Update training logs at step 4500

training_logs.jsonl

@@ -167,3 +167,24 @@
 {"step": 3975, "epoch": 2.5078864353312302, "time": "2026-07-08 14:39:01", "logs": {"loss": 0.026499390602111816, "grad_norm": 0.36902672052383423, "learning_rate": 7.350702929022739e-06, "epoch": 2.5078864353312302}}
 {"step": 4000, "epoch": 2.5236593059936907, "time": "2026-07-08 14:39:24", "logs": {"loss": 0.025588753223419188, "grad_norm": 0.01817493699491024, "learning_rate": 7.2182321627378e-06, "epoch": 2.5236593059936907}}
 {"step": 4000, "epoch": 2.5236593059936907, "time": "2026-07-08 14:40:26", "logs": {"eval_loss": 0.06328697502613068, "eval_accuracy": 0.9851735015772871, "eval_precision": 0.977639751552795, "eval_recall": 0.9930599369085173, "eval_f1": 0.9852895148669797, "eval_runtime": 61.6662, "eval_samples_per_second": 51.406, "eval_steps_per_second": 3.227, "epoch": 2.5236593059936907}}
+{"step": 4025, "epoch": 2.5394321766561516, "time": "2026-07-08 14:41:13", "logs": {"loss": 0.041892838478088376, "grad_norm": 13.168465614318848, "learning_rate": 7.0862884202829075e-06, "epoch": 2.5394321766561516}}
+{"step": 4050, "epoch": 2.555205047318612, "time": "2026-07-08 14:41:38", "logs": {"loss": 0.07447029113769531, "grad_norm": 5.61896276473999, "learning_rate": 6.954896699249725e-06, "epoch": 2.555205047318612}}
+{"step": 4075, "epoch": 2.5709779179810726, "time": "2026-07-08 14:42:01", "logs": {"loss": 0.046046819686889645, "grad_norm": 0.027618594467639923, "learning_rate": 6.824081892645902e-06, "epoch": 2.5709779179810726}}
+{"step": 4100, "epoch": 2.586750788643533, "time": "2026-07-08 14:42:24", "logs": {"loss": 0.045093264579772946, "grad_norm": 0.049199484288692474, "learning_rate": 6.693868784178934e-06, "epoch": 2.586750788643533}}
+{"step": 4125, "epoch": 2.6025236593059935, "time": "2026-07-08 14:42:48", "logs": {"loss": 0.07267766952514648, "grad_norm": 0.1187095046043396, "learning_rate": 6.564282043560731e-06, "epoch": 2.6025236593059935}}
+{"step": 4150, "epoch": 2.6182965299684544, "time": "2026-07-08 14:43:12", "logs": {"loss": 0.028906166553497314, "grad_norm": 0.08217153698205948, "learning_rate": 6.435346221833788e-06, "epoch": 2.6182965299684544}}
+{"step": 4175, "epoch": 2.634069400630915, "time": "2026-07-08 14:43:35", "logs": {"loss": 0.0493446683883667, "grad_norm": 0.07539000362157822, "learning_rate": 6.307085746719808e-06, "epoch": 2.634069400630915}}
+{"step": 4200, "epoch": 2.6498422712933754, "time": "2026-07-08 14:43:59", "logs": {"loss": 0.07266830444335938, "grad_norm": 0.10210157185792923, "learning_rate": 6.179524917991746e-06, "epoch": 2.6498422712933754}}
+{"step": 4225, "epoch": 2.665615141955836, "time": "2026-07-08 14:44:23", "logs": {"loss": 0.048619318008422854, "grad_norm": 0.36085325479507446, "learning_rate": 6.05268790287003e-06, "epoch": 2.665615141955836}}
+{"step": 4250, "epoch": 2.6813880126182967, "time": "2026-07-08 14:44:46", "logs": {"loss": 0.014012279510498047, "grad_norm": 0.10123010724782944, "learning_rate": 5.926598731443951e-06, "epoch": 2.6813880126182967}}
+{"step": 4275, "epoch": 2.697160883280757, "time": "2026-07-08 14:45:10", "logs": {"loss": 0.036032071113586424, "grad_norm": 0.03025141730904579, "learning_rate": 5.801281292118985e-06, "epoch": 2.697160883280757}}
+{"step": 4300, "epoch": 2.7129337539432177, "time": "2026-07-08 14:45:33", "logs": {"loss": 0.026038434505462647, "grad_norm": 0.07237054407596588, "learning_rate": 5.676759327090997e-06, "epoch": 2.7129337539432177}}
+{"step": 4325, "epoch": 2.728706624605678, "time": "2026-07-08 14:45:57", "logs": {"loss": 0.023030064105987548, "grad_norm": 0.028059201315045357, "learning_rate": 5.553056427848136e-06, "epoch": 2.728706624605678}}
+{"step": 4350, "epoch": 2.7444794952681386, "time": "2026-07-08 14:46:20", "logs": {"loss": 0.046771907806396486, "grad_norm": 8.2833890914917, "learning_rate": 5.430196030701236e-06, "epoch": 2.7444794952681386}}
+{"step": 4375, "epoch": 2.7602523659305995, "time": "2026-07-08 14:46:44", "logs": {"loss": 0.06948490142822265, "grad_norm": 0.07120615243911743, "learning_rate": 5.308201412343714e-06, "epoch": 2.7602523659305995}}
+{"step": 4400, "epoch": 2.77602523659306, "time": "2026-07-08 14:47:08", "logs": {"loss": 0.06183769702911377, "grad_norm": 0.032343458384275436, "learning_rate": 5.187095685441592e-06, "epoch": 2.77602523659306}}
+{"step": 4425, "epoch": 2.7917981072555205, "time": "2026-07-08 14:47:31", "logs": {"loss": 0.06815167903900146, "grad_norm": 2.7497501373291016, "learning_rate": 5.066901794254707e-06, "epoch": 2.7917981072555205}}
+{"step": 4450, "epoch": 2.807570977917981, "time": "2026-07-08 14:47:55", "logs": {"loss": 0.04611932754516602, "grad_norm": 6.535846710205078, "learning_rate": 4.947642510289735e-06, "epoch": 2.807570977917981}}
+{"step": 4475, "epoch": 2.823343848580442, "time": "2026-07-08 14:48:18", "logs": {"loss": 0.017079710960388184, "grad_norm": 19.74146842956543, "learning_rate": 4.829340427985988e-06, "epoch": 2.823343848580442}}
+{"step": 4500, "epoch": 2.8391167192429023, "time": "2026-07-08 14:48:42", "logs": {"loss": 0.04784326076507568, "grad_norm": 17.29274559020996, "learning_rate": 4.71201796043478e-06, "epoch": 2.8391167192429023}}
+{"step": 4500, "epoch": 2.8391167192429023, "time": "2026-07-08 14:49:44", "logs": {"eval_loss": 0.07868093252182007, "eval_accuracy": 0.9823343848580441, "eval_precision": 0.968156766687079, "eval_recall": 0.9974763406940063, "eval_f1": 0.9825978868862648, "eval_runtime": 61.7337, "eval_samples_per_second": 51.35, "eval_steps_per_second": 3.224, "epoch": 2.8391167192429023}}