diff --git "a/trainer_state.json" "b/trainer_state.json" new file mode 100644--- /dev/null +++ "b/trainer_state.json" @@ -0,0 +1,3130 @@ +{ + "best_metric": null, + "best_model_checkpoint": null, + "epoch": 100.0, + "eval_steps": 500, + "global_step": 9600, + "is_hyper_param_search": false, + "is_local_process_zero": true, + "is_world_process_zero": true, + "log_history": [ + { + "epoch": 1.0, + "grad_norm": 2.1603243350982666, + "learning_rate": 4.9500000000000004e-05, + "loss": 1.0769, + "step": 96 + }, + { + "epoch": 1.0, + "eval_LOCATION_f1": 0.0, + "eval_LOCATION_number": 103, + "eval_LOCATION_precision": 0.0, + "eval_LOCATION_recall": 0.0, + "eval_ORGANIZATION_f1": 0.0, + "eval_ORGANIZATION_number": 171, + "eval_ORGANIZATION_precision": 0.0, + "eval_ORGANIZATION_recall": 0.0, + "eval_PERSON_f1": 0.0, + "eval_PERSON_number": 131, + "eval_PERSON_precision": 0.0, + "eval_PERSON_recall": 0.0, + "eval_loss": 0.6692458391189575, + "eval_overall_accuracy": 0.8373377520022093, + "eval_overall_f1": 0.0, + "eval_overall_precision": 0.0, + "eval_overall_recall": 0.0, + "eval_runtime": 0.5807, + "eval_samples_per_second": 292.757, + "eval_steps_per_second": 5.166, + "step": 96 + }, + { + "epoch": 2.0, + "grad_norm": 0.9606540203094482, + "learning_rate": 4.9e-05, + "loss": 0.6366, + "step": 192 + }, + { + "epoch": 2.0, + "eval_LOCATION_f1": 0.0, + "eval_LOCATION_number": 103, + "eval_LOCATION_precision": 0.0, + "eval_LOCATION_recall": 0.0, + "eval_ORGANIZATION_f1": 0.0, + "eval_ORGANIZATION_number": 171, + "eval_ORGANIZATION_precision": 0.0, + "eval_ORGANIZATION_recall": 0.0, + "eval_PERSON_f1": 0.0, + "eval_PERSON_number": 131, + "eval_PERSON_precision": 0.0, + "eval_PERSON_recall": 0.0, + "eval_loss": 0.5158160924911499, + "eval_overall_accuracy": 0.8381662524164596, + "eval_overall_f1": 0.0, + "eval_overall_precision": 0.0, + "eval_overall_recall": 0.0, + "eval_runtime": 0.6003, + "eval_samples_per_second": 283.189, + "eval_steps_per_second": 4.997, + "step": 192 + }, + { + "epoch": 3.0, + "grad_norm": 1.0227470397949219, + "learning_rate": 4.85e-05, + "loss": 0.4812, + "step": 288 + }, + { + "epoch": 3.0, + "eval_LOCATION_f1": 0.048, + "eval_LOCATION_number": 103, + "eval_LOCATION_precision": 0.13636363636363635, + "eval_LOCATION_recall": 0.02912621359223301, + "eval_ORGANIZATION_f1": 0.27636363636363637, + "eval_ORGANIZATION_number": 171, + "eval_ORGANIZATION_precision": 0.36538461538461536, + "eval_ORGANIZATION_recall": 0.2222222222222222, + "eval_PERSON_f1": 0.2954545454545454, + "eval_PERSON_number": 131, + "eval_PERSON_precision": 0.2932330827067669, + "eval_PERSON_recall": 0.29770992366412213, + "eval_loss": 0.359359472990036, + "eval_overall_accuracy": 0.8762772714719691, + "eval_overall_f1": 0.2409638554216867, + "eval_overall_precision": 0.3088803088803089, + "eval_overall_recall": 0.19753086419753085, + "eval_runtime": 0.6263, + "eval_samples_per_second": 271.426, + "eval_steps_per_second": 4.79, + "step": 288 + }, + { + "epoch": 4.0, + "grad_norm": 1.0585229396820068, + "learning_rate": 4.8e-05, + "loss": 0.3408, + "step": 384 + }, + { + "epoch": 4.0, + "eval_LOCATION_f1": 0.34831460674157305, + "eval_LOCATION_number": 103, + "eval_LOCATION_precision": 0.41333333333333333, + "eval_LOCATION_recall": 0.30097087378640774, + "eval_ORGANIZATION_f1": 0.5435356200527705, + "eval_ORGANIZATION_number": 171, + "eval_ORGANIZATION_precision": 0.4951923076923077, + "eval_ORGANIZATION_recall": 0.6023391812865497, + "eval_PERSON_f1": 0.5657894736842104, + "eval_PERSON_number": 131, + "eval_PERSON_precision": 0.49710982658959535, + "eval_PERSON_recall": 0.6564885496183206, + "eval_loss": 0.2543448805809021, + "eval_overall_accuracy": 0.9207401270367301, + "eval_overall_f1": 0.5110336817653891, + "eval_overall_precision": 0.4824561403508772, + "eval_overall_recall": 0.5432098765432098, + "eval_runtime": 0.6321, + "eval_samples_per_second": 268.959, + "eval_steps_per_second": 4.746, + "step": 384 + }, + { + "epoch": 5.0, + "grad_norm": 0.6169285178184509, + "learning_rate": 4.75e-05, + "loss": 0.2457, + "step": 480 + }, + { + "epoch": 5.0, + "eval_LOCATION_f1": 0.6, + "eval_LOCATION_number": 103, + "eval_LOCATION_precision": 0.6551724137931034, + "eval_LOCATION_recall": 0.5533980582524272, + "eval_ORGANIZATION_f1": 0.712468193384224, + "eval_ORGANIZATION_number": 171, + "eval_ORGANIZATION_precision": 0.6306306306306306, + "eval_ORGANIZATION_recall": 0.8187134502923976, + "eval_PERSON_f1": 0.8592057761732852, + "eval_PERSON_number": 131, + "eval_PERSON_precision": 0.815068493150685, + "eval_PERSON_recall": 0.9083969465648855, + "eval_loss": 0.18795940279960632, + "eval_overall_accuracy": 0.951946975973488, + "eval_overall_f1": 0.7348837209302326, + "eval_overall_precision": 0.6945054945054945, + "eval_overall_recall": 0.7802469135802469, + "eval_runtime": 0.6105, + "eval_samples_per_second": 278.451, + "eval_steps_per_second": 4.914, + "step": 480 + }, + { + "epoch": 6.0, + "grad_norm": 1.268639087677002, + "learning_rate": 4.7e-05, + "loss": 0.1942, + "step": 576 + }, + { + "epoch": 6.0, + "eval_LOCATION_f1": 0.7352941176470589, + "eval_LOCATION_number": 103, + "eval_LOCATION_precision": 0.7425742574257426, + "eval_LOCATION_recall": 0.7281553398058253, + "eval_ORGANIZATION_f1": 0.7913279132791328, + "eval_ORGANIZATION_number": 171, + "eval_ORGANIZATION_precision": 0.7373737373737373, + "eval_ORGANIZATION_recall": 0.8538011695906432, + "eval_PERSON_f1": 0.9264705882352942, + "eval_PERSON_number": 131, + "eval_PERSON_precision": 0.8936170212765957, + "eval_PERSON_recall": 0.9618320610687023, + "eval_loss": 0.146920308470726, + "eval_overall_accuracy": 0.9624413145539906, + "eval_overall_f1": 0.821301775147929, + "eval_overall_precision": 0.7886363636363637, + "eval_overall_recall": 0.8567901234567902, + "eval_runtime": 0.597, + "eval_samples_per_second": 284.775, + "eval_steps_per_second": 5.025, + "step": 576 + }, + { + "epoch": 7.0, + "grad_norm": 1.4306912422180176, + "learning_rate": 4.6500000000000005e-05, + "loss": 0.1609, + "step": 672 + }, + { + "epoch": 7.0, + "eval_LOCATION_f1": 0.8038277511961723, + "eval_LOCATION_number": 103, + "eval_LOCATION_precision": 0.7924528301886793, + "eval_LOCATION_recall": 0.8155339805825242, + "eval_ORGANIZATION_f1": 0.821917808219178, + "eval_ORGANIZATION_number": 171, + "eval_ORGANIZATION_precision": 0.7731958762886598, + "eval_ORGANIZATION_recall": 0.8771929824561403, + "eval_PERSON_f1": 0.9304029304029305, + "eval_PERSON_number": 131, + "eval_PERSON_precision": 0.8943661971830986, + "eval_PERSON_recall": 0.9694656488549618, + "eval_loss": 0.12500987946987152, + "eval_overall_accuracy": 0.9660314830157415, + "eval_overall_f1": 0.8524203069657617, + "eval_overall_precision": 0.8167420814479638, + "eval_overall_recall": 0.891358024691358, + "eval_runtime": 0.5824, + "eval_samples_per_second": 291.909, + "eval_steps_per_second": 5.151, + "step": 672 + }, + { + "epoch": 8.0, + "grad_norm": 0.9121193289756775, + "learning_rate": 4.600000000000001e-05, + "loss": 0.1441, + "step": 768 + }, + { + "epoch": 8.0, + "eval_LOCATION_f1": 0.7924528301886793, + "eval_LOCATION_number": 103, + "eval_LOCATION_precision": 0.7706422018348624, + "eval_LOCATION_recall": 0.8155339805825242, + "eval_ORGANIZATION_f1": 0.8179271708683473, + "eval_ORGANIZATION_number": 171, + "eval_ORGANIZATION_precision": 0.7849462365591398, + "eval_ORGANIZATION_recall": 0.8538011695906432, + "eval_PERSON_f1": 0.951310861423221, + "eval_PERSON_number": 131, + "eval_PERSON_precision": 0.9338235294117647, + "eval_PERSON_recall": 0.9694656488549618, + "eval_loss": 0.1037909984588623, + "eval_overall_accuracy": 0.9693454846727423, + "eval_overall_f1": 0.854066985645933, + "eval_overall_precision": 0.8283062645011601, + "eval_overall_recall": 0.8814814814814815, + "eval_runtime": 0.5885, + "eval_samples_per_second": 288.868, + "eval_steps_per_second": 5.098, + "step": 768 + }, + { + "epoch": 9.0, + "grad_norm": 0.6778863668441772, + "learning_rate": 4.55e-05, + "loss": 0.1317, + "step": 864 + }, + { + "epoch": 9.0, + "eval_LOCATION_f1": 0.8585365853658536, + "eval_LOCATION_number": 103, + "eval_LOCATION_precision": 0.8627450980392157, + "eval_LOCATION_recall": 0.8543689320388349, + "eval_ORGANIZATION_f1": 0.830601092896175, + "eval_ORGANIZATION_number": 171, + "eval_ORGANIZATION_precision": 0.7794871794871795, + "eval_ORGANIZATION_recall": 0.8888888888888888, + "eval_PERSON_f1": 0.9442379182156133, + "eval_PERSON_number": 131, + "eval_PERSON_precision": 0.9202898550724637, + "eval_PERSON_recall": 0.9694656488549618, + "eval_loss": 0.09429998695850372, + "eval_overall_accuracy": 0.9710024855012428, + "eval_overall_f1": 0.8738095238095238, + "eval_overall_precision": 0.8436781609195402, + "eval_overall_recall": 0.9061728395061729, + "eval_runtime": 0.5949, + "eval_samples_per_second": 285.77, + "eval_steps_per_second": 5.043, + "step": 864 + }, + { + "epoch": 10.0, + "grad_norm": 1.0111494064331055, + "learning_rate": 4.5e-05, + "loss": 0.1184, + "step": 960 + }, + { + "epoch": 10.0, + "eval_LOCATION_f1": 0.8380952380952381, + "eval_LOCATION_number": 103, + "eval_LOCATION_precision": 0.822429906542056, + "eval_LOCATION_recall": 0.8543689320388349, + "eval_ORGANIZATION_f1": 0.8484848484848485, + "eval_ORGANIZATION_number": 171, + "eval_ORGANIZATION_precision": 0.8020833333333334, + "eval_ORGANIZATION_recall": 0.9005847953216374, + "eval_PERSON_f1": 0.9407407407407407, + "eval_PERSON_number": 131, + "eval_PERSON_precision": 0.9136690647482014, + "eval_PERSON_recall": 0.9694656488549618, + "eval_loss": 0.08724513649940491, + "eval_overall_accuracy": 0.9712786523059929, + "eval_overall_f1": 0.8754448398576513, + "eval_overall_precision": 0.8424657534246576, + "eval_overall_recall": 0.9111111111111111, + "eval_runtime": 0.6024, + "eval_samples_per_second": 282.214, + "eval_steps_per_second": 4.98, + "step": 960 + }, + { + "epoch": 11.0, + "grad_norm": 1.130439281463623, + "learning_rate": 4.4500000000000004e-05, + "loss": 0.1103, + "step": 1056 + }, + { + "epoch": 11.0, + "eval_LOCATION_f1": 0.8768472906403941, + "eval_LOCATION_number": 103, + "eval_LOCATION_precision": 0.89, + "eval_LOCATION_recall": 0.8640776699029126, + "eval_ORGANIZATION_f1": 0.8555240793201134, + "eval_ORGANIZATION_number": 171, + "eval_ORGANIZATION_precision": 0.8296703296703297, + "eval_ORGANIZATION_recall": 0.8830409356725146, + "eval_PERSON_f1": 0.9438202247191011, + "eval_PERSON_number": 131, + "eval_PERSON_precision": 0.9264705882352942, + "eval_PERSON_recall": 0.9618320610687023, + "eval_loss": 0.07618498057126999, + "eval_overall_accuracy": 0.9748688207677437, + "eval_overall_f1": 0.8894289185905225, + "eval_overall_precision": 0.8755980861244019, + "eval_overall_recall": 0.9037037037037037, + "eval_runtime": 0.5964, + "eval_samples_per_second": 285.062, + "eval_steps_per_second": 5.031, + "step": 1056 + }, + { + "epoch": 12.0, + "grad_norm": 1.6433125734329224, + "learning_rate": 4.4000000000000006e-05, + "loss": 0.1049, + "step": 1152 + }, + { + "epoch": 12.0, + "eval_LOCATION_f1": 0.8792270531400966, + "eval_LOCATION_number": 103, + "eval_LOCATION_precision": 0.875, + "eval_LOCATION_recall": 0.883495145631068, + "eval_ORGANIZATION_f1": 0.8619718309859156, + "eval_ORGANIZATION_number": 171, + "eval_ORGANIZATION_precision": 0.8315217391304348, + "eval_ORGANIZATION_recall": 0.8947368421052632, + "eval_PERSON_f1": 0.9548872180451129, + "eval_PERSON_number": 131, + "eval_PERSON_precision": 0.9407407407407408, + "eval_PERSON_recall": 0.9694656488549618, + "eval_loss": 0.07063060998916626, + "eval_overall_accuracy": 0.9779066556199945, + "eval_overall_f1": 0.8961352657004832, + "eval_overall_precision": 0.8770685579196218, + "eval_overall_recall": 0.9160493827160494, + "eval_runtime": 0.6031, + "eval_samples_per_second": 281.862, + "eval_steps_per_second": 4.974, + "step": 1152 + }, + { + "epoch": 13.0, + "grad_norm": 1.8723031282424927, + "learning_rate": 4.35e-05, + "loss": 0.0961, + "step": 1248 + }, + { + "epoch": 13.0, + "eval_LOCATION_f1": 0.8750000000000001, + "eval_LOCATION_number": 103, + "eval_LOCATION_precision": 0.8666666666666667, + "eval_LOCATION_recall": 0.883495145631068, + "eval_ORGANIZATION_f1": 0.8703170028818444, + "eval_ORGANIZATION_number": 171, + "eval_ORGANIZATION_precision": 0.8579545454545454, + "eval_ORGANIZATION_recall": 0.8830409356725146, + "eval_PERSON_f1": 0.9584905660377357, + "eval_PERSON_number": 131, + "eval_PERSON_precision": 0.9477611940298507, + "eval_PERSON_recall": 0.9694656488549618, + "eval_loss": 0.06398220360279083, + "eval_overall_accuracy": 0.9779066556199945, + "eval_overall_f1": 0.8999999999999999, + "eval_overall_precision": 0.8891566265060241, + "eval_overall_recall": 0.9111111111111111, + "eval_runtime": 0.5978, + "eval_samples_per_second": 284.362, + "eval_steps_per_second": 5.018, + "step": 1248 + }, + { + "epoch": 14.0, + "grad_norm": 0.660153329372406, + "learning_rate": 4.3e-05, + "loss": 0.0909, + "step": 1344 + }, + { + "epoch": 14.0, + "eval_LOCATION_f1": 0.8653846153846153, + "eval_LOCATION_number": 103, + "eval_LOCATION_precision": 0.8571428571428571, + "eval_LOCATION_recall": 0.8737864077669902, + "eval_ORGANIZATION_f1": 0.8802395209580838, + "eval_ORGANIZATION_number": 171, + "eval_ORGANIZATION_precision": 0.901840490797546, + "eval_ORGANIZATION_recall": 0.8596491228070176, + "eval_PERSON_f1": 0.9694656488549618, + "eval_PERSON_number": 131, + "eval_PERSON_precision": 0.9694656488549618, + "eval_PERSON_recall": 0.9694656488549618, + "eval_loss": 0.06527247279882431, + "eval_overall_accuracy": 0.980115990057995, + "eval_overall_f1": 0.9054726368159204, + "eval_overall_precision": 0.9122807017543859, + "eval_overall_recall": 0.8987654320987655, + "eval_runtime": 0.5953, + "eval_samples_per_second": 285.592, + "eval_steps_per_second": 5.04, + "step": 1344 + }, + { + "epoch": 15.0, + "grad_norm": 1.0047082901000977, + "learning_rate": 4.25e-05, + "loss": 0.0888, + "step": 1440 + }, + { + "epoch": 15.0, + "eval_LOCATION_f1": 0.8975609756097561, + "eval_LOCATION_number": 103, + "eval_LOCATION_precision": 0.9019607843137255, + "eval_LOCATION_recall": 0.8932038834951457, + "eval_ORGANIZATION_f1": 0.8838526912181304, + "eval_ORGANIZATION_number": 171, + "eval_ORGANIZATION_precision": 0.8571428571428571, + "eval_ORGANIZATION_recall": 0.9122807017543859, + "eval_PERSON_f1": 0.9657794676806084, + "eval_PERSON_number": 131, + "eval_PERSON_precision": 0.9621212121212122, + "eval_PERSON_recall": 0.9694656488549618, + "eval_loss": 0.05794135853648186, + "eval_overall_accuracy": 0.9809444904722452, + "eval_overall_f1": 0.9135200974421438, + "eval_overall_precision": 0.9014423076923077, + "eval_overall_recall": 0.9259259259259259, + "eval_runtime": 0.5911, + "eval_samples_per_second": 287.609, + "eval_steps_per_second": 5.075, + "step": 1440 + }, + { + "epoch": 16.0, + "grad_norm": 1.2689850330352783, + "learning_rate": 4.2e-05, + "loss": 0.0873, + "step": 1536 + }, + { + "epoch": 16.0, + "eval_LOCATION_f1": 0.8899521531100479, + "eval_LOCATION_number": 103, + "eval_LOCATION_precision": 0.8773584905660378, + "eval_LOCATION_recall": 0.9029126213592233, + "eval_ORGANIZATION_f1": 0.8851540616246498, + "eval_ORGANIZATION_number": 171, + "eval_ORGANIZATION_precision": 0.8494623655913979, + "eval_ORGANIZATION_recall": 0.9239766081871345, + "eval_PERSON_f1": 0.9581749049429659, + "eval_PERSON_number": 131, + "eval_PERSON_precision": 0.9545454545454546, + "eval_PERSON_recall": 0.9618320610687023, + "eval_loss": 0.05623210594058037, + "eval_overall_accuracy": 0.9814968240817453, + "eval_overall_f1": 0.9095295536791316, + "eval_overall_precision": 0.8891509433962265, + "eval_overall_recall": 0.9308641975308642, + "eval_runtime": 0.602, + "eval_samples_per_second": 282.375, + "eval_steps_per_second": 4.983, + "step": 1536 + }, + { + "epoch": 17.0, + "grad_norm": 0.6952976584434509, + "learning_rate": 4.15e-05, + "loss": 0.0827, + "step": 1632 + }, + { + "epoch": 17.0, + "eval_LOCATION_f1": 0.8921568627450982, + "eval_LOCATION_number": 103, + "eval_LOCATION_precision": 0.900990099009901, + "eval_LOCATION_recall": 0.883495145631068, + "eval_ORGANIZATION_f1": 0.877906976744186, + "eval_ORGANIZATION_number": 171, + "eval_ORGANIZATION_precision": 0.8728323699421965, + "eval_ORGANIZATION_recall": 0.8830409356725146, + "eval_PERSON_f1": 0.9657794676806084, + "eval_PERSON_number": 131, + "eval_PERSON_precision": 0.9621212121212122, + "eval_PERSON_recall": 0.9694656488549618, + "eval_loss": 0.055690497159957886, + "eval_overall_accuracy": 0.9806683236674951, + "eval_overall_f1": 0.9099876695437731, + "eval_overall_precision": 0.9088669950738916, + "eval_overall_recall": 0.9111111111111111, + "eval_runtime": 0.6671, + "eval_samples_per_second": 254.852, + "eval_steps_per_second": 4.497, + "step": 1632 + }, + { + "epoch": 18.0, + "grad_norm": 0.47416993975639343, + "learning_rate": 4.1e-05, + "loss": 0.0798, + "step": 1728 + }, + { + "epoch": 18.0, + "eval_LOCATION_f1": 0.8942307692307693, + "eval_LOCATION_number": 103, + "eval_LOCATION_precision": 0.8857142857142857, + "eval_LOCATION_recall": 0.9029126213592233, + "eval_ORGANIZATION_f1": 0.904899135446686, + "eval_ORGANIZATION_number": 171, + "eval_ORGANIZATION_precision": 0.8920454545454546, + "eval_ORGANIZATION_recall": 0.9181286549707602, + "eval_PERSON_f1": 0.9545454545454546, + "eval_PERSON_number": 131, + "eval_PERSON_precision": 0.9473684210526315, + "eval_PERSON_recall": 0.9618320610687023, + "eval_loss": 0.051382821053266525, + "eval_overall_accuracy": 0.9845346589339962, + "eval_overall_f1": 0.9181929181929182, + "eval_overall_precision": 0.9082125603864735, + "eval_overall_recall": 0.928395061728395, + "eval_runtime": 0.6057, + "eval_samples_per_second": 280.66, + "eval_steps_per_second": 4.953, + "step": 1728 + }, + { + "epoch": 19.0, + "grad_norm": 0.9612790942192078, + "learning_rate": 4.05e-05, + "loss": 0.076, + "step": 1824 + }, + { + "epoch": 19.0, + "eval_LOCATION_f1": 0.903846153846154, + "eval_LOCATION_number": 103, + "eval_LOCATION_precision": 0.8952380952380953, + "eval_LOCATION_recall": 0.912621359223301, + "eval_ORGANIZATION_f1": 0.8979591836734693, + "eval_ORGANIZATION_number": 171, + "eval_ORGANIZATION_precision": 0.8953488372093024, + "eval_ORGANIZATION_recall": 0.9005847953216374, + "eval_PERSON_f1": 0.9545454545454546, + "eval_PERSON_number": 131, + "eval_PERSON_precision": 0.9473684210526315, + "eval_PERSON_recall": 0.9618320610687023, + "eval_loss": 0.05274493247270584, + "eval_overall_accuracy": 0.9834299917149959, + "eval_overall_f1": 0.9177914110429448, + "eval_overall_precision": 0.9121951219512195, + "eval_overall_recall": 0.9234567901234568, + "eval_runtime": 0.5991, + "eval_samples_per_second": 283.754, + "eval_steps_per_second": 5.007, + "step": 1824 + }, + { + "epoch": 20.0, + "grad_norm": 0.6812459826469421, + "learning_rate": 4e-05, + "loss": 0.0712, + "step": 1920 + }, + { + "epoch": 20.0, + "eval_LOCATION_f1": 0.9090909090909091, + "eval_LOCATION_number": 103, + "eval_LOCATION_precision": 0.8962264150943396, + "eval_LOCATION_recall": 0.9223300970873787, + "eval_ORGANIZATION_f1": 0.8977272727272728, + "eval_ORGANIZATION_number": 171, + "eval_ORGANIZATION_precision": 0.8729281767955801, + "eval_ORGANIZATION_recall": 0.9239766081871345, + "eval_PERSON_f1": 0.9584905660377357, + "eval_PERSON_number": 131, + "eval_PERSON_precision": 0.9477611940298507, + "eval_PERSON_recall": 0.9694656488549618, + "eval_loss": 0.05241977423429489, + "eval_overall_accuracy": 0.983706158519746, + "eval_overall_f1": 0.920096852300242, + "eval_overall_precision": 0.9026128266033254, + "eval_overall_recall": 0.9382716049382716, + "eval_runtime": 0.6023, + "eval_samples_per_second": 282.233, + "eval_steps_per_second": 4.981, + "step": 1920 + }, + { + "epoch": 21.0, + "grad_norm": 1.0114864110946655, + "learning_rate": 3.9500000000000005e-05, + "loss": 0.072, + "step": 2016 + }, + { + "epoch": 21.0, + "eval_LOCATION_f1": 0.903846153846154, + "eval_LOCATION_number": 103, + "eval_LOCATION_precision": 0.8952380952380953, + "eval_LOCATION_recall": 0.912621359223301, + "eval_ORGANIZATION_f1": 0.8914956011730205, + "eval_ORGANIZATION_number": 171, + "eval_ORGANIZATION_precision": 0.8941176470588236, + "eval_ORGANIZATION_recall": 0.8888888888888888, + "eval_PERSON_f1": 0.962121212121212, + "eval_PERSON_number": 131, + "eval_PERSON_precision": 0.9548872180451128, + "eval_PERSON_recall": 0.9694656488549618, + "eval_loss": 0.050774551928043365, + "eval_overall_accuracy": 0.983706158519746, + "eval_overall_f1": 0.9175891758917589, + "eval_overall_precision": 0.9142156862745098, + "eval_overall_recall": 0.9209876543209876, + "eval_runtime": 0.5964, + "eval_samples_per_second": 285.065, + "eval_steps_per_second": 5.031, + "step": 2016 + }, + { + "epoch": 22.0, + "grad_norm": 0.9129711985588074, + "learning_rate": 3.9000000000000006e-05, + "loss": 0.0717, + "step": 2112 + }, + { + "epoch": 22.0, + "eval_LOCATION_f1": 0.8952380952380954, + "eval_LOCATION_number": 103, + "eval_LOCATION_precision": 0.8785046728971962, + "eval_LOCATION_recall": 0.912621359223301, + "eval_ORGANIZATION_f1": 0.8863636363636364, + "eval_ORGANIZATION_number": 171, + "eval_ORGANIZATION_precision": 0.861878453038674, + "eval_ORGANIZATION_recall": 0.9122807017543859, + "eval_PERSON_f1": 0.9694656488549618, + "eval_PERSON_number": 131, + "eval_PERSON_precision": 0.9694656488549618, + "eval_PERSON_recall": 0.9694656488549618, + "eval_loss": 0.04809549078345299, + "eval_overall_accuracy": 0.9828776581054958, + "eval_overall_f1": 0.9150485436893205, + "eval_overall_precision": 0.8997613365155132, + "eval_overall_recall": 0.9308641975308642, + "eval_runtime": 0.598, + "eval_samples_per_second": 284.294, + "eval_steps_per_second": 5.017, + "step": 2112 + }, + { + "epoch": 23.0, + "grad_norm": 2.053520679473877, + "learning_rate": 3.85e-05, + "loss": 0.0644, + "step": 2208 + }, + { + "epoch": 23.0, + "eval_LOCATION_f1": 0.8857142857142858, + "eval_LOCATION_number": 103, + "eval_LOCATION_precision": 0.8691588785046729, + "eval_LOCATION_recall": 0.9029126213592233, + "eval_ORGANIZATION_f1": 0.9064327485380118, + "eval_ORGANIZATION_number": 171, + "eval_ORGANIZATION_precision": 0.9064327485380117, + "eval_ORGANIZATION_recall": 0.9064327485380117, + "eval_PERSON_f1": 0.9694656488549618, + "eval_PERSON_number": 131, + "eval_PERSON_precision": 0.9694656488549618, + "eval_PERSON_recall": 0.9694656488549618, + "eval_loss": 0.04921582341194153, + "eval_overall_accuracy": 0.9842584921292461, + "eval_overall_f1": 0.9213759213759213, + "eval_overall_precision": 0.9168704156479217, + "eval_overall_recall": 0.9259259259259259, + "eval_runtime": 0.6297, + "eval_samples_per_second": 269.953, + "eval_steps_per_second": 4.764, + "step": 2208 + }, + { + "epoch": 24.0, + "grad_norm": 0.9503189325332642, + "learning_rate": 3.8e-05, + "loss": 0.0647, + "step": 2304 + }, + { + "epoch": 24.0, + "eval_LOCATION_f1": 0.8857142857142858, + "eval_LOCATION_number": 103, + "eval_LOCATION_precision": 0.8691588785046729, + "eval_LOCATION_recall": 0.9029126213592233, + "eval_ORGANIZATION_f1": 0.8882352941176469, + "eval_ORGANIZATION_number": 171, + "eval_ORGANIZATION_precision": 0.893491124260355, + "eval_ORGANIZATION_recall": 0.8830409356725146, + "eval_PERSON_f1": 0.9694656488549618, + "eval_PERSON_number": 131, + "eval_PERSON_precision": 0.9694656488549618, + "eval_PERSON_recall": 0.9694656488549618, + "eval_loss": 0.04936998710036278, + "eval_overall_accuracy": 0.9826014913007457, + "eval_overall_f1": 0.9137931034482758, + "eval_overall_precision": 0.9115479115479116, + "eval_overall_recall": 0.9160493827160494, + "eval_runtime": 0.6072, + "eval_samples_per_second": 279.994, + "eval_steps_per_second": 4.941, + "step": 2304 + }, + { + "epoch": 25.0, + "grad_norm": 0.8211380839347839, + "learning_rate": 3.7500000000000003e-05, + "loss": 0.0652, + "step": 2400 + }, + { + "epoch": 25.0, + "eval_LOCATION_f1": 0.8857142857142858, + "eval_LOCATION_number": 103, + "eval_LOCATION_precision": 0.8691588785046729, + "eval_LOCATION_recall": 0.9029126213592233, + "eval_ORGANIZATION_f1": 0.8802395209580838, + "eval_ORGANIZATION_number": 171, + "eval_ORGANIZATION_precision": 0.901840490797546, + "eval_ORGANIZATION_recall": 0.8596491228070176, + "eval_PERSON_f1": 0.9657794676806084, + "eval_PERSON_number": 131, + "eval_PERSON_precision": 0.9621212121212122, + "eval_PERSON_recall": 0.9694656488549618, + "eval_loss": 0.051058411598205566, + "eval_overall_accuracy": 0.9814968240817453, + "eval_overall_f1": 0.9095415117719949, + "eval_overall_precision": 0.9129353233830846, + "eval_overall_recall": 0.9061728395061729, + "eval_runtime": 0.6072, + "eval_samples_per_second": 279.989, + "eval_steps_per_second": 4.941, + "step": 2400 + }, + { + "epoch": 26.0, + "grad_norm": 0.9145235419273376, + "learning_rate": 3.7e-05, + "loss": 0.0635, + "step": 2496 + }, + { + "epoch": 26.0, + "eval_LOCATION_f1": 0.8985507246376813, + "eval_LOCATION_number": 103, + "eval_LOCATION_precision": 0.8942307692307693, + "eval_LOCATION_recall": 0.9029126213592233, + "eval_ORGANIZATION_f1": 0.9085714285714285, + "eval_ORGANIZATION_number": 171, + "eval_ORGANIZATION_precision": 0.888268156424581, + "eval_ORGANIZATION_recall": 0.9298245614035088, + "eval_PERSON_f1": 0.9694656488549618, + "eval_PERSON_number": 131, + "eval_PERSON_precision": 0.9694656488549618, + "eval_PERSON_recall": 0.9694656488549618, + "eval_loss": 0.046471431851387024, + "eval_overall_accuracy": 0.9845346589339962, + "eval_overall_f1": 0.9255189255189255, + "eval_overall_precision": 0.9154589371980676, + "eval_overall_recall": 0.9358024691358025, + "eval_runtime": 0.6076, + "eval_samples_per_second": 279.789, + "eval_steps_per_second": 4.937, + "step": 2496 + }, + { + "epoch": 27.0, + "grad_norm": 1.1205835342407227, + "learning_rate": 3.65e-05, + "loss": 0.0597, + "step": 2592 + }, + { + "epoch": 27.0, + "eval_LOCATION_f1": 0.8995215311004785, + "eval_LOCATION_number": 103, + "eval_LOCATION_precision": 0.8867924528301887, + "eval_LOCATION_recall": 0.912621359223301, + "eval_ORGANIZATION_f1": 0.9127906976744187, + "eval_ORGANIZATION_number": 171, + "eval_ORGANIZATION_precision": 0.9075144508670521, + "eval_ORGANIZATION_recall": 0.9181286549707602, + "eval_PERSON_f1": 0.9545454545454546, + "eval_PERSON_number": 131, + "eval_PERSON_precision": 0.9473684210526315, + "eval_PERSON_recall": 0.9618320610687023, + "eval_loss": 0.045045264065265656, + "eval_overall_accuracy": 0.9859154929577465, + "eval_overall_f1": 0.922888616891065, + "eval_overall_precision": 0.9150485436893204, + "eval_overall_recall": 0.9308641975308642, + "eval_runtime": 0.603, + "eval_samples_per_second": 281.922, + "eval_steps_per_second": 4.975, + "step": 2592 + }, + { + "epoch": 28.0, + "grad_norm": 0.4941413402557373, + "learning_rate": 3.6e-05, + "loss": 0.0596, + "step": 2688 + }, + { + "epoch": 28.0, + "eval_LOCATION_f1": 0.8952380952380954, + "eval_LOCATION_number": 103, + "eval_LOCATION_precision": 0.8785046728971962, + "eval_LOCATION_recall": 0.912621359223301, + "eval_ORGANIZATION_f1": 0.9106628242074928, + "eval_ORGANIZATION_number": 171, + "eval_ORGANIZATION_precision": 0.8977272727272727, + "eval_ORGANIZATION_recall": 0.9239766081871345, + "eval_PERSON_f1": 0.9808429118773947, + "eval_PERSON_number": 131, + "eval_PERSON_precision": 0.9846153846153847, + "eval_PERSON_recall": 0.9770992366412213, + "eval_loss": 0.04560532048344612, + "eval_overall_accuracy": 0.9864678265672466, + "eval_overall_f1": 0.9290953545232274, + "eval_overall_precision": 0.9200968523002422, + "eval_overall_recall": 0.9382716049382716, + "eval_runtime": 0.6062, + "eval_samples_per_second": 280.448, + "eval_steps_per_second": 4.949, + "step": 2688 + }, + { + "epoch": 29.0, + "grad_norm": 0.585677981376648, + "learning_rate": 3.55e-05, + "loss": 0.0588, + "step": 2784 + }, + { + "epoch": 29.0, + "eval_LOCATION_f1": 0.8888888888888888, + "eval_LOCATION_number": 103, + "eval_LOCATION_precision": 0.8846153846153846, + "eval_LOCATION_recall": 0.8932038834951457, + "eval_ORGANIZATION_f1": 0.8850574712643677, + "eval_ORGANIZATION_number": 171, + "eval_ORGANIZATION_precision": 0.8700564971751412, + "eval_ORGANIZATION_recall": 0.9005847953216374, + "eval_PERSON_f1": 0.9581749049429659, + "eval_PERSON_number": 131, + "eval_PERSON_precision": 0.9545454545454546, + "eval_PERSON_recall": 0.9618320610687023, + "eval_loss": 0.04385360702872276, + "eval_overall_accuracy": 0.9845346589339962, + "eval_overall_f1": 0.9095354523227382, + "eval_overall_precision": 0.9007263922518159, + "eval_overall_recall": 0.9185185185185185, + "eval_runtime": 0.6062, + "eval_samples_per_second": 280.425, + "eval_steps_per_second": 4.949, + "step": 2784 + }, + { + "epoch": 30.0, + "grad_norm": 0.7528411746025085, + "learning_rate": 3.5e-05, + "loss": 0.0546, + "step": 2880 + }, + { + "epoch": 30.0, + "eval_LOCATION_f1": 0.8899521531100479, + "eval_LOCATION_number": 103, + "eval_LOCATION_precision": 0.8773584905660378, + "eval_LOCATION_recall": 0.9029126213592233, + "eval_ORGANIZATION_f1": 0.8901734104046243, + "eval_ORGANIZATION_number": 171, + "eval_ORGANIZATION_precision": 0.88, + "eval_ORGANIZATION_recall": 0.9005847953216374, + "eval_PERSON_f1": 0.9694656488549618, + "eval_PERSON_number": 131, + "eval_PERSON_precision": 0.9694656488549618, + "eval_PERSON_recall": 0.9694656488549618, + "eval_loss": 0.045274246484041214, + "eval_overall_accuracy": 0.9848108257387462, + "eval_overall_f1": 0.9155446756425949, + "eval_overall_precision": 0.9077669902912622, + "eval_overall_recall": 0.9234567901234568, + "eval_runtime": 0.6061, + "eval_samples_per_second": 280.481, + "eval_steps_per_second": 4.95, + "step": 2880 + }, + { + "epoch": 31.0, + "grad_norm": 0.530910849571228, + "learning_rate": 3.45e-05, + "loss": 0.0543, + "step": 2976 + }, + { + "epoch": 31.0, + "eval_LOCATION_f1": 0.8703703703703703, + "eval_LOCATION_number": 103, + "eval_LOCATION_precision": 0.831858407079646, + "eval_LOCATION_recall": 0.912621359223301, + "eval_ORGANIZATION_f1": 0.8842729970326408, + "eval_ORGANIZATION_number": 171, + "eval_ORGANIZATION_precision": 0.8975903614457831, + "eval_ORGANIZATION_recall": 0.8713450292397661, + "eval_PERSON_f1": 0.9694656488549618, + "eval_PERSON_number": 131, + "eval_PERSON_precision": 0.9694656488549618, + "eval_PERSON_recall": 0.9694656488549618, + "eval_loss": 0.04553670436143875, + "eval_overall_accuracy": 0.9831538249102458, + "eval_overall_f1": 0.9079754601226994, + "eval_overall_precision": 0.9024390243902439, + "eval_overall_recall": 0.9135802469135802, + "eval_runtime": 0.5995, + "eval_samples_per_second": 283.552, + "eval_steps_per_second": 5.004, + "step": 2976 + }, + { + "epoch": 32.0, + "grad_norm": 0.8336060643196106, + "learning_rate": 3.4000000000000007e-05, + "loss": 0.0536, + "step": 3072 + }, + { + "epoch": 32.0, + "eval_LOCATION_f1": 0.8651162790697675, + "eval_LOCATION_number": 103, + "eval_LOCATION_precision": 0.8303571428571429, + "eval_LOCATION_recall": 0.9029126213592233, + "eval_ORGANIZATION_f1": 0.8862275449101796, + "eval_ORGANIZATION_number": 171, + "eval_ORGANIZATION_precision": 0.9079754601226994, + "eval_ORGANIZATION_recall": 0.8654970760233918, + "eval_PERSON_f1": 0.9694656488549618, + "eval_PERSON_number": 131, + "eval_PERSON_precision": 0.9694656488549618, + "eval_PERSON_recall": 0.9694656488549618, + "eval_loss": 0.04577063396573067, + "eval_overall_accuracy": 0.9834299917149959, + "eval_overall_f1": 0.9075215782983971, + "eval_overall_precision": 0.9064039408866995, + "eval_overall_recall": 0.908641975308642, + "eval_runtime": 0.605, + "eval_samples_per_second": 280.986, + "eval_steps_per_second": 4.959, + "step": 3072 + }, + { + "epoch": 33.0, + "grad_norm": 0.7508697509765625, + "learning_rate": 3.35e-05, + "loss": 0.0529, + "step": 3168 + }, + { + "epoch": 33.0, + "eval_LOCATION_f1": 0.8675799086757991, + "eval_LOCATION_number": 103, + "eval_LOCATION_precision": 0.8189655172413793, + "eval_LOCATION_recall": 0.9223300970873787, + "eval_ORGANIZATION_f1": 0.8821752265861027, + "eval_ORGANIZATION_number": 171, + "eval_ORGANIZATION_precision": 0.9125, + "eval_ORGANIZATION_recall": 0.8538011695906432, + "eval_PERSON_f1": 0.9694656488549618, + "eval_PERSON_number": 131, + "eval_PERSON_precision": 0.9694656488549618, + "eval_PERSON_recall": 0.9694656488549618, + "eval_loss": 0.04717917740345001, + "eval_overall_accuracy": 0.9831538249102458, + "eval_overall_f1": 0.9064039408866995, + "eval_overall_precision": 0.9041769041769042, + "eval_overall_recall": 0.908641975308642, + "eval_runtime": 0.6092, + "eval_samples_per_second": 279.064, + "eval_steps_per_second": 4.925, + "step": 3168 + }, + { + "epoch": 34.0, + "grad_norm": 0.6504380702972412, + "learning_rate": 3.3e-05, + "loss": 0.054, + "step": 3264 + }, + { + "epoch": 34.0, + "eval_LOCATION_f1": 0.8995215311004785, + "eval_LOCATION_number": 103, + "eval_LOCATION_precision": 0.8867924528301887, + "eval_LOCATION_recall": 0.912621359223301, + "eval_ORGANIZATION_f1": 0.8843930635838151, + "eval_ORGANIZATION_number": 171, + "eval_ORGANIZATION_precision": 0.8742857142857143, + "eval_ORGANIZATION_recall": 0.8947368421052632, + "eval_PERSON_f1": 0.9694656488549618, + "eval_PERSON_number": 131, + "eval_PERSON_precision": 0.9694656488549618, + "eval_PERSON_recall": 0.9694656488549618, + "eval_loss": 0.04478117451071739, + "eval_overall_accuracy": 0.9850869925434963, + "eval_overall_f1": 0.9155446756425949, + "eval_overall_precision": 0.9077669902912622, + "eval_overall_recall": 0.9234567901234568, + "eval_runtime": 0.5952, + "eval_samples_per_second": 285.597, + "eval_steps_per_second": 5.04, + "step": 3264 + }, + { + "epoch": 35.0, + "grad_norm": 1.1786144971847534, + "learning_rate": 3.2500000000000004e-05, + "loss": 0.052, + "step": 3360 + }, + { + "epoch": 35.0, + "eval_LOCATION_f1": 0.8773584905660378, + "eval_LOCATION_number": 103, + "eval_LOCATION_precision": 0.8532110091743119, + "eval_LOCATION_recall": 0.9029126213592233, + "eval_ORGANIZATION_f1": 0.8823529411764705, + "eval_ORGANIZATION_number": 171, + "eval_ORGANIZATION_precision": 0.8875739644970414, + "eval_ORGANIZATION_recall": 0.8771929824561403, + "eval_PERSON_f1": 0.9694656488549618, + "eval_PERSON_number": 131, + "eval_PERSON_precision": 0.9694656488549618, + "eval_PERSON_recall": 0.9694656488549618, + "eval_loss": 0.044395700097084045, + "eval_overall_accuracy": 0.9845346589339962, + "eval_overall_f1": 0.9090909090909091, + "eval_overall_precision": 0.9046454767726161, + "eval_overall_recall": 0.9135802469135802, + "eval_runtime": 0.6073, + "eval_samples_per_second": 279.931, + "eval_steps_per_second": 4.94, + "step": 3360 + }, + { + "epoch": 36.0, + "grad_norm": 1.2726478576660156, + "learning_rate": 3.2000000000000005e-05, + "loss": 0.0513, + "step": 3456 + }, + { + "epoch": 36.0, + "eval_LOCATION_f1": 0.8744186046511627, + "eval_LOCATION_number": 103, + "eval_LOCATION_precision": 0.8392857142857143, + "eval_LOCATION_recall": 0.912621359223301, + "eval_ORGANIZATION_f1": 0.8746355685131195, + "eval_ORGANIZATION_number": 171, + "eval_ORGANIZATION_precision": 0.872093023255814, + "eval_ORGANIZATION_recall": 0.8771929824561403, + "eval_PERSON_f1": 0.9694656488549618, + "eval_PERSON_number": 131, + "eval_PERSON_precision": 0.9694656488549618, + "eval_PERSON_recall": 0.9694656488549618, + "eval_loss": 0.04507197067141533, + "eval_overall_accuracy": 0.983706158519746, + "eval_overall_f1": 0.9048780487804877, + "eval_overall_precision": 0.8939759036144578, + "eval_overall_recall": 0.9160493827160494, + "eval_runtime": 0.6039, + "eval_samples_per_second": 281.521, + "eval_steps_per_second": 4.968, + "step": 3456 + }, + { + "epoch": 37.0, + "grad_norm": 1.9155858755111694, + "learning_rate": 3.15e-05, + "loss": 0.0501, + "step": 3552 + }, + { + "epoch": 37.0, + "eval_LOCATION_f1": 0.9073170731707318, + "eval_LOCATION_number": 103, + "eval_LOCATION_precision": 0.9117647058823529, + "eval_LOCATION_recall": 0.9029126213592233, + "eval_ORGANIZATION_f1": 0.8908045977011494, + "eval_ORGANIZATION_number": 171, + "eval_ORGANIZATION_precision": 0.8757062146892656, + "eval_ORGANIZATION_recall": 0.9064327485380117, + "eval_PERSON_f1": 0.9694656488549618, + "eval_PERSON_number": 131, + "eval_PERSON_precision": 0.9694656488549618, + "eval_PERSON_recall": 0.9694656488549618, + "eval_loss": 0.0453326515853405, + "eval_overall_accuracy": 0.9850869925434963, + "eval_overall_f1": 0.9202453987730062, + "eval_overall_precision": 0.9146341463414634, + "eval_overall_recall": 0.9259259259259259, + "eval_runtime": 0.6018, + "eval_samples_per_second": 282.497, + "eval_steps_per_second": 4.985, + "step": 3552 + }, + { + "epoch": 38.0, + "grad_norm": 0.8169270157814026, + "learning_rate": 3.1e-05, + "loss": 0.048, + "step": 3648 + }, + { + "epoch": 38.0, + "eval_LOCATION_f1": 0.8867924528301887, + "eval_LOCATION_number": 103, + "eval_LOCATION_precision": 0.8623853211009175, + "eval_LOCATION_recall": 0.912621359223301, + "eval_ORGANIZATION_f1": 0.8862275449101796, + "eval_ORGANIZATION_number": 171, + "eval_ORGANIZATION_precision": 0.9079754601226994, + "eval_ORGANIZATION_recall": 0.8654970760233918, + "eval_PERSON_f1": 0.9694656488549618, + "eval_PERSON_number": 131, + "eval_PERSON_precision": 0.9694656488549618, + "eval_PERSON_recall": 0.9694656488549618, + "eval_loss": 0.04849234223365784, + "eval_overall_accuracy": 0.983706158519746, + "eval_overall_f1": 0.9133663366336634, + "eval_overall_precision": 0.9156327543424317, + "eval_overall_recall": 0.9111111111111111, + "eval_runtime": 0.6022, + "eval_samples_per_second": 282.312, + "eval_steps_per_second": 4.982, + "step": 3648 + }, + { + "epoch": 39.0, + "grad_norm": 0.8261823058128357, + "learning_rate": 3.05e-05, + "loss": 0.0484, + "step": 3744 + }, + { + "epoch": 39.0, + "eval_LOCATION_f1": 0.8899521531100479, + "eval_LOCATION_number": 103, + "eval_LOCATION_precision": 0.8773584905660378, + "eval_LOCATION_recall": 0.9029126213592233, + "eval_ORGANIZATION_f1": 0.8825214899713466, + "eval_ORGANIZATION_number": 171, + "eval_ORGANIZATION_precision": 0.8651685393258427, + "eval_ORGANIZATION_recall": 0.9005847953216374, + "eval_PERSON_f1": 0.9694656488549618, + "eval_PERSON_number": 131, + "eval_PERSON_precision": 0.9694656488549618, + "eval_PERSON_recall": 0.9694656488549618, + "eval_loss": 0.045127272605895996, + "eval_overall_accuracy": 0.9845346589339962, + "eval_overall_f1": 0.9121951219512195, + "eval_overall_precision": 0.9012048192771084, + "eval_overall_recall": 0.9234567901234568, + "eval_runtime": 0.599, + "eval_samples_per_second": 283.814, + "eval_steps_per_second": 5.008, + "step": 3744 + }, + { + "epoch": 40.0, + "grad_norm": 1.6085362434387207, + "learning_rate": 3e-05, + "loss": 0.0498, + "step": 3840 + }, + { + "epoch": 40.0, + "eval_LOCATION_f1": 0.912621359223301, + "eval_LOCATION_number": 103, + "eval_LOCATION_precision": 0.912621359223301, + "eval_LOCATION_recall": 0.912621359223301, + "eval_ORGANIZATION_f1": 0.881159420289855, + "eval_ORGANIZATION_number": 171, + "eval_ORGANIZATION_precision": 0.8735632183908046, + "eval_ORGANIZATION_recall": 0.8888888888888888, + "eval_PERSON_f1": 0.9657794676806084, + "eval_PERSON_number": 131, + "eval_PERSON_precision": 0.9621212121212122, + "eval_PERSON_recall": 0.9694656488549618, + "eval_loss": 0.04475143179297447, + "eval_overall_accuracy": 0.9850869925434963, + "eval_overall_f1": 0.9164619164619164, + "eval_overall_precision": 0.9119804400977995, + "eval_overall_recall": 0.9209876543209876, + "eval_runtime": 0.6018, + "eval_samples_per_second": 282.508, + "eval_steps_per_second": 4.985, + "step": 3840 + }, + { + "epoch": 41.0, + "grad_norm": 0.6581994295120239, + "learning_rate": 2.95e-05, + "loss": 0.047, + "step": 3936 + }, + { + "epoch": 41.0, + "eval_LOCATION_f1": 0.8899521531100479, + "eval_LOCATION_number": 103, + "eval_LOCATION_precision": 0.8773584905660378, + "eval_LOCATION_recall": 0.9029126213592233, + "eval_ORGANIZATION_f1": 0.8908045977011494, + "eval_ORGANIZATION_number": 171, + "eval_ORGANIZATION_precision": 0.8757062146892656, + "eval_ORGANIZATION_recall": 0.9064327485380117, + "eval_PERSON_f1": 0.9694656488549618, + "eval_PERSON_number": 131, + "eval_PERSON_precision": 0.9694656488549618, + "eval_PERSON_recall": 0.9694656488549618, + "eval_loss": 0.04271620139479637, + "eval_overall_accuracy": 0.9853631593482464, + "eval_overall_f1": 0.9157509157509157, + "eval_overall_precision": 0.9057971014492754, + "eval_overall_recall": 0.9259259259259259, + "eval_runtime": 0.5986, + "eval_samples_per_second": 284.002, + "eval_steps_per_second": 5.012, + "step": 3936 + }, + { + "epoch": 42.0, + "grad_norm": 0.7113787531852722, + "learning_rate": 2.9e-05, + "loss": 0.0465, + "step": 4032 + }, + { + "epoch": 42.0, + "eval_LOCATION_f1": 0.8985507246376813, + "eval_LOCATION_number": 103, + "eval_LOCATION_precision": 0.8942307692307693, + "eval_LOCATION_recall": 0.9029126213592233, + "eval_ORGANIZATION_f1": 0.8914285714285715, + "eval_ORGANIZATION_number": 171, + "eval_ORGANIZATION_precision": 0.8715083798882681, + "eval_ORGANIZATION_recall": 0.9122807017543859, + "eval_PERSON_f1": 0.9694656488549618, + "eval_PERSON_number": 131, + "eval_PERSON_precision": 0.9694656488549618, + "eval_PERSON_recall": 0.9694656488549618, + "eval_loss": 0.04163723438978195, + "eval_overall_accuracy": 0.9859154929577465, + "eval_overall_f1": 0.9181929181929182, + "eval_overall_precision": 0.9082125603864735, + "eval_overall_recall": 0.928395061728395, + "eval_runtime": 0.6001, + "eval_samples_per_second": 283.294, + "eval_steps_per_second": 4.999, + "step": 4032 + }, + { + "epoch": 43.0, + "grad_norm": 1.796217918395996, + "learning_rate": 2.8499999999999998e-05, + "loss": 0.0443, + "step": 4128 + }, + { + "epoch": 43.0, + "eval_LOCATION_f1": 0.8952380952380954, + "eval_LOCATION_number": 103, + "eval_LOCATION_precision": 0.8785046728971962, + "eval_LOCATION_recall": 0.912621359223301, + "eval_ORGANIZATION_f1": 0.8843930635838151, + "eval_ORGANIZATION_number": 171, + "eval_ORGANIZATION_precision": 0.8742857142857143, + "eval_ORGANIZATION_recall": 0.8947368421052632, + "eval_PERSON_f1": 0.9694656488549618, + "eval_PERSON_number": 131, + "eval_PERSON_precision": 0.9694656488549618, + "eval_PERSON_recall": 0.9694656488549618, + "eval_loss": 0.04234285280108452, + "eval_overall_accuracy": 0.9853631593482464, + "eval_overall_f1": 0.9144254278728606, + "eval_overall_precision": 0.9055690072639225, + "eval_overall_recall": 0.9234567901234568, + "eval_runtime": 0.5917, + "eval_samples_per_second": 287.289, + "eval_steps_per_second": 5.07, + "step": 4128 + }, + { + "epoch": 44.0, + "grad_norm": 0.6815195083618164, + "learning_rate": 2.8000000000000003e-05, + "loss": 0.0428, + "step": 4224 + }, + { + "epoch": 44.0, + "eval_LOCATION_f1": 0.8867924528301887, + "eval_LOCATION_number": 103, + "eval_LOCATION_precision": 0.8623853211009175, + "eval_LOCATION_recall": 0.912621359223301, + "eval_ORGANIZATION_f1": 0.8856304985337242, + "eval_ORGANIZATION_number": 171, + "eval_ORGANIZATION_precision": 0.888235294117647, + "eval_ORGANIZATION_recall": 0.8830409356725146, + "eval_PERSON_f1": 0.9694656488549618, + "eval_PERSON_number": 131, + "eval_PERSON_precision": 0.9694656488549618, + "eval_PERSON_recall": 0.9694656488549618, + "eval_loss": 0.04329132288694382, + "eval_overall_accuracy": 0.9853631593482464, + "eval_overall_f1": 0.9128834355828221, + "eval_overall_precision": 0.9073170731707317, + "eval_overall_recall": 0.9185185185185185, + "eval_runtime": 0.6015, + "eval_samples_per_second": 282.633, + "eval_steps_per_second": 4.988, + "step": 4224 + }, + { + "epoch": 45.0, + "grad_norm": 0.8750056028366089, + "learning_rate": 2.7500000000000004e-05, + "loss": 0.0437, + "step": 4320 + }, + { + "epoch": 45.0, + "eval_LOCATION_f1": 0.8744186046511627, + "eval_LOCATION_number": 103, + "eval_LOCATION_precision": 0.8392857142857143, + "eval_LOCATION_recall": 0.912621359223301, + "eval_ORGANIZATION_f1": 0.8823529411764705, + "eval_ORGANIZATION_number": 171, + "eval_ORGANIZATION_precision": 0.8875739644970414, + "eval_ORGANIZATION_recall": 0.8771929824561403, + "eval_PERSON_f1": 0.9694656488549618, + "eval_PERSON_number": 131, + "eval_PERSON_precision": 0.9694656488549618, + "eval_PERSON_recall": 0.9694656488549618, + "eval_loss": 0.04397225007414818, + "eval_overall_accuracy": 0.9848108257387462, + "eval_overall_f1": 0.908200734394125, + "eval_overall_precision": 0.9004854368932039, + "eval_overall_recall": 0.9160493827160494, + "eval_runtime": 0.5973, + "eval_samples_per_second": 284.623, + "eval_steps_per_second": 5.023, + "step": 4320 + }, + { + "epoch": 46.0, + "grad_norm": 0.48337626457214355, + "learning_rate": 2.7000000000000002e-05, + "loss": 0.0447, + "step": 4416 + }, + { + "epoch": 46.0, + "eval_LOCATION_f1": 0.9, + "eval_LOCATION_number": 103, + "eval_LOCATION_precision": 0.8461538461538461, + "eval_LOCATION_recall": 0.9611650485436893, + "eval_ORGANIZATION_f1": 0.8821752265861027, + "eval_ORGANIZATION_number": 171, + "eval_ORGANIZATION_precision": 0.9125, + "eval_ORGANIZATION_recall": 0.8538011695906432, + "eval_PERSON_f1": 0.9694656488549618, + "eval_PERSON_number": 131, + "eval_PERSON_precision": 0.9694656488549618, + "eval_PERSON_recall": 0.9694656488549618, + "eval_loss": 0.04790230840444565, + "eval_overall_accuracy": 0.983982325324496, + "eval_overall_f1": 0.915129151291513, + "eval_overall_precision": 0.9117647058823529, + "eval_overall_recall": 0.9185185185185185, + "eval_runtime": 0.6839, + "eval_samples_per_second": 248.588, + "eval_steps_per_second": 4.387, + "step": 4416 + }, + { + "epoch": 47.0, + "grad_norm": 1.842751383781433, + "learning_rate": 2.6500000000000004e-05, + "loss": 0.0421, + "step": 4512 + }, + { + "epoch": 47.0, + "eval_LOCATION_f1": 0.9004739336492891, + "eval_LOCATION_number": 103, + "eval_LOCATION_precision": 0.8796296296296297, + "eval_LOCATION_recall": 0.9223300970873787, + "eval_ORGANIZATION_f1": 0.8850574712643677, + "eval_ORGANIZATION_number": 171, + "eval_ORGANIZATION_precision": 0.8700564971751412, + "eval_ORGANIZATION_recall": 0.9005847953216374, + "eval_PERSON_f1": 0.9694656488549618, + "eval_PERSON_number": 131, + "eval_PERSON_precision": 0.9694656488549618, + "eval_PERSON_recall": 0.9694656488549618, + "eval_loss": 0.043716877698898315, + "eval_overall_accuracy": 0.9850869925434963, + "eval_overall_f1": 0.9159561510353228, + "eval_overall_precision": 0.9038461538461539, + "eval_overall_recall": 0.928395061728395, + "eval_runtime": 0.6022, + "eval_samples_per_second": 282.284, + "eval_steps_per_second": 4.981, + "step": 4512 + }, + { + "epoch": 48.0, + "grad_norm": 0.8989099264144897, + "learning_rate": 2.6000000000000002e-05, + "loss": 0.0403, + "step": 4608 + }, + { + "epoch": 48.0, + "eval_LOCATION_f1": 0.8990825688073394, + "eval_LOCATION_number": 103, + "eval_LOCATION_precision": 0.8521739130434782, + "eval_LOCATION_recall": 0.9514563106796117, + "eval_ORGANIZATION_f1": 0.8922155688622756, + "eval_ORGANIZATION_number": 171, + "eval_ORGANIZATION_precision": 0.9141104294478528, + "eval_ORGANIZATION_recall": 0.8713450292397661, + "eval_PERSON_f1": 0.9694656488549618, + "eval_PERSON_number": 131, + "eval_PERSON_precision": 0.9694656488549618, + "eval_PERSON_recall": 0.9694656488549618, + "eval_loss": 0.044955264776945114, + "eval_overall_accuracy": 0.9856393261529964, + "eval_overall_f1": 0.918918918918919, + "eval_overall_precision": 0.9144254278728606, + "eval_overall_recall": 0.9234567901234568, + "eval_runtime": 0.5976, + "eval_samples_per_second": 284.471, + "eval_steps_per_second": 5.02, + "step": 4608 + }, + { + "epoch": 49.0, + "grad_norm": 0.10502011328935623, + "learning_rate": 2.5500000000000003e-05, + "loss": 0.0423, + "step": 4704 + }, + { + "epoch": 49.0, + "eval_LOCATION_f1": 0.8949771689497718, + "eval_LOCATION_number": 103, + "eval_LOCATION_precision": 0.8448275862068966, + "eval_LOCATION_recall": 0.9514563106796117, + "eval_ORGANIZATION_f1": 0.8828828828828829, + "eval_ORGANIZATION_number": 171, + "eval_ORGANIZATION_precision": 0.9074074074074074, + "eval_ORGANIZATION_recall": 0.8596491228070176, + "eval_PERSON_f1": 0.9694656488549618, + "eval_PERSON_number": 131, + "eval_PERSON_precision": 0.9694656488549618, + "eval_PERSON_recall": 0.9694656488549618, + "eval_loss": 0.047507476061582565, + "eval_overall_accuracy": 0.9850869925434963, + "eval_overall_f1": 0.914004914004914, + "eval_overall_precision": 0.9095354523227384, + "eval_overall_recall": 0.9185185185185185, + "eval_runtime": 0.5985, + "eval_samples_per_second": 284.048, + "eval_steps_per_second": 5.013, + "step": 4704 + }, + { + "epoch": 50.0, + "grad_norm": 0.5192816853523254, + "learning_rate": 2.5e-05, + "loss": 0.039, + "step": 4800 + }, + { + "epoch": 50.0, + "eval_LOCATION_f1": 0.8899082568807339, + "eval_LOCATION_number": 103, + "eval_LOCATION_precision": 0.8434782608695652, + "eval_LOCATION_recall": 0.941747572815534, + "eval_ORGANIZATION_f1": 0.8882175226586103, + "eval_ORGANIZATION_number": 171, + "eval_ORGANIZATION_precision": 0.91875, + "eval_ORGANIZATION_recall": 0.8596491228070176, + "eval_PERSON_f1": 0.9694656488549618, + "eval_PERSON_number": 131, + "eval_PERSON_precision": 0.9694656488549618, + "eval_PERSON_recall": 0.9694656488549618, + "eval_loss": 0.050250228494405746, + "eval_overall_accuracy": 0.983706158519746, + "eval_overall_f1": 0.9149198520345252, + "eval_overall_precision": 0.9137931034482759, + "eval_overall_recall": 0.9160493827160494, + "eval_runtime": 0.6017, + "eval_samples_per_second": 282.548, + "eval_steps_per_second": 4.986, + "step": 4800 + }, + { + "epoch": 51.0, + "grad_norm": 0.5883473753929138, + "learning_rate": 2.45e-05, + "loss": 0.0395, + "step": 4896 + }, + { + "epoch": 51.0, + "eval_LOCATION_f1": 0.8949771689497718, + "eval_LOCATION_number": 103, + "eval_LOCATION_precision": 0.8448275862068966, + "eval_LOCATION_recall": 0.9514563106796117, + "eval_ORGANIZATION_f1": 0.8855421686746987, + "eval_ORGANIZATION_number": 171, + "eval_ORGANIZATION_precision": 0.9130434782608695, + "eval_ORGANIZATION_recall": 0.8596491228070176, + "eval_PERSON_f1": 0.9694656488549618, + "eval_PERSON_number": 131, + "eval_PERSON_precision": 0.9694656488549618, + "eval_PERSON_recall": 0.9694656488549618, + "eval_loss": 0.051189180463552475, + "eval_overall_accuracy": 0.983982325324496, + "eval_overall_f1": 0.915129151291513, + "eval_overall_precision": 0.9117647058823529, + "eval_overall_recall": 0.9185185185185185, + "eval_runtime": 0.6003, + "eval_samples_per_second": 283.207, + "eval_steps_per_second": 4.998, + "step": 4896 + }, + { + "epoch": 52.0, + "grad_norm": 0.886803925037384, + "learning_rate": 2.4e-05, + "loss": 0.0405, + "step": 4992 + }, + { + "epoch": 52.0, + "eval_LOCATION_f1": 0.8878504672897196, + "eval_LOCATION_number": 103, + "eval_LOCATION_precision": 0.8558558558558559, + "eval_LOCATION_recall": 0.9223300970873787, + "eval_ORGANIZATION_f1": 0.8713450292397662, + "eval_ORGANIZATION_number": 171, + "eval_ORGANIZATION_precision": 0.8713450292397661, + "eval_ORGANIZATION_recall": 0.8713450292397661, + "eval_PERSON_f1": 0.9694656488549618, + "eval_PERSON_number": 131, + "eval_PERSON_precision": 0.9694656488549618, + "eval_PERSON_recall": 0.9694656488549618, + "eval_loss": 0.04509918391704559, + "eval_overall_accuracy": 0.9842584921292461, + "eval_overall_f1": 0.9070904645476773, + "eval_overall_precision": 0.8983050847457628, + "eval_overall_recall": 0.9160493827160494, + "eval_runtime": 0.6017, + "eval_samples_per_second": 282.539, + "eval_steps_per_second": 4.986, + "step": 4992 + }, + { + "epoch": 53.0, + "grad_norm": 0.3778795003890991, + "learning_rate": 2.35e-05, + "loss": 0.0383, + "step": 5088 + }, + { + "epoch": 53.0, + "eval_LOCATION_f1": 0.8899082568807339, + "eval_LOCATION_number": 103, + "eval_LOCATION_precision": 0.8434782608695652, + "eval_LOCATION_recall": 0.941747572815534, + "eval_ORGANIZATION_f1": 0.876876876876877, + "eval_ORGANIZATION_number": 171, + "eval_ORGANIZATION_precision": 0.9012345679012346, + "eval_ORGANIZATION_recall": 0.8538011695906432, + "eval_PERSON_f1": 0.9694656488549618, + "eval_PERSON_number": 131, + "eval_PERSON_precision": 0.9694656488549618, + "eval_PERSON_recall": 0.9694656488549618, + "eval_loss": 0.047354765236377716, + "eval_overall_accuracy": 0.9834299917149959, + "eval_overall_f1": 0.9102091020910209, + "eval_overall_precision": 0.9068627450980392, + "eval_overall_recall": 0.9135802469135802, + "eval_runtime": 0.6072, + "eval_samples_per_second": 279.966, + "eval_steps_per_second": 4.941, + "step": 5088 + }, + { + "epoch": 54.0, + "grad_norm": 0.46652740240097046, + "learning_rate": 2.3000000000000003e-05, + "loss": 0.0355, + "step": 5184 + }, + { + "epoch": 54.0, + "eval_LOCATION_f1": 0.9142857142857143, + "eval_LOCATION_number": 103, + "eval_LOCATION_precision": 0.897196261682243, + "eval_LOCATION_recall": 0.9320388349514563, + "eval_ORGANIZATION_f1": 0.9037900874635568, + "eval_ORGANIZATION_number": 171, + "eval_ORGANIZATION_precision": 0.9011627906976745, + "eval_ORGANIZATION_recall": 0.9064327485380117, + "eval_PERSON_f1": 0.9694656488549618, + "eval_PERSON_number": 131, + "eval_PERSON_precision": 0.9694656488549618, + "eval_PERSON_recall": 0.9694656488549618, + "eval_loss": 0.045011453330516815, + "eval_overall_accuracy": 0.9864678265672466, + "eval_overall_f1": 0.9276073619631902, + "eval_overall_precision": 0.9219512195121952, + "eval_overall_recall": 0.9333333333333333, + "eval_runtime": 0.608, + "eval_samples_per_second": 279.59, + "eval_steps_per_second": 4.934, + "step": 5184 + }, + { + "epoch": 55.0, + "grad_norm": 1.0857142210006714, + "learning_rate": 2.25e-05, + "loss": 0.0404, + "step": 5280 + }, + { + "epoch": 55.0, + "eval_LOCATION_f1": 0.9014084507042254, + "eval_LOCATION_number": 103, + "eval_LOCATION_precision": 0.8727272727272727, + "eval_LOCATION_recall": 0.9320388349514563, + "eval_ORGANIZATION_f1": 0.8682634730538922, + "eval_ORGANIZATION_number": 171, + "eval_ORGANIZATION_precision": 0.8895705521472392, + "eval_ORGANIZATION_recall": 0.847953216374269, + "eval_PERSON_f1": 0.9694656488549618, + "eval_PERSON_number": 131, + "eval_PERSON_precision": 0.9694656488549618, + "eval_PERSON_recall": 0.9694656488549618, + "eval_loss": 0.04946543648838997, + "eval_overall_accuracy": 0.9828776581054958, + "eval_overall_f1": 0.9097651421508034, + "eval_overall_precision": 0.9108910891089109, + "eval_overall_recall": 0.908641975308642, + "eval_runtime": 0.601, + "eval_samples_per_second": 282.885, + "eval_steps_per_second": 4.992, + "step": 5280 + }, + { + "epoch": 56.0, + "grad_norm": 0.6342605948448181, + "learning_rate": 2.2000000000000003e-05, + "loss": 0.0367, + "step": 5376 + }, + { + "epoch": 56.0, + "eval_LOCATION_f1": 0.8930232558139534, + "eval_LOCATION_number": 103, + "eval_LOCATION_precision": 0.8571428571428571, + "eval_LOCATION_recall": 0.9320388349514563, + "eval_ORGANIZATION_f1": 0.8875739644970414, + "eval_ORGANIZATION_number": 171, + "eval_ORGANIZATION_precision": 0.8982035928143712, + "eval_ORGANIZATION_recall": 0.8771929824561403, + "eval_PERSON_f1": 0.9694656488549618, + "eval_PERSON_number": 131, + "eval_PERSON_precision": 0.9694656488549618, + "eval_PERSON_recall": 0.9694656488549618, + "eval_loss": 0.04730325564742088, + "eval_overall_accuracy": 0.9853631593482464, + "eval_overall_f1": 0.9153374233128834, + "eval_overall_precision": 0.9097560975609756, + "eval_overall_recall": 0.9209876543209876, + "eval_runtime": 0.6002, + "eval_samples_per_second": 283.241, + "eval_steps_per_second": 4.998, + "step": 5376 + }, + { + "epoch": 57.0, + "grad_norm": 1.5249769687652588, + "learning_rate": 2.15e-05, + "loss": 0.0383, + "step": 5472 + }, + { + "epoch": 57.0, + "eval_LOCATION_f1": 0.888888888888889, + "eval_LOCATION_number": 103, + "eval_LOCATION_precision": 0.8495575221238938, + "eval_LOCATION_recall": 0.9320388349514563, + "eval_ORGANIZATION_f1": 0.8816568047337278, + "eval_ORGANIZATION_number": 171, + "eval_ORGANIZATION_precision": 0.8922155688622755, + "eval_ORGANIZATION_recall": 0.8713450292397661, + "eval_PERSON_f1": 0.9694656488549618, + "eval_PERSON_number": 131, + "eval_PERSON_precision": 0.9694656488549618, + "eval_PERSON_recall": 0.9694656488549618, + "eval_loss": 0.04862285032868385, + "eval_overall_accuracy": 0.9848108257387462, + "eval_overall_f1": 0.9117647058823529, + "eval_overall_precision": 0.9051094890510949, + "eval_overall_recall": 0.9185185185185185, + "eval_runtime": 0.6085, + "eval_samples_per_second": 279.376, + "eval_steps_per_second": 4.93, + "step": 5472 + }, + { + "epoch": 58.0, + "grad_norm": 0.3260681629180908, + "learning_rate": 2.1e-05, + "loss": 0.0353, + "step": 5568 + }, + { + "epoch": 58.0, + "eval_LOCATION_f1": 0.8930232558139534, + "eval_LOCATION_number": 103, + "eval_LOCATION_precision": 0.8571428571428571, + "eval_LOCATION_recall": 0.9320388349514563, + "eval_ORGANIZATION_f1": 0.8869047619047619, + "eval_ORGANIZATION_number": 171, + "eval_ORGANIZATION_precision": 0.9030303030303031, + "eval_ORGANIZATION_recall": 0.8713450292397661, + "eval_PERSON_f1": 0.9694656488549618, + "eval_PERSON_number": 131, + "eval_PERSON_precision": 0.9694656488549618, + "eval_PERSON_recall": 0.9694656488549618, + "eval_loss": 0.04823388531804085, + "eval_overall_accuracy": 0.9842584921292461, + "eval_overall_f1": 0.915129151291513, + "eval_overall_precision": 0.9117647058823529, + "eval_overall_recall": 0.9185185185185185, + "eval_runtime": 0.6379, + "eval_samples_per_second": 266.516, + "eval_steps_per_second": 4.703, + "step": 5568 + }, + { + "epoch": 59.0, + "grad_norm": 1.2659943103790283, + "learning_rate": 2.05e-05, + "loss": 0.0362, + "step": 5664 + }, + { + "epoch": 59.0, + "eval_LOCATION_f1": 0.8971962616822431, + "eval_LOCATION_number": 103, + "eval_LOCATION_precision": 0.8648648648648649, + "eval_LOCATION_recall": 0.9320388349514563, + "eval_ORGANIZATION_f1": 0.875739644970414, + "eval_ORGANIZATION_number": 171, + "eval_ORGANIZATION_precision": 0.8862275449101796, + "eval_ORGANIZATION_recall": 0.8654970760233918, + "eval_PERSON_f1": 0.9694656488549618, + "eval_PERSON_number": 131, + "eval_PERSON_precision": 0.9694656488549618, + "eval_PERSON_recall": 0.9694656488549618, + "eval_loss": 0.046954311430454254, + "eval_overall_accuracy": 0.9845346589339962, + "eval_overall_f1": 0.9115479115479117, + "eval_overall_precision": 0.9070904645476773, + "eval_overall_recall": 0.9160493827160494, + "eval_runtime": 0.6026, + "eval_samples_per_second": 282.118, + "eval_steps_per_second": 4.979, + "step": 5664 + }, + { + "epoch": 60.0, + "grad_norm": 0.4821118116378784, + "learning_rate": 2e-05, + "loss": 0.0351, + "step": 5760 + }, + { + "epoch": 60.0, + "eval_LOCATION_f1": 0.9014084507042254, + "eval_LOCATION_number": 103, + "eval_LOCATION_precision": 0.8727272727272727, + "eval_LOCATION_recall": 0.9320388349514563, + "eval_ORGANIZATION_f1": 0.8724035608308606, + "eval_ORGANIZATION_number": 171, + "eval_ORGANIZATION_precision": 0.8855421686746988, + "eval_ORGANIZATION_recall": 0.8596491228070176, + "eval_PERSON_f1": 0.9694656488549618, + "eval_PERSON_number": 131, + "eval_PERSON_precision": 0.9694656488549618, + "eval_PERSON_recall": 0.9694656488549618, + "eval_loss": 0.048675838857889175, + "eval_overall_accuracy": 0.9842584921292461, + "eval_overall_f1": 0.9113300492610836, + "eval_overall_precision": 0.9090909090909091, + "eval_overall_recall": 0.9135802469135802, + "eval_runtime": 0.606, + "eval_samples_per_second": 280.545, + "eval_steps_per_second": 4.951, + "step": 5760 + }, + { + "epoch": 61.0, + "grad_norm": 1.131179690361023, + "learning_rate": 1.9500000000000003e-05, + "loss": 0.0399, + "step": 5856 + }, + { + "epoch": 61.0, + "eval_LOCATION_f1": 0.909952606635071, + "eval_LOCATION_number": 103, + "eval_LOCATION_precision": 0.8888888888888888, + "eval_LOCATION_recall": 0.9320388349514563, + "eval_ORGANIZATION_f1": 0.8790560471976402, + "eval_ORGANIZATION_number": 171, + "eval_ORGANIZATION_precision": 0.8869047619047619, + "eval_ORGANIZATION_recall": 0.8713450292397661, + "eval_PERSON_f1": 0.9694656488549618, + "eval_PERSON_number": 131, + "eval_PERSON_precision": 0.9694656488549618, + "eval_PERSON_recall": 0.9694656488549618, + "eval_loss": 0.04870187118649483, + "eval_overall_accuracy": 0.9842584921292461, + "eval_overall_f1": 0.916256157635468, + "eval_overall_precision": 0.914004914004914, + "eval_overall_recall": 0.9185185185185185, + "eval_runtime": 0.6128, + "eval_samples_per_second": 277.402, + "eval_steps_per_second": 4.895, + "step": 5856 + }, + { + "epoch": 62.0, + "grad_norm": 1.2458444833755493, + "learning_rate": 1.9e-05, + "loss": 0.0364, + "step": 5952 + }, + { + "epoch": 62.0, + "eval_LOCATION_f1": 0.9074074074074074, + "eval_LOCATION_number": 103, + "eval_LOCATION_precision": 0.8672566371681416, + "eval_LOCATION_recall": 0.9514563106796117, + "eval_ORGANIZATION_f1": 0.8821752265861027, + "eval_ORGANIZATION_number": 171, + "eval_ORGANIZATION_precision": 0.9125, + "eval_ORGANIZATION_recall": 0.8538011695906432, + "eval_PERSON_f1": 0.9694656488549618, + "eval_PERSON_number": 131, + "eval_PERSON_precision": 0.9694656488549618, + "eval_PERSON_recall": 0.9694656488549618, + "eval_loss": 0.050012677907943726, + "eval_overall_accuracy": 0.9834299917149959, + "eval_overall_f1": 0.9171817058096416, + "eval_overall_precision": 0.9183168316831684, + "eval_overall_recall": 0.9160493827160494, + "eval_runtime": 0.6008, + "eval_samples_per_second": 282.959, + "eval_steps_per_second": 4.993, + "step": 5952 + }, + { + "epoch": 63.0, + "grad_norm": 2.4960782527923584, + "learning_rate": 1.85e-05, + "loss": 0.0342, + "step": 6048 + }, + { + "epoch": 63.0, + "eval_LOCATION_f1": 0.8971962616822431, + "eval_LOCATION_number": 103, + "eval_LOCATION_precision": 0.8648648648648649, + "eval_LOCATION_recall": 0.9320388349514563, + "eval_ORGANIZATION_f1": 0.8705882352941177, + "eval_ORGANIZATION_number": 171, + "eval_ORGANIZATION_precision": 0.8757396449704142, + "eval_ORGANIZATION_recall": 0.8654970760233918, + "eval_PERSON_f1": 0.9694656488549618, + "eval_PERSON_number": 131, + "eval_PERSON_precision": 0.9694656488549618, + "eval_PERSON_recall": 0.9694656488549618, + "eval_loss": 0.04661116749048233, + "eval_overall_accuracy": 0.9842584921292461, + "eval_overall_f1": 0.9093137254901962, + "eval_overall_precision": 0.902676399026764, + "eval_overall_recall": 0.9160493827160494, + "eval_runtime": 0.6666, + "eval_samples_per_second": 255.011, + "eval_steps_per_second": 4.5, + "step": 6048 + }, + { + "epoch": 64.0, + "grad_norm": 1.4916831254959106, + "learning_rate": 1.8e-05, + "loss": 0.0347, + "step": 6144 + }, + { + "epoch": 64.0, + "eval_LOCATION_f1": 0.9056603773584906, + "eval_LOCATION_number": 103, + "eval_LOCATION_precision": 0.8807339449541285, + "eval_LOCATION_recall": 0.9320388349514563, + "eval_ORGANIZATION_f1": 0.8783382789317508, + "eval_ORGANIZATION_number": 171, + "eval_ORGANIZATION_precision": 0.891566265060241, + "eval_ORGANIZATION_recall": 0.8654970760233918, + "eval_PERSON_f1": 0.9694656488549618, + "eval_PERSON_number": 131, + "eval_PERSON_precision": 0.9694656488549618, + "eval_PERSON_recall": 0.9694656488549618, + "eval_loss": 0.04867375269532204, + "eval_overall_accuracy": 0.983982325324496, + "eval_overall_f1": 0.9149198520345252, + "eval_overall_precision": 0.9137931034482759, + "eval_overall_recall": 0.9160493827160494, + "eval_runtime": 0.5955, + "eval_samples_per_second": 285.467, + "eval_steps_per_second": 5.038, + "step": 6144 + }, + { + "epoch": 65.0, + "grad_norm": 0.5264370441436768, + "learning_rate": 1.75e-05, + "loss": 0.0357, + "step": 6240 + }, + { + "epoch": 65.0, + "eval_LOCATION_f1": 0.903225806451613, + "eval_LOCATION_number": 103, + "eval_LOCATION_precision": 0.8596491228070176, + "eval_LOCATION_recall": 0.9514563106796117, + "eval_ORGANIZATION_f1": 0.8835820895522388, + "eval_ORGANIZATION_number": 171, + "eval_ORGANIZATION_precision": 0.9024390243902439, + "eval_ORGANIZATION_recall": 0.8654970760233918, + "eval_PERSON_f1": 0.9694656488549618, + "eval_PERSON_number": 131, + "eval_PERSON_precision": 0.9694656488549618, + "eval_PERSON_recall": 0.9694656488549618, + "eval_loss": 0.04691191017627716, + "eval_overall_accuracy": 0.9856393261529964, + "eval_overall_f1": 0.9164619164619164, + "eval_overall_precision": 0.9119804400977995, + "eval_overall_recall": 0.9209876543209876, + "eval_runtime": 0.6062, + "eval_samples_per_second": 280.449, + "eval_steps_per_second": 4.949, + "step": 6240 + }, + { + "epoch": 66.0, + "grad_norm": 0.6101003289222717, + "learning_rate": 1.7000000000000003e-05, + "loss": 0.035, + "step": 6336 + }, + { + "epoch": 66.0, + "eval_LOCATION_f1": 0.9023255813953488, + "eval_LOCATION_number": 103, + "eval_LOCATION_precision": 0.8660714285714286, + "eval_LOCATION_recall": 0.941747572815534, + "eval_ORGANIZATION_f1": 0.8787878787878787, + "eval_ORGANIZATION_number": 171, + "eval_ORGANIZATION_precision": 0.9119496855345912, + "eval_ORGANIZATION_recall": 0.847953216374269, + "eval_PERSON_f1": 0.9694656488549618, + "eval_PERSON_number": 131, + "eval_PERSON_precision": 0.9694656488549618, + "eval_PERSON_recall": 0.9694656488549618, + "eval_loss": 0.051512736827135086, + "eval_overall_accuracy": 0.9831538249102458, + "eval_overall_f1": 0.9144981412639405, + "eval_overall_precision": 0.917910447761194, + "eval_overall_recall": 0.9111111111111111, + "eval_runtime": 0.596, + "eval_samples_per_second": 285.237, + "eval_steps_per_second": 5.034, + "step": 6336 + }, + { + "epoch": 67.0, + "grad_norm": 0.2658337354660034, + "learning_rate": 1.65e-05, + "loss": 0.0341, + "step": 6432 + }, + { + "epoch": 67.0, + "eval_LOCATION_f1": 0.888888888888889, + "eval_LOCATION_number": 103, + "eval_LOCATION_precision": 0.8495575221238938, + "eval_LOCATION_recall": 0.9320388349514563, + "eval_ORGANIZATION_f1": 0.875, + "eval_ORGANIZATION_number": 171, + "eval_ORGANIZATION_precision": 0.8909090909090909, + "eval_ORGANIZATION_recall": 0.8596491228070176, + "eval_PERSON_f1": 0.9694656488549618, + "eval_PERSON_number": 131, + "eval_PERSON_precision": 0.9694656488549618, + "eval_PERSON_recall": 0.9694656488549618, + "eval_loss": 0.049565933644771576, + "eval_overall_accuracy": 0.9828776581054958, + "eval_overall_f1": 0.9090909090909091, + "eval_overall_precision": 0.9046454767726161, + "eval_overall_recall": 0.9135802469135802, + "eval_runtime": 0.6031, + "eval_samples_per_second": 281.883, + "eval_steps_per_second": 4.974, + "step": 6432 + }, + { + "epoch": 68.0, + "grad_norm": 0.8890864849090576, + "learning_rate": 1.6000000000000003e-05, + "loss": 0.0336, + "step": 6528 + }, + { + "epoch": 68.0, + "eval_LOCATION_f1": 0.8878504672897196, + "eval_LOCATION_number": 103, + "eval_LOCATION_precision": 0.8558558558558559, + "eval_LOCATION_recall": 0.9223300970873787, + "eval_ORGANIZATION_f1": 0.8739002932551321, + "eval_ORGANIZATION_number": 171, + "eval_ORGANIZATION_precision": 0.8764705882352941, + "eval_ORGANIZATION_recall": 0.8713450292397661, + "eval_PERSON_f1": 0.9694656488549618, + "eval_PERSON_number": 131, + "eval_PERSON_precision": 0.9694656488549618, + "eval_PERSON_recall": 0.9694656488549618, + "eval_loss": 0.04961269721388817, + "eval_overall_accuracy": 0.983982325324496, + "eval_overall_f1": 0.908200734394125, + "eval_overall_precision": 0.9004854368932039, + "eval_overall_recall": 0.9160493827160494, + "eval_runtime": 0.6041, + "eval_samples_per_second": 281.409, + "eval_steps_per_second": 4.966, + "step": 6528 + }, + { + "epoch": 69.0, + "grad_norm": 0.6407134532928467, + "learning_rate": 1.55e-05, + "loss": 0.033, + "step": 6624 + }, + { + "epoch": 69.0, + "eval_LOCATION_f1": 0.9023255813953488, + "eval_LOCATION_number": 103, + "eval_LOCATION_precision": 0.8660714285714286, + "eval_LOCATION_recall": 0.941747572815534, + "eval_ORGANIZATION_f1": 0.8716417910447761, + "eval_ORGANIZATION_number": 171, + "eval_ORGANIZATION_precision": 0.8902439024390244, + "eval_ORGANIZATION_recall": 0.8538011695906432, + "eval_PERSON_f1": 0.9694656488549618, + "eval_PERSON_number": 131, + "eval_PERSON_precision": 0.9694656488549618, + "eval_PERSON_recall": 0.9694656488549618, + "eval_loss": 0.05003252625465393, + "eval_overall_accuracy": 0.983706158519746, + "eval_overall_f1": 0.9113300492610836, + "eval_overall_precision": 0.9090909090909091, + "eval_overall_recall": 0.9135802469135802, + "eval_runtime": 0.6133, + "eval_samples_per_second": 277.194, + "eval_steps_per_second": 4.892, + "step": 6624 + }, + { + "epoch": 70.0, + "grad_norm": 0.38998541235923767, + "learning_rate": 1.5e-05, + "loss": 0.0327, + "step": 6720 + }, + { + "epoch": 70.0, + "eval_LOCATION_f1": 0.8971962616822431, + "eval_LOCATION_number": 103, + "eval_LOCATION_precision": 0.8648648648648649, + "eval_LOCATION_recall": 0.9320388349514563, + "eval_ORGANIZATION_f1": 0.8664688427299703, + "eval_ORGANIZATION_number": 171, + "eval_ORGANIZATION_precision": 0.8795180722891566, + "eval_ORGANIZATION_recall": 0.8538011695906432, + "eval_PERSON_f1": 0.9694656488549618, + "eval_PERSON_number": 131, + "eval_PERSON_precision": 0.9694656488549618, + "eval_PERSON_recall": 0.9694656488549618, + "eval_loss": 0.04880837723612785, + "eval_overall_accuracy": 0.983706158519746, + "eval_overall_f1": 0.9077490774907748, + "eval_overall_precision": 0.9044117647058824, + "eval_overall_recall": 0.9111111111111111, + "eval_runtime": 0.6045, + "eval_samples_per_second": 281.241, + "eval_steps_per_second": 4.963, + "step": 6720 + }, + { + "epoch": 71.0, + "grad_norm": 1.0783237218856812, + "learning_rate": 1.45e-05, + "loss": 0.0331, + "step": 6816 + }, + { + "epoch": 71.0, + "eval_LOCATION_f1": 0.8930232558139534, + "eval_LOCATION_number": 103, + "eval_LOCATION_precision": 0.8571428571428571, + "eval_LOCATION_recall": 0.9320388349514563, + "eval_ORGANIZATION_f1": 0.8816568047337278, + "eval_ORGANIZATION_number": 171, + "eval_ORGANIZATION_precision": 0.8922155688622755, + "eval_ORGANIZATION_recall": 0.8713450292397661, + "eval_PERSON_f1": 0.9694656488549618, + "eval_PERSON_number": 131, + "eval_PERSON_precision": 0.9694656488549618, + "eval_PERSON_recall": 0.9694656488549618, + "eval_loss": 0.04777364805340767, + "eval_overall_accuracy": 0.9850869925434963, + "eval_overall_f1": 0.9128834355828221, + "eval_overall_precision": 0.9073170731707317, + "eval_overall_recall": 0.9185185185185185, + "eval_runtime": 0.6008, + "eval_samples_per_second": 282.978, + "eval_steps_per_second": 4.994, + "step": 6816 + }, + { + "epoch": 72.0, + "grad_norm": 0.44706445932388306, + "learning_rate": 1.4000000000000001e-05, + "loss": 0.0311, + "step": 6912 + }, + { + "epoch": 72.0, + "eval_LOCATION_f1": 0.903225806451613, + "eval_LOCATION_number": 103, + "eval_LOCATION_precision": 0.8596491228070176, + "eval_LOCATION_recall": 0.9514563106796117, + "eval_ORGANIZATION_f1": 0.8868501529051989, + "eval_ORGANIZATION_number": 171, + "eval_ORGANIZATION_precision": 0.9294871794871795, + "eval_ORGANIZATION_recall": 0.847953216374269, + "eval_PERSON_f1": 0.9694656488549618, + "eval_PERSON_number": 131, + "eval_PERSON_precision": 0.9694656488549618, + "eval_PERSON_recall": 0.9694656488549618, + "eval_loss": 0.05447942018508911, + "eval_overall_accuracy": 0.9831538249102458, + "eval_overall_f1": 0.9181141439205954, + "eval_overall_precision": 0.9226932668329177, + "eval_overall_recall": 0.9135802469135802, + "eval_runtime": 0.7507, + "eval_samples_per_second": 226.457, + "eval_steps_per_second": 3.996, + "step": 6912 + }, + { + "epoch": 73.0, + "grad_norm": 1.0067603588104248, + "learning_rate": 1.3500000000000001e-05, + "loss": 0.033, + "step": 7008 + }, + { + "epoch": 73.0, + "eval_LOCATION_f1": 0.9124423963133641, + "eval_LOCATION_number": 103, + "eval_LOCATION_precision": 0.868421052631579, + "eval_LOCATION_recall": 0.9611650485436893, + "eval_ORGANIZATION_f1": 0.8835820895522388, + "eval_ORGANIZATION_number": 171, + "eval_ORGANIZATION_precision": 0.9024390243902439, + "eval_ORGANIZATION_recall": 0.8654970760233918, + "eval_PERSON_f1": 0.9694656488549618, + "eval_PERSON_number": 131, + "eval_PERSON_precision": 0.9694656488549618, + "eval_PERSON_recall": 0.9694656488549618, + "eval_loss": 0.05023306608200073, + "eval_overall_accuracy": 0.9859154929577465, + "eval_overall_f1": 0.918918918918919, + "eval_overall_precision": 0.9144254278728606, + "eval_overall_recall": 0.9234567901234568, + "eval_runtime": 0.6046, + "eval_samples_per_second": 281.177, + "eval_steps_per_second": 4.962, + "step": 7008 + }, + { + "epoch": 74.0, + "grad_norm": 0.11179722845554352, + "learning_rate": 1.3000000000000001e-05, + "loss": 0.0321, + "step": 7104 + }, + { + "epoch": 74.0, + "eval_LOCATION_f1": 0.9074074074074074, + "eval_LOCATION_number": 103, + "eval_LOCATION_precision": 0.8672566371681416, + "eval_LOCATION_recall": 0.9514563106796117, + "eval_ORGANIZATION_f1": 0.8828828828828829, + "eval_ORGANIZATION_number": 171, + "eval_ORGANIZATION_precision": 0.9074074074074074, + "eval_ORGANIZATION_recall": 0.8596491228070176, + "eval_PERSON_f1": 0.9694656488549618, + "eval_PERSON_number": 131, + "eval_PERSON_precision": 0.9694656488549618, + "eval_PERSON_recall": 0.9694656488549618, + "eval_loss": 0.051882192492485046, + "eval_overall_accuracy": 0.983982325324496, + "eval_overall_f1": 0.9173859432799014, + "eval_overall_precision": 0.916256157635468, + "eval_overall_recall": 0.9185185185185185, + "eval_runtime": 0.6045, + "eval_samples_per_second": 281.236, + "eval_steps_per_second": 4.963, + "step": 7104 + }, + { + "epoch": 75.0, + "grad_norm": 0.5369475483894348, + "learning_rate": 1.25e-05, + "loss": 0.032, + "step": 7200 + }, + { + "epoch": 75.0, + "eval_LOCATION_f1": 0.892018779342723, + "eval_LOCATION_number": 103, + "eval_LOCATION_precision": 0.8636363636363636, + "eval_LOCATION_recall": 0.9223300970873787, + "eval_ORGANIZATION_f1": 0.8690476190476191, + "eval_ORGANIZATION_number": 171, + "eval_ORGANIZATION_precision": 0.8848484848484849, + "eval_ORGANIZATION_recall": 0.8538011695906432, + "eval_PERSON_f1": 0.9694656488549618, + "eval_PERSON_number": 131, + "eval_PERSON_precision": 0.9694656488549618, + "eval_PERSON_recall": 0.9694656488549618, + "eval_loss": 0.051246125251054764, + "eval_overall_accuracy": 0.9828776581054958, + "eval_overall_f1": 0.9075215782983971, + "eval_overall_precision": 0.9064039408866995, + "eval_overall_recall": 0.908641975308642, + "eval_runtime": 0.6018, + "eval_samples_per_second": 282.469, + "eval_steps_per_second": 4.985, + "step": 7200 + }, + { + "epoch": 76.0, + "grad_norm": 0.09722655266523361, + "learning_rate": 1.2e-05, + "loss": 0.0316, + "step": 7296 + }, + { + "epoch": 76.0, + "eval_LOCATION_f1": 0.8981481481481481, + "eval_LOCATION_number": 103, + "eval_LOCATION_precision": 0.8584070796460177, + "eval_LOCATION_recall": 0.941747572815534, + "eval_ORGANIZATION_f1": 0.8705882352941177, + "eval_ORGANIZATION_number": 171, + "eval_ORGANIZATION_precision": 0.8757396449704142, + "eval_ORGANIZATION_recall": 0.8654970760233918, + "eval_PERSON_f1": 0.9694656488549618, + "eval_PERSON_number": 131, + "eval_PERSON_precision": 0.9694656488549618, + "eval_PERSON_recall": 0.9694656488549618, + "eval_loss": 0.0496387705206871, + "eval_overall_accuracy": 0.9842584921292461, + "eval_overall_f1": 0.9095354523227382, + "eval_overall_precision": 0.9007263922518159, + "eval_overall_recall": 0.9185185185185185, + "eval_runtime": 0.5987, + "eval_samples_per_second": 283.941, + "eval_steps_per_second": 5.011, + "step": 7296 + }, + { + "epoch": 77.0, + "grad_norm": 0.33898091316223145, + "learning_rate": 1.1500000000000002e-05, + "loss": 0.0318, + "step": 7392 + }, + { + "epoch": 77.0, + "eval_LOCATION_f1": 0.9065420560747663, + "eval_LOCATION_number": 103, + "eval_LOCATION_precision": 0.8738738738738738, + "eval_LOCATION_recall": 0.941747572815534, + "eval_ORGANIZATION_f1": 0.875, + "eval_ORGANIZATION_number": 171, + "eval_ORGANIZATION_precision": 0.8909090909090909, + "eval_ORGANIZATION_recall": 0.8596491228070176, + "eval_PERSON_f1": 0.9694656488549618, + "eval_PERSON_number": 131, + "eval_PERSON_precision": 0.9694656488549618, + "eval_PERSON_recall": 0.9694656488549618, + "eval_loss": 0.050782710313797, + "eval_overall_accuracy": 0.9834299917149959, + "eval_overall_f1": 0.9137931034482758, + "eval_overall_precision": 0.9115479115479116, + "eval_overall_recall": 0.9160493827160494, + "eval_runtime": 0.6078, + "eval_samples_per_second": 279.711, + "eval_steps_per_second": 4.936, + "step": 7392 + }, + { + "epoch": 78.0, + "grad_norm": 0.5207917094230652, + "learning_rate": 1.1000000000000001e-05, + "loss": 0.0294, + "step": 7488 + }, + { + "epoch": 78.0, + "eval_LOCATION_f1": 0.9065420560747663, + "eval_LOCATION_number": 103, + "eval_LOCATION_precision": 0.8738738738738738, + "eval_LOCATION_recall": 0.941747572815534, + "eval_ORGANIZATION_f1": 0.8716417910447761, + "eval_ORGANIZATION_number": 171, + "eval_ORGANIZATION_precision": 0.8902439024390244, + "eval_ORGANIZATION_recall": 0.8538011695906432, + "eval_PERSON_f1": 0.9694656488549618, + "eval_PERSON_number": 131, + "eval_PERSON_precision": 0.9694656488549618, + "eval_PERSON_recall": 0.9694656488549618, + "eval_loss": 0.051793355494737625, + "eval_overall_accuracy": 0.9828776581054958, + "eval_overall_f1": 0.9124537607891491, + "eval_overall_precision": 0.9113300492610837, + "eval_overall_recall": 0.9135802469135802, + "eval_runtime": 0.5986, + "eval_samples_per_second": 284.001, + "eval_steps_per_second": 5.012, + "step": 7488 + }, + { + "epoch": 79.0, + "grad_norm": 1.5169590711593628, + "learning_rate": 1.05e-05, + "loss": 0.0307, + "step": 7584 + }, + { + "epoch": 79.0, + "eval_LOCATION_f1": 0.9014084507042254, + "eval_LOCATION_number": 103, + "eval_LOCATION_precision": 0.8727272727272727, + "eval_LOCATION_recall": 0.9320388349514563, + "eval_ORGANIZATION_f1": 0.8690476190476191, + "eval_ORGANIZATION_number": 171, + "eval_ORGANIZATION_precision": 0.8848484848484849, + "eval_ORGANIZATION_recall": 0.8538011695906432, + "eval_PERSON_f1": 0.9694656488549618, + "eval_PERSON_number": 131, + "eval_PERSON_precision": 0.9694656488549618, + "eval_PERSON_recall": 0.9694656488549618, + "eval_loss": 0.051547568291425705, + "eval_overall_accuracy": 0.9826014913007457, + "eval_overall_f1": 0.9099876695437731, + "eval_overall_precision": 0.9088669950738916, + "eval_overall_recall": 0.9111111111111111, + "eval_runtime": 0.5952, + "eval_samples_per_second": 285.614, + "eval_steps_per_second": 5.04, + "step": 7584 + }, + { + "epoch": 80.0, + "grad_norm": 0.9668756127357483, + "learning_rate": 1e-05, + "loss": 0.0314, + "step": 7680 + }, + { + "epoch": 80.0, + "eval_LOCATION_f1": 0.9074074074074074, + "eval_LOCATION_number": 103, + "eval_LOCATION_precision": 0.8672566371681416, + "eval_LOCATION_recall": 0.9514563106796117, + "eval_ORGANIZATION_f1": 0.875, + "eval_ORGANIZATION_number": 171, + "eval_ORGANIZATION_precision": 0.8909090909090909, + "eval_ORGANIZATION_recall": 0.8596491228070176, + "eval_PERSON_f1": 0.9694656488549618, + "eval_PERSON_number": 131, + "eval_PERSON_precision": 0.9694656488549618, + "eval_PERSON_recall": 0.9694656488549618, + "eval_loss": 0.05063919723033905, + "eval_overall_accuracy": 0.9845346589339962, + "eval_overall_f1": 0.914004914004914, + "eval_overall_precision": 0.9095354523227384, + "eval_overall_recall": 0.9185185185185185, + "eval_runtime": 0.5977, + "eval_samples_per_second": 284.404, + "eval_steps_per_second": 5.019, + "step": 7680 + }, + { + "epoch": 81.0, + "grad_norm": 0.464497447013855, + "learning_rate": 9.5e-06, + "loss": 0.0323, + "step": 7776 + }, + { + "epoch": 81.0, + "eval_LOCATION_f1": 0.9166666666666666, + "eval_LOCATION_number": 103, + "eval_LOCATION_precision": 0.8761061946902655, + "eval_LOCATION_recall": 0.9611650485436893, + "eval_ORGANIZATION_f1": 0.8802395209580838, + "eval_ORGANIZATION_number": 171, + "eval_ORGANIZATION_precision": 0.901840490797546, + "eval_ORGANIZATION_recall": 0.8596491228070176, + "eval_PERSON_f1": 0.9694656488549618, + "eval_PERSON_number": 131, + "eval_PERSON_precision": 0.9694656488549618, + "eval_PERSON_recall": 0.9694656488549618, + "eval_loss": 0.051717810332775116, + "eval_overall_accuracy": 0.983706158519746, + "eval_overall_f1": 0.91871921182266, + "eval_overall_precision": 0.9164619164619164, + "eval_overall_recall": 0.9209876543209876, + "eval_runtime": 0.602, + "eval_samples_per_second": 282.378, + "eval_steps_per_second": 4.983, + "step": 7776 + }, + { + "epoch": 82.0, + "grad_norm": 1.350043535232544, + "learning_rate": 9e-06, + "loss": 0.0314, + "step": 7872 + }, + { + "epoch": 82.0, + "eval_LOCATION_f1": 0.9056603773584906, + "eval_LOCATION_number": 103, + "eval_LOCATION_precision": 0.8807339449541285, + "eval_LOCATION_recall": 0.9320388349514563, + "eval_ORGANIZATION_f1": 0.8739002932551321, + "eval_ORGANIZATION_number": 171, + "eval_ORGANIZATION_precision": 0.8764705882352941, + "eval_ORGANIZATION_recall": 0.8713450292397661, + "eval_PERSON_f1": 0.9694656488549618, + "eval_PERSON_number": 131, + "eval_PERSON_precision": 0.9694656488549618, + "eval_PERSON_recall": 0.9694656488549618, + "eval_loss": 0.04948532581329346, + "eval_overall_accuracy": 0.9842584921292461, + "eval_overall_f1": 0.9128834355828221, + "eval_overall_precision": 0.9073170731707317, + "eval_overall_recall": 0.9185185185185185, + "eval_runtime": 0.6085, + "eval_samples_per_second": 279.397, + "eval_steps_per_second": 4.931, + "step": 7872 + }, + { + "epoch": 83.0, + "grad_norm": 1.0389440059661865, + "learning_rate": 8.500000000000002e-06, + "loss": 0.0301, + "step": 7968 + }, + { + "epoch": 83.0, + "eval_LOCATION_f1": 0.9065420560747663, + "eval_LOCATION_number": 103, + "eval_LOCATION_precision": 0.8738738738738738, + "eval_LOCATION_recall": 0.941747572815534, + "eval_ORGANIZATION_f1": 0.8776119402985074, + "eval_ORGANIZATION_number": 171, + "eval_ORGANIZATION_precision": 0.8963414634146342, + "eval_ORGANIZATION_recall": 0.8596491228070176, + "eval_PERSON_f1": 0.9694656488549618, + "eval_PERSON_number": 131, + "eval_PERSON_precision": 0.9694656488549618, + "eval_PERSON_recall": 0.9694656488549618, + "eval_loss": 0.052480973303318024, + "eval_overall_accuracy": 0.9828776581054958, + "eval_overall_f1": 0.9149198520345252, + "eval_overall_precision": 0.9137931034482759, + "eval_overall_recall": 0.9160493827160494, + "eval_runtime": 0.5957, + "eval_samples_per_second": 285.401, + "eval_steps_per_second": 5.036, + "step": 7968 + }, + { + "epoch": 84.0, + "grad_norm": 0.759078323841095, + "learning_rate": 8.000000000000001e-06, + "loss": 0.0309, + "step": 8064 + }, + { + "epoch": 84.0, + "eval_LOCATION_f1": 0.9116279069767441, + "eval_LOCATION_number": 103, + "eval_LOCATION_precision": 0.875, + "eval_LOCATION_recall": 0.9514563106796117, + "eval_ORGANIZATION_f1": 0.8802395209580838, + "eval_ORGANIZATION_number": 171, + "eval_ORGANIZATION_precision": 0.901840490797546, + "eval_ORGANIZATION_recall": 0.8596491228070176, + "eval_PERSON_f1": 0.9694656488549618, + "eval_PERSON_number": 131, + "eval_PERSON_precision": 0.9694656488549618, + "eval_PERSON_recall": 0.9694656488549618, + "eval_loss": 0.05260142683982849, + "eval_overall_accuracy": 0.9831538249102458, + "eval_overall_f1": 0.9173859432799014, + "eval_overall_precision": 0.916256157635468, + "eval_overall_recall": 0.9185185185185185, + "eval_runtime": 0.6048, + "eval_samples_per_second": 281.092, + "eval_steps_per_second": 4.96, + "step": 8064 + }, + { + "epoch": 85.0, + "grad_norm": 0.8793176412582397, + "learning_rate": 7.5e-06, + "loss": 0.0305, + "step": 8160 + }, + { + "epoch": 85.0, + "eval_LOCATION_f1": 0.9116279069767441, + "eval_LOCATION_number": 103, + "eval_LOCATION_precision": 0.875, + "eval_LOCATION_recall": 0.9514563106796117, + "eval_ORGANIZATION_f1": 0.8724035608308606, + "eval_ORGANIZATION_number": 171, + "eval_ORGANIZATION_precision": 0.8855421686746988, + "eval_ORGANIZATION_recall": 0.8596491228070176, + "eval_PERSON_f1": 0.9694656488549618, + "eval_PERSON_number": 131, + "eval_PERSON_precision": 0.9694656488549618, + "eval_PERSON_recall": 0.9694656488549618, + "eval_loss": 0.05188142508268356, + "eval_overall_accuracy": 0.983982325324496, + "eval_overall_f1": 0.914004914004914, + "eval_overall_precision": 0.9095354523227384, + "eval_overall_recall": 0.9185185185185185, + "eval_runtime": 0.6012, + "eval_samples_per_second": 282.78, + "eval_steps_per_second": 4.99, + "step": 8160 + }, + { + "epoch": 86.0, + "grad_norm": 0.38757383823394775, + "learning_rate": 7.000000000000001e-06, + "loss": 0.0295, + "step": 8256 + }, + { + "epoch": 86.0, + "eval_LOCATION_f1": 0.9116279069767441, + "eval_LOCATION_number": 103, + "eval_LOCATION_precision": 0.875, + "eval_LOCATION_recall": 0.9514563106796117, + "eval_ORGANIZATION_f1": 0.875, + "eval_ORGANIZATION_number": 171, + "eval_ORGANIZATION_precision": 0.8909090909090909, + "eval_ORGANIZATION_recall": 0.8596491228070176, + "eval_PERSON_f1": 0.9694656488549618, + "eval_PERSON_number": 131, + "eval_PERSON_precision": 0.9694656488549618, + "eval_PERSON_recall": 0.9694656488549618, + "eval_loss": 0.051886074244976044, + "eval_overall_accuracy": 0.983706158519746, + "eval_overall_f1": 0.915129151291513, + "eval_overall_precision": 0.9117647058823529, + "eval_overall_recall": 0.9185185185185185, + "eval_runtime": 0.6639, + "eval_samples_per_second": 256.06, + "eval_steps_per_second": 4.519, + "step": 8256 + }, + { + "epoch": 87.0, + "grad_norm": 1.0249720811843872, + "learning_rate": 6.5000000000000004e-06, + "loss": 0.0316, + "step": 8352 + }, + { + "epoch": 87.0, + "eval_LOCATION_f1": 0.9065420560747663, + "eval_LOCATION_number": 103, + "eval_LOCATION_precision": 0.8738738738738738, + "eval_LOCATION_recall": 0.941747572815534, + "eval_ORGANIZATION_f1": 0.8776119402985074, + "eval_ORGANIZATION_number": 171, + "eval_ORGANIZATION_precision": 0.8963414634146342, + "eval_ORGANIZATION_recall": 0.8596491228070176, + "eval_PERSON_f1": 0.9694656488549618, + "eval_PERSON_number": 131, + "eval_PERSON_precision": 0.9694656488549618, + "eval_PERSON_recall": 0.9694656488549618, + "eval_loss": 0.05171561613678932, + "eval_overall_accuracy": 0.9828776581054958, + "eval_overall_f1": 0.9149198520345252, + "eval_overall_precision": 0.9137931034482759, + "eval_overall_recall": 0.9160493827160494, + "eval_runtime": 0.6115, + "eval_samples_per_second": 278.002, + "eval_steps_per_second": 4.906, + "step": 8352 + }, + { + "epoch": 88.0, + "grad_norm": 0.5740599632263184, + "learning_rate": 6e-06, + "loss": 0.0298, + "step": 8448 + }, + { + "epoch": 88.0, + "eval_LOCATION_f1": 0.9014084507042254, + "eval_LOCATION_number": 103, + "eval_LOCATION_precision": 0.8727272727272727, + "eval_LOCATION_recall": 0.9320388349514563, + "eval_ORGANIZATION_f1": 0.8698224852071007, + "eval_ORGANIZATION_number": 171, + "eval_ORGANIZATION_precision": 0.8802395209580839, + "eval_ORGANIZATION_recall": 0.8596491228070176, + "eval_PERSON_f1": 0.9694656488549618, + "eval_PERSON_number": 131, + "eval_PERSON_precision": 0.9694656488549618, + "eval_PERSON_recall": 0.9694656488549618, + "eval_loss": 0.05182284861803055, + "eval_overall_accuracy": 0.9831538249102458, + "eval_overall_f1": 0.9102091020910209, + "eval_overall_precision": 0.9068627450980392, + "eval_overall_recall": 0.9135802469135802, + "eval_runtime": 0.6039, + "eval_samples_per_second": 281.507, + "eval_steps_per_second": 4.968, + "step": 8448 + }, + { + "epoch": 89.0, + "grad_norm": 0.25633862614631653, + "learning_rate": 5.500000000000001e-06, + "loss": 0.0299, + "step": 8544 + }, + { + "epoch": 89.0, + "eval_LOCATION_f1": 0.8971962616822431, + "eval_LOCATION_number": 103, + "eval_LOCATION_precision": 0.8648648648648649, + "eval_LOCATION_recall": 0.9320388349514563, + "eval_ORGANIZATION_f1": 0.8802395209580838, + "eval_ORGANIZATION_number": 171, + "eval_ORGANIZATION_precision": 0.901840490797546, + "eval_ORGANIZATION_recall": 0.8596491228070176, + "eval_PERSON_f1": 0.9694656488549618, + "eval_PERSON_number": 131, + "eval_PERSON_precision": 0.9694656488549618, + "eval_PERSON_recall": 0.9694656488549618, + "eval_loss": 0.053376831114292145, + "eval_overall_accuracy": 0.9828776581054958, + "eval_overall_f1": 0.9135802469135802, + "eval_overall_precision": 0.9135802469135802, + "eval_overall_recall": 0.9135802469135802, + "eval_runtime": 0.5978, + "eval_samples_per_second": 284.372, + "eval_steps_per_second": 5.018, + "step": 8544 + }, + { + "epoch": 90.0, + "grad_norm": 0.3275259733200073, + "learning_rate": 5e-06, + "loss": 0.0292, + "step": 8640 + }, + { + "epoch": 90.0, + "eval_LOCATION_f1": 0.9014084507042254, + "eval_LOCATION_number": 103, + "eval_LOCATION_precision": 0.8727272727272727, + "eval_LOCATION_recall": 0.9320388349514563, + "eval_ORGANIZATION_f1": 0.8698224852071007, + "eval_ORGANIZATION_number": 171, + "eval_ORGANIZATION_precision": 0.8802395209580839, + "eval_ORGANIZATION_recall": 0.8596491228070176, + "eval_PERSON_f1": 0.9694656488549618, + "eval_PERSON_number": 131, + "eval_PERSON_precision": 0.9694656488549618, + "eval_PERSON_recall": 0.9694656488549618, + "eval_loss": 0.051723167300224304, + "eval_overall_accuracy": 0.9831538249102458, + "eval_overall_f1": 0.9102091020910209, + "eval_overall_precision": 0.9068627450980392, + "eval_overall_recall": 0.9135802469135802, + "eval_runtime": 0.6033, + "eval_samples_per_second": 281.799, + "eval_steps_per_second": 4.973, + "step": 8640 + }, + { + "epoch": 91.0, + "grad_norm": 1.8535882234573364, + "learning_rate": 4.5e-06, + "loss": 0.0301, + "step": 8736 + }, + { + "epoch": 91.0, + "eval_LOCATION_f1": 0.9014084507042254, + "eval_LOCATION_number": 103, + "eval_LOCATION_precision": 0.8727272727272727, + "eval_LOCATION_recall": 0.9320388349514563, + "eval_ORGANIZATION_f1": 0.8672566371681416, + "eval_ORGANIZATION_number": 171, + "eval_ORGANIZATION_precision": 0.875, + "eval_ORGANIZATION_recall": 0.8596491228070176, + "eval_PERSON_f1": 0.9694656488549618, + "eval_PERSON_number": 131, + "eval_PERSON_precision": 0.9694656488549618, + "eval_PERSON_recall": 0.9694656488549618, + "eval_loss": 0.051088012754917145, + "eval_overall_accuracy": 0.9834299917149959, + "eval_overall_f1": 0.9090909090909091, + "eval_overall_precision": 0.9046454767726161, + "eval_overall_recall": 0.9135802469135802, + "eval_runtime": 0.6066, + "eval_samples_per_second": 280.273, + "eval_steps_per_second": 4.946, + "step": 8736 + }, + { + "epoch": 92.0, + "grad_norm": 0.26752015948295593, + "learning_rate": 4.000000000000001e-06, + "loss": 0.0294, + "step": 8832 + }, + { + "epoch": 92.0, + "eval_LOCATION_f1": 0.9014084507042254, + "eval_LOCATION_number": 103, + "eval_LOCATION_precision": 0.8727272727272727, + "eval_LOCATION_recall": 0.9320388349514563, + "eval_ORGANIZATION_f1": 0.8698224852071007, + "eval_ORGANIZATION_number": 171, + "eval_ORGANIZATION_precision": 0.8802395209580839, + "eval_ORGANIZATION_recall": 0.8596491228070176, + "eval_PERSON_f1": 0.9694656488549618, + "eval_PERSON_number": 131, + "eval_PERSON_precision": 0.9694656488549618, + "eval_PERSON_recall": 0.9694656488549618, + "eval_loss": 0.05176582932472229, + "eval_overall_accuracy": 0.9831538249102458, + "eval_overall_f1": 0.9102091020910209, + "eval_overall_precision": 0.9068627450980392, + "eval_overall_recall": 0.9135802469135802, + "eval_runtime": 0.6749, + "eval_samples_per_second": 251.872, + "eval_steps_per_second": 4.445, + "step": 8832 + }, + { + "epoch": 93.0, + "grad_norm": 0.5065526366233826, + "learning_rate": 3.5000000000000004e-06, + "loss": 0.0296, + "step": 8928 + }, + { + "epoch": 93.0, + "eval_LOCATION_f1": 0.9014084507042254, + "eval_LOCATION_number": 103, + "eval_LOCATION_precision": 0.8727272727272727, + "eval_LOCATION_recall": 0.9320388349514563, + "eval_ORGANIZATION_f1": 0.8672566371681416, + "eval_ORGANIZATION_number": 171, + "eval_ORGANIZATION_precision": 0.875, + "eval_ORGANIZATION_recall": 0.8596491228070176, + "eval_PERSON_f1": 0.9694656488549618, + "eval_PERSON_number": 131, + "eval_PERSON_precision": 0.9694656488549618, + "eval_PERSON_recall": 0.9694656488549618, + "eval_loss": 0.05158179625868797, + "eval_overall_accuracy": 0.9834299917149959, + "eval_overall_f1": 0.9090909090909091, + "eval_overall_precision": 0.9046454767726161, + "eval_overall_recall": 0.9135802469135802, + "eval_runtime": 0.6001, + "eval_samples_per_second": 283.295, + "eval_steps_per_second": 4.999, + "step": 8928 + }, + { + "epoch": 94.0, + "grad_norm": 0.13080176711082458, + "learning_rate": 3e-06, + "loss": 0.0293, + "step": 9024 + }, + { + "epoch": 94.0, + "eval_LOCATION_f1": 0.9023255813953488, + "eval_LOCATION_number": 103, + "eval_LOCATION_precision": 0.8660714285714286, + "eval_LOCATION_recall": 0.941747572815534, + "eval_ORGANIZATION_f1": 0.875, + "eval_ORGANIZATION_number": 171, + "eval_ORGANIZATION_precision": 0.8909090909090909, + "eval_ORGANIZATION_recall": 0.8596491228070176, + "eval_PERSON_f1": 0.9694656488549618, + "eval_PERSON_number": 131, + "eval_PERSON_precision": 0.9694656488549618, + "eval_PERSON_recall": 0.9694656488549618, + "eval_loss": 0.05225222557783127, + "eval_overall_accuracy": 0.983982325324496, + "eval_overall_f1": 0.9126691266912669, + "eval_overall_precision": 0.9093137254901961, + "eval_overall_recall": 0.9160493827160494, + "eval_runtime": 0.6008, + "eval_samples_per_second": 282.96, + "eval_steps_per_second": 4.993, + "step": 9024 + }, + { + "epoch": 95.0, + "grad_norm": 0.2286527156829834, + "learning_rate": 2.5e-06, + "loss": 0.0295, + "step": 9120 + }, + { + "epoch": 95.0, + "eval_LOCATION_f1": 0.9014084507042254, + "eval_LOCATION_number": 103, + "eval_LOCATION_precision": 0.8727272727272727, + "eval_LOCATION_recall": 0.9320388349514563, + "eval_ORGANIZATION_f1": 0.8672566371681416, + "eval_ORGANIZATION_number": 171, + "eval_ORGANIZATION_precision": 0.875, + "eval_ORGANIZATION_recall": 0.8596491228070176, + "eval_PERSON_f1": 0.9694656488549618, + "eval_PERSON_number": 131, + "eval_PERSON_precision": 0.9694656488549618, + "eval_PERSON_recall": 0.9694656488549618, + "eval_loss": 0.05151774361729622, + "eval_overall_accuracy": 0.9834299917149959, + "eval_overall_f1": 0.9090909090909091, + "eval_overall_precision": 0.9046454767726161, + "eval_overall_recall": 0.9135802469135802, + "eval_runtime": 0.6086, + "eval_samples_per_second": 279.342, + "eval_steps_per_second": 4.93, + "step": 9120 + }, + { + "epoch": 96.0, + "grad_norm": 0.8817101120948792, + "learning_rate": 2.0000000000000003e-06, + "loss": 0.0284, + "step": 9216 + }, + { + "epoch": 96.0, + "eval_LOCATION_f1": 0.9014084507042254, + "eval_LOCATION_number": 103, + "eval_LOCATION_precision": 0.8727272727272727, + "eval_LOCATION_recall": 0.9320388349514563, + "eval_ORGANIZATION_f1": 0.8672566371681416, + "eval_ORGANIZATION_number": 171, + "eval_ORGANIZATION_precision": 0.875, + "eval_ORGANIZATION_recall": 0.8596491228070176, + "eval_PERSON_f1": 0.9694656488549618, + "eval_PERSON_number": 131, + "eval_PERSON_precision": 0.9694656488549618, + "eval_PERSON_recall": 0.9694656488549618, + "eval_loss": 0.051495932042598724, + "eval_overall_accuracy": 0.9834299917149959, + "eval_overall_f1": 0.9090909090909091, + "eval_overall_precision": 0.9046454767726161, + "eval_overall_recall": 0.9135802469135802, + "eval_runtime": 0.5975, + "eval_samples_per_second": 284.531, + "eval_steps_per_second": 5.021, + "step": 9216 + }, + { + "epoch": 97.0, + "grad_norm": 0.6794419288635254, + "learning_rate": 1.5e-06, + "loss": 0.0289, + "step": 9312 + }, + { + "epoch": 97.0, + "eval_LOCATION_f1": 0.892018779342723, + "eval_LOCATION_number": 103, + "eval_LOCATION_precision": 0.8636363636363636, + "eval_LOCATION_recall": 0.9223300970873787, + "eval_ORGANIZATION_f1": 0.8698224852071007, + "eval_ORGANIZATION_number": 171, + "eval_ORGANIZATION_precision": 0.8802395209580839, + "eval_ORGANIZATION_recall": 0.8596491228070176, + "eval_PERSON_f1": 0.9694656488549618, + "eval_PERSON_number": 131, + "eval_PERSON_precision": 0.9694656488549618, + "eval_PERSON_recall": 0.9694656488549618, + "eval_loss": 0.0521596297621727, + "eval_overall_accuracy": 0.9834299917149959, + "eval_overall_f1": 0.9077490774907748, + "eval_overall_precision": 0.9044117647058824, + "eval_overall_recall": 0.9111111111111111, + "eval_runtime": 0.6054, + "eval_samples_per_second": 280.816, + "eval_steps_per_second": 4.956, + "step": 9312 + }, + { + "epoch": 98.0, + "grad_norm": 1.6478160619735718, + "learning_rate": 1.0000000000000002e-06, + "loss": 0.0282, + "step": 9408 + }, + { + "epoch": 98.0, + "eval_LOCATION_f1": 0.892018779342723, + "eval_LOCATION_number": 103, + "eval_LOCATION_precision": 0.8636363636363636, + "eval_LOCATION_recall": 0.9223300970873787, + "eval_ORGANIZATION_f1": 0.8698224852071007, + "eval_ORGANIZATION_number": 171, + "eval_ORGANIZATION_precision": 0.8802395209580839, + "eval_ORGANIZATION_recall": 0.8596491228070176, + "eval_PERSON_f1": 0.9694656488549618, + "eval_PERSON_number": 131, + "eval_PERSON_precision": 0.9694656488549618, + "eval_PERSON_recall": 0.9694656488549618, + "eval_loss": 0.05200561136007309, + "eval_overall_accuracy": 0.9834299917149959, + "eval_overall_f1": 0.9077490774907748, + "eval_overall_precision": 0.9044117647058824, + "eval_overall_recall": 0.9111111111111111, + "eval_runtime": 0.595, + "eval_samples_per_second": 285.72, + "eval_steps_per_second": 5.042, + "step": 9408 + }, + { + "epoch": 99.0, + "grad_norm": 0.42531368136405945, + "learning_rate": 5.000000000000001e-07, + "loss": 0.0287, + "step": 9504 + }, + { + "epoch": 99.0, + "eval_LOCATION_f1": 0.9014084507042254, + "eval_LOCATION_number": 103, + "eval_LOCATION_precision": 0.8727272727272727, + "eval_LOCATION_recall": 0.9320388349514563, + "eval_ORGANIZATION_f1": 0.8672566371681416, + "eval_ORGANIZATION_number": 171, + "eval_ORGANIZATION_precision": 0.875, + "eval_ORGANIZATION_recall": 0.8596491228070176, + "eval_PERSON_f1": 0.9694656488549618, + "eval_PERSON_number": 131, + "eval_PERSON_precision": 0.9694656488549618, + "eval_PERSON_recall": 0.9694656488549618, + "eval_loss": 0.051944032311439514, + "eval_overall_accuracy": 0.9834299917149959, + "eval_overall_f1": 0.9090909090909091, + "eval_overall_precision": 0.9046454767726161, + "eval_overall_recall": 0.9135802469135802, + "eval_runtime": 0.604, + "eval_samples_per_second": 281.445, + "eval_steps_per_second": 4.967, + "step": 9504 + }, + { + "epoch": 100.0, + "grad_norm": 0.9529930949211121, + "learning_rate": 0.0, + "loss": 0.0301, + "step": 9600 + }, + { + "epoch": 100.0, + "eval_LOCATION_f1": 0.9014084507042254, + "eval_LOCATION_number": 103, + "eval_LOCATION_precision": 0.8727272727272727, + "eval_LOCATION_recall": 0.9320388349514563, + "eval_ORGANIZATION_f1": 0.8672566371681416, + "eval_ORGANIZATION_number": 171, + "eval_ORGANIZATION_precision": 0.875, + "eval_ORGANIZATION_recall": 0.8596491228070176, + "eval_PERSON_f1": 0.9694656488549618, + "eval_PERSON_number": 131, + "eval_PERSON_precision": 0.9694656488549618, + "eval_PERSON_recall": 0.9694656488549618, + "eval_loss": 0.051663465797901154, + "eval_overall_accuracy": 0.9834299917149959, + "eval_overall_f1": 0.9090909090909091, + "eval_overall_precision": 0.9046454767726161, + "eval_overall_recall": 0.9135802469135802, + "eval_runtime": 0.6039, + "eval_samples_per_second": 281.497, + "eval_steps_per_second": 4.968, + "step": 9600 + }, + { + "epoch": 100.0, + "step": 9600, + "total_flos": 3903436930778652.0, + "train_loss": 0.07638429149985314, + "train_runtime": 1043.5343, + "train_samples_per_second": 146.713, + "train_steps_per_second": 9.2 + } + ], + "logging_steps": 500, + "max_steps": 9600, + "num_input_tokens_seen": 0, + "num_train_epochs": 100, + "save_steps": 500, + "total_flos": 3903436930778652.0, + "train_batch_size": 16, + "trial_name": null, + "trial_params": null +}