Update training logs at step 4000

training_logs.jsonl

@@ -19,3 +19,24 @@
 {"step": 3475, "epoch": 2.192429022082019, "time": "2026-07-08 23:50:51", "logs": {"loss": 0.019525480270385743, "grad_norm": 0.027517961338162422, "learning_rate": 1.0071574405515545e-05, "epoch": 2.192429022082019}}
 {"step": 3500, "epoch": 2.2082018927444795, "time": "2026-07-08 23:51:15", "logs": {"loss": 0.07426059722900391, "grad_norm": 0.06598785519599915, "learning_rate": 9.93393123453644e-06, "epoch": 2.2082018927444795}}
 {"step": 3500, "epoch": 2.2082018927444795, "time": "2026-07-08 23:52:15", "logs": {"eval_loss": 0.07981470227241516, "eval_accuracy": 0.9820189274447949, "eval_precision": 0.9769038701622972, "eval_recall": 0.9873817034700315, "eval_f1": 0.9821148415437716, "eval_runtime": 59.5362, "eval_samples_per_second": 53.245, "eval_steps_per_second": 3.343, "epoch": 2.2082018927444795}}
+{"step": 3525, "epoch": 2.22397476340694, "time": "2026-07-08 23:52:56", "logs": {"loss": 0.06443918228149415, "grad_norm": 0.22194409370422363, "learning_rate": 9.796300580710401e-06, "epoch": 2.22397476340694}}
+{"step": 3550, "epoch": 2.2397476340694005, "time": "2026-07-08 23:53:21", "logs": {"loss": 0.01147290587425232, "grad_norm": 0.19404132664203644, "learning_rate": 9.658708519051078e-06, "epoch": 2.2397476340694005}}
+{"step": 3575, "epoch": 2.2555205047318614, "time": "2026-07-08 23:53:45", "logs": {"loss": 0.03980299949645996, "grad_norm": 0.3892022967338562, "learning_rate": 9.521181117260576e-06, "epoch": 2.2555205047318614}}
+{"step": 3600, "epoch": 2.271293375394322, "time": "2026-07-08 23:54:08", "logs": {"loss": 0.05841335296630859, "grad_norm": 0.09898478537797928, "learning_rate": 9.38374443079078e-06, "epoch": 2.271293375394322}}
+{"step": 3625, "epoch": 2.2870662460567823, "time": "2026-07-08 23:54:32", "logs": {"loss": 0.019510600566864014, "grad_norm": 0.0590379536151886, "learning_rate": 9.246424497906963e-06, "epoch": 2.2870662460567823}}
+{"step": 3650, "epoch": 2.302839116719243, "time": "2026-07-08 23:54:56", "logs": {"loss": 0.08494356155395508, "grad_norm": 0.6535356640815735, "learning_rate": 9.10924733475469e-06, "epoch": 2.302839116719243}}
+{"step": 3675, "epoch": 2.3186119873817033, "time": "2026-07-08 23:55:20", "logs": {"loss": 0.04381181716918945, "grad_norm": 3.5332858562469482, "learning_rate": 8.972238930430877e-06, "epoch": 2.3186119873817033}}
+{"step": 3700, "epoch": 2.334384858044164, "time": "2026-07-08 23:55:44", "logs": {"loss": 0.03998601913452148, "grad_norm": 0.237799733877182, "learning_rate": 8.83542524205999e-06, "epoch": 2.334384858044164}}
+{"step": 3725, "epoch": 2.3501577287066246, "time": "2026-07-08 23:56:08", "logs": {"loss": 0.039485976696014405, "grad_norm": 0.04083641991019249, "learning_rate": 8.698832189876313e-06, "epoch": 2.3501577287066246}}
+{"step": 3750, "epoch": 2.365930599369085, "time": "2026-07-08 23:56:32", "logs": {"loss": 0.017637826204299926, "grad_norm": 0.033618684858083725, "learning_rate": 8.562485652313168e-06, "epoch": 2.365930599369085}}
+{"step": 3775, "epoch": 2.3817034700315456, "time": "2026-07-08 23:56:56", "logs": {"loss": 0.048912029266357425, "grad_norm": 4.868691444396973, "learning_rate": 8.42641146110011e-06, "epoch": 2.3817034700315456}}
+{"step": 3800, "epoch": 2.3974763406940065, "time": "2026-07-08 23:57:20", "logs": {"loss": 0.019228267669677734, "grad_norm": 0.06035134196281433, "learning_rate": 8.290635396368916e-06, "epoch": 2.3974763406940065}}
+{"step": 3825, "epoch": 2.413249211356467, "time": "2026-07-08 23:57:44", "logs": {"loss": 0.04954765319824219, "grad_norm": 2.0009303092956543, "learning_rate": 8.15518318176938e-06, "epoch": 2.413249211356467}}
+{"step": 3850, "epoch": 2.4290220820189274, "time": "2026-07-08 23:58:08", "logs": {"loss": 0.048542871475219726, "grad_norm": 0.032643403857946396, "learning_rate": 8.020080479595818e-06, "epoch": 2.4290220820189274}}
+{"step": 3875, "epoch": 2.444794952681388, "time": "2026-07-08 23:58:32", "logs": {"loss": 0.030491657257080078, "grad_norm": 0.16234640777111053, "learning_rate": 7.885352885925139e-06, "epoch": 2.444794952681388}}
+{"step": 3900, "epoch": 2.4605678233438484, "time": "2026-07-08 23:58:56", "logs": {"loss": 0.03745485067367554, "grad_norm": 0.28245651721954346, "learning_rate": 7.75102592576756e-06, "epoch": 2.4605678233438484}}
+{"step": 3925, "epoch": 2.4763406940063093, "time": "2026-07-08 23:59:19", "logs": {"loss": 0.027678358554840087, "grad_norm": 0.06313358247280121, "learning_rate": 7.617125048230678e-06, "epoch": 2.4763406940063093}}
+{"step": 3950, "epoch": 2.4921135646687698, "time": "2026-07-08 23:59:43", "logs": {"loss": 0.03676470041275024, "grad_norm": 0.034601207822561264, "learning_rate": 7.483675621697999e-06, "epoch": 2.4921135646687698}}
+{"step": 3975, "epoch": 2.5078864353312302, "time": "2026-07-09 00:00:07", "logs": {"loss": 0.04815303325653076, "grad_norm": 0.04799091815948486, "learning_rate": 7.350702929022739e-06, "epoch": 2.5078864353312302}}
+{"step": 4000, "epoch": 2.5236593059936907, "time": "2026-07-09 00:00:31", "logs": {"loss": 0.03499504089355469, "grad_norm": 0.6451889276504517, "learning_rate": 7.2182321627378e-06, "epoch": 2.5236593059936907}}
+{"step": 4000, "epoch": 2.5236593059936907, "time": "2026-07-09 00:01:30", "logs": {"eval_loss": 0.09398859739303589, "eval_accuracy": 0.9817034700315458, "eval_precision": 0.9698461538461538, "eval_recall": 0.9943217665615142, "eval_f1": 0.9819314641744549, "eval_runtime": 59.5237, "eval_samples_per_second": 53.256, "eval_steps_per_second": 3.343, "epoch": 2.5236593059936907}}