| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 10.0, | |
| "eval_steps": 500, | |
| "global_step": 4340, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.12, | |
| "learning_rate": 4.8847926267281106e-05, | |
| "loss": 0.3456, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.23, | |
| "learning_rate": 4.7695852534562216e-05, | |
| "loss": 0.286, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.35, | |
| "learning_rate": 4.654377880184332e-05, | |
| "loss": 0.2749, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.46, | |
| "learning_rate": 4.539170506912442e-05, | |
| "loss": 0.2639, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.58, | |
| "learning_rate": 4.423963133640553e-05, | |
| "loss": 0.2596, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.69, | |
| "learning_rate": 4.308755760368664e-05, | |
| "loss": 0.2545, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.81, | |
| "learning_rate": 4.1935483870967746e-05, | |
| "loss": 0.2534, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 0.92, | |
| "learning_rate": 4.078341013824885e-05, | |
| "loss": 0.2542, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 1.04, | |
| "learning_rate": 3.963133640552996e-05, | |
| "loss": 0.2447, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 1.15, | |
| "learning_rate": 3.847926267281106e-05, | |
| "loss": 0.2421, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 1.15, | |
| "eval_accuracy": 0.9349465545650428, | |
| "eval_f1": 0.08678013061547594, | |
| "eval_loss": 0.23889660835266113, | |
| "eval_precision": 0.16307177389364075, | |
| "eval_recall": 0.059120938384791696, | |
| "eval_runtime": 7.6728, | |
| "eval_samples_per_second": 521.324, | |
| "eval_steps_per_second": 65.166, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 1.27, | |
| "learning_rate": 3.7327188940092166e-05, | |
| "loss": 0.2375, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 1.38, | |
| "learning_rate": 3.6175115207373276e-05, | |
| "loss": 0.2319, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 1.5, | |
| "learning_rate": 3.502304147465438e-05, | |
| "loss": 0.2361, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 1.61, | |
| "learning_rate": 3.387096774193548e-05, | |
| "loss": 0.2315, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 1.73, | |
| "learning_rate": 3.271889400921659e-05, | |
| "loss": 0.2313, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 1.84, | |
| "learning_rate": 3.15668202764977e-05, | |
| "loss": 0.2315, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 1.96, | |
| "learning_rate": 3.0414746543778806e-05, | |
| "loss": 0.2238, | |
| "step": 850 | |
| }, | |
| { | |
| "epoch": 2.07, | |
| "learning_rate": 2.926267281105991e-05, | |
| "loss": 0.2145, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 2.19, | |
| "learning_rate": 2.8110599078341016e-05, | |
| "loss": 0.2068, | |
| "step": 950 | |
| }, | |
| { | |
| "epoch": 2.3, | |
| "learning_rate": 2.6958525345622122e-05, | |
| "loss": 0.2065, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 2.3, | |
| "eval_accuracy": 0.938073006886645, | |
| "eval_f1": 0.21391349661281917, | |
| "eval_loss": 0.2181781381368637, | |
| "eval_precision": 0.3005857944837686, | |
| "eval_recall": 0.16603748146150735, | |
| "eval_runtime": 7.74, | |
| "eval_samples_per_second": 516.799, | |
| "eval_steps_per_second": 64.6, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 2.42, | |
| "learning_rate": 2.5806451612903226e-05, | |
| "loss": 0.2053, | |
| "step": 1050 | |
| }, | |
| { | |
| "epoch": 2.53, | |
| "learning_rate": 2.4654377880184332e-05, | |
| "loss": 0.2025, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 2.65, | |
| "learning_rate": 2.350230414746544e-05, | |
| "loss": 0.2018, | |
| "step": 1150 | |
| }, | |
| { | |
| "epoch": 2.76, | |
| "learning_rate": 2.2350230414746546e-05, | |
| "loss": 0.1997, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 2.88, | |
| "learning_rate": 2.1198156682027652e-05, | |
| "loss": 0.2001, | |
| "step": 1250 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "learning_rate": 2.0046082949308755e-05, | |
| "loss": 0.1961, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 3.11, | |
| "learning_rate": 1.8894009216589862e-05, | |
| "loss": 0.1786, | |
| "step": 1350 | |
| }, | |
| { | |
| "epoch": 3.23, | |
| "learning_rate": 1.774193548387097e-05, | |
| "loss": 0.1781, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 3.34, | |
| "learning_rate": 1.6589861751152075e-05, | |
| "loss": 0.176, | |
| "step": 1450 | |
| }, | |
| { | |
| "epoch": 3.46, | |
| "learning_rate": 1.543778801843318e-05, | |
| "loss": 0.1729, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 3.46, | |
| "eval_accuracy": 0.9446110946808061, | |
| "eval_f1": 0.3065833333333333, | |
| "eval_loss": 0.19863183796405792, | |
| "eval_precision": 0.40137464542875845, | |
| "eval_recall": 0.24801132533369286, | |
| "eval_runtime": 7.8899, | |
| "eval_samples_per_second": 506.98, | |
| "eval_steps_per_second": 63.372, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 3.57, | |
| "learning_rate": 1.4285714285714285e-05, | |
| "loss": 0.1724, | |
| "step": 1550 | |
| }, | |
| { | |
| "epoch": 3.69, | |
| "learning_rate": 1.313364055299539e-05, | |
| "loss": 0.1752, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 3.8, | |
| "learning_rate": 1.1981566820276497e-05, | |
| "loss": 0.1749, | |
| "step": 1650 | |
| }, | |
| { | |
| "epoch": 3.92, | |
| "learning_rate": 1.0829493087557604e-05, | |
| "loss": 0.1712, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 4.03, | |
| "learning_rate": 9.67741935483871e-06, | |
| "loss": 0.1639, | |
| "step": 1750 | |
| }, | |
| { | |
| "epoch": 4.15, | |
| "learning_rate": 8.525345622119817e-06, | |
| "loss": 0.1566, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 4.26, | |
| "learning_rate": 7.373271889400922e-06, | |
| "loss": 0.157, | |
| "step": 1850 | |
| }, | |
| { | |
| "epoch": 4.38, | |
| "learning_rate": 6.221198156682028e-06, | |
| "loss": 0.1554, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 4.49, | |
| "learning_rate": 5.0691244239631346e-06, | |
| "loss": 0.1566, | |
| "step": 1950 | |
| }, | |
| { | |
| "epoch": 4.61, | |
| "learning_rate": 3.9170506912442395e-06, | |
| "loss": 0.1558, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 4.61, | |
| "eval_accuracy": 0.9485297224217331, | |
| "eval_f1": 0.3556011489536316, | |
| "eval_loss": 0.18988750874996185, | |
| "eval_precision": 0.454383389261745, | |
| "eval_recall": 0.2920992314952137, | |
| "eval_runtime": 7.6742, | |
| "eval_samples_per_second": 521.226, | |
| "eval_steps_per_second": 65.153, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 4.72, | |
| "learning_rate": 2.7649769585253458e-06, | |
| "loss": 0.1545, | |
| "step": 2050 | |
| }, | |
| { | |
| "epoch": 4.84, | |
| "learning_rate": 1.6129032258064516e-06, | |
| "loss": 0.156, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 4.95, | |
| "learning_rate": 4.6082949308755763e-07, | |
| "loss": 0.1547, | |
| "step": 2150 | |
| }, | |
| { | |
| "epoch": 5.07, | |
| "learning_rate": 2.4654377880184332e-05, | |
| "loss": 0.1501, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 5.18, | |
| "learning_rate": 2.4078341013824887e-05, | |
| "loss": 0.1533, | |
| "step": 2250 | |
| }, | |
| { | |
| "epoch": 5.3, | |
| "learning_rate": 2.350230414746544e-05, | |
| "loss": 0.1555, | |
| "step": 2300 | |
| }, | |
| { | |
| "epoch": 5.41, | |
| "learning_rate": 2.2926267281105994e-05, | |
| "loss": 0.1521, | |
| "step": 2350 | |
| }, | |
| { | |
| "epoch": 5.53, | |
| "learning_rate": 2.2350230414746546e-05, | |
| "loss": 0.1549, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 5.65, | |
| "learning_rate": 2.1774193548387097e-05, | |
| "loss": 0.1529, | |
| "step": 2450 | |
| }, | |
| { | |
| "epoch": 5.76, | |
| "learning_rate": 2.1198156682027652e-05, | |
| "loss": 0.1546, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 5.76, | |
| "eval_accuracy": 0.9504309434281127, | |
| "eval_f1": 0.38409736308316433, | |
| "eval_loss": 0.1856723576784134, | |
| "eval_precision": 0.48227383863080686, | |
| "eval_recall": 0.31913172441688015, | |
| "eval_runtime": 7.5018, | |
| "eval_samples_per_second": 533.207, | |
| "eval_steps_per_second": 66.651, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 5.88, | |
| "learning_rate": 2.0622119815668204e-05, | |
| "loss": 0.152, | |
| "step": 2550 | |
| }, | |
| { | |
| "epoch": 5.99, | |
| "learning_rate": 2.0046082949308755e-05, | |
| "loss": 0.1527, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 6.11, | |
| "learning_rate": 1.9470046082949307e-05, | |
| "loss": 0.1373, | |
| "step": 2650 | |
| }, | |
| { | |
| "epoch": 6.22, | |
| "learning_rate": 1.8894009216589862e-05, | |
| "loss": 0.1373, | |
| "step": 2700 | |
| }, | |
| { | |
| "epoch": 6.34, | |
| "learning_rate": 1.8317972350230417e-05, | |
| "loss": 0.1355, | |
| "step": 2750 | |
| }, | |
| { | |
| "epoch": 6.45, | |
| "learning_rate": 1.774193548387097e-05, | |
| "loss": 0.1367, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 6.57, | |
| "learning_rate": 1.7165898617511524e-05, | |
| "loss": 0.1344, | |
| "step": 2850 | |
| }, | |
| { | |
| "epoch": 6.68, | |
| "learning_rate": 1.6589861751152075e-05, | |
| "loss": 0.135, | |
| "step": 2900 | |
| }, | |
| { | |
| "epoch": 6.8, | |
| "learning_rate": 1.6013824884792627e-05, | |
| "loss": 0.1348, | |
| "step": 2950 | |
| }, | |
| { | |
| "epoch": 6.91, | |
| "learning_rate": 1.543778801843318e-05, | |
| "loss": 0.1343, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 6.91, | |
| "eval_accuracy": 0.9534517723604715, | |
| "eval_f1": 0.4423137378182962, | |
| "eval_loss": 0.17842046916484833, | |
| "eval_precision": 0.5302430751837196, | |
| "eval_recall": 0.3793986787110692, | |
| "eval_runtime": 7.7899, | |
| "eval_samples_per_second": 513.483, | |
| "eval_steps_per_second": 64.185, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 7.03, | |
| "learning_rate": 1.4861751152073732e-05, | |
| "loss": 0.1315, | |
| "step": 3050 | |
| }, | |
| { | |
| "epoch": 7.14, | |
| "learning_rate": 1.4285714285714285e-05, | |
| "loss": 0.1194, | |
| "step": 3100 | |
| }, | |
| { | |
| "epoch": 7.26, | |
| "learning_rate": 1.3709677419354839e-05, | |
| "loss": 0.1237, | |
| "step": 3150 | |
| }, | |
| { | |
| "epoch": 7.37, | |
| "learning_rate": 1.313364055299539e-05, | |
| "loss": 0.1183, | |
| "step": 3200 | |
| }, | |
| { | |
| "epoch": 7.49, | |
| "learning_rate": 1.2557603686635947e-05, | |
| "loss": 0.1231, | |
| "step": 3250 | |
| }, | |
| { | |
| "epoch": 7.6, | |
| "learning_rate": 1.1981566820276497e-05, | |
| "loss": 0.1209, | |
| "step": 3300 | |
| }, | |
| { | |
| "epoch": 7.72, | |
| "learning_rate": 1.1405529953917052e-05, | |
| "loss": 0.1209, | |
| "step": 3350 | |
| }, | |
| { | |
| "epoch": 7.83, | |
| "learning_rate": 1.0829493087557604e-05, | |
| "loss": 0.1213, | |
| "step": 3400 | |
| }, | |
| { | |
| "epoch": 7.95, | |
| "learning_rate": 1.0253456221198157e-05, | |
| "loss": 0.1218, | |
| "step": 3450 | |
| }, | |
| { | |
| "epoch": 8.06, | |
| "learning_rate": 9.67741935483871e-06, | |
| "loss": 0.1163, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 8.06, | |
| "eval_accuracy": 0.9556170518399595, | |
| "eval_f1": 0.47168932038834954, | |
| "eval_loss": 0.17665258049964905, | |
| "eval_precision": 0.55633931843166, | |
| "eval_recall": 0.4093973304570581, | |
| "eval_runtime": 7.4971, | |
| "eval_samples_per_second": 533.536, | |
| "eval_steps_per_second": 66.692, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 8.18, | |
| "learning_rate": 9.101382488479262e-06, | |
| "loss": 0.111, | |
| "step": 3550 | |
| }, | |
| { | |
| "epoch": 8.29, | |
| "learning_rate": 8.525345622119817e-06, | |
| "loss": 0.112, | |
| "step": 3600 | |
| }, | |
| { | |
| "epoch": 8.41, | |
| "learning_rate": 7.949308755760369e-06, | |
| "loss": 0.1106, | |
| "step": 3650 | |
| }, | |
| { | |
| "epoch": 8.53, | |
| "learning_rate": 7.373271889400922e-06, | |
| "loss": 0.1098, | |
| "step": 3700 | |
| }, | |
| { | |
| "epoch": 8.64, | |
| "learning_rate": 6.7972350230414745e-06, | |
| "loss": 0.1125, | |
| "step": 3750 | |
| }, | |
| { | |
| "epoch": 8.76, | |
| "learning_rate": 6.221198156682028e-06, | |
| "loss": 0.1099, | |
| "step": 3800 | |
| }, | |
| { | |
| "epoch": 8.87, | |
| "learning_rate": 5.64516129032258e-06, | |
| "loss": 0.1113, | |
| "step": 3850 | |
| }, | |
| { | |
| "epoch": 8.99, | |
| "learning_rate": 5.0691244239631346e-06, | |
| "loss": 0.1079, | |
| "step": 3900 | |
| }, | |
| { | |
| "epoch": 9.1, | |
| "learning_rate": 4.493087557603687e-06, | |
| "loss": 0.1048, | |
| "step": 3950 | |
| }, | |
| { | |
| "epoch": 9.22, | |
| "learning_rate": 3.9170506912442395e-06, | |
| "loss": 0.1045, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 9.22, | |
| "eval_accuracy": 0.9554269297393215, | |
| "eval_f1": 0.48804591584628837, | |
| "eval_loss": 0.17828913033008575, | |
| "eval_precision": 0.5594771241830065, | |
| "eval_recall": 0.4327895375488742, | |
| "eval_runtime": 7.85, | |
| "eval_samples_per_second": 509.557, | |
| "eval_steps_per_second": 63.695, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 9.33, | |
| "learning_rate": 3.341013824884793e-06, | |
| "loss": 0.1054, | |
| "step": 4050 | |
| }, | |
| { | |
| "epoch": 9.45, | |
| "learning_rate": 2.7649769585253458e-06, | |
| "loss": 0.1061, | |
| "step": 4100 | |
| }, | |
| { | |
| "epoch": 9.56, | |
| "learning_rate": 2.1889400921658987e-06, | |
| "loss": 0.1008, | |
| "step": 4150 | |
| }, | |
| { | |
| "epoch": 9.68, | |
| "learning_rate": 1.6129032258064516e-06, | |
| "loss": 0.1024, | |
| "step": 4200 | |
| }, | |
| { | |
| "epoch": 9.79, | |
| "learning_rate": 1.0368663594470047e-06, | |
| "loss": 0.1034, | |
| "step": 4250 | |
| }, | |
| { | |
| "epoch": 9.91, | |
| "learning_rate": 4.6082949308755763e-07, | |
| "loss": 0.1048, | |
| "step": 4300 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "step": 4340, | |
| "total_flos": 1.3513697075311152e+16, | |
| "train_loss": 0.06244980893376786, | |
| "train_runtime": 304.7861, | |
| "train_samples_per_second": 1822.196, | |
| "train_steps_per_second": 14.239 | |
| } | |
| ], | |
| "logging_steps": 50, | |
| "max_steps": 4340, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 10, | |
| "save_steps": 500, | |
| "total_flos": 1.3513697075311152e+16, | |
| "train_batch_size": 128, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |