Update training logs at step 4000

training_logs.jsonl

@@ -63,3 +63,24 @@
 {"step": 3475, "epoch": 2.192429022082019, "time": "2026-07-09 00:11:47", "logs": {"loss": 0.019525556564331054, "grad_norm": 0.02751922979950905, "learning_rate": 1.0071574405515545e-05, "epoch": 2.192429022082019}}
 {"step": 3500, "epoch": 2.2082018927444795, "time": "2026-07-09 00:12:11", "logs": {"loss": 0.07425961017608643, "grad_norm": 0.06599944829940796, "learning_rate": 9.93393123453644e-06, "epoch": 2.2082018927444795}}
 {"step": 3500, "epoch": 2.2082018927444795, "time": "2026-07-09 00:13:11", "logs": {"eval_loss": 0.0798133835196495, "eval_accuracy": 0.9820189274447949, "eval_precision": 0.9769038701622972, "eval_recall": 0.9873817034700315, "eval_f1": 0.9821148415437716, "eval_runtime": 59.7613, "eval_samples_per_second": 53.044, "eval_steps_per_second": 3.33, "epoch": 2.2082018927444795}}
+{"step": 3525, "epoch": 2.22397476340694, "time": "2026-07-09 00:13:55", "logs": {"loss": 0.06443671703338623, "grad_norm": 0.22195865213871002, "learning_rate": 9.796300580710401e-06, "epoch": 2.22397476340694}}
+{"step": 3550, "epoch": 2.2397476340694005, "time": "2026-07-09 00:14:20", "logs": {"loss": 0.011473581790924073, "grad_norm": 0.19403228163719177, "learning_rate": 9.658708519051078e-06, "epoch": 2.2397476340694005}}
+{"step": 3575, "epoch": 2.2555205047318614, "time": "2026-07-09 00:14:43", "logs": {"loss": 0.03980313539505005, "grad_norm": 0.3892849087715149, "learning_rate": 9.521181117260576e-06, "epoch": 2.2555205047318614}}
+{"step": 3600, "epoch": 2.271293375394322, "time": "2026-07-09 00:15:07", "logs": {"loss": 0.05841365337371826, "grad_norm": 0.09898291528224945, "learning_rate": 9.38374443079078e-06, "epoch": 2.271293375394322}}
+{"step": 3625, "epoch": 2.2870662460567823, "time": "2026-07-09 00:15:31", "logs": {"loss": 0.019511160850524904, "grad_norm": 0.059039290994405746, "learning_rate": 9.246424497906963e-06, "epoch": 2.2870662460567823}}
+{"step": 3650, "epoch": 2.302839116719243, "time": "2026-07-09 00:15:55", "logs": {"loss": 0.08494296073913574, "grad_norm": 0.653461217880249, "learning_rate": 9.10924733475469e-06, "epoch": 2.302839116719243}}
+{"step": 3675, "epoch": 2.3186119873817033, "time": "2026-07-09 00:16:19", "logs": {"loss": 0.043811373710632324, "grad_norm": 3.5332391262054443, "learning_rate": 8.972238930430877e-06, "epoch": 2.3186119873817033}}
+{"step": 3700, "epoch": 2.334384858044164, "time": "2026-07-09 00:16:43", "logs": {"loss": 0.03998267650604248, "grad_norm": 0.237837016582489, "learning_rate": 8.83542524205999e-06, "epoch": 2.334384858044164}}
+{"step": 3725, "epoch": 2.3501577287066246, "time": "2026-07-09 00:17:07", "logs": {"loss": 0.03948731899261475, "grad_norm": 0.04083343222737312, "learning_rate": 8.698832189876313e-06, "epoch": 2.3501577287066246}}
+{"step": 3750, "epoch": 2.365930599369085, "time": "2026-07-09 00:17:31", "logs": {"loss": 0.017638446092605592, "grad_norm": 0.03361862152814865, "learning_rate": 8.562485652313168e-06, "epoch": 2.365930599369085}}
+{"step": 3775, "epoch": 2.3817034700315456, "time": "2026-07-09 00:17:55", "logs": {"loss": 0.04891120433807373, "grad_norm": 4.868694305419922, "learning_rate": 8.42641146110011e-06, "epoch": 2.3817034700315456}}
+{"step": 3800, "epoch": 2.3974763406940065, "time": "2026-07-09 00:18:19", "logs": {"loss": 0.019228975772857666, "grad_norm": 0.060350608080625534, "learning_rate": 8.290635396368916e-06, "epoch": 2.3974763406940065}}
+{"step": 3825, "epoch": 2.413249211356467, "time": "2026-07-09 00:18:43", "logs": {"loss": 0.04954701423645019, "grad_norm": 2.000920295715332, "learning_rate": 8.15518318176938e-06, "epoch": 2.413249211356467}}
+{"step": 3850, "epoch": 2.4290220820189274, "time": "2026-07-09 00:19:06", "logs": {"loss": 0.048543753623962405, "grad_norm": 0.03264353796839714, "learning_rate": 8.020080479595818e-06, "epoch": 2.4290220820189274}}
+{"step": 3875, "epoch": 2.444794952681388, "time": "2026-07-09 00:19:30", "logs": {"loss": 0.03049187421798706, "grad_norm": 0.16232901811599731, "learning_rate": 7.885352885925139e-06, "epoch": 2.444794952681388}}
+{"step": 3900, "epoch": 2.4605678233438484, "time": "2026-07-09 00:19:54", "logs": {"loss": 0.03745495319366455, "grad_norm": 0.2823891341686249, "learning_rate": 7.75102592576756e-06, "epoch": 2.4605678233438484}}
+{"step": 3925, "epoch": 2.4763406940063093, "time": "2026-07-09 00:20:18", "logs": {"loss": 0.027678728103637695, "grad_norm": 0.06313452869653702, "learning_rate": 7.617125048230678e-06, "epoch": 2.4763406940063093}}
+{"step": 3950, "epoch": 2.4921135646687698, "time": "2026-07-09 00:20:42", "logs": {"loss": 0.036765682697296145, "grad_norm": 0.03460308164358139, "learning_rate": 7.483675621697999e-06, "epoch": 2.4921135646687698}}
+{"step": 3975, "epoch": 2.5078864353312302, "time": "2026-07-09 00:21:05", "logs": {"loss": 0.048152894973754884, "grad_norm": 0.04799109697341919, "learning_rate": 7.350702929022739e-06, "epoch": 2.5078864353312302}}
+{"step": 4000, "epoch": 2.5236593059936907, "time": "2026-07-09 00:21:29", "logs": {"loss": 0.03499542236328125, "grad_norm": 0.6452091336250305, "learning_rate": 7.2182321627378e-06, "epoch": 2.5236593059936907}}
+{"step": 4000, "epoch": 2.5236593059936907, "time": "2026-07-09 00:22:28", "logs": {"eval_loss": 0.09398876130580902, "eval_accuracy": 0.9817034700315458, "eval_precision": 0.9698461538461538, "eval_recall": 0.9943217665615142, "eval_f1": 0.9819314641744549, "eval_runtime": 59.2692, "eval_samples_per_second": 53.485, "eval_steps_per_second": 3.358, "epoch": 2.5236593059936907}}