] Validation | Batch 810/1463 | Loss: 0.7917 | LM_LOSS: 0.2914 | LB_LOSS: 1.0006 [2026-05-23 20:19:22] Validation | Batch 820/1463 | Loss: 0.7918 | LM_LOSS: 0.2915 | LB_LOSS: 1.0006 [2026-05-23 20:19:25] Validation | Batch 830/1463 | Loss: 0.7920 | LM_LOSS: 0.2917 | LB_LOSS: 1.0006 [2026-05-23 20:19:29] Validation | Batch 840/1463 | Loss: 0.7921 | LM_LOSS: 0.2918 | LB_LOSS: 1.0006 [2026-05-23 20:19:32] Validation | Batch 850/1463 | Loss: 0.7924 | LM_LOSS: 0.2921 | LB_LOSS: 1.0006 [2026-05-23 20:19:35] Validation | Batch 860/1463 | Loss: 0.7926 | LM_LOSS: 0.2923 | LB_LOSS: 1.0006 [2026-05-23 20:19:38] Validation | Batch 870/1463 | Loss: 0.7927 | LM_LOSS: 0.2924 | LB_LOSS: 1.0006 [2026-05-23 20:19:41] Validation | Batch 880/1463 | Loss: 0.7919 | LM_LOSS: 0.2916 | LB_LOSS: 1.0006 [2026-05-23 20:19:44] Validation | Batch 890/1463 | Loss: 0.7919 | LM_LOSS: 0.2916 | LB_LOSS: 1.0006 [2026-05-23 20:19:47] Validation | Batch 900/1463 | Loss: 0.7916 | LM_LOSS: 0.2913 | LB_LOSS: 1.0006 [2026-05-23 20:19:50] Validation | Batch 910/1463 | Loss: 0.7916 | LM_LOSS: 0.2913 | LB_LOSS: 1.0006 [2026-05-23 20:19:54] Validation | Batch 920/1463 | Loss: 0.7918 | LM_LOSS: 0.2915 | LB_LOSS: 1.0006 [2026-05-23 20:19:56] Validation | Batch 930/1463 | Loss: 0.7916 | LM_LOSS: 0.2913 | LB_LOSS: 1.0006 [2026-05-23 20:20:00] Validation | Batch 940/1463 | Loss: 0.7914 | LM_LOSS: 0.2911 | LB_LOSS: 1.0006 [2026-05-23 20:20:03] Validation | Batch 950/1463 | Loss: 0.7916 | LM_LOSS: 0.2913 | LB_LOSS: 1.0006 [2026-05-23 20:20:06] Validation | Batch 960/1463 | Loss: 0.7917 | LM_LOSS: 0.2914 | LB_LOSS: 1.0006 [2026-05-23 20:20:10] Validation | Batch 970/1463 | Loss: 0.7917 | LM_LOSS: 0.2914 | LB_LOSS: 1.0006 [2026-05-23 20:20:13] Validation | Batch 980/1463 | Loss: 0.7920 | LM_LOSS: 0.2917 | LB_LOSS: 1.0006 [2026-05-23 20:20:16] Validation | Batch 990/1463 | Loss: 0.7922 | LM_LOSS: 0.2919 | LB_LOSS: 1.0006 [2026-05-23 20:20:19] Validation | Batch 1000/1463 | Loss: 0.7928 | LM_LOSS: 0.2925 | LB_LOSS: 1.0006 [2026-05-23 20:20:22] Validation | Batch 1010/1463 | Loss: 0.7928 | LM_LOSS: 0.2925 | LB_LOSS: 1.0006 [2026-05-23 20:20:25] Validation | Batch 1020/1463 | Loss: 0.7929 | LM_LOSS: 0.2926 | LB_LOSS: 1.0006 [2026-05-23 20:20:28] Validation | Batch 1030/1463 | Loss: 0.7933 | LM_LOSS: 0.2930 | LB_LOSS: 1.0006 [2026-05-23 20:20:31] Validation | Batch 1040/1463 | Loss: 0.7930 | LM_LOSS: 0.2927 | LB_LOSS: 1.0006 [2026-05-23 20:20:34] Validation | Batch 1050/1463 | Loss: 0.7930 | LM_LOSS: 0.2927 | LB_LOSS: 1.0006 [2026-05-23 20:20:38] Validation | Batch 1060/1463 | Loss: 0.7929 | LM_LOSS: 0.2926 | LB_LOSS: 1.0006 [2026-05-23 20:20:41] Validation | Batch 1070/1463 | Loss: 0.7929 | LM_LOSS: 0.2927 | LB_LOSS: 1.0006 [2026-05-23 20:20:43] Validation | Batch 1080/1463 | Loss: 0.7928 | LM_LOSS: 0.2925 | LB_LOSS: 1.0006 [2026-05-23 20:20:46] Validation | Batch 1090/1463 | Loss: 0.7928 | LM_LOSS: 0.2925 | LB_LOSS: 1.0006 [2026-05-23 20:20:49] Validation | Batch 1100/1463 | Loss: 0.7926 | LM_LOSS: 0.2923 | LB_LOSS: 1.0006 [2026-05-23 20:20:51] Validation | Batch 1110/1463 | Loss: 0.7926 | LM_LOSS: 0.2923 | LB_LOSS: 1.0006 [2026-05-23 20:20:54] Validation | Batch 1120/1463 | Loss: 0.7924 | LM_LOSS: 0.2921 | LB_LOSS: 1.0006 [2026-05-23 20:20:58] Validation | Batch 1130/1463 | Loss: 0.7922 | LM_LOSS: 0.2919 | LB_LOSS: 1.0006 [2026-05-23 20:21:01] Validation | Batch 1140/1463 | Loss: 0.7924 | LM_LOSS: 0.2921 | LB_LOSS: 1.0006 [2026-05-23 20:21:05] Validation | Batch 1150/1463 | Loss: 0.7923 | LM_LOSS: 0.2920 | LB_LOSS: 1.0006 [2026-05-23 20:21:08] Validation | Batch 1160/1463 | Loss: 0.7922 | LM_LOSS: 0.2919 | LB_LOSS: 1.0006 [2026-05-23 20:21:11] Validation | Batch 1170/1463 | Loss: 0.7922 | LM_LOSS: 0.2919 | LB_LOSS: 1.0006 [2026-05-23 20:21:14] Validation | Batch 1180/1463 | Loss: 0.7923 | LM_LOSS: 0.2920 | LB_LOSS: 1.0006 [2026-05-23 20:21:17] Validation | Batch 1190/1463 | Loss: 0.7925 | LM_LOSS: 0.2922 | LB_LOSS: 1.0006 [2026-05-23 20:21:20] Validation | Batch 1200/1463 | Loss: 0.7924 | LM_LOSS: 0.2921 | LB_LOSS: 1.0006 [2026-05-23 20:21:23] Validation | Batch 1210/1463 | Loss: 0.7926 | LM_LOSS: 0.2923 | LB_LOSS: 1.0006 [2026-05-23 20:21:26] Validation | Batch 1220/1463 | Loss: 0.7930 | LM_LOSS: 0.2927 | LB_LOSS: 1.0006 [2026-05-23 20:21:29] Validation | Batch 1230/1463 | Loss: 0.7930 | LM_LOSS: 0.2927 | LB_LOSS: 1.0006 [2026-05-23 20:21:32] Validation | Batch 1240/1463 | Loss: 0.7932 | LM_LOSS: 0.2929 | LB_LOSS: 1.0006 [2026-05-23 20:21:35] Validation | Batch 1250/1463 | Loss: 0.7933 | LM_LOSS: 0.2930 | LB_LOSS: 1.0006 [2026-05-23 20:21:39] Validation | Batch 1260/1463 | Loss: 0.7933 | LM_LOSS: 0.2930 | LB_LOSS: 1.0006 [2026-05-23 20:21:41] Validation | Batch 1270/1463 | Loss: 0.7931 | LM_LOSS: 0.2928 | LB_LOSS: 1.0006 [2026-05-23 20:21:45] Validation | Batch 1280/1463 | Loss: 0.7930 | LM_LOSS: 0.2927 | LB_LOSS: 1.0006 [2026-05-23 20:21:48] Validation | Batch 1290/1463 | Loss: 0.7929 | LM_LOSS: 0.2926 | LB_LOSS: 1.0006 [2026-05-23 20:21:51] Validation | Batch 1300/1463 | Loss: 0.7927 | LM_LOSS: 0.2924 | LB_LOSS: 1.0006 [2026-05-23 20:21:54] Validation | Batch 1310/1463 | Loss: 0.7926 | LM_LOSS: 0.2923 | LB_LOSS: 1.0006 [2026-05-23 20:21:56] Validation | Batch 1320/1463 | Loss: 0.7927 | LM_LOSS: 0.2924 | LB_LOSS: 1.0006 [2026-05-23 20:21:59] Validation | Batch 1330/1463 | Loss: 0.7927 | LM_LOSS: 0.2924 | LB_LOSS: 1.0006 [2026-05-23 20:22:02] Validation | Batch 1340/1463 | Loss: 0.7929 | LM_LOSS: 0.2927 | LB_LOSS: 1.0006 [2026-05-23 20:22:05] Validation | Batch 1350/1463 | Loss: 0.7929 | LM_LOSS: 0.2926 | LB_LOSS: 1.0006 [2026-05-23 20:22:09] Validation | Batch 1360/1463 | Loss: 0.7932 | LM_LOSS: 0.2929 | LB_LOSS: 1.0006 [2026-05-23 20:22:12] Validation | Batch 1370/1463 | Loss: 0.7935 | LM_LOSS: 0.2932 | LB_LOSS: 1.0006 [2026-05-23 20:22:15] Validation | Batch 1380/1463 | Loss: 0.7933 | LM_LOSS: 0.2930 | LB_LOSS: 1.0006 [2026-05-23 20:22:17] Validation | Batch 1390/1463 | Loss: 0.7932 | LM_LOSS: 0.2929 | LB_LOSS: 1.0006 [2026-05-23 20:22:21] Validation | Batch 1400/1463 | Loss: 0.7935 | LM_LOSS: 0.2932 | LB_LOSS: 1.0006 [2026-05-23 20:22:23] Validation | Batch 1410/1463 | Loss: 0.7938 | LM_LOSS: 0.2935 | LB_LOSS: 1.0006 [2026-05-23 20:22:26] Validation | Batch 1420/1463 | Loss: 0.7937 | LM_LOSS: 0.2935 | LB_LOSS: 1.0006 [2026-05-23 20:22:30] Validation | Batch 1430/1463 | Loss: 0.7938 | LM_LOSS: 0.2935 | LB_LOSS: 1.0006 [2026-05-23 20:22:33] Validation | Batch 1440/1463 | Loss: 0.7942 | LM_LOSS: 0.2939 | LB_LOSS: 1.0006 [2026-05-23 20:22:36] Validation | Batch 1450/1463 | Loss: 0.7939 | LM_LOSS: 0.2937 | LB_LOSS: 1.0006 [2026-05-23 20:22:39] Validation | Batch 1460/1463 | Loss: 0.7942 | LM_LOSS: 0.2940 | LB_LOSS: 1.0006 [2026-05-23 20:22:40] Validation | Batch 1463/1463 | Loss: 0.7941 | LM_LOSS: 0.2939 | LB_LOSS: 1.0006 [2026-05-23 20:22:40] Validation | Loss: 0.7941 | LM_LOSS: 0.2939 | LB_LOSS: 1.0006 | PPL: 1.34 | Time: 445.16s [2026-05-23 20:22:52] Epoch 2 | Step 6010 | Loss: 0.7183 | LM: 0.2190 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.389 | LR: 1.00e-04 [2026-05-23 20:23:04] Epoch 2 | Step 6020 | Loss: 0.7180 | LM: 0.2189 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.389 | LR: 1.00e-04 [2026-05-23 20:23:16] Epoch 2 | Step 6030 | Loss: 0.7181 | LM: 0.2190 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.389 | LR: 1.00e-04 [2026-05-23 20:23:28] Epoch 2 | Step 6040 | Loss: 0.7182 | LM: 0.2188 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.389 | LR: 1.00e-04 [2026-05-23 20:23:42] Epoch 2 | Step 6050 | Loss: 0.7183 | LM: 0.2192 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.389 | LR: 1.00e-04 [2026-05-23 20:23:54] Epoch 2 | Step 6060 | Loss: 0.7183 | LM: 0.2193 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.389 | LR: 1.00e-04 [2026-05-23 20:24:06] Epoch 2 | Step 6070 | Loss: 0.7182 | LM: 0.2192 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.389 | LR: 1.00e-04 [2026-05-23 20:24:19] Epoch 2 | Step 6080 | Loss: 0.7185 | LM: 0.2194 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.389 | LR: 1.00e-04 [2026-05-23 20:24:32] Epoch 2 | Step 6090 | Loss: 0.7185 | LM: 0.2194 | LB: 1.0013 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.389 | LR: 1.00e-04 [2026-05-23 20:24:44] Epoch 2 | Step 6100 | Loss: 0.7186 | LM: 0.2195 | LB: 1.0013 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.389 | LR: 1.00e-04 [2026-05-23 20:24:57] Epoch 2 | Step 6110 | Loss: 0.7185 | LM: 0.2198 | LB: 1.0013 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.389 | LR: 1.00e-04 [2026-05-23 20:25:09] Epoch 2 | Step 6120 | Loss: 0.7185 | LM: 0.2197 | LB: 1.0013 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.389 | LR: 1.00e-04 [2026-05-23 20:25:22] Epoch 2 | Step 6130 | Loss: 0.7185 | LM: 0.2198 | LB: 1.0013 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.389 | LR: 1.00e-04 [2026-05-23 20:25:35] Epoch 2 | Step 6140 | Loss: 0.7183 | LM: 0.2196 | LB: 1.0013 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.389 | LR: 1.00e-04 [2026-05-23 20:25:50] Epoch 2 | Step 6150 | Loss: 0.7184 | LM: 0.2198 | LB: 1.0013 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.389 | LR: 1.00e-04 [2026-05-23 20:26:02] Epoch 2 | Step 6160 | Loss: 0.7183 | LM: 0.2196 | LB: 1.0013 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.389 | LR: 1.00e-04 [2026-05-23 20:26:14] Epoch 2 | Step 6170 | Loss: 0.7183 | LM: 0.2193 | LB: 1.0013 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.389 | LR: 1.00e-04 [2026-05-23 20:26:26] Epoch 2 | Step 6180 | Loss: 0.7184 | LM: 0.2192 | LB: 1.0013 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.389 | LR: 1.00e-04 [2026-05-23 20:26:38] Epoch 2 | Step 6190 | Loss: 0.7182 | LM: 0.2190 | LB: 1.0013 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.389 | LR: 1.00e-04 [2026-05-23 20:26:50] Epoch 2 | Step 6200 | Loss: 0.7183 | LM: 0.2192 | LB: 1.0013 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.389 | LR: 1.00e-04 [2026-05-23 20:27:02] Epoch 2 | Step 6210 | Loss: 0.7183 | LM: 0.2191 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.389 | LR: 1.00e-04 [2026-05-23 20:27:15] Epoch 2 | Step 6220 | Loss: 0.7184 | LM: 0.2194 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.389 | LR: 1.00e-04 [2026-05-23 20:27:27] Epoch 2 | Step 6230 | Loss: 0.7184 | LM: 0.2194 | LB: 1.0013 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.389 | LR: 1.00e-04 [2026-05-23 20:27:39] Epoch 2 | Step 6240 | Loss: 0.7186 | LM: 0.2196 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.389 | LR: 1.00e-04 [2026-05-23 20:27:51] Epoch 2 | Step 6250 | Loss: 0.7185 | LM: 0.2197 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.389 | LR: 1.00e-04 [2026-05-23 20:28:03] Epoch 2 | Step 6260 | Loss: 0.7185 | LM: 0.2197 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.389 | LR: 1.00e-04 [2026-05-23 20:28:16] Epoch 2 | Step 6270 | Loss: 0.7186 | LM: 0.2197 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.389 | LR: 1.00e-04 [2026-05-23 20:28:27] Epoch 2 | Step 6280 | Loss: 0.7185 | LM: 0.2194 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.389 | LR: 1.00e-04 [2026-05-23 20:28:41] Epoch 2 | Step 6290 | Loss: 0.7185 | LM: 0.2198 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.389 | LR: 1.00e-04 [2026-05-23 20:28:52] Epoch 2 | Step 6300 | Loss: 0.7185 | LM: 0.2196 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.389 | LR: 1.00e-04 [2026-05-23 20:29:04] Epoch 2 | Step 6310 | Loss: 0.7184 | LM: 0.2193 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.389 | LR: 1.00e-04 [2026-05-23 20:29:17] Epoch 2 | Step 6320 | Loss: 0.7185 | LM: 0.2191 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.389 | LR: 1.00e-04 [2026-05-23 20:29:29] Epoch 2 | Step 6330 | Loss: 0.7184 | LM: 0.2190 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.389 | LR: 1.00e-04 [2026-05-23 20:29:41] Epoch 2 | Step 6340 | Loss: 0.7184 | LM: 0.2190 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.389 | LR: 1.00e-04 [2026-05-23 20:29:54] Epoch 2 | Step 6350 | Loss: 0.7186 | LM: 0.2192 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.389 | LR: 1.00e-04 [2026-05-23 20:30:05] Epoch 2 | Step 6360 | Loss: 0.7186 | LM: 0.2192 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.389 | LR: 1.00e-04 [2026-05-23 20:30:17] Epoch 2 | Step 6370 | Loss: 0.7185 | LM: 0.2190 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.389 | LR: 1.00e-04 [2026-05-23 20:30:29] Epoch 2 | Step 6380 | Loss: 0.7186 | LM: 0.2190 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.389 | LR: 1.00e-04 [2026-05-23 20:30:42] Epoch 2 | Step 6390 | Loss: 0.7185 | LM: 0.2189 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.389 | LR: 1.00e-04 [2026-05-23 20:30:56] Epoch 2 | Step 6400 | Loss: 0.7187 | LM: 0.2190 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.389 | LR: 1.00e-04 [2026-05-23 20:31:09] Epoch 2 | Step 6410 | Loss: 0.7186 | LM: 0.2188 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.389 | LR: 1.00e-04 [2026-05-23 20:31:21] Epoch 2 | Step 6420 | Loss: 0.7186 | LM: 0.2189 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.389 | LR: 1.00e-04 [2026-05-23 20:31:33] Epoch 2 | Step 6430 | Loss: 0.7185 | LM: 0.2183 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.389 | LR: 1.00e-04 [2026-05-23 20:31:46] Epoch 2 | Step 6440 | Loss: 0.7185 | LM: 0.2183 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.389 | LR: 1.00e-04 [2026-05-23 20:31:59] Epoch 2 | Step 6450 | Loss: 0.7184 | LM: 0.2180 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.389 | LR: 1.00e-04 [2026-05-23 20:32:12] Epoch 2 | Step 6460 | Loss: 0.7184 | LM: 0.2177 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.389 | LR: 1.00e-04 [2026-05-23 20:32:25] Epoch 2 | Step 6470 | Loss: 0.7185 | LM: 0.2179 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.389 | LR: 1.00e-04 [2026-05-23 20:32:37] Epoch 2 | Step 6480 | Loss: 0.7185 | LM: 0.2177 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.389 | LR: 1.00e-04 [2026-05-23 20:32:49] Epoch 2 | Step 6490 | Loss: 0.7184 | LM: 0.2177 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.389 | LR: 1.00e-04 [2026-05-23 20:33:02] Epoch 2 | Step 6500 | Loss: 0.7185 | LM: 0.2175 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.389 | LR: 1.00e-04 [2026-05-23 20:33:15] Epoch 2 | Step 6510 | Loss: 0.7186 | LM: 0.2176 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.389 | LR: 1.00e-04 [2026-05-23 20:33:27] Epoch 2 | Step 6520 | Loss: 0.7185 | LM: 0.2175 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.389 | LR: 1.00e-04 [2026-05-23 20:33:40] Epoch 2 | Step 6530 | Loss: 0.7185 | LM: 0.2171 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.389 | LR: 1.00e-04 [2026-05-23 20:33:53] Epoch 2 | Step 6540 | Loss: 0.7184 | LM: 0.2171 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.389 | LR: 1.00e-04 [2026-05-23 20:34:05] Epoch 2 | Step 6550 | Loss: 0.7184 | LM: 0.2172 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.389 | LR: 1.00e-04 [2026-05-23 20:34:16] Epoch 2 | Step 6560 | Loss: 0.7184 | LM: 0.2171 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.389 | LR: 1.00e-04 [2026-05-23 20:34:29] Epoch 2 | Step 6570 | Loss: 0.7182 | LM: 0.2166 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.389 | LR: 1.00e-04 [2026-05-23 20:34:41] Epoch 2 | Step 6580 | Loss: 0.7182 | LM: 0.2170 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.389 | LR: 1.00e-04 [2026-05-23 20:34:54] Epoch 2 | Step 6590 | Loss: 0.7182 | LM: 0.2169 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.389 | LR: 1.00e-04 [2026-05-23 20:35:07] Epoch 2 | Step 6600 | Loss: 0.7183 | LM: 0.2167 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.389 | LR: 1.00e-04 [2026-05-23 20:35:20] Epoch 2 | Step 6610 | Loss: 0.7183 | LM: 0.2169 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.389 | LR: 1.00e-04 [2026-05-23 20:35:32] Epoch 2 | Step 6620 | Loss: 0.7183 | LM: 0.2168 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.389 | LR: 1.00e-04 [2026-05-23 20:35:44] Epoch 2 | Step 6630 | Loss: 0.7184 | LM: 0.2169 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.389 | LR: 1.00e-04 [2026-05-23 20:35:57] Epoch 2 | Step 6640 | Loss: 0.7184 | LM: 0.2170 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.389 | LR: 1.00e-04 [2026-05-23 20:36:11] Epoch 2 | Step 6650 | Loss: 0.7184 | LM: 0.2169 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.389 | LR: 1.00e-04 [2026-05-23 20:36:23] Epoch 2 | Step 6660 | Loss: 0.7183 | LM: 0.2168 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.389 | LR: 1.00e-04 [2026-05-23 20:36:35] Epoch 2 | Step 6670 | Loss: 0.7183 | LM: 0.2166 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.389 | LR: 1.00e-04 [2026-05-23 20:36:48] Epoch 2 | Step 6680 | Loss: 0.7184 | LM: 0.2169 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.389 | LR: 1.00e-04 [2026-05-23 20:36:59] Epoch 2 | Step 6690 | Loss: 0.7184 | LM: 0.2167 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.389 | LR: 1.00e-04 [2026-05-23 20:37:12] Epoch 2 | Step 6700 | Loss: 0.7186 | LM: 0.2170 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.389 | LR: 1.00e-04 [2026-05-23 20:37:25] Epoch 2 | Step 6710 | Loss: 0.7185 | LM: 0.2170 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.389 | LR: 1.00e-04 [2026-05-23 20:37:38] Epoch 2 | Step 6720 | Loss: 0.7184 | LM: 0.2168 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.389 | LR: 1.00e-04 [2026-05-23 20:37:51] Epoch 2 | Step 6730 | Loss: 0.7184 | LM: 0.2168 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.389 | LR: 1.00e-04 [2026-05-23 20:38:03] Epoch 2 | Step 6740 | Loss: 0.7183 | LM: 0.2168 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.389 | LR: 1.00e-04 [2026-05-23 20:38:15] Epoch 2 | Step 6750 | Loss: 0.7184 | LM: 0.2170 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.389 | LR: 1.00e-04 [2026-05-23 20:38:28] Epoch 2 | Step 6760 | Loss: 0.7183 | LM: 0.2169 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.389 | LR: 1.00e-04 [2026-05-23 20:38:41] Epoch 2 | Step 6770 | Loss: 0.7182 | LM: 0.2167 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.389 | LR: 1.00e-04 [2026-05-23 20:38:53] Epoch 2 | Step 6780 | Loss: 0.7181 | LM: 0.2164 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.389 | LR: 1.00e-04 [2026-05-23 20:39:05] Epoch 2 | Step 6790 | Loss: 0.7182 | LM: 0.2166 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.389 | LR: 1.00e-04 [2026-05-23 20:39:18] Epoch 2 | Step 6800 | Loss: 0.7182 | LM: 0.2168 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.389 | LR: 1.00e-04 [2026-05-23 20:39:29] Epoch 2 | Step 6810 | Loss: 0.7182 | LM: 0.2170 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.389 | LR: 1.00e-04 [2026-05-23 20:39:42] Epoch 2 | Step 6820 | Loss: 0.7181 | LM: 0.2169 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.389 | LR: 1.00e-04 [2026-05-23 20:39:53] Epoch 2 | Step 6830 | Loss: 0.7181 | LM: 0.2168 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.389 | LR: 1.00e-04 [2026-05-23 20:40:06] Epoch 2 | Step 6840 | Loss: 0.7181 | LM: 0.2167 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.389 | LR: 1.00e-04 [2026-05-23 20:40:18] Epoch 2 | Step 6850 | Loss: 0.7180 | LM: 0.2170 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.389 | LR: 1.00e-04 [2026-05-23 20:40:30] Epoch 2 | Step 6860 | Loss: 0.7181 | LM: 0.2170 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.389 | LR: 1.00e-04 [2026-05-23 20:40:41] Epoch 2 | Step 6870 | Loss: 0.7181 | LM: 0.2170 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.389 | LR: 1.00e-04 [2026-05-23 20:40:54] Epoch 2 | Step 6880 | Loss: 0.7182 | LM: 0.2168 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.389 | LR: 1.00e-04 [2026-05-23 20:41:06] Epoch 2 | Step 6890 | Loss: 0.7180 | LM: 0.2170 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.389 | LR: 1.00e-04 [2026-05-23 20:41:19] Epoch 2 | Step 6900 | Loss: 0.7179 | LM: 0.2168 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.389 | LR: 1.00e-04 [2026-05-23 20:41:31] Epoch 2 | Step 6910 | Loss: 0.7179 | LM: 0.2167 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.389 | LR: 1.00e-04 [2026-05-23 20:41:43] Epoch 2 | Step 6920 | Loss: 0.7179 | LM: 0.2168 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.389 | LR: 1.00e-04 [2026-05-23 20:41:55] Epoch 2 | Step 6930 | Loss: 0.7180 | LM: 0.2168 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.389 | LR: 1.00e-04 [2026-05-23 20:42:08] Epoch 2 | Step 6940 | Loss: 0.7180 | LM: 0.2168 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.389 | LR: 1.00e-04 [2026-05-23 20:42:20] Epoch 2 | Step 6950 | Loss: 0.7180 | LM: 0.2168 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.389 | LR: 1.00e-04 [2026-05-23 20:42:33] Epoch 2 | Step 6960 | Loss: 0.7182 | LM: 0.2169 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.389 | LR: 1.00e-04 [2026-05-23 20:42:46] Epoch 2 | Step 6970 | Loss: 0.7182 | LM: 0.2170 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.389 | LR: 1.00e-04 [2026-05-23 20:42:59] Epoch 2 | Step 6980 | Loss: 0.7182 | LM: 0.2171 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.389 | LR: 1.00e-04 [2026-05-23 20:43:11] Epoch 2 | Step 6990 | Loss: 0.7181 | LM: 0.2168 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.389 | LR: 1.00e-04 [2026-05-23 20:43:23] Epoch 2 | Step 7000 | Loss: 0.7180 | LM: 0.2168 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.389 | LR: 1.00e-04 [2026-05-23 20:43:25] Validation | Batch 10/1463 | Loss: 0.7628 | LM_LOSS: 0.2629 | LB_LOSS: 0.9996 [2026-05-23 20:43:28] Validation | Batch 20/1463 | Loss: 0.7732 | LM_LOSS: 0.2731 | LB_LOSS: 1.0000 [2026-05-23 20:43:32] Validation | Batch 30/1463 | Loss: 0.7745 | LM_LOSS: 0.2744 | LB_LOSS: 1.0003 [2026-05-23 20:43:34] Validation | Batch 40/1463 | Loss: 0.7815 | LM_LOSS: 0.2814 | LB_LOSS: 1.0003 [2026-05-23 20:43:37] Validation | Batch 50/1463 | Loss: 0.7901 | LM_LOSS: 0.2900 | LB_LOSS: 1.0002 [2026-05-23 20:43:41] Validation | Batch 60/1463 | Loss: 0.7903 | LM_LOSS: 0.2901 | LB_LOSS: 1.0003 [2026-05-23 20:43:45] Validation | Batch 70/1463 | Loss: 0.7907 | LM_LOSS: 0.2906 | LB_LOSS: 1.0003 [2026-05-23 20:43:49] Validation | Batch 80/1463 | Loss: 0.7926 | LM_LOSS: 0.2925 | LB_LOSS: 1.0003 [2026-05-23 20:43:51] Validation | Batch 90/1463 | Loss: 0.7926 | LM_LOSS: 0.2924 | LB_LOSS: 1.0003 [2026-05-23 20:43:54] Validation | Batch 100/1463 | Loss: 0.7924 | LM_LOSS: 0.2922 | LB_LOSS: 1.0003 [2026-05-23 20:43:58] Validation | Batch 110/1463 | Loss: 0.7902 | LM_LOSS: 0.2900 | LB_LOSS: 1.0003 [2026-05-23 20:44:00] Validation | Batch 120/1463 | Loss: 0.7921 | LM_LOSS: 0.2920 | LB_LOSS: 1.0003 [2026-05-23 20:44:02] Validation | Batch 130/1463 | Loss: 0.7936 | LM_LOSS: 0.2934 | LB_LOSS: 1.0003 [2026-05-23 20:44:06] Validation | Batch 140/1463 | Loss: 0.7950 | LM_LOSS: 0.2948 | LB_LOSS: 1.0003 [2026-05-23 20:44:09] Validation | Batch 150/1463 | Loss: 0.7928 | LM_LOSS: 0.2926 | LB_LOSS: 1.0004 [2026-05-23 20:44:13] Validation | Batch 160/1463 | Loss: 0.7933 | LM_LOSS: 0.2930 | LB_LOSS: 1.0004 [2026-05-23 20:44:16] Validation | Batch 170/1463 | Loss: 0.7925 | LM_LOSS: 0.2923 | LB_LOSS: 1.0004 [2026-05-23 20:44:19] Validation | Batch 180/1463 | Loss: 0.7934 | LM_LOSS: 0.2933 | LB_LOSS: 1.0004 [2026-05-23 20:44:22] Validation | Batch 190/1463 | Loss: 0.7945 | LM_LOSS: 0.2943 | LB_LOSS: 1.0004 [2026-05-23 20:44:25] Validation | Batch 200/1463 | Loss: 0.7944 | LM_LOSS: 0.2942 | LB_LOSS: 1.0004 [2026-05-23 20:44:28] Validation | Batch 210/1463 | Loss: 0.7952 | LM_LOSS: 0.2950 | LB_LOSS: 1.0004 [2026-05-23 20:44:32] Validation | Batch 220/1463 | Loss: 0.7946 | LM_LOSS: 0.2944 | LB_LOSS: 1.0004 [2026-05-23 20:44:35] Validation | Batch 230/1463 | Loss: 0.7960 | LM_LOSS: 0.2958 | LB_LOSS: 1.0004 [2026-05-23 20:44:38] Validation | Batch 240/1463 | Loss: 0.7977 | LM_LOSS: 0.2975 | LB_LOSS: 1.0004 [2026-05-23 20:44:42] Validation | Batch 250/1463 | Loss: 0.7966 | LM_LOSS: 0.2964 | LB_LOSS: 1.0004 [2026-05-23 20:44:44] Validation | Batch 260/1463 | Loss: 0.7982 | LM_LOSS: 0.2980 | LB_LOSS: 1.0004 [2026-05-23 20:44:47] Validation | Batch 270/1463 | Loss: 0.7967 | LM_LOSS: 0.2965 | LB_LOSS: 1.0004 [2026-05-23 20:44:50] Validation | Batch 280/1463 | Loss: 0.7971 | LM_LOSS: 0.2969 | LB_LOSS: 1.0004 [2026-05-23 20:44:54] Validation | Batch 290/1463 | Loss: 0.7960 | LM_LOSS: 0.2958 | LB_LOSS: 1.0004 [2026-05-23 20:44:56] Validation | Batch 300/1463 | Loss: 0.7964 | LM_LOSS: 0.2962 | LB_LOSS: 1.0004 [2026-05-23 20:44:59] Validation | Batch 310/1463 | Loss: 0.7974 | LM_LOSS: 0.2972 | LB_LOSS: 1.0004 [2026-05-23 20:45:02] Validation | Batch 320/1463 | Loss: 0.7962 | LM_LOSS: 0.2960 | LB_LOSS: 1.0004 [2026-05-23 20:45:05] Validation | Batch 330/1463 | Loss: 0.7960 | LM_LOSS: 0.2957 | LB_LOSS: 1.0004 [2026-05-23 20:45:08] Validation | Batch 340/1463 | Loss: 0.7957 | LM_LOSS: 0.2955 | LB_LOSS: 1.0004 [2026-05-23 20:45:12] Validation | Batch 350/1463 | Loss: 0.7965 | LM_LOSS: 0.2963 | LB_LOSS: 1.0004 [2026-05-23 20:45:14] Validation | Batch 360/1463 | Loss: 0.7962 | LM_LOSS: 0.2960 | LB_LOSS: 1.0004 [2026-05-23 20:45:17] Validation | Batch 370/1463 | Loss: 0.7960 | LM_LOSS: 0.2958 | LB_LOSS: 1.0004 [2026-05-23 20:45:20] Validation | Batch 380/1463 | Loss: 0.7958 | LM_LOSS: 0.2956 | LB_LOSS: 1.0004 [2026-05-23 20:45:23] Validation | Batch 390/1463 | Loss: 0.7964 | LM_LOSS: 0.2962 | LB_LOSS: 1.0004 [2026-05-23 20:45:26] Validation | Batch 400/1463 | Loss: 0.7963 | LM_LOSS: 0.2961 | LB_LOSS: 1.0004 [2026-05-23 20:45:29] Validation | Batch 410/1463 | Loss: 0.7960 | LM_LOSS: 0.2958 | LB_LOSS: 1.0004 [2026-05-23 20:45:32] Validation | Batch 420/1463 | Loss: 0.7962 | LM_LOSS: 0.2960 | LB_LOSS: 1.0004 [2026-05-23 20:45:36] Validation | Batch 430/1463 | Loss: 0.7957 | LM_LOSS: 0.2955 | LB_LOSS: 1.0004 [2026-05-23 20:45:38] Validation | Batch 440/1463 | Loss: 0.7959 | LM_LOSS: 0.2957 | LB_LOSS: 1.0004 [2026-05-23 20:45:41] Validation | Batch 450/1463 | Loss: 0.7956 | LM_LOSS: 0.2954 | LB_LOSS: 1.0004 [2026-05-23 20:45:44] Validation | Batch 460/1463 | Loss: 0.7956 | LM_LOSS: 0.2954 | LB_LOSS: 1.0004 [2026-05-23 20:45:47] Validation | Batch 470/1463 | Loss: 0.7962 | LM_LOSS: 0.2960 | LB_LOSS: 1.0004 [2026-05-23 20:45:51] Validation | Batch 480/1463 | Loss: 0.7959 | LM_LOSS: 0.2957 | LB_LOSS: 1.0004 [2026-05-23 20:45:54] Validation | Batch 490/1463 | Loss: 0.7958 | LM_LOSS: 0.2956 | LB_LOSS: 1.0004 [2026-05-23 20:45:58] Validation | Batch 500/1463 | Loss: 0.7957 | LM_LOSS: 0.2955 | LB_LOSS: 1.0004 [2026-05-23 20:46:01] Validation | Batch 510/1463 | Loss: 0.7950 | LM_LOSS: 0.2948 | LB_LOSS: 1.0004 [2026-05-23 20:46:03] Validation | Batch 520/1463 | Loss: 0.7948 | LM_LOSS: 0.2946 | LB_LOSS: 1.0004 [2026-05-23 20:46:06] Validation | Batch 530/1463 | Loss: 0.7952 | LM_LOSS: 0.2950 | LB_LOSS: 1.0004 [2026-05-23 20:46:09] Validation | Batch 540/1463 | Loss: 0.7954 | LM_LOSS: 0.2952 | LB_LOSS: 1.0004 [2026-05-23 20:46:11] Validation | Batch 550/1463 | Loss: 0.7950 | LM_LOSS: 0.2948 | LB_LOSS: 1.0004 [2026-05-23 20:46:14] Validation | Batch 560/1463 | Loss: 0.7951 | LM_LOSS: 0.2949 | LB_LOSS: 1.0004 [2026-05-23 20:46:18] Validation | Batch 570/1463 | Loss: 0.7951 | LM_LOSS: 0.2948 | LB_LOSS: 1.0004 [2026-05-23 20:46:21] Validation | Batch 580/1463 | Loss: 0.7946 | LM_LOSS: 0.2944 | LB_LOSS: 1.0004 [2026-05-23 20:46:24] Validation | Batch 590/1463 | Loss: 0.7943 | LM_LOSS: 0.2941 | LB_LOSS: 1.0004 [2026-05-23 20:46:28] Validation | Batch 600/1463 | Loss: 0.7946 | LM_LOSS: 0.2944 | LB_LOSS: 1.0004 [2026-05-23 20:46:31] Validation | Batch 610/1463 | Loss: 0.7949 | LM_LOSS: 0.2947 | LB_LOSS: 1.0004 [2026-05-23 20:46:34] Validation | Batch 620/1463 | Loss: 0.7944 | LM_LOSS: 0.2941 | LB_LOSS: 1.0004 [2026-05-23 20:46:37] Validation | Batch 630/1463 | Loss: 0.7937 | LM_LOSS: 0.2935 | LB_LOSS: 1.0004 [2026-05-23 20:46:40] Validation | Batch 640/1463 | Loss: 0.7936 | LM_LOSS: 0.2933 | LB_LOSS: 1.0004 [2026-05-23 20:46:42] Validation | Batch 650/1463 | Loss: 0.7931 | LM_LOSS: 0.2929 | LB_LOSS: 1.0004 [2026-05-23 20:46:45] Validation | Batch 660/1463 | Loss: 0.7927 | LM_LOSS: 0.2925 | LB_LOSS: 1.0004 [2026-05-23 20:46:48] Validation | Batch 670/1463 | Loss: 0.7924 | LM_LOSS: 0.2921 | LB_LOSS: 1.0004 [2026-05-23 20:46:51] Validation | Batch 680/1463 | Loss: 0.7923 | LM_LOSS: 0.2921 | LB_LOSS: 1.0004 [2026-05-23 20:46:54] Validation | Batch 690/1463 | Loss: 0.7926 | LM_LOSS: 0.2924 | LB_LOSS: 1.0004 [2026-05-23 20:46:57] Validation | Batch 700/1463 | Loss: 0.7925 | LM_LOSS: 0.2923 | LB_LOSS: 1.0004 [2026-05-23 20:47:00] Validation | Batch 710/1463 | Loss: 0.7927 | LM_LOSS: 0.2924 | LB_LOSS: 1.0004 [2026-05-23 20:47:03] Validation | Batch 720/1463 | Loss: 0.7934 | LM_LOSS: 0.2932 | LB_LOSS: 1.0004 [2026-05-23 20:47:05] Validation | Batch 730/1463 | Loss: 0.7933 | LM_LOSS: 0.2931 | LB_LOSS: 1.0004 [2026-05-23 20:47:08] Validation | Batch 740/1463 | Loss: 0.7931 | LM_LOSS: 0.2928 | LB_LOSS: 1.0004 [2026-05-23 20:47:10] Validation | Batch 750/1463 | Loss: 0.7930 | LM_LOSS: 0.2928 | LB_LOSS: 1.0004 [2026-05-23 20:47:13] Validation | Batch 760/1463 | Loss: 0.7929 | LM_LOSS: 0.2927 | LB_LOSS: 1.0004 [2026-05-23 20:47:16] Validation | Batch 770/1463 | Loss: 0.7929 | LM_LOSS: 0.2927 | LB_LOSS: 1.0004 [2026-05-23 20:47:18] Validation | Batch 780/1463 | Loss: 0.7928 | LM_LOSS: 0.2926 | LB_LOSS: 1.0004 [2026-05-23 20:47:20] Validation | Batch 790/1463 | Loss: 0.7930 | LM_LOSS: 0.2928 | LB_LOSS: 1.0004 [2026-05-23 20:47:23] Validation | Batch 800/1463 | Loss: 0.7930 | LM_LOSS: 0.2928 | LB_LOSS: 1.0004 [2026-05-23 20:47:26] Validation | Batch 810/1463 | Loss: 0.7927 | LM_LOSS: 0.2925 | LB_LOSS: 1.0004 [2026-05-23 20:47:29] Validation | Batch 820/1463 | Loss: 0.7929 | LM_LOSS: 0.2926 | LB_LOSS: 1.0004 [2026-05-23 20:47:32] Validation | Batch 830/1463 | Loss: 0.7931 | LM_LOSS: 0.2929 | LB_LOSS: 1.0004 [2026-05-23 20:47:36] Validation | Batch 840/1463 | Loss: 0.7932 | LM_LOSS: 0.2930 | LB_LOSS: 1.0004 [2026-05-23 20:47:39] Validation | Batch 850/1463 | Loss: 0.7935 | LM_LOSS: 0.2933 | LB_LOSS: 1.0004 [2026-05-23 20:47:42] Validation | Batch 860/1463 | Loss: 0.7936 | LM_LOSS: 0.2934 | LB_LOSS: 1.0004 [2026-05-23 20:47:45] Validation | Batch 870/1463 | Loss: 0.7937 | LM_LOSS: 0.2935 | LB_LOSS: 1.0004 [2026-05-23 20:47:48] Validation | Batch 880/1463 | Loss: 0.7929 | LM_LOSS: 0.2927 | LB_LOSS: 1.0004 [2026-05-23 20:47:51] Validation | Batch 890/1463 | Loss: 0.7929 | LM_LOSS: 0.2927 | LB_LOSS: 1.0004 [2026-05-23 20:47:54] Validation | Batch 900/1463 | Loss: 0.7926 | LM_LOSS: 0.2924 | LB_LOSS: 1.0004 [2026-05-23 20:47:57] Validation | Batch 910/1463 | Loss: 0.7926 | LM_LOSS: 0.2924 | LB_LOSS: 1.0004 [2026-05-23 20:48:01] Validation | Batch 920/1463 | Loss: 0.7927 | LM_LOSS: 0.2925 | LB_LOSS: 1.0004 [2026-05-23 20:48:04] Validation | Batch 930/1463 | Loss: 0.7925 | LM_LOSS: 0.2923 | LB_LOSS: 1.0004 [2026-05-23 20:48:07] Validation | Batch 940/1463 | Loss: 0.7923 | LM_LOSS: 0.2921 | LB_LOSS: 1.0004 [2026-05-23 20:48:10] Validation | Batch 950/1463 | Loss: 0.7925 | LM_LOSS: 0.2923 | LB_LOSS: 1.0004 [2026-05-23 20:48:13] Validation | Batch 960/1463 | Loss: 0.7925 | LM_LOSS: 0.2923 | LB_LOSS: 1.0004 [2026-05-23 20:48:17] Validation | Batch 970/1463 | Loss: 0.7925 | LM_LOSS: 0.2923 | LB_LOSS: 1.0004 [2026-05-23 20:48:20] Validation | Batch 980/1463 | Loss: 0.7929 | LM_LOSS: 0.2926 | LB_LOSS: 1.0004 [2026-05-23 20:48:23] Validation | Batch 990/1463 | Loss: 0.7931 | LM_LOSS: 0.2929 | LB_LOSS: 1.0004 [2026-05-23 20:48:26] Validation | Batch 1000/1463 | Loss: 0.7936 | LM_LOSS: 0.2934 | LB_LOSS: 1.0004 [2026-05-23 20:48:29] Validation | Batch 1010/1463 | Loss: 0.7937 | LM_LOSS: 0.2935 | LB_LOSS: 1.0004 [2026-05-23 20:48:32] Validation | Batch 1020/1463 | Loss: 0.7937 | LM_LOSS: 0.2935 | LB_LOSS: 1.0004 [2026-05-23 20:48:35] Validation | Batch 1030/1463 | Loss: 0.7942 | LM_LOSS: 0.2939 | LB_LOSS: 1.0004 [2026-05-23 20:48:38] Validation | Batch 1040/1463 | Loss: 0.7938 | LM_LOSS: 0.2936 | LB_LOSS: 1.0004 [2026-05-23 20:48:41] Validation | Batch 1050/1463 | Loss: 0.7937 | LM_LOSS: 0.2935 | LB_LOSS: 1.0004 [2026-05-23 20:48:45] Validation | Batch 1060/1463 | Loss: 0.7937 | LM_LOSS: 0.2935 | LB_LOSS: 1.0004 [2026-05-23 20:48:48] Validation | Batch 1070/1463 | Loss: 0.7937 | LM_LOSS: 0.2935 | LB_LOSS: 1.0004 [2026-05-23 20:48:50] Validation | Batch 1080/1463 | Loss: 0.7936 | LM_LOSS: 0.2933 | LB_LOSS: 1.0004 [2026-05-23 20:48:53] Validation | Batch 1090/1463 | Loss: 0.7936 | LM_LOSS: 0.2934 | LB_LOSS: 1.0004 [2026-05-23 20:48:56] Validation | Batch 1100/1463 | Loss: 0.7934 | LM_LOSS: 0.2932 | LB_LOSS: 1.0004 [2026-05-23 20:48:58] Validation | Batch 1110/1463 | Loss: 0.7934 | LM_LOSS: 0.2932 | LB_LOSS: 1.0004 [2026-05-23 20:49:01] Validation | Batch 1120/1463 | Loss: 0.7931 | LM_LOSS: 0.2929 | LB_LOSS: 1.0004 [2026-05-23 20:49:05] Validation | Batch 1130/1463 | Loss: 0.7929 | LM_LOSS: 0.2927 | LB_LOSS: 1.0004 [2026-05-23 20:49:08] Validation | Batch 1140/1463 | Loss: 0.7931 | LM_LOSS: 0.2929 | LB_LOSS: 1.0004 [2026-05-23 20:49:12] Validation | Batch 1150/1463 | Loss: 0.7930 | LM_LOSS: 0.2928 | LB_LOSS: 1.0004 [2026-05-23 20:49:14] Validation | Batch 1160/1463 | Loss: 0.7929 | LM_LOSS: 0.2927 | LB_LOSS: 1.0004 [2026-05-23 20:49:18] Validation | Batch 1170/1463 | Loss: 0.7929 | LM_LOSS: 0.2927 | LB_LOSS: 1.0004 [2026-05-23 20:49:20] Validation | Batch 1180/1463 | Loss: 0.7930 | LM_LOSS: 0.2928 | LB_LOSS: 1.0004 [2026-05-23 20:49:24] Validation | Batch 1190/1463 | Loss: 0.7931 | LM_LOSS: 0.2929 | LB_LOSS: 1.0004 [2026-05-23 20:49:27] Validation | Batch 1200/1463 | Loss: 0.7931 | LM_LOSS: 0.2929 | LB_LOSS: 1.0004 [2026-05-23 20:49:30] Validation | Batch 1210/1463 | Loss: 0.7933 | LM_LOSS: 0.2931 | LB_LOSS: 1.0004 [2026-05-23 20:49:33] Validation | Batch 1220/1463 | Loss: 0.7936 | LM_LOSS: 0.2934 | LB_LOSS: 1.0004 [2026-05-23 20:49:35] Validation | Batch 1230/1463 | Loss: 0.7937 | LM_LOSS: 0.2934 | LB_LOSS: 1.0004 [2026-05-23 20:49:39] Validation | Batch 1240/1463 | Loss: 0.7938 | LM_LOSS: 0.2936 | LB_LOSS: 1.0004 [2026-05-23 20:49:42] Validation | Batch 1250/1463 | Loss: 0.7939 | LM_LOSS: 0.2937 | LB_LOSS: 1.0004 [2026-05-23 20:49:45] Validation | Batch 1260/1463 | Loss: 0.7938 | LM_LOSS: 0.2936 | LB_LOSS: 1.0004 [2026-05-23 20:49:48] Validation | Batch 1270/1463 | Loss: 0.7936 | LM_LOSS: 0.2934 | LB_LOSS: 1.0004 [2026-05-23 20:49:52] Validation | Batch 1280/1463 | Loss: 0.7935 | LM_LOSS: 0.2933 | LB_LOSS: 1.0004 [2026-05-23 20:49:54] Validation | Batch 1290/1463 | Loss: 0.7935 | LM_LOSS: 0.2933 | LB_LOSS: 1.0004 [2026-05-23 20:49:57] Validation | Batch 1300/1463 | Loss: 0.7932 | LM_LOSS: 0.2930 | LB_LOSS: 1.0004 [2026-05-23 20:50:01] Validation | Batch 1310/1463 | Loss: 0.7931 | LM_LOSS: 0.2929 | LB_LOSS: 1.0004 [2026-05-23 20:50:03] Validation | Batch 1320/1463 | Loss: 0.7932 | LM_LOSS: 0.2930 | LB_LOSS: 1.0004 [2026-05-23 20:50:05] Validation | Batch 1330/1463 | Loss: 0.7932 | LM_LOSS: 0.2930 | LB_LOSS: 1.0004 [2026-05-23 20:50:08] Validation | Batch 1340/1463 | Loss: 0.7935 | LM_LOSS: 0.2933 | LB_LOSS: 1.0004 [2026-05-23 20:50:11] Validation | Batch 1350/1463 | Loss: 0.7935 | LM_LOSS: 0.2933 | LB_LOSS: 1.0004 [2026-05-23 20:50:15] Validation | Batch 1360/1463 | Loss: 0.7938 | LM_LOSS: 0.2936 | LB_LOSS: 1.0004 [2026-05-23 20:50:18] Validation | Batch 1370/1463 | Loss: 0.7941 | LM_LOSS: 0.2939 | LB_LOSS: 1.0004 [2026-05-23 20:50:21] Validation | Batch 1380/1463 | Loss: 0.7938 | LM_LOSS: 0.2936 | LB_LOSS: 1.0004 [2026-05-23 20:50:24] Validation | Batch 1390/1463 | Loss: 0.7938 | LM_LOSS: 0.2936 | LB_LOSS: 1.0004 [2026-05-23 20:50:27] Validation | Batch 1400/1463 | Loss: 0.7941 | LM_LOSS: 0.2939 | LB_LOSS: 1.0004 [2026-05-23 20:50:29] Validation | Batch 1410/1463 | Loss: 0.7943 | LM_LOSS: 0.2941 | LB_LOSS: 1.0004 [2026-05-23 20:50:33] Validation | Batch 1420/1463 | Loss: 0.7943 | LM_LOSS: 0.2941 | LB_LOSS: 1.0004 [2026-05-23 20:50:36] Validation | Batch 1430/1463 | Loss: 0.7943 | LM_LOSS: 0.2941 | LB_LOSS: 1.0004 [2026-05-23 20:50:39] Validation | Batch 1440/1463 | Loss: 0.7947 | LM_LOSS: 0.2945 | LB_LOSS: 1.0004 [2026-05-23 20:50:42] Validation | Batch 1450/1463 | Loss: 0.7945 | LM_LOSS: 0.2943 | LB_LOSS: 1.0004 [2026-05-23 20:50:45] Validation | Batch 1460/1463 | Loss: 0.7948 | LM_LOSS: 0.2946 | LB_LOSS: 1.0004 [2026-05-23 20:50:47] Validation | Batch 1463/1463 | Loss: 0.7947 | LM_LOSS: 0.2945 | LB_LOSS: 1.0004 [2026-05-23 20:50:47] Validation | Loss: 0.7947 | LM_LOSS: 0.2945 | LB_LOSS: 1.0004 | PPL: 1.34 | Time: 443.52s [2026-05-23 20:51:00] Epoch 2 | Step 7010 | Loss: 0.7181 | LM: 0.2172 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.389 | LR: 1.00e-04 [2026-05-23 20:51:13] Epoch 2 | Step 7020 | Loss: 0.7181 | LM: 0.2174 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.389 | LR: 1.00e-04 [2026-05-23 20:51:26] Epoch 2 | Step 7030 | Loss: 0.7181 | LM: 0.2175 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.389 | LR: 1.00e-04 [2026-05-23 20:51:37] Epoch 2 | Step 7040 | Loss: 0.7181 | LM: 0.2174 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.389 | LR: 1.00e-04 [2026-05-23 20:51:50] Epoch 2 | Step 7050 | Loss: 0.7182 | LM: 0.2175 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.389 | LR: 1.00e-04 [2026-05-23 20:52:02] Epoch 2 | Step 7060 | Loss: 0.7181 | LM: 0.2174 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.389 | LR: 1.00e-04 [2026-05-23 20:52:15] Epoch 2 | Step 7070 | Loss: 0.7180 | LM: 0.2176 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.389 | LR: 1.00e-04 [2026-05-23 20:52:28] Epoch 2 | Step 7080 | Loss: 0.7181 | LM: 0.2176 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.389 | LR: 1.00e-04 [2026-05-23 20:52:39] Epoch 2 | Step 7090 | Loss: 0.7180 | LM: 0.2175 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.389 | LR: 1.00e-04 [2026-05-23 20:52:52] Epoch 2 | Step 7100 | Loss: 0.7181 | LM: 0.2175 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.389 | LR: 1.00e-04 [2026-05-23 20:53:04] Epoch 2 | Step 7110 | Loss: 0.7180 | LM: 0.2172 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.389 | LR: 1.00e-04 [2026-05-23 20:53:17] Epoch 2 | Step 7120 | Loss: 0.7179 | LM: 0.2170 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.389 | LR: 1.00e-04 [2026-05-23 20:53:29] Epoch 2 | Step 7130 | Loss: 0.7178 | LM: 0.2169 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.390 | LR: 1.00e-04 [2026-05-23 20:53:41] Epoch 2 | Step 7140 | Loss: 0.7179 | LM: 0.2169 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.389 | LR: 1.00e-04 [2026-05-23 20:53:53] Epoch 2 | Step 7150 | Loss: 0.7179 | LM: 0.2170 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.390 | LR: 1.00e-04 [2026-05-23 20:54:06] Epoch 2 | Step 7160 | Loss: 0.7180 | LM: 0.2169 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.390 | LR: 1.00e-04 [2026-05-23 20:54:18] Epoch 2 | Step 7170 | Loss: 0.7179 | LM: 0.2168 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.390 | LR: 1.00e-04 [2026-05-23 20:54:31] Epoch 2 | Step 7180 | Loss: 0.7179 | LM: 0.2167 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.390 | LR: 1.00e-04 [2026-05-23 20:54:44] Epoch 2 | Step 7190 | Loss: 0.7178 | LM: 0.2166 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.390 | LR: 1.00e-04 [2026-05-23 20:54:56] Epoch 2 | Step 7200 | Loss: 0.7178 | LM: 0.2166 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.390 | LR: 1.00e-04 [2026-05-23 20:55:07] Epoch 2 | Step 7210 | Loss: 0.7178 | LM: 0.2165 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.390 | LR: 1.00e-04 [2026-05-23 20:55:19] Epoch 2 | Step 7220 | Loss: 0.7178 | LM: 0.2168 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.390 | LR: 1.00e-04 [2026-05-23 20:55:32] Epoch 2 | Step 7230 | Loss: 0.7178 | LM: 0.2167 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.390 | LR: 1.00e-04 [2026-05-23 20:55:45] Epoch 2 | Step 7240 | Loss: 0.7178 | LM: 0.2168 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.390 | LR: 1.00e-04 [2026-05-23 20:55:57] Epoch 2 | Step 7250 | Loss: 0.7178 | LM: 0.2168 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.390 | LR: 1.00e-04 [2026-05-23 20:56:11] Epoch 2 | Step 7260 | Loss: 0.7179 | LM: 0.2170 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.390 | LR: 1.00e-04 [2026-05-23 20:56:23] Epoch 2 | Step 7270 | Loss: 0.7179 | LM: 0.2170 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.390 | LR: 1.00e-04 [2026-05-23 20:56:35] Epoch 2 | Step 7280 | Loss: 0.7179 | LM: 0.2171 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.390 | LR: 1.00e-04 [2026-05-23 20:56:47] Epoch 2 | Step 7290 | Loss: 0.7179 | LM: 0.2172 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.390 | LR: 1.00e-04 [2026-05-23 20:57:00] Epoch 2 | Step 7300 | Loss: 0.7180 | LM: 0.2173 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.390 | LR: 1.00e-04 [2026-05-23 20:57:12] Epoch 2 | Step 7310 | Loss: 0.7180 | LM: 0.2170 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.390 | LR: 1.00e-04 [2026-05-23 20:57:24] Epoch 2 | Step 7320 | Loss: 0.7179 | LM: 0.2170 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.390 | LR: 1.00e-04 [2026-05-23 20:57:37] Epoch 2 | Step 7330 | Loss: 0.7178 | LM: 0.2169 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.390 | LR: 1.00e-04 [2026-05-23 20:57:49] Epoch 2 | Step 7340 | Loss: 0.7178 | LM: 0.2169 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.390 | LR: 1.00e-04 [2026-05-23 20:58:01] Epoch 2 | Step 7350 | Loss: 0.7177 | LM: 0.2170 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.390 | LR: 1.00e-04 [2026-05-23 20:58:13] Epoch 2 | Step 7360 | Loss: 0.7177 | LM: 0.2171 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.390 | LR: 1.00e-04 [2026-05-23 20:58:25] Epoch 2 | Step 7370 | Loss: 0.7177 | LM: 0.2170 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.390 | LR: 1.00e-04 [2026-05-23 20:58:37] Epoch 2 | Step 7380 | Loss: 0.7177 | LM: 0.2171 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.390 | LR: 1.00e-04 [2026-05-23 20:58:51] Epoch 2 | Step 7390 | Loss: 0.7178 | LM: 0.2173 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.390 | LR: 1.00e-04 [2026-05-23 20:59:04] Epoch 2 | Step 7400 | Loss: 0.7178 | LM: 0.2172 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.390 | LR: 1.00e-04 [2026-05-23 20:59:18] Epoch 2 | Step 7410 | Loss: 0.7178 | LM: 0.2171 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.390 | LR: 1.00e-04 [2026-05-23 20:59:31] Epoch 2 | Step 7420 | Loss: 0.7177 | LM: 0.2170 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.390 | LR: 1.00e-04 [2026-05-23 20:59:43] Epoch 2 | Step 7430 | Loss: 0.7177 | LM: 0.2171 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.390 | LR: 1.00e-04 [2026-05-23 20:59:55] Epoch 2 | Step 7440 | Loss: 0.7177 | LM: 0.2170 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.390 | LR: 1.00e-04 [2026-05-23 21:00:08] Epoch 2 | Step 7450 | Loss: 0.7177 | LM: 0.2169 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.390 | LR: 1.00e-04 [2026-05-23 21:00:21] Epoch 2 | Step 7460 | Loss: 0.7177 | LM: 0.2170 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.390 | LR: 1.00e-04 [2026-05-23 21:00:33] Epoch 2 | Step 7470 | Loss: 0.7177 | LM: 0.2170 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.390 | LR: 1.00e-04 [2026-05-23 21:00:46] Epoch 2 | Step 7480 | Loss: 0.7177 | LM: 0.2171 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.390 | LR: 1.00e-04 [2026-05-23 21:00:59] Epoch 2 | Step 7490 | Loss: 0.7177 | LM: 0.2172 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.390 | LR: 1.00e-04 [2026-05-23 21:01:12] Epoch 2 | Step 7500 | Loss: 0.7177 | LM: 0.2172 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.390 | LR: 1.00e-04 [2026-05-23 21:01:24] Epoch 2 | Step 7510 | Loss: 0.7177 | LM: 0.2170 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.390 | LR: 1.00e-04 [2026-05-23 21:01:37] Epoch 2 | Step 7520 | Loss: 0.7177 | LM: 0.2170 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.390 | LR: 1.00e-04 [2026-05-23 21:01:49] Epoch 2 | Step 7530 | Loss: 0.7177 | LM: 0.2170 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.390 | LR: 1.00e-04 [2026-05-23 21:02:03] Epoch 2 | Step 7540 | Loss: 0.7176 | LM: 0.2170 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.390 | LR: 1.00e-04 [2026-05-23 21:02:17] Epoch 2 | Step 7550 | Loss: 0.7177 | LM: 0.2172 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.390 | LR: 1.00e-04 [2026-05-23 21:02:29] Epoch 2 | Step 7560 | Loss: 0.7177 | LM: 0.2172 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.390 | LR: 1.00e-04 [2026-05-23 21:02:41] Epoch 2 | Step 7570 | Loss: 0.7176 | LM: 0.2171 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.390 | LR: 1.00e-04 [2026-05-23 21:02:53] Epoch 2 | Step 7580 | Loss: 0.7176 | LM: 0.2172 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.390 | LR: 1.00e-04 [2026-05-23 21:03:05] Epoch 2 | Step 7590 | Loss: 0.7177 | LM: 0.2175 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.390 | LR: 1.00e-04 [2026-05-23 21:03:16] Epoch 2 | Step 7600 | Loss: 0.7177 | LM: 0.2174 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.390 | LR: 1.00e-04 [2026-05-23 21:03:29] Epoch 2 | Step 7610 | Loss: 0.7177 | LM: 0.2174 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.390 | LR: 1.00e-04 [2026-05-23 21:03:41] Epoch 2 | Step 7620 | Loss: 0.7176 | LM: 0.2174 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.390 | LR: 1.00e-04 [2026-05-23 21:03:53] Epoch 2 | Step 7630 | Loss: 0.7176 | LM: 0.2173 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.390 | LR: 1.00e-04 [2026-05-23 21:04:05] Epoch 2 | Step 7640 | Loss: 0.7176 | LM: 0.2174 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.390 | LR: 1.00e-04 [2026-05-23 21:04:17] Epoch 2 | Step 7650 | Loss: 0.7177 | LM: 0.2175 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.390 | LR: 1.00e-04 [2026-05-23 21:04:29] Epoch 2 | Step 7660 | Loss: 0.7176 | LM: 0.2174 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.390 | LR: 1.00e-04 [2026-05-23 21:04:42] Epoch 2 | Step 7670 | Loss: 0.7175 | LM: 0.2172 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.390 | LR: 1.00e-04 [2026-05-23 21:04:54] Epoch 2 | Step 7680 | Loss: 0.7175 | LM: 0.2173 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.390 | LR: 1.00e-04 [2026-05-23 21:05:05] Epoch 2 | Step 7690 | Loss: 0.7175 | LM: 0.2173 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.390 | LR: 1.00e-04 [2026-05-23 21:05:18] Epoch 2 | Step 7700 | Loss: 0.7175 | LM: 0.2172 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.390 | LR: 1.00e-04 [2026-05-23 21:05:29] Epoch 2 | Step 7710 | Loss: 0.7175 | LM: 0.2173 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.390 | LR: 1.00e-04 [2026-05-23 21:05:42] Epoch 2 | Step 7720 | Loss: 0.7175 | LM: 0.2175 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.390 | LR: 1.00e-04 [2026-05-23 21:05:56] Epoch 2 | Step 7730 | Loss: 0.7176 | LM: 0.2173 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.390 | LR: 1.00e-04 [2026-05-23 21:06:07] Epoch 2 | Step 7740 | Loss: 0.7177 | LM: 0.2175 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.390 | LR: 1.00e-04 [2026-05-23 21:06:20] Epoch 2 | Step 7750 | Loss: 0.7177 | LM: 0.2177 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.390 | LR: 1.00e-04 [2026-05-23 21:06:31] Epoch 2 | Step 7760 | Loss: 0.7178 | LM: 0.2178 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.390 | LR: 1.00e-04 [2026-05-23 21:06:43] Epoch 2 | Step 7770 | Loss: 0.7178 | LM: 0.2178 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.390 | LR: 1.00e-04 [2026-05-23 21:06:55] Epoch 2 | Step 7780 | Loss: 0.7178 | LM: 0.2178 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.390 | LR: 1.00e-04 [2026-05-23 21:07:07] Epoch 2 | Step 7790 | Loss: 0.7179 | LM: 0.2180 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.390 | LR: 1.00e-04 [2026-05-23 21:07:18] Epoch 2 | Step 7800 | Loss: 0.7179 | LM: 0.2180 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.390 | LR: 1.00e-04 [2026-05-23 21:07:29] Epoch 2 | Step 7810 | Loss: 0.7179 | LM: 0.2180 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.390 | LR: 1.00e-04 [2026-05-23 21:07:42] Epoch 2 | Step 7820 | Loss: 0.7180 | LM: 0.2180 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.390 | LR: 1.00e-04 [2026-05-23 21:07:55] Epoch 2 | Step 7830 | Loss: 0.7180 | LM: 0.2181 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.390 | LR: 1.00e-04 [2026-05-23 21:08:07] Epoch 2 | Step 7840 | Loss: 0.7181 | LM: 0.2181 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.390 | LR: 1.00e-04 [2026-05-23 21:08:18] Epoch 2 | Step 7850 | Loss: 0.7181 | LM: 0.2182 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.390 | LR: 1.00e-04 [2026-05-23 21:08:31] Epoch 2 | Step 7860 | Loss: 0.7180 | LM: 0.2180 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.390 | LR: 1.00e-04 [2026-05-23 21:08:43] Epoch 2 | Step 7870 | Loss: 0.7180 | LM: 0.2181 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.390 | LR: 1.00e-04 [2026-05-23 21:08:55] Epoch 2 | Step 7880 | Loss: 0.7180 | LM: 0.2182 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.390 | LR: 1.00e-04 [2026-05-23 21:09:06] Epoch 2 | Step 7890 | Loss: 0.7179 | LM: 0.2180 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.390 | LR: 1.00e-04 [2026-05-23 21:09:18] Epoch 2 | Step 7900 | Loss: 0.7179 | LM: 0.2180 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.390 | LR: 1.00e-04 [2026-05-23 21:09:31] Epoch 2 | Step 7910 | Loss: 0.7179 | LM: 0.2180 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.390 | LR: 1.00e-04 [2026-05-23 21:09:43] Epoch 2 | Step 7920 | Loss: 0.7178 | LM: 0.2181 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.390 | LR: 1.00e-04 [2026-05-23 21:09:54] Epoch 2 | Step 7930 | Loss: 0.7178 | LM: 0.2179 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.390 | LR: 1.00e-04 [2026-05-23 21:10:07] Epoch 2 | Step 7940 | Loss: 0.7177 | LM: 0.2179 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.390 | LR: 1.00e-04 [2026-05-23 21:10:19] Epoch 2 | Step 7950 | Loss: 0.7177 | LM: 0.2179 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.390 | LR: 1.00e-04 [2026-05-23 21:10:31] Epoch 2 | Step 7960 | Loss: 0.7178 | LM: 0.2181 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.390 | LR: 1.00e-04 [2026-05-23 21:10:43] Epoch 2 | Step 7970 | Loss: 0.7179 | LM: 0.2181 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.390 | LR: 1.00e-04 [2026-05-23 21:10:55] Epoch 2 | Step 7980 | Loss: 0.7179 | LM: 0.2181 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.390 | LR: 1.00e-04 [2026-05-23 21:11:08] Epoch 2 | Step 7990 | Loss: 0.7178 | LM: 0.2179 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.390 | LR: 1.00e-04 [2026-05-23 21:11:20] Epoch 2 | Step 8000 | Loss: 0.7178 | LM: 0.2179 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.390 | LR: 1.00e-04 [2026-05-23 21:11:22] Validation | Batch 10/1463 | Loss: 0.7619 | LM_LOSS: 0.2621 | LB_LOSS: 0.9996 [2026-05-23 21:11:25] Validation | Batch 20/1463 | Loss: 0.7713 | LM_LOSS: 0.2713 | LB_LOSS: 1.0000 [2026-05-23 21:11:28] Validation | Batch 30/1463 | Loss: 0.7728 | LM_LOSS: 0.2726 | LB_LOSS: 1.0003 [2026-05-23 21:11:31] Validation | Batch 40/1463 | Loss: 0.7794 | LM_LOSS: 0.2793 | LB_LOSS: 1.0003 [2026-05-23 21:11:34] Validation | Batch 50/1463 | Loss: 0.7877 | LM_LOSS: 0.2876 | LB_LOSS: 1.0002 [2026-05-23 21:11:38] Validation | Batch 60/1463 | Loss: 0.7875 | LM_LOSS: 0.2874 | LB_LOSS: 1.0003 [2026-05-23 21:11:42] Validation | Batch 70/1463 | Loss: 0.7880 | LM_LOSS: 0.2878 | LB_LOSS: 1.0003 [2026-05-23 21:11:45] Validation | Batch 80/1463 | Loss: 0.7902 | LM_LOSS: 0.2901 | LB_LOSS: 1.0003 [2026-05-23 21:11:48] Validation | Batch 90/1463 | Loss: 0.7903 | LM_LOSS: 0.2902 | LB_LOSS: 1.0003 [2026-05-23 21:11:51] Validation | Batch 100/1463 | Loss: 0.7904 | LM_LOSS: 0.2903 | LB_LOSS: 1.0003 [2026-05-23 21:11:54] Validation | Batch 110/1463 | Loss: 0.7884 | LM_LOSS: 0.2882 | LB_LOSS: 1.0003 [2026-05-23 21:11:57] Validation | Batch 120/1463 | Loss: 0.7903 | LM_LOSS: 0.2902 | LB_LOSS: 1.0003 [2026-05-23 21:11:59] Validation | Batch 130/1463 | Loss: 0.7917 | LM_LOSS: 0.2915 | LB_LOSS: 1.0003 [2026-05-23 21:12:03] Validation | Batch 140/1463 | Loss: 0.7931 | LM_LOSS: 0.2929 | LB_LOSS: 1.0003 [2026-05-23 21:12:06] Validation | Batch 150/1463 | Loss: 0.7910 | LM_LOSS: 0.2908 | LB_LOSS: 1.0004 [2026-05-23 21:12:09] Validation | Batch 160/1463 | Loss: 0.7915 | LM_LOSS: 0.2913 | LB_LOSS: 1.0004 [2026-05-23 21:12:13] Validation | Batch 170/1463 | Loss: 0.7907 | LM_LOSS: 0.2905 | LB_LOSS: 1.0004 [2026-05-23 21:12:16] Validation | Batch 180/1463 | Loss: 0.7916 | LM_LOSS: 0.2914 | LB_LOSS: 1.0004 [2026-05-23 21:12:19] Validation | Batch 190/1463 | Loss: 0.7928 | LM_LOSS: 0.2926 | LB_LOSS: 1.0004 [2026-05-23 21:12:22] Validation | Batch 200/1463 | Loss: 0.7928 | LM_LOSS: 0.2927 | LB_LOSS: 1.0004 [2026-05-23 21:12:25] Validation | Batch 210/1463 | Loss: 0.7938 | LM_LOSS: 0.2936 | LB_LOSS: 1.0004 [2026-05-23 21:12:28] Validation | Batch 220/1463 | Loss: 0.7932 | LM_LOSS: 0.2930 | LB_LOSS: 1.0004 [2026-05-23 21:12:32] Validation | Batch 230/1463 | Loss: 0.7945 | LM_LOSS: 0.2943 | LB_LOSS: 1.0004 [2026-05-23 21:12:35] Validation | Batch 240/1463 | Loss: 0.7961 | LM_LOSS: 0.2959 | LB_LOSS: 1.0004 [2026-05-23 21:12:38] Validation | Batch 250/1463 | Loss: 0.7951 | LM_LOSS: 0.2949 | LB_LOSS: 1.0004 [2026-05-23 21:12:41] Validation | Batch 260/1463 | Loss: 0.7968 | LM_LOSS: 0.2966 | LB_LOSS: 1.0004 [2026-05-23 21:12:44] Validation | Batch 270/1463 | Loss: 0.7953 | LM_LOSS: 0.2951 | LB_LOSS: 1.0004 [2026-05-23 21:12:47] Validation | Batch 280/1463 | Loss: 0.7957 | LM_LOSS: 0.2955 | LB_LOSS: 1.0004 [2026-05-23 21:12:50] Validation | Batch 290/1463 | Loss: 0.7946 | LM_LOSS: 0.2944 | LB_LOSS: 1.0004 [2026-05-23 21:12:53] Validation | Batch 300/1463 | Loss: 0.7950 | LM_LOSS: 0.2948 | LB_LOSS: 1.0004 [2026-05-23 21:12:56] Validation | Batch 310/1463 | Loss: 0.7960 | LM_LOSS: 0.2957 | LB_LOSS: 1.0004 [2026-05-23 21:12:59] Validation | Batch 320/1463 | Loss: 0.7947 | LM_LOSS: 0.2946 | LB_LOSS: 1.0004 [2026-05-23 21:13:02] Validation | Batch 330/1463 | Loss: 0.7945 | LM_LOSS: 0.2943 | LB_LOSS: 1.0004 [2026-05-23 21:13:05] Validation | Batch 340/1463 | Loss: 0.7942 | LM_LOSS: 0.2940 | LB_LOSS: 1.0004 [2026-05-23 21:13:09] Validation | Batch 350/1463 | Loss: 0.7950 | LM_LOSS: 0.2948 | LB_LOSS: 1.0004 [2026-05-23 21:13:11] Validation | Batch 360/1463 | Loss: 0.7947 | LM_LOSS: 0.2945 | LB_LOSS: 1.0004 [2026-05-23 21:13:14] Validation | Batch 370/1463 | Loss: 0.7945 | LM_LOSS: 0.2943 | LB_LOSS: 1.0004 [2026-05-23 21:13:17] Validation | Batch 380/1463 | Loss: 0.7943 | LM_LOSS: 0.2941 | LB_LOSS: 1.0004 [2026-05-23 21:13:20] Validation | Batch 390/1463 | Loss: 0.7950 | LM_LOSS: 0.2948 | LB_LOSS: 1.0004 [2026-05-23 21:13:23] Validation | Batch 400/1463 | Loss: 0.7949 | LM_LOSS: 0.2947 | LB_LOSS: 1.0004 [2026-05-23 21:13:26] Validation | Batch 410/1463 | Loss: 0.7945 | LM_LOSS: 0.2943 | LB_LOSS: 1.0004 [2026-05-23 21:13:29] Validation | Batch 420/1463 | Loss: 0.7946 | LM_LOSS: 0.2944 | LB_LOSS: 1.0004 [2026-05-23 21:13:33] Validation | Batch 430/1463 | Loss: 0.7942 | LM_LOSS: 0.2940 | LB_LOSS: 1.0004 [2026-05-23 21:13:35] Validation | Batch 440/1463 | Loss: 0.7943 | LM_LOSS: 0.2941 | LB_LOSS: 1.0004 [2026-05-23 21:13:38] Validation | Batch 450/1463 | Loss: 0.7940 | LM_LOSS: 0.2938 | LB_LOSS: 1.0004 [2026-05-23 21:13:41] Validation | Batch 460/1463 | Loss: 0.7940 | LM_LOSS: 0.2937 | LB_LOSS: 1.0004 [2026-05-23 21:13:44] Validation | Batch 470/1463 | Loss: 0.7946 | LM_LOSS: 0.2944 | LB_LOSS: 1.0004 [2026-05-23 21:13:48] Validation | Batch 480/1463 | Loss: 0.7944 | LM_LOSS: 0.2942 | LB_LOSS: 1.0004 [2026-05-23 21:13:51] Validation | Batch 490/1463 | Loss: 0.7943 | LM_LOSS: 0.2941 | LB_LOSS: 1.0004 [2026-05-23 21:13:55] Validation | Batch 500/1463 | Loss: 0.7942 | LM_LOSS: 0.2940 | LB_LOSS: 1.0004 [2026-05-23 21:13:58] Validation | Batch 510/1463 | Loss: 0.7935 | LM_LOSS: 0.2933 | LB_LOSS: 1.0004 [2026-05-23 21:14:00] Validation | Batch 520/1463 | Loss: 0.7932 | LM_LOSS: 0.2930 | LB_LOSS: 1.0004 [2026-05-23 21:14:03] Validation | Batch 530/1463 | Loss: 0.7937 | LM_LOSS: 0.2935 | LB_LOSS: 1.0004 [2026-05-23 21:14:06] Validation | Batch 540/1463 | Loss: 0.7939 | LM_LOSS: 0.2937 | LB_LOSS: 1.0004 [2026-05-23 21:14:08] Validation | Batch 550/1463 | Loss: 0.7935 | LM_LOSS: 0.2933 | LB_LOSS: 1.0004 [2026-05-23 21:14:11] Validation | Batch 560/1463 | Loss: 0.7936 | LM_LOSS: 0.2934 | LB_LOSS: 1.0004 [2026-05-23 21:14:14] Validation | Batch 570/1463 | Loss: 0.7936 | LM_LOSS: 0.2934 | LB_LOSS: 1.0004 [2026-05-23 21:14:18] Validation | Batch 580/1463 | Loss: 0.7931 | LM_LOSS: 0.2929 | LB_LOSS: 1.0004 [2026-05-23 21:14:21] Validation | Batch 590/1463 | Loss: 0.7928 | LM_LOSS: 0.2926 | LB_LOSS: 1.0004 [2026-05-23 21:14:24] Validation | Batch 600/1463 | Loss: 0.7931 | LM_LOSS: 0.2929 | LB_LOSS: 1.0004 [2026-05-23 21:14:28] Validation | Batch 610/1463 | Loss: 0.7934 | LM_LOSS: 0.2931 | LB_LOSS: 1.0004 [2026-05-23 21:14:31] Validation | Batch 620/1463 | Loss: 0.7928 | LM_LOSS: 0.2926 | LB_LOSS: 1.0004 [2026-05-23 21:14:34] Validation | Batch 630/1463 | Loss: 0.7922 | LM_LOSS: 0.2920 | LB_LOSS: 1.0004 [2026-05-23 21:14:37] Validation | Batch 640/1463 | Loss: 0.7920 | LM_LOSS: 0.2918 | LB_LOSS: 1.0004 [2026-05-23 21:14:39] Validation | Batch 650/1463 | Loss: 0.7916 | LM_LOSS: 0.2914 | LB_LOSS: 1.0004 [2026-05-23 21:14:42] Validation | Batch 660/1463 | Loss: 0.7912 | LM_LOSS: 0.2910 | LB_LOSS: 1.0004 [2026-05-23 21:14:45] Validation | Batch 670/1463 | Loss: 0.7908 | LM_LOSS: 0.2906 | LB_LOSS: 1.0004 [2026-05-23 21:14:47] Validation | Batch 680/1463 | Loss: 0.7908 | LM_LOSS: 0.2906 | LB_LOSS: 1.0004 [2026-05-23 21:14:51] Validation | Batch 690/1463 | Loss: 0.7910 | LM_LOSS: 0.2908 | LB_LOSS: 1.0004 [2026-05-23 21:14:54] Validation | Batch 700/1463 | Loss: 0.7909 | LM_LOSS: 0.2907 | LB_LOSS: 1.0004 [2026-05-23 21:14:56] Validation | Batch 710/1463 | Loss: 0.7911 | LM_LOSS: 0.2909 | LB_LOSS: 1.0004 [2026-05-23 21:15:00] Validation | Batch 720/1463 | Loss: 0.7919 | LM_LOSS: 0.2917 | LB_LOSS: 1.0004 [2026-05-23 21:15:02] Validation | Batch 730/1463 | Loss: 0.7918 | LM_LOSS: 0.2916 | LB_LOSS: 1.0004 [2026-05-23 21:15:05] Validation | Batch 740/1463 | Loss: 0.7915 | LM_LOSS: 0.2913 | LB_LOSS: 1.0004 [2026-05-23 21:15:07] Validation | Batch 750/1463 | Loss: 0.7915 | LM_LOSS: 0.2913 | LB_LOSS: 1.0004 [2026-05-23 21:15:10] Validation | Batch 760/1463 | Loss: 0.7913 | LM_LOSS: 0.2911 | LB_LOSS: 1.0004 [2026-05-23 21:15:12] Validation | Batch 770/1463 | Loss: 0.7913 | LM_LOSS: 0.2911 | LB_LOSS: 1.0004 [2026-05-23 21:15:15] Validation | Batch 780/1463 | Loss: 0.7913 | LM_LOSS: 0.2911 | LB_LOSS: 1.0004 [2026-05-23 21:15:17] Validation | Batch 790/1463 | Loss: 0.7915 | LM_LOSS: 0.2913 | LB_LOSS: 1.0004 [2026-05-23 21:15:20] Validation | Batch 800/1463 | Loss: 0.7914 | LM_LOSS: 0.2912 | LB_LOSS: 1.0004 [2026-05-23 21:15:23] Validation | Batch 810/1463 | Loss: 0.7911 | LM_LOSS: 0.2909 | LB_LOSS: 1.0004 [2026-05-23 21:15:26] Validation | Batch 820/1463 | Loss: 0.7912 | LM_LOSS: 0.2910 | LB_LOSS: 1.0004 [2026-05-23 21:15:29] Validation | Batch 830/1463 | Loss: 0.7914 | LM_LOSS: 0.2912 | LB_LOSS: 1.0004 [2026-05-23 21:15:33] Validation | Batch 840/1463 | Loss: 0.7915 | LM_LOSS: 0.2913 | LB_LOSS: 1.0004 [2026-05-23 21:15:36] Validation | Batch 850/1463 | Loss: 0.7918 | LM_LOSS: 0.2916 | LB_LOSS: 1.0004 [2026-05-23 21:15:39] Validation | Batch 860/1463 | Loss: 0.7920 | LM_LOSS: 0.2918 | LB_LOSS: 1.0004 [2026-05-23 21:15:42] Validation | Batch 870/1463 | Loss: 0.7920 | LM_LOSS: 0.2918 | LB_LOSS: 1.0004 [2026-05-23 21:15:45] Validation | Batch 880/1463 | Loss: 0.7913 | LM_LOSS: 0.2911 | LB_LOSS: 1.0004 [2026-05-23 21:15:48] Validation | Batch 890/1463 | Loss: 0.7913 | LM_LOSS: 0.2911 | LB_LOSS: 1.0004 [2026-05-23 21:15:51] Validation | Batch 900/1463 | Loss: 0.7910 | LM_LOSS: 0.2908 | LB_LOSS: 1.0004 [2026-05-23 21:15:54] Validation | Batch 910/1463 | Loss: 0.7910 | LM_LOSS: 0.2908 | LB_LOSS: 1.0004 [2026-05-23 21:15:58] Validation | Batch 920/1463 | Loss: 0.7912 | LM_LOSS: 0.2910 | LB_LOSS: 1.0004 [2026-05-23 21:16:00] Validation | Batch 930/1463 | Loss: 0.7909 | LM_LOSS: 0.2907 | LB_LOSS: 1.0004 [2026-05-23 21:16:04] Validation | Batch 940/1463 | Loss: 0.7907 | LM_LOSS: 0.2905 | LB_LOSS: 1.0004 [2026-05-23 21:16:07] Validation | Batch 950/1463 | Loss: 0.7909 | LM_LOSS: 0.2907 | LB_LOSS: 1.0004 [2026-05-23 21:16:10] Validation | Batch 960/1463 | Loss: 0.7910 | LM_LOSS: 0.2908 | LB_LOSS: 1.0004 [2026-05-23 21:16:14] Validation | Batch 970/1463 | Loss: 0.7910 | LM_LOSS: 0.2908 | LB_LOSS: 1.0004 [2026-05-23 21:16:17] Validation | Batch 980/1463 | Loss: 0.7913 | LM_LOSS: 0.2911 | LB_LOSS: 1.0004 [2026-05-23 21:16:20] Validation | Batch 990/1463 | Loss: 0.7915 | LM_LOSS: 0.2913 | LB_LOSS: 1.0004 [2026-05-23 21:16:23] Validation | Batch 1000/1463 | Loss: 0.7921 | LM_LOSS: 0.2919 | LB_LOSS: 1.0004 [2026-05-23 21:16:26] Validation | Batch 1010/1463 | Loss: 0.7921 | LM_LOSS: 0.2919 | LB_LOSS: 1.0004 [2026-05-23 21:16:29] Validation | Batch 1020/1463 | Loss: 0.7922 | LM_LOSS: 0.2920 | LB_LOSS: 1.0004 [2026-05-23 21:16:32] Validation | Batch 1030/1463 | Loss: 0.7926 | LM_LOSS: 0.2924 | LB_LOSS: 1.0004 [2026-05-23 21:16:35] Validation | Batch 1040/1463 | Loss: 0.7923 | LM_LOSS: 0.2921 | LB_LOSS: 1.0004 [2026-05-23 21:16:38] Validation | Batch 1050/1463 | Loss: 0.7922 | LM_LOSS: 0.2920 | LB_LOSS: 1.0004 [2026-05-23 21:16:41] Validation | Batch 1060/1463 | Loss: 0.7922 | LM_LOSS: 0.2920 | LB_LOSS: 1.0004 [2026-05-23 21:16:44] Validation | Batch 1070/1463 | Loss: 0.7922 | LM_LOSS: 0.2920 | LB_LOSS: 1.0004 [2026-05-23 21:16:47] Validation | Batch 1080/1463 | Loss: 0.7920 | LM_LOSS: 0.2918 | LB_LOSS: 1.0004 [2026-05-23 21:16:50] Validation | Batch 1090/1463 | Loss: 0.7920 | LM_LOSS: 0.2918 | LB_LOSS: 1.0004 [2026-05-23 21:16:53] Validation | Batch 1100/1463 | Loss: 0.7918 | LM_LOSS: 0.2916 | LB_LOSS: 1.0004 [2026-05-23 21:16:55] Validation | Batch 1110/1463 | Loss: 0.7918 | LM_LOSS: 0.2916 | LB_LOSS: 1.0004 [2026-05-23 21:16:58] Validation | Batch 1120/1463 | Loss: 0.7916 | LM_LOSS: 0.2914 | LB_LOSS: 1.0004 [2026-05-23 21:17:02] Validation | Batch 1130/1463 | Loss: 0.7914 | LM_LOSS: 0.2912 | LB_LOSS: 1.0004 [2026-05-23 21:17:05] Validation | Batch 1140/1463 | Loss: 0.7916 | LM_LOSS: 0.2914 | LB_LOSS: 1.0004 [2026-05-23 21:17:08] Validation | Batch 1150/1463 | Loss: 0.7915 | LM_LOSS: 0.2913 | LB_LOSS: 1.0004 [2026-05-23 21:17:11] Validation | Batch 1160/1463 | Loss: 0.7914 | LM_LOSS: 0.2912 | LB_LOSS: 1.0004 [2026-05-23 21:17:14] Validation | Batch 1170/1463 | Loss: 0.7914 | LM_LOSS: 0.2912 | LB_LOSS: 1.0004 [2026-05-23 21:17:17] Validation | Batch 1180/1463 | Loss: 0.7915 | LM_LOSS: 0.2913 | LB_LOSS: 1.0004 [2026-05-23 21:17:21] Validation | Batch 1190/1463 | Loss: 0.7917 | LM_LOSS: 0.2915 | LB_LOSS: 1.0004 [2026-05-23 21:17:24] Validation | Batch 1200/1463 | Loss: 0.7916 | LM_LOSS: 0.2914 | LB_LOSS: 1.0004 [2026-05-23 21:17:27] Validation | Batch 1210/1463 | Loss: 0.7919 | LM_LOSS: 0.2917 | LB_LOSS: 1.0004 [2026-05-23 21:17:30] Validation | Batch 1220/1463 | Loss: 0.7922 | LM_LOSS: 0.2920 | LB_LOSS: 1.0004 [2026-05-23 21:17:32] Validation | Batch 1230/1463 | Loss: 0.7922 | LM_LOSS: 0.2920 | LB_LOSS: 1.0004 [2026-05-23 21:17:35] Validation | Batch 1240/1463 | Loss: 0.7924 | LM_LOSS: 0.2922 | LB_LOSS: 1.0004 [2026-05-23 21:17:39] Validation | Batch 1250/1463 | Loss: 0.7925 | LM_LOSS: 0.2923 | LB_LOSS: 1.0004 [2026-05-23 21:17:42] Validation | Batch 1260/1463 | Loss: 0.7924 | LM_LOSS: 0.2922 | LB_LOSS: 1.0004 [2026-05-23 21:17:45] Validation | Batch 1270/1463 | Loss: 0.7922 | LM_LOSS: 0.2920 | LB_LOSS: 1.0004 [2026-05-23 21:17:48] Validation | Batch 1280/1463 | Loss: 0.7921 | LM_LOSS: 0.2919 | LB_LOSS: 1.0004 [2026-05-23 21:17:51] Validation | Batch 1290/1463 | Loss: 0.7921 | LM_LOSS: 0.2919 | LB_LOSS: 1.0004 [2026-05-23 21:17:54] Validation | Batch 1300/1463 | Loss: 0.7919 | LM_LOSS: 0.2917 | LB_LOSS: 1.0004 [2026-05-23 21:17:57] Validation | Batch 1310/1463 | Loss: 0.7917 | LM_LOSS: 0.2915 | LB_LOSS: 1.0004 [2026-05-23 21:18:00] Validation | Batch 1320/1463 | Loss: 0.7918 | LM_LOSS: 0.2916 | LB_LOSS: 1.0004 [2026-05-23 21:18:02] Validation | Batch 1330/1463 | Loss: 0.7918 | LM_LOSS: 0.2916 | LB_LOSS: 1.0004 [2026-05-23 21:18:05] Validation | Batch 1340/1463 | Loss: 0.7921 | LM_LOSS: 0.2919 | LB_LOSS: 1.0004 [2026-05-23 21:18:08] Validation | Batch 1350/1463 | Loss: 0.7920 | LM_LOSS: 0.2918 | LB_LOSS: 1.0004 [2026-05-23 21:18:12] Validation | Batch 1360/1463 | Loss: 0.7923 | LM_LOSS: 0.2921 | LB_LOSS: 1.0004 [2026-05-23 21:18:15] Validation | Batch 1370/1463 | Loss: 0.7926 | LM_LOSS: 0.2924 | LB_LOSS: 1.0004 [2026-05-23 21:18:18] Validation | Batch 1380/1463 | Loss: 0.7923 | LM_LOSS: 0.2921 | LB_LOSS: 1.0004 [2026-05-23 21:18:21] Validation | Batch 1390/1463 | Loss: 0.7923 | LM_LOSS: 0.2921 | LB_LOSS: 1.0004 [2026-05-23 21:18:24] Validation | Batch 1400/1463 | Loss: 0.7926 | LM_LOSS: 0.2924 | LB_LOSS: 1.0004 [2026-05-23 21:18:26] Validation | Batch 1410/1463 | Loss: 0.7929 | LM_LOSS: 0.2927 | LB_LOSS: 1.0004 [2026-05-23 21:18:29] Validation | Batch 1420/1463 | Loss: 0.7929 | LM_LOSS: 0.2927 | LB_LOSS: 1.0004 [2026-05-23 21:18:33] Validation | Batch 1430/1463 | Loss: 0.7929 | LM_LOSS: 0.2927 | LB_LOSS: 1.0004 [2026-05-23 21:18:36] Validation | Batch 1440/1463 | Loss: 0.7933 | LM_LOSS: 0.2931 | LB_LOSS: 1.0004 [2026-05-23 21:18:39] Validation | Batch 1450/1463 | Loss: 0.7931 | LM_LOSS: 0.2929 | LB_LOSS: 1.0004 [2026-05-23 21:18:42] Validation | Batch 1460/1463 | Loss: 0.7934 | LM_LOSS: 0.2932 | LB_LOSS: 1.0004 [2026-05-23 21:18:43] Validation | Batch 1463/1463 | Loss: 0.7933 | LM_LOSS: 0.2931 | LB_LOSS: 1.0004 [2026-05-23 21:18:43] Validation | Loss: 0.7933 | LM_LOSS: 0.2931 | LB_LOSS: 1.0004 | PPL: 1.34 | Time: 443.27s [2026-05-23 21:18:55] Epoch 2 | Step 8010 | Loss: 0.7178 | LM: 0.2178 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.390 | LR: 1.00e-04 [2026-05-23 21:19:07] Epoch 2 | Step 8020 | Loss: 0.7178 | LM: 0.2179 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.390 | LR: 1.00e-04 [2026-05-23 21:19:19] Epoch 2 | Step 8030 | Loss: 0.7178 | LM: 0.2178 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.390 | LR: 1.00e-04 [2026-05-23 21:19:31] Epoch 2 | Step 8040 | Loss: 0.7179 | LM: 0.2179 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.390 | LR: 1.00e-04 [2026-05-23 21:19:43] Epoch 2 | Step 8050 | Loss: 0.7178 | LM: 0.2178 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.390 | LR: 1.00e-04 [2026-05-23 21:19:55] Epoch 2 | Step 8060 | Loss: 0.7178 | LM: 0.2177 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.390 | LR: 1.00e-04 [2026-05-23 21:20:07] Epoch 2 | Step 8070 | Loss: 0.7178 | LM: 0.2176 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.390 | LR: 1.00e-04 [2026-05-23 21:20:20] Epoch 2 | Step 8080 | Loss: 0.7178 | LM: 0.2177 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.390 | LR: 1.00e-04 [2026-05-23 21:20:32] Epoch 2 | Step 8090 | Loss: 0.7178 | LM: 0.2176 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.390 | LR: 1.00e-04 [2026-05-23 21:20:44] Epoch 2 | Step 8100 | Loss: 0.7178 | LM: 0.2176 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.390 | LR: 1.00e-04 [2026-05-23 21:20:56] Epoch 2 | Step 8110 | Loss: 0.7178 | LM: 0.2177 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.390 | LR: 1.00e-04 [2026-05-23 21:21:08] Epoch 2 | Step 8120 | Loss: 0.7178 | LM: 0.2177 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.390 | LR: 1.00e-04 [2026-05-23 21:21:22] Epoch 2 | Step 8130 | Loss: 0.7178 | LM: 0.2176 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.390 | LR: 1.00e-04 [2026-05-23 21:21:35] Epoch 2 | Step 8140 | Loss: 0.7179 | LM: 0.2177 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.390 | LR: 1.00e-04 [2026-05-23 21:21:47] Epoch 2 | Step 8150 | Loss: 0.7179 | LM: 0.2176 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.390 | LR: 1.00e-04 [2026-05-23 21:21:59] Epoch 2 | Step 8160 | Loss: 0.7179 | LM: 0.2176 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.390 | LR: 1.00e-04 [2026-05-23 21:22:11] Epoch 2 | Step 8170 | Loss: 0.7179 | LM: 0.2178 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.390 | LR: 1.00e-04 [2026-05-23 21:22:25] Epoch 2 | Step 8180 | Loss: 0.7179 | LM: 0.2179 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.390 | LR: 1.00e-04 [2026-05-23 21:22:37] Epoch 2 | Step 8190 | Loss: 0.7180 | LM: 0.2181 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.390 | LR: 1.00e-04 [2026-05-23 21:22:50] Epoch 2 | Step 8200 | Loss: 0.7181 | LM: 0.2181 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.390 | LR: 1.00e-04 [2026-05-23 21:23:05] Epoch 2 | Step 8210 | Loss: 0.7181 | LM: 0.2183 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.390 | LR: 1.00e-04 [2026-05-23 21:23:17] Epoch 2 | Step 8220 | Loss: 0.7181 | LM: 0.2182 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.390 | LR: 1.00e-04 [2026-05-23 21:23:29] Epoch 2 | Step 8230 | Loss: 0.7181 | LM: 0.2181 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.438 | HR1: 0.400/SR1: 0.390 | LR: 1.00e-04 [2026-05-23 21:23:42] Epoch 2 | Step 8240 | Loss: 0.7181 | LM: 0.2181 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.438 | HR1: 0.400/SR1: 0.390 | LR: 1.00e-04 [2026-05-23 21:23:55] Epoch 2 | Step 8250 | Loss: 0.7181 | LM: 0.2183 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.438 | HR1: 0.400/SR1: 0.390 | LR: 1.00e-04 [2026-05-23 21:24:07] Epoch 2 | Step 8260 | Loss: 0.7182 | LM: 0.2183 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.438 | HR1: 0.400/SR1: 0.390 | LR: 1.00e-04 [2026-05-23 21:24:20] Epoch 2 | Step 8270 | Loss: 0.7183 | LM: 0.2183 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.438 | HR1: 0.400/SR1: 0.390 | LR: 1.00e-04 [2026-05-23 21:24:33] Epoch 2 | Step 8280 | Loss: 0.7183 | LM: 0.2183 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.438 | HR1: 0.400/SR1: 0.390 | LR: 1.00e-04 [2026-05-23 21:24:46] Epoch 2 | Step 8290 | Loss: 0.7183 | LM: 0.2183 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.438 | HR1: 0.400/SR1: 0.390 | LR: 1.00e-04 [2026-05-23 21:24:58] Epoch 2 | Step 8300 | Loss: 0.7183 | LM: 0.2183 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.438 | HR1: 0.400/SR1: 0.390 | LR: 1.00e-04 [2026-05-23 21:25:11] Epoch 2 | Step 8310 | Loss: 0.7183 | LM: 0.2183 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.390 | LR: 1.00e-04 [2026-05-23 21:25:22] Epoch 2 | Step 8320 | Loss: 0.7183 | LM: 0.2184 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.390 | LR: 1.00e-04 [2026-05-23 21:25:35] Epoch 2 | Step 8330 | Loss: 0.7183 | LM: 0.2185 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.390 | LR: 1.00e-04 [2026-05-23 21:25:47] Epoch 2 | Step 8340 | Loss: 0.7183 | LM: 0.2183 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.390 | LR: 1.00e-04 [2026-05-23 21:25:59] Epoch 2 | Step 8350 | Loss: 0.7183 | LM: 0.2183 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.390 | LR: 1.00e-04 [2026-05-23 21:26:12] Epoch 2 | Step 8360 | Loss: 0.7184 | LM: 0.2185 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.390 | LR: 1.00e-04 [2026-05-23 21:26:24] Epoch 2 | Step 8370 | Loss: 0.7184 | LM: 0.2184 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.390 | LR: 1.00e-04 [2026-05-23 21:26:36] Epoch 2 | Step 8380 | Loss: 0.7183 | LM: 0.2183 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.390 | LR: 1.00e-04 [2026-05-23 21:26:48] Epoch 2 | Step 8390 | Loss: 0.7184 | LM: 0.2183 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.390 | LR: 1.00e-04 [2026-05-23 21:27:00] Epoch 2 | Step 8400 | Loss: 0.7184 | LM: 0.2184 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.390 | LR: 1.00e-04 [2026-05-23 21:27:13] Epoch 2 | Step 8410 | Loss: 0.7184 | LM: 0.2183 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.390 | LR: 1.00e-04 [2026-05-23 21:27:25] Epoch 2 | Step 8420 | Loss: 0.7184 | LM: 0.2183 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.390 | LR: 1.00e-04 [2026-05-23 21:27:38] Epoch 2 | Step 8430 | Loss: 0.7185 | LM: 0.2183 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.390 | LR: 1.00e-04 [2026-05-23 21:27:50] Epoch 2 | Step 8440 | Loss: 0.7185 | LM: 0.2184 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.390 | LR: 1.00e-04 [2026-05-23 21:28:01] Epoch 2 | Step 8450 | Loss: 0.7185 | LM: 0.2185 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.390 | LR: 1.00e-04 [2026-05-23 21:28:13] Epoch 2 | Step 8460 | Loss: 0.7185 | LM: 0.2185 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.390 | LR: 1.00e-04 [2026-05-23 21:28:24] Epoch 2 | Step 8470 | Loss: 0.7185 | LM: 0.2184 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.390 | LR: 1.00e-04 [2026-05-23 21:28:37] Epoch 2 | Step 8480 | Loss: 0.7184 | LM: 0.2182 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.390 | LR: 1.00e-04 [2026-05-23 21:28:49] Epoch 2 | Step 8490 | Loss: 0.7185 | LM: 0.2184 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.390 | LR: 1.00e-04 [2026-05-23 21:29:02] Epoch 2 | Step 8500 | Loss: 0.7185 | LM: 0.2184 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.390 | LR: 1.00e-04 [2026-05-23 21:29:14] Epoch 2 | Step 8510 | Loss: 0.7185 | LM: 0.2185 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.390 | LR: 1.00e-04 [2026-05-23 21:29:27] Epoch 2 | Step 8520 | Loss: 0.7185 | LM: 0.2186 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.390 | LR: 1.00e-04 [2026-05-23 21:29:40] Epoch 2 | Step 8530 | Loss: 0.7186 | LM: 0.2187 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.390 | LR: 1.00e-04 [2026-05-23 21:29:54] Epoch 2 | Step 8540 | Loss: 0.7186 | LM: 0.2185 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.390 | LR: 1.00e-04 [2026-05-23 21:30:07] Epoch 2 | Step 8550 | Loss: 0.7186 | LM: 0.2185 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.390 | LR: 1.00e-04 [2026-05-23 21:30:20] Epoch 2 | Step 8560 | Loss: 0.7185 | LM: 0.2185 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.390 | LR: 1.00e-04 [2026-05-23 21:30:32] Epoch 2 | Step 8570 | Loss: 0.7185 | LM: 0.2185 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.390 | LR: 1.00e-04 [2026-05-23 21:30:45] Epoch 2 | Step 8580 | Loss: 0.7184 | LM: 0.2184 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.390 | LR: 1.00e-04 [2026-05-23 21:30:56] Epoch 2 | Step 8590 | Loss: 0.7185 | LM: 0.2185 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.390 | LR: 1.00e-04 [2026-05-23 21:31:08] Epoch 2 | Step 8600 | Loss: 0.7185 | LM: 0.2185 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.390 | LR: 1.00e-04 [2026-05-23 21:31:20] Epoch 2 | Step 8610 | Loss: 0.7185 | LM: 0.2186 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.390 | LR: 1.00e-04 [2026-05-23 21:31:33] Epoch 2 | Step 8620 | Loss: 0.7186 | LM: 0.2187 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.390 | LR: 1.00e-04 [2026-05-23 21:31:44] Epoch 2 | Step 8630 | Loss: 0.7185 | LM: 0.2187 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.390 | LR: 1.00e-04 [2026-05-23 21:31:56] Epoch 2 | Step 8640 | Loss: 0.7185 | LM: 0.2186 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.390 | LR: 1.00e-04 [2026-05-23 21:32:09] Epoch 2 | Step 8650 | Loss: 0.7185 | LM: 0.2186 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.390 | LR: 1.00e-04 [2026-05-23 21:32:21] Epoch 2 | Step 8660 | Loss: 0.7185 | LM: 0.2186 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.391 | LR: 1.00e-04 [2026-05-23 21:32:33] Epoch 2 | Step 8670 | Loss: 0.7185 | LM: 0.2185 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.390 | LR: 1.00e-04 [2026-05-23 21:32:45] Epoch 2 | Step 8680 | Loss: 0.7185 | LM: 0.2184 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.391 | LR: 1.00e-04 [2026-05-23 21:32:58] Epoch 2 | Step 8690 | Loss: 0.7185 | LM: 0.2184 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.390 | LR: 1.00e-04 [2026-05-23 21:33:09] Epoch 2 | Step 8700 | Loss: 0.7185 | LM: 0.2186 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.391 | LR: 1.00e-04 [2026-05-23 21:33:21] Epoch 2 | Step 8710 | Loss: 0.7186 | LM: 0.2186 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.391 | LR: 1.00e-04 [2026-05-23 21:33:33] Epoch 2 | Step 8720 | Loss: 0.7186 | LM: 0.2187 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.391 | LR: 1.00e-04 [2026-05-23 21:33:44] Epoch 2 | Step 8730 | Loss: 0.7187 | LM: 0.2188 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.391 | LR: 1.00e-04 [2026-05-23 21:33:56] Epoch 2 | Step 8740 | Loss: 0.7187 | LM: 0.2189 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.391 | LR: 1.00e-04 [2026-05-23 21:34:07] Epoch 2 | Step 8750 | Loss: 0.7187 | LM: 0.2188 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.391 | LR: 1.00e-04 [2026-05-23 21:34:21] Epoch 2 | Step 8760 | Loss: 0.7188 | LM: 0.2189 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.391 | LR: 1.00e-04 [2026-05-23 21:34:33] Epoch 2 | Step 8770 | Loss: 0.7188 | LM: 0.2190 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.391 | LR: 1.00e-04 [2026-05-23 21:34:45] Epoch 2 | Step 8780 | Loss: 0.7188 | LM: 0.2188 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.391 | LR: 1.00e-04 [2026-05-23 21:34:57] Epoch 2 | Step 8790 | Loss: 0.7188 | LM: 0.2189 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.391 | LR: 1.00e-04 [2026-05-23 21:35:10] Epoch 2 | Step 8800 | Loss: 0.7189 | LM: 0.2190 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.391 | LR: 1.00e-04 [2026-05-23 21:35:21] Epoch 2 | Step 8810 | Loss: 0.7189 | LM: 0.2191 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.391 | LR: 1.00e-04 [2026-05-23 21:35:33] Epoch 2 | Step 8820 | Loss: 0.7190 | LM: 0.2193 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.391 | LR: 1.00e-04 [2026-05-23 21:35:44] Epoch 2 | Step 8830 | Loss: 0.7190 | LM: 0.2194 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.391 | LR: 1.00e-04 [2026-05-23 21:35:56] Epoch 2 | Step 8840 | Loss: 0.7189 | LM: 0.2195 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.391 | LR: 1.00e-04 [2026-05-23 21:36:09] Epoch 2 | Step 8850 | Loss: 0.7190 | LM: 0.2195 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.391 | LR: 1.00e-04 [2026-05-23 21:36:21] Epoch 2 | Step 8860 | Loss: 0.7190 | LM: 0.2195 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.391 | LR: 1.00e-04 [2026-05-23 21:36:33] Epoch 2 | Step 8870 | Loss: 0.7190 | LM: 0.2195 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.391 | LR: 1.00e-04 [2026-05-23 21:36:45] Epoch 2 | Step 8880 | Loss: 0.7190 | LM: 0.2196 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.391 | LR: 1.00e-04 [2026-05-23 21:36:58] Epoch 2 | Step 8890 | Loss: 0.7190 | LM: 0.2197 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.391 | LR: 1.00e-04 [2026-05-23 21:37:10] Epoch 2 | Step 8900 | Loss: 0.7190 | LM: 0.2196 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.391 | LR: 1.00e-04 [2026-05-23 21:37:23] Epoch 2 | Step 8910 | Loss: 0.7190 | LM: 0.2198 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.391 | LR: 1.00e-04 [2026-05-23 21:37:35] Epoch 2 | Step 8920 | Loss: 0.7190 | LM: 0.2198 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.391 | LR: 1.00e-04 [2026-05-23 21:37:47] Epoch 2 | Step 8930 | Loss: 0.7190 | LM: 0.2198 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.391 | LR: 1.00e-04 [2026-05-23 21:37:59] Epoch 2 | Step 8940 | Loss: 0.7190 | LM: 0.2198 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.391 | LR: 1.00e-04 [2026-05-23 21:38:11] Epoch 2 | Step 8950 | Loss: 0.7190 | LM: 0.2196 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.391 | LR: 1.00e-04 [2026-05-23 21:38:23] Epoch 2 | Step 8960 | Loss: 0.7189 | LM: 0.2196 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.391 | LR: 1.00e-04 [2026-05-23 21:38:37] Epoch 2 | Step 8970 | Loss: 0.7190 | LM: 0.2198 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.391 | LR: 1.00e-04 [2026-05-23 21:38:49] Epoch 2 | Step 8980 | Loss: 0.7191 | LM: 0.2199 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.391 | LR: 1.00e-04 [2026-05-23 21:39:00] Epoch 2 | Step 8990 | Loss: 0.7191 | LM: 0.2199 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.391 | LR: 1.00e-04 [2026-05-23 21:39:12] Epoch 2 | Step 9000 | Loss: 0.7191 | LM: 0.2198 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.391 | LR: 1.00e-04 [2026-05-23 21:39:16] Checkpoint saved: outputs/2026-05-23/N_2.5/checkpoints/checkpoint_step_9000.pt [2026-05-23 21:39:25] Validation | Batch 10/1463 | Loss: 0.7639 | LM_LOSS: 0.2641 | LB_LOSS: 0.9996 [2026-05-23 21:39:27] Validation | Batch 20/1463 | Loss: 0.7717 | LM_LOSS: 0.2717 | LB_LOSS: 1.0001 [2026-05-23 21:39:31] Validation | Batch 30/1463 | Loss: 0.7726 | LM_LOSS: 0.2724 | LB_LOSS: 1.0004 [2026-05-23 21:39:34] Validation | Batch 40/1463 | Loss: 0.7793 | LM_LOSS: 0.2791 | LB_LOSS: 1.0004 [2026-05-23 21:39:37] Validation | Batch 50/1463 | Loss: 0.7877 | LM_LOSS: 0.2876 | LB_LOSS: 1.0003 [2026-05-23 21:39:41] Validation | Batch 60/1463 | Loss: 0.7878 | LM_LOSS: 0.2876 | LB_LOSS: 1.0004 [2026-05-23 21:39:44] Validation | Batch 70/1463 | Loss: 0.7880 | LM_LOSS: 0.2878 | LB_LOSS: 1.0004 [2026-05-23 21:39:48] Validation | Batch 80/1463 | Loss: 0.7903 | LM_LOSS: 0.2901 | LB_LOSS: 1.0004 [2026-05-23 21:39:51] Validation | Batch 90/1463 | Loss: 0.7905 | LM_LOSS: 0.2903 | LB_LOSS: 1.0004 [2026-05-23 21:39:54] Validation | Batch 100/1463 | Loss: 0.7904 | LM_LOSS: 0.2902 | LB_LOSS: 1.0004 [2026-05-23 21:39:57] Validation | Batch 110/1463 | Loss: 0.7879 | LM_LOSS: 0.2877 | LB_LOSS: 1.0004 [2026-05-23 21:39:59] Validation | Batch 120/1463 | Loss: 0.7896 | LM_LOSS: 0.2894 | LB_LOSS: 1.0004 [2026-05-23 21:40:02] Validation | Batch 130/1463 | Loss: 0.7911 | LM_LOSS: 0.2909 | LB_LOSS: 1.0004 [2026-05-23 21:40:05] Validation | Batch 140/1463 | Loss: 0.7923 | LM_LOSS: 0.2921 | LB_LOSS: 1.0004 [2026-05-23 21:40:09] Validation | Batch 150/1463 | Loss: 0.7904 | LM_LOSS: 0.2901 | LB_LOSS: 1.0005 [2026-05-23 21:40:12] Validation | Batch 160/1463 | Loss: 0.7910 | LM_LOSS: 0.2908 | LB_LOSS: 1.0005 [2026-05-23 21:40:15] Validation | Batch 170/1463 | Loss: 0.7902 | LM_LOSS: 0.2900 | LB_LOSS: 1.0005 [2026-05-23 21:40:18] Validation | Batch 180/1463 | Loss: 0.7911 | LM_LOSS: 0.2908 | LB_LOSS: 1.0005 [2026-05-23 21:40:21] Validation | Batch 190/1463 | Loss: 0.7922 | LM_LOSS: 0.2920 | LB_LOSS: 1.0005 [2026-05-23 21:40:24] Validation | Batch 200/1463 | Loss: 0.7922 | LM_LOSS: 0.2919 | LB_LOSS: 1.0005 [2026-05-23 21:40:27] Validation | Batch 210/1463 | Loss: 0.7931 | LM_LOSS: 0.2928 | LB_LOSS: 1.0005 [2026-05-23 21:40:31] Validation | Batch 220/1463 | Loss: 0.7922 | LM_LOSS: 0.2920 | LB_LOSS: 1.0005 [2026-05-23 21:40:34] Validation | Batch 230/1463 | Loss: 0.7935 | LM_LOSS: 0.2932 | LB_LOSS: 1.0005 [2026-05-23 21:40:37] Validation | Batch 240/1463 | Loss: 0.7952 | LM_LOSS: 0.2949 | LB_LOSS: 1.0005 [2026-05-23 21:40:41] Validation | Batch 250/1463 | Loss: 0.7941 | LM_LOSS: 0.2938 | LB_LOSS: 1.0005 [2026-05-23 21:40:43] Validation | Batch 260/1463 | Loss: 0.7958 | LM_LOSS: 0.2955 | LB_LOSS: 1.0005 [2026-05-23 21:40:47] Validation | Batch 270/1463 | Loss: 0.7943 | LM_LOSS: 0.2940 | LB_LOSS: 1.0005 [2026-05-23 21:40:49] Validation | Batch 280/1463 | Loss: 0.7947 | LM_LOSS: 0.2945 | LB_LOSS: 1.0005 [2026-05-23 21:40:53] Validation | Batch 290/1463 | Loss: 0.7936 | LM_LOSS: 0.2934 | LB_LOSS: 1.0005 [2026-05-23 21:40:55] Validation | Batch 300/1463 | Loss: 0.7940 | LM_LOSS: 0.2937 | LB_LOSS: 1.0005 [2026-05-23 21:40:58] Validation | Batch 310/1463 | Loss: 0.7950 | LM_LOSS: 0.2948 | LB_LOSS: 1.0005 [2026-05-23 21:41:01] Validation | Batch 320/1463 | Loss: 0.7938 | LM_LOSS: 0.2936 | LB_LOSS: 1.0005 [2026-05-23 21:41:04] Validation | Batch 330/1463 | Loss: 0.7935 | LM_LOSS: 0.2933 | LB_LOSS: 1.0005 [2026-05-23 21:41:07] Validation | Batch 340/1463 | Loss: 0.7932 | LM_LOSS: 0.2929 | LB_LOSS: 1.0005 [2026-05-23 21:41:11] Validation | Batch 350/1463 | Loss: 0.7939 | LM_LOSS: 0.2936 | LB_LOSS: 1.0005 [2026-05-23 21:41:13] Validation | Batch 360/1463 | Loss: 0.7937 | LM_LOSS: 0.2934 | LB_LOSS: 1.0005 [2026-05-23 21:41:16] Validation | Batch 370/1463 | Loss: 0.7935 | LM_LOSS: 0.2932 | LB_LOSS: 1.0005 [2026-05-23 21:41:19] Validation | Batch 380/1463 | Loss: 0.7932 | LM_LOSS: 0.2929 | LB_LOSS: 1.0005 [2026-05-23 21:41:22] Validation | Batch 390/1463 | Loss: 0.7937 | LM_LOSS: 0.2935 | LB_LOSS: 1.0005 [2026-05-23 21:41:25] Validation | Batch 400/1463 | Loss: 0.7935 | LM_LOSS: 0.2933 | LB_LOSS: 1.0005 [2026-05-23 21:41:28] Validation | Batch 410/1463 | Loss: 0.7932 | LM_LOSS: 0.2929 | LB_LOSS: 1.0005 [2026-05-23 21:41:31] Validation | Batch 420/1463 | Loss: 0.7933 | LM_LOSS: 0.2931 | LB_LOSS: 1.0005 [2026-05-23 21:41:35] Validation | Batch 430/1463 | Loss: 0.7929 | LM_LOSS: 0.2926 | LB_LOSS: 1.0005 [2026-05-23 21:41:37] Validation | Batch 440/1463 | Loss: 0.7931 | LM_LOSS: 0.2928 | LB_LOSS: 1.0005 [2026-05-23 21:41:40] Validation | Batch 450/1463 | Loss: 0.7929 | LM_LOSS: 0.2926 | LB_LOSS: 1.0005 [2026-05-23 21:41:43] Validation | Batch 460/1463 | Loss: 0.7928 | LM_LOSS: 0.2926 | LB_LOSS: 1.0005 [2026-05-23 21:41:46] Validation | Batch 470/1463 | Loss: 0.7935 | LM_LOSS: 0.2933 | LB_LOSS: 1.0005 [2026-05-23 21:41:50] Validation | Batch 480/1463 | Loss: 0.7932 | LM_LOSS: 0.2930 | LB_LOSS: 1.0005 [2026-05-23 21:41:53] Validation | Batch 490/1463 | Loss: 0.7932 | LM_LOSS: 0.2929 | LB_LOSS: 1.0005 [2026-05-23 21:41:57] Validation | Batch 500/1463 | Loss: 0.7931 | LM_LOSS: 0.2928 | LB_LOSS: 1.0005 [2026-05-23 21:42:00] Validation | Batch 510/1463 | Loss: 0.7924 | LM_LOSS: 0.2922 | LB_LOSS: 1.0005 [2026-05-23 21:42:02] Validation | Batch 520/1463 | Loss: 0.7922 | LM_LOSS: 0.2919 | LB_LOSS: 1.0005 [2026-05-23 21:42:05] Validation | Batch 530/1463 | Loss: 0.7926 | LM_LOSS: 0.2923 | LB_LOSS: 1.0005 [2026-05-23 21:42:08] Validation | Batch 540/1463 | Loss: 0.7928 | LM_LOSS: 0.2925 | LB_LOSS: 1.0005 [2026-05-23 21:42:10] Validation | Batch 550/1463 | Loss: 0.7924 | LM_LOSS: 0.2921 | LB_LOSS: 1.0005 [2026-05-23 21:42:13] Validation | Batch 560/1463 | Loss: 0.7924 | LM_LOSS: 0.2921 | LB_LOSS: 1.0005 [2026-05-23 21:42:16] Validation | Batch 570/1463 | Loss: 0.7923 | LM_LOSS: 0.2921 | LB_LOSS: 1.0005 [2026-05-23 21:42:20] Validation | Batch 580/1463 | Loss: 0.7918 | LM_LOSS: 0.2916 | LB_LOSS: 1.0005 [2026-05-23 21:42:23] Validation | Batch 590/1463 | Loss: 0.7916 | LM_LOSS: 0.2914 | LB_LOSS: 1.0005 [2026-05-23 21:42:26] Validation | Batch 600/1463 | Loss: 0.7919 | LM_LOSS: 0.2917 | LB_LOSS: 1.0005 [2026-05-23 21:42:30] Validation | Batch 610/1463 | Loss: 0.7922 | LM_LOSS: 0.2919 | LB_LOSS: 1.0005 [2026-05-23 21:42:33] Validation | Batch 620/1463 | Loss: 0.7917 | LM_LOSS: 0.2914 | LB_LOSS: 1.0005 [2026-05-23 21:42:36] Validation | Batch 630/1463 | Loss: 0.7911 | LM_LOSS: 0.2908 | LB_LOSS: 1.0005 [2026-05-23 21:42:39] Validation | Batch 640/1463 | Loss: 0.7909 | LM_LOSS: 0.2907 | LB_LOSS: 1.0005 [2026-05-23 21:42:41] Validation | Batch 650/1463 | Loss: 0.7905 | LM_LOSS: 0.2902 | LB_LOSS: 1.0005 [2026-05-23 21:42:44] Validation | Batch 660/1463 | Loss: 0.7901 | LM_LOSS: 0.2898 | LB_LOSS: 1.0005 [2026-05-23 21:42:47] Validation | Batch 670/1463 | Loss: 0.7896 | LM_LOSS: 0.2894 | LB_LOSS: 1.0005 [2026-05-23 21:42:49] Validation | Batch 680/1463 | Loss: 0.7897 | LM_LOSS: 0.2895 | LB_LOSS: 1.0005 [2026-05-23 21:42:53] Validation | Batch 690/1463 | Loss: 0.7899 | LM_LOSS: 0.2896 | LB_LOSS: 1.0005 [2026-05-23 21:42:56] Validation | Batch 700/1463 | Loss: 0.7898 | LM_LOSS: 0.2896 | LB_LOSS: 1.0005 [2026-05-23 21:42:58] Validation | Batch 710/1463 | Loss: 0.7900 | LM_LOSS: 0.2898 | LB_LOSS: 1.0005 [2026-05-23 21:43:02] Validation | Batch 720/1463 | Loss: 0.7907 | LM_LOSS: 0.2905 | LB_LOSS: 1.0005 [2026-05-23 21:43:04] Validation | Batch 730/1463 | Loss: 0.7906 | LM_LOSS: 0.2903 | LB_LOSS: 1.0005 [2026-05-23 21:43:06] Validation | Batch 740/1463 | Loss: 0.7904 | LM_LOSS: 0.2901 | LB_LOSS: 1.0005 [2026-05-23 21:43:09] Validation | Batch 750/1463 | Loss: 0.7903 | LM_LOSS: 0.2900 | LB_LOSS: 1.0005 [2026-05-23 21:43:12] Validation | Batch 760/1463 | Loss: 0.7902 | LM_LOSS: 0.2899 | LB_LOSS: 1.0005 [2026-05-23 21:43:14] Validation | Batch 770/1463 | Loss: 0.7902 | LM_LOSS: 0.2899 | LB_LOSS: 1.0005 [2026-05-23 21:43:17] Validation | Batch 780/1463 | Loss: 0.7901 | LM_LOSS: 0.2899 | LB_LOSS: 1.0005 [2026-05-23 21:43:19] Validation | Batch 790/1463 | Loss: 0.7903 | LM_LOSS: 0.2901 | LB_LOSS: 1.0005 [2026-05-23 21:43:22] Validation | Batch 800/1463 | Loss: 0.7902 | LM_LOSS: 0.2899 | LB_LOSS: 1.0005 [2026-05-23 21:43:25] Validation | Batch 810/1463 | Loss: 0.7899 | LM_LOSS: 0.2896 | LB_LOSS: 1.0005 [2026-05-23 21:43:28] Validation | Batch 820/1463 | Loss: 0.7901 | LM_LOSS: 0.2898 | LB_LOSS: 1.0005 [2026-05-23 21:43:31] Validation | Batch 830/1463 | Loss: 0.7903 | LM_LOSS: 0.2901 | LB_LOSS: 1.0005 [2026-05-23 21:43:35] Validation | Batch 840/1463 | Loss: 0.7904 | LM_LOSS: 0.2901 | LB_LOSS: 1.0005 [2026-05-23 21:43:38] Validation | Batch 850/1463 | Loss: 0.7907 | LM_LOSS: 0.2904 | LB_LOSS: 1.0005 [2026-05-23 21:43:41] Validation | Batch 860/1463 | Loss: 0.7908 | LM_LOSS: 0.2906 | LB_LOSS: 1.0005 [2026-05-23 21:43:44] Validation | Batch 870/1463 | Loss: 0.7909 | LM_LOSS: 0.2907 | LB_LOSS: 1.0005 [2026-05-23 21:43:47] Validation | Batch 880/1463 | Loss: 0.7902 | LM_LOSS: 0.2899 | LB_LOSS: 1.0005 [2026-05-23 21:43:49] Validation | Batch 890/1463 | Loss: 0.7901 | LM_LOSS: 0.2899 | LB_LOSS: 1.0005 [2026-05-23 21:43:53] Validation | Batch 900/1463 | Loss: 0.7899 | LM_LOSS: 0.2896 | LB_LOSS: 1.0005 [2026-05-23 21:43:56] Validation | Batch 910/1463 | Loss: 0.7899 | LM_LOSS: 0.2896 | LB_LOSS: 1.0005 [2026-05-23 21:43:59] Validation | Batch 920/1463 | Loss: 0.7900 | LM_LOSS: 0.2898 | LB_LOSS: 1.0005 [2026-05-23 21:44:02] Validation | Batch 930/1463 | Loss: 0.7898 | LM_LOSS: 0.2896 | LB_LOSS: 1.0005 [2026-05-23 21:44:05] Validation | Batch 940/1463 | Loss: 0.7896 | LM_LOSS: 0.2894 | LB_LOSS: 1.0005 [2026-05-23 21:44:09] Validation | Batch 950/1463 | Loss: 0.7898 | LM_LOSS: 0.2896 | LB_LOSS: 1.0005 [2026-05-23 21:44:12] Validation | Batch 960/1463 | Loss: 0.7899 | LM_LOSS: 0.2897 | LB_LOSS: 1.0005 [2026-05-23 21:44:15] Validation | Batch 970/1463 | Loss: 0.7899 | LM_LOSS: 0.2896 | LB_LOSS: 1.0005 [2026-05-23 21:44:19] Validation | Batch 980/1463 | Loss: 0.7902 | LM_LOSS: 0.2899 | LB_LOSS: 1.0005 [2026-05-23 21:44:21] Validation | Batch 990/1463 | Loss: 0.7904 | LM_LOSS: 0.2902 | LB_LOSS: 1.0005 [2026-05-23 21:44:24] Validation | Batch 1000/1463 | Loss: 0.7910 | LM_LOSS: 0.2907 | LB_LOSS: 1.0005 [2026-05-23 21:44:27] Validation | Batch 1010/1463 | Loss: 0.7910 | LM_LOSS: 0.2908 | LB_LOSS: 1.0005 [2026-05-23 21:44:31] Validation | Batch 1020/1463 | Loss: 0.7911 | LM_LOSS: 0.2908 | LB_LOSS: 1.0005 [2026-05-23 21:44:34] Validation | Batch 1030/1463 | Loss: 0.7916 | LM_LOSS: 0.2913 | LB_LOSS: 1.0005 [2026-05-23 21:44:36] Validation | Batch 1040/1463 | Loss: 0.7912 | LM_LOSS: 0.2909 | LB_LOSS: 1.0005 [2026-05-23 21:44:40] Validation | Batch 1050/1463 | Loss: 0.7911 | LM_LOSS: 0.2909 | LB_LOSS: 1.0005 [2026-05-23 21:44:43] Validation | Batch 1060/1463 | Loss: 0.7911 | LM_LOSS: 0.2908 | LB_LOSS: 1.0005 [2026-05-23 21:44:46] Validation | Batch 1070/1463 | Loss: 0.7911 | LM_LOSS: 0.2909 | LB_LOSS: 1.0005 [2026-05-23 21:44:49] Validation | Batch 1080/1463 | Loss: 0.7909 | LM_LOSS: 0.2907 | LB_LOSS: 1.0005 [2026-05-23 21:44:51] Validation | Batch 1090/1463 | Loss: 0.7910 | LM_LOSS: 0.2907 | LB_LOSS: 1.0005 [2026-05-23 21:44:54] Validation | Batch 1100/1463 | Loss: 0.7907 | LM_LOSS: 0.2905 | LB_LOSS: 1.0005 [2026-05-23 21:44:57] Validation | Batch 1110/1463 | Loss: 0.7908 | LM_LOSS: 0.2905 | LB_LOSS: 1.0005 [2026-05-23 21:45:00] Validation | Batch 1120/1463 | Loss: 0.7905 | LM_LOSS: 0.2903 | LB_LOSS: 1.0005 [2026-05-23 21:45:03] Validation | Batch 1130/1463 | Loss: 0.7904 | LM_LOSS: 0.2901 | LB_LOSS: 1.0005 [2026-05-23 21:45:06] Validation | Batch 1140/1463 | Loss: 0.7906 | LM_LOSS: 0.2903 | LB_LOSS: 1.0005 [2026-05-23 21:45:10] Validation | Batch 1150/1463 | Loss: 0.7905 | LM_LOSS: 0.2902 | LB_LOSS: 1.0005 [2026-05-23 21:45:13] Validation | Batch 1160/1463 | Loss: 0.7904 | LM_LOSS: 0.2901 | LB_LOSS: 1.0005 [2026-05-23 21:45:16] Validation | Batch 1170/1463 | Loss: 0.7904 | LM_LOSS: 0.2902 | LB_LOSS: 1.0005 [2026-05-23 21:45:19] Validation | Batch 1180/1463 | Loss: 0.7905 | LM_LOSS: 0.2902 | LB_LOSS: 1.0005 [2026-05-23 21:45:22] Validation | Batch 1190/1463 | Loss: 0.7907 | LM_LOSS: 0.2904 | LB_LOSS: 1.0005 [2026-05-23 21:45:25] Validation | Batch 1200/1463 | Loss: 0.7907 | LM_LOSS: 0.2904 | LB_LOSS: 1.0005 [2026-05-23 21:45:28] Validation | Batch 1210/1463 | Loss: 0.7909 | LM_LOSS: 0.2906 | LB_LOSS: 1.0005 [2026-05-23 21:45:31] Validation | Batch 1220/1463 | Loss: 0.7912 | LM_LOSS: 0.2909 | LB_LOSS: 1.0005 [2026-05-23 21:45:34] Validation | Batch 1230/1463 | Loss: 0.7912 | LM_LOSS: 0.2909 | LB_LOSS: 1.0005 [2026-05-23 21:45:37] Validation | Batch 1240/1463 | Loss: 0.7914 | LM_LOSS: 0.2911 | LB_LOSS: 1.0005 [2026-05-23 21:45:40] Validation | Batch 1250/1463 | Loss: 0.7914 | LM_LOSS: 0.2912 | LB_LOSS: 1.0005 [2026-05-23 21:45:43] Validation | Batch 1260/1463 | Loss: 0.7914 | LM_LOSS: 0.2911 | LB_LOSS: 1.0005 [2026-05-23 21:45:46] Validation | Batch 1270/1463 | Loss: 0.7911 | LM_LOSS: 0.2909 | LB_LOSS: 1.0005 [2026-05-23 21:45:50] Validation | Batch 1280/1463 | Loss: 0.7910 | LM_LOSS: 0.2908 | LB_LOSS: 1.0005 [2026-05-23 21:45:53] Validation | Batch 1290/1463 | Loss: 0.7910 | LM_LOSS: 0.2907 | LB_LOSS: 1.0005 [2026-05-23 21:45:56] Validation | Batch 1300/1463 | Loss: 0.7908 | LM_LOSS: 0.2905 | LB_LOSS: 1.0005 [2026-05-23 21:45:59] Validation | Batch 1310/1463 | Loss: 0.7907 | LM_LOSS: 0.2904 | LB_LOSS: 1.0005 [2026-05-23 21:46:01] Validation | Batch 1320/1463 | Loss: 0.7907 | LM_LOSS: 0.2905 | LB_LOSS: 1.0005 [2026-05-23 21:46:04] Validation | Batch 1330/1463 | Loss: 0.7908 | LM_LOSS: 0.2905 | LB_LOSS: 1.0005 [2026-05-23 21:46:06] Validation | Batch 1340/1463 | Loss: 0.7910 | LM_LOSS: 0.2908 | LB_LOSS: 1.0005 [2026-05-23 21:46:10] Validation | Batch 1350/1463 | Loss: 0.7910 | LM_LOSS: 0.2907 | LB_LOSS: 1.0005 [2026-05-23 21:46:14] Validation | Batch 1360/1463 | Loss: 0.7912 | LM_LOSS: 0.2910 | LB_LOSS: 1.0005 [2026-05-23 21:46:17] Validation | Batch 1370/1463 | Loss: 0.7915 | LM_LOSS: 0.2913 | LB_LOSS: 1.0005 [2026-05-23 21:46:19] Validation | Batch 1380/1463 | Loss: 0.7913 | LM_LOSS: 0.2910 | LB_LOSS: 1.0005 [2026-05-23 21:46:22] Validation | Batch 1390/1463 | Loss: 0.7912 | LM_LOSS: 0.2910 | LB_LOSS: 1.0005 [2026-05-23 21:46:25] Validation | Batch 1400/1463 | Loss: 0.7915 | LM_LOSS: 0.2913 | LB_LOSS: 1.0005 [2026-05-23 21:46:28] Validation | Batch 1410/1463 | Loss: 0.7918 | LM_LOSS: 0.2916 | LB_LOSS: 1.0005 [2026-05-23 21:46:31] Validation | Batch 1420/1463 | Loss: 0.7918 | LM_LOSS: 0.2915 | LB_LOSS: 1.0005 [2026-05-23 21:46:34] Validation | Batch 1430/1463 | Loss: 0.7918 | LM_LOSS: 0.2916 | LB_LOSS: 1.0005 [2026-05-23 21:46:37] Validation | Batch 1440/1463 | Loss: 0.7922 | LM_LOSS: 0.2920 | LB_LOSS: 1.0005 [2026-05-23 21:46:40] Validation | Batch 1450/1463 | Loss: 0.7920 | LM_LOSS: 0.2917 | LB_LOSS: 1.0005 [2026-05-23 21:46:44] Validation | Batch 1460/1463 | Loss: 0.7923 | LM_LOSS: 0.2920 | LB_LOSS: 1.0005 [2026-05-23 21:46:45] Validation | Batch 1463/1463 | Loss: 0.7922 | LM_LOSS: 0.2919 | LB_LOSS: 1.0005 [2026-05-23 21:46:45] Validation | Loss: 0.7922 | LM_LOSS: 0.2919 | LB_LOSS: 1.0005 | PPL: 1.34 | Time: 442.24s [2026-05-23 21:46:57] Epoch 2 | Step 9010 | Loss: 0.7191 | LM: 0.2198 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.391 | LR: 1.00e-04 [2026-05-23 21:47:09] Epoch 2 | Step 9020 | Loss: 0.7190 | LM: 0.2197 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.391 | LR: 1.00e-04 [2026-05-23 21:47:22] Epoch 2 | Step 9030 | Loss: 0.7190 | LM: 0.2197 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.391 | LR: 1.00e-04 [2026-05-23 21:47:34] Epoch 2 | Step 9040 | Loss: 0.7190 | LM: 0.2197 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.391 | LR: 1.00e-04 [2026-05-23 21:47:46] Epoch 2 | Step 9050 | Loss: 0.7190 | LM: 0.2197 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.391 | LR: 1.00e-04 [2026-05-23 21:47:59] Epoch 2 | Step 9060 | Loss: 0.7190 | LM: 0.2198 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.391 | LR: 1.00e-04 [2026-05-23 21:48:11] Epoch 2 | Step 9070 | Loss: 0.7190 | LM: 0.2197 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.391 | LR: 1.00e-04 [2026-05-23 21:48:23] Epoch 2 | Step 9080 | Loss: 0.7190 | LM: 0.2197 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.391 | LR: 1.00e-04 [2026-05-23 21:48:35] Epoch 2 | Step 9090 | Loss: 0.7190 | LM: 0.2197 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.391 | LR: 1.00e-04 [2026-05-23 21:48:47] Epoch 2 | Step 9100 | Loss: 0.7190 | LM: 0.2197 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.391 | LR: 1.00e-04 [2026-05-23 21:49:00] Epoch 2 | Step 9110 | Loss: 0.7190 | LM: 0.2197 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.391 | LR: 1.00e-04 [2026-05-23 21:49:12] Epoch 2 | Step 9120 | Loss: 0.7191 | LM: 0.2197 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.391 | LR: 1.00e-04 [2026-05-23 21:49:25] Epoch 2 | Step 9130 | Loss: 0.7190 | LM: 0.2196 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.391 | LR: 1.00e-04 [2026-05-23 21:49:39] Epoch 2 | Step 9140 | Loss: 0.7191 | LM: 0.2197 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.391 | LR: 1.00e-04 [2026-05-23 21:49:52] Epoch 2 | Step 9150 | Loss: 0.7191 | LM: 0.2198 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.391 | LR: 1.00e-04 [2026-05-23 21:50:04] Epoch 2 | Step 9160 | Loss: 0.7190 | LM: 0.2197 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.391 | LR: 1.00e-04 [2026-05-23 21:50:17] Epoch 2 | Step 9170 | Loss: 0.7190 | LM: 0.2197 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.391 | LR: 1.00e-04 [2026-05-23 21:50:28] Epoch 2 | Step 9180 | Loss: 0.7190 | LM: 0.2197 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.391 | LR: 1.00e-04 [2026-05-23 21:50:42] Epoch 2 | Step 9190 | Loss: 0.7190 | LM: 0.2198 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.391 | LR: 1.00e-04 [2026-05-23 21:50:54] Epoch 2 | Step 9200 | Loss: 0.7190 | LM: 0.2198 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.391 | LR: 1.00e-04 [2026-05-23 21:51:07] Epoch 2 | Step 9210 | Loss: 0.7190 | LM: 0.2197 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.391 | LR: 1.00e-04 [2026-05-23 21:51:19] Epoch 2 | Step 9220 | Loss: 0.7190 | LM: 0.2197 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.391 | LR: 1.00e-04 [2026-05-23 21:51:32] Epoch 2 | Step 9230 | Loss: 0.7191 | LM: 0.2197 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.391 | LR: 1.00e-04 [2026-05-23 21:51:45] Epoch 2 | Step 9240 | Loss: 0.7191 | LM: 0.2197 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.391 | LR: 1.00e-04 [2026-05-23 21:51:57] Epoch 2 | Step 9250 | Loss: 0.7191 | LM: 0.2196 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.391 | LR: 1.00e-04 [2026-05-23 21:52:10] Epoch 2 | Step 9260 | Loss: 0.7191 | LM: 0.2197 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.391 | LR: 1.00e-04 [2026-05-23 21:52:22] Epoch 2 | Step 9270 | Loss: 0.7191 | LM: 0.2196 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.391 | LR: 1.00e-04 [2026-05-23 21:52:34] Epoch 2 | Step 9280 | Loss: 0.7191 | LM: 0.2197 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.391 | LR: 1.00e-04 [2026-05-23 21:52:46] Epoch 2 | Step 9290 | Loss: 0.7192 | LM: 0.2197 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.391 | LR: 1.00e-04 [2026-05-23 21:53:00] Epoch 2 | Step 9300 | Loss: 0.7192 | LM: 0.2197 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.391 | LR: 1.00e-04 [2026-05-23 21:53:12] Epoch 2 | Step 9310 | Loss: 0.7193 | LM: 0.2197 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.391 | LR: 1.00e-04 [2026-05-23 21:53:25] Epoch 2 | Step 9320 | Loss: 0.7193 | LM: 0.2198 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.391 | LR: 1.00e-04 [2026-05-23 21:53:37] Epoch 2 | Step 9330 | Loss: 0.7193 | LM: 0.2199 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.391 | LR: 1.00e-04 [2026-05-23 21:53:50] Epoch 2 | Step 9340 | Loss: 0.7194 | LM: 0.2198 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.391 | LR: 1.00e-04 [2026-05-23 21:54:01] Epoch 2 | Step 9350 | Loss: 0.7194 | LM: 0.2198 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.391 | LR: 1.00e-04 [2026-05-23 21:54:13] Epoch 2 | Step 9360 | Loss: 0.7194 | LM: 0.2198 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.391 | LR: 1.00e-04 [2026-05-23 21:54:27] Epoch 2 | Step 9370 | Loss: 0.7194 | LM: 0.2199 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.400/SR1: 0.391 | LR: 1.00e-04 [2026-05-23 21:54:32] Epoch 2 completed in 8035.21s | Loss: 0.7194 | CL0: 2.5 | CL1: 2.5 [2026-05-23 21:54:32] ============================================================ [2026-05-23 21:54:32] EPOCH 3/3 (step 9374) [2026-05-23 21:54:32] ============================================================ [2026-05-23 21:54:39] Epoch 3 | Step 9380 | Loss: 0.6458 | LM: 0.1401 | LB: 1.0006 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.438 | HR1: 0.399/SR1: 0.396 | LR: 1.00e-04 [2026-05-23 21:54:51] Epoch 3 | Step 9390 | Loss: 0.6502 | LM: 0.1350 | LB: 1.0009 | CL0: 2.5 | CL1: 2.5 | HR0: 0.397/SR0: 0.434 | HR1: 0.396/SR1: 0.390 | LR: 1.00e-04 [2026-05-23 21:55:03] Epoch 3 | Step 9400 | Loss: 0.6440 | LM: 0.1351 | LB: 1.0004 | CL0: 2.5 | CL1: 2.5 | HR0: 0.393/SR0: 0.433 | HR1: 0.401/SR1: 0.394 | LR: 1.00e-04 [2026-05-23 21:55:16] Epoch 3 | Step 9410 | Loss: 0.6416 | LM: 0.1280 | LB: 1.0006 | CL0: 2.5 | CL1: 2.5 | HR0: 0.395/SR0: 0.435 | HR1: 0.398/SR1: 0.392 | LR: 1.00e-04 [2026-05-23 21:55:28] Epoch 3 | Step 9420 | Loss: 0.6312 | LM: 0.1145 | LB: 1.0009 | CL0: 2.5 | CL1: 2.5 | HR0: 0.396/SR0: 0.436 | HR1: 0.396/SR1: 0.390 | LR: 1.00e-04 [2026-05-23 21:55:40] Epoch 3 | Step 9430 | Loss: 0.6296 | LM: 0.1178 | LB: 1.0010 | CL0: 2.5 | CL1: 2.5 | HR0: 0.398/SR0: 0.438 | HR1: 0.397/SR1: 0.390 | LR: 1.00e-04 [2026-05-23 21:55:53] Epoch 3 | Step 9440 | Loss: 0.6268 | LM: 0.1128 | LB: 1.0011 | CL0: 2.5 | CL1: 2.5 | HR0: 0.399/SR0: 0.438 | HR1: 0.396/SR1: 0.390 | LR: 1.00e-04 [2026-05-23 21:56:05] Epoch 3 | Step 9450 | Loss: 0.6305 | LM: 0.1179 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.439 | HR1: 0.398/SR1: 0.391 | LR: 1.00e-04 [2026-05-23 21:56:18] Epoch 3 | Step 9460 | Loss: 0.6291 | LM: 0.1159 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.438 | HR1: 0.397/SR1: 0.391 | LR: 1.00e-04 [2026-05-23 21:56:31] Epoch 3 | Step 9470 | Loss: 0.6276 | LM: 0.1166 | LB: 1.0013 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.439 | HR1: 0.398/SR1: 0.391 | LR: 1.00e-04 [2026-05-23 21:56:43] Epoch 3 | Step 9480 | Loss: 0.6272 | LM: 0.1217 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.439 | HR1: 0.398/SR1: 0.392 | LR: 1.00e-04 [2026-05-23 21:56:55] Epoch 3 | Step 9490 | Loss: 0.6267 | LM: 0.1203 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.439 | HR1: 0.398/SR1: 0.391 | LR: 1.00e-04 [2026-05-23 21:57:07] Epoch 3 | Step 9500 | Loss: 0.6258 | LM: 0.1204 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.439 | HR1: 0.398/SR1: 0.391 | LR: 1.00e-04 [2026-05-23 21:57:19] Epoch 3 | Step 9510 | Loss: 0.6248 | LM: 0.1176 | LB: 1.0014 | CL0: 2.5 | CL1: 2.5 | HR0: 0.402/SR0: 0.440 | HR1: 0.398/SR1: 0.391 | LR: 1.00e-04 [2026-05-23 21:57:32] Epoch 3 | Step 9520 | Loss: 0.6248 | LM: 0.1175 | LB: 1.0014 | CL0: 2.5 | CL1: 2.5 | HR0: 0.402/SR0: 0.441 | HR1: 0.398/SR1: 0.392 | LR: 1.00e-04 [2026-05-23 21:57:44] Epoch 3 | Step 9530 | Loss: 0.6257 | LM: 0.1202 | LB: 1.0015 | CL0: 2.5 | CL1: 2.5 | HR0: 0.402/SR0: 0.441 | HR1: 0.398/SR1: 0.391 | LR: 1.00e-04 [2026-05-23 21:57:56] Epoch 3 | Step 9540 | Loss: 0.6250 | LM: 0.1195 | LB: 1.0015 | CL0: 2.5 | CL1: 2.5 | HR0: 0.402/SR0: 0.440 | HR1: 0.398/SR1: 0.391 | LR: 1.00e-04 [2026-05-23 21:58:09] Epoch 3 | Step 9550 | Loss: 0.6250 | LM: 0.1184 | LB: 1.0014 | CL0: 2.5 | CL1: 2.5 | HR0: 0.402/SR0: 0.441 | HR1: 0.398/SR1: 0.391 | LR: 1.00e-04 [2026-05-23 21:58:21] Epoch 3 | Step 9560 | Loss: 0.6239 | LM: 0.1177 | LB: 1.0014 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.440 | HR1: 0.399/SR1: 0.392 | LR: 1.00e-04 [2026-05-23 21:58:34] Epoch 3 | Step 9570 | Loss: 0.6248 | LM: 0.1231 | LB: 1.0014 | CL0: 2.5 | CL1: 2.5 | HR0: 0.402/SR0: 0.440 | HR1: 0.399/SR1: 0.392 | LR: 1.00e-04 [2026-05-23 21:58:46] Epoch 3 | Step 9580 | Loss: 0.6245 | LM: 0.1228 | LB: 1.0014 | CL0: 2.5 | CL1: 2.5 | HR0: 0.402/SR0: 0.440 | HR1: 0.400/SR1: 0.392 | LR: 1.00e-04 [2026-05-23 21:58:58] Epoch 3 | Step 9590 | Loss: 0.6237 | LM: 0.1221 | LB: 1.0014 | CL0: 2.5 | CL1: 2.5 | HR0: 0.402/SR0: 0.440 | HR1: 0.400/SR1: 0.393 | LR: 1.00e-04 [2026-05-23 21:59:09] Epoch 3 | Step 9600 | Loss: 0.6239 | LM: 0.1210 | LB: 1.0014 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.440 | HR1: 0.400/SR1: 0.393 | LR: 1.00e-04 [2026-05-23 21:59:22] Epoch 3 | Step 9610 | Loss: 0.6234 | LM: 0.1206 | LB: 1.0013 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.440 | HR1: 0.400/SR1: 0.393 | LR: 1.00e-04 [2026-05-23 21:59:34] Epoch 3 | Step 9620 | Loss: 0.6224 | LM: 0.1221 | LB: 1.0013 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.440 | HR1: 0.400/SR1: 0.393 | LR: 1.00e-04 [2026-05-23 21:59:46] Epoch 3 | Step 9630 | Loss: 0.6225 | LM: 0.1220 | LB: 1.0013 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.440 | HR1: 0.400/SR1: 0.393 | LR: 1.00e-04 [2026-05-23 21:59:58] Epoch 3 | Step 9640 | Loss: 0.6225 | LM: 0.1223 | LB: 1.0014 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.440 | HR1: 0.399/SR1: 0.393 | LR: 1.00e-04 [2026-05-23 22:00:11] Epoch 3 | Step 9650 | Loss: 0.6220 | LM: 0.1216 | LB: 1.0013 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.440 | HR1: 0.400/SR1: 0.393 | LR: 1.00e-04 [2026-05-23 22:00:23] Epoch 3 | Step 9660 | Loss: 0.6217 | LM: 0.1214 | LB: 1.0013 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.440 | HR1: 0.399/SR1: 0.393 | LR: 1.00e-04 [2026-05-23 22:00:35] Epoch 3 | Step 9670 | Loss: 0.6215 | LM: 0.1210 | LB: 1.0013 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.440 | HR1: 0.399/SR1: 0.393 | LR: 1.00e-04 [2026-05-23 22:00:46] Epoch 3 | Step 9680 | Loss: 0.6213 | LM: 0.1211 | LB: 1.0013 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.440 | HR1: 0.400/SR1: 0.394 | LR: 1.00e-04 [2026-05-23 22:00:58] Epoch 3 | Step 9690 | Loss: 0.6212 | LM: 0.1206 | LB: 1.0013 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.440 | HR1: 0.400/SR1: 0.393 | LR: 1.00e-04 [2026-05-23 22:01:10] Epoch 3 | Step 9700 | Loss: 0.6220 | LM: 0.1211 | LB: 1.0013 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.439 | HR1: 0.400/SR1: 0.393 | LR: 1.00e-04 [2026-05-23 22:01:22] Epoch 3 | Step 9710 | Loss: 0.6219 | LM: 0.1210 | LB: 1.0013 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.439 | HR1: 0.400/SR1: 0.394 | LR: 1.00e-04 [2026-05-23 22:01:35] Epoch 3 | Step 9720 | Loss: 0.6220 | LM: 0.1203 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.439 | HR1: 0.400/SR1: 0.394 | LR: 1.00e-04 [2026-05-23 22:01:46] Epoch 3 | Step 9730 | Loss: 0.6220 | LM: 0.1210 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.439 | HR1: 0.400/SR1: 0.394 | LR: 1.00e-04 [2026-05-23 22:01:59] Epoch 3 | Step 9740 | Loss: 0.6218 | LM: 0.1217 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.439 | HR1: 0.400/SR1: 0.394 | LR: 1.00e-04 [2026-05-23 22:02:11] Epoch 3 | Step 9750 | Loss: 0.6216 | LM: 0.1210 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.439 | HR1: 0.400/SR1: 0.394 | LR: 1.00e-04 [2026-05-23 22:02:23] Epoch 3 | Step 9760 | Loss: 0.6217 | LM: 0.1210 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.439 | HR1: 0.400/SR1: 0.394 | LR: 1.00e-04 [2026-05-23 22:02:36] Epoch 3 | Step 9770 | Loss: 0.6220 | LM: 0.1220 | LB: 1.0013 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.439 | HR1: 0.400/SR1: 0.394 | LR: 1.00e-04 [2026-05-23 22:02:47] Epoch 3 | Step 9780 | Loss: 0.6222 | LM: 0.1218 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.439 | HR1: 0.400/SR1: 0.394 | LR: 1.00e-04 [2026-05-23 22:02:59] Epoch 3 | Step 9790 | Loss: 0.6219 | LM: 0.1225 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.439 | HR1: 0.400/SR1: 0.394 | LR: 1.00e-04 [2026-05-23 22:03:12] Epoch 3 | Step 9800 | Loss: 0.6218 | LM: 0.1223 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.439 | HR1: 0.400/SR1: 0.394 | LR: 1.00e-04 [2026-05-23 22:03:24] Epoch 3 | Step 9810 | Loss: 0.6217 | LM: 0.1221 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.439 | HR1: 0.400/SR1: 0.394 | LR: 1.00e-04 [2026-05-23 22:03:37] Epoch 3 | Step 9820 | Loss: 0.6216 | LM: 0.1221 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.439 | HR1: 0.400/SR1: 0.394 | LR: 1.00e-04 [2026-05-23 22:03:49] Epoch 3 | Step 9830 | Loss: 0.6219 | LM: 0.1227 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.439 | HR1: 0.400/SR1: 0.394 | LR: 1.00e-04 [2026-05-23 22:04:00] Epoch 3 | Step 9840 | Loss: 0.6220 | LM: 0.1224 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.439 | HR1: 0.400/SR1: 0.394 | LR: 1.00e-04 [2026-05-23 22:04:12] Epoch 3 | Step 9850 | Loss: 0.6220 | LM: 0.1218 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.439 | HR1: 0.400/SR1: 0.395 | LR: 1.00e-04 [2026-05-23 22:04:23] Epoch 3 | Step 9860 | Loss: 0.6222 | LM: 0.1214 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.439 | HR1: 0.400/SR1: 0.394 | LR: 1.00e-04 [2026-05-23 22:04:36] Epoch 3 | Step 9870 | Loss: 0.6222 | LM: 0.1210 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.439 | HR1: 0.400/SR1: 0.394 | LR: 1.00e-04 [2026-05-23 22:04:49] Epoch 3 | Step 9880 | Loss: 0.6222 | LM: 0.1221 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.439 | HR1: 0.400/SR1: 0.394 | LR: 1.00e-04 [2026-05-23 22:05:01] Epoch 3 | Step 9890 | Loss: 0.6219 | LM: 0.1216 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.439 | HR1: 0.400/SR1: 0.394 | LR: 1.00e-04 [2026-05-23 22:05:13] Epoch 3 | Step 9900 | Loss: 0.6216 | LM: 0.1215 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.439 | HR1: 0.400/SR1: 0.394 | LR: 1.00e-04 [2026-05-23 22:05:25] Epoch 3 | Step 9910 | Loss: 0.6217 | LM: 0.1221 | LB: 1.0011 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.439 | HR1: 0.400/SR1: 0.394 | LR: 1.00e-04 [2026-05-23 22:05:38] Epoch 3 | Step 9920 | Loss: 0.6215 | LM: 0.1221 | LB: 1.0011 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.439 | HR1: 0.400/SR1: 0.394 | LR: 1.00e-04 [2026-05-23 22:05:50] Epoch 3 | Step 9930 | Loss: 0.6212 | LM: 0.1214 | LB: 1.0011 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.439 | HR1: 0.400/SR1: 0.394 | LR: 1.00e-04 [2026-05-23 22:06:01] Epoch 3 | Step 9940 | Loss: 0.6214 | LM: 0.1216 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.439 | HR1: 0.400/SR1: 0.394 | LR: 1.00e-04 [2026-05-23 22:06:13] Epoch 3 | Step 9950 | Loss: 0.6219 | LM: 0.1224 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.439 | HR1: 0.400/SR1: 0.394 | LR: 1.00e-04 [2026-05-23 22:06:25] Epoch 3 | Step 9960 | Loss: 0.6218 | LM: 0.1224 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.439 | HR1: 0.400/SR1: 0.394 | LR: 1.00e-04 [2026-05-23 22:06:38] Epoch 3 | Step 9970 | Loss: 0.6217 | LM: 0.1219 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.439 | HR1: 0.400/SR1: 0.394 | LR: 1.00e-04 [2026-05-23 22:06:51] Epoch 3 | Step 9980 | Loss: 0.6215 | LM: 0.1219 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.439 | HR1: 0.400/SR1: 0.394 | LR: 1.00e-04 [2026-05-23 22:07:04] Epoch 3 | Step 9990 | Loss: 0.6217 | LM: 0.1220 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.439 | HR1: 0.400/SR1: 0.394 | LR: 1.00e-04 [2026-05-23 22:07:16] Epoch 3 | Step 10000 | Loss: 0.6215 | LM: 0.1217 | LB: 1.0011 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.439 | HR1: 0.400/SR1: 0.394 | LR: 1.00e-04 [2026-05-23 22:07:18] Validation | Batch 10/1463 | Loss: 0.8259 | LM_LOSS: 0.3261 | LB_LOSS: 0.9997 [2026-05-23 22:07:20] Validation | Batch 20/1463 | Loss: 0.8365 | LM_LOSS: 0.3364 | LB_LOSS: 1.0001 [2026-05-23 22:07:24] Validation | Batch 30/1463 | Loss: 0.8342 | LM_LOSS: 0.3340 | LB_LOSS: 1.0005 [2026-05-23 22:07:27] Validation | Batch 40/1463 | Loss: 0.8413 | LM_LOSS: 0.3411 | LB_LOSS: 1.0004 [2026-05-23 22:07:30] Validation | Batch 50/1463 | Loss: 0.8506 | LM_LOSS: 0.3505 | LB_LOSS: 1.0003 [2026-05-23 22:07:34] Validation | Batch 60/1463 | Loss: 0.8518 | LM_LOSS: 0.3516 | LB_LOSS: 1.0004 [2026-05-23 22:07:38] Validation | Batch 70/1463 | Loss: 0.8526 | LM_LOSS: 0.3524 | LB_LOSS: 1.0004 [2026-05-23 22:07:41] Validation | Batch 80/1463 | Loss: 0.8554 | LM_LOSS: 0.3551 | LB_LOSS: 1.0004 [2026-05-23 22:07:44] Validation | Batch 90/1463 | Loss: 0.8553 | LM_LOSS: 0.3551 | LB_LOSS: 1.0004 [2026-05-23 22:07:47] Validation | Batch 100/1463 | Loss: 0.8546 | LM_LOSS: 0.3544 | LB_LOSS: 1.0004 [2026-05-23 22:07:50] Validation | Batch 110/1463 | Loss: 0.8518 | LM_LOSS: 0.3516 | LB_LOSS: 1.0004 [2026-05-23 22:07:52] Validation | Batch 120/1463 | Loss: 0.8546 | LM_LOSS: 0.3544 | LB_LOSS: 1.0004 [2026-05-23 22:07:55] Validation | Batch 130/1463 | Loss: 0.8563 | LM_LOSS: 0.3561 | LB_LOSS: 1.0004 [2026-05-23 22:07:59] Validation | Batch 140/1463 | Loss: 0.8580 | LM_LOSS: 0.3578 | LB_LOSS: 1.0004 [2026-05-23 22:08:02] Validation | Batch 150/1463 | Loss: 0.8550 | LM_LOSS: 0.3548 | LB_LOSS: 1.0005 [2026-05-23 22:08:05] Validation | Batch 160/1463 | Loss: 0.8554 | LM_LOSS: 0.3552 | LB_LOSS: 1.0005 [2026-05-23 22:08:08] Validation | Batch 170/1463 | Loss: 0.8542 | LM_LOSS: 0.3540 | LB_LOSS: 1.0005 [2026-05-23 22:08:12] Validation | Batch 180/1463 | Loss: 0.8553 | LM_LOSS: 0.3550 | LB_LOSS: 1.0005 [2026-05-23 22:08:15] Validation | Batch 190/1463 | Loss: 0.8568 | LM_LOSS: 0.3566 | LB_LOSS: 1.0005 [2026-05-23 22:08:18] Validation | Batch 200/1463 | Loss: 0.8567 | LM_LOSS: 0.3565 | LB_LOSS: 1.0005 [2026-05-23 22:08:21] Validation | Batch 210/1463 | Loss: 0.8578 | LM_LOSS: 0.3576 | LB_LOSS: 1.0005 [2026-05-23 22:08:24] Validation | Batch 220/1463 | Loss: 0.8572 | LM_LOSS: 0.3569 | LB_LOSS: 1.0005 [2026-05-23 22:08:28] Validation | Batch 230/1463 | Loss: 0.8584 | LM_LOSS: 0.3581 | LB_LOSS: 1.0005 [2026-05-23 22:08:31] Validation | Batch 240/1463 | Loss: 0.8603 | LM_LOSS: 0.3601 | LB_LOSS: 1.0005 [2026-05-23 22:08:34] Validation | Batch 250/1463 | Loss: 0.8591 | LM_LOSS: 0.3589 | LB_LOSS: 1.0005 [2026-05-23 22:08:37] Validation | Batch 260/1463 | Loss: 0.8612 | LM_LOSS: 0.3609 | LB_LOSS: 1.0005 [2026-05-23 22:08:40] Validation | Batch 270/1463 | Loss: 0.8595 | LM_LOSS: 0.3593 | LB_LOSS: 1.0005 [2026-05-23 22:08:43] Validation | Batch 280/1463 | Loss: 0.8597 | LM_LOSS: 0.3595 | LB_LOSS: 1.0005 [2026-05-23 22:08:46] Validation | Batch 290/1463 | Loss: 0.8581 | LM_LOSS: 0.3579 | LB_LOSS: 1.0005 [2026-05-23 22:08:49] Validation | Batch 300/1463 | Loss: 0.8589 | LM_LOSS: 0.3586 | LB_LOSS: 1.0005 [2026-05-23 22:08:52] Validation | Batch 310/1463 | Loss: 0.8599 | LM_LOSS: 0.3596 | LB_LOSS: 1.0005 [2026-05-23 22:08:55] Validation | Batch 320/1463 | Loss: 0.8586 | LM_LOSS: 0.3584 | LB_LOSS: 1.0005 [2026-05-23 22:08:58] Validation | Batch 330/1463 | Loss: 0.8580 | LM_LOSS: 0.3577 | LB_LOSS: 1.0005 [2026-05-23 22:09:01] Validation | Batch 340/1463 | Loss: 0.8576 | LM_LOSS: 0.3573 | LB_LOSS: 1.0005 [2026-05-23 22:09:04] Validation | Batch 350/1463 | Loss: 0.8583 | LM_LOSS: 0.3581 | LB_LOSS: 1.0005 [2026-05-23 22:09:07] Validation | Batch 360/1463 | Loss: 0.8583 | LM_LOSS: 0.3581 | LB_LOSS: 1.0005 [2026-05-23 22:09:09] Validation | Batch 370/1463 | Loss: 0.8583 | LM_LOSS: 0.3580 | LB_LOSS: 1.0005 [2026-05-23 22:09:13] Validation | Batch 380/1463 | Loss: 0.8577 | LM_LOSS: 0.3575 | LB_LOSS: 1.0005 [2026-05-23 22:09:16] Validation | Batch 390/1463 | Loss: 0.8585 | LM_LOSS: 0.3582 | LB_LOSS: 1.0005 [2026-05-23 22:09:18] Validation | Batch 400/1463 | Loss: 0.8584 | LM_LOSS: 0.3581 | LB_LOSS: 1.0005 [2026-05-23 22:09:21] Validation | Batch 410/1463 | Loss: 0.8580 | LM_LOSS: 0.3577 | LB_LOSS: 1.0005 [2026-05-23 22:09:25] Validation | Batch 420/1463 | Loss: 0.8581 | LM_LOSS: 0.3579 | LB_LOSS: 1.0005 [2026-05-23 22:09:28] Validation | Batch 430/1463 | Loss: 0.8576 | LM_LOSS: 0.3573 | LB_LOSS: 1.0005 [2026-05-23 22:09:31] Validation | Batch 440/1463 | Loss: 0.8579 | LM_LOSS: 0.3576 | LB_LOSS: 1.0005 [2026-05-23 22:09:33] Validation | Batch 450/1463 | Loss: 0.8574 | LM_LOSS: 0.3571 | LB_LOSS: 1.0006 [2026-05-23 22:09:37] Validation | Batch 460/1463 | Loss: 0.8573 | LM_LOSS: 0.3570 | LB_LOSS: 1.0005 [2026-05-23 22:09:40] Validation | Batch 470/1463 | Loss: 0.8581 | LM_LOSS: 0.3579 | LB_LOSS: 1.0005 [2026-05-23 22:09:43] Validation | Batch 480/1463 | Loss: 0.8577 | LM_LOSS: 0.3575 | LB_LOSS: 1.0005 [2026-05-23 22:09:46] Validation | Batch 490/1463 | Loss: 0.8576 | LM_LOSS: 0.3574 | LB_LOSS: 1.0005 [2026-05-23 22:09:50] Validation | Batch 500/1463 | Loss: 0.8576 | LM_LOSS: 0.3574 | LB_LOSS: 1.0005 [2026-05-23 22:09:53] Validation | Batch 510/1463 | Loss: 0.8567 | LM_LOSS: 0.3564 | LB_LOSS: 1.0005 [2026-05-23 22:09:56] Validation | Batch 520/1463 | Loss: 0.8565 | LM_LOSS: 0.3562 | LB_LOSS: 1.0005 [2026-05-23 22:09:59] Validation | Batch 530/1463 | Loss: 0.8572 | LM_LOSS: 0.3569 | LB_LOSS: 1.0005 [2026-05-23 22:10:01] Validation | Batch 540/1463 | Loss: 0.8574 | LM_LOSS: 0.3571 | LB_LOSS: 1.0005 [2026-05-23 22:10:03] Validation | Batch 550/1463 | Loss: 0.8569 | LM_LOSS: 0.3567 | LB_LOSS: 1.0005 [2026-05-23 22:10:07] Validation | Batch 560/1463 | Loss: 0.8569 | LM_LOSS: 0.3567 | LB_LOSS: 1.0005 [2026-05-23 22:10:10] Validation | Batch 570/1463 | Loss: 0.8568 | LM_LOSS: 0.3566 | LB_LOSS: 1.0005 [2026-05-23 22:10:13] Validation | Batch 580/1463 | Loss: 0.8560 | LM_LOSS: 0.3557 | LB_LOSS: 1.0005 [2026-05-23 22:10:17] Validation | Batch 590/1463 | Loss: 0.8557 | LM_LOSS: 0.3554 | LB_LOSS: 1.0005 [2026-05-23 22:10:20] Validation | Batch 600/1463 | Loss: 0.8560 | LM_LOSS: 0.3557 | LB_LOSS: 1.0005 [2026-05-23 22:10:24] Validation | Batch 610/1463 | Loss: 0.8563 | LM_LOSS: 0.3560 | LB_LOSS: 1.0005 [2026-05-23 22:10:27] Validation | Batch 620/1463 | Loss: 0.8556 | LM_LOSS: 0.3554 | LB_LOSS: 1.0005 [2026-05-23 22:10:30] Validation | Batch 630/1463 | Loss: 0.8550 | LM_LOSS: 0.3547 | LB_LOSS: 1.0005 [2026-05-23 22:10:32] Validation | Batch 640/1463 | Loss: 0.8548 | LM_LOSS: 0.3545 | LB_LOSS: 1.0005 [2026-05-23 22:10:35] Validation | Batch 650/1463 | Loss: 0.8543 | LM_LOSS: 0.3540 | LB_LOSS: 1.0005 [2026-05-23 22:10:38] Validation | Batch 660/1463 | Loss: 0.8538 | LM_LOSS: 0.3535 | LB_LOSS: 1.0005 [2026-05-23 22:10:41] Validation | Batch 670/1463 | Loss: 0.8531 | LM_LOSS: 0.3529 | LB_LOSS: 1.0005 [2026-05-23 22:10:43] Validation | Batch 680/1463 | Loss: 0.8532 | LM_LOSS: 0.3529 | LB_LOSS: 1.0005 [2026-05-23 22:10:47] Validation | Batch 690/1463 | Loss: 0.8534 | LM_LOSS: 0.3531 | LB_LOSS: 1.0005 [2026-05-23 22:10:50] Validation | Batch 700/1463 | Loss: 0.8531 | LM_LOSS: 0.3529 | LB_LOSS: 1.0005 [2026-05-23 22:10:52] Validation | Batch 710/1463 | Loss: 0.8534 | LM_LOSS: 0.3531 | LB_LOSS: 1.0005 [2026-05-23 22:10:56] Validation | Batch 720/1463 | Loss: 0.8541 | LM_LOSS: 0.3539 | LB_LOSS: 1.0005 [2026-05-23 22:10:58] Validation | Batch 730/1463 | Loss: 0.8540 | LM_LOSS: 0.3537 | LB_LOSS: 1.0005 [2026-05-23 22:11:00] Validation | Batch 740/1463 | Loss: 0.8537 | LM_LOSS: 0.3534 | LB_LOSS: 1.0005 [2026-05-23 22:11:03] Validation | Batch 750/1463 | Loss: 0.8536 | LM_LOSS: 0.3534 | LB_LOSS: 1.0005 [2026-05-23 22:11:06] Validation | Batch 760/1463 | Loss: 0.8534 | LM_LOSS: 0.3532 | LB_LOSS: 1.0005 [2026-05-23 22:11:08] Validation | Batch 770/1463 | Loss: 0.8535 | LM_LOSS: 0.3533 | LB_LOSS: 1.0005 [2026-05-23 22:11:11] Validation | Batch 780/1463 | Loss: 0.8535 | LM_LOSS: 0.3532 | LB_LOSS: 1.0005 [2026-05-23 22:11:13] Validation | Batch 790/1463 | Loss: 0.8538 | LM_LOSS: 0.3536 | LB_LOSS: 1.0005 [2026-05-23 22:11:16] Validation | Batch 800/1463 | Loss: 0.8538 | LM_LOSS: 0.3535 | LB_LOSS: 1.0005 [2026-05-23 22:11:19] Validation | Batch 810/1463 | Loss: 0.8534 | LM_LOSS: 0.3532 | LB_LOSS: 1.0005 [2026-05-23 22:11:22] Validation | Batch 820/1463 | Loss: 0.8536 | LM_LOSS: 0.3534 | LB_LOSS: 1.0005 [2026-05-23 22:11:25] Validation | Batch 830/1463 | Loss: 0.8538 | LM_LOSS: 0.3536 | LB_LOSS: 1.0005 [2026-05-23 22:11:29] Validation | Batch 840/1463 | Loss: 0.8539 | LM_LOSS: 0.3536 | LB_LOSS: 1.0005 [2026-05-23 22:11:32] Validation | Batch 850/1463 | Loss: 0.8543 | LM_LOSS: 0.3540 | LB_LOSS: 1.0005 [2026-05-23 22:11:35] Validation | Batch 860/1463 | Loss: 0.8546 | LM_LOSS: 0.3544 | LB_LOSS: 1.0005 [2026-05-23 22:11:38] Validation | Batch 870/1463 | Loss: 0.8548 | LM_LOSS: 0.3545 | LB_LOSS: 1.0005 [2026-05-23 22:11:41] Validation | Batch 880/1463 | Loss: 0.8540 | LM_LOSS: 0.3537 | LB_LOSS: 1.0005 [2026-05-23 22:11:44] Validation | Batch 890/1463 | Loss: 0.8540 | LM_LOSS: 0.3537 | LB_LOSS: 1.0005 [2026-05-23 22:11:47] Validation | Batch 900/1463 | Loss: 0.8536 | LM_LOSS: 0.3533 | LB_LOSS: 1.0005 [2026-05-23 22:11:50] Validation | Batch 910/1463 | Loss: 0.8537 | LM_LOSS: 0.3534 | LB_LOSS: 1.0005 [2026-05-23 22:11:54] Validation | Batch 920/1463 | Loss: 0.8540 | LM_LOSS: 0.3537 | LB_LOSS: 1.0005 [2026-05-23 22:11:56] Validation | Batch 930/1463 | Loss: 0.8537 | LM_LOSS: 0.3535 | LB_LOSS: 1.0005 [2026-05-23 22:11:59] Validation | Batch 940/1463 | Loss: 0.8535 | LM_LOSS: 0.3532 | LB_LOSS: 1.0005 [2026-05-23 22:12:03] Validation | Batch 950/1463 | Loss: 0.8537 | LM_LOSS: 0.3535 | LB_LOSS: 1.0005 [2026-05-23 22:12:06] Validation | Batch 960/1463 | Loss: 0.8538 | LM_LOSS: 0.3535 | LB_LOSS: 1.0005 [2026-05-23 22:12:09] Validation | Batch 970/1463 | Loss: 0.8538 | LM_LOSS: 0.3535 | LB_LOSS: 1.0005 [2026-05-23 22:12:13] Validation | Batch 980/1463 | Loss: 0.8541 | LM_LOSS: 0.3539 | LB_LOSS: 1.0005 [2026-05-23 22:12:15] Validation | Batch 990/1463 | Loss: 0.8544 | LM_LOSS: 0.3542 | LB_LOSS: 1.0005 [2026-05-23 22:12:18] Validation | Batch 1000/1463 | Loss: 0.8551 | LM_LOSS: 0.3548 | LB_LOSS: 1.0005 [2026-05-23 22:12:22] Validation | Batch 1010/1463 | Loss: 0.8550 | LM_LOSS: 0.3548 | LB_LOSS: 1.0005 [2026-05-23 22:12:25] Validation | Batch 1020/1463 | Loss: 0.8551 | LM_LOSS: 0.3548 | LB_LOSS: 1.0005 [2026-05-23 22:12:28] Validation | Batch 1030/1463 | Loss: 0.8556 | LM_LOSS: 0.3553 | LB_LOSS: 1.0005 [2026-05-23 22:12:30] Validation | Batch 1040/1463 | Loss: 0.8552 | LM_LOSS: 0.3549 | LB_LOSS: 1.0005 [2026-05-23 22:12:34] Validation | Batch 1050/1463 | Loss: 0.8551 | LM_LOSS: 0.3549 | LB_LOSS: 1.0005 [2026-05-23 22:12:37] Validation | Batch 1060/1463 | Loss: 0.8551 | LM_LOSS: 0.3549 | LB_LOSS: 1.0005 [2026-05-23 22:12:40] Validation | Batch 1070/1463 | Loss: 0.8552 | LM_LOSS: 0.3549 | LB_LOSS: 1.0005 [2026-05-23 22:12:43] Validation | Batch 1080/1463 | Loss: 0.8550 | LM_LOSS: 0.3547 | LB_LOSS: 1.0005 [2026-05-23 22:12:46] Validation | Batch 1090/1463 | Loss: 0.8549 | LM_LOSS: 0.3547 | LB_LOSS: 1.0005 [2026-05-23 22:12:48] Validation | Batch 1100/1463 | Loss: 0.8546 | LM_LOSS: 0.3544 | LB_LOSS: 1.0005 [2026-05-23 22:12:51] Validation | Batch 1110/1463 | Loss: 0.8547 | LM_LOSS: 0.3544 | LB_LOSS: 1.0005 [2026-05-23 22:12:54] Validation | Batch 1120/1463 | Loss: 0.8543 | LM_LOSS: 0.3541 | LB_LOSS: 1.0005 [2026-05-23 22:12:58] Validation | Batch 1130/1463 | Loss: 0.8541 | LM_LOSS: 0.3538 | LB_LOSS: 1.0005 [2026-05-23 22:13:01] Validation | Batch 1140/1463 | Loss: 0.8544 | LM_LOSS: 0.3541 | LB_LOSS: 1.0005 [2026-05-23 22:13:04] Validation | Batch 1150/1463 | Loss: 0.8542 | LM_LOSS: 0.3539 | LB_LOSS: 1.0005 [2026-05-23 22:13:07] Validation | Batch 1160/1463 | Loss: 0.8541 | LM_LOSS: 0.3538 | LB_LOSS: 1.0005 [2026-05-23 22:13:10] Validation | Batch 1170/1463 | Loss: 0.8541 | LM_LOSS: 0.3539 | LB_LOSS: 1.0005 [2026-05-23 22:13:13] Validation | Batch 1180/1463 | Loss: 0.8542 | LM_LOSS: 0.3539 | LB_LOSS: 1.0005 [2026-05-23 22:13:16] Validation | Batch 1190/1463 | Loss: 0.8544 | LM_LOSS: 0.3541 | LB_LOSS: 1.0005 [2026-05-23 22:13:20] Validation | Batch 1200/1463 | Loss: 0.8544 | LM_LOSS: 0.3541 | LB_LOSS: 1.0005 [2026-05-23 22:13:23] Validation | Batch 1210/1463 | Loss: 0.8547 | LM_LOSS: 0.3545 | LB_LOSS: 1.0005 [2026-05-23 22:13:25] Validation | Batch 1220/1463 | Loss: 0.8551 | LM_LOSS: 0.3548 | LB_LOSS: 1.0005 [2026-05-23 22:13:28] Validation | Batch 1230/1463 | Loss: 0.8552 | LM_LOSS: 0.3549 | LB_LOSS: 1.0005 [2026-05-23 22:13:31] Validation | Batch 1240/1463 | Loss: 0.8553 | LM_LOSS: 0.3550 | LB_LOSS: 1.0005 [2026-05-23 22:13:35] Validation | Batch 1250/1463 | Loss: 0.8554 | LM_LOSS: 0.3551 | LB_LOSS: 1.0005 [2026-05-23 22:13:38] Validation | Batch 1260/1463 | Loss: 0.8553 | LM_LOSS: 0.3551 | LB_LOSS: 1.0005 [2026-05-23 22:13:41] Validation | Batch 1270/1463 | Loss: 0.8551 | LM_LOSS: 0.3548 | LB_LOSS: 1.0005 [2026-05-23 22:13:44] Validation | Batch 1280/1463 | Loss: 0.8550 | LM_LOSS: 0.3547 | LB_LOSS: 1.0005 [2026-05-23 22:13:47] Validation | Batch 1290/1463 | Loss: 0.8550 | LM_LOSS: 0.3547 | LB_LOSS: 1.0005 [2026-05-23 22:13:50] Validation | Batch 1300/1463 | Loss: 0.8547 | LM_LOSS: 0.3545 | LB_LOSS: 1.0005 [2026-05-23 22:13:53] Validation | Batch 1310/1463 | Loss: 0.8545 | LM_LOSS: 0.3543 | LB_LOSS: 1.0005 [2026-05-23 22:13:56] Validation | Batch 1320/1463 | Loss: 0.8546 | LM_LOSS: 0.3544 | LB_LOSS: 1.0005 [2026-05-23 22:13:58] Validation | Batch 1330/1463 | Loss: 0.8546 | LM_LOSS: 0.3543 | LB_LOSS: 1.0005 [2026-05-23 22:14:01] Validation | Batch 1340/1463 | Loss: 0.8550 | LM_LOSS: 0.3547 | LB_LOSS: 1.0005 [2026-05-23 22:14:04] Validation | Batch 1350/1463 | Loss: 0.8549 | LM_LOSS: 0.3546 | LB_LOSS: 1.0005 [2026-05-23 22:14:08] Validation | Batch 1360/1463 | Loss: 0.8552 | LM_LOSS: 0.3549 | LB_LOSS: 1.0005 [2026-05-23 22:14:11] Validation | Batch 1370/1463 | Loss: 0.8556 | LM_LOSS: 0.3553 | LB_LOSS: 1.0005 [2026-05-23 22:14:14] Validation | Batch 1380/1463 | Loss: 0.8552 | LM_LOSS: 0.3550 | LB_LOSS: 1.0005 [2026-05-23 22:14:17] Validation | Batch 1390/1463 | Loss: 0.8552 | LM_LOSS: 0.3549 | LB_LOSS: 1.0005 [2026-05-23 22:14:20] Validation | Batch 1400/1463 | Loss: 0.8555 | LM_LOSS: 0.3553 | LB_LOSS: 1.0005 [2026-05-23 22:14:22] Validation | Batch 1410/1463 | Loss: 0.8559 | LM_LOSS: 0.3557 | LB_LOSS: 1.0005 [2026-05-23 22:14:25] Validation | Batch 1420/1463 | Loss: 0.8558 | LM_LOSS: 0.3556 | LB_LOSS: 1.0005 [2026-05-23 22:14:29] Validation | Batch 1430/1463 | Loss: 0.8559 | LM_LOSS: 0.3556 | LB_LOSS: 1.0005 [2026-05-23 22:14:32] Validation | Batch 1440/1463 | Loss: 0.8564 | LM_LOSS: 0.3561 | LB_LOSS: 1.0005 [2026-05-23 22:14:35] Validation | Batch 1450/1463 | Loss: 0.8561 | LM_LOSS: 0.3559 | LB_LOSS: 1.0005 [2026-05-23 22:14:38] Validation | Batch 1460/1463 | Loss: 0.8565 | LM_LOSS: 0.3563 | LB_LOSS: 1.0005 [2026-05-23 22:14:39] Validation | Batch 1463/1463 | Loss: 0.8564 | LM_LOSS: 0.3562 | LB_LOSS: 1.0005 [2026-05-23 22:14:39] Validation | Loss: 0.8564 | LM_LOSS: 0.3562 | LB_LOSS: 1.0005 | PPL: 1.43 | Time: 443.59s [2026-05-23 22:14:52] Epoch 3 | Step 10010 | Loss: 0.6214 | LM: 0.1219 | LB: 1.0011 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.439 | HR1: 0.400/SR1: 0.394 | LR: 1.00e-04 [2026-05-23 22:15:05] Epoch 3 | Step 10020 | Loss: 0.6211 | LM: 0.1217 | LB: 1.0011 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.439 | HR1: 0.400/SR1: 0.394 | LR: 1.00e-04 [2026-05-23 22:15:18] Epoch 3 | Step 10030 | Loss: 0.6210 | LM: 0.1213 | LB: 1.0011 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.439 | HR1: 0.400/SR1: 0.394 | LR: 1.00e-04 [2026-05-23 22:15:30] Epoch 3 | Step 10040 | Loss: 0.6209 | LM: 0.1210 | LB: 1.0011 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.439 | HR1: 0.400/SR1: 0.394 | LR: 1.00e-04 [2026-05-23 22:15:42] Epoch 3 | Step 10050 | Loss: 0.6207 | LM: 0.1210 | LB: 1.0011 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.439 | HR1: 0.400/SR1: 0.394 | LR: 1.00e-04 [2026-05-23 22:15:56] Epoch 3 | Step 10060 | Loss: 0.6209 | LM: 0.1216 | LB: 1.0011 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.439 | HR1: 0.400/SR1: 0.394 | LR: 1.00e-04 [2026-05-23 22:16:09] Epoch 3 | Step 10070 | Loss: 0.6212 | LM: 0.1220 | LB: 1.0011 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.439 | HR1: 0.400/SR1: 0.394 | LR: 1.00e-04 [2026-05-23 22:16:22] Epoch 3 | Step 10080 | Loss: 0.6216 | LM: 0.1226 | LB: 1.0011 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.439 | HR1: 0.400/SR1: 0.394 | LR: 1.00e-04 [2026-05-23 22:16:35] Epoch 3 | Step 10090 | Loss: 0.6219 | LM: 0.1229 | LB: 1.0011 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.439 | HR1: 0.400/SR1: 0.394 | LR: 1.00e-04 [2026-05-23 22:16:47] Epoch 3 | Step 10100 | Loss: 0.6218 | LM: 0.1226 | LB: 1.0011 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.439 | HR1: 0.400/SR1: 0.394 | LR: 1.00e-04 [2026-05-23 22:16:59] Epoch 3 | Step 10110 | Loss: 0.6219 | LM: 0.1225 | LB: 1.0011 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.439 | HR1: 0.400/SR1: 0.394 | LR: 1.00e-04 [2026-05-23 22:17:11] Epoch 3 | Step 10120 | Loss: 0.6223 | LM: 0.1225 | LB: 1.0011 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.439 | HR1: 0.400/SR1: 0.394 | LR: 1.00e-04 [2026-05-23 22:17:23] Epoch 3 | Step 10130 | Loss: 0.6224 | LM: 0.1229 | LB: 1.0011 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.439 | HR1: 0.400/SR1: 0.394 | LR: 1.00e-04 [2026-05-23 22:17:37] Epoch 3 | Step 10140 | Loss: 0.6222 | LM: 0.1222 | LB: 1.0011 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.439 | HR1: 0.400/SR1: 0.394 | LR: 1.00e-04 [2026-05-23 22:17:50] Epoch 3 | Step 10150 | Loss: 0.6224 | LM: 0.1220 | LB: 1.0011 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.439 | HR1: 0.400/SR1: 0.394 | LR: 1.00e-04 [2026-05-23 22:18:03] Epoch 3 | Step 10160 | Loss: 0.6224 | LM: 0.1221 | LB: 1.0011 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.439 | HR1: 0.400/SR1: 0.394 | LR: 1.00e-04 [2026-05-23 22:18:16] Epoch 3 | Step 10170 | Loss: 0.6223 | LM: 0.1219 | LB: 1.0011 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.439 | HR1: 0.400/SR1: 0.394 | LR: 1.00e-04 [2026-05-23 22:18:29] Epoch 3 | Step 10180 | Loss: 0.6222 | LM: 0.1220 | LB: 1.0011 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.439 | HR1: 0.400/SR1: 0.394 | LR: 1.00e-04 [2026-05-23 22:18:41] Epoch 3 | Step 10190 | Loss: 0.6223 | LM: 0.1217 | LB: 1.0011 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.439 | HR1: 0.400/SR1: 0.394 | LR: 1.00e-04 [2026-05-23 22:18:55] Epoch 3 | Step 10200 | Loss: 0.6222 | LM: 0.1213 | LB: 1.0011 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.439 | HR1: 0.400/SR1: 0.394 | LR: 1.00e-04 [2026-05-23 22:19:07] Epoch 3 | Step 10210 | Loss: 0.6223 | LM: 0.1212 | LB: 1.0011 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.439 | HR1: 0.400/SR1: 0.394 | LR: 1.00e-04 [2026-05-23 22:19:20] Epoch 3 | Step 10220 | Loss: 0.6224 | LM: 0.1215 | LB: 1.0011 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.439 | HR1: 0.400/SR1: 0.394 | LR: 1.00e-04 [2026-05-23 22:19:32] Epoch 3 | Step 10230 | Loss: 0.6225 | LM: 0.1216 | LB: 1.0011 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.439 | HR1: 0.400/SR1: 0.394 | LR: 1.00e-04 [2026-05-23 22:19:44] Epoch 3 | Step 10240 | Loss: 0.6228 | LM: 0.1219 | LB: 1.0011 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.439 | HR1: 0.400/SR1: 0.394 | LR: 1.00e-04 [2026-05-23 22:19:56] Epoch 3 | Step 10250 | Loss: 0.6226 | LM: 0.1218 | LB: 1.0011 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.439 | HR1: 0.400/SR1: 0.394 | LR: 1.00e-04 [2026-05-23 22:20:08] Epoch 3 | Step 10260 | Loss: 0.6226 | LM: 0.1215 | LB: 1.0011 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.439 | HR1: 0.400/SR1: 0.394 | LR: 1.00e-04 [2026-05-23 22:20:21] Epoch 3 | Step 10270 | Loss: 0.6225 | LM: 0.1215 | LB: 1.0011 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.439 | HR1: 0.400/SR1: 0.394 | LR: 1.00e-04 [2026-05-23 22:20:35] Epoch 3 | Step 10280 | Loss: 0.6226 | LM: 0.1214 | LB: 1.0011 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.439 | HR1: 0.400/SR1: 0.394 | LR: 1.00e-04 [2026-05-23 22:20:48] Epoch 3 | Step 10290 | Loss: 0.6227 | LM: 0.1216 | LB: 1.0011 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.439 | HR1: 0.400/SR1: 0.394 | LR: 1.00e-04 [2026-05-23 22:21:00] Epoch 3 | Step 10300 | Loss: 0.6227 | LM: 0.1218 | LB: 1.0011 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.439 | HR1: 0.400/SR1: 0.394 | LR: 1.00e-04 [2026-05-23 22:21:12] Epoch 3 | Step 10310 | Loss: 0.6224 | LM: 0.1216 | LB: 1.0011 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.439 | HR1: 0.400/SR1: 0.394 | LR: 1.00e-04 [2026-05-23 22:21:26] Epoch 3 | Step 10320 | Loss: 0.6224 | LM: 0.1220 | LB: 1.0011 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.439 | HR1: 0.400/SR1: 0.394 | LR: 1.00e-04 [2026-05-23 22:21:39] Epoch 3 | Step 10330 | Loss: 0.6224 | LM: 0.1219 | LB: 1.0011 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.439 | HR1: 0.400/SR1: 0.394 | LR: 1.00e-04 [2026-05-23 22:21:50] Epoch 3 | Step 10340 | Loss: 0.6225 | LM: 0.1220 | LB: 1.0011 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.439 | HR1: 0.400/SR1: 0.394 | LR: 1.00e-04 [2026-05-23 22:22:02] Epoch 3 | Step 10350 | Loss: 0.6224 | LM: 0.1221 | LB: 1.0011 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.439 | HR1: 0.400/SR1: 0.394 | LR: 1.00e-04 [2026-05-23 22:22:14] Epoch 3 | Step 10360 | Loss: 0.6223 | LM: 0.1222 | LB: 1.0011 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.439 | HR1: 0.400/SR1: 0.394 | LR: 1.00e-04 [2026-05-23 22:22:26] Epoch 3 | Step 10370 | Loss: 0.6224 | LM: 0.1220 | LB: 1.0011 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.439 | HR1: 0.400/SR1: 0.394 | LR: 1.00e-04 [2026-05-23 22:22:37] Epoch 3 | Step 10380 | Loss: 0.6223 | LM: 0.1217 | LB: 1.0011 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.439 | HR1: 0.401/SR1: 0.394 | LR: 1.00e-04 [2026-05-23 22:22:50] Epoch 3 | Step 10390 | Loss: 0.6223 | LM: 0.1217 | LB: 1.0011 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.439 | HR1: 0.401/SR1: 0.394 | LR: 1.00e-04 [2026-05-23 22:23:02] Epoch 3 | Step 10400 | Loss: 0.6224 | LM: 0.1222 | LB: 1.0011 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.439 | HR1: 0.401/SR1: 0.394 | LR: 1.00e-04 [2026-05-23 22:23:15] Epoch 3 | Step 10410 | Loss: 0.6223 | LM: 0.1221 | LB: 1.0011 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.439 | HR1: 0.401/SR1: 0.394 | LR: 1.00e-04 [2026-05-23 22:23:27] Epoch 3 | Step 10420 | Loss: 0.6224 | LM: 0.1228 | LB: 1.0011 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.439 | HR1: 0.401/SR1: 0.394 | LR: 1.00e-04 [2026-05-23 22:23:39] Epoch 3 | Step 10430 | Loss: 0.6222 | LM: 0.1227 | LB: 1.0011 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.439 | HR1: 0.401/SR1: 0.394 | LR: 1.00e-04 [2026-05-23 22:23:51] Epoch 3 | Step 10440 | Loss: 0.6219 | LM: 0.1224 | LB: 1.0011 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.439 | HR1: 0.401/SR1: 0.394 | LR: 1.00e-04 [2026-05-23 22:24:03] Epoch 3 | Step 10450 | Loss: 0.6218 | LM: 0.1223 | LB: 1.0011 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.439 | HR1: 0.401/SR1: 0.394 | LR: 1.00e-04 [2026-05-23 22:24:16] Epoch 3 | Step 10460 | Loss: 0.6220 | LM: 0.1223 | LB: 1.0011 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.439 | HR1: 0.401/SR1: 0.394 | LR: 1.00e-04 [2026-05-23 22:24:28] Epoch 3 | Step 10470 | Loss: 0.6220 | LM: 0.1223 | LB: 1.0011 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.439 | HR1: 0.401/SR1: 0.394 | LR: 1.00e-04 [2026-05-23 22:24:40] Epoch 3 | Step 10480 | Loss: 0.6219 | LM: 0.1222 | LB: 1.0011 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.439 | HR1: 0.401/SR1: 0.394 | LR: 1.00e-04 [2026-05-23 22:24:52] Epoch 3 | Step 10490 | Loss: 0.6219 | LM: 0.1220 | LB: 1.0011 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.439 | HR1: 0.400/SR1: 0.394 | LR: 1.00e-04 [2026-05-23 22:25:05] Epoch 3 | Step 10500 | Loss: 0.6218 | LM: 0.1218 | LB: 1.0011 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.439 | HR1: 0.400/SR1: 0.394 | LR: 1.00e-04 [2026-05-23 22:25:17] Epoch 3 | Step 10510 | Loss: 0.6217 | LM: 0.1218 | LB: 1.0011 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.439 | HR1: 0.400/SR1: 0.394 | LR: 1.00e-04 [2026-05-23 22:25:30] Epoch 3 | Step 10520 | Loss: 0.6220 | LM: 0.1220 | LB: 1.0011 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.439 | HR1: 0.400/SR1: 0.394 | LR: 1.00e-04 [2026-05-23 22:25:43] Epoch 3 | Step 10530 | Loss: 0.6221 | LM: 0.1225 | LB: 1.0011 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.439 | HR1: 0.400/SR1: 0.394 | LR: 1.00e-04 [2026-05-23 22:25:55] Epoch 3 | Step 10540 | Loss: 0.6220 | LM: 0.1224 | LB: 1.0011 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.439 | HR1: 0.400/SR1: 0.394 | LR: 1.00e-04 [2026-05-23 22:26:08] Epoch 3 | Step 10550 | Loss: 0.6221 | LM: 0.1226 | LB: 1.0011 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.439 | HR1: 0.401/SR1: 0.394 | LR: 1.00e-04 [2026-05-23 22:26:22] Epoch 3 | Step 10560 | Loss: 0.6221 | LM: 0.1226 | LB: 1.0011 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.439 | HR1: 0.401/SR1: 0.394 | LR: 1.00e-04 [2026-05-23 22:26:33] Epoch 3 | Step 10570 | Loss: 0.6221 | LM: 0.1227 | LB: 1.0011 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.439 | HR1: 0.401/SR1: 0.394 | LR: 1.00e-04 [2026-05-23 22:26:46] Epoch 3 | Step 10580 | Loss: 0.6222 | LM: 0.1227 | LB: 1.0011 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.439 | HR1: 0.401/SR1: 0.394 | LR: 1.00e-04 [2026-05-23 22:26:58] Epoch 3 | Step 10590 | Loss: 0.6222 | LM: 0.1227 | LB: 1.0011 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.439 | HR1: 0.401/SR1: 0.394 | LR: 1.00e-04 [2026-05-23 22:27:10] Epoch 3 | Step 10600 | Loss: 0.6220 | LM: 0.1225 | LB: 1.0011 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.439 | HR1: 0.401/SR1: 0.394 | LR: 1.00e-04 [2026-05-23 22:27:23] Epoch 3 | Step 10610 | Loss: 0.6220 | LM: 0.1224 | LB: 1.0011 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.439 | HR1: 0.401/SR1: 0.394 | LR: 1.00e-04 [2026-05-23 22:27:36] Epoch 3 | Step 10620 | Loss: 0.6220 | LM: 0.1224 | LB: 1.0011 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.439 | HR1: 0.401/SR1: 0.394 | LR: 1.00e-04 [2026-05-23 22:27:48] Epoch 3 | Step 10630 | Loss: 0.6220 | LM: 0.1225 | LB: 1.0011 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.439 | HR1: 0.401/SR1: 0.394 | LR: 1.00e-04 [2026-05-23 22:27:59] Epoch 3 | Step 10640 | Loss: 0.6221 | LM: 0.1231 | LB: 1.0011 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.439 | HR1: 0.401/SR1: 0.394 | LR: 1.00e-04 [2026-05-23 22:28:12] Epoch 3 | Step 10650 | Loss: 0.6222 | LM: 0.1231 | LB: 1.0011 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.439 | HR1: 0.401/SR1: 0.394 | LR: 1.00e-04 [2026-05-23 22:28:26] Epoch 3 | Step 10660 | Loss: 0.6222 | LM: 0.1233 | LB: 1.0011 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.439 | HR1: 0.401/SR1: 0.394 | LR: 1.00e-04 [2026-05-23 22:28:37] Epoch 3 | Step 10670 | Loss: 0.6222 | LM: 0.1233 | LB: 1.0011 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.439 | HR1: 0.401/SR1: 0.394 | LR: 1.00e-04 [2026-05-23 22:28:50] Epoch 3 | Step 10680 | Loss: 0.6222 | LM: 0.1233 | LB: 1.0011 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.439 | HR1: 0.401/SR1: 0.394 | LR: 1.00e-04 [2026-05-23 22:29:03] Epoch 3 | Step 10690 | Loss: 0.6222 | LM: 0.1236 | LB: 1.0011 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.439 | HR1: 0.401/SR1: 0.394 | LR: 1.00e-04 [2026-05-23 22:29:15] Epoch 3 | Step 10700 | Loss: 0.6225 | LM: 0.1239 | LB: 1.0011 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.439 | HR1: 0.401/SR1: 0.394 | LR: 1.00e-04 [2026-05-23 22:29:27] Epoch 3 | Step 10710 | Loss: 0.6224 | LM: 0.1239 | LB: 1.0011 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.439 | HR1: 0.401/SR1: 0.394 | LR: 1.00e-04 [2026-05-23 22:29:39] Epoch 3 | Step 10720 | Loss: 0.6224 | LM: 0.1238 | LB: 1.0011 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.439 | HR1: 0.401/SR1: 0.394 | LR: 1.00e-04 [2026-05-23 22:29:51] Epoch 3 | Step 10730 | Loss: 0.6224 | LM: 0.1238 | LB: 1.0011 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.439 | HR1: 0.401/SR1: 0.394 | LR: 1.00e-04 [2026-05-23 22:30:03] Epoch 3 | Step 10740 | Loss: 0.6223 | LM: 0.1237 | LB: 1.0011 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.439 | HR1: 0.401/SR1: 0.394 | LR: 1.00e-04 [2026-05-23 22:30:15] Epoch 3 | Step 10750 | Loss: 0.6224 | LM: 0.1236 | LB: 1.0011 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.439 | HR1: 0.401/SR1: 0.394 | LR: 1.00e-04 [2026-05-23 22:30:26] Epoch 3 | Step 10760 | Loss: 0.6223 | LM: 0.1236 | LB: 1.0011 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.439 | HR1: 0.401/SR1: 0.395 | LR: 1.00e-04 [2026-05-23 22:30:38] Epoch 3 | Step 10770 | Loss: 0.6225 | LM: 0.1241 | LB: 1.0011 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.439 | HR1: 0.401/SR1: 0.395 | LR: 1.00e-04 [2026-05-23 22:30:50] Epoch 3 | Step 10780 | Loss: 0.6225 | LM: 0.1238 | LB: 1.0011 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.439 | HR1: 0.401/SR1: 0.395 | LR: 1.00e-04 [2026-05-23 22:31:03] Epoch 3 | Step 10790 | Loss: 0.6224 | LM: 0.1238 | LB: 1.0011 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.439 | HR1: 0.401/SR1: 0.395 | LR: 1.00e-04 [2026-05-23 22:31:14] Epoch 3 | Step 10800 | Loss: 0.6224 | LM: 0.1237 | LB: 1.0011 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.439 | HR1: 0.401/SR1: 0.395 | LR: 1.00e-04 [2026-05-23 22:31:25] Epoch 3 | Step 10810 | Loss: 0.6223 | LM: 0.1236 | LB: 1.0011 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.439 | HR1: 0.401/SR1: 0.395 | LR: 1.00e-04 [2026-05-23 22:31:37] Epoch 3 | Step 10820 | Loss: 0.6224 | LM: 0.1235 | LB: 1.0011 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.439 | HR1: 0.401/SR1: 0.395 | LR: 1.00e-04 [2026-05-23 22:31:49] Epoch 3 | Step 10830 | Loss: 0.6224 | LM: 0.1233 | LB: 1.0011 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.439 | HR1: 0.401/SR1: 0.395 | LR: 1.00e-04 [2026-05-23 22:32:01] Epoch 3 | Step 10840 | Loss: 0.6223 | LM: 0.1232 | LB: 1.0011 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.439 | HR1: 0.401/SR1: 0.395 | LR: 1.00e-04 [2026-05-23 22:32:12] Epoch 3 | Step 10850 | Loss: 0.6223 | LM: 0.1232 | LB: 1.0011 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.439 | HR1: 0.401/SR1: 0.395 | LR: 1.00e-04 [2026-05-23 22:32:24] Epoch 3 | Step 10860 | Loss: 0.6223 | LM: 0.1232 | LB: 1.0011 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.439 | HR1: 0.401/SR1: 0.395 | LR: 1.00e-04 [2026-05-23 22:32:38] Epoch 3 | Step 10870 | Loss: 0.6224 | LM: 0.1231 | LB: 1.0011 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.439 | HR1: 0.401/SR1: 0.395 | LR: 1.00e-04 [2026-05-23 22:32:50] Epoch 3 | Step 10880 | Loss: 0.6225 | LM: 0.1230 | LB: 1.0011 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.439 | HR1: 0.401/SR1: 0.395 | LR: 1.00e-04 [2026-05-23 22:33:01] Epoch 3 | Step 10890 | Loss: 0.6223 | LM: 0.1229 | LB: 1.0011 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.439 | HR1: 0.401/SR1: 0.395 | LR: 1.00e-04 [2026-05-23 22:33:12] Epoch 3 | Step 10900 | Loss: 0.6223 | LM: 0.1230 | LB: 1.0011 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.439 | HR1: 0.401/SR1: 0.395 | LR: 1.00e-04 [2026-05-23 22:33:24] Epoch 3 | Step 10910 | Loss: 0.6223 | LM: 0.1228 | LB: 1.0011 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.439 | HR1: 0.401/SR1: 0.395 | LR: 1.00e-04 [2026-05-23 22:33:36] Epoch 3 | Step 10920 | Loss: 0.6223 | LM: 0.1230 | LB: 1.0011 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.439 | HR1: 0.401/SR1: 0.395 | LR: 1.00e-04 [2026-05-23 22:33:48] Epoch 3 | Step 10930 | Loss: 0.6224 | LM: 0.1231 | LB: 1.0011 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.439 | HR1: 0.401/SR1: 0.395 | LR: 1.00e-04 [2026-05-23 22:34:00] Epoch 3 | Step 10940 | Loss: 0.6225 | LM: 0.1231 | LB: 1.0011 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.439 | HR1: 0.401/SR1: 0.395 | LR: 1.00e-04 [2026-05-23 22:34:12] Epoch 3 | Step 10950 | Loss: 0.6225 | LM: 0.1231 | LB: 1.0011 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.439 | HR1: 0.401/SR1: 0.395 | LR: 1.00e-04 [2026-05-23 22:34:24] Epoch 3 | Step 10960 | Loss: 0.6226 | LM: 0.1232 | LB: 1.0011 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.439 | HR1: 0.401/SR1: 0.395 | LR: 1.00e-04 [2026-05-23 22:34:35] Epoch 3 | Step 10970 | Loss: 0.6226 | LM: 0.1231 | LB: 1.0011 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.439 | HR1: 0.401/SR1: 0.395 | LR: 1.00e-04 [2026-05-23 22:34:47] Epoch 3 | Step 10980 | Loss: 0.6226 | LM: 0.1230 | LB: 1.0011 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.439 | HR1: 0.401/SR1: 0.395 | LR: 1.00e-04 [2026-05-23 22:34:59] Epoch 3 | Step 10990 | Loss: 0.6227 | LM: 0.1231 | LB: 1.0011 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.439 | HR1: 0.401/SR1: 0.395 | LR: 1.00e-04 [2026-05-23 22:35:12] Epoch 3 | Step 11000 | Loss: 0.6226 | LM: 0.1229 | LB: 1.0011 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.439 | HR1: 0.401/SR1: 0.395 | LR: 1.00e-04 [2026-05-23 22:35:14] Validation | Batch 10/1463 | Loss: 0.8242 | LM_LOSS: 0.3244 | LB_LOSS: 0.9996 [2026-05-23 22:35:17] Validation | Batch 20/1463 | Loss: 0.8344 | LM_LOSS: 0.3343 | LB_LOSS: 1.0001 [2026-05-23 22:35:20] Validation | Batch 30/1463 | Loss: 0.8330 | LM_LOSS: 0.3328 | LB_LOSS: 1.0004 [2026-05-23 22:35:23] Validation | Batch 40/1463 | Loss: 0.8398 | LM_LOSS: 0.3396 | LB_LOSS: 1.0004 [2026-05-23 22:35:26] Validation | Batch 50/1463 | Loss: 0.8487 | LM_LOSS: 0.3486 | LB_LOSS: 1.0003 [2026-05-23 22:35:30] Validation | Batch 60/1463 | Loss: 0.8497 | LM_LOSS: 0.3495 | LB_LOSS: 1.0004 [2026-05-23 22:35:34] Validation | Batch 70/1463 | Loss: 0.8506 | LM_LOSS: 0.3504 | LB_LOSS: 1.0004 [2026-05-23 22:35:37] Validation | Batch 80/1463 | Loss: 0.8539 | LM_LOSS: 0.3537 | LB_LOSS: 1.0004 [2026-05-23 22:35:40] Validation | Batch 90/1463 | Loss: 0.8537 | LM_LOSS: 0.3535 | LB_LOSS: 1.0004 [2026-05-23 22:35:43] Validation | Batch 100/1463 | Loss: 0.8528 | LM_LOSS: 0.3526 | LB_LOSS: 1.0004 [2026-05-23 22:35:46] Validation | Batch 110/1463 | Loss: 0.8501 | LM_LOSS: 0.3499 | LB_LOSS: 1.0004 [2026-05-23 22:35:49] Validation | Batch 120/1463 | Loss: 0.8531 | LM_LOSS: 0.3529 | LB_LOSS: 1.0004 [2026-05-23 22:35:51] Validation | Batch 130/1463 | Loss: 0.8545 | LM_LOSS: 0.3543 | LB_LOSS: 1.0004 [2026-05-23 22:35:55] Validation | Batch 140/1463 | Loss: 0.8562 | LM_LOSS: 0.3560 | LB_LOSS: 1.0004 [2026-05-23 22:35:58] Validation | Batch 150/1463 | Loss: 0.8532 | LM_LOSS: 0.3530 | LB_LOSS: 1.0005 [2026-05-23 22:36:01] Validation | Batch 160/1463 | Loss: 0.8537 | LM_LOSS: 0.3534 | LB_LOSS: 1.0005 [2026-05-23 22:36:04] Validation | Batch 170/1463 | Loss: 0.8525 | LM_LOSS: 0.3523 | LB_LOSS: 1.0005 [2026-05-23 22:36:08] Validation | Batch 180/1463 | Loss: 0.8536 | LM_LOSS: 0.3534 | LB_LOSS: 1.0005 [2026-05-23 22:36:11] Validation | Batch 190/1463 | Loss: 0.8552 | LM_LOSS: 0.3549 | LB_LOSS: 1.0005 [2026-05-23 22:36:14] Validation | Batch 200/1463 | Loss: 0.8551 | LM_LOSS: 0.3549 | LB_LOSS: 1.0005 [2026-05-23 22:36:17] Validation | Batch 210/1463 | Loss: 0.8561 | LM_LOSS: 0.3559 | LB_LOSS: 1.0005 [2026-05-23 22:36:20] Validation | Batch 220/1463 | Loss: 0.8557 | LM_LOSS: 0.3554 | LB_LOSS: 1.0005 [2026-05-23 22:36:24] Validation | Batch 230/1463 | Loss: 0.8569 | LM_LOSS: 0.3566 | LB_LOSS: 1.0005 [2026-05-23 22:36:27] Validation | Batch 240/1463 | Loss: 0.8588 | LM_LOSS: 0.3586 | LB_LOSS: 1.0005 [2026-05-23 22:36:30] Validation | Batch 250/1463 | Loss: 0.8576 | LM_LOSS: 0.3573 | LB_LOSS: 1.0005 [2026-05-23 22:36:33] Validation | Batch 260/1463 | Loss: 0.8598 | LM_LOSS: 0.3595 | LB_LOSS: 1.0005 [2026-05-23 22:36:36] Validation | Batch 270/1463 | Loss: 0.8580 | LM_LOSS: 0.3577 | LB_LOSS: 1.0005 [2026-05-23 22:36:39] Validation | Batch 280/1463 | Loss: 0.8582 | LM_LOSS: 0.3579 | LB_LOSS: 1.0005 [2026-05-23 22:36:42] Validation | Batch 290/1463 | Loss: 0.8569 | LM_LOSS: 0.3567 | LB_LOSS: 1.0005 [2026-05-23 22:36:45] Validation | Batch 300/1463 | Loss: 0.8580 | LM_LOSS: 0.3577 | LB_LOSS: 1.0005 [2026-05-23 22:36:48] Validation | Batch 310/1463 | Loss: 0.8589 | LM_LOSS: 0.3587 | LB_LOSS: 1.0005 [2026-05-23 22:36:51] Validation | Batch 320/1463 | Loss: 0.8577 | LM_LOSS: 0.3574 | LB_LOSS: 1.0005 [2026-05-23 22:36:54] Validation | Batch 330/1463 | Loss: 0.8571 | LM_LOSS: 0.3568 | LB_LOSS: 1.0005 [2026-05-23 22:36:57] Validation | Batch 340/1463 | Loss: 0.8567 | LM_LOSS: 0.3564 | LB_LOSS: 1.0005 [2026-05-23 22:37:00] Validation | Batch 350/1463 | Loss: 0.8577 | LM_LOSS: 0.3574 | LB_LOSS: 1.0005 [2026-05-23 22:37:03] Validation | Batch 360/1463 | Loss: 0.8576 | LM_LOSS: 0.3574 | LB_LOSS: 1.0005 [2026-05-23 22:37:06] Validation | Batch 370/1463 | Loss: 0.8574 | LM_LOSS: 0.3572 | LB_LOSS: 1.0005 [2026-05-23 22:37:09] Validation | Batch 380/1463 | Loss: 0.8568 | LM_LOSS: 0.3566 | LB_LOSS: 1.0005 [2026-05-23 22:37:12] Validation | Batch 390/1463 | Loss: 0.8577 | LM_LOSS: 0.3575 | LB_LOSS: 1.0005 [2026-05-23 22:37:15] Validation | Batch 400/1463 | Loss: 0.8575 | LM_LOSS: 0.3573 | LB_LOSS: 1.0005 [2026-05-23 22:37:18] Validation | Batch 410/1463 | Loss: 0.8571 | LM_LOSS: 0.3568 | LB_LOSS: 1.0005 [2026-05-23 22:37:21] Validation | Batch 420/1463 | Loss: 0.8573 | LM_LOSS: 0.3570 | LB_LOSS: 1.0005 [2026-05-23 22:37:24] Validation | Batch 430/1463 | Loss: 0.8567 | LM_LOSS: 0.3564 | LB_LOSS: 1.0005 [2026-05-23 22:37:27] Validation | Batch 440/1463 | Loss: 0.8569 | LM_LOSS: 0.3567 | LB_LOSS: 1.0005 [2026-05-23 22:37:30] Validation | Batch 450/1463 | Loss: 0.8565 | LM_LOSS: 0.3562 | LB_LOSS: 1.0005 [2026-05-23 22:37:33] Validation | Batch 460/1463 | Loss: 0.8565 | LM_LOSS: 0.3562 | LB_LOSS: 1.0005 [2026-05-23 22:37:36] Validation | Batch 470/1463 | Loss: 0.8574 | LM_LOSS: 0.3572 | LB_LOSS: 1.0005 [2026-05-23 22:37:40] Validation | Batch 480/1463 | Loss: 0.8571 | LM_LOSS: 0.3569 | LB_LOSS: 1.0005 [2026-05-23 22:37:43] Validation | Batch 490/1463 | Loss: 0.8571 | LM_LOSS: 0.3568 | LB_LOSS: 1.0005 [2026-05-23 22:37:46] Validation | Batch 500/1463 | Loss: 0.8570 | LM_LOSS: 0.3567 | LB_LOSS: 1.0005 [2026-05-23 22:37:50] Validation | Batch 510/1463 | Loss: 0.8561 | LM_LOSS: 0.3559 | LB_LOSS: 1.0005 [2026-05-23 22:37:52] Validation | Batch 520/1463 | Loss: 0.8559 | LM_LOSS: 0.3557 | LB_LOSS: 1.0005 [2026-05-23 22:37:55] Validation | Batch 530/1463 | Loss: 0.8567 | LM_LOSS: 0.3564 | LB_LOSS: 1.0005 [2026-05-23 22:37:57] Validation | Batch 540/1463 | Loss: 0.8569 | LM_LOSS: 0.3566 | LB_LOSS: 1.0005 [2026-05-23 22:38:00] Validation | Batch 550/1463 | Loss: 0.8564 | LM_LOSS: 0.3562 | LB_LOSS: 1.0005 [2026-05-23 22:38:03] Validation | Batch 560/1463 | Loss: 0.8564 | LM_LOSS: 0.3561 | LB_LOSS: 1.0005 [2026-05-23 22:38:06] Validation | Batch 570/1463 | Loss: 0.8563 | LM_LOSS: 0.3560 | LB_LOSS: 1.0005 [2026-05-23 22:38:10] Validation | Batch 580/1463 | Loss: 0.8555 | LM_LOSS: 0.3552 | LB_LOSS: 1.0005 [2026-05-23 22:38:13] Validation | Batch 590/1463 | Loss: 0.8552 | LM_LOSS: 0.3549 | LB_LOSS: 1.0005 [2026-05-23 22:38:16] Validation | Batch 600/1463 | Loss: 0.8556 | LM_LOSS: 0.3553 | LB_LOSS: 1.0005 [2026-05-23 22:38:20] Validation | Batch 610/1463 | Loss: 0.8558 | LM_LOSS: 0.3556 | LB_LOSS: 1.0005 [2026-05-23 22:38:23] Validation | Batch 620/1463 | Loss: 0.8552 | LM_LOSS: 0.3550 | LB_LOSS: 1.0005 [2026-05-23 22:38:26] Validation | Batch 630/1463 | Loss: 0.8546 | LM_LOSS: 0.3544 | LB_LOSS: 1.0005 [2026-05-23 22:38:29] Validation | Batch 640/1463 | Loss: 0.8544 | LM_LOSS: 0.3541 | LB_LOSS: 1.0005 [2026-05-23 22:38:31] Validation | Batch 650/1463 | Loss: 0.8540 | LM_LOSS: 0.3537 | LB_LOSS: 1.0005 [2026-05-23 22:38:34] Validation | Batch 660/1463 | Loss: 0.8535 | LM_LOSS: 0.3532 | LB_LOSS: 1.0005 [2026-05-23 22:38:37] Validation | Batch 670/1463 | Loss: 0.8528 | LM_LOSS: 0.3525 | LB_LOSS: 1.0005 [2026-05-23 22:38:39] Validation | Batch 680/1463 | Loss: 0.8529 | LM_LOSS: 0.3526 | LB_LOSS: 1.0005 [2026-05-23 22:38:43] Validation | Batch 690/1463 | Loss: 0.8530 | LM_LOSS: 0.3528 | LB_LOSS: 1.0005 [2026-05-23 22:38:46] Validation | Batch 700/1463 | Loss: 0.8527 | LM_LOSS: 0.3525 | LB_LOSS: 1.0005 [2026-05-23 22:38:48] Validation | Batch 710/1463 | Loss: 0.8530 | LM_LOSS: 0.3528 | LB_LOSS: 1.0005 [2026-05-23 22:38:52] Validation | Batch 720/1463 | Loss: 0.8538 | LM_LOSS: 0.3535 | LB_LOSS: 1.0005 [2026-05-23 22:38:54] Validation | Batch 730/1463 | Loss: 0.8537 | LM_LOSS: 0.3535 | LB_LOSS: 1.0005 [2026-05-23 22:38:57] Validation | Batch 740/1463 | Loss: 0.8534 | LM_LOSS: 0.3532 | LB_LOSS: 1.0005 [2026-05-23 22:38:59] Validation | Batch 750/1463 | Loss: 0.8533 | LM_LOSS: 0.3531 | LB_LOSS: 1.0005 [2026-05-23 22:39:02] Validation | Batch 760/1463 | Loss: 0.8531 | LM_LOSS: 0.3528 | LB_LOSS: 1.0005 [2026-05-23 22:39:04] Validation | Batch 770/1463 | Loss: 0.8530 | LM_LOSS: 0.3528 | LB_LOSS: 1.0005 [2026-05-23 22:39:07] Validation | Batch 780/1463 | Loss: 0.8530 | LM_LOSS: 0.3528 | LB_LOSS: 1.0005 [2026-05-23 22:39:09] Validation | Batch 790/1463 | Loss: 0.8533 | LM_LOSS: 0.3531 | LB_LOSS: 1.0005 [2026-05-23 22:39:12] Validation | Batch 800/1463 | Loss: 0.8533 | LM_LOSS: 0.3530 | LB_LOSS: 1.0005 [2026-05-23 22:39:15] Validation | Batch 810/1463 | Loss: 0.8529 | LM_LOSS: 0.3526 | LB_LOSS: 1.0005 [2026-05-23 22:39:18] Validation | Batch 820/1463 | Loss: 0.8531 | LM_LOSS: 0.3529 | LB_LOSS: 1.0005 [2026-05-23 22:39:21] Validation | Batch 830/1463 | Loss: 0.8533 | LM_LOSS: 0.3530 | LB_LOSS: 1.0005 [2026-05-23 22:39:25] Validation | Batch 840/1463 | Loss: 0.8533 | LM_LOSS: 0.3531 | LB_LOSS: 1.0005 [2026-05-23 22:39:28] Validation | Batch 850/1463 | Loss: 0.8537 | LM_LOSS: 0.3535 | LB_LOSS: 1.0005 [2026-05-23 22:39:31] Validation | Batch 860/1463 | Loss: 0.8541 | LM_LOSS: 0.3538 | LB_LOSS: 1.0005 [2026-05-23 22:39:34] Validation | Batch 870/1463 | Loss: 0.8542 | LM_LOSS: 0.3539 | LB_LOSS: 1.0005 [2026-05-23 22:39:37] Validation | Batch 880/1463 | Loss: 0.8534 | LM_LOSS: 0.3531 | LB_LOSS: 1.0005 [2026-05-23 22:39:40] Validation | Batch 890/1463 | Loss: 0.8533 | LM_LOSS: 0.3530 | LB_LOSS: 1.0005 [2026-05-23 22:39:43] Validation | Batch 900/1463 | Loss: 0.8529 | LM_LOSS: 0.3526 | LB_LOSS: 1.0005 [2026-05-23 22:39:46] Validation | Batch 910/1463 | Loss: 0.8529 | LM_LOSS: 0.3526 | LB_LOSS: 1.0005 [2026-05-23 22:39:50] Validation | Batch 920/1463 | Loss: 0.8532 | LM_LOSS: 0.3530 | LB_LOSS: 1.0005 [2026-05-23 22:39:52] Validation | Batch 930/1463 | Loss: 0.8529 | LM_LOSS: 0.3527 | LB_LOSS: 1.0005 [2026-05-23 22:39:56] Validation | Batch 940/1463 | Loss: 0.8527 | LM_LOSS: 0.3524 | LB_LOSS: 1.0005 [2026-05-23 22:39:59] Validation | Batch 950/1463 | Loss: 0.8529 | LM_LOSS: 0.3526 | LB_LOSS: 1.0005 [2026-05-23 22:40:02] Validation | Batch 960/1463 | Loss: 0.8530 | LM_LOSS: 0.3527 | LB_LOSS: 1.0005 [2026-05-23 22:40:06] Validation | Batch 970/1463 | Loss: 0.8529 | LM_LOSS: 0.3527 | LB_LOSS: 1.0005 [2026-05-23 22:40:09] Validation | Batch 980/1463 | Loss: 0.8534 | LM_LOSS: 0.3531 | LB_LOSS: 1.0005 [2026-05-23 22:40:12] Validation | Batch 990/1463 | Loss: 0.8536 | LM_LOSS: 0.3534 | LB_LOSS: 1.0005 [2026-05-23 22:40:15] Validation | Batch 1000/1463 | Loss: 0.8544 | LM_LOSS: 0.3541 | LB_LOSS: 1.0005 [2026-05-23 22:40:18] Validation | Batch 1010/1463 | Loss: 0.8544 | LM_LOSS: 0.3541 | LB_LOSS: 1.0005 [2026-05-23 22:40:21] Validation | Batch 1020/1463 | Loss: 0.8544 | LM_LOSS: 0.3541 | LB_LOSS: 1.0005 [2026-05-23 22:40:24] Validation | Batch 1030/1463 | Loss: 0.8549 | LM_LOSS: 0.3546 | LB_LOSS: 1.0005 [2026-05-23 22:40:27] Validation | Batch 1040/1463 | Loss: 0.8545 | LM_LOSS: 0.3542 | LB_LOSS: 1.0005 [2026-05-23 22:40:30] Validation | Batch 1050/1463 | Loss: 0.8544 | LM_LOSS: 0.3541 | LB_LOSS: 1.0005 [2026-05-23 22:40:33] Validation | Batch 1060/1463 | Loss: 0.8544 | LM_LOSS: 0.3542 | LB_LOSS: 1.0005 [2026-05-23 22:40:37] Validation | Batch 1070/1463 | Loss: 0.8544 | LM_LOSS: 0.3542 | LB_LOSS: 1.0005 [2026-05-23 22:40:39] Validation | Batch 1080/1463 | Loss: 0.8542 | LM_LOSS: 0.3539 | LB_LOSS: 1.0005 [2026-05-23 22:40:42] Validation | Batch 1090/1463 | Loss: 0.8541 | LM_LOSS: 0.3539 | LB_LOSS: 1.0005 [2026-05-23 22:40:45] Validation | Batch 1100/1463 | Loss: 0.8538 | LM_LOSS: 0.3536 | LB_LOSS: 1.0005 [2026-05-23 22:40:47] Validation | Batch 1110/1463 | Loss: 0.8538 | LM_LOSS: 0.3536 | LB_LOSS: 1.0005 [2026-05-23 22:40:50] Validation | Batch 1120/1463 | Loss: 0.8535 | LM_LOSS: 0.3533 | LB_LOSS: 1.0005 [2026-05-23 22:40:54] Validation | Batch 1130/1463 | Loss: 0.8533 | LM_LOSS: 0.3531 | LB_LOSS: 1.0005 [2026-05-23 22:40:57] Validation | Batch 1140/1463 | Loss: 0.8536 | LM_LOSS: 0.3533 | LB_LOSS: 1.0005 [2026-05-23 22:41:01] Validation | Batch 1150/1463 | Loss: 0.8533 | LM_LOSS: 0.3531 | LB_LOSS: 1.0005 [2026-05-23 22:41:03] Validation | Batch 1160/1463 | Loss: 0.8532 | LM_LOSS: 0.3530 | LB_LOSS: 1.0005 [2026-05-23 22:41:07] Validation | Batch 1170/1463 | Loss: 0.8533 | LM_LOSS: 0.3530 | LB_LOSS: 1.0005 [2026-05-23 22:41:09] Validation | Batch 1180/1463 | Loss: 0.8533 | LM_LOSS: 0.3530 | LB_LOSS: 1.0005 [2026-05-23 22:41:13] Validation | Batch 1190/1463 | Loss: 0.8535 | LM_LOSS: 0.3533 | LB_LOSS: 1.0005 [2026-05-23 22:41:16] Validation | Batch 1200/1463 | Loss: 0.8535 | LM_LOSS: 0.3533 | LB_LOSS: 1.0005 [2026-05-23 22:41:19] Validation | Batch 1210/1463 | Loss: 0.8539 | LM_LOSS: 0.3536 | LB_LOSS: 1.0005 [2026-05-23 22:41:22] Validation | Batch 1220/1463 | Loss: 0.8543 | LM_LOSS: 0.3540 | LB_LOSS: 1.0005 [2026-05-23 22:41:24] Validation | Batch 1230/1463 | Loss: 0.8543 | LM_LOSS: 0.3541 | LB_LOSS: 1.0005 [2026-05-23 22:41:28] Validation | Batch 1240/1463 | Loss: 0.8545 | LM_LOSS: 0.3542 | LB_LOSS: 1.0005 [2026-05-23 22:41:31] Validation | Batch 1250/1463 | Loss: 0.8545 | LM_LOSS: 0.3543 | LB_LOSS: 1.0005 [2026-05-23 22:41:34] Validation | Batch 1260/1463 | Loss: 0.8545 | LM_LOSS: 0.3543 | LB_LOSS: 1.0005 [2026-05-23 22:41:37] Validation | Batch 1270/1463 | Loss: 0.8543 | LM_LOSS: 0.3540 | LB_LOSS: 1.0005 [2026-05-23 22:41:40] Validation | Batch 1280/1463 | Loss: 0.8541 | LM_LOSS: 0.3539 | LB_LOSS: 1.0005 [2026-05-23 22:41:43] Validation | Batch 1290/1463 | Loss: 0.8541 | LM_LOSS: 0.3539 | LB_LOSS: 1.0005 [2026-05-23 22:41:46] Validation | Batch 1300/1463 | Loss: 0.8538 | LM_LOSS: 0.3536 | LB_LOSS: 1.0005 [2026-05-23 22:41:50] Validation | Batch 1310/1463 | Loss: 0.8537 | LM_LOSS: 0.3534 | LB_LOSS: 1.0005 [2026-05-23 22:41:52] Validation | Batch 1320/1463 | Loss: 0.8538 | LM_LOSS: 0.3535 | LB_LOSS: 1.0005 [2026-05-23 22:41:54] Validation | Batch 1330/1463 | Loss: 0.8538 | LM_LOSS: 0.3535 | LB_LOSS: 1.0005 [2026-05-23 22:41:57] Validation | Batch 1340/1463 | Loss: 0.8542 | LM_LOSS: 0.3539 | LB_LOSS: 1.0005 [2026-05-23 22:42:00] Validation | Batch 1350/1463 | Loss: 0.8541 | LM_LOSS: 0.3538 | LB_LOSS: 1.0005 [2026-05-23 22:42:04] Validation | Batch 1360/1463 | Loss: 0.8544 | LM_LOSS: 0.3542 | LB_LOSS: 1.0005 [2026-05-23 22:42:07] Validation | Batch 1370/1463 | Loss: 0.8548 | LM_LOSS: 0.3545 | LB_LOSS: 1.0005 [2026-05-23 22:42:10] Validation | Batch 1380/1463 | Loss: 0.8544 | LM_LOSS: 0.3542 | LB_LOSS: 1.0005 [2026-05-23 22:42:13] Validation | Batch 1390/1463 | Loss: 0.8544 | LM_LOSS: 0.3542 | LB_LOSS: 1.0005 [2026-05-23 22:42:16] Validation | Batch 1400/1463 | Loss: 0.8547 | LM_LOSS: 0.3545 | LB_LOSS: 1.0005 [2026-05-23 22:42:18] Validation | Batch 1410/1463 | Loss: 0.8551 | LM_LOSS: 0.3549 | LB_LOSS: 1.0005 [2026-05-23 22:42:22] Validation | Batch 1420/1463 | Loss: 0.8551 | LM_LOSS: 0.3548 | LB_LOSS: 1.0005 [2026-05-23 22:42:25] Validation | Batch 1430/1463 | Loss: 0.8551 | LM_LOSS: 0.3549 | LB_LOSS: 1.0005 [2026-05-23 22:42:28] Validation | Batch 1440/1463 | Loss: 0.8556 | LM_LOSS: 0.3554 | LB_LOSS: 1.0005 [2026-05-23 22:42:31] Validation | Batch 1450/1463 | Loss: 0.8554 | LM_LOSS: 0.3551 | LB_LOSS: 1.0005 [2026-05-23 22:42:34] Validation | Batch 1460/1463 | Loss: 0.8557 | LM_LOSS: 0.3555 | LB_LOSS: 1.0005 [2026-05-23 22:42:35] Validation | Batch 1463/1463 | Loss: 0.8556 | LM_LOSS: 0.3554 | LB_LOSS: 1.0005 [2026-05-23 22:42:35] Validation | Loss: 0.8556 | LM_LOSS: 0.3554 | LB_LOSS: 1.0005 | PPL: 1.43 | Time: 443.86s [2026-05-23 22:42:47] Epoch 3 | Step 11010 | Loss: 0.6227 | LM: 0.1229 | LB: 1.0011 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.439 | HR1: 0.401/SR1: 0.395 | LR: 1.00e-04 [2026-05-23 22:42:59] Epoch 3 | Step 11020 | Loss: 0.6227 | LM: 0.1231 | LB: 1.0011 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.439 | HR1: 0.401/SR1: 0.395 | LR: 1.00e-04 [2026-05-23 22:43:11] Epoch 3 | Step 11030 | Loss: 0.6226 | LM: 0.1230 | LB: 1.0011 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.439 | HR1: 0.401/SR1: 0.395 | LR: 1.00e-04 [2026-05-23 22:43:23] Epoch 3 | Step 11040 | Loss: 0.6227 | LM: 0.1233 | LB: 1.0011 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.439 | HR1: 0.401/SR1: 0.395 | LR: 1.00e-04 [2026-05-23 22:43:35] Epoch 3 | Step 11050 | Loss: 0.6227 | LM: 0.1234 | LB: 1.0011 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.439 | HR1: 0.401/SR1: 0.395 | LR: 1.00e-04 [2026-05-23 22:43:47] Epoch 3 | Step 11060 | Loss: 0.6228 | LM: 0.1234 | LB: 1.0011 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.439 | HR1: 0.401/SR1: 0.395 | LR: 1.00e-04 [2026-05-23 22:43:58] Epoch 3 | Step 11070 | Loss: 0.6228 | LM: 0.1234 | LB: 1.0011 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.439 | HR1: 0.401/SR1: 0.395 | LR: 1.00e-04 [2026-05-23 22:44:10] Epoch 3 | Step 11080 | Loss: 0.6230 | LM: 0.1236 | LB: 1.0011 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.439 | HR1: 0.401/SR1: 0.395 | LR: 1.00e-04 [2026-05-23 22:44:23] Epoch 3 | Step 11090 | Loss: 0.6230 | LM: 0.1236 | LB: 1.0011 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.439 | HR1: 0.402/SR1: 0.395 | LR: 1.00e-04 [2026-05-23 22:44:37] Epoch 3 | Step 11100 | Loss: 0.6231 | LM: 0.1239 | LB: 1.0011 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.439 | HR1: 0.402/SR1: 0.395 | LR: 1.00e-04 [2026-05-23 22:44:48] Epoch 3 | Step 11110 | Loss: 0.6231 | LM: 0.1239 | LB: 1.0011 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.439 | HR1: 0.402/SR1: 0.395 | LR: 1.00e-04 [2026-05-23 22:45:00] Epoch 3 | Step 11120 | Loss: 0.6231 | LM: 0.1241 | LB: 1.0011 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.439 | HR1: 0.402/SR1: 0.395 | LR: 1.00e-04 [2026-05-23 22:45:11] Epoch 3 | Step 11130 | Loss: 0.6231 | LM: 0.1241 | LB: 1.0011 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.439 | HR1: 0.402/SR1: 0.395 | LR: 1.00e-04 [2026-05-23 22:45:23] Epoch 3 | Step 11140 | Loss: 0.6232 | LM: 0.1240 | LB: 1.0011 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.439 | HR1: 0.402/SR1: 0.395 | LR: 1.00e-04 [2026-05-23 22:45:35] Epoch 3 | Step 11150 | Loss: 0.6233 | LM: 0.1242 | LB: 1.0011 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.439 | HR1: 0.401/SR1: 0.395 | LR: 1.00e-04 [2026-05-23 22:45:47] Epoch 3 | Step 11160 | Loss: 0.6234 | LM: 0.1241 | LB: 1.0011 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.439 | HR1: 0.401/SR1: 0.395 | LR: 1.00e-04 [2026-05-23 22:46:01] Epoch 3 | Step 11170 | Loss: 0.6234 | LM: 0.1242 | LB: 1.0011 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.439 | HR1: 0.401/SR1: 0.395 | LR: 1.00e-04 [2026-05-23 22:46:14] Epoch 3 | Step 11180 | Loss: 0.6234 | LM: 0.1242 | LB: 1.0011 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.439 | HR1: 0.401/SR1: 0.395 | LR: 1.00e-04 [2026-05-23 22:46:26] Epoch 3 | Step 11190 | Loss: 0.6233 | LM: 0.1243 | LB: 1.0011 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.439 | HR1: 0.401/SR1: 0.395 | LR: 1.00e-04 [2026-05-23 22:46:39] Epoch 3 | Step 11200 | Loss: 0.6233 | LM: 0.1242 | LB: 1.0011 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.439 | HR1: 0.401/SR1: 0.395 | LR: 1.00e-04 [2026-05-23 22:46:51] Epoch 3 | Step 11210 | Loss: 0.6233 | LM: 0.1240 | LB: 1.0011 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.439 | HR1: 0.401/SR1: 0.395 | LR: 1.00e-04 [2026-05-23 22:47:03] Epoch 3 | Step 11220 | Loss: 0.6233 | LM: 0.1237 | LB: 1.0011 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.439 | HR1: 0.401/SR1: 0.395 | LR: 1.00e-04 [2026-05-23 22:47:16] Epoch 3 | Step 11230 | Loss: 0.6233 | LM: 0.1235 | LB: 1.0011 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.439 | HR1: 0.401/SR1: 0.395 | LR: 1.00e-04 [2026-05-23 22:47:29] Epoch 3 | Step 11240 | Loss: 0.6234 | LM: 0.1235 | LB: 1.0011 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.439 | HR1: 0.401/SR1: 0.395 | LR: 1.00e-04 [2026-05-23 22:47:41] Epoch 3 | Step 11250 | Loss: 0.6234 | LM: 0.1236 | LB: 1.0011 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.439 | HR1: 0.401/SR1: 0.395 | LR: 1.00e-04 [2026-05-23 22:47:54] Epoch 3 | Step 11260 | Loss: 0.6234 | LM: 0.1236 | LB: 1.0011 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.439 | HR1: 0.401/SR1: 0.395 | LR: 1.00e-04 [2026-05-23 22:48:06] Epoch 3 | Step 11270 | Loss: 0.6234 | LM: 0.1236 | LB: 1.0011 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.439 | HR1: 0.401/SR1: 0.395 | LR: 1.00e-04 [2026-05-23 22:48:18] Epoch 3 | Step 11280 | Loss: 0.6234 | LM: 0.1235 | LB: 1.0011 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.439 | HR1: 0.401/SR1: 0.395 | LR: 1.00e-04 [2026-05-23 22:48:30] Epoch 3 | Step 11290 | Loss: 0.6235 | LM: 0.1236 | LB: 1.0011 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.439 | HR1: 0.401/SR1: 0.395 | LR: 1.00e-04 [2026-05-23 22:48:41] Epoch 3 | Step 11300 | Loss: 0.6234 | LM: 0.1236 | LB: 1.0011 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.439 | HR1: 0.401/SR1: 0.395 | LR: 9.99e-05 [2026-05-23 22:48:54] Epoch 3 | Step 11310 | Loss: 0.6235 | LM: 0.1235 | LB: 1.0011 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.439 | HR1: 0.401/SR1: 0.395 | LR: 9.99e-05 [2026-05-23 22:49:06] Epoch 3 | Step 11320 | Loss: 0.6234 | LM: 0.1234 | LB: 1.0011 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.439 | HR1: 0.401/SR1: 0.395 | LR: 9.99e-05 [2026-05-23 22:49:18] Epoch 3 | Step 11330 | Loss: 0.6234 | LM: 0.1236 | LB: 1.0011 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.439 | HR1: 0.401/SR1: 0.395 | LR: 9.98e-05 [2026-05-23 22:49:30] Epoch 3 | Step 11340 | Loss: 0.6234 | LM: 0.1236 | LB: 1.0011 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.439 | HR1: 0.401/SR1: 0.395 | LR: 9.98e-05 [2026-05-23 22:49:41] Epoch 3 | Step 11350 | Loss: 0.6233 | LM: 0.1235 | LB: 1.0011 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.439 | HR1: 0.401/SR1: 0.395 | LR: 9.97e-05 [2026-05-23 22:49:53] Epoch 3 | Step 11360 | Loss: 0.6233 | LM: 0.1234 | LB: 1.0011 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.439 | HR1: 0.401/SR1: 0.395 | LR: 9.97e-05 [2026-05-23 22:50:05] Epoch 3 | Step 11370 | Loss: 0.6233 | LM: 0.1232 | LB: 1.0011 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.439 | HR1: 0.401/SR1: 0.395 | LR: 9.96e-05 [2026-05-23 22:50:16] Epoch 3 | Step 11380 | Loss: 0.6232 | LM: 0.1230 | LB: 1.0011 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.439 | HR1: 0.401/SR1: 0.395 | LR: 9.95e-05 [2026-05-23 22:50:29] Epoch 3 | Step 11390 | Loss: 0.6234 | LM: 0.1232 | LB: 1.0011 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.439 | HR1: 0.401/SR1: 0.395 | LR: 9.95e-05 [2026-05-23 22:50:40] Epoch 3 | Step 11400 | Loss: 0.6234 | LM: 0.1232 | LB: 1.0011 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.439 | HR1: 0.401/SR1: 0.395 | LR: 9.94e-05 [2026-05-23 22:50:53] Epoch 3 | Step 11410 | Loss: 0.6235 | LM: 0.1232 | LB: 1.0011 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.439 | HR1: 0.401/SR1: 0.395 | LR: 9.93e-05 [2026-05-23 22:51:06] Epoch 3 | Step 11420 | Loss: 0.6235 | LM: 0.1233 | LB: 1.0011 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.439 | HR1: 0.401/SR1: 0.395 | LR: 9.92e-05 [2026-05-23 22:51:18] Epoch 3 | Step 11430 | Loss: 0.6235 | LM: 0.1234 | LB: 1.0011 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.439 | HR1: 0.401/SR1: 0.395 | LR: 9.91e-05 [2026-05-23 22:51:31] Epoch 3 | Step 11440 | Loss: 0.6235 | LM: 0.1236 | LB: 1.0011 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.439 | HR1: 0.401/SR1: 0.395 | LR: 9.90e-05 [2026-05-23 22:51:44] Epoch 3 | Step 11450 | Loss: 0.6235 | LM: 0.1234 | LB: 1.0011 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.439 | HR1: 0.401/SR1: 0.395 | LR: 9.89e-05 [2026-05-23 22:51:57] Epoch 3 | Step 11460 | Loss: 0.6236 | LM: 0.1235 | LB: 1.0011 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.439 | HR1: 0.401/SR1: 0.395 | LR: 9.88e-05 [2026-05-23 22:52:09] Epoch 3 | Step 11470 | Loss: 0.6236 | LM: 0.1235 | LB: 1.0011 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.439 | HR1: 0.401/SR1: 0.395 | LR: 9.86e-05 [2026-05-23 22:52:21] Epoch 3 | Step 11480 | Loss: 0.6237 | LM: 0.1235 | LB: 1.0011 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.439 | HR1: 0.401/SR1: 0.395 | LR: 9.85e-05 [2026-05-23 22:52:33] Epoch 3 | Step 11490 | Loss: 0.6237 | LM: 0.1236 | LB: 1.0011 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.439 | HR1: 0.401/SR1: 0.395 | LR: 9.84e-05 [2026-05-23 22:52:45] Epoch 3 | Step 11500 | Loss: 0.6237 | LM: 0.1236 | LB: 1.0011 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.439 | HR1: 0.401/SR1: 0.395 | LR: 9.83e-05 [2026-05-23 22:52:58] Epoch 3 | Step 11510 | Loss: 0.6237 | LM: 0.1236 | LB: 1.0011 | CL0: 2.5 | CL1: 2.5 | HR0: 0.400/SR0: 0.439 | HR1: 0.401/SR1: 0.395 | LR: 9.81e-05 [2026-05-23 22:53:10] Epoch 3 | Step 11520 | Loss: 0.6237 | LM: 0.1236 | LB: 1.0011 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.439 | HR1: 0.401/SR1: 0.395 | LR: 9.80e-05 [2026-05-23 22:53:23] Epoch 3 | Step 11530 | Loss: 0.6239 | LM: 0.1237 | LB: 1.0011 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.439 | HR1: 0.401/SR1: 0.395 | LR: 9.78e-05 [2026-05-23 22:53:36] Epoch 3 | Step 11540 | Loss: 0.6239 | LM: 0.1237 | LB: 1.0011 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.439 | HR1: 0.401/SR1: 0.395 | LR: 9.77e-05 [2026-05-23 22:53:49] Epoch 3 | Step 11550 | Loss: 0.6240 | LM: 0.1238 | LB: 1.0011 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.439 | HR1: 0.401/SR1: 0.395 | LR: 9.75e-05 [2026-05-23 22:54:02] Epoch 3 | Step 11560 | Loss: 0.6240 | LM: 0.1237 | LB: 1.0011 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.439 | HR1: 0.401/SR1: 0.395 | LR: 9.73e-05 [2026-05-23 22:54:14] Epoch 3 | Step 11570 | Loss: 0.6241 | LM: 0.1238 | LB: 1.0011 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.439 | HR1: 0.401/SR1: 0.395 | LR: 9.72e-05 [2026-05-23 22:54:26] Epoch 3 | Step 11580 | Loss: 0.6241 | LM: 0.1239 | LB: 1.0011 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.439 | HR1: 0.401/SR1: 0.395 | LR: 9.70e-05 [2026-05-23 22:54:40] Epoch 3 | Step 11590 | Loss: 0.6242 | LM: 0.1242 | LB: 1.0011 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.439 | HR1: 0.401/SR1: 0.395 | LR: 9.68e-05 [2026-05-23 22:54:53] Epoch 3 | Step 11600 | Loss: 0.6241 | LM: 0.1240 | LB: 1.0011 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.439 | HR1: 0.401/SR1: 0.395 | LR: 9.66e-05 [2026-05-23 22:55:05] Epoch 3 | Step 11610 | Loss: 0.6241 | LM: 0.1240 | LB: 1.0011 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.439 | HR1: 0.401/SR1: 0.395 | LR: 9.64e-05 [2026-05-23 22:55:17] Epoch 3 | Step 11620 | Loss: 0.6241 | LM: 0.1238 | LB: 1.0011 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.439 | HR1: 0.401/SR1: 0.395 | LR: 9.62e-05 [2026-05-23 22:55:29] Epoch 3 | Step 11630 | Loss: 0.6241 | LM: 0.1240 | LB: 1.0011 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.439 | HR1: 0.401/SR1: 0.395 | LR: 9.60e-05 [2026-05-23 22:55:41] Epoch 3 | Step 11640 | Loss: 0.6241 | LM: 0.1240 | LB: 1.0011 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.439 | HR1: 0.401/SR1: 0.395 | LR: 9.58e-05 [2026-05-23 22:55:53] Epoch 3 | Step 11650 | Loss: 0.6240 | LM: 0.1240 | LB: 1.0011 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.439 | HR1: 0.401/SR1: 0.395 | LR: 9.56e-05 [2026-05-23 22:56:06] Epoch 3 | Step 11660 | Loss: 0.6241 | LM: 0.1239 | LB: 1.0011 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.439 | HR1: 0.401/SR1: 0.395 | LR: 9.54e-05 [2026-05-23 22:56:20] Epoch 3 | Step 11670 | Loss: 0.6241 | LM: 0.1239 | LB: 1.0011 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.439 | HR1: 0.401/SR1: 0.395 | LR: 9.51e-05 [2026-05-23 22:56:33] Epoch 3 | Step 11680 | Loss: 0.6241 | LM: 0.1239 | LB: 1.0011 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.439 | HR1: 0.401/SR1: 0.395 | LR: 9.49e-05 [2026-05-23 22:56:45] Epoch 3 | Step 11690 | Loss: 0.6242 | LM: 0.1239 | LB: 1.0011 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.439 | HR1: 0.401/SR1: 0.395 | LR: 9.47e-05 [2026-05-23 22:56:58] Epoch 3 | Step 11700 | Loss: 0.6242 | LM: 0.1239 | LB: 1.0011 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.439 | HR1: 0.401/SR1: 0.395 | LR: 9.44e-05 [2026-05-23 22:57:11] Epoch 3 | Step 11710 | Loss: 0.6242 | LM: 0.1238 | LB: 1.0011 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.439 | HR1: 0.401/SR1: 0.395 | LR: 9.42e-05 [2026-05-23 22:57:23] Epoch 3 | Step 11720 | Loss: 0.6243 | LM: 0.1238 | LB: 1.0011 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.439 | HR1: 0.401/SR1: 0.395 | LR: 9.39e-05 [2026-05-23 22:57:35] Epoch 3 | Step 11730 | Loss: 0.6244 | LM: 0.1238 | LB: 1.0011 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.439 | HR1: 0.401/SR1: 0.395 | LR: 9.37e-05 [2026-05-23 22:57:47] Epoch 3 | Step 11740 | Loss: 0.6244 | LM: 0.1237 | LB: 1.0011 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.439 | HR1: 0.402/SR1: 0.395 | LR: 9.34e-05 [2026-05-23 22:57:59] Epoch 3 | Step 11750 | Loss: 0.6244 | LM: 0.1237 | LB: 1.0011 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.439 | HR1: 0.401/SR1: 0.395 | LR: 9.32e-05 [2026-05-23 22:58:12] Epoch 3 | Step 11760 | Loss: 0.6244 | LM: 0.1237 | LB: 1.0011 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.439 | HR1: 0.401/SR1: 0.395 | LR: 9.29e-05 [2026-05-23 22:58:24] Epoch 3 | Step 11770 | Loss: 0.6245 | LM: 0.1239 | LB: 1.0011 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.439 | HR1: 0.401/SR1: 0.395 | LR: 9.26e-05 [2026-05-23 22:58:37] Epoch 3 | Step 11780 | Loss: 0.6245 | LM: 0.1240 | LB: 1.0011 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.439 | HR1: 0.401/SR1: 0.395 | LR: 9.23e-05 [2026-05-23 22:58:48] Epoch 3 | Step 11790 | Loss: 0.6245 | LM: 0.1240 | LB: 1.0011 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.439 | HR1: 0.401/SR1: 0.395 | LR: 9.21e-05 [2026-05-23 22:59:01] Epoch 3 | Step 11800 | Loss: 0.6246 | LM: 0.1241 | LB: 1.0011 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.439 | HR1: 0.401/SR1: 0.395 | LR: 9.18e-05 [2026-05-23 22:59:14] Epoch 3 | Step 11810 | Loss: 0.6246 | LM: 0.1240 | LB: 1.0011 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.439 | HR1: 0.401/SR1: 0.395 | LR: 9.15e-05 [2026-05-23 22:59:27] Epoch 3 | Step 11820 | Loss: 0.6246 | LM: 0.1240 | LB: 1.0011 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.439 | HR1: 0.401/SR1: 0.395 | LR: 9.12e-05 [2026-05-23 22:59:40] Epoch 3 | Step 11830 | Loss: 0.6247 | LM: 0.1239 | LB: 1.0011 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.439 | HR1: 0.401/SR1: 0.395 | LR: 9.09e-05 [2026-05-23 22:59:53] Epoch 3 | Step 11840 | Loss: 0.6246 | LM: 0.1240 | LB: 1.0011 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.439 | HR1: 0.401/SR1: 0.395 | LR: 9.06e-05 [2026-05-23 23:00:06] Epoch 3 | Step 11850 | Loss: 0.6247 | LM: 0.1240 | LB: 1.0011 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.439 | HR1: 0.401/SR1: 0.395 | LR: 9.03e-05 [2026-05-23 23:00:19] Epoch 3 | Step 11860 | Loss: 0.6247 | LM: 0.1239 | LB: 1.0011 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.439 | HR1: 0.401/SR1: 0.395 | LR: 8.99e-05 [2026-05-23 23:00:31] Epoch 3 | Step 11870 | Loss: 0.6247 | LM: 0.1240 | LB: 1.0011 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.439 | HR1: 0.401/SR1: 0.395 | LR: 8.96e-05 [2026-05-23 23:00:45] Epoch 3 | Step 11880 | Loss: 0.6247 | LM: 0.1239 | LB: 1.0011 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.439 | HR1: 0.401/SR1: 0.395 | LR: 8.93e-05 [2026-05-23 23:00:57] Epoch 3 | Step 11890 | Loss: 0.6247 | LM: 0.1239 | LB: 1.0011 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.439 | HR1: 0.401/SR1: 0.395 | LR: 8.90e-05 [2026-05-23 23:01:10] Epoch 3 | Step 11900 | Loss: 0.6249 | LM: 0.1240 | LB: 1.0011 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.439 | HR1: 0.401/SR1: 0.395 | LR: 8.86e-05 [2026-05-23 23:01:22] Epoch 3 | Step 11910 | Loss: 0.6249 | LM: 0.1242 | LB: 1.0011 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.439 | HR1: 0.401/SR1: 0.395 | LR: 8.83e-05 [2026-05-23 23:01:34] Epoch 3 | Step 11920 | Loss: 0.6249 | LM: 0.1242 | LB: 1.0011 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.439 | HR1: 0.401/SR1: 0.395 | LR: 8.80e-05 [2026-05-23 23:01:46] Epoch 3 | Step 11930 | Loss: 0.6249 | LM: 0.1241 | LB: 1.0011 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.439 | HR1: 0.401/SR1: 0.395 | LR: 8.76e-05 [2026-05-23 23:01:59] Epoch 3 | Step 11940 | Loss: 0.6249 | LM: 0.1241 | LB: 1.0011 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.439 | HR1: 0.401/SR1: 0.395 | LR: 8.73e-05 [2026-05-23 23:02:12] Epoch 3 | Step 11950 | Loss: 0.6250 | LM: 0.1241 | LB: 1.0011 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.439 | HR1: 0.401/SR1: 0.395 | LR: 8.69e-05 [2026-05-23 23:02:25] Epoch 3 | Step 11960 | Loss: 0.6250 | LM: 0.1240 | LB: 1.0011 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.439 | HR1: 0.401/SR1: 0.395 | LR: 8.66e-05 [2026-05-23 23:02:37] Epoch 3 | Step 11970 | Loss: 0.6250 | LM: 0.1239 | LB: 1.0011 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.439 | HR1: 0.401/SR1: 0.395 | LR: 8.62e-05 [2026-05-23 23:02:50] Epoch 3 | Step 11980 | Loss: 0.6250 | LM: 0.1243 | LB: 1.0011 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.439 | HR1: 0.401/SR1: 0.395 | LR: 8.58e-05 [2026-05-23 23:03:02] Epoch 3 | Step 11990 | Loss: 0.6250 | LM: 0.1242 | LB: 1.0011 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.439 | HR1: 0.401/SR1: 0.395 | LR: 8.55e-05 [2026-05-23 23:03:14] Epoch 3 | Step 12000 | Loss: 0.6250 | LM: 0.1241 | LB: 1.0011 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.439 | HR1: 0.401/SR1: 0.395 | LR: 8.51e-05 [2026-05-23 23:03:18] Checkpoint saved: outputs/2026-05-23/N_2.5/checkpoints/checkpoint_step_12000.pt [2026-05-23 23:03:26] Validation | Batch 10/1463 | Loss: 0.8276 | LM_LOSS: 0.3278 | LB_LOSS: 0.9996 [2026-05-23 23:03:28] Validation | Batch 20/1463 | Loss: 0.8383 | LM_LOSS: 0.3382 | LB_LOSS: 1.0001 [2026-05-23 23:03:32] Validation | Batch 30/1463 | Loss: 0.8368 | LM_LOSS: 0.3366 | LB_LOSS: 1.0004 [2026-05-23 23:03:35] Validation | Batch 40/1463 | Loss: 0.8437 | LM_LOSS: 0.3435 | LB_LOSS: 1.0004 [2026-05-23 23:03:38] Validation | Batch 50/1463 | Loss: 0.8520 | LM_LOSS: 0.3519 | LB_LOSS: 1.0003 [2026-05-23 23:03:42] Validation | Batch 60/1463 | Loss: 0.8528 | LM_LOSS: 0.3526 | LB_LOSS: 1.0003 [2026-05-23 23:03:45] Validation | Batch 70/1463 | Loss: 0.8548 | LM_LOSS: 0.3546 | LB_LOSS: 1.0004 [2026-05-23 23:03:49] Validation | Batch 80/1463 | Loss: 0.8582 | LM_LOSS: 0.3580 | LB_LOSS: 1.0004 [2026-05-23 23:03:52] Validation | Batch 90/1463 | Loss: 0.8585 | LM_LOSS: 0.3583 | LB_LOSS: 1.0003 [2026-05-23 23:03:55] Validation | Batch 100/1463 | Loss: 0.8580 | LM_LOSS: 0.3578 | LB_LOSS: 1.0003 [2026-05-23 23:03:58] Validation | Batch 110/1463 | Loss: 0.8551 | LM_LOSS: 0.3550 | LB_LOSS: 1.0003 [2026-05-23 23:04:00] Validation | Batch 120/1463 | Loss: 0.8582 | LM_LOSS: 0.3580 | LB_LOSS: 1.0003 [2026-05-23 23:04:03] Validation | Batch 130/1463 | Loss: 0.8597 | LM_LOSS: 0.3596 | LB_LOSS: 1.0004 [2026-05-23 23:04:07] Validation | Batch 140/1463 | Loss: 0.8615 | LM_LOSS: 0.3613 | LB_LOSS: 1.0004 [2026-05-23 23:04:10] Validation | Batch 150/1463 | Loss: 0.8580 | LM_LOSS: 0.3578 | LB_LOSS: 1.0004 [2026-05-23 23:04:13] Validation | Batch 160/1463 | Loss: 0.8582 | LM_LOSS: 0.3580 | LB_LOSS: 1.0004 [2026-05-23 23:04:16] Validation | Batch 170/1463 | Loss: 0.8570 | LM_LOSS: 0.3568 | LB_LOSS: 1.0004 [2026-05-23 23:04:19] Validation | Batch 180/1463 | Loss: 0.8580 | LM_LOSS: 0.3578 | LB_LOSS: 1.0004 [2026-05-23 23:04:23] Validation | Batch 190/1463 | Loss: 0.8597 | LM_LOSS: 0.3595 | LB_LOSS: 1.0004 [2026-05-23 23:04:26] Validation | Batch 200/1463 | Loss: 0.8596 | LM_LOSS: 0.3594 | LB_LOSS: 1.0004 [2026-05-23 23:04:28] Validation | Batch 210/1463 | Loss: 0.8609 | LM_LOSS: 0.3607 | LB_LOSS: 1.0004 [2026-05-23 23:04:32] Validation | Batch 220/1463 | Loss: 0.8604 | LM_LOSS: 0.3602 | LB_LOSS: 1.0004 [2026-05-23 23:04:35] Validation | Batch 230/1463 | Loss: 0.8617 | LM_LOSS: 0.3615 | LB_LOSS: 1.0004 [2026-05-23 23:04:39] Validation | Batch 240/1463 | Loss: 0.8636 | LM_LOSS: 0.3633 | LB_LOSS: 1.0004 [2026-05-23 23:04:42] Validation | Batch 250/1463 | Loss: 0.8622 | LM_LOSS: 0.3620 | LB_LOSS: 1.0004 [2026-05-23 23:04:45] Validation | Batch 260/1463 | Loss: 0.8643 | LM_LOSS: 0.3641 | LB_LOSS: 1.0004 [2026-05-23 23:04:48] Validation | Batch 270/1463 | Loss: 0.8624 | LM_LOSS: 0.3622 | LB_LOSS: 1.0005 [2026-05-23 23:04:51] Validation | Batch 280/1463 | Loss: 0.8627 | LM_LOSS: 0.3625 | LB_LOSS: 1.0004 [2026-05-23 23:04:54] Validation | Batch 290/1463 | Loss: 0.8612 | LM_LOSS: 0.3610 | LB_LOSS: 1.0004 [2026-05-23 23:04:57] Validation | Batch 300/1463 | Loss: 0.8623 | LM_LOSS: 0.3620 | LB_LOSS: 1.0005 [2026-05-23 23:05:00] Validation | Batch 310/1463 | Loss: 0.8633 | LM_LOSS: 0.3630 | LB_LOSS: 1.0005 [2026-05-23 23:05:03] Validation | Batch 320/1463 | Loss: 0.8621 | LM_LOSS: 0.3619 | LB_LOSS: 1.0004 [2026-05-23 23:05:06] Validation | Batch 330/1463 | Loss: 0.8615 | LM_LOSS: 0.3613 | LB_LOSS: 1.0005 [2026-05-23 23:05:08] Validation | Batch 340/1463 | Loss: 0.8611 | LM_LOSS: 0.3609 | LB_LOSS: 1.0005 [2026-05-23 23:05:12] Validation | Batch 350/1463 | Loss: 0.8622 | LM_LOSS: 0.3619 | LB_LOSS: 1.0005 [2026-05-23 23:05:15] Validation | Batch 360/1463 | Loss: 0.8620 | LM_LOSS: 0.3618 | LB_LOSS: 1.0005 [2026-05-23 23:05:17] Validation | Batch 370/1463 | Loss: 0.8620 | LM_LOSS: 0.3618 | LB_LOSS: 1.0005 [2026-05-23 23:05:21] Validation | Batch 380/1463 | Loss: 0.8613 | LM_LOSS: 0.3611 | LB_LOSS: 1.0004 [2026-05-23 23:05:24] Validation | Batch 390/1463 | Loss: 0.8620 | LM_LOSS: 0.3618 | LB_LOSS: 1.0004 [2026-05-23 23:05:26] Validation | Batch 400/1463 | Loss: 0.8618 | LM_LOSS: 0.3616 | LB_LOSS: 1.0005 [2026-05-23 23:05:29] Validation | Batch 410/1463 | Loss: 0.8614 | LM_LOSS: 0.3611 | LB_LOSS: 1.0005 [2026-05-23 23:05:33] Validation | Batch 420/1463 | Loss: 0.8617 | LM_LOSS: 0.3614 | LB_LOSS: 1.0005 [2026-05-23 23:05:36] Validation | Batch 430/1463 | Loss: 0.8610 | LM_LOSS: 0.3608 | LB_LOSS: 1.0005 [2026-05-23 23:05:39] Validation | Batch 440/1463 | Loss: 0.8613 | LM_LOSS: 0.3611 | LB_LOSS: 1.0005 [2026-05-23 23:05:41] Validation | Batch 450/1463 | Loss: 0.8608 | LM_LOSS: 0.3606 | LB_LOSS: 1.0005 [2026-05-23 23:05:45] Validation | Batch 460/1463 | Loss: 0.8608 | LM_LOSS: 0.3605 | LB_LOSS: 1.0005 [2026-05-23 23:05:48] Validation | Batch 470/1463 | Loss: 0.8618 | LM_LOSS: 0.3615 | LB_LOSS: 1.0004 [2026-05-23 23:05:52] Validation | Batch 480/1463 | Loss: 0.8615 | LM_LOSS: 0.3612 | LB_LOSS: 1.0005 [2026-05-23 23:05:54] Validation | Batch 490/1463 | Loss: 0.8613 | LM_LOSS: 0.3611 | LB_LOSS: 1.0004 [2026-05-23 23:05:58] Validation | Batch 500/1463 | Loss: 0.8613 | LM_LOSS: 0.3610 | LB_LOSS: 1.0004 [2026-05-23 23:06:01] Validation | Batch 510/1463 | Loss: 0.8604 | LM_LOSS: 0.3602 | LB_LOSS: 1.0004 [2026-05-23 23:06:04] Validation | Batch 520/1463 | Loss: 0.8602 | LM_LOSS: 0.3600 | LB_LOSS: 1.0005 [2026-05-23 23:06:07] Validation | Batch 530/1463 | Loss: 0.8611 | LM_LOSS: 0.3609 | LB_LOSS: 1.0004 [2026-05-23 23:06:09] Validation | Batch 540/1463 | Loss: 0.8611 | LM_LOSS: 0.3609 | LB_LOSS: 1.0004 [2026-05-23 23:06:12] Validation | Batch 550/1463 | Loss: 0.8607 | LM_LOSS: 0.3604 | LB_LOSS: 1.0004 [2026-05-23 23:06:15] Validation | Batch 560/1463 | Loss: 0.8607 | LM_LOSS: 0.3605 | LB_LOSS: 1.0005 [2026-05-23 23:06:18] Validation | Batch 570/1463 | Loss: 0.8607 | LM_LOSS: 0.3604 | LB_LOSS: 1.0005 [2026-05-23 23:06:22] Validation | Batch 580/1463 | Loss: 0.8598 | LM_LOSS: 0.3596 | LB_LOSS: 1.0005 [2026-05-23 23:06:25] Validation | Batch 590/1463 | Loss: 0.8596 | LM_LOSS: 0.3593 | LB_LOSS: 1.0005 [2026-05-23 23:06:28] Validation | Batch 600/1463 | Loss: 0.8599 | LM_LOSS: 0.3597 | LB_LOSS: 1.0005 [2026-05-23 23:06:32] Validation | Batch 610/1463 | Loss: 0.8602 | LM_LOSS: 0.3599 | LB_LOSS: 1.0005 [2026-05-23 23:06:35] Validation | Batch 620/1463 | Loss: 0.8595 | LM_LOSS: 0.3593 | LB_LOSS: 1.0005 [2026-05-23 23:06:38] Validation | Batch 630/1463 | Loss: 0.8588 | LM_LOSS: 0.3586 | LB_LOSS: 1.0005 [2026-05-23 23:06:41] Validation | Batch 640/1463 | Loss: 0.8586 | LM_LOSS: 0.3584 | LB_LOSS: 1.0005 [2026-05-23 23:06:43] Validation | Batch 650/1463 | Loss: 0.8580 | LM_LOSS: 0.3578 | LB_LOSS: 1.0005 [2026-05-23 23:06:46] Validation | Batch 660/1463 | Loss: 0.8576 | LM_LOSS: 0.3573 | LB_LOSS: 1.0005 [2026-05-23 23:06:49] Validation | Batch 670/1463 | Loss: 0.8569 | LM_LOSS: 0.3567 | LB_LOSS: 1.0005 [2026-05-23 23:06:51] Validation | Batch 680/1463 | Loss: 0.8570 | LM_LOSS: 0.3568 | LB_LOSS: 1.0004 [2026-05-23 23:06:55] Validation | Batch 690/1463 | Loss: 0.8573 | LM_LOSS: 0.3570 | LB_LOSS: 1.0005 [2026-05-23 23:06:58] Validation | Batch 700/1463 | Loss: 0.8570 | LM_LOSS: 0.3568 | LB_LOSS: 1.0004 [2026-05-23 23:07:00] Validation | Batch 710/1463 | Loss: 0.8573 | LM_LOSS: 0.3571 | LB_LOSS: 1.0004 [2026-05-23 23:07:04] Validation | Batch 720/1463 | Loss: 0.8581 | LM_LOSS: 0.3578 | LB_LOSS: 1.0005 [2026-05-23 23:07:06] Validation | Batch 730/1463 | Loss: 0.8580 | LM_LOSS: 0.3578 | LB_LOSS: 1.0005 [2026-05-23 23:07:09] Validation | Batch 740/1463 | Loss: 0.8577 | LM_LOSS: 0.3575 | LB_LOSS: 1.0005 [2026-05-23 23:07:11] Validation | Batch 750/1463 | Loss: 0.8577 | LM_LOSS: 0.3575 | LB_LOSS: 1.0005 [2026-05-23 23:07:14] Validation | Batch 760/1463 | Loss: 0.8574 | LM_LOSS: 0.3572 | LB_LOSS: 1.0005 [2026-05-23 23:07:16] Validation | Batch 770/1463 | Loss: 0.8575 | LM_LOSS: 0.3573 | LB_LOSS: 1.0005 [2026-05-23 23:07:19] Validation | Batch 780/1463 | Loss: 0.8575 | LM_LOSS: 0.3573 | LB_LOSS: 1.0005 [2026-05-23 23:07:21] Validation | Batch 790/1463 | Loss: 0.8577 | LM_LOSS: 0.3575 | LB_LOSS: 1.0004 [2026-05-23 23:07:24] Validation | Batch 800/1463 | Loss: 0.8577 | LM_LOSS: 0.3574 | LB_LOSS: 1.0004 [2026-05-23 23:07:27] Validation | Batch 810/1463 | Loss: 0.8573 | LM_LOSS: 0.3571 | LB_LOSS: 1.0004 [2026-05-23 23:07:30] Validation | Batch 820/1463 | Loss: 0.8575 | LM_LOSS: 0.3573 | LB_LOSS: 1.0004 [2026-05-23 23:07:33] Validation | Batch 830/1463 | Loss: 0.8576 | LM_LOSS: 0.3574 | LB_LOSS: 1.0004 [2026-05-23 23:07:37] Validation | Batch 840/1463 | Loss: 0.8577 | LM_LOSS: 0.3575 | LB_LOSS: 1.0004 [2026-05-23 23:07:40] Validation | Batch 850/1463 | Loss: 0.8580 | LM_LOSS: 0.3578 | LB_LOSS: 1.0004 [2026-05-23 23:07:43] Validation | Batch 860/1463 | Loss: 0.8584 | LM_LOSS: 0.3582 | LB_LOSS: 1.0004 [2026-05-23 23:07:46] Validation | Batch 870/1463 | Loss: 0.8586 | LM_LOSS: 0.3584 | LB_LOSS: 1.0005 [2026-05-23 23:07:49] Validation | Batch 880/1463 | Loss: 0.8578 | LM_LOSS: 0.3576 | LB_LOSS: 1.0004 [2026-05-23 23:07:52] Validation | Batch 890/1463 | Loss: 0.8577 | LM_LOSS: 0.3575 | LB_LOSS: 1.0004 [2026-05-23 23:07:55] Validation | Batch 900/1463 | Loss: 0.8573 | LM_LOSS: 0.3571 | LB_LOSS: 1.0004 [2026-05-23 23:07:58] Validation | Batch 910/1463 | Loss: 0.8573 | LM_LOSS: 0.3571 | LB_LOSS: 1.0004 [2026-05-23 23:08:02] Validation | Batch 920/1463 | Loss: 0.8577 | LM_LOSS: 0.3575 | LB_LOSS: 1.0004 [2026-05-23 23:08:05] Validation | Batch 930/1463 | Loss: 0.8574 | LM_LOSS: 0.3572 | LB_LOSS: 1.0004 [2026-05-23 23:08:08] Validation | Batch 940/1463 | Loss: 0.8571 | LM_LOSS: 0.3569 | LB_LOSS: 1.0004 [2026-05-23 23:08:11] Validation | Batch 950/1463 | Loss: 0.8573 | LM_LOSS: 0.3571 | LB_LOSS: 1.0004 [2026-05-23 23:08:14] Validation | Batch 960/1463 | Loss: 0.8574 | LM_LOSS: 0.3571 | LB_LOSS: 1.0004 [2026-05-23 23:08:18] Validation | Batch 970/1463 | Loss: 0.8574 | LM_LOSS: 0.3571 | LB_LOSS: 1.0004 [2026-05-23 23:08:21] Validation | Batch 980/1463 | Loss: 0.8577 | LM_LOSS: 0.3575 | LB_LOSS: 1.0004 [2026-05-23 23:08:24] Validation | Batch 990/1463 | Loss: 0.8580 | LM_LOSS: 0.3578 | LB_LOSS: 1.0005 [2026-05-23 23:08:27] Validation | Batch 1000/1463 | Loss: 0.8587 | LM_LOSS: 0.3584 | LB_LOSS: 1.0005 [2026-05-23 23:08:30] Validation | Batch 1010/1463 | Loss: 0.8586 | LM_LOSS: 0.3584 | LB_LOSS: 1.0005 [2026-05-23 23:08:33] Validation | Batch 1020/1463 | Loss: 0.8587 | LM_LOSS: 0.3584 | LB_LOSS: 1.0005 [2026-05-23 23:08:37] Validation | Batch 1030/1463 | Loss: 0.8591 | LM_LOSS: 0.3589 | LB_LOSS: 1.0005 [2026-05-23 23:08:39] Validation | Batch 1040/1463 | Loss: 0.8587 | LM_LOSS: 0.3585 | LB_LOSS: 1.0005 [2026-05-23 23:08:42] Validation | Batch 1050/1463 | Loss: 0.8587 | LM_LOSS: 0.3585 | LB_LOSS: 1.0004 [2026-05-23 23:08:46] Validation | Batch 1060/1463 | Loss: 0.8587 | LM_LOSS: 0.3585 | LB_LOSS: 1.0004 [2026-05-23 23:08:49] Validation | Batch 1070/1463 | Loss: 0.8587 | LM_LOSS: 0.3585 | LB_LOSS: 1.0004 [2026-05-23 23:08:52] Validation | Batch 1080/1463 | Loss: 0.8585 | LM_LOSS: 0.3583 | LB_LOSS: 1.0004 [2026-05-23 23:08:54] Validation | Batch 1090/1463 | Loss: 0.8585 | LM_LOSS: 0.3583 | LB_LOSS: 1.0004 [2026-05-23 23:08:57] Validation | Batch 1100/1463 | Loss: 0.8581 | LM_LOSS: 0.3579 | LB_LOSS: 1.0004 [2026-05-23 23:09:00] Validation | Batch 1110/1463 | Loss: 0.8582 | LM_LOSS: 0.3579 | LB_LOSS: 1.0004 [2026-05-23 23:09:03] Validation | Batch 1120/1463 | Loss: 0.8578 | LM_LOSS: 0.3576 | LB_LOSS: 1.0004 [2026-05-23 23:09:06] Validation | Batch 1130/1463 | Loss: 0.8576 | LM_LOSS: 0.3574 | LB_LOSS: 1.0004 [2026-05-23 23:09:09] Validation | Batch 1140/1463 | Loss: 0.8579 | LM_LOSS: 0.3577 | LB_LOSS: 1.0004 [2026-05-23 23:09:13] Validation | Batch 1150/1463 | Loss: 0.8576 | LM_LOSS: 0.3574 | LB_LOSS: 1.0004 [2026-05-23 23:09:16] Validation | Batch 1160/1463 | Loss: 0.8576 | LM_LOSS: 0.3573 | LB_LOSS: 1.0004 [2026-05-23 23:09:19] Validation | Batch 1170/1463 | Loss: 0.8576 | LM_LOSS: 0.3574 | LB_LOSS: 1.0004 [2026-05-23 23:09:22] Validation | Batch 1180/1463 | Loss: 0.8577 | LM_LOSS: 0.3575 | LB_LOSS: 1.0004 [2026-05-23 23:09:25] Validation | Batch 1190/1463 | Loss: 0.8579 | LM_LOSS: 0.3577 | LB_LOSS: 1.0004 [2026-05-23 23:09:29] Validation | Batch 1200/1463 | Loss: 0.8579 | LM_LOSS: 0.3577 | LB_LOSS: 1.0004 [2026-05-23 23:09:31] Validation | Batch 1210/1463 | Loss: 0.8582 | LM_LOSS: 0.3580 | LB_LOSS: 1.0004 [2026-05-23 23:09:34] Validation | Batch 1220/1463 | Loss: 0.8586 | LM_LOSS: 0.3583 | LB_LOSS: 1.0004 [2026-05-23 23:09:37] Validation | Batch 1230/1463 | Loss: 0.8586 | LM_LOSS: 0.3584 | LB_LOSS: 1.0004 [2026-05-23 23:09:40] Validation | Batch 1240/1463 | Loss: 0.8588 | LM_LOSS: 0.3585 | LB_LOSS: 1.0004 [2026-05-23 23:09:43] Validation | Batch 1250/1463 | Loss: 0.8588 | LM_LOSS: 0.3586 | LB_LOSS: 1.0004 [2026-05-23 23:09:47] Validation | Batch 1260/1463 | Loss: 0.8588 | LM_LOSS: 0.3586 | LB_LOSS: 1.0004 [2026-05-23 23:09:49] Validation | Batch 1270/1463 | Loss: 0.8585 | LM_LOSS: 0.3583 | LB_LOSS: 1.0004 [2026-05-23 23:09:53] Validation | Batch 1280/1463 | Loss: 0.8584 | LM_LOSS: 0.3582 | LB_LOSS: 1.0004 [2026-05-23 23:09:56] Validation | Batch 1290/1463 | Loss: 0.8584 | LM_LOSS: 0.3581 | LB_LOSS: 1.0004 [2026-05-23 23:09:59] Validation | Batch 1300/1463 | Loss: 0.8581 | LM_LOSS: 0.3578 | LB_LOSS: 1.0004 [2026-05-23 23:10:02] Validation | Batch 1310/1463 | Loss: 0.8578 | LM_LOSS: 0.3576 | LB_LOSS: 1.0004 [2026-05-23 23:10:04] Validation | Batch 1320/1463 | Loss: 0.8579 | LM_LOSS: 0.3577 | LB_LOSS: 1.0004 [2026-05-23 23:10:07] Validation | Batch 1330/1463 | Loss: 0.8579 | LM_LOSS: 0.3577 | LB_LOSS: 1.0004 [2026-05-23 23:10:10] Validation | Batch 1340/1463 | Loss: 0.8583 | LM_LOSS: 0.3581 | LB_LOSS: 1.0004 [2026-05-23 23:10:13] Validation | Batch 1350/1463 | Loss: 0.8583 | LM_LOSS: 0.3581 | LB_LOSS: 1.0004 [2026-05-23 23:10:17] Validation | Batch 1360/1463 | Loss: 0.8586 | LM_LOSS: 0.3584 | LB_LOSS: 1.0004 [2026-05-23 23:10:20] Validation | Batch 1370/1463 | Loss: 0.8589 | LM_LOSS: 0.3587 | LB_LOSS: 1.0004 [2026-05-23 23:10:23] Validation | Batch 1380/1463 | Loss: 0.8586 | LM_LOSS: 0.3584 | LB_LOSS: 1.0004 [2026-05-23 23:10:25] Validation | Batch 1390/1463 | Loss: 0.8586 | LM_LOSS: 0.3584 | LB_LOSS: 1.0004 [2026-05-23 23:10:29] Validation | Batch 1400/1463 | Loss: 0.8589 | LM_LOSS: 0.3587 | LB_LOSS: 1.0004 [2026-05-23 23:10:31] Validation | Batch 1410/1463 | Loss: 0.8593 | LM_LOSS: 0.3591 | LB_LOSS: 1.0004 [2026-05-23 23:10:34] Validation | Batch 1420/1463 | Loss: 0.8592 | LM_LOSS: 0.3590 | LB_LOSS: 1.0004 [2026-05-23 23:10:38] Validation | Batch 1430/1463 | Loss: 0.8593 | LM_LOSS: 0.3591 | LB_LOSS: 1.0004 [2026-05-23 23:10:41] Validation | Batch 1440/1463 | Loss: 0.8598 | LM_LOSS: 0.3595 | LB_LOSS: 1.0004 [2026-05-23 23:10:44] Validation | Batch 1450/1463 | Loss: 0.8595 | LM_LOSS: 0.3593 | LB_LOSS: 1.0004 [2026-05-23 23:10:47] Validation | Batch 1460/1463 | Loss: 0.8599 | LM_LOSS: 0.3597 | LB_LOSS: 1.0004 [2026-05-23 23:10:48] Validation | Batch 1463/1463 | Loss: 0.8598 | LM_LOSS: 0.3596 | LB_LOSS: 1.0004 [2026-05-23 23:10:48] Validation | Loss: 0.8598 | LM_LOSS: 0.3596 | LB_LOSS: 1.0004 | PPL: 1.43 | Time: 444.84s [2026-05-23 23:11:00] Epoch 3 | Step 12010 | Loss: 0.6250 | LM: 0.1243 | LB: 1.0011 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.439 | HR1: 0.401/SR1: 0.395 | LR: 8.47e-05 [2026-05-23 23:11:13] Epoch 3 | Step 12020 | Loss: 0.6251 | LM: 0.1244 | LB: 1.0011 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.439 | HR1: 0.401/SR1: 0.395 | LR: 8.44e-05 [2026-05-23 23:11:25] Epoch 3 | Step 12030 | Loss: 0.6250 | LM: 0.1243 | LB: 1.0011 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.439 | HR1: 0.401/SR1: 0.395 | LR: 8.40e-05 [2026-05-23 23:11:38] Epoch 3 | Step 12040 | Loss: 0.6252 | LM: 0.1245 | LB: 1.0011 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.439 | HR1: 0.401/SR1: 0.395 | LR: 8.36e-05 [2026-05-23 23:11:50] Epoch 3 | Step 12050 | Loss: 0.6252 | LM: 0.1246 | LB: 1.0011 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.439 | HR1: 0.401/SR1: 0.395 | LR: 8.32e-05 [2026-05-23 23:12:03] Epoch 3 | Step 12060 | Loss: 0.6251 | LM: 0.1245 | LB: 1.0011 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.439 | HR1: 0.401/SR1: 0.395 | LR: 8.28e-05 [2026-05-23 23:12:15] Epoch 3 | Step 12070 | Loss: 0.6251 | LM: 0.1245 | LB: 1.0011 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.439 | HR1: 0.401/SR1: 0.395 | LR: 8.24e-05 [2026-05-23 23:12:28] Epoch 3 | Step 12080 | Loss: 0.6252 | LM: 0.1247 | LB: 1.0011 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.439 | HR1: 0.401/SR1: 0.395 | LR: 8.20e-05 [2026-05-23 23:12:41] Epoch 3 | Step 12090 | Loss: 0.6251 | LM: 0.1246 | LB: 1.0011 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.439 | HR1: 0.401/SR1: 0.395 | LR: 8.16e-05 [2026-05-23 23:12:55] Epoch 3 | Step 12100 | Loss: 0.6250 | LM: 0.1245 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.440 | HR1: 0.401/SR1: 0.395 | LR: 8.12e-05 [2026-05-23 23:13:07] Epoch 3 | Step 12110 | Loss: 0.6250 | LM: 0.1245 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.440 | HR1: 0.401/SR1: 0.395 | LR: 8.08e-05 [2026-05-23 23:13:19] Epoch 3 | Step 12120 | Loss: 0.6250 | LM: 0.1244 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.440 | HR1: 0.401/SR1: 0.395 | LR: 8.04e-05 [2026-05-23 23:13:33] Epoch 3 | Step 12130 | Loss: 0.6251 | LM: 0.1244 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.440 | HR1: 0.401/SR1: 0.395 | LR: 7.99e-05 [2026-05-23 23:13:45] Epoch 3 | Step 12140 | Loss: 0.6251 | LM: 0.1245 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.440 | HR1: 0.401/SR1: 0.395 | LR: 7.95e-05 [2026-05-23 23:13:58] Epoch 3 | Step 12150 | Loss: 0.6251 | LM: 0.1244 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.440 | HR1: 0.401/SR1: 0.395 | LR: 7.91e-05 [2026-05-23 23:14:10] Epoch 3 | Step 12160 | Loss: 0.6251 | LM: 0.1244 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.440 | HR1: 0.401/SR1: 0.395 | LR: 7.87e-05 [2026-05-23 23:14:22] Epoch 3 | Step 12170 | Loss: 0.6252 | LM: 0.1245 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.440 | HR1: 0.401/SR1: 0.395 | LR: 7.83e-05 [2026-05-23 23:14:35] Epoch 3 | Step 12180 | Loss: 0.6252 | LM: 0.1245 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.440 | HR1: 0.401/SR1: 0.395 | LR: 7.78e-05 [2026-05-23 23:14:48] Epoch 3 | Step 12190 | Loss: 0.6252 | LM: 0.1245 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.440 | HR1: 0.401/SR1: 0.395 | LR: 7.74e-05 [2026-05-23 23:15:00] Epoch 3 | Step 12200 | Loss: 0.6252 | LM: 0.1245 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.440 | HR1: 0.401/SR1: 0.395 | LR: 7.69e-05 [2026-05-23 23:15:14] Epoch 3 | Step 12210 | Loss: 0.6253 | LM: 0.1245 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.440 | HR1: 0.401/SR1: 0.395 | LR: 7.65e-05 [2026-05-23 23:15:26] Epoch 3 | Step 12220 | Loss: 0.6254 | LM: 0.1247 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.440 | HR1: 0.401/SR1: 0.395 | LR: 7.61e-05 [2026-05-23 23:15:39] Epoch 3 | Step 12230 | Loss: 0.6254 | LM: 0.1247 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.440 | HR1: 0.401/SR1: 0.395 | LR: 7.56e-05 [2026-05-23 23:15:52] Epoch 3 | Step 12240 | Loss: 0.6254 | LM: 0.1247 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.440 | HR1: 0.401/SR1: 0.395 | LR: 7.52e-05 [2026-05-23 23:16:05] Epoch 3 | Step 12250 | Loss: 0.6255 | LM: 0.1248 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.440 | HR1: 0.401/SR1: 0.395 | LR: 7.47e-05 [2026-05-23 23:16:17] Epoch 3 | Step 12260 | Loss: 0.6254 | LM: 0.1248 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.440 | HR1: 0.401/SR1: 0.395 | LR: 7.43e-05 [2026-05-23 23:16:29] Epoch 3 | Step 12270 | Loss: 0.6255 | LM: 0.1247 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.440 | HR1: 0.401/SR1: 0.395 | LR: 7.38e-05 [2026-05-23 23:16:42] Epoch 3 | Step 12280 | Loss: 0.6255 | LM: 0.1248 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.440 | HR1: 0.401/SR1: 0.395 | LR: 7.34e-05 [2026-05-23 23:16:54] Epoch 3 | Step 12290 | Loss: 0.6255 | LM: 0.1247 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.440 | HR1: 0.401/SR1: 0.395 | LR: 7.29e-05 [2026-05-23 23:17:06] Epoch 3 | Step 12300 | Loss: 0.6255 | LM: 0.1248 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.440 | HR1: 0.401/SR1: 0.395 | LR: 7.24e-05 [2026-05-23 23:17:19] Epoch 3 | Step 12310 | Loss: 0.6256 | LM: 0.1250 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.440 | HR1: 0.401/SR1: 0.395 | LR: 7.20e-05 [2026-05-23 23:17:32] Epoch 3 | Step 12320 | Loss: 0.6256 | LM: 0.1251 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.440 | HR1: 0.401/SR1: 0.395 | LR: 7.15e-05 [2026-05-23 23:17:45] Epoch 3 | Step 12330 | Loss: 0.6256 | LM: 0.1251 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.440 | HR1: 0.401/SR1: 0.395 | LR: 7.10e-05 [2026-05-23 23:17:59] Epoch 3 | Step 12340 | Loss: 0.6257 | LM: 0.1253 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.439 | HR1: 0.401/SR1: 0.395 | LR: 7.06e-05 [2026-05-23 23:18:11] Epoch 3 | Step 12350 | Loss: 0.6257 | LM: 0.1255 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.439 | HR1: 0.401/SR1: 0.395 | LR: 7.01e-05 [2026-05-23 23:18:24] Epoch 3 | Step 12360 | Loss: 0.6258 | LM: 0.1254 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.440 | HR1: 0.401/SR1: 0.395 | LR: 6.96e-05 [2026-05-23 23:18:37] Epoch 3 | Step 12370 | Loss: 0.6257 | LM: 0.1254 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.440 | HR1: 0.401/SR1: 0.395 | LR: 6.91e-05 [2026-05-23 23:18:50] Epoch 3 | Step 12380 | Loss: 0.6257 | LM: 0.1254 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.440 | HR1: 0.401/SR1: 0.395 | LR: 6.87e-05 [2026-05-23 23:19:02] Epoch 3 | Step 12390 | Loss: 0.6258 | LM: 0.1254 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.440 | HR1: 0.401/SR1: 0.395 | LR: 6.82e-05 [2026-05-23 23:19:16] Epoch 3 | Step 12400 | Loss: 0.6258 | LM: 0.1254 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.440 | HR1: 0.401/SR1: 0.395 | LR: 6.77e-05 [2026-05-23 23:19:29] Epoch 3 | Step 12410 | Loss: 0.6258 | LM: 0.1255 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.440 | HR1: 0.401/SR1: 0.395 | LR: 6.72e-05 [2026-05-23 23:19:41] Epoch 3 | Step 12420 | Loss: 0.6258 | LM: 0.1255 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.440 | HR1: 0.401/SR1: 0.395 | LR: 6.67e-05 [2026-05-23 23:19:54] Epoch 3 | Step 12430 | Loss: 0.6259 | LM: 0.1256 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.440 | HR1: 0.401/SR1: 0.395 | LR: 6.62e-05 [2026-05-23 23:20:06] Epoch 3 | Step 12440 | Loss: 0.6259 | LM: 0.1255 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.440 | HR1: 0.401/SR1: 0.395 | LR: 6.58e-05 [2026-05-23 23:20:19] Epoch 3 | Step 12450 | Loss: 0.6258 | LM: 0.1255 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.440 | HR1: 0.401/SR1: 0.395 | LR: 6.53e-05 [2026-05-23 23:20:31] Epoch 3 | Step 12460 | Loss: 0.6258 | LM: 0.1254 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.439 | HR1: 0.401/SR1: 0.395 | LR: 6.48e-05 [2026-05-23 23:20:44] Epoch 3 | Step 12470 | Loss: 0.6258 | LM: 0.1256 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.439 | HR1: 0.401/SR1: 0.395 | LR: 6.43e-05 [2026-05-23 23:20:58] Epoch 3 | Step 12480 | Loss: 0.6259 | LM: 0.1256 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.440 | HR1: 0.401/SR1: 0.395 | LR: 6.38e-05 [2026-05-23 23:21:09] Epoch 3 | Step 12490 | Loss: 0.6258 | LM: 0.1255 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.440 | HR1: 0.401/SR1: 0.395 | LR: 6.33e-05 [2026-05-23 23:21:22] Epoch 3 | Step 12500 | Loss: 0.6258 | LM: 0.1254 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.439 | HR1: 0.401/SR1: 0.395 | LR: 6.28e-05 [2026-05-23 23:21:33] Epoch 3 | Step 12510 | Loss: 0.6258 | LM: 0.1255 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.440 | HR1: 0.401/SR1: 0.395 | LR: 6.23e-05 [2026-05-23 23:21:45] Epoch 3 | Step 12520 | Loss: 0.6258 | LM: 0.1254 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.440 | HR1: 0.401/SR1: 0.395 | LR: 6.18e-05 [2026-05-23 23:21:57] Epoch 3 | Step 12530 | Loss: 0.6258 | LM: 0.1254 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.439 | HR1: 0.401/SR1: 0.395 | LR: 6.13e-05 [2026-05-23 23:22:09] Epoch 3 | Step 12540 | Loss: 0.6258 | LM: 0.1253 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.440 | HR1: 0.401/SR1: 0.395 | LR: 6.08e-05 [2026-05-23 23:22:22] Epoch 3 | Step 12550 | Loss: 0.6258 | LM: 0.1253 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.440 | HR1: 0.401/SR1: 0.395 | LR: 6.03e-05 [2026-05-23 23:22:34] Epoch 3 | Step 12560 | Loss: 0.6259 | LM: 0.1254 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.440 | HR1: 0.401/SR1: 0.395 | LR: 5.98e-05 [2026-05-23 23:22:48] Epoch 3 | Step 12570 | Loss: 0.6259 | LM: 0.1254 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.440 | HR1: 0.401/SR1: 0.395 | LR: 5.93e-05 [2026-05-23 23:22:59] Epoch 3 | Step 12580 | Loss: 0.6258 | LM: 0.1253 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.440 | HR1: 0.401/SR1: 0.395 | LR: 5.88e-05 [2026-05-23 23:23:11] Epoch 3 | Step 12590 | Loss: 0.6258 | LM: 0.1253 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.440 | HR1: 0.401/SR1: 0.395 | LR: 5.83e-05 [2026-05-23 23:23:23] Epoch 3 | Step 12600 | Loss: 0.6258 | LM: 0.1253 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.440 | HR1: 0.401/SR1: 0.395 | LR: 5.78e-05 [2026-05-23 23:23:35] Epoch 3 | Step 12610 | Loss: 0.6259 | LM: 0.1255 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.440 | HR1: 0.401/SR1: 0.395 | LR: 5.73e-05 [2026-05-23 23:23:48] Epoch 3 | Step 12620 | Loss: 0.6259 | LM: 0.1254 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.440 | HR1: 0.401/SR1: 0.395 | LR: 5.68e-05 [2026-05-23 23:24:00] Epoch 3 | Step 12630 | Loss: 0.6258 | LM: 0.1253 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.440 | HR1: 0.401/SR1: 0.395 | LR: 5.63e-05 [2026-05-23 23:24:13] Epoch 3 | Step 12640 | Loss: 0.6259 | LM: 0.1253 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.440 | HR1: 0.401/SR1: 0.395 | LR: 5.58e-05 [2026-05-23 23:24:24] Epoch 3 | Step 12650 | Loss: 0.6259 | LM: 0.1253 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.440 | HR1: 0.401/SR1: 0.395 | LR: 5.53e-05 [2026-05-23 23:24:35] Epoch 3 | Step 12660 | Loss: 0.6259 | LM: 0.1253 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.440 | HR1: 0.401/SR1: 0.395 | LR: 5.48e-05 [2026-05-23 23:24:47] Epoch 3 | Step 12670 | Loss: 0.6259 | LM: 0.1253 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.440 | HR1: 0.401/SR1: 0.395 | LR: 5.43e-05 [2026-05-23 23:24:59] Epoch 3 | Step 12680 | Loss: 0.6258 | LM: 0.1252 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.440 | HR1: 0.401/SR1: 0.395 | LR: 5.38e-05 [2026-05-23 23:25:11] Epoch 3 | Step 12690 | Loss: 0.6258 | LM: 0.1251 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.440 | HR1: 0.401/SR1: 0.395 | LR: 5.33e-05 [2026-05-23 23:25:24] Epoch 3 | Step 12700 | Loss: 0.6257 | LM: 0.1251 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.440 | HR1: 0.401/SR1: 0.395 | LR: 5.28e-05 [2026-05-23 23:25:37] Epoch 3 | Step 12710 | Loss: 0.6257 | LM: 0.1251 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.440 | HR1: 0.401/SR1: 0.395 | LR: 5.23e-05 [2026-05-23 23:25:49] Epoch 3 | Step 12720 | Loss: 0.6256 | LM: 0.1249 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.440 | HR1: 0.401/SR1: 0.395 | LR: 5.18e-05 [2026-05-23 23:26:00] Epoch 3 | Step 12730 | Loss: 0.6257 | LM: 0.1249 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.440 | HR1: 0.401/SR1: 0.395 | LR: 5.13e-05 [2026-05-23 23:26:11] Epoch 3 | Step 12740 | Loss: 0.6257 | LM: 0.1249 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.440 | HR1: 0.401/SR1: 0.395 | LR: 5.08e-05 [2026-05-23 23:26:23] Epoch 3 | Step 12750 | Loss: 0.6257 | LM: 0.1249 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.440 | HR1: 0.401/SR1: 0.395 | LR: 5.03e-05 [2026-05-23 23:26:36] Epoch 3 | Step 12760 | Loss: 0.6257 | LM: 0.1250 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.440 | HR1: 0.401/SR1: 0.395 | LR: 4.98e-05 [2026-05-23 23:26:48] Epoch 3 | Step 12770 | Loss: 0.6257 | LM: 0.1249 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.440 | HR1: 0.401/SR1: 0.395 | LR: 4.93e-05 [2026-05-23 23:27:00] Epoch 3 | Step 12780 | Loss: 0.6258 | LM: 0.1250 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.440 | HR1: 0.401/SR1: 0.395 | LR: 4.88e-05 [2026-05-23 23:27:13] Epoch 3 | Step 12790 | Loss: 0.6258 | LM: 0.1250 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.440 | HR1: 0.401/SR1: 0.395 | LR: 4.83e-05 [2026-05-23 23:27:26] Epoch 3 | Step 12800 | Loss: 0.6259 | LM: 0.1251 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.440 | HR1: 0.401/SR1: 0.395 | LR: 4.78e-05 [2026-05-23 23:27:38] Epoch 3 | Step 12810 | Loss: 0.6259 | LM: 0.1251 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.440 | HR1: 0.401/SR1: 0.395 | LR: 4.73e-05 [2026-05-23 23:27:51] Epoch 3 | Step 12820 | Loss: 0.6259 | LM: 0.1252 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.440 | HR1: 0.401/SR1: 0.395 | LR: 4.68e-05 [2026-05-23 23:28:04] Epoch 3 | Step 12830 | Loss: 0.6259 | LM: 0.1252 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.440 | HR1: 0.401/SR1: 0.395 | LR: 4.63e-05 [2026-05-23 23:28:15] Epoch 3 | Step 12840 | Loss: 0.6259 | LM: 0.1252 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.440 | HR1: 0.401/SR1: 0.395 | LR: 4.58e-05 [2026-05-23 23:28:27] Epoch 3 | Step 12850 | Loss: 0.6259 | LM: 0.1252 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.440 | HR1: 0.401/SR1: 0.395 | LR: 4.53e-05 [2026-05-23 23:28:39] Epoch 3 | Step 12860 | Loss: 0.6260 | LM: 0.1252 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.440 | HR1: 0.401/SR1: 0.395 | LR: 4.48e-05 [2026-05-23 23:28:51] Epoch 3 | Step 12870 | Loss: 0.6259 | LM: 0.1251 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.440 | HR1: 0.401/SR1: 0.395 | LR: 4.43e-05 [2026-05-23 23:29:04] Epoch 3 | Step 12880 | Loss: 0.6260 | LM: 0.1252 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.440 | HR1: 0.401/SR1: 0.395 | LR: 4.39e-05 [2026-05-23 23:29:15] Epoch 3 | Step 12890 | Loss: 0.6260 | LM: 0.1253 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.440 | HR1: 0.402/SR1: 0.395 | LR: 4.34e-05 [2026-05-23 23:29:27] Epoch 3 | Step 12900 | Loss: 0.6259 | LM: 0.1253 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.440 | HR1: 0.402/SR1: 0.395 | LR: 4.29e-05 [2026-05-23 23:29:40] Epoch 3 | Step 12910 | Loss: 0.6260 | LM: 0.1253 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.440 | HR1: 0.402/SR1: 0.395 | LR: 4.24e-05 [2026-05-23 23:29:52] Epoch 3 | Step 12920 | Loss: 0.6260 | LM: 0.1254 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.440 | HR1: 0.402/SR1: 0.395 | LR: 4.19e-05 [2026-05-23 23:30:04] Epoch 3 | Step 12930 | Loss: 0.6261 | LM: 0.1254 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.440 | HR1: 0.402/SR1: 0.395 | LR: 4.14e-05 [2026-05-23 23:30:17] Epoch 3 | Step 12940 | Loss: 0.6260 | LM: 0.1254 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.440 | HR1: 0.401/SR1: 0.395 | LR: 4.10e-05 [2026-05-23 23:30:28] Epoch 3 | Step 12950 | Loss: 0.6260 | LM: 0.1256 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.440 | HR1: 0.401/SR1: 0.395 | LR: 4.05e-05 [2026-05-23 23:30:41] Epoch 3 | Step 12960 | Loss: 0.6261 | LM: 0.1256 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.440 | HR1: 0.401/SR1: 0.395 | LR: 4.00e-05 [2026-05-23 23:30:53] Epoch 3 | Step 12970 | Loss: 0.6260 | LM: 0.1255 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.440 | HR1: 0.401/SR1: 0.395 | LR: 3.95e-05 [2026-05-23 23:31:04] Epoch 3 | Step 12980 | Loss: 0.6260 | LM: 0.1254 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.440 | HR1: 0.401/SR1: 0.395 | LR: 3.91e-05 [2026-05-23 23:31:16] Epoch 3 | Step 12990 | Loss: 0.6260 | LM: 0.1254 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.440 | HR1: 0.401/SR1: 0.395 | LR: 3.86e-05 [2026-05-23 23:31:29] Epoch 3 | Step 13000 | Loss: 0.6260 | LM: 0.1255 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.440 | HR1: 0.401/SR1: 0.395 | LR: 3.81e-05 [2026-05-23 23:31:31] Validation | Batch 10/1463 | Loss: 0.8236 | LM_LOSS: 0.3238 | LB_LOSS: 0.9996 [2026-05-23 23:31:34] Validation | Batch 20/1463 | Loss: 0.8335 | LM_LOSS: 0.3334 | LB_LOSS: 1.0001 [2026-05-23 23:31:38] Validation | Batch 30/1463 | Loss: 0.8315 | LM_LOSS: 0.3312 | LB_LOSS: 1.0005 [2026-05-23 23:31:40] Validation | Batch 40/1463 | Loss: 0.8396 | LM_LOSS: 0.3394 | LB_LOSS: 1.0004 [2026-05-23 23:31:43] Validation | Batch 50/1463 | Loss: 0.8479 | LM_LOSS: 0.3477 | LB_LOSS: 1.0003 [2026-05-23 23:31:47] Validation | Batch 60/1463 | Loss: 0.8486 | LM_LOSS: 0.3484 | LB_LOSS: 1.0004 [2026-05-23 23:31:51] Validation | Batch 70/1463 | Loss: 0.8502 | LM_LOSS: 0.3500 | LB_LOSS: 1.0004 [2026-05-23 23:31:54] Validation | Batch 80/1463 | Loss: 0.8537 | LM_LOSS: 0.3535 | LB_LOSS: 1.0004 [2026-05-23 23:31:57] Validation | Batch 90/1463 | Loss: 0.8540 | LM_LOSS: 0.3538 | LB_LOSS: 1.0004 [2026-05-23 23:32:00] Validation | Batch 100/1463 | Loss: 0.8534 | LM_LOSS: 0.3532 | LB_LOSS: 1.0004 [2026-05-23 23:32:04] Validation | Batch 110/1463 | Loss: 0.8503 | LM_LOSS: 0.3501 | LB_LOSS: 1.0004 [2026-05-23 23:32:06] Validation | Batch 120/1463 | Loss: 0.8535 | LM_LOSS: 0.3533 | LB_LOSS: 1.0004 [2026-05-23 23:32:08] Validation | Batch 130/1463 | Loss: 0.8550 | LM_LOSS: 0.3548 | LB_LOSS: 1.0004 [2026-05-23 23:32:12] Validation | Batch 140/1463 | Loss: 0.8568 | LM_LOSS: 0.3566 | LB_LOSS: 1.0005 [2026-05-23 23:32:15] Validation | Batch 150/1463 | Loss: 0.8538 | LM_LOSS: 0.3535 | LB_LOSS: 1.0005 [2026-05-23 23:32:19] Validation | Batch 160/1463 | Loss: 0.8541 | LM_LOSS: 0.3539 | LB_LOSS: 1.0005 [2026-05-23 23:32:22] Validation | Batch 170/1463 | Loss: 0.8530 | LM_LOSS: 0.3528 | LB_LOSS: 1.0005 [2026-05-23 23:32:25] Validation | Batch 180/1463 | Loss: 0.8541 | LM_LOSS: 0.3538 | LB_LOSS: 1.0005 [2026-05-23 23:32:28] Validation | Batch 190/1463 | Loss: 0.8559 | LM_LOSS: 0.3557 | LB_LOSS: 1.0005 [2026-05-23 23:32:31] Validation | Batch 200/1463 | Loss: 0.8559 | LM_LOSS: 0.3556 | LB_LOSS: 1.0005 [2026-05-23 23:32:34] Validation | Batch 210/1463 | Loss: 0.8570 | LM_LOSS: 0.3568 | LB_LOSS: 1.0005 [2026-05-23 23:32:38] Validation | Batch 220/1463 | Loss: 0.8565 | LM_LOSS: 0.3563 | LB_LOSS: 1.0005 [2026-05-23 23:32:41] Validation | Batch 230/1463 | Loss: 0.8579 | LM_LOSS: 0.3576 | LB_LOSS: 1.0005 [2026-05-23 23:32:44] Validation | Batch 240/1463 | Loss: 0.8598 | LM_LOSS: 0.3595 | LB_LOSS: 1.0005 [2026-05-23 23:32:47] Validation | Batch 250/1463 | Loss: 0.8584 | LM_LOSS: 0.3581 | LB_LOSS: 1.0005 [2026-05-23 23:32:50] Validation | Batch 260/1463 | Loss: 0.8605 | LM_LOSS: 0.3602 | LB_LOSS: 1.0005 [2026-05-23 23:32:53] Validation | Batch 270/1463 | Loss: 0.8585 | LM_LOSS: 0.3583 | LB_LOSS: 1.0005 [2026-05-23 23:32:56] Validation | Batch 280/1463 | Loss: 0.8587 | LM_LOSS: 0.3584 | LB_LOSS: 1.0005 [2026-05-23 23:33:00] Validation | Batch 290/1463 | Loss: 0.8573 | LM_LOSS: 0.3570 | LB_LOSS: 1.0005 [2026-05-23 23:33:02] Validation | Batch 300/1463 | Loss: 0.8582 | LM_LOSS: 0.3580 | LB_LOSS: 1.0005 [2026-05-23 23:33:05] Validation | Batch 310/1463 | Loss: 0.8591 | LM_LOSS: 0.3588 | LB_LOSS: 1.0005 [2026-05-23 23:33:08] Validation | Batch 320/1463 | Loss: 0.8579 | LM_LOSS: 0.3576 | LB_LOSS: 1.0005 [2026-05-23 23:33:11] Validation | Batch 330/1463 | Loss: 0.8572 | LM_LOSS: 0.3569 | LB_LOSS: 1.0005 [2026-05-23 23:33:14] Validation | Batch 340/1463 | Loss: 0.8568 | LM_LOSS: 0.3565 | LB_LOSS: 1.0005 [2026-05-23 23:33:18] Validation | Batch 350/1463 | Loss: 0.8578 | LM_LOSS: 0.3576 | LB_LOSS: 1.0006 [2026-05-23 23:33:20] Validation | Batch 360/1463 | Loss: 0.8578 | LM_LOSS: 0.3576 | LB_LOSS: 1.0006 [2026-05-23 23:33:23] Validation | Batch 370/1463 | Loss: 0.8578 | LM_LOSS: 0.3575 | LB_LOSS: 1.0005 [2026-05-23 23:33:26] Validation | Batch 380/1463 | Loss: 0.8572 | LM_LOSS: 0.3569 | LB_LOSS: 1.0005 [2026-05-23 23:33:29] Validation | Batch 390/1463 | Loss: 0.8578 | LM_LOSS: 0.3576 | LB_LOSS: 1.0005 [2026-05-23 23:33:32] Validation | Batch 400/1463 | Loss: 0.8577 | LM_LOSS: 0.3574 | LB_LOSS: 1.0005 [2026-05-23 23:33:35] Validation | Batch 410/1463 | Loss: 0.8574 | LM_LOSS: 0.3571 | LB_LOSS: 1.0005 [2026-05-23 23:33:38] Validation | Batch 420/1463 | Loss: 0.8577 | LM_LOSS: 0.3575 | LB_LOSS: 1.0005 [2026-05-23 23:33:42] Validation | Batch 430/1463 | Loss: 0.8572 | LM_LOSS: 0.3570 | LB_LOSS: 1.0005 [2026-05-23 23:33:44] Validation | Batch 440/1463 | Loss: 0.8576 | LM_LOSS: 0.3573 | LB_LOSS: 1.0005 [2026-05-23 23:33:47] Validation | Batch 450/1463 | Loss: 0.8572 | LM_LOSS: 0.3569 | LB_LOSS: 1.0006 [2026-05-23 23:33:50] Validation | Batch 460/1463 | Loss: 0.8570 | LM_LOSS: 0.3567 | LB_LOSS: 1.0005 [2026-05-23 23:33:53] Validation | Batch 470/1463 | Loss: 0.8580 | LM_LOSS: 0.3578 | LB_LOSS: 1.0005 [2026-05-23 23:33:57] Validation | Batch 480/1463 | Loss: 0.8577 | LM_LOSS: 0.3574 | LB_LOSS: 1.0005 [2026-05-23 23:34:00] Validation | Batch 490/1463 | Loss: 0.8575 | LM_LOSS: 0.3572 | LB_LOSS: 1.0005 [2026-05-23 23:34:04] Validation | Batch 500/1463 | Loss: 0.8575 | LM_LOSS: 0.3572 | LB_LOSS: 1.0005 [2026-05-23 23:34:07] Validation | Batch 510/1463 | Loss: 0.8566 | LM_LOSS: 0.3564 | LB_LOSS: 1.0005 [2026-05-23 23:34:09] Validation | Batch 520/1463 | Loss: 0.8565 | LM_LOSS: 0.3562 | LB_LOSS: 1.0005 [2026-05-23 23:34:12] Validation | Batch 530/1463 | Loss: 0.8572 | LM_LOSS: 0.3569 | LB_LOSS: 1.0005 [2026-05-23 23:34:15] Validation | Batch 540/1463 | Loss: 0.8573 | LM_LOSS: 0.3570 | LB_LOSS: 1.0005 [2026-05-23 23:34:17] Validation | Batch 550/1463 | Loss: 0.8569 | LM_LOSS: 0.3566 | LB_LOSS: 1.0005 [2026-05-23 23:34:20] Validation | Batch 560/1463 | Loss: 0.8569 | LM_LOSS: 0.3566 | LB_LOSS: 1.0005 [2026-05-23 23:34:23] Validation | Batch 570/1463 | Loss: 0.8568 | LM_LOSS: 0.3566 | LB_LOSS: 1.0005 [2026-05-23 23:34:27] Validation | Batch 580/1463 | Loss: 0.8560 | LM_LOSS: 0.3557 | LB_LOSS: 1.0005 [2026-05-23 23:34:30] Validation | Batch 590/1463 | Loss: 0.8558 | LM_LOSS: 0.3555 | LB_LOSS: 1.0005 [2026-05-23 23:34:33] Validation | Batch 600/1463 | Loss: 0.8561 | LM_LOSS: 0.3558 | LB_LOSS: 1.0005 [2026-05-23 23:34:37] Validation | Batch 610/1463 | Loss: 0.8563 | LM_LOSS: 0.3560 | LB_LOSS: 1.0005 [2026-05-23 23:34:40] Validation | Batch 620/1463 | Loss: 0.8557 | LM_LOSS: 0.3555 | LB_LOSS: 1.0005 [2026-05-23 23:34:43] Validation | Batch 630/1463 | Loss: 0.8551 | LM_LOSS: 0.3548 | LB_LOSS: 1.0005 [2026-05-23 23:34:46] Validation | Batch 640/1463 | Loss: 0.8549 | LM_LOSS: 0.3546 | LB_LOSS: 1.0006 [2026-05-23 23:34:48] Validation | Batch 650/1463 | Loss: 0.8544 | LM_LOSS: 0.3541 | LB_LOSS: 1.0006 [2026-05-23 23:34:51] Validation | Batch 660/1463 | Loss: 0.8538 | LM_LOSS: 0.3536 | LB_LOSS: 1.0005 [2026-05-23 23:34:54] Validation | Batch 670/1463 | Loss: 0.8532 | LM_LOSS: 0.3529 | LB_LOSS: 1.0005 [2026-05-23 23:34:56] Validation | Batch 680/1463 | Loss: 0.8533 | LM_LOSS: 0.3531 | LB_LOSS: 1.0005 [2026-05-23 23:35:00] Validation | Batch 690/1463 | Loss: 0.8535 | LM_LOSS: 0.3533 | LB_LOSS: 1.0005 [2026-05-23 23:35:03] Validation | Batch 700/1463 | Loss: 0.8533 | LM_LOSS: 0.3530 | LB_LOSS: 1.0005 [2026-05-23 23:35:06] Validation | Batch 710/1463 | Loss: 0.8536 | LM_LOSS: 0.3533 | LB_LOSS: 1.0005 [2026-05-23 23:35:09] Validation | Batch 720/1463 | Loss: 0.8543 | LM_LOSS: 0.3540 | LB_LOSS: 1.0005 [2026-05-23 23:35:11] Validation | Batch 730/1463 | Loss: 0.8542 | LM_LOSS: 0.3539 | LB_LOSS: 1.0005 [2026-05-23 23:35:14] Validation | Batch 740/1463 | Loss: 0.8540 | LM_LOSS: 0.3537 | LB_LOSS: 1.0005 [2026-05-23 23:35:16] Validation | Batch 750/1463 | Loss: 0.8538 | LM_LOSS: 0.3536 | LB_LOSS: 1.0005 [2026-05-23 23:35:19] Validation | Batch 760/1463 | Loss: 0.8536 | LM_LOSS: 0.3533 | LB_LOSS: 1.0005 [2026-05-23 23:35:21] Validation | Batch 770/1463 | Loss: 0.8537 | LM_LOSS: 0.3534 | LB_LOSS: 1.0005 [2026-05-23 23:35:24] Validation | Batch 780/1463 | Loss: 0.8536 | LM_LOSS: 0.3533 | LB_LOSS: 1.0005 [2026-05-23 23:35:26] Validation | Batch 790/1463 | Loss: 0.8539 | LM_LOSS: 0.3536 | LB_LOSS: 1.0005 [2026-05-23 23:35:29] Validation | Batch 800/1463 | Loss: 0.8538 | LM_LOSS: 0.3536 | LB_LOSS: 1.0005 [2026-05-23 23:35:32] Validation | Batch 810/1463 | Loss: 0.8535 | LM_LOSS: 0.3532 | LB_LOSS: 1.0005 [2026-05-23 23:35:35] Validation | Batch 820/1463 | Loss: 0.8537 | LM_LOSS: 0.3534 | LB_LOSS: 1.0005 [2026-05-23 23:35:38] Validation | Batch 830/1463 | Loss: 0.8538 | LM_LOSS: 0.3536 | LB_LOSS: 1.0005 [2026-05-23 23:35:42] Validation | Batch 840/1463 | Loss: 0.8539 | LM_LOSS: 0.3536 | LB_LOSS: 1.0005 [2026-05-23 23:35:45] Validation | Batch 850/1463 | Loss: 0.8543 | LM_LOSS: 0.3540 | LB_LOSS: 1.0005 [2026-05-23 23:35:48] Validation | Batch 860/1463 | Loss: 0.8547 | LM_LOSS: 0.3544 | LB_LOSS: 1.0005 [2026-05-23 23:35:51] Validation | Batch 870/1463 | Loss: 0.8548 | LM_LOSS: 0.3546 | LB_LOSS: 1.0005 [2026-05-23 23:35:54] Validation | Batch 880/1463 | Loss: 0.8540 | LM_LOSS: 0.3537 | LB_LOSS: 1.0005 [2026-05-23 23:35:57] Validation | Batch 890/1463 | Loss: 0.8540 | LM_LOSS: 0.3537 | LB_LOSS: 1.0005 [2026-05-23 23:36:00] Validation | Batch 900/1463 | Loss: 0.8536 | LM_LOSS: 0.3533 | LB_LOSS: 1.0005 [2026-05-23 23:36:03] Validation | Batch 910/1463 | Loss: 0.8535 | LM_LOSS: 0.3533 | LB_LOSS: 1.0005 [2026-05-23 23:36:07] Validation | Batch 920/1463 | Loss: 0.8539 | LM_LOSS: 0.3536 | LB_LOSS: 1.0005 [2026-05-23 23:36:09] Validation | Batch 930/1463 | Loss: 0.8536 | LM_LOSS: 0.3533 | LB_LOSS: 1.0005 [2026-05-23 23:36:13] Validation | Batch 940/1463 | Loss: 0.8533 | LM_LOSS: 0.3531 | LB_LOSS: 1.0005 [2026-05-23 23:36:16] Validation | Batch 950/1463 | Loss: 0.8535 | LM_LOSS: 0.3533 | LB_LOSS: 1.0005 [2026-05-23 23:36:19] Validation | Batch 960/1463 | Loss: 0.8536 | LM_LOSS: 0.3533 | LB_LOSS: 1.0005 [2026-05-23 23:36:23] Validation | Batch 970/1463 | Loss: 0.8536 | LM_LOSS: 0.3533 | LB_LOSS: 1.0005 [2026-05-23 23:36:26] Validation | Batch 980/1463 | Loss: 0.8540 | LM_LOSS: 0.3537 | LB_LOSS: 1.0005 [2026-05-23 23:36:29] Validation | Batch 990/1463 | Loss: 0.8543 | LM_LOSS: 0.3540 | LB_LOSS: 1.0005 [2026-05-23 23:36:32] Validation | Batch 1000/1463 | Loss: 0.8549 | LM_LOSS: 0.3547 | LB_LOSS: 1.0005 [2026-05-23 23:36:35] Validation | Batch 1010/1463 | Loss: 0.8549 | LM_LOSS: 0.3546 | LB_LOSS: 1.0005 [2026-05-23 23:36:38] Validation | Batch 1020/1463 | Loss: 0.8549 | LM_LOSS: 0.3546 | LB_LOSS: 1.0005 [2026-05-23 23:36:41] Validation | Batch 1030/1463 | Loss: 0.8554 | LM_LOSS: 0.3551 | LB_LOSS: 1.0005 [2026-05-23 23:36:44] Validation | Batch 1040/1463 | Loss: 0.8550 | LM_LOSS: 0.3547 | LB_LOSS: 1.0005 [2026-05-23 23:36:47] Validation | Batch 1050/1463 | Loss: 0.8550 | LM_LOSS: 0.3547 | LB_LOSS: 1.0005 [2026-05-23 23:36:50] Validation | Batch 1060/1463 | Loss: 0.8550 | LM_LOSS: 0.3547 | LB_LOSS: 1.0005 [2026-05-23 23:36:54] Validation | Batch 1070/1463 | Loss: 0.8550 | LM_LOSS: 0.3548 | LB_LOSS: 1.0005 [2026-05-23 23:36:56] Validation | Batch 1080/1463 | Loss: 0.8548 | LM_LOSS: 0.3546 | LB_LOSS: 1.0005 [2026-05-23 23:36:59] Validation | Batch 1090/1463 | Loss: 0.8548 | LM_LOSS: 0.3545 | LB_LOSS: 1.0005 [2026-05-23 23:37:02] Validation | Batch 1100/1463 | Loss: 0.8544 | LM_LOSS: 0.3542 | LB_LOSS: 1.0005 [2026-05-23 23:37:04] Validation | Batch 1110/1463 | Loss: 0.8545 | LM_LOSS: 0.3542 | LB_LOSS: 1.0005 [2026-05-23 23:37:07] Validation | Batch 1120/1463 | Loss: 0.8542 | LM_LOSS: 0.3539 | LB_LOSS: 1.0005 [2026-05-23 23:37:11] Validation | Batch 1130/1463 | Loss: 0.8540 | LM_LOSS: 0.3537 | LB_LOSS: 1.0005 [2026-05-23 23:37:14] Validation | Batch 1140/1463 | Loss: 0.8543 | LM_LOSS: 0.3540 | LB_LOSS: 1.0005 [2026-05-23 23:37:18] Validation | Batch 1150/1463 | Loss: 0.8540 | LM_LOSS: 0.3538 | LB_LOSS: 1.0005 [2026-05-23 23:37:20] Validation | Batch 1160/1463 | Loss: 0.8539 | LM_LOSS: 0.3537 | LB_LOSS: 1.0005 [2026-05-23 23:37:24] Validation | Batch 1170/1463 | Loss: 0.8540 | LM_LOSS: 0.3538 | LB_LOSS: 1.0005 [2026-05-23 23:37:26] Validation | Batch 1180/1463 | Loss: 0.8541 | LM_LOSS: 0.3538 | LB_LOSS: 1.0005 [2026-05-23 23:37:30] Validation | Batch 1190/1463 | Loss: 0.8543 | LM_LOSS: 0.3541 | LB_LOSS: 1.0005 [2026-05-23 23:37:33] Validation | Batch 1200/1463 | Loss: 0.8543 | LM_LOSS: 0.3541 | LB_LOSS: 1.0005 [2026-05-23 23:37:36] Validation | Batch 1210/1463 | Loss: 0.8546 | LM_LOSS: 0.3544 | LB_LOSS: 1.0005 [2026-05-23 23:37:39] Validation | Batch 1220/1463 | Loss: 0.8550 | LM_LOSS: 0.3547 | LB_LOSS: 1.0005 [2026-05-23 23:37:41] Validation | Batch 1230/1463 | Loss: 0.8551 | LM_LOSS: 0.3548 | LB_LOSS: 1.0005 [2026-05-23 23:37:44] Validation | Batch 1240/1463 | Loss: 0.8552 | LM_LOSS: 0.3549 | LB_LOSS: 1.0005 [2026-05-23 23:37:48] Validation | Batch 1250/1463 | Loss: 0.8553 | LM_LOSS: 0.3550 | LB_LOSS: 1.0005 [2026-05-23 23:37:51] Validation | Batch 1260/1463 | Loss: 0.8552 | LM_LOSS: 0.3550 | LB_LOSS: 1.0005 [2026-05-23 23:37:54] Validation | Batch 1270/1463 | Loss: 0.8550 | LM_LOSS: 0.3547 | LB_LOSS: 1.0005 [2026-05-23 23:37:57] Validation | Batch 1280/1463 | Loss: 0.8548 | LM_LOSS: 0.3546 | LB_LOSS: 1.0005 [2026-05-23 23:38:00] Validation | Batch 1290/1463 | Loss: 0.8548 | LM_LOSS: 0.3545 | LB_LOSS: 1.0005 [2026-05-23 23:38:03] Validation | Batch 1300/1463 | Loss: 0.8545 | LM_LOSS: 0.3542 | LB_LOSS: 1.0005 [2026-05-23 23:38:07] Validation | Batch 1310/1463 | Loss: 0.8543 | LM_LOSS: 0.3540 | LB_LOSS: 1.0005 [2026-05-23 23:38:09] Validation | Batch 1320/1463 | Loss: 0.8543 | LM_LOSS: 0.3541 | LB_LOSS: 1.0005 [2026-05-23 23:38:11] Validation | Batch 1330/1463 | Loss: 0.8543 | LM_LOSS: 0.3541 | LB_LOSS: 1.0005 [2026-05-23 23:38:14] Validation | Batch 1340/1463 | Loss: 0.8547 | LM_LOSS: 0.3545 | LB_LOSS: 1.0005 [2026-05-23 23:38:17] Validation | Batch 1350/1463 | Loss: 0.8547 | LM_LOSS: 0.3544 | LB_LOSS: 1.0005 [2026-05-23 23:38:21] Validation | Batch 1360/1463 | Loss: 0.8550 | LM_LOSS: 0.3547 | LB_LOSS: 1.0005 [2026-05-23 23:38:24] Validation | Batch 1370/1463 | Loss: 0.8553 | LM_LOSS: 0.3551 | LB_LOSS: 1.0005 [2026-05-23 23:38:27] Validation | Batch 1380/1463 | Loss: 0.8550 | LM_LOSS: 0.3547 | LB_LOSS: 1.0005 [2026-05-23 23:38:30] Validation | Batch 1390/1463 | Loss: 0.8550 | LM_LOSS: 0.3548 | LB_LOSS: 1.0005 [2026-05-23 23:38:33] Validation | Batch 1400/1463 | Loss: 0.8553 | LM_LOSS: 0.3551 | LB_LOSS: 1.0005 [2026-05-23 23:38:35] Validation | Batch 1410/1463 | Loss: 0.8557 | LM_LOSS: 0.3554 | LB_LOSS: 1.0005 [2026-05-23 23:38:38] Validation | Batch 1420/1463 | Loss: 0.8556 | LM_LOSS: 0.3554 | LB_LOSS: 1.0005 [2026-05-23 23:38:42] Validation | Batch 1430/1463 | Loss: 0.8556 | LM_LOSS: 0.3554 | LB_LOSS: 1.0005 [2026-05-23 23:38:45] Validation | Batch 1440/1463 | Loss: 0.8561 | LM_LOSS: 0.3559 | LB_LOSS: 1.0005 [2026-05-23 23:38:48] Validation | Batch 1450/1463 | Loss: 0.8558 | LM_LOSS: 0.3556 | LB_LOSS: 1.0005 [2026-05-23 23:38:51] Validation | Batch 1460/1463 | Loss: 0.8562 | LM_LOSS: 0.3560 | LB_LOSS: 1.0005 [2026-05-23 23:38:52] Validation | Batch 1463/1463 | Loss: 0.8561 | LM_LOSS: 0.3559 | LB_LOSS: 1.0005 [2026-05-23 23:38:52] Validation | Loss: 0.8561 | LM_LOSS: 0.3559 | LB_LOSS: 1.0005 | PPL: 1.43 | Time: 443.49s [2026-05-23 23:39:05] Epoch 3 | Step 13010 | Loss: 0.6260 | LM: 0.1256 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.440 | HR1: 0.401/SR1: 0.395 | LR: 3.77e-05 [2026-05-23 23:39:18] Epoch 3 | Step 13020 | Loss: 0.6259 | LM: 0.1256 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.440 | HR1: 0.401/SR1: 0.395 | LR: 3.72e-05 [2026-05-23 23:39:31] Epoch 3 | Step 13030 | Loss: 0.6259 | LM: 0.1256 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.440 | HR1: 0.401/SR1: 0.395 | LR: 3.67e-05 [2026-05-23 23:39:45] Epoch 3 | Step 13040 | Loss: 0.6259 | LM: 0.1257 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.440 | HR1: 0.401/SR1: 0.395 | LR: 3.63e-05 [2026-05-23 23:39:57] Epoch 3 | Step 13050 | Loss: 0.6259 | LM: 0.1256 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.440 | HR1: 0.401/SR1: 0.395 | LR: 3.58e-05 [2026-05-23 23:40:09] Epoch 3 | Step 13060 | Loss: 0.6259 | LM: 0.1256 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.440 | HR1: 0.401/SR1: 0.395 | LR: 3.54e-05 [2026-05-23 23:40:22] Epoch 3 | Step 13070 | Loss: 0.6259 | LM: 0.1257 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.440 | HR1: 0.401/SR1: 0.395 | LR: 3.49e-05 [2026-05-23 23:40:35] Epoch 3 | Step 13080 | Loss: 0.6258 | LM: 0.1255 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.440 | HR1: 0.402/SR1: 0.395 | LR: 3.45e-05 [2026-05-23 23:40:47] Epoch 3 | Step 13090 | Loss: 0.6259 | LM: 0.1257 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.440 | HR1: 0.402/SR1: 0.395 | LR: 3.40e-05 [2026-05-23 23:40:59] Epoch 3 | Step 13100 | Loss: 0.6259 | LM: 0.1257 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.440 | HR1: 0.401/SR1: 0.395 | LR: 3.36e-05 [2026-05-23 23:41:13] Epoch 3 | Step 13110 | Loss: 0.6259 | LM: 0.1258 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.440 | HR1: 0.402/SR1: 0.395 | LR: 3.31e-05 [2026-05-23 23:41:27] Epoch 3 | Step 13120 | Loss: 0.6259 | LM: 0.1258 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.440 | HR1: 0.402/SR1: 0.395 | LR: 3.27e-05 [2026-05-23 23:41:40] Epoch 3 | Step 13130 | Loss: 0.6259 | LM: 0.1258 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.440 | HR1: 0.402/SR1: 0.395 | LR: 3.23e-05 [2026-05-23 23:41:54] Epoch 3 | Step 13140 | Loss: 0.6259 | LM: 0.1257 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.440 | HR1: 0.402/SR1: 0.395 | LR: 3.18e-05 [2026-05-23 23:42:06] Epoch 3 | Step 13150 | Loss: 0.6259 | LM: 0.1257 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.440 | HR1: 0.402/SR1: 0.395 | LR: 3.14e-05 [2026-05-23 23:42:18] Epoch 3 | Step 13160 | Loss: 0.6258 | LM: 0.1256 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.440 | HR1: 0.402/SR1: 0.395 | LR: 3.10e-05 [2026-05-23 23:42:32] Epoch 3 | Step 13170 | Loss: 0.6258 | LM: 0.1257 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.440 | HR1: 0.402/SR1: 0.395 | LR: 3.06e-05 [2026-05-23 23:42:45] Epoch 3 | Step 13180 | Loss: 0.6258 | LM: 0.1258 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.440 | HR1: 0.401/SR1: 0.395 | LR: 3.01e-05 [2026-05-23 23:42:59] Epoch 3 | Step 13190 | Loss: 0.6257 | LM: 0.1257 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.440 | HR1: 0.402/SR1: 0.395 | LR: 2.97e-05 [2026-05-23 23:43:11] Epoch 3 | Step 13200 | Loss: 0.6257 | LM: 0.1256 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.440 | HR1: 0.402/SR1: 0.395 | LR: 2.93e-05 [2026-05-23 23:43:25] Epoch 3 | Step 13210 | Loss: 0.6258 | LM: 0.1256 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.440 | HR1: 0.402/SR1: 0.395 | LR: 2.89e-05 [2026-05-23 23:43:37] Epoch 3 | Step 13220 | Loss: 0.6258 | LM: 0.1258 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.440 | HR1: 0.402/SR1: 0.395 | LR: 2.85e-05 [2026-05-23 23:43:50] Epoch 3 | Step 13230 | Loss: 0.6258 | LM: 0.1257 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.440 | HR1: 0.401/SR1: 0.395 | LR: 2.81e-05 [2026-05-23 23:44:03] Epoch 3 | Step 13240 | Loss: 0.6258 | LM: 0.1257 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.440 | HR1: 0.401/SR1: 0.395 | LR: 2.77e-05 [2026-05-23 23:44:15] Epoch 3 | Step 13250 | Loss: 0.6258 | LM: 0.1256 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.440 | HR1: 0.401/SR1: 0.395 | LR: 2.73e-05 [2026-05-23 23:44:28] Epoch 3 | Step 13260 | Loss: 0.6258 | LM: 0.1257 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.440 | HR1: 0.402/SR1: 0.395 | LR: 2.69e-05 [2026-05-23 23:44:40] Epoch 3 | Step 13270 | Loss: 0.6258 | LM: 0.1257 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.440 | HR1: 0.401/SR1: 0.395 | LR: 2.65e-05 [2026-05-23 23:44:53] Epoch 3 | Step 13280 | Loss: 0.6257 | LM: 0.1257 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.440 | HR1: 0.401/SR1: 0.395 | LR: 2.61e-05 [2026-05-23 23:45:06] Epoch 3 | Step 13290 | Loss: 0.6257 | LM: 0.1257 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.440 | HR1: 0.401/SR1: 0.395 | LR: 2.57e-05 [2026-05-23 23:45:18] Epoch 3 | Step 13300 | Loss: 0.6257 | LM: 0.1256 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.440 | HR1: 0.401/SR1: 0.395 | LR: 2.53e-05 [2026-05-23 23:45:30] Epoch 3 | Step 13310 | Loss: 0.6257 | LM: 0.1256 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.440 | HR1: 0.401/SR1: 0.395 | LR: 2.50e-05 [2026-05-23 23:45:42] Epoch 3 | Step 13320 | Loss: 0.6257 | LM: 0.1256 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.440 | HR1: 0.401/SR1: 0.395 | LR: 2.46e-05 [2026-05-23 23:45:54] Epoch 3 | Step 13330 | Loss: 0.6257 | LM: 0.1255 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.440 | HR1: 0.401/SR1: 0.395 | LR: 2.42e-05 [2026-05-23 23:46:07] Epoch 3 | Step 13340 | Loss: 0.6257 | LM: 0.1256 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.440 | HR1: 0.401/SR1: 0.395 | LR: 2.39e-05 [2026-05-23 23:46:19] Epoch 3 | Step 13350 | Loss: 0.6257 | LM: 0.1256 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.440 | HR1: 0.401/SR1: 0.395 | LR: 2.35e-05 [2026-05-23 23:46:32] Epoch 3 | Step 13360 | Loss: 0.6257 | LM: 0.1257 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.440 | HR1: 0.401/SR1: 0.395 | LR: 2.31e-05 [2026-05-23 23:46:45] Epoch 3 | Step 13370 | Loss: 0.6257 | LM: 0.1256 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.440 | HR1: 0.401/SR1: 0.395 | LR: 2.28e-05 [2026-05-23 23:46:58] Epoch 3 | Step 13380 | Loss: 0.6257 | LM: 0.1258 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.440 | HR1: 0.401/SR1: 0.395 | LR: 2.24e-05 [2026-05-23 23:47:10] Epoch 3 | Step 13390 | Loss: 0.6257 | LM: 0.1257 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.440 | HR1: 0.401/SR1: 0.395 | LR: 2.21e-05 [2026-05-23 23:47:23] Epoch 3 | Step 13400 | Loss: 0.6257 | LM: 0.1257 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.440 | HR1: 0.401/SR1: 0.395 | LR: 2.18e-05 [2026-05-23 23:47:36] Epoch 3 | Step 13410 | Loss: 0.6257 | LM: 0.1258 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.440 | HR1: 0.401/SR1: 0.395 | LR: 2.14e-05 [2026-05-23 23:47:49] Epoch 3 | Step 13420 | Loss: 0.6258 | LM: 0.1257 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.440 | HR1: 0.401/SR1: 0.395 | LR: 2.11e-05 [2026-05-23 23:48:03] Epoch 3 | Step 13430 | Loss: 0.6258 | LM: 0.1257 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.440 | HR1: 0.401/SR1: 0.395 | LR: 2.08e-05 [2026-05-23 23:48:16] Epoch 3 | Step 13440 | Loss: 0.6257 | LM: 0.1257 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.440 | HR1: 0.401/SR1: 0.395 | LR: 2.04e-05 [2026-05-23 23:48:28] Epoch 3 | Step 13450 | Loss: 0.6257 | LM: 0.1256 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.440 | HR1: 0.401/SR1: 0.395 | LR: 2.01e-05 [2026-05-23 23:48:42] Epoch 3 | Step 13460 | Loss: 0.6257 | LM: 0.1257 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.440 | HR1: 0.402/SR1: 0.395 | LR: 1.98e-05 [2026-05-23 23:48:54] Epoch 3 | Step 13470 | Loss: 0.6257 | LM: 0.1257 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.440 | HR1: 0.402/SR1: 0.395 | LR: 1.95e-05 [2026-05-23 23:49:07] Epoch 3 | Step 13480 | Loss: 0.6257 | LM: 0.1256 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.440 | HR1: 0.401/SR1: 0.395 | LR: 1.92e-05 [2026-05-23 23:49:19] Epoch 3 | Step 13490 | Loss: 0.6258 | LM: 0.1258 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.440 | HR1: 0.401/SR1: 0.395 | LR: 1.89e-05 [2026-05-23 23:49:32] Epoch 3 | Step 13500 | Loss: 0.6258 | LM: 0.1257 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.440 | HR1: 0.401/SR1: 0.395 | LR: 1.86e-05 [2026-05-23 23:49:44] Epoch 3 | Step 13510 | Loss: 0.6258 | LM: 0.1257 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.440 | HR1: 0.401/SR1: 0.395 | LR: 1.83e-05 [2026-05-23 23:49:56] Epoch 3 | Step 13520 | Loss: 0.6258 | LM: 0.1256 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.440 | HR1: 0.401/SR1: 0.395 | LR: 1.80e-05 [2026-05-23 23:50:09] Epoch 3 | Step 13530 | Loss: 0.6258 | LM: 0.1256 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.440 | HR1: 0.401/SR1: 0.395 | LR: 1.77e-05 [2026-05-23 23:50:22] Epoch 3 | Step 13540 | Loss: 0.6258 | LM: 0.1256 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.440 | HR1: 0.401/SR1: 0.395 | LR: 1.74e-05 [2026-05-23 23:50:34] Epoch 3 | Step 13550 | Loss: 0.6258 | LM: 0.1257 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.440 | HR1: 0.401/SR1: 0.395 | LR: 1.72e-05 [2026-05-23 23:50:46] Epoch 3 | Step 13560 | Loss: 0.6258 | LM: 0.1257 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.440 | HR1: 0.401/SR1: 0.395 | LR: 1.69e-05 [2026-05-23 23:50:59] Epoch 3 | Step 13570 | Loss: 0.6258 | LM: 0.1256 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.440 | HR1: 0.401/SR1: 0.395 | LR: 1.66e-05 [2026-05-23 23:51:12] Epoch 3 | Step 13580 | Loss: 0.6258 | LM: 0.1257 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.440 | HR1: 0.401/SR1: 0.395 | LR: 1.64e-05 [2026-05-23 23:51:25] Epoch 3 | Step 13590 | Loss: 0.6258 | LM: 0.1257 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.440 | HR1: 0.401/SR1: 0.395 | LR: 1.61e-05 [2026-05-23 23:51:37] Epoch 3 | Step 13600 | Loss: 0.6259 | LM: 0.1257 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.440 | HR1: 0.401/SR1: 0.395 | LR: 1.59e-05 [2026-05-23 23:51:48] Epoch 3 | Step 13610 | Loss: 0.6258 | LM: 0.1256 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.440 | HR1: 0.401/SR1: 0.395 | LR: 1.56e-05 [2026-05-23 23:52:01] Epoch 3 | Step 13620 | Loss: 0.6259 | LM: 0.1257 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.440 | HR1: 0.401/SR1: 0.395 | LR: 1.54e-05 [2026-05-23 23:52:14] Epoch 3 | Step 13630 | Loss: 0.6259 | LM: 0.1257 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.440 | HR1: 0.401/SR1: 0.395 | LR: 1.51e-05 [2026-05-23 23:52:26] Epoch 3 | Step 13640 | Loss: 0.6258 | LM: 0.1256 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.440 | HR1: 0.401/SR1: 0.395 | LR: 1.49e-05 [2026-05-23 23:52:39] Epoch 3 | Step 13650 | Loss: 0.6258 | LM: 0.1257 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.440 | HR1: 0.401/SR1: 0.395 | LR: 1.47e-05 [2026-05-23 23:52:51] Epoch 3 | Step 13660 | Loss: 0.6258 | LM: 0.1256 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.440 | HR1: 0.401/SR1: 0.395 | LR: 1.45e-05 [2026-05-23 23:53:04] Epoch 3 | Step 13670 | Loss: 0.6258 | LM: 0.1256 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.440 | HR1: 0.401/SR1: 0.395 | LR: 1.42e-05 [2026-05-23 23:53:18] Epoch 3 | Step 13680 | Loss: 0.6259 | LM: 0.1257 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.440 | HR1: 0.401/SR1: 0.395 | LR: 1.40e-05 [2026-05-23 23:53:31] Epoch 3 | Step 13690 | Loss: 0.6258 | LM: 0.1257 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.440 | HR1: 0.401/SR1: 0.395 | LR: 1.38e-05 [2026-05-23 23:53:43] Epoch 3 | Step 13700 | Loss: 0.6258 | LM: 0.1257 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.440 | HR1: 0.401/SR1: 0.395 | LR: 1.36e-05 [2026-05-23 23:53:57] Epoch 3 | Step 13710 | Loss: 0.6258 | LM: 0.1257 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.440 | HR1: 0.401/SR1: 0.395 | LR: 1.34e-05 [2026-05-23 23:54:09] Epoch 3 | Step 13720 | Loss: 0.6258 | LM: 0.1258 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.440 | HR1: 0.401/SR1: 0.395 | LR: 1.32e-05 [2026-05-23 23:54:22] Epoch 3 | Step 13730 | Loss: 0.6258 | LM: 0.1257 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.440 | HR1: 0.401/SR1: 0.395 | LR: 1.31e-05 [2026-05-23 23:54:34] Epoch 3 | Step 13740 | Loss: 0.6258 | LM: 0.1257 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.440 | HR1: 0.401/SR1: 0.395 | LR: 1.29e-05 [2026-05-23 23:54:46] Epoch 3 | Step 13750 | Loss: 0.6258 | LM: 0.1257 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.440 | HR1: 0.401/SR1: 0.395 | LR: 1.27e-05 [2026-05-23 23:54:58] Epoch 3 | Step 13760 | Loss: 0.6258 | LM: 0.1257 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.440 | HR1: 0.401/SR1: 0.395 | LR: 1.25e-05 [2026-05-23 23:55:09] Epoch 3 | Step 13770 | Loss: 0.6258 | LM: 0.1258 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.440 | HR1: 0.401/SR1: 0.395 | LR: 1.24e-05 [2026-05-23 23:55:21] Epoch 3 | Step 13780 | Loss: 0.6258 | LM: 0.1259 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.440 | HR1: 0.401/SR1: 0.395 | LR: 1.22e-05 [2026-05-23 23:55:33] Epoch 3 | Step 13790 | Loss: 0.6258 | LM: 0.1259 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.440 | HR1: 0.401/SR1: 0.395 | LR: 1.21e-05 [2026-05-23 23:55:44] Epoch 3 | Step 13800 | Loss: 0.6258 | LM: 0.1258 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.440 | HR1: 0.401/SR1: 0.395 | LR: 1.19e-05 [2026-05-23 23:55:57] Epoch 3 | Step 13810 | Loss: 0.6258 | LM: 0.1258 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.440 | HR1: 0.401/SR1: 0.395 | LR: 1.18e-05 [2026-05-23 23:56:10] Epoch 3 | Step 13820 | Loss: 0.6258 | LM: 0.1260 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.440 | HR1: 0.401/SR1: 0.395 | LR: 1.16e-05 [2026-05-23 23:56:22] Epoch 3 | Step 13830 | Loss: 0.6258 | LM: 0.1259 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.440 | HR1: 0.401/SR1: 0.395 | LR: 1.15e-05 [2026-05-23 23:56:33] Epoch 3 | Step 13840 | Loss: 0.6258 | LM: 0.1259 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.440 | HR1: 0.401/SR1: 0.395 | LR: 1.14e-05 [2026-05-23 23:56:46] Epoch 3 | Step 13850 | Loss: 0.6258 | LM: 0.1259 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.440 | HR1: 0.401/SR1: 0.395 | LR: 1.13e-05 [2026-05-23 23:56:58] Epoch 3 | Step 13860 | Loss: 0.6257 | LM: 0.1258 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.440 | HR1: 0.402/SR1: 0.395 | LR: 1.11e-05 [2026-05-23 23:57:11] Epoch 3 | Step 13870 | Loss: 0.6257 | LM: 0.1259 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.440 | HR1: 0.402/SR1: 0.395 | LR: 1.10e-05 [2026-05-23 23:57:24] Epoch 3 | Step 13880 | Loss: 0.6257 | LM: 0.1258 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.440 | HR1: 0.402/SR1: 0.395 | LR: 1.09e-05 [2026-05-23 23:57:37] Epoch 3 | Step 13890 | Loss: 0.6257 | LM: 0.1258 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.440 | HR1: 0.402/SR1: 0.395 | LR: 1.08e-05 [2026-05-23 23:57:50] Epoch 3 | Step 13900 | Loss: 0.6256 | LM: 0.1258 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.440 | HR1: 0.402/SR1: 0.395 | LR: 1.07e-05 [2026-05-23 23:58:04] Epoch 3 | Step 13910 | Loss: 0.6256 | LM: 0.1258 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.440 | HR1: 0.402/SR1: 0.395 | LR: 1.06e-05 [2026-05-23 23:58:17] Epoch 3 | Step 13920 | Loss: 0.6256 | LM: 0.1258 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.440 | HR1: 0.402/SR1: 0.395 | LR: 1.06e-05 [2026-05-23 23:58:28] Epoch 3 | Step 13930 | Loss: 0.6256 | LM: 0.1258 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.440 | HR1: 0.402/SR1: 0.395 | LR: 1.05e-05 [2026-05-23 23:58:41] Epoch 3 | Step 13940 | Loss: 0.6256 | LM: 0.1258 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.440 | HR1: 0.402/SR1: 0.395 | LR: 1.04e-05 [2026-05-23 23:58:53] Epoch 3 | Step 13950 | Loss: 0.6256 | LM: 0.1258 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.440 | HR1: 0.402/SR1: 0.395 | LR: 1.04e-05 [2026-05-23 23:59:05] Epoch 3 | Step 13960 | Loss: 0.6256 | LM: 0.1257 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.440 | HR1: 0.402/SR1: 0.395 | LR: 1.03e-05 [2026-05-23 23:59:16] Epoch 3 | Step 13970 | Loss: 0.6256 | LM: 0.1256 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.440 | HR1: 0.402/SR1: 0.395 | LR: 1.02e-05 [2026-05-23 23:59:29] Epoch 3 | Step 13980 | Loss: 0.6256 | LM: 0.1256 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.440 | HR1: 0.402/SR1: 0.395 | LR: 1.02e-05 [2026-05-23 23:59:41] Epoch 3 | Step 13990 | Loss: 0.6256 | LM: 0.1256 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.440 | HR1: 0.401/SR1: 0.395 | LR: 1.01e-05 [2026-05-23 23:59:54] Epoch 3 | Step 14000 | Loss: 0.6256 | LM: 0.1255 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.440 | HR1: 0.402/SR1: 0.395 | LR: 1.01e-05 [2026-05-23 23:59:56] Validation | Batch 10/1463 | Loss: 0.8212 | LM_LOSS: 0.3214 | LB_LOSS: 0.9997 [2026-05-23 23:59:59] Validation | Batch 20/1463 | Loss: 0.8314 | LM_LOSS: 0.3313 | LB_LOSS: 1.0002 [2026-05-24 00:00:02] Validation | Batch 30/1463 | Loss: 0.8295 | LM_LOSS: 0.3292 | LB_LOSS: 1.0005 [2026-05-24 00:00:05] Validation | Batch 40/1463 | Loss: 0.8376 | LM_LOSS: 0.3374 | LB_LOSS: 1.0005 [2026-05-24 00:00:08] Validation | Batch 50/1463 | Loss: 0.8459 | LM_LOSS: 0.3457 | LB_LOSS: 1.0004 [2026-05-24 00:00:12] Validation | Batch 60/1463 | Loss: 0.8467 | LM_LOSS: 0.3464 | LB_LOSS: 1.0004 [2026-05-24 00:00:16] Validation | Batch 70/1463 | Loss: 0.8482 | LM_LOSS: 0.3479 | LB_LOSS: 1.0005 [2026-05-24 00:00:19] Validation | Batch 80/1463 | Loss: 0.8517 | LM_LOSS: 0.3515 | LB_LOSS: 1.0005 [2026-05-24 00:00:22] Validation | Batch 90/1463 | Loss: 0.8521 | LM_LOSS: 0.3518 | LB_LOSS: 1.0005 [2026-05-24 00:00:25] Validation | Batch 100/1463 | Loss: 0.8515 | LM_LOSS: 0.3513 | LB_LOSS: 1.0004 [2026-05-24 00:00:28] Validation | Batch 110/1463 | Loss: 0.8487 | LM_LOSS: 0.3484 | LB_LOSS: 1.0005 [2026-05-24 00:00:30] Validation | Batch 120/1463 | Loss: 0.8518 | LM_LOSS: 0.3516 | LB_LOSS: 1.0004 [2026-05-24 00:00:33] Validation | Batch 130/1463 | Loss: 0.8534 | LM_LOSS: 0.3531 | LB_LOSS: 1.0005 [2026-05-24 00:00:37] Validation | Batch 140/1463 | Loss: 0.8552 | LM_LOSS: 0.3550 | LB_LOSS: 1.0005 [2026-05-24 00:00:40] Validation | Batch 150/1463 | Loss: 0.8523 | LM_LOSS: 0.3520 | LB_LOSS: 1.0005 [2026-05-24 00:00:43] Validation | Batch 160/1463 | Loss: 0.8526 | LM_LOSS: 0.3524 | LB_LOSS: 1.0005 [2026-05-24 00:00:46] Validation | Batch 170/1463 | Loss: 0.8516 | LM_LOSS: 0.3513 | LB_LOSS: 1.0005 [2026-05-24 00:00:50] Validation | Batch 180/1463 | Loss: 0.8526 | LM_LOSS: 0.3523 | LB_LOSS: 1.0005 [2026-05-24 00:00:53] Validation | Batch 190/1463 | Loss: 0.8543 | LM_LOSS: 0.3540 | LB_LOSS: 1.0005 [2026-05-24 00:00:56] Validation | Batch 200/1463 | Loss: 0.8543 | LM_LOSS: 0.3540 | LB_LOSS: 1.0005 [2026-05-24 00:00:58] Validation | Batch 210/1463 | Loss: 0.8554 | LM_LOSS: 0.3551 | LB_LOSS: 1.0005 [2026-05-24 00:01:02] Validation | Batch 220/1463 | Loss: 0.8549 | LM_LOSS: 0.3546 | LB_LOSS: 1.0006 [2026-05-24 00:01:06] Validation | Batch 230/1463 | Loss: 0.8562 | LM_LOSS: 0.3560 | LB_LOSS: 1.0005 [2026-05-24 00:01:09] Validation | Batch 240/1463 | Loss: 0.8581 | LM_LOSS: 0.3578 | LB_LOSS: 1.0005 [2026-05-24 00:01:12] Validation | Batch 250/1463 | Loss: 0.8568 | LM_LOSS: 0.3565 | LB_LOSS: 1.0005 [2026-05-24 00:01:15] Validation | Batch 260/1463 | Loss: 0.8589 | LM_LOSS: 0.3586 | LB_LOSS: 1.0006 [2026-05-24 00:01:18] Validation | Batch 270/1463 | Loss: 0.8570 | LM_LOSS: 0.3567 | LB_LOSS: 1.0006 [2026-05-24 00:01:21] Validation | Batch 280/1463 | Loss: 0.8572 | LM_LOSS: 0.3569 | LB_LOSS: 1.0006 [2026-05-24 00:01:24] Validation | Batch 290/1463 | Loss: 0.8558 | LM_LOSS: 0.3555 | LB_LOSS: 1.0006 [2026-05-24 00:01:27] Validation | Batch 300/1463 | Loss: 0.8567 | LM_LOSS: 0.3564 | LB_LOSS: 1.0006 [2026-05-24 00:01:30] Validation | Batch 310/1463 | Loss: 0.8576 | LM_LOSS: 0.3573 | LB_LOSS: 1.0006 [2026-05-24 00:01:33] Validation | Batch 320/1463 | Loss: 0.8564 | LM_LOSS: 0.3561 | LB_LOSS: 1.0005 [2026-05-24 00:01:36] Validation | Batch 330/1463 | Loss: 0.8558 | LM_LOSS: 0.3555 | LB_LOSS: 1.0006 [2026-05-24 00:01:38] Validation | Batch 340/1463 | Loss: 0.8553 | LM_LOSS: 0.3550 | LB_LOSS: 1.0006 [2026-05-24 00:01:42] Validation | Batch 350/1463 | Loss: 0.8564 | LM_LOSS: 0.3561 | LB_LOSS: 1.0006 [2026-05-24 00:01:45] Validation | Batch 360/1463 | Loss: 0.8564 | LM_LOSS: 0.3561 | LB_LOSS: 1.0006 [2026-05-24 00:01:47] Validation | Batch 370/1463 | Loss: 0.8563 | LM_LOSS: 0.3560 | LB_LOSS: 1.0006 [2026-05-24 00:01:50] Validation | Batch 380/1463 | Loss: 0.8557 | LM_LOSS: 0.3554 | LB_LOSS: 1.0006 [2026-05-24 00:01:53] Validation | Batch 390/1463 | Loss: 0.8563 | LM_LOSS: 0.3561 | LB_LOSS: 1.0006 [2026-05-24 00:01:56] Validation | Batch 400/1463 | Loss: 0.8562 | LM_LOSS: 0.3559 | LB_LOSS: 1.0006 [2026-05-24 00:01:59] Validation | Batch 410/1463 | Loss: 0.8558 | LM_LOSS: 0.3555 | LB_LOSS: 1.0006 [2026-05-24 00:02:03] Validation | Batch 420/1463 | Loss: 0.8562 | LM_LOSS: 0.3559 | LB_LOSS: 1.0006 [2026-05-24 00:02:06] Validation | Batch 430/1463 | Loss: 0.8557 | LM_LOSS: 0.3554 | LB_LOSS: 1.0006 [2026-05-24 00:02:08] Validation | Batch 440/1463 | Loss: 0.8560 | LM_LOSS: 0.3557 | LB_LOSS: 1.0006 [2026-05-24 00:02:11] Validation | Batch 450/1463 | Loss: 0.8556 | LM_LOSS: 0.3553 | LB_LOSS: 1.0006 [2026-05-24 00:02:15] Validation | Batch 460/1463 | Loss: 0.8555 | LM_LOSS: 0.3552 | LB_LOSS: 1.0006 [2026-05-24 00:02:18] Validation | Batch 470/1463 | Loss: 0.8565 | LM_LOSS: 0.3562 | LB_LOSS: 1.0006 [2026-05-24 00:02:21] Validation | Batch 480/1463 | Loss: 0.8561 | LM_LOSS: 0.3559 | LB_LOSS: 1.0006 [2026-05-24 00:02:24] Validation | Batch 490/1463 | Loss: 0.8560 | LM_LOSS: 0.3557 | LB_LOSS: 1.0006 [2026-05-24 00:02:28] Validation | Batch 500/1463 | Loss: 0.8560 | LM_LOSS: 0.3557 | LB_LOSS: 1.0006 [2026-05-24 00:02:31] Validation | Batch 510/1463 | Loss: 0.8551 | LM_LOSS: 0.3548 | LB_LOSS: 1.0006 [2026-05-24 00:02:34] Validation | Batch 520/1463 | Loss: 0.8550 | LM_LOSS: 0.3547 | LB_LOSS: 1.0006 [2026-05-24 00:02:36] Validation | Batch 530/1463 | Loss: 0.8557 | LM_LOSS: 0.3554 | LB_LOSS: 1.0006 [2026-05-24 00:02:39] Validation | Batch 540/1463 | Loss: 0.8558 | LM_LOSS: 0.3555 | LB_LOSS: 1.0006 [2026-05-24 00:02:41] Validation | Batch 550/1463 | Loss: 0.8554 | LM_LOSS: 0.3551 | LB_LOSS: 1.0006 [2026-05-24 00:02:44] Validation | Batch 560/1463 | Loss: 0.8553 | LM_LOSS: 0.3551 | LB_LOSS: 1.0006 [2026-05-24 00:02:48] Validation | Batch 570/1463 | Loss: 0.8553 | LM_LOSS: 0.3550 | LB_LOSS: 1.0006 [2026-05-24 00:02:51] Validation | Batch 580/1463 | Loss: 0.8545 | LM_LOSS: 0.3542 | LB_LOSS: 1.0006 [2026-05-24 00:02:55] Validation | Batch 590/1463 | Loss: 0.8543 | LM_LOSS: 0.3540 | LB_LOSS: 1.0006 [2026-05-24 00:02:58] Validation | Batch 600/1463 | Loss: 0.8545 | LM_LOSS: 0.3543 | LB_LOSS: 1.0006 [2026-05-24 00:03:01] Validation | Batch 610/1463 | Loss: 0.8548 | LM_LOSS: 0.3545 | LB_LOSS: 1.0006 [2026-05-24 00:03:04] Validation | Batch 620/1463 | Loss: 0.8542 | LM_LOSS: 0.3540 | LB_LOSS: 1.0006 [2026-05-24 00:03:08] Validation | Batch 630/1463 | Loss: 0.8536 | LM_LOSS: 0.3533 | LB_LOSS: 1.0006 [2026-05-24 00:03:10] Validation | Batch 640/1463 | Loss: 0.8534 | LM_LOSS: 0.3531 | LB_LOSS: 1.0006 [2026-05-24 00:03:13] Validation | Batch 650/1463 | Loss: 0.8529 | LM_LOSS: 0.3527 | LB_LOSS: 1.0006 [2026-05-24 00:03:16] Validation | Batch 660/1463 | Loss: 0.8524 | LM_LOSS: 0.3521 | LB_LOSS: 1.0006 [2026-05-24 00:03:19] Validation | Batch 670/1463 | Loss: 0.8518 | LM_LOSS: 0.3515 | LB_LOSS: 1.0006 [2026-05-24 00:03:21] Validation | Batch 680/1463 | Loss: 0.8519 | LM_LOSS: 0.3516 | LB_LOSS: 1.0006 [2026-05-24 00:03:25] Validation | Batch 690/1463 | Loss: 0.8521 | LM_LOSS: 0.3518 | LB_LOSS: 1.0006 [2026-05-24 00:03:28] Validation | Batch 700/1463 | Loss: 0.8519 | LM_LOSS: 0.3516 | LB_LOSS: 1.0006 [2026-05-24 00:03:30] Validation | Batch 710/1463 | Loss: 0.8522 | LM_LOSS: 0.3519 | LB_LOSS: 1.0006 [2026-05-24 00:03:34] Validation | Batch 720/1463 | Loss: 0.8529 | LM_LOSS: 0.3526 | LB_LOSS: 1.0006 [2026-05-24 00:03:36] Validation | Batch 730/1463 | Loss: 0.8528 | LM_LOSS: 0.3525 | LB_LOSS: 1.0006 [2026-05-24 00:03:38] Validation | Batch 740/1463 | Loss: 0.8525 | LM_LOSS: 0.3523 | LB_LOSS: 1.0006 [2026-05-24 00:03:41] Validation | Batch 750/1463 | Loss: 0.8525 | LM_LOSS: 0.3522 | LB_LOSS: 1.0006 [2026-05-24 00:03:44] Validation | Batch 760/1463 | Loss: 0.8523 | LM_LOSS: 0.3520 | LB_LOSS: 1.0006 [2026-05-24 00:03:46] Validation | Batch 770/1463 | Loss: 0.8523 | LM_LOSS: 0.3520 | LB_LOSS: 1.0006 [2026-05-24 00:03:49] Validation | Batch 780/1463 | Loss: 0.8523 | LM_LOSS: 0.3520 | LB_LOSS: 1.0006 [2026-05-24 00:03:51] Validation | Batch 790/1463 | Loss: 0.8525 | LM_LOSS: 0.3522 | LB_LOSS: 1.0006 [2026-05-24 00:03:54] Validation | Batch 800/1463 | Loss: 0.8525 | LM_LOSS: 0.3522 | LB_LOSS: 1.0006 [2026-05-24 00:03:56] Validation | Batch 810/1463 | Loss: 0.8521 | LM_LOSS: 0.3518 | LB_LOSS: 1.0006 [2026-05-24 00:04:00] Validation | Batch 820/1463 | Loss: 0.8523 | LM_LOSS: 0.3520 | LB_LOSS: 1.0006 [2026-05-24 00:04:03] Validation | Batch 830/1463 | Loss: 0.8524 | LM_LOSS: 0.3521 | LB_LOSS: 1.0006 [2026-05-24 00:04:06] Validation | Batch 840/1463 | Loss: 0.8525 | LM_LOSS: 0.3522 | LB_LOSS: 1.0006 [2026-05-24 00:04:09] Validation | Batch 850/1463 | Loss: 0.8529 | LM_LOSS: 0.3526 | LB_LOSS: 1.0006 [2026-05-24 00:04:13] Validation | Batch 860/1463 | Loss: 0.8533 | LM_LOSS: 0.3530 | LB_LOSS: 1.0006 [2026-05-24 00:04:16] Validation | Batch 870/1463 | Loss: 0.8534 | LM_LOSS: 0.3531 | LB_LOSS: 1.0006 [2026-05-24 00:04:18] Validation | Batch 880/1463 | Loss: 0.8526 | LM_LOSS: 0.3523 | LB_LOSS: 1.0006 [2026-05-24 00:04:21] Validation | Batch 890/1463 | Loss: 0.8525 | LM_LOSS: 0.3522 | LB_LOSS: 1.0006 [2026-05-24 00:04:24] Validation | Batch 900/1463 | Loss: 0.8521 | LM_LOSS: 0.3518 | LB_LOSS: 1.0006 [2026-05-24 00:04:28] Validation | Batch 910/1463 | Loss: 0.8521 | LM_LOSS: 0.3518 | LB_LOSS: 1.0006 [2026-05-24 00:04:31] Validation | Batch 920/1463 | Loss: 0.8524 | LM_LOSS: 0.3522 | LB_LOSS: 1.0006 [2026-05-24 00:04:34] Validation | Batch 930/1463 | Loss: 0.8522 | LM_LOSS: 0.3519 | LB_LOSS: 1.0006 [2026-05-24 00:04:37] Validation | Batch 940/1463 | Loss: 0.8519 | LM_LOSS: 0.3516 | LB_LOSS: 1.0006 [2026-05-24 00:04:41] Validation | Batch 950/1463 | Loss: 0.8521 | LM_LOSS: 0.3518 | LB_LOSS: 1.0006 [2026-05-24 00:04:44] Validation | Batch 960/1463 | Loss: 0.8522 | LM_LOSS: 0.3519 | LB_LOSS: 1.0006 [2026-05-24 00:04:47] Validation | Batch 970/1463 | Loss: 0.8522 | LM_LOSS: 0.3519 | LB_LOSS: 1.0006 [2026-05-24 00:04:51] Validation | Batch 980/1463 | Loss: 0.8525 | LM_LOSS: 0.3522 | LB_LOSS: 1.0006 [2026-05-24 00:04:53] Validation | Batch 990/1463 | Loss: 0.8528 | LM_LOSS: 0.3525 | LB_LOSS: 1.0006 [2026-05-24 00:04:56] Validation | Batch 1000/1463 | Loss: 0.8535 | LM_LOSS: 0.3532 | LB_LOSS: 1.0006 [2026-05-24 00:04:59] Validation | Batch 1010/1463 | Loss: 0.8534 | LM_LOSS: 0.3532 | LB_LOSS: 1.0006 [2026-05-24 00:05:02] Validation | Batch 1020/1463 | Loss: 0.8535 | LM_LOSS: 0.3532 | LB_LOSS: 1.0006 [2026-05-24 00:05:06] Validation | Batch 1030/1463 | Loss: 0.8539 | LM_LOSS: 0.3537 | LB_LOSS: 1.0006 [2026-05-24 00:05:08] Validation | Batch 1040/1463 | Loss: 0.8536 | LM_LOSS: 0.3533 | LB_LOSS: 1.0006 [2026-05-24 00:05:11] Validation | Batch 1050/1463 | Loss: 0.8536 | LM_LOSS: 0.3533 | LB_LOSS: 1.0006 [2026-05-24 00:05:15] Validation | Batch 1060/1463 | Loss: 0.8535 | LM_LOSS: 0.3533 | LB_LOSS: 1.0006 [2026-05-24 00:05:18] Validation | Batch 1070/1463 | Loss: 0.8536 | LM_LOSS: 0.3533 | LB_LOSS: 1.0006 [2026-05-24 00:05:21] Validation | Batch 1080/1463 | Loss: 0.8534 | LM_LOSS: 0.3531 | LB_LOSS: 1.0006 [2026-05-24 00:05:23] Validation | Batch 1090/1463 | Loss: 0.8533 | LM_LOSS: 0.3531 | LB_LOSS: 1.0006 [2026-05-24 00:05:26] Validation | Batch 1100/1463 | Loss: 0.8530 | LM_LOSS: 0.3527 | LB_LOSS: 1.0006 [2026-05-24 00:05:29] Validation | Batch 1110/1463 | Loss: 0.8531 | LM_LOSS: 0.3528 | LB_LOSS: 1.0006 [2026-05-24 00:05:32] Validation | Batch 1120/1463 | Loss: 0.8527 | LM_LOSS: 0.3525 | LB_LOSS: 1.0006 [2026-05-24 00:05:35] Validation | Batch 1130/1463 | Loss: 0.8525 | LM_LOSS: 0.3523 | LB_LOSS: 1.0006 [2026-05-24 00:05:38] Validation | Batch 1140/1463 | Loss: 0.8528 | LM_LOSS: 0.3526 | LB_LOSS: 1.0006 [2026-05-24 00:05:42] Validation | Batch 1150/1463 | Loss: 0.8526 | LM_LOSS: 0.3523 | LB_LOSS: 1.0006 [2026-05-24 00:05:45] Validation | Batch 1160/1463 | Loss: 0.8525 | LM_LOSS: 0.3522 | LB_LOSS: 1.0006 [2026-05-24 00:05:48] Validation | Batch 1170/1463 | Loss: 0.8526 | LM_LOSS: 0.3523 | LB_LOSS: 1.0006 [2026-05-24 00:05:51] Validation | Batch 1180/1463 | Loss: 0.8527 | LM_LOSS: 0.3524 | LB_LOSS: 1.0006 [2026-05-24 00:05:54] Validation | Batch 1190/1463 | Loss: 0.8529 | LM_LOSS: 0.3526 | LB_LOSS: 1.0006 [2026-05-24 00:05:57] Validation | Batch 1200/1463 | Loss: 0.8529 | LM_LOSS: 0.3526 | LB_LOSS: 1.0006 [2026-05-24 00:06:00] Validation | Batch 1210/1463 | Loss: 0.8532 | LM_LOSS: 0.3530 | LB_LOSS: 1.0006 [2026-05-24 00:06:03] Validation | Batch 1220/1463 | Loss: 0.8536 | LM_LOSS: 0.3533 | LB_LOSS: 1.0006 [2026-05-24 00:06:06] Validation | Batch 1230/1463 | Loss: 0.8536 | LM_LOSS: 0.3534 | LB_LOSS: 1.0006 [2026-05-24 00:06:09] Validation | Batch 1240/1463 | Loss: 0.8538 | LM_LOSS: 0.3535 | LB_LOSS: 1.0006 [2026-05-24 00:06:12] Validation | Batch 1250/1463 | Loss: 0.8539 | LM_LOSS: 0.3536 | LB_LOSS: 1.0006 [2026-05-24 00:06:15] Validation | Batch 1260/1463 | Loss: 0.8538 | LM_LOSS: 0.3535 | LB_LOSS: 1.0006 [2026-05-24 00:06:18] Validation | Batch 1270/1463 | Loss: 0.8536 | LM_LOSS: 0.3533 | LB_LOSS: 1.0006 [2026-05-24 00:06:22] Validation | Batch 1280/1463 | Loss: 0.8534 | LM_LOSS: 0.3531 | LB_LOSS: 1.0006 [2026-05-24 00:06:25] Validation | Batch 1290/1463 | Loss: 0.8534 | LM_LOSS: 0.3531 | LB_LOSS: 1.0006 [2026-05-24 00:06:28] Validation | Batch 1300/1463 | Loss: 0.8531 | LM_LOSS: 0.3528 | LB_LOSS: 1.0006 [2026-05-24 00:06:31] Validation | Batch 1310/1463 | Loss: 0.8528 | LM_LOSS: 0.3526 | LB_LOSS: 1.0006 [2026-05-24 00:06:33] Validation | Batch 1320/1463 | Loss: 0.8529 | LM_LOSS: 0.3527 | LB_LOSS: 1.0006 [2026-05-24 00:06:35] Validation | Batch 1330/1463 | Loss: 0.8529 | LM_LOSS: 0.3527 | LB_LOSS: 1.0006 [2026-05-24 00:06:38] Validation | Batch 1340/1463 | Loss: 0.8533 | LM_LOSS: 0.3531 | LB_LOSS: 1.0006 [2026-05-24 00:06:41] Validation | Batch 1350/1463 | Loss: 0.8533 | LM_LOSS: 0.3530 | LB_LOSS: 1.0006 [2026-05-24 00:06:45] Validation | Batch 1360/1463 | Loss: 0.8536 | LM_LOSS: 0.3533 | LB_LOSS: 1.0006 [2026-05-24 00:06:48] Validation | Batch 1370/1463 | Loss: 0.8539 | LM_LOSS: 0.3536 | LB_LOSS: 1.0006 [2026-05-24 00:06:51] Validation | Batch 1380/1463 | Loss: 0.8536 | LM_LOSS: 0.3533 | LB_LOSS: 1.0005 [2026-05-24 00:06:54] Validation | Batch 1390/1463 | Loss: 0.8536 | LM_LOSS: 0.3533 | LB_LOSS: 1.0005 [2026-05-24 00:06:57] Validation | Batch 1400/1463 | Loss: 0.8539 | LM_LOSS: 0.3536 | LB_LOSS: 1.0005 [2026-05-24 00:06:59] Validation | Batch 1410/1463 | Loss: 0.8543 | LM_LOSS: 0.3540 | LB_LOSS: 1.0005 [2026-05-24 00:07:03] Validation | Batch 1420/1463 | Loss: 0.8542 | LM_LOSS: 0.3539 | LB_LOSS: 1.0005 [2026-05-24 00:07:06] Validation | Batch 1430/1463 | Loss: 0.8542 | LM_LOSS: 0.3540 | LB_LOSS: 1.0005 [2026-05-24 00:07:09] Validation | Batch 1440/1463 | Loss: 0.8547 | LM_LOSS: 0.3544 | LB_LOSS: 1.0005 [2026-05-24 00:07:12] Validation | Batch 1450/1463 | Loss: 0.8544 | LM_LOSS: 0.3542 | LB_LOSS: 1.0005 [2026-05-24 00:07:15] Validation | Batch 1460/1463 | Loss: 0.8548 | LM_LOSS: 0.3545 | LB_LOSS: 1.0005 [2026-05-24 00:07:17] Validation | Batch 1463/1463 | Loss: 0.8547 | LM_LOSS: 0.3544 | LB_LOSS: 1.0005 [2026-05-24 00:07:17] Validation | Loss: 0.8547 | LM_LOSS: 0.3544 | LB_LOSS: 1.0005 | PPL: 1.43 | Time: 442.93s [2026-05-24 00:07:29] Epoch 3 | Step 14010 | Loss: 0.6256 | LM: 0.1255 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.440 | HR1: 0.402/SR1: 0.395 | LR: 1.01e-05 [2026-05-24 00:07:42] Epoch 3 | Step 14020 | Loss: 0.6256 | LM: 0.1254 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.440 | HR1: 0.401/SR1: 0.395 | LR: 1.00e-05 [2026-05-24 00:07:56] Epoch 3 | Step 14030 | Loss: 0.6257 | LM: 0.1255 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.440 | HR1: 0.401/SR1: 0.395 | LR: 1.00e-05 [2026-05-24 00:08:08] Epoch 3 | Step 14040 | Loss: 0.6257 | LM: 0.1254 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.440 | HR1: 0.401/SR1: 0.395 | LR: 1.00e-05 [2026-05-24 00:08:22] Epoch 3 | Step 14050 | Loss: 0.6257 | LM: 0.1255 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.440 | HR1: 0.401/SR1: 0.395 | LR: 1.00e-05 [2026-05-24 00:08:35] Epoch 3 | Step 14060 | Loss: 0.6257 | LM: 0.1255 | LB: 1.0012 | CL0: 2.5 | CL1: 2.5 | HR0: 0.401/SR0: 0.440 | HR1: 0.402/SR1: 0.395 | LR: 1.00e-05 [2026-05-24 00:08:37] Epoch 3 completed in 8044.96s | Loss: 0.6257 | CL0: 2.5 | CL1: 2.5 [2026-05-24 00:08:37] Training completed! [2026-05-24 00:08:42] Final model: outputs/2026-05-23/N_2.5/model_final.pt