Update training logs at step 3500

training_logs.jsonl

@@ -125,3 +125,24 @@
 {"step": 2975, "epoch": 2.406957928802589, "time": "2026-06-26 22:38:53", "logs": {"loss": 0.08545310974121094, "grad_norm": 7.103508472442627, "learning_rate": 8.209469632429811e-06, "epoch": 2.406957928802589}}
 {"step": 3000, "epoch": 2.4271844660194173, "time": "2026-06-26 22:39:15", "logs": {"loss": 0.2049958038330078, "grad_norm": 6.692376136779785, "learning_rate": 8.036115791272697e-06, "epoch": 2.4271844660194173}}
 {"step": 3000, "epoch": 2.4271844660194173, "time": "2026-06-26 22:39:50", "logs": {"eval_loss": 0.1759410947561264, "eval_accuracy": 0.9631126831733199, "eval_precision": 0.9638009049773756, "eval_recall": 0.9887267904509284, "eval_f1": 0.9761047463175123, "eval_runtime": 34.2085, "eval_samples_per_second": 57.851, "eval_steps_per_second": 3.625, "epoch": 2.4271844660194173}}
+{"step": 3025, "epoch": 2.4474110032362457, "time": "2026-06-26 22:40:25", "logs": {"loss": 0.21097312927246092, "grad_norm": 11.302281379699707, "learning_rate": 7.863373686546868e-06, "epoch": 2.4474110032362457}}
+{"step": 3050, "epoch": 2.4676375404530746, "time": "2026-06-26 22:40:48", "logs": {"loss": 0.12400970458984376, "grad_norm": 5.9209113121032715, "learning_rate": 7.691297126230515e-06, "epoch": 2.4676375404530746}}
+{"step": 3075, "epoch": 2.487864077669903, "time": "2026-06-26 22:41:10", "logs": {"loss": 0.20074039459228515, "grad_norm": 15.600579261779785, "learning_rate": 7.519939710989326e-06, "epoch": 2.487864077669903}}
+{"step": 3100, "epoch": 2.5080906148867315, "time": "2026-06-26 22:41:32", "logs": {"loss": 0.0825264072418213, "grad_norm": 14.571202278137207, "learning_rate": 7.349354817480234e-06, "epoch": 2.5080906148867315}}
+{"step": 3125, "epoch": 2.52831715210356, "time": "2026-06-26 22:41:54", "logs": {"loss": 0.13173922538757324, "grad_norm": 0.2681007981300354, "learning_rate": 7.179595581724971e-06, "epoch": 2.52831715210356}}
+{"step": 3150, "epoch": 2.5485436893203883, "time": "2026-06-26 22:42:16", "logs": {"loss": 0.09000961303710937, "grad_norm": 0.1719094216823578, "learning_rate": 7.010714882558551e-06, "epoch": 2.5485436893203883}}
+{"step": 3175, "epoch": 2.5687702265372168, "time": "2026-06-26 22:42:39", "logs": {"loss": 0.13644008636474608, "grad_norm": 0.06844039261341095, "learning_rate": 6.842765325157874e-06, "epoch": 2.5687702265372168}}
+{"step": 3200, "epoch": 2.588996763754045, "time": "2026-06-26 22:43:01", "logs": {"loss": 0.07055030345916748, "grad_norm": 11.590834617614746, "learning_rate": 6.675799224655588e-06, "epoch": 2.588996763754045}}
+{"step": 3225, "epoch": 2.6092233009708736, "time": "2026-06-26 22:43:23", "logs": {"loss": 0.10780407905578614, "grad_norm": 0.16210342943668365, "learning_rate": 6.509868589844274e-06, "epoch": 2.6092233009708736}}
+{"step": 3250, "epoch": 2.6294498381877025, "time": "2026-06-26 22:43:45", "logs": {"loss": 0.06451849937438965, "grad_norm": 0.1514226347208023, "learning_rate": 6.34502510697607e-06, "epoch": 2.6294498381877025}}
+{"step": 3275, "epoch": 2.649676375404531, "time": "2026-06-26 22:44:07", "logs": {"loss": 0.13269135475158692, "grad_norm": 0.14009274542331696, "learning_rate": 6.181320123662755e-06, "epoch": 2.649676375404531}}
+{"step": 3300, "epoch": 2.6699029126213594, "time": "2026-06-26 22:44:29", "logs": {"loss": 0.1452858829498291, "grad_norm": 1.227739930152893, "learning_rate": 6.018804632881326e-06, "epoch": 2.6699029126213594}}
+{"step": 3325, "epoch": 2.690129449838188, "time": "2026-06-26 22:44:51", "logs": {"loss": 0.11763744354248047, "grad_norm": 0.12412651628255844, "learning_rate": 5.857529257090027e-06, "epoch": 2.690129449838188}}
+{"step": 3350, "epoch": 2.7103559870550162, "time": "2026-06-26 22:45:13", "logs": {"loss": 0.10433626174926758, "grad_norm": 9.471817016601562, "learning_rate": 5.6975442324598165e-06, "epoch": 2.7103559870550162}}
+{"step": 3375, "epoch": 2.7305825242718447, "time": "2026-06-26 22:45:35", "logs": {"loss": 0.14967796325683594, "grad_norm": 5.368896961212158, "learning_rate": 5.538899393226122e-06, "epoch": 2.7305825242718447}}
+{"step": 3400, "epoch": 2.750809061488673, "time": "2026-06-26 22:45:57", "logs": {"loss": 0.13424870491027832, "grad_norm": 0.7815220355987549, "learning_rate": 5.381644156165844e-06, "epoch": 2.750809061488673}}
+{"step": 3425, "epoch": 2.7710355987055015, "time": "2026-06-26 22:46:19", "logs": {"loss": 0.12580464363098146, "grad_norm": 17.132827758789062, "learning_rate": 5.225827505204355e-06, "epoch": 2.7710355987055015}}
+{"step": 3450, "epoch": 2.79126213592233, "time": "2026-06-26 22:46:42", "logs": {"loss": 0.18594202041625976, "grad_norm": 2.3049182891845703, "learning_rate": 5.071497976157346e-06, "epoch": 2.79126213592233}}
+{"step": 3475, "epoch": 2.8114886731391584, "time": "2026-06-26 22:47:04", "logs": {"loss": 0.14881797790527343, "grad_norm": 9.519274711608887, "learning_rate": 4.918703641612255e-06, "epoch": 2.8114886731391584}}
+{"step": 3500, "epoch": 2.831715210355987, "time": "2026-06-26 22:47:26", "logs": {"loss": 0.10424537658691406, "grad_norm": 7.568735599517822, "learning_rate": 4.767492095953985e-06, "epoch": 2.831715210355987}}
+{"step": 3500, "epoch": 2.831715210355987, "time": "2026-06-26 22:48:00", "logs": {"eval_loss": 0.1581990271806717, "eval_accuracy": 0.9676604345629105, "eval_precision": 0.9718954248366013, "eval_recall": 0.9860742705570292, "eval_f1": 0.9789335088874259, "eval_runtime": 34.2504, "eval_samples_per_second": 57.78, "eval_steps_per_second": 3.62, "epoch": 2.831715210355987}}