Update training logs at step 3500

training_logs.jsonl

@@ -42,3 +42,24 @@
 {"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}}
 {"step": 4000, "epoch": 2.5236593059936907, "time": "2026-07-09 00:01:47", "logs": {"train_runtime": 1101.9647, "train_samples_per_second": 92.032, "train_steps_per_second": 5.753, "total_flos": 1.67654364475392e+16, "train_loss": 0.012189213857054711, "epoch": 2.5236593059936907}}
 {"step": 4000, "epoch": 2.5236593059936907, "time": "2026-07-09 00:02:48", "logs": {"eval_loss": 0.09398859739303589, "eval_accuracy": 0.9817034700315458, "eval_precision": 0.9698461538461538, "eval_recall": 0.9943217665615142, "eval_f1": 0.9819314641744549, "eval_runtime": 61.337, "eval_samples_per_second": 51.682, "eval_steps_per_second": 3.244, "epoch": 2.5236593059936907}}
+{"step": 3025, "epoch": 1.9085173501577288, "time": "2026-07-09 00:04:37", "logs": {"loss": 0.0519749116897583, "grad_norm": 32.33957290649414, "learning_rate": 1.2521652607281511e-05, "epoch": 1.9085173501577288}}
+{"step": 3050, "epoch": 1.9242902208201893, "time": "2026-07-09 00:05:02", "logs": {"loss": 0.08039166450500489, "grad_norm": 5.458426475524902, "learning_rate": 1.2388221765257922e-05, "epoch": 1.9242902208201893}}
+{"step": 3075, "epoch": 1.9400630914826498, "time": "2026-07-09 00:05:25", "logs": {"loss": 0.03573695421218872, "grad_norm": 0.032594215124845505, "learning_rate": 1.2254338459241878e-05, "epoch": 1.9400630914826498}}
+{"step": 3100, "epoch": 1.9558359621451105, "time": "2026-07-09 00:05:48", "logs": {"loss": 0.07872512817382812, "grad_norm": 13.032034873962402, "learning_rate": 1.2120028054287853e-05, "epoch": 1.9558359621451105}}
+{"step": 3125, "epoch": 1.971608832807571, "time": "2026-07-09 00:06:12", "logs": {"loss": 0.07527520179748536, "grad_norm": 3.8453073501586914, "learning_rate": 1.198531599636696e-05, "epoch": 1.971608832807571}}
+{"step": 3150, "epoch": 1.9873817034700316, "time": "2026-07-09 00:06:36", "logs": {"loss": 0.08369292259216309, "grad_norm": 0.0618828721344471, "learning_rate": 1.1850227807546056e-05, "epoch": 1.9873817034700316}}
+{"step": 3175, "epoch": 2.003154574132492, "time": "2026-07-09 00:07:00", "logs": {"loss": 0.06914762020111084, "grad_norm": 0.2873757481575012, "learning_rate": 1.1714789081152414e-05, "epoch": 2.003154574132492}}
+{"step": 3200, "epoch": 2.0189274447949526, "time": "2026-07-09 00:07:24", "logs": {"loss": 0.04823594570159912, "grad_norm": 0.24084864556789398, "learning_rate": 1.1579025476924912e-05, "epoch": 2.0189274447949526}}
+{"step": 3225, "epoch": 2.034700315457413, "time": "2026-07-09 00:07:47", "logs": {"loss": 0.05293301582336426, "grad_norm": 0.06181693077087402, "learning_rate": 1.1442962716152626e-05, "epoch": 2.034700315457413}}
+{"step": 3250, "epoch": 2.050473186119874, "time": "2026-07-09 00:08:12", "logs": {"loss": 0.042997422218322756, "grad_norm": 0.23160123825073242, "learning_rate": 1.1306626576801762e-05, "epoch": 2.050473186119874}}
+{"step": 3275, "epoch": 2.0662460567823344, "time": "2026-07-09 00:08:36", "logs": {"loss": 0.03015615463256836, "grad_norm": 0.11183146387338638, "learning_rate": 1.1170042888631865e-05, "epoch": 2.0662460567823344}}
+{"step": 3300, "epoch": 2.082018927444795, "time": "2026-07-09 00:09:00", "logs": {"loss": 0.06211026668548584, "grad_norm": 0.020436899736523628, "learning_rate": 1.103323752830219e-05, "epoch": 2.082018927444795}}
+{"step": 3325, "epoch": 2.0977917981072554, "time": "2026-07-09 00:09:24", "logs": {"loss": 0.06024351596832275, "grad_norm": 0.09942758828401566, "learning_rate": 1.0896236414469217e-05, "epoch": 2.0977917981072554}}
+{"step": 3350, "epoch": 2.1135646687697163, "time": "2026-07-09 00:09:47", "logs": {"loss": 0.09793644905090332, "grad_norm": 0.06320168823003769, "learning_rate": 1.075906550287618e-05, "epoch": 2.1135646687697163}}
+{"step": 3375, "epoch": 2.1293375394321767, "time": "2026-07-09 00:10:11", "logs": {"loss": 0.03903674602508545, "grad_norm": 0.14571192860603333, "learning_rate": 1.0621750781435597e-05, "epoch": 2.1293375394321767}}
+{"step": 3400, "epoch": 2.145110410094637, "time": "2026-07-09 00:10:35", "logs": {"loss": 0.024111902713775633, "grad_norm": 30.196697235107422, "learning_rate": 1.048431826530569e-05, "epoch": 2.145110410094637}}
+{"step": 3425, "epoch": 2.1608832807570977, "time": "2026-07-09 00:10:59", "logs": {"loss": 0.09379005432128906, "grad_norm": 0.08068069070577621, "learning_rate": 1.0346793991961636e-05, "epoch": 2.1608832807570977}}
+{"step": 3450, "epoch": 2.176656151419558, "time": "2026-07-09 00:11:23", "logs": {"loss": 0.028721418380737305, "grad_norm": 0.041268039494752884, "learning_rate": 1.0209204016262617e-05, "epoch": 2.176656151419558}}
+{"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}}