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": 1.3216755390167236, + "learning_rate": 4.9500000000000004e-05, + "loss": 0.834, + "step": 96 + }, + { + "epoch": 1.0, + "eval_LOCATION_f1": 0.0, + "eval_LOCATION_number": 86, + "eval_LOCATION_precision": 0.0, + "eval_LOCATION_recall": 0.0, + "eval_ORGANIZATION_f1": 0.021739130434782608, + "eval_ORGANIZATION_number": 178, + "eval_ORGANIZATION_precision": 0.3333333333333333, + "eval_ORGANIZATION_recall": 0.011235955056179775, + "eval_PERSON_f1": 0.0, + "eval_PERSON_number": 128, + "eval_PERSON_precision": 0.0, + "eval_PERSON_recall": 0.0, + "eval_loss": 0.519976794719696, + "eval_overall_accuracy": 0.8437246963562753, + "eval_overall_f1": 0.009999999999999998, + "eval_overall_precision": 0.25, + "eval_overall_recall": 0.00510204081632653, + "eval_runtime": 0.2471, + "eval_samples_per_second": 687.937, + "eval_steps_per_second": 12.14, + "step": 96 + }, + { + "epoch": 2.0, + "grad_norm": 0.9635790586471558, + "learning_rate": 4.9e-05, + "loss": 0.4501, + "step": 192 + }, + { + "epoch": 2.0, + "eval_LOCATION_f1": 0.296875, + "eval_LOCATION_number": 86, + "eval_LOCATION_precision": 0.4523809523809524, + "eval_LOCATION_recall": 0.22093023255813954, + "eval_ORGANIZATION_f1": 0.4215686274509804, + "eval_ORGANIZATION_number": 178, + "eval_ORGANIZATION_precision": 0.3739130434782609, + "eval_ORGANIZATION_recall": 0.48314606741573035, + "eval_PERSON_f1": 0.427299703264095, + "eval_PERSON_number": 128, + "eval_PERSON_precision": 0.3444976076555024, + "eval_PERSON_recall": 0.5625, + "eval_loss": 0.30531638860702515, + "eval_overall_accuracy": 0.9041835357624831, + "eval_overall_f1": 0.4054982817869416, + "eval_overall_precision": 0.367983367983368, + "eval_overall_recall": 0.45153061224489793, + "eval_runtime": 0.2475, + "eval_samples_per_second": 686.861, + "eval_steps_per_second": 12.121, + "step": 192 + }, + { + "epoch": 3.0, + "grad_norm": 1.0305564403533936, + "learning_rate": 4.85e-05, + "loss": 0.309, + "step": 288 + }, + { + "epoch": 3.0, + "eval_LOCATION_f1": 0.5398773006134969, + "eval_LOCATION_number": 86, + "eval_LOCATION_precision": 0.5714285714285714, + "eval_LOCATION_recall": 0.5116279069767442, + "eval_ORGANIZATION_f1": 0.6542553191489362, + "eval_ORGANIZATION_number": 178, + "eval_ORGANIZATION_precision": 0.6212121212121212, + "eval_ORGANIZATION_recall": 0.6910112359550562, + "eval_PERSON_f1": 0.7342657342657343, + "eval_PERSON_number": 128, + "eval_PERSON_precision": 0.6645569620253164, + "eval_PERSON_recall": 0.8203125, + "eval_loss": 0.21796205639839172, + "eval_overall_accuracy": 0.9425101214574899, + "eval_overall_f1": 0.6593939393939394, + "eval_overall_precision": 0.628175519630485, + "eval_overall_recall": 0.6938775510204082, + "eval_runtime": 0.2476, + "eval_samples_per_second": 686.487, + "eval_steps_per_second": 12.114, + "step": 288 + }, + { + "epoch": 4.0, + "grad_norm": 0.8546921610832214, + "learning_rate": 4.8e-05, + "loss": 0.2169, + "step": 384 + }, + { + "epoch": 4.0, + "eval_LOCATION_f1": 0.7065217391304347, + "eval_LOCATION_number": 86, + "eval_LOCATION_precision": 0.6632653061224489, + "eval_LOCATION_recall": 0.7558139534883721, + "eval_ORGANIZATION_f1": 0.7457627118644068, + "eval_ORGANIZATION_number": 178, + "eval_ORGANIZATION_precision": 0.75, + "eval_ORGANIZATION_recall": 0.7415730337078652, + "eval_PERSON_f1": 0.9022556390977444, + "eval_PERSON_number": 128, + "eval_PERSON_precision": 0.8695652173913043, + "eval_PERSON_recall": 0.9375, + "eval_loss": 0.1386880725622177, + "eval_overall_accuracy": 0.9624831309041836, + "eval_overall_f1": 0.7885572139303483, + "eval_overall_precision": 0.7694174757281553, + "eval_overall_recall": 0.8086734693877551, + "eval_runtime": 0.2495, + "eval_samples_per_second": 681.233, + "eval_steps_per_second": 12.022, + "step": 384 + }, + { + "epoch": 5.0, + "grad_norm": 0.6915740966796875, + "learning_rate": 4.75e-05, + "loss": 0.145, + "step": 480 + }, + { + "epoch": 5.0, + "eval_LOCATION_f1": 0.7807486631016043, + "eval_LOCATION_number": 86, + "eval_LOCATION_precision": 0.7227722772277227, + "eval_LOCATION_recall": 0.8488372093023255, + "eval_ORGANIZATION_f1": 0.7819148936170214, + "eval_ORGANIZATION_number": 178, + "eval_ORGANIZATION_precision": 0.7424242424242424, + "eval_ORGANIZATION_recall": 0.8258426966292135, + "eval_PERSON_f1": 0.9389312977099238, + "eval_PERSON_number": 128, + "eval_PERSON_precision": 0.917910447761194, + "eval_PERSON_recall": 0.9609375, + "eval_loss": 0.10832899063825607, + "eval_overall_accuracy": 0.9670715249662618, + "eval_overall_f1": 0.8315151515151514, + "eval_overall_precision": 0.792147806004619, + "eval_overall_recall": 0.875, + "eval_runtime": 0.2512, + "eval_samples_per_second": 676.782, + "eval_steps_per_second": 11.943, + "step": 480 + }, + { + "epoch": 6.0, + "grad_norm": 0.8912493586540222, + "learning_rate": 4.7e-05, + "loss": 0.1191, + "step": 576 + }, + { + "epoch": 6.0, + "eval_LOCATION_f1": 0.774869109947644, + "eval_LOCATION_number": 86, + "eval_LOCATION_precision": 0.7047619047619048, + "eval_LOCATION_recall": 0.8604651162790697, + "eval_ORGANIZATION_f1": 0.8225352112676056, + "eval_ORGANIZATION_number": 178, + "eval_ORGANIZATION_precision": 0.8248587570621468, + "eval_ORGANIZATION_recall": 0.8202247191011236, + "eval_PERSON_f1": 0.9461538461538462, + "eval_PERSON_number": 128, + "eval_PERSON_precision": 0.9318181818181818, + "eval_PERSON_recall": 0.9609375, + "eval_loss": 0.09309709072113037, + "eval_overall_accuracy": 0.970310391363023, + "eval_overall_f1": 0.8511166253101737, + "eval_overall_precision": 0.8285024154589372, + "eval_overall_recall": 0.875, + "eval_runtime": 0.2486, + "eval_samples_per_second": 683.849, + "eval_steps_per_second": 12.068, + "step": 576 + }, + { + "epoch": 7.0, + "grad_norm": 1.663607120513916, + "learning_rate": 4.6500000000000005e-05, + "loss": 0.1063, + "step": 672 + }, + { + "epoch": 7.0, + "eval_LOCATION_f1": 0.7938144329896906, + "eval_LOCATION_number": 86, + "eval_LOCATION_precision": 0.7129629629629629, + "eval_LOCATION_recall": 0.8953488372093024, + "eval_ORGANIZATION_f1": 0.8228882833787466, + "eval_ORGANIZATION_number": 178, + "eval_ORGANIZATION_precision": 0.798941798941799, + "eval_ORGANIZATION_recall": 0.848314606741573, + "eval_PERSON_f1": 0.9545454545454545, + "eval_PERSON_number": 128, + "eval_PERSON_precision": 0.9264705882352942, + "eval_PERSON_recall": 0.984375, + "eval_loss": 0.0903930515050888, + "eval_overall_accuracy": 0.9697705802968961, + "eval_overall_f1": 0.8581818181818182, + "eval_overall_precision": 0.8175519630484989, + "eval_overall_recall": 0.9030612244897959, + "eval_runtime": 0.2477, + "eval_samples_per_second": 686.203, + "eval_steps_per_second": 12.109, + "step": 672 + }, + { + "epoch": 8.0, + "grad_norm": 0.6487865447998047, + "learning_rate": 4.600000000000001e-05, + "loss": 0.0947, + "step": 768 + }, + { + "epoch": 8.0, + "eval_LOCATION_f1": 0.823529411764706, + "eval_LOCATION_number": 86, + "eval_LOCATION_precision": 0.7623762376237624, + "eval_LOCATION_recall": 0.8953488372093024, + "eval_ORGANIZATION_f1": 0.8333333333333334, + "eval_ORGANIZATION_number": 178, + "eval_ORGANIZATION_precision": 0.8241758241758241, + "eval_ORGANIZATION_recall": 0.8426966292134831, + "eval_PERSON_f1": 0.9571984435797667, + "eval_PERSON_number": 128, + "eval_PERSON_precision": 0.9534883720930233, + "eval_PERSON_recall": 0.9609375, + "eval_loss": 0.0733816996216774, + "eval_overall_accuracy": 0.9767881241565453, + "eval_overall_f1": 0.8706467661691543, + "eval_overall_precision": 0.8495145631067961, + "eval_overall_recall": 0.8928571428571429, + "eval_runtime": 0.2484, + "eval_samples_per_second": 684.488, + "eval_steps_per_second": 12.079, + "step": 768 + }, + { + "epoch": 9.0, + "grad_norm": 0.4733569920063019, + "learning_rate": 4.55e-05, + "loss": 0.085, + "step": 864 + }, + { + "epoch": 9.0, + "eval_LOCATION_f1": 0.8478260869565216, + "eval_LOCATION_number": 86, + "eval_LOCATION_precision": 0.7959183673469388, + "eval_LOCATION_recall": 0.9069767441860465, + "eval_ORGANIZATION_f1": 0.8657534246575344, + "eval_ORGANIZATION_number": 178, + "eval_ORGANIZATION_precision": 0.8449197860962567, + "eval_ORGANIZATION_recall": 0.8876404494382022, + "eval_PERSON_f1": 0.9575289575289575, + "eval_PERSON_number": 128, + "eval_PERSON_precision": 0.9465648854961832, + "eval_PERSON_recall": 0.96875, + "eval_loss": 0.06920091062784195, + "eval_overall_accuracy": 0.9789473684210527, + "eval_overall_f1": 0.8910891089108911, + "eval_overall_precision": 0.8653846153846154, + "eval_overall_recall": 0.9183673469387755, + "eval_runtime": 0.2503, + "eval_samples_per_second": 679.314, + "eval_steps_per_second": 11.988, + "step": 864 + }, + { + "epoch": 10.0, + "grad_norm": 1.6128334999084473, + "learning_rate": 4.5e-05, + "loss": 0.082, + "step": 960 + }, + { + "epoch": 10.0, + "eval_LOCATION_f1": 0.8478260869565216, + "eval_LOCATION_number": 86, + "eval_LOCATION_precision": 0.7959183673469388, + "eval_LOCATION_recall": 0.9069767441860465, + "eval_ORGANIZATION_f1": 0.8698060941828255, + "eval_ORGANIZATION_number": 178, + "eval_ORGANIZATION_precision": 0.8579234972677595, + "eval_ORGANIZATION_recall": 0.8820224719101124, + "eval_PERSON_f1": 0.9689922480620154, + "eval_PERSON_number": 128, + "eval_PERSON_precision": 0.9615384615384616, + "eval_PERSON_recall": 0.9765625, + "eval_loss": 0.06693387776613235, + "eval_overall_accuracy": 0.9789473684210527, + "eval_overall_f1": 0.8966376089663761, + "eval_overall_precision": 0.8759124087591241, + "eval_overall_recall": 0.9183673469387755, + "eval_runtime": 0.244, + "eval_samples_per_second": 696.829, + "eval_steps_per_second": 12.297, + "step": 960 + }, + { + "epoch": 11.0, + "grad_norm": 0.9450006484985352, + "learning_rate": 4.4500000000000004e-05, + "loss": 0.0769, + "step": 1056 + }, + { + "epoch": 11.0, + "eval_LOCATION_f1": 0.8756756756756757, + "eval_LOCATION_number": 86, + "eval_LOCATION_precision": 0.8181818181818182, + "eval_LOCATION_recall": 0.9418604651162791, + "eval_ORGANIZATION_f1": 0.8736263736263736, + "eval_ORGANIZATION_number": 178, + "eval_ORGANIZATION_precision": 0.8548387096774194, + "eval_ORGANIZATION_recall": 0.8932584269662921, + "eval_PERSON_f1": 0.9649805447470817, + "eval_PERSON_number": 128, + "eval_PERSON_precision": 0.9612403100775194, + "eval_PERSON_recall": 0.96875, + "eval_loss": 0.06316138803958893, + "eval_overall_accuracy": 0.9802968960863697, + "eval_overall_f1": 0.9032258064516129, + "eval_overall_precision": 0.8792270531400966, + "eval_overall_recall": 0.9285714285714286, + "eval_runtime": 0.2433, + "eval_samples_per_second": 698.829, + "eval_steps_per_second": 12.332, + "step": 1056 + }, + { + "epoch": 12.0, + "grad_norm": 1.495387315750122, + "learning_rate": 4.4000000000000006e-05, + "loss": 0.0749, + "step": 1152 + }, + { + "epoch": 12.0, + "eval_LOCATION_f1": 0.8839779005524862, + "eval_LOCATION_number": 86, + "eval_LOCATION_precision": 0.8421052631578947, + "eval_LOCATION_recall": 0.9302325581395349, + "eval_ORGANIZATION_f1": 0.893150684931507, + "eval_ORGANIZATION_number": 178, + "eval_ORGANIZATION_precision": 0.8716577540106952, + "eval_ORGANIZATION_recall": 0.9157303370786517, + "eval_PERSON_f1": 0.9727626459143969, + "eval_PERSON_number": 128, + "eval_PERSON_precision": 0.9689922480620154, + "eval_PERSON_recall": 0.9765625, + "eval_loss": 0.060925450176000595, + "eval_overall_accuracy": 0.9819163292847504, + "eval_overall_f1": 0.916562889165629, + "eval_overall_precision": 0.8953771289537713, + "eval_overall_recall": 0.9387755102040817, + "eval_runtime": 0.2418, + "eval_samples_per_second": 703.179, + "eval_steps_per_second": 12.409, + "step": 1152 + }, + { + "epoch": 13.0, + "grad_norm": 1.075753092765808, + "learning_rate": 4.35e-05, + "loss": 0.0648, + "step": 1248 + }, + { + "epoch": 13.0, + "eval_LOCATION_f1": 0.8729281767955801, + "eval_LOCATION_number": 86, + "eval_LOCATION_precision": 0.8315789473684211, + "eval_LOCATION_recall": 0.9186046511627907, + "eval_ORGANIZATION_f1": 0.8833333333333334, + "eval_ORGANIZATION_number": 178, + "eval_ORGANIZATION_precision": 0.8736263736263736, + "eval_ORGANIZATION_recall": 0.8932584269662921, + "eval_PERSON_f1": 0.9649805447470817, + "eval_PERSON_number": 128, + "eval_PERSON_precision": 0.9612403100775194, + "eval_PERSON_recall": 0.96875, + "eval_loss": 0.05693093314766884, + "eval_overall_accuracy": 0.9811066126855601, + "eval_overall_f1": 0.9072681704260652, + "eval_overall_precision": 0.8916256157635468, + "eval_overall_recall": 0.923469387755102, + "eval_runtime": 0.2491, + "eval_samples_per_second": 682.437, + "eval_steps_per_second": 12.043, + "step": 1248 + }, + { + "epoch": 14.0, + "grad_norm": 0.4505135715007782, + "learning_rate": 4.3e-05, + "loss": 0.0639, + "step": 1344 + }, + { + "epoch": 14.0, + "eval_LOCATION_f1": 0.861878453038674, + "eval_LOCATION_number": 86, + "eval_LOCATION_precision": 0.8210526315789474, + "eval_LOCATION_recall": 0.9069767441860465, + "eval_ORGANIZATION_f1": 0.8857142857142857, + "eval_ORGANIZATION_number": 178, + "eval_ORGANIZATION_precision": 0.9011627906976745, + "eval_ORGANIZATION_recall": 0.8707865168539326, + "eval_PERSON_f1": 0.9649805447470817, + "eval_PERSON_number": 128, + "eval_PERSON_precision": 0.9612403100775194, + "eval_PERSON_recall": 0.96875, + "eval_loss": 0.059763651341199875, + "eval_overall_accuracy": 0.9811066126855601, + "eval_overall_f1": 0.9060913705583755, + "eval_overall_precision": 0.9015151515151515, + "eval_overall_recall": 0.9107142857142857, + "eval_runtime": 0.2481, + "eval_samples_per_second": 685.305, + "eval_steps_per_second": 12.094, + "step": 1344 + }, + { + "epoch": 15.0, + "grad_norm": 0.8750821352005005, + "learning_rate": 4.25e-05, + "loss": 0.0627, + "step": 1440 + }, + { + "epoch": 15.0, + "eval_LOCATION_f1": 0.88268156424581, + "eval_LOCATION_number": 86, + "eval_LOCATION_precision": 0.8494623655913979, + "eval_LOCATION_recall": 0.9186046511627907, + "eval_ORGANIZATION_f1": 0.8925619834710743, + "eval_ORGANIZATION_number": 178, + "eval_ORGANIZATION_precision": 0.8756756756756757, + "eval_ORGANIZATION_recall": 0.9101123595505618, + "eval_PERSON_f1": 0.9803921568627452, + "eval_PERSON_number": 128, + "eval_PERSON_precision": 0.984251968503937, + "eval_PERSON_recall": 0.9765625, + "eval_loss": 0.053729888051748276, + "eval_overall_accuracy": 0.9819163292847504, + "eval_overall_f1": 0.918444165621079, + "eval_overall_precision": 0.9037037037037037, + "eval_overall_recall": 0.9336734693877551, + "eval_runtime": 0.2483, + "eval_samples_per_second": 684.761, + "eval_steps_per_second": 12.084, + "step": 1440 + }, + { + "epoch": 16.0, + "grad_norm": 0.6922814846038818, + "learning_rate": 4.2e-05, + "loss": 0.0567, + "step": 1536 + }, + { + "epoch": 16.0, + "eval_LOCATION_f1": 0.8743169398907104, + "eval_LOCATION_number": 86, + "eval_LOCATION_precision": 0.8247422680412371, + "eval_LOCATION_recall": 0.9302325581395349, + "eval_ORGANIZATION_f1": 0.8846153846153846, + "eval_ORGANIZATION_number": 178, + "eval_ORGANIZATION_precision": 0.8655913978494624, + "eval_ORGANIZATION_recall": 0.9044943820224719, + "eval_PERSON_f1": 0.9689922480620154, + "eval_PERSON_number": 128, + "eval_PERSON_precision": 0.9615384615384616, + "eval_PERSON_recall": 0.9765625, + "eval_loss": 0.05485454946756363, + "eval_overall_accuracy": 0.9792172739541161, + "eval_overall_f1": 0.9093167701863355, + "eval_overall_precision": 0.8861985472154964, + "eval_overall_recall": 0.9336734693877551, + "eval_runtime": 0.2497, + "eval_samples_per_second": 680.691, + "eval_steps_per_second": 12.012, + "step": 1536 + }, + { + "epoch": 17.0, + "grad_norm": 0.5154015421867371, + "learning_rate": 4.15e-05, + "loss": 0.0546, + "step": 1632 + }, + { + "epoch": 17.0, + "eval_LOCATION_f1": 0.898876404494382, + "eval_LOCATION_number": 86, + "eval_LOCATION_precision": 0.8695652173913043, + "eval_LOCATION_recall": 0.9302325581395349, + "eval_ORGANIZATION_f1": 0.8925619834710743, + "eval_ORGANIZATION_number": 178, + "eval_ORGANIZATION_precision": 0.8756756756756757, + "eval_ORGANIZATION_recall": 0.9101123595505618, + "eval_PERSON_f1": 0.9803921568627452, + "eval_PERSON_number": 128, + "eval_PERSON_precision": 0.984251968503937, + "eval_PERSON_recall": 0.9765625, + "eval_loss": 0.05112185701727867, + "eval_overall_accuracy": 0.982995951417004, + "eval_overall_f1": 0.922110552763819, + "eval_overall_precision": 0.9084158415841584, + "eval_overall_recall": 0.9362244897959183, + "eval_runtime": 0.2489, + "eval_samples_per_second": 683.073, + "eval_steps_per_second": 12.054, + "step": 1632 + }, + { + "epoch": 18.0, + "grad_norm": 0.3619915544986725, + "learning_rate": 4.1e-05, + "loss": 0.0527, + "step": 1728 + }, + { + "epoch": 18.0, + "eval_LOCATION_f1": 0.8961748633879781, + "eval_LOCATION_number": 86, + "eval_LOCATION_precision": 0.845360824742268, + "eval_LOCATION_recall": 0.9534883720930233, + "eval_ORGANIZATION_f1": 0.895184135977337, + "eval_ORGANIZATION_number": 178, + "eval_ORGANIZATION_precision": 0.9028571428571428, + "eval_ORGANIZATION_recall": 0.8876404494382022, + "eval_PERSON_f1": 0.96875, + "eval_PERSON_number": 128, + "eval_PERSON_precision": 0.96875, + "eval_PERSON_recall": 0.96875, + "eval_loss": 0.05289569869637489, + "eval_overall_accuracy": 0.9827260458839406, + "eval_overall_f1": 0.9191919191919193, + "eval_overall_precision": 0.91, + "eval_overall_recall": 0.9285714285714286, + "eval_runtime": 0.242, + "eval_samples_per_second": 702.406, + "eval_steps_per_second": 12.395, + "step": 1728 + }, + { + "epoch": 19.0, + "grad_norm": 0.6516693234443665, + "learning_rate": 4.05e-05, + "loss": 0.0492, + "step": 1824 + }, + { + "epoch": 19.0, + "eval_LOCATION_f1": 0.893854748603352, + "eval_LOCATION_number": 86, + "eval_LOCATION_precision": 0.8602150537634409, + "eval_LOCATION_recall": 0.9302325581395349, + "eval_ORGANIZATION_f1": 0.8926553672316383, + "eval_ORGANIZATION_number": 178, + "eval_ORGANIZATION_precision": 0.8977272727272727, + "eval_ORGANIZATION_recall": 0.8876404494382022, + "eval_PERSON_f1": 0.96875, + "eval_PERSON_number": 128, + "eval_PERSON_precision": 0.96875, + "eval_PERSON_recall": 0.96875, + "eval_loss": 0.05347622185945511, + "eval_overall_accuracy": 0.9821862348178138, + "eval_overall_f1": 0.917617237008872, + "eval_overall_precision": 0.9118387909319899, + "eval_overall_recall": 0.923469387755102, + "eval_runtime": 0.2422, + "eval_samples_per_second": 701.922, + "eval_steps_per_second": 12.387, + "step": 1824 + }, + { + "epoch": 20.0, + "grad_norm": 0.7471908330917358, + "learning_rate": 4e-05, + "loss": 0.0474, + "step": 1920 + }, + { + "epoch": 20.0, + "eval_LOCATION_f1": 0.8999999999999999, + "eval_LOCATION_number": 86, + "eval_LOCATION_precision": 0.8617021276595744, + "eval_LOCATION_recall": 0.9418604651162791, + "eval_ORGANIZATION_f1": 0.913649025069638, + "eval_ORGANIZATION_number": 178, + "eval_ORGANIZATION_precision": 0.9060773480662984, + "eval_ORGANIZATION_recall": 0.9213483146067416, + "eval_PERSON_f1": 0.9803921568627452, + "eval_PERSON_number": 128, + "eval_PERSON_precision": 0.984251968503937, + "eval_PERSON_recall": 0.9765625, + "eval_loss": 0.05013725161552429, + "eval_overall_accuracy": 0.9827260458839406, + "eval_overall_f1": 0.9319899244332494, + "eval_overall_precision": 0.9203980099502488, + "eval_overall_recall": 0.9438775510204082, + "eval_runtime": 0.243, + "eval_samples_per_second": 699.537, + "eval_steps_per_second": 12.345, + "step": 1920 + }, + { + "epoch": 21.0, + "grad_norm": 1.2585930824279785, + "learning_rate": 3.9500000000000005e-05, + "loss": 0.0447, + "step": 2016 + }, + { + "epoch": 21.0, + "eval_LOCATION_f1": 0.9180327868852459, + "eval_LOCATION_number": 86, + "eval_LOCATION_precision": 0.865979381443299, + "eval_LOCATION_recall": 0.9767441860465116, + "eval_ORGANIZATION_f1": 0.9044943820224719, + "eval_ORGANIZATION_number": 178, + "eval_ORGANIZATION_precision": 0.9044943820224719, + "eval_ORGANIZATION_recall": 0.9044943820224719, + "eval_PERSON_f1": 0.9803921568627452, + "eval_PERSON_number": 128, + "eval_PERSON_precision": 0.984251968503937, + "eval_PERSON_recall": 0.9765625, + "eval_loss": 0.05279025062918663, + "eval_overall_accuracy": 0.9835357624831309, + "eval_overall_f1": 0.9319899244332494, + "eval_overall_precision": 0.9203980099502488, + "eval_overall_recall": 0.9438775510204082, + "eval_runtime": 0.2473, + "eval_samples_per_second": 687.519, + "eval_steps_per_second": 12.133, + "step": 2016 + }, + { + "epoch": 22.0, + "grad_norm": 0.8391615748405457, + "learning_rate": 3.9000000000000006e-05, + "loss": 0.0453, + "step": 2112 + }, + { + "epoch": 22.0, + "eval_LOCATION_f1": 0.8829787234042553, + "eval_LOCATION_number": 86, + "eval_LOCATION_precision": 0.8137254901960784, + "eval_LOCATION_recall": 0.9651162790697675, + "eval_ORGANIZATION_f1": 0.9106145251396648, + "eval_ORGANIZATION_number": 178, + "eval_ORGANIZATION_precision": 0.9055555555555556, + "eval_ORGANIZATION_recall": 0.9157303370786517, + "eval_PERSON_f1": 0.9803921568627452, + "eval_PERSON_number": 128, + "eval_PERSON_precision": 0.984251968503937, + "eval_PERSON_recall": 0.9765625, + "eval_loss": 0.05221620574593544, + "eval_overall_accuracy": 0.9819163292847504, + "eval_overall_f1": 0.9263420724094882, + "eval_overall_precision": 0.9070904645476773, + "eval_overall_recall": 0.9464285714285714, + "eval_runtime": 0.2477, + "eval_samples_per_second": 686.445, + "eval_steps_per_second": 12.114, + "step": 2112 + }, + { + "epoch": 23.0, + "grad_norm": 0.5329034924507141, + "learning_rate": 3.85e-05, + "loss": 0.039, + "step": 2208 + }, + { + "epoch": 23.0, + "eval_LOCATION_f1": 0.8972972972972972, + "eval_LOCATION_number": 86, + "eval_LOCATION_precision": 0.8383838383838383, + "eval_LOCATION_recall": 0.9651162790697675, + "eval_ORGANIZATION_f1": 0.9121813031161473, + "eval_ORGANIZATION_number": 178, + "eval_ORGANIZATION_precision": 0.92, + "eval_ORGANIZATION_recall": 0.9044943820224719, + "eval_PERSON_f1": 0.9765625, + "eval_PERSON_number": 128, + "eval_PERSON_precision": 0.9765625, + "eval_PERSON_recall": 0.9765625, + "eval_loss": 0.04938729107379913, + "eval_overall_accuracy": 0.9835357624831309, + "eval_overall_f1": 0.929471032745592, + "eval_overall_precision": 0.917910447761194, + "eval_overall_recall": 0.9413265306122449, + "eval_runtime": 0.2413, + "eval_samples_per_second": 704.628, + "eval_steps_per_second": 12.435, + "step": 2208 + }, + { + "epoch": 24.0, + "grad_norm": 0.5503416657447815, + "learning_rate": 3.8e-05, + "loss": 0.0414, + "step": 2304 + }, + { + "epoch": 24.0, + "eval_LOCATION_f1": 0.8924731182795699, + "eval_LOCATION_number": 86, + "eval_LOCATION_precision": 0.83, + "eval_LOCATION_recall": 0.9651162790697675, + "eval_ORGANIZATION_f1": 0.906515580736544, + "eval_ORGANIZATION_number": 178, + "eval_ORGANIZATION_precision": 0.9142857142857143, + "eval_ORGANIZATION_recall": 0.898876404494382, + "eval_PERSON_f1": 0.9765625, + "eval_PERSON_number": 128, + "eval_PERSON_precision": 0.9765625, + "eval_PERSON_recall": 0.9765625, + "eval_loss": 0.0536591075360775, + "eval_overall_accuracy": 0.9819163292847504, + "eval_overall_f1": 0.9257861635220126, + "eval_overall_precision": 0.913151364764268, + "eval_overall_recall": 0.9387755102040817, + "eval_runtime": 0.248, + "eval_samples_per_second": 685.464, + "eval_steps_per_second": 12.096, + "step": 2304 + }, + { + "epoch": 25.0, + "grad_norm": 0.6665748953819275, + "learning_rate": 3.7500000000000003e-05, + "loss": 0.038, + "step": 2400 + }, + { + "epoch": 25.0, + "eval_LOCATION_f1": 0.9060773480662985, + "eval_LOCATION_number": 86, + "eval_LOCATION_precision": 0.8631578947368421, + "eval_LOCATION_recall": 0.9534883720930233, + "eval_ORGANIZATION_f1": 0.905982905982906, + "eval_ORGANIZATION_number": 178, + "eval_ORGANIZATION_precision": 0.9190751445086706, + "eval_ORGANIZATION_recall": 0.8932584269662921, + "eval_PERSON_f1": 0.9803921568627452, + "eval_PERSON_number": 128, + "eval_PERSON_precision": 0.984251968503937, + "eval_PERSON_recall": 0.9765625, + "eval_loss": 0.05137418955564499, + "eval_overall_accuracy": 0.9838056680161943, + "eval_overall_f1": 0.9301143583227446, + "eval_overall_precision": 0.9265822784810127, + "eval_overall_recall": 0.9336734693877551, + "eval_runtime": 0.248, + "eval_samples_per_second": 685.611, + "eval_steps_per_second": 12.099, + "step": 2400 + }, + { + "epoch": 26.0, + "grad_norm": 0.4098527729511261, + "learning_rate": 3.7e-05, + "loss": 0.038, + "step": 2496 + }, + { + "epoch": 26.0, + "eval_LOCATION_f1": 0.923076923076923, + "eval_LOCATION_number": 86, + "eval_LOCATION_precision": 0.875, + "eval_LOCATION_recall": 0.9767441860465116, + "eval_ORGANIZATION_f1": 0.9085714285714286, + "eval_ORGANIZATION_number": 178, + "eval_ORGANIZATION_precision": 0.9244186046511628, + "eval_ORGANIZATION_recall": 0.8932584269662921, + "eval_PERSON_f1": 0.9765625, + "eval_PERSON_number": 128, + "eval_PERSON_precision": 0.9765625, + "eval_PERSON_recall": 0.9765625, + "eval_loss": 0.0504906140267849, + "eval_overall_accuracy": 0.9832658569500675, + "eval_overall_f1": 0.934010152284264, + "eval_overall_precision": 0.9292929292929293, + "eval_overall_recall": 0.9387755102040817, + "eval_runtime": 0.2482, + "eval_samples_per_second": 685.063, + "eval_steps_per_second": 12.089, + "step": 2496 + }, + { + "epoch": 27.0, + "grad_norm": 0.31108349561691284, + "learning_rate": 3.65e-05, + "loss": 0.037, + "step": 2592 + }, + { + "epoch": 27.0, + "eval_LOCATION_f1": 0.9021739130434784, + "eval_LOCATION_number": 86, + "eval_LOCATION_precision": 0.8469387755102041, + "eval_LOCATION_recall": 0.9651162790697675, + "eval_ORGANIZATION_f1": 0.9090909090909091, + "eval_ORGANIZATION_number": 178, + "eval_ORGANIZATION_precision": 0.9195402298850575, + "eval_ORGANIZATION_recall": 0.898876404494382, + "eval_PERSON_f1": 0.9803921568627452, + "eval_PERSON_number": 128, + "eval_PERSON_precision": 0.984251968503937, + "eval_PERSON_recall": 0.9765625, + "eval_loss": 0.05084557086229324, + "eval_overall_accuracy": 0.9824561403508771, + "eval_overall_f1": 0.9304677623261693, + "eval_overall_precision": 0.9223057644110275, + "eval_overall_recall": 0.9387755102040817, + "eval_runtime": 0.2421, + "eval_samples_per_second": 702.05, + "eval_steps_per_second": 12.389, + "step": 2592 + }, + { + "epoch": 28.0, + "grad_norm": 1.1318433284759521, + "learning_rate": 3.6e-05, + "loss": 0.0375, + "step": 2688 + }, + { + "epoch": 28.0, + "eval_LOCATION_f1": 0.923076923076923, + "eval_LOCATION_number": 86, + "eval_LOCATION_precision": 0.875, + "eval_LOCATION_recall": 0.9767441860465116, + "eval_ORGANIZATION_f1": 0.9101123595505618, + "eval_ORGANIZATION_number": 178, + "eval_ORGANIZATION_precision": 0.9101123595505618, + "eval_ORGANIZATION_recall": 0.9101123595505618, + "eval_PERSON_f1": 0.9803921568627452, + "eval_PERSON_number": 128, + "eval_PERSON_precision": 0.984251968503937, + "eval_PERSON_recall": 0.9765625, + "eval_loss": 0.04823584854602814, + "eval_overall_accuracy": 0.9840755735492578, + "eval_overall_f1": 0.935687263556116, + "eval_overall_precision": 0.9251870324189526, + "eval_overall_recall": 0.9464285714285714, + "eval_runtime": 0.2486, + "eval_samples_per_second": 683.7, + "eval_steps_per_second": 12.065, + "step": 2688 + }, + { + "epoch": 29.0, + "grad_norm": 0.8521238565444946, + "learning_rate": 3.55e-05, + "loss": 0.0342, + "step": 2784 + }, + { + "epoch": 29.0, + "eval_LOCATION_f1": 0.9222222222222223, + "eval_LOCATION_number": 86, + "eval_LOCATION_precision": 0.8829787234042553, + "eval_LOCATION_recall": 0.9651162790697675, + "eval_ORGANIZATION_f1": 0.9070422535211267, + "eval_ORGANIZATION_number": 178, + "eval_ORGANIZATION_precision": 0.9096045197740112, + "eval_ORGANIZATION_recall": 0.9044943820224719, + "eval_PERSON_f1": 0.9765625, + "eval_PERSON_number": 128, + "eval_PERSON_precision": 0.9765625, + "eval_PERSON_recall": 0.9765625, + "eval_loss": 0.04844392463564873, + "eval_overall_accuracy": 0.9832658569500675, + "eval_overall_f1": 0.9329962073324906, + "eval_overall_precision": 0.924812030075188, + "eval_overall_recall": 0.9413265306122449, + "eval_runtime": 0.2486, + "eval_samples_per_second": 683.711, + "eval_steps_per_second": 12.065, + "step": 2784 + }, + { + "epoch": 30.0, + "grad_norm": 0.2059481143951416, + "learning_rate": 3.5e-05, + "loss": 0.0314, + "step": 2880 + }, + { + "epoch": 30.0, + "eval_LOCATION_f1": 0.907103825136612, + "eval_LOCATION_number": 86, + "eval_LOCATION_precision": 0.8556701030927835, + "eval_LOCATION_recall": 0.9651162790697675, + "eval_ORGANIZATION_f1": 0.9060773480662982, + "eval_ORGANIZATION_number": 178, + "eval_ORGANIZATION_precision": 0.8913043478260869, + "eval_ORGANIZATION_recall": 0.9213483146067416, + "eval_PERSON_f1": 0.980544747081712, + "eval_PERSON_number": 128, + "eval_PERSON_precision": 0.9767441860465116, + "eval_PERSON_recall": 0.984375, + "eval_loss": 0.048888128250837326, + "eval_overall_accuracy": 0.9840755735492578, + "eval_overall_f1": 0.9301745635910225, + "eval_overall_precision": 0.9097560975609756, + "eval_overall_recall": 0.951530612244898, + "eval_runtime": 0.2412, + "eval_samples_per_second": 704.721, + "eval_steps_per_second": 12.436, + "step": 2880 + }, + { + "epoch": 31.0, + "grad_norm": 0.7870827317237854, + "learning_rate": 3.45e-05, + "loss": 0.0308, + "step": 2976 + }, + { + "epoch": 31.0, + "eval_LOCATION_f1": 0.9180327868852459, + "eval_LOCATION_number": 86, + "eval_LOCATION_precision": 0.865979381443299, + "eval_LOCATION_recall": 0.9767441860465116, + "eval_ORGANIZATION_f1": 0.895184135977337, + "eval_ORGANIZATION_number": 178, + "eval_ORGANIZATION_precision": 0.9028571428571428, + "eval_ORGANIZATION_recall": 0.8876404494382022, + "eval_PERSON_f1": 0.9765625, + "eval_PERSON_number": 128, + "eval_PERSON_precision": 0.9765625, + "eval_PERSON_recall": 0.9765625, + "eval_loss": 0.052150849252939224, + "eval_overall_accuracy": 0.9824561403508771, + "eval_overall_f1": 0.9267676767676768, + "eval_overall_precision": 0.9175, + "eval_overall_recall": 0.9362244897959183, + "eval_runtime": 0.2424, + "eval_samples_per_second": 701.228, + "eval_steps_per_second": 12.375, + "step": 2976 + }, + { + "epoch": 32.0, + "grad_norm": 0.24624638259410858, + "learning_rate": 3.4000000000000007e-05, + "loss": 0.0284, + "step": 3072 + }, + { + "epoch": 32.0, + "eval_LOCATION_f1": 0.923076923076923, + "eval_LOCATION_number": 86, + "eval_LOCATION_precision": 0.875, + "eval_LOCATION_recall": 0.9767441860465116, + "eval_ORGANIZATION_f1": 0.906515580736544, + "eval_ORGANIZATION_number": 178, + "eval_ORGANIZATION_precision": 0.9142857142857143, + "eval_ORGANIZATION_recall": 0.898876404494382, + "eval_PERSON_f1": 0.9765625, + "eval_PERSON_number": 128, + "eval_PERSON_precision": 0.9765625, + "eval_PERSON_recall": 0.9765625, + "eval_loss": 0.0507497675716877, + "eval_overall_accuracy": 0.9832658569500675, + "eval_overall_f1": 0.9329962073324906, + "eval_overall_precision": 0.924812030075188, + "eval_overall_recall": 0.9413265306122449, + "eval_runtime": 0.2466, + "eval_samples_per_second": 689.376, + "eval_steps_per_second": 12.165, + "step": 3072 + }, + { + "epoch": 33.0, + "grad_norm": 0.924070417881012, + "learning_rate": 3.35e-05, + "loss": 0.029, + "step": 3168 + }, + { + "epoch": 33.0, + "eval_LOCATION_f1": 0.908108108108108, + "eval_LOCATION_number": 86, + "eval_LOCATION_precision": 0.8484848484848485, + "eval_LOCATION_recall": 0.9767441860465116, + "eval_ORGANIZATION_f1": 0.9002849002849002, + "eval_ORGANIZATION_number": 178, + "eval_ORGANIZATION_precision": 0.9132947976878613, + "eval_ORGANIZATION_recall": 0.8876404494382022, + "eval_PERSON_f1": 0.9765625, + "eval_PERSON_number": 128, + "eval_PERSON_precision": 0.9765625, + "eval_PERSON_recall": 0.9765625, + "eval_loss": 0.05186566710472107, + "eval_overall_accuracy": 0.9819163292847504, + "eval_overall_f1": 0.9267676767676768, + "eval_overall_precision": 0.9175, + "eval_overall_recall": 0.9362244897959183, + "eval_runtime": 0.2479, + "eval_samples_per_second": 685.778, + "eval_steps_per_second": 12.102, + "step": 3168 + }, + { + "epoch": 34.0, + "grad_norm": 0.6430802345275879, + "learning_rate": 3.3e-05, + "loss": 0.0273, + "step": 3264 + }, + { + "epoch": 34.0, + "eval_LOCATION_f1": 0.9180327868852459, + "eval_LOCATION_number": 86, + "eval_LOCATION_precision": 0.865979381443299, + "eval_LOCATION_recall": 0.9767441860465116, + "eval_ORGANIZATION_f1": 0.9121813031161473, + "eval_ORGANIZATION_number": 178, + "eval_ORGANIZATION_precision": 0.92, + "eval_ORGANIZATION_recall": 0.9044943820224719, + "eval_PERSON_f1": 0.9765625, + "eval_PERSON_number": 128, + "eval_PERSON_precision": 0.9765625, + "eval_PERSON_recall": 0.9765625, + "eval_loss": 0.05143400654196739, + "eval_overall_accuracy": 0.9827260458839406, + "eval_overall_f1": 0.9343434343434343, + "eval_overall_precision": 0.925, + "eval_overall_recall": 0.9438775510204082, + "eval_runtime": 0.2483, + "eval_samples_per_second": 684.535, + "eval_steps_per_second": 12.08, + "step": 3264 + }, + { + "epoch": 35.0, + "grad_norm": 0.4695442020893097, + "learning_rate": 3.2500000000000004e-05, + "loss": 0.0273, + "step": 3360 + }, + { + "epoch": 35.0, + "eval_LOCATION_f1": 0.9180327868852459, + "eval_LOCATION_number": 86, + "eval_LOCATION_precision": 0.865979381443299, + "eval_LOCATION_recall": 0.9767441860465116, + "eval_ORGANIZATION_f1": 0.9002849002849002, + "eval_ORGANIZATION_number": 178, + "eval_ORGANIZATION_precision": 0.9132947976878613, + "eval_ORGANIZATION_recall": 0.8876404494382022, + "eval_PERSON_f1": 0.9765625, + "eval_PERSON_number": 128, + "eval_PERSON_precision": 0.9765625, + "eval_PERSON_recall": 0.9765625, + "eval_loss": 0.05347920209169388, + "eval_overall_accuracy": 0.9827260458839406, + "eval_overall_f1": 0.929113924050633, + "eval_overall_precision": 0.9221105527638191, + "eval_overall_recall": 0.9362244897959183, + "eval_runtime": 0.2467, + "eval_samples_per_second": 689.161, + "eval_steps_per_second": 12.162, + "step": 3360 + }, + { + "epoch": 36.0, + "grad_norm": 0.5556744933128357, + "learning_rate": 3.2000000000000005e-05, + "loss": 0.0275, + "step": 3456 + }, + { + "epoch": 36.0, + "eval_LOCATION_f1": 0.8972972972972972, + "eval_LOCATION_number": 86, + "eval_LOCATION_precision": 0.8383838383838383, + "eval_LOCATION_recall": 0.9651162790697675, + "eval_ORGANIZATION_f1": 0.8969359331476323, + "eval_ORGANIZATION_number": 178, + "eval_ORGANIZATION_precision": 0.8895027624309392, + "eval_ORGANIZATION_recall": 0.9044943820224719, + "eval_PERSON_f1": 0.9689922480620154, + "eval_PERSON_number": 128, + "eval_PERSON_precision": 0.9615384615384616, + "eval_PERSON_recall": 0.9765625, + "eval_loss": 0.05185906961560249, + "eval_overall_accuracy": 0.9813765182186235, + "eval_overall_f1": 0.9201995012468829, + "eval_overall_precision": 0.9, + "eval_overall_recall": 0.9413265306122449, + "eval_runtime": 0.2496, + "eval_samples_per_second": 681.016, + "eval_steps_per_second": 12.018, + "step": 3456 + }, + { + "epoch": 37.0, + "grad_norm": 1.0498524904251099, + "learning_rate": 3.15e-05, + "loss": 0.0244, + "step": 3552 + }, + { + "epoch": 37.0, + "eval_LOCATION_f1": 0.8901098901098902, + "eval_LOCATION_number": 86, + "eval_LOCATION_precision": 0.84375, + "eval_LOCATION_recall": 0.9418604651162791, + "eval_ORGANIZATION_f1": 0.898876404494382, + "eval_ORGANIZATION_number": 178, + "eval_ORGANIZATION_precision": 0.898876404494382, + "eval_ORGANIZATION_recall": 0.898876404494382, + "eval_PERSON_f1": 0.9765625, + "eval_PERSON_number": 128, + "eval_PERSON_precision": 0.9765625, + "eval_PERSON_recall": 0.9765625, + "eval_loss": 0.05113949626684189, + "eval_overall_accuracy": 0.9819163292847504, + "eval_overall_f1": 0.9219143576826196, + "eval_overall_precision": 0.9104477611940298, + "eval_overall_recall": 0.9336734693877551, + "eval_runtime": 0.2473, + "eval_samples_per_second": 687.363, + "eval_steps_per_second": 12.13, + "step": 3552 + }, + { + "epoch": 38.0, + "grad_norm": 0.8411348462104797, + "learning_rate": 3.1e-05, + "loss": 0.0272, + "step": 3648 + }, + { + "epoch": 38.0, + "eval_LOCATION_f1": 0.9180327868852459, + "eval_LOCATION_number": 86, + "eval_LOCATION_precision": 0.865979381443299, + "eval_LOCATION_recall": 0.9767441860465116, + "eval_ORGANIZATION_f1": 0.895184135977337, + "eval_ORGANIZATION_number": 178, + "eval_ORGANIZATION_precision": 0.9028571428571428, + "eval_ORGANIZATION_recall": 0.8876404494382022, + "eval_PERSON_f1": 0.9765625, + "eval_PERSON_number": 128, + "eval_PERSON_precision": 0.9765625, + "eval_PERSON_recall": 0.9765625, + "eval_loss": 0.05628972873091698, + "eval_overall_accuracy": 0.9827260458839406, + "eval_overall_f1": 0.9267676767676768, + "eval_overall_precision": 0.9175, + "eval_overall_recall": 0.9362244897959183, + "eval_runtime": 0.2479, + "eval_samples_per_second": 685.727, + "eval_steps_per_second": 12.101, + "step": 3648 + }, + { + "epoch": 39.0, + "grad_norm": 0.41838157176971436, + "learning_rate": 3.05e-05, + "loss": 0.028, + "step": 3744 + }, + { + "epoch": 39.0, + "eval_LOCATION_f1": 0.923076923076923, + "eval_LOCATION_number": 86, + "eval_LOCATION_precision": 0.875, + "eval_LOCATION_recall": 0.9767441860465116, + "eval_ORGANIZATION_f1": 0.9101123595505618, + "eval_ORGANIZATION_number": 178, + "eval_ORGANIZATION_precision": 0.9101123595505618, + "eval_ORGANIZATION_recall": 0.9101123595505618, + "eval_PERSON_f1": 0.9765625, + "eval_PERSON_number": 128, + "eval_PERSON_precision": 0.9765625, + "eval_PERSON_recall": 0.9765625, + "eval_loss": 0.051703643053770065, + "eval_overall_accuracy": 0.9835357624831309, + "eval_overall_f1": 0.9345088161209069, + "eval_overall_precision": 0.9228855721393034, + "eval_overall_recall": 0.9464285714285714, + "eval_runtime": 0.2477, + "eval_samples_per_second": 686.448, + "eval_steps_per_second": 12.114, + "step": 3744 + }, + { + "epoch": 40.0, + "grad_norm": 0.9456387758255005, + "learning_rate": 3e-05, + "loss": 0.0236, + "step": 3840 + }, + { + "epoch": 40.0, + "eval_LOCATION_f1": 0.9111111111111112, + "eval_LOCATION_number": 86, + "eval_LOCATION_precision": 0.8723404255319149, + "eval_LOCATION_recall": 0.9534883720930233, + "eval_ORGANIZATION_f1": 0.9106145251396648, + "eval_ORGANIZATION_number": 178, + "eval_ORGANIZATION_precision": 0.9055555555555556, + "eval_ORGANIZATION_recall": 0.9157303370786517, + "eval_PERSON_f1": 0.9765625, + "eval_PERSON_number": 128, + "eval_PERSON_precision": 0.9765625, + "eval_PERSON_recall": 0.9765625, + "eval_loss": 0.05246607959270477, + "eval_overall_accuracy": 0.9832658569500675, + "eval_overall_f1": 0.9319899244332494, + "eval_overall_precision": 0.9203980099502488, + "eval_overall_recall": 0.9438775510204082, + "eval_runtime": 0.2488, + "eval_samples_per_second": 683.399, + "eval_steps_per_second": 12.06, + "step": 3840 + }, + { + "epoch": 41.0, + "grad_norm": 0.31706151366233826, + "learning_rate": 2.95e-05, + "loss": 0.0223, + "step": 3936 + }, + { + "epoch": 41.0, + "eval_LOCATION_f1": 0.9152542372881357, + "eval_LOCATION_number": 86, + "eval_LOCATION_precision": 0.8901098901098901, + "eval_LOCATION_recall": 0.9418604651162791, + "eval_ORGANIZATION_f1": 0.9157303370786517, + "eval_ORGANIZATION_number": 178, + "eval_ORGANIZATION_precision": 0.9157303370786517, + "eval_ORGANIZATION_recall": 0.9157303370786517, + "eval_PERSON_f1": 0.9765625, + "eval_PERSON_number": 128, + "eval_PERSON_precision": 0.9765625, + "eval_PERSON_recall": 0.9765625, + "eval_loss": 0.054678838700056076, + "eval_overall_accuracy": 0.9835357624831309, + "eval_overall_f1": 0.935361216730038, + "eval_overall_precision": 0.929471032745592, + "eval_overall_recall": 0.9413265306122449, + "eval_runtime": 0.2485, + "eval_samples_per_second": 684.014, + "eval_steps_per_second": 12.071, + "step": 3936 + }, + { + "epoch": 42.0, + "grad_norm": 0.6095432639122009, + "learning_rate": 2.9e-05, + "loss": 0.0228, + "step": 4032 + }, + { + "epoch": 42.0, + "eval_LOCATION_f1": 0.9101123595505618, + "eval_LOCATION_number": 86, + "eval_LOCATION_precision": 0.8804347826086957, + "eval_LOCATION_recall": 0.9418604651162791, + "eval_ORGANIZATION_f1": 0.9111111111111112, + "eval_ORGANIZATION_number": 178, + "eval_ORGANIZATION_precision": 0.9010989010989011, + "eval_ORGANIZATION_recall": 0.9213483146067416, + "eval_PERSON_f1": 0.9765625, + "eval_PERSON_number": 128, + "eval_PERSON_precision": 0.9765625, + "eval_PERSON_recall": 0.9765625, + "eval_loss": 0.05048713833093643, + "eval_overall_accuracy": 0.9840755735492578, + "eval_overall_f1": 0.9319899244332494, + "eval_overall_precision": 0.9203980099502488, + "eval_overall_recall": 0.9438775510204082, + "eval_runtime": 0.2475, + "eval_samples_per_second": 686.767, + "eval_steps_per_second": 12.119, + "step": 4032 + }, + { + "epoch": 43.0, + "grad_norm": 0.06603565812110901, + "learning_rate": 2.8499999999999998e-05, + "loss": 0.0233, + "step": 4128 + }, + { + "epoch": 43.0, + "eval_LOCATION_f1": 0.9120879120879122, + "eval_LOCATION_number": 86, + "eval_LOCATION_precision": 0.8645833333333334, + "eval_LOCATION_recall": 0.9651162790697675, + "eval_ORGANIZATION_f1": 0.913649025069638, + "eval_ORGANIZATION_number": 178, + "eval_ORGANIZATION_precision": 0.9060773480662984, + "eval_ORGANIZATION_recall": 0.9213483146067416, + "eval_PERSON_f1": 0.9765625, + "eval_PERSON_number": 128, + "eval_PERSON_precision": 0.9765625, + "eval_PERSON_recall": 0.9765625, + "eval_loss": 0.0556485578417778, + "eval_overall_accuracy": 0.9835357624831309, + "eval_overall_f1": 0.9335006273525721, + "eval_overall_precision": 0.9185185185185185, + "eval_overall_recall": 0.9489795918367347, + "eval_runtime": 0.2475, + "eval_samples_per_second": 686.995, + "eval_steps_per_second": 12.123, + "step": 4128 + }, + { + "epoch": 44.0, + "grad_norm": 0.4065522253513336, + "learning_rate": 2.8000000000000003e-05, + "loss": 0.0219, + "step": 4224 + }, + { + "epoch": 44.0, + "eval_LOCATION_f1": 0.9130434782608695, + "eval_LOCATION_number": 86, + "eval_LOCATION_precision": 0.8571428571428571, + "eval_LOCATION_recall": 0.9767441860465116, + "eval_ORGANIZATION_f1": 0.9035812672176309, + "eval_ORGANIZATION_number": 178, + "eval_ORGANIZATION_precision": 0.8864864864864865, + "eval_ORGANIZATION_recall": 0.9213483146067416, + "eval_PERSON_f1": 0.9727626459143969, + "eval_PERSON_number": 128, + "eval_PERSON_precision": 0.9689922480620154, + "eval_PERSON_recall": 0.9765625, + "eval_loss": 0.0528954342007637, + "eval_overall_accuracy": 0.9835357624831309, + "eval_overall_f1": 0.927860696517413, + "eval_overall_precision": 0.9053398058252428, + "eval_overall_recall": 0.951530612244898, + "eval_runtime": 0.249, + "eval_samples_per_second": 682.73, + "eval_steps_per_second": 12.048, + "step": 4224 + }, + { + "epoch": 45.0, + "grad_norm": 0.8652843832969666, + "learning_rate": 2.7500000000000004e-05, + "loss": 0.0194, + "step": 4320 + }, + { + "epoch": 45.0, + "eval_LOCATION_f1": 0.9180327868852459, + "eval_LOCATION_number": 86, + "eval_LOCATION_precision": 0.865979381443299, + "eval_LOCATION_recall": 0.9767441860465116, + "eval_ORGANIZATION_f1": 0.9090909090909091, + "eval_ORGANIZATION_number": 178, + "eval_ORGANIZATION_precision": 0.9195402298850575, + "eval_ORGANIZATION_recall": 0.898876404494382, + "eval_PERSON_f1": 0.9765625, + "eval_PERSON_number": 128, + "eval_PERSON_precision": 0.9765625, + "eval_PERSON_recall": 0.9765625, + "eval_loss": 0.05724570155143738, + "eval_overall_accuracy": 0.9819163292847504, + "eval_overall_f1": 0.9329962073324906, + "eval_overall_precision": 0.924812030075188, + "eval_overall_recall": 0.9413265306122449, + "eval_runtime": 0.248, + "eval_samples_per_second": 685.477, + "eval_steps_per_second": 12.097, + "step": 4320 + }, + { + "epoch": 46.0, + "grad_norm": 0.5592328906059265, + "learning_rate": 2.7000000000000002e-05, + "loss": 0.0212, + "step": 4416 + }, + { + "epoch": 46.0, + "eval_LOCATION_f1": 0.8961748633879781, + "eval_LOCATION_number": 86, + "eval_LOCATION_precision": 0.845360824742268, + "eval_LOCATION_recall": 0.9534883720930233, + "eval_ORGANIZATION_f1": 0.9090909090909092, + "eval_ORGANIZATION_number": 178, + "eval_ORGANIZATION_precision": 0.8918918918918919, + "eval_ORGANIZATION_recall": 0.9269662921348315, + "eval_PERSON_f1": 0.9689922480620154, + "eval_PERSON_number": 128, + "eval_PERSON_precision": 0.9615384615384616, + "eval_PERSON_recall": 0.9765625, + "eval_loss": 0.05504947155714035, + "eval_overall_accuracy": 0.9827260458839406, + "eval_overall_f1": 0.9253731343283583, + "eval_overall_precision": 0.9029126213592233, + "eval_overall_recall": 0.9489795918367347, + "eval_runtime": 0.2477, + "eval_samples_per_second": 686.443, + "eval_steps_per_second": 12.114, + "step": 4416 + }, + { + "epoch": 47.0, + "grad_norm": 0.38780418038368225, + "learning_rate": 2.6500000000000004e-05, + "loss": 0.019, + "step": 4512 + }, + { + "epoch": 47.0, + "eval_LOCATION_f1": 0.888888888888889, + "eval_LOCATION_number": 86, + "eval_LOCATION_precision": 0.851063829787234, + "eval_LOCATION_recall": 0.9302325581395349, + "eval_ORGANIZATION_f1": 0.9116022099447514, + "eval_ORGANIZATION_number": 178, + "eval_ORGANIZATION_precision": 0.8967391304347826, + "eval_ORGANIZATION_recall": 0.9269662921348315, + "eval_PERSON_f1": 0.9727626459143969, + "eval_PERSON_number": 128, + "eval_PERSON_precision": 0.9689922480620154, + "eval_PERSON_recall": 0.9765625, + "eval_loss": 0.051470689475536346, + "eval_overall_accuracy": 0.9843454790823212, + "eval_overall_f1": 0.9261576971214017, + "eval_overall_precision": 0.9090909090909091, + "eval_overall_recall": 0.9438775510204082, + "eval_runtime": 0.2483, + "eval_samples_per_second": 684.586, + "eval_steps_per_second": 12.081, + "step": 4512 + }, + { + "epoch": 48.0, + "grad_norm": 0.4960862994194031, + "learning_rate": 2.6000000000000002e-05, + "loss": 0.0182, + "step": 4608 + }, + { + "epoch": 48.0, + "eval_LOCATION_f1": 0.9060773480662985, + "eval_LOCATION_number": 86, + "eval_LOCATION_precision": 0.8631578947368421, + "eval_LOCATION_recall": 0.9534883720930233, + "eval_ORGANIZATION_f1": 0.9080779944289694, + "eval_ORGANIZATION_number": 178, + "eval_ORGANIZATION_precision": 0.9005524861878453, + "eval_ORGANIZATION_recall": 0.9157303370786517, + "eval_PERSON_f1": 0.9765625, + "eval_PERSON_number": 128, + "eval_PERSON_precision": 0.9765625, + "eval_PERSON_recall": 0.9765625, + "eval_loss": 0.05240018665790558, + "eval_overall_accuracy": 0.9840755735492578, + "eval_overall_f1": 0.9296482412060302, + "eval_overall_precision": 0.9158415841584159, + "eval_overall_recall": 0.9438775510204082, + "eval_runtime": 0.2469, + "eval_samples_per_second": 688.445, + "eval_steps_per_second": 12.149, + "step": 4608 + }, + { + "epoch": 49.0, + "grad_norm": 0.2815532386302948, + "learning_rate": 2.5500000000000003e-05, + "loss": 0.0185, + "step": 4704 + }, + { + "epoch": 49.0, + "eval_LOCATION_f1": 0.9222222222222223, + "eval_LOCATION_number": 86, + "eval_LOCATION_precision": 0.8829787234042553, + "eval_LOCATION_recall": 0.9651162790697675, + "eval_ORGANIZATION_f1": 0.9096045197740114, + "eval_ORGANIZATION_number": 178, + "eval_ORGANIZATION_precision": 0.9147727272727273, + "eval_ORGANIZATION_recall": 0.9044943820224719, + "eval_PERSON_f1": 0.9765625, + "eval_PERSON_number": 128, + "eval_PERSON_precision": 0.9765625, + "eval_PERSON_recall": 0.9765625, + "eval_loss": 0.052264876663684845, + "eval_overall_accuracy": 0.9843454790823212, + "eval_overall_f1": 0.9341772151898734, + "eval_overall_precision": 0.9271356783919598, + "eval_overall_recall": 0.9413265306122449, + "eval_runtime": 0.241, + "eval_samples_per_second": 705.288, + "eval_steps_per_second": 12.446, + "step": 4704 + }, + { + "epoch": 50.0, + "grad_norm": 0.5738477110862732, + "learning_rate": 2.5e-05, + "loss": 0.0181, + "step": 4800 + }, + { + "epoch": 50.0, + "eval_LOCATION_f1": 0.923076923076923, + "eval_LOCATION_number": 86, + "eval_LOCATION_precision": 0.875, + "eval_LOCATION_recall": 0.9767441860465116, + "eval_ORGANIZATION_f1": 0.9050279329608939, + "eval_ORGANIZATION_number": 178, + "eval_ORGANIZATION_precision": 0.9, + "eval_ORGANIZATION_recall": 0.9101123595505618, + "eval_PERSON_f1": 0.9765625, + "eval_PERSON_number": 128, + "eval_PERSON_precision": 0.9765625, + "eval_PERSON_recall": 0.9765625, + "eval_loss": 0.053917378187179565, + "eval_overall_accuracy": 0.9838056680161943, + "eval_overall_f1": 0.9321608040201005, + "eval_overall_precision": 0.9183168316831684, + "eval_overall_recall": 0.9464285714285714, + "eval_runtime": 0.2469, + "eval_samples_per_second": 688.451, + "eval_steps_per_second": 12.149, + "step": 4800 + }, + { + "epoch": 51.0, + "grad_norm": 0.5479133129119873, + "learning_rate": 2.45e-05, + "loss": 0.0181, + "step": 4896 + }, + { + "epoch": 51.0, + "eval_LOCATION_f1": 0.9273743016759777, + "eval_LOCATION_number": 86, + "eval_LOCATION_precision": 0.8924731182795699, + "eval_LOCATION_recall": 0.9651162790697675, + "eval_ORGANIZATION_f1": 0.9152542372881356, + "eval_ORGANIZATION_number": 178, + "eval_ORGANIZATION_precision": 0.9204545454545454, + "eval_ORGANIZATION_recall": 0.9101123595505618, + "eval_PERSON_f1": 0.9765625, + "eval_PERSON_number": 128, + "eval_PERSON_precision": 0.9765625, + "eval_PERSON_recall": 0.9765625, + "eval_loss": 0.05642145872116089, + "eval_overall_accuracy": 0.9840755735492578, + "eval_overall_f1": 0.9378960709759189, + "eval_overall_precision": 0.9319899244332494, + "eval_overall_recall": 0.9438775510204082, + "eval_runtime": 0.2486, + "eval_samples_per_second": 683.917, + "eval_steps_per_second": 12.069, + "step": 4896 + }, + { + "epoch": 52.0, + "grad_norm": 0.19654129445552826, + "learning_rate": 2.4e-05, + "loss": 0.0177, + "step": 4992 + }, + { + "epoch": 52.0, + "eval_LOCATION_f1": 0.9222222222222223, + "eval_LOCATION_number": 86, + "eval_LOCATION_precision": 0.8829787234042553, + "eval_LOCATION_recall": 0.9651162790697675, + "eval_ORGANIZATION_f1": 0.9235127478753541, + "eval_ORGANIZATION_number": 178, + "eval_ORGANIZATION_precision": 0.9314285714285714, + "eval_ORGANIZATION_recall": 0.9157303370786517, + "eval_PERSON_f1": 0.9765625, + "eval_PERSON_number": 128, + "eval_PERSON_precision": 0.9765625, + "eval_PERSON_recall": 0.9765625, + "eval_loss": 0.058501098304986954, + "eval_overall_accuracy": 0.9835357624831309, + "eval_overall_f1": 0.9404309252217997, + "eval_overall_precision": 0.9345088161209067, + "eval_overall_recall": 0.9464285714285714, + "eval_runtime": 0.2483, + "eval_samples_per_second": 684.607, + "eval_steps_per_second": 12.081, + "step": 4992 + }, + { + "epoch": 53.0, + "grad_norm": 1.0167557001113892, + "learning_rate": 2.35e-05, + "loss": 0.0168, + "step": 5088 + }, + { + "epoch": 53.0, + "eval_LOCATION_f1": 0.8791208791208791, + "eval_LOCATION_number": 86, + "eval_LOCATION_precision": 0.8333333333333334, + "eval_LOCATION_recall": 0.9302325581395349, + "eval_ORGANIZATION_f1": 0.9095890410958904, + "eval_ORGANIZATION_number": 178, + "eval_ORGANIZATION_precision": 0.8877005347593583, + "eval_ORGANIZATION_recall": 0.9325842696629213, + "eval_PERSON_f1": 0.9727626459143969, + "eval_PERSON_number": 128, + "eval_PERSON_precision": 0.9689922480620154, + "eval_PERSON_recall": 0.9765625, + "eval_loss": 0.054052095860242844, + "eval_overall_accuracy": 0.9835357624831309, + "eval_overall_f1": 0.9228855721393036, + "eval_overall_precision": 0.9004854368932039, + "eval_overall_recall": 0.9464285714285714, + "eval_runtime": 0.247, + "eval_samples_per_second": 688.128, + "eval_steps_per_second": 12.143, + "step": 5088 + }, + { + "epoch": 54.0, + "grad_norm": 1.0977023839950562, + "learning_rate": 2.3000000000000003e-05, + "loss": 0.0172, + "step": 5184 + }, + { + "epoch": 54.0, + "eval_LOCATION_f1": 0.898876404494382, + "eval_LOCATION_number": 86, + "eval_LOCATION_precision": 0.8695652173913043, + "eval_LOCATION_recall": 0.9302325581395349, + "eval_ORGANIZATION_f1": 0.915068493150685, + "eval_ORGANIZATION_number": 178, + "eval_ORGANIZATION_precision": 0.893048128342246, + "eval_ORGANIZATION_recall": 0.9382022471910112, + "eval_PERSON_f1": 0.9727626459143969, + "eval_PERSON_number": 128, + "eval_PERSON_precision": 0.9689922480620154, + "eval_PERSON_recall": 0.9765625, + "eval_loss": 0.051402755081653595, + "eval_overall_accuracy": 0.9846153846153847, + "eval_overall_f1": 0.9300000000000002, + "eval_overall_precision": 0.9117647058823529, + "eval_overall_recall": 0.9489795918367347, + "eval_runtime": 0.2481, + "eval_samples_per_second": 685.29, + "eval_steps_per_second": 12.093, + "step": 5184 + }, + { + "epoch": 55.0, + "grad_norm": 0.5542155504226685, + "learning_rate": 2.25e-05, + "loss": 0.0165, + "step": 5280 + }, + { + "epoch": 55.0, + "eval_LOCATION_f1": 0.9325842696629213, + "eval_LOCATION_number": 86, + "eval_LOCATION_precision": 0.9021739130434783, + "eval_LOCATION_recall": 0.9651162790697675, + "eval_ORGANIZATION_f1": 0.9162011173184358, + "eval_ORGANIZATION_number": 178, + "eval_ORGANIZATION_precision": 0.9111111111111111, + "eval_ORGANIZATION_recall": 0.9213483146067416, + "eval_PERSON_f1": 0.9803921568627452, + "eval_PERSON_number": 128, + "eval_PERSON_precision": 0.984251968503937, + "eval_PERSON_recall": 0.9765625, + "eval_loss": 0.05044760927557945, + "eval_overall_accuracy": 0.9854251012145749, + "eval_overall_f1": 0.9405815423514539, + "eval_overall_precision": 0.9323308270676691, + "eval_overall_recall": 0.9489795918367347, + "eval_runtime": 0.2414, + "eval_samples_per_second": 704.273, + "eval_steps_per_second": 12.428, + "step": 5280 + }, + { + "epoch": 56.0, + "grad_norm": 0.8049263954162598, + "learning_rate": 2.2000000000000003e-05, + "loss": 0.017, + "step": 5376 + }, + { + "epoch": 56.0, + "eval_LOCATION_f1": 0.9222222222222223, + "eval_LOCATION_number": 86, + "eval_LOCATION_precision": 0.8829787234042553, + "eval_LOCATION_recall": 0.9651162790697675, + "eval_ORGANIZATION_f1": 0.9256198347107438, + "eval_ORGANIZATION_number": 178, + "eval_ORGANIZATION_precision": 0.9081081081081082, + "eval_ORGANIZATION_recall": 0.9438202247191011, + "eval_PERSON_f1": 0.9765625, + "eval_PERSON_number": 128, + "eval_PERSON_precision": 0.9765625, + "eval_PERSON_recall": 0.9765625, + "eval_loss": 0.051421601325273514, + "eval_overall_accuracy": 0.9851551956815114, + "eval_overall_f1": 0.9411764705882352, + "eval_overall_precision": 0.9238329238329238, + "eval_overall_recall": 0.9591836734693877, + "eval_runtime": 0.2479, + "eval_samples_per_second": 685.841, + "eval_steps_per_second": 12.103, + "step": 5376 + }, + { + "epoch": 57.0, + "grad_norm": 0.5076385736465454, + "learning_rate": 2.15e-05, + "loss": 0.0153, + "step": 5472 + }, + { + "epoch": 57.0, + "eval_LOCATION_f1": 0.9171270718232045, + "eval_LOCATION_number": 86, + "eval_LOCATION_precision": 0.8736842105263158, + "eval_LOCATION_recall": 0.9651162790697675, + "eval_ORGANIZATION_f1": 0.901639344262295, + "eval_ORGANIZATION_number": 178, + "eval_ORGANIZATION_precision": 0.8776595744680851, + "eval_ORGANIZATION_recall": 0.9269662921348315, + "eval_PERSON_f1": 0.9765625, + "eval_PERSON_number": 128, + "eval_PERSON_precision": 0.9765625, + "eval_PERSON_recall": 0.9765625, + "eval_loss": 0.05126916617155075, + "eval_overall_accuracy": 0.9865047233468286, + "eval_overall_f1": 0.9290161892901618, + "eval_overall_precision": 0.9075425790754258, + "eval_overall_recall": 0.951530612244898, + "eval_runtime": 0.2476, + "eval_samples_per_second": 686.651, + "eval_steps_per_second": 12.117, + "step": 5472 + }, + { + "epoch": 58.0, + "grad_norm": 0.2657437324523926, + "learning_rate": 2.1e-05, + "loss": 0.0157, + "step": 5568 + }, + { + "epoch": 58.0, + "eval_LOCATION_f1": 0.9333333333333332, + "eval_LOCATION_number": 86, + "eval_LOCATION_precision": 0.8936170212765957, + "eval_LOCATION_recall": 0.9767441860465116, + "eval_ORGANIZATION_f1": 0.9090909090909091, + "eval_ORGANIZATION_number": 178, + "eval_ORGANIZATION_precision": 0.9195402298850575, + "eval_ORGANIZATION_recall": 0.898876404494382, + "eval_PERSON_f1": 0.9765625, + "eval_PERSON_number": 128, + "eval_PERSON_precision": 0.9765625, + "eval_PERSON_recall": 0.9765625, + "eval_loss": 0.05684962496161461, + "eval_overall_accuracy": 0.9840755735492578, + "eval_overall_f1": 0.9365482233502538, + "eval_overall_precision": 0.9318181818181818, + "eval_overall_recall": 0.9413265306122449, + "eval_runtime": 0.2471, + "eval_samples_per_second": 688.109, + "eval_steps_per_second": 12.143, + "step": 5568 + }, + { + "epoch": 59.0, + "grad_norm": 0.2525281310081482, + "learning_rate": 2.05e-05, + "loss": 0.0139, + "step": 5664 + }, + { + "epoch": 59.0, + "eval_LOCATION_f1": 0.9257142857142857, + "eval_LOCATION_number": 86, + "eval_LOCATION_precision": 0.9101123595505618, + "eval_LOCATION_recall": 0.9418604651162791, + "eval_ORGANIZATION_f1": 0.8994413407821229, + "eval_ORGANIZATION_number": 178, + "eval_ORGANIZATION_precision": 0.8944444444444445, + "eval_ORGANIZATION_recall": 0.9044943820224719, + "eval_PERSON_f1": 0.9765625, + "eval_PERSON_number": 128, + "eval_PERSON_precision": 0.9765625, + "eval_PERSON_recall": 0.9765625, + "eval_loss": 0.05300280079245567, + "eval_overall_accuracy": 0.9838056680161943, + "eval_overall_f1": 0.9302915082382762, + "eval_overall_precision": 0.924433249370277, + "eval_overall_recall": 0.9362244897959183, + "eval_runtime": 0.2478, + "eval_samples_per_second": 686.144, + "eval_steps_per_second": 12.108, + "step": 5664 + }, + { + "epoch": 60.0, + "grad_norm": 0.33274856209754944, + "learning_rate": 2e-05, + "loss": 0.0156, + "step": 5760 + }, + { + "epoch": 60.0, + "eval_LOCATION_f1": 0.9120879120879122, + "eval_LOCATION_number": 86, + "eval_LOCATION_precision": 0.8645833333333334, + "eval_LOCATION_recall": 0.9651162790697675, + "eval_ORGANIZATION_f1": 0.8950276243093923, + "eval_ORGANIZATION_number": 178, + "eval_ORGANIZATION_precision": 0.8804347826086957, + "eval_ORGANIZATION_recall": 0.9101123595505618, + "eval_PERSON_f1": 0.9727626459143969, + "eval_PERSON_number": 128, + "eval_PERSON_precision": 0.9689922480620154, + "eval_PERSON_recall": 0.9765625, + "eval_loss": 0.05184905230998993, + "eval_overall_accuracy": 0.9840755735492578, + "eval_overall_f1": 0.9238451935081149, + "eval_overall_precision": 0.9046454767726161, + "eval_overall_recall": 0.9438775510204082, + "eval_runtime": 0.2473, + "eval_samples_per_second": 687.413, + "eval_steps_per_second": 12.131, + "step": 5760 + }, + { + "epoch": 61.0, + "grad_norm": 0.16549794375896454, + "learning_rate": 1.9500000000000003e-05, + "loss": 0.0134, + "step": 5856 + }, + { + "epoch": 61.0, + "eval_LOCATION_f1": 0.9257142857142857, + "eval_LOCATION_number": 86, + "eval_LOCATION_precision": 0.9101123595505618, + "eval_LOCATION_recall": 0.9418604651162791, + "eval_ORGANIZATION_f1": 0.9201101928374654, + "eval_ORGANIZATION_number": 178, + "eval_ORGANIZATION_precision": 0.9027027027027027, + "eval_ORGANIZATION_recall": 0.9382022471910112, + "eval_PERSON_f1": 0.9727626459143969, + "eval_PERSON_number": 128, + "eval_PERSON_precision": 0.9689922480620154, + "eval_PERSON_recall": 0.9765625, + "eval_loss": 0.0490739643573761, + "eval_overall_accuracy": 0.9851551956815114, + "eval_overall_f1": 0.9383647798742139, + "eval_overall_precision": 0.9255583126550868, + "eval_overall_recall": 0.951530612244898, + "eval_runtime": 0.249, + "eval_samples_per_second": 682.822, + "eval_steps_per_second": 12.05, + "step": 5856 + }, + { + "epoch": 62.0, + "grad_norm": 0.5557262897491455, + "learning_rate": 1.9e-05, + "loss": 0.0128, + "step": 5952 + }, + { + "epoch": 62.0, + "eval_LOCATION_f1": 0.9213483146067417, + "eval_LOCATION_number": 86, + "eval_LOCATION_precision": 0.8913043478260869, + "eval_LOCATION_recall": 0.9534883720930233, + "eval_ORGANIZATION_f1": 0.8986301369863013, + "eval_ORGANIZATION_number": 178, + "eval_ORGANIZATION_precision": 0.8770053475935828, + "eval_ORGANIZATION_recall": 0.9213483146067416, + "eval_PERSON_f1": 0.9727626459143969, + "eval_PERSON_number": 128, + "eval_PERSON_precision": 0.9689922480620154, + "eval_PERSON_recall": 0.9765625, + "eval_loss": 0.051324497908353806, + "eval_overall_accuracy": 0.9843454790823212, + "eval_overall_f1": 0.9275, + "eval_overall_precision": 0.9093137254901961, + "eval_overall_recall": 0.9464285714285714, + "eval_runtime": 0.2482, + "eval_samples_per_second": 684.886, + "eval_steps_per_second": 12.086, + "step": 5952 + }, + { + "epoch": 63.0, + "grad_norm": 1.3744993209838867, + "learning_rate": 1.85e-05, + "loss": 0.0126, + "step": 6048 + }, + { + "epoch": 63.0, + "eval_LOCATION_f1": 0.9378531073446328, + "eval_LOCATION_number": 86, + "eval_LOCATION_precision": 0.9120879120879121, + "eval_LOCATION_recall": 0.9651162790697675, + "eval_ORGANIZATION_f1": 0.9090909090909092, + "eval_ORGANIZATION_number": 178, + "eval_ORGANIZATION_precision": 0.8918918918918919, + "eval_ORGANIZATION_recall": 0.9269662921348315, + "eval_PERSON_f1": 0.9727626459143969, + "eval_PERSON_number": 128, + "eval_PERSON_precision": 0.9689922480620154, + "eval_PERSON_recall": 0.9765625, + "eval_loss": 0.05115790665149689, + "eval_overall_accuracy": 0.9854251012145749, + "eval_overall_f1": 0.9360100376411542, + "eval_overall_precision": 0.9209876543209876, + "eval_overall_recall": 0.951530612244898, + "eval_runtime": 0.2472, + "eval_samples_per_second": 687.609, + "eval_steps_per_second": 12.134, + "step": 6048 + }, + { + "epoch": 64.0, + "grad_norm": 0.0732036605477333, + "learning_rate": 1.8e-05, + "loss": 0.0134, + "step": 6144 + }, + { + "epoch": 64.0, + "eval_LOCATION_f1": 0.9265536723163842, + "eval_LOCATION_number": 86, + "eval_LOCATION_precision": 0.9010989010989011, + "eval_LOCATION_recall": 0.9534883720930233, + "eval_ORGANIZATION_f1": 0.9055555555555556, + "eval_ORGANIZATION_number": 178, + "eval_ORGANIZATION_precision": 0.8956043956043956, + "eval_ORGANIZATION_recall": 0.9157303370786517, + "eval_PERSON_f1": 0.9765625, + "eval_PERSON_number": 128, + "eval_PERSON_precision": 0.9765625, + "eval_PERSON_recall": 0.9765625, + "eval_loss": 0.05195976048707962, + "eval_overall_accuracy": 0.9843454790823212, + "eval_overall_f1": 0.9331651954602775, + "eval_overall_precision": 0.9226932668329177, + "eval_overall_recall": 0.9438775510204082, + "eval_runtime": 0.2481, + "eval_samples_per_second": 685.08, + "eval_steps_per_second": 12.09, + "step": 6144 + }, + { + "epoch": 65.0, + "grad_norm": 0.16457298398017883, + "learning_rate": 1.75e-05, + "loss": 0.0128, + "step": 6240 + }, + { + "epoch": 65.0, + "eval_LOCATION_f1": 0.9325842696629213, + "eval_LOCATION_number": 86, + "eval_LOCATION_precision": 0.9021739130434783, + "eval_LOCATION_recall": 0.9651162790697675, + "eval_ORGANIZATION_f1": 0.9046321525885558, + "eval_ORGANIZATION_number": 178, + "eval_ORGANIZATION_precision": 0.8783068783068783, + "eval_ORGANIZATION_recall": 0.9325842696629213, + "eval_PERSON_f1": 0.9727626459143969, + "eval_PERSON_number": 128, + "eval_PERSON_precision": 0.9689922480620154, + "eval_PERSON_recall": 0.9765625, + "eval_loss": 0.04940816015005112, + "eval_overall_accuracy": 0.9859649122807017, + "eval_overall_f1": 0.9326683291770573, + "eval_overall_precision": 0.9121951219512195, + "eval_overall_recall": 0.9540816326530612, + "eval_runtime": 0.248, + "eval_samples_per_second": 685.437, + "eval_steps_per_second": 12.096, + "step": 6240 + }, + { + "epoch": 66.0, + "grad_norm": 0.8374859690666199, + "learning_rate": 1.7000000000000003e-05, + "loss": 0.0119, + "step": 6336 + }, + { + "epoch": 66.0, + "eval_LOCATION_f1": 0.9222222222222223, + "eval_LOCATION_number": 86, + "eval_LOCATION_precision": 0.8829787234042553, + "eval_LOCATION_recall": 0.9651162790697675, + "eval_ORGANIZATION_f1": 0.8980716253443526, + "eval_ORGANIZATION_number": 178, + "eval_ORGANIZATION_precision": 0.8810810810810811, + "eval_ORGANIZATION_recall": 0.9157303370786517, + "eval_PERSON_f1": 0.9727626459143969, + "eval_PERSON_number": 128, + "eval_PERSON_precision": 0.9689922480620154, + "eval_PERSON_recall": 0.9765625, + "eval_loss": 0.05106525495648384, + "eval_overall_accuracy": 0.9851551956815114, + "eval_overall_f1": 0.9275, + "eval_overall_precision": 0.9093137254901961, + "eval_overall_recall": 0.9464285714285714, + "eval_runtime": 0.2476, + "eval_samples_per_second": 686.465, + "eval_steps_per_second": 12.114, + "step": 6336 + }, + { + "epoch": 67.0, + "grad_norm": 0.22659751772880554, + "learning_rate": 1.65e-05, + "loss": 0.012, + "step": 6432 + }, + { + "epoch": 67.0, + "eval_LOCATION_f1": 0.9222222222222223, + "eval_LOCATION_number": 86, + "eval_LOCATION_precision": 0.8829787234042553, + "eval_LOCATION_recall": 0.9651162790697675, + "eval_ORGANIZATION_f1": 0.9121813031161473, + "eval_ORGANIZATION_number": 178, + "eval_ORGANIZATION_precision": 0.92, + "eval_ORGANIZATION_recall": 0.9044943820224719, + "eval_PERSON_f1": 0.9765625, + "eval_PERSON_number": 128, + "eval_PERSON_precision": 0.9765625, + "eval_PERSON_recall": 0.9765625, + "eval_loss": 0.05494025722146034, + "eval_overall_accuracy": 0.9851551956815114, + "eval_overall_f1": 0.935361216730038, + "eval_overall_precision": 0.929471032745592, + "eval_overall_recall": 0.9413265306122449, + "eval_runtime": 0.2482, + "eval_samples_per_second": 684.843, + "eval_steps_per_second": 12.085, + "step": 6432 + }, + { + "epoch": 68.0, + "grad_norm": 0.7559866905212402, + "learning_rate": 1.6000000000000003e-05, + "loss": 0.0124, + "step": 6528 + }, + { + "epoch": 68.0, + "eval_LOCATION_f1": 0.9213483146067417, + "eval_LOCATION_number": 86, + "eval_LOCATION_precision": 0.8913043478260869, + "eval_LOCATION_recall": 0.9534883720930233, + "eval_ORGANIZATION_f1": 0.9035812672176309, + "eval_ORGANIZATION_number": 178, + "eval_ORGANIZATION_precision": 0.8864864864864865, + "eval_ORGANIZATION_recall": 0.9213483146067416, + "eval_PERSON_f1": 0.9727626459143969, + "eval_PERSON_number": 128, + "eval_PERSON_precision": 0.9689922480620154, + "eval_PERSON_recall": 0.9765625, + "eval_loss": 0.052313096821308136, + "eval_overall_accuracy": 0.9838056680161943, + "eval_overall_f1": 0.9298245614035087, + "eval_overall_precision": 0.9137931034482759, + "eval_overall_recall": 0.9464285714285714, + "eval_runtime": 0.2481, + "eval_samples_per_second": 685.296, + "eval_steps_per_second": 12.093, + "step": 6528 + }, + { + "epoch": 69.0, + "grad_norm": 0.5614367723464966, + "learning_rate": 1.55e-05, + "loss": 0.0139, + "step": 6624 + }, + { + "epoch": 69.0, + "eval_LOCATION_f1": 0.9162011173184358, + "eval_LOCATION_number": 86, + "eval_LOCATION_precision": 0.8817204301075269, + "eval_LOCATION_recall": 0.9534883720930233, + "eval_ORGANIZATION_f1": 0.9070422535211267, + "eval_ORGANIZATION_number": 178, + "eval_ORGANIZATION_precision": 0.9096045197740112, + "eval_ORGANIZATION_recall": 0.9044943820224719, + "eval_PERSON_f1": 0.9727626459143969, + "eval_PERSON_number": 128, + "eval_PERSON_precision": 0.9689922480620154, + "eval_PERSON_recall": 0.9765625, + "eval_loss": 0.055041369050741196, + "eval_overall_accuracy": 0.9840755735492578, + "eval_overall_f1": 0.9304677623261693, + "eval_overall_precision": 0.9223057644110275, + "eval_overall_recall": 0.9387755102040817, + "eval_runtime": 0.2439, + "eval_samples_per_second": 696.924, + "eval_steps_per_second": 12.299, + "step": 6624 + }, + { + "epoch": 70.0, + "grad_norm": 0.11021454632282257, + "learning_rate": 1.5e-05, + "loss": 0.0121, + "step": 6720 + }, + { + "epoch": 70.0, + "eval_LOCATION_f1": 0.9325842696629213, + "eval_LOCATION_number": 86, + "eval_LOCATION_precision": 0.9021739130434783, + "eval_LOCATION_recall": 0.9651162790697675, + "eval_ORGANIZATION_f1": 0.9065934065934066, + "eval_ORGANIZATION_number": 178, + "eval_ORGANIZATION_precision": 0.8870967741935484, + "eval_ORGANIZATION_recall": 0.9269662921348315, + "eval_PERSON_f1": 0.9727626459143969, + "eval_PERSON_number": 128, + "eval_PERSON_precision": 0.9689922480620154, + "eval_PERSON_recall": 0.9765625, + "eval_loss": 0.05283116549253464, + "eval_overall_accuracy": 0.9851551956815114, + "eval_overall_f1": 0.9336670838548186, + "eval_overall_precision": 0.9164619164619164, + "eval_overall_recall": 0.951530612244898, + "eval_runtime": 0.2468, + "eval_samples_per_second": 688.687, + "eval_steps_per_second": 12.153, + "step": 6720 + }, + { + "epoch": 71.0, + "grad_norm": 0.6227138638496399, + "learning_rate": 1.45e-05, + "loss": 0.0113, + "step": 6816 + }, + { + "epoch": 71.0, + "eval_LOCATION_f1": 0.9152542372881357, + "eval_LOCATION_number": 86, + "eval_LOCATION_precision": 0.8901098901098901, + "eval_LOCATION_recall": 0.9418604651162791, + "eval_ORGANIZATION_f1": 0.8991825613079019, + "eval_ORGANIZATION_number": 178, + "eval_ORGANIZATION_precision": 0.873015873015873, + "eval_ORGANIZATION_recall": 0.9269662921348315, + "eval_PERSON_f1": 0.9727626459143969, + "eval_PERSON_number": 128, + "eval_PERSON_precision": 0.9689922480620154, + "eval_PERSON_recall": 0.9765625, + "eval_loss": 0.04995880275964737, + "eval_overall_accuracy": 0.9856950067476383, + "eval_overall_f1": 0.9263420724094882, + "eval_overall_precision": 0.9070904645476773, + "eval_overall_recall": 0.9464285714285714, + "eval_runtime": 0.2476, + "eval_samples_per_second": 686.613, + "eval_steps_per_second": 12.117, + "step": 6816 + }, + { + "epoch": 72.0, + "grad_norm": 0.25081753730773926, + "learning_rate": 1.4000000000000001e-05, + "loss": 0.0114, + "step": 6912 + }, + { + "epoch": 72.0, + "eval_LOCATION_f1": 0.9101123595505618, + "eval_LOCATION_number": 86, + "eval_LOCATION_precision": 0.8804347826086957, + "eval_LOCATION_recall": 0.9418604651162791, + "eval_ORGANIZATION_f1": 0.8980716253443526, + "eval_ORGANIZATION_number": 178, + "eval_ORGANIZATION_precision": 0.8810810810810811, + "eval_ORGANIZATION_recall": 0.9157303370786517, + "eval_PERSON_f1": 0.9727626459143969, + "eval_PERSON_number": 128, + "eval_PERSON_precision": 0.9689922480620154, + "eval_PERSON_recall": 0.9765625, + "eval_loss": 0.05604882910847664, + "eval_overall_accuracy": 0.9843454790823212, + "eval_overall_f1": 0.924812030075188, + "eval_overall_precision": 0.9088669950738916, + "eval_overall_recall": 0.9413265306122449, + "eval_runtime": 0.2479, + "eval_samples_per_second": 685.886, + "eval_steps_per_second": 12.104, + "step": 6912 + }, + { + "epoch": 73.0, + "grad_norm": 0.3885822296142578, + "learning_rate": 1.3500000000000001e-05, + "loss": 0.01, + "step": 7008 + }, + { + "epoch": 73.0, + "eval_LOCATION_f1": 0.9171270718232045, + "eval_LOCATION_number": 86, + "eval_LOCATION_precision": 0.8736842105263158, + "eval_LOCATION_recall": 0.9651162790697675, + "eval_ORGANIZATION_f1": 0.9065934065934066, + "eval_ORGANIZATION_number": 178, + "eval_ORGANIZATION_precision": 0.8870967741935484, + "eval_ORGANIZATION_recall": 0.9269662921348315, + "eval_PERSON_f1": 0.9727626459143969, + "eval_PERSON_number": 128, + "eval_PERSON_precision": 0.9689922480620154, + "eval_PERSON_recall": 0.9765625, + "eval_loss": 0.05322219058871269, + "eval_overall_accuracy": 0.9856950067476383, + "eval_overall_f1": 0.9301745635910225, + "eval_overall_precision": 0.9097560975609756, + "eval_overall_recall": 0.951530612244898, + "eval_runtime": 0.2471, + "eval_samples_per_second": 687.924, + "eval_steps_per_second": 12.14, + "step": 7008 + }, + { + "epoch": 74.0, + "grad_norm": 0.020958004519343376, + "learning_rate": 1.3000000000000001e-05, + "loss": 0.0138, + "step": 7104 + }, + { + "epoch": 74.0, + "eval_LOCATION_f1": 0.9273743016759777, + "eval_LOCATION_number": 86, + "eval_LOCATION_precision": 0.8924731182795699, + "eval_LOCATION_recall": 0.9651162790697675, + "eval_ORGANIZATION_f1": 0.901098901098901, + "eval_ORGANIZATION_number": 178, + "eval_ORGANIZATION_precision": 0.8817204301075269, + "eval_ORGANIZATION_recall": 0.9213483146067416, + "eval_PERSON_f1": 0.9727626459143969, + "eval_PERSON_number": 128, + "eval_PERSON_precision": 0.9689922480620154, + "eval_PERSON_recall": 0.9765625, + "eval_loss": 0.052465103566646576, + "eval_overall_accuracy": 0.9851551956815114, + "eval_overall_f1": 0.9300000000000002, + "eval_overall_precision": 0.9117647058823529, + "eval_overall_recall": 0.9489795918367347, + "eval_runtime": 0.2474, + "eval_samples_per_second": 687.252, + "eval_steps_per_second": 12.128, + "step": 7104 + }, + { + "epoch": 75.0, + "grad_norm": 0.15891754627227783, + "learning_rate": 1.25e-05, + "loss": 0.0108, + "step": 7200 + }, + { + "epoch": 75.0, + "eval_LOCATION_f1": 0.9120879120879122, + "eval_LOCATION_number": 86, + "eval_LOCATION_precision": 0.8645833333333334, + "eval_LOCATION_recall": 0.9651162790697675, + "eval_ORGANIZATION_f1": 0.9014084507042253, + "eval_ORGANIZATION_number": 178, + "eval_ORGANIZATION_precision": 0.903954802259887, + "eval_ORGANIZATION_recall": 0.898876404494382, + "eval_PERSON_f1": 0.9727626459143969, + "eval_PERSON_number": 128, + "eval_PERSON_precision": 0.9689922480620154, + "eval_PERSON_recall": 0.9765625, + "eval_loss": 0.0572768896818161, + "eval_overall_accuracy": 0.9827260458839406, + "eval_overall_f1": 0.9269521410579346, + "eval_overall_precision": 0.9154228855721394, + "eval_overall_recall": 0.9387755102040817, + "eval_runtime": 0.2477, + "eval_samples_per_second": 686.447, + "eval_steps_per_second": 12.114, + "step": 7200 + }, + { + "epoch": 76.0, + "grad_norm": 0.42831161618232727, + "learning_rate": 1.2e-05, + "loss": 0.0113, + "step": 7296 + }, + { + "epoch": 76.0, + "eval_LOCATION_f1": 0.9325842696629213, + "eval_LOCATION_number": 86, + "eval_LOCATION_precision": 0.9021739130434783, + "eval_LOCATION_recall": 0.9651162790697675, + "eval_ORGANIZATION_f1": 0.902506963788301, + "eval_ORGANIZATION_number": 178, + "eval_ORGANIZATION_precision": 0.8950276243093923, + "eval_ORGANIZATION_recall": 0.9101123595505618, + "eval_PERSON_f1": 0.9727626459143969, + "eval_PERSON_number": 128, + "eval_PERSON_precision": 0.9689922480620154, + "eval_PERSON_recall": 0.9765625, + "eval_loss": 0.054985810071229935, + "eval_overall_accuracy": 0.9848852901484481, + "eval_overall_f1": 0.9319899244332494, + "eval_overall_precision": 0.9203980099502488, + "eval_overall_recall": 0.9438775510204082, + "eval_runtime": 0.2472, + "eval_samples_per_second": 687.566, + "eval_steps_per_second": 12.134, + "step": 7296 + }, + { + "epoch": 77.0, + "grad_norm": 0.14717106521129608, + "learning_rate": 1.1500000000000002e-05, + "loss": 0.0111, + "step": 7392 + }, + { + "epoch": 77.0, + "eval_LOCATION_f1": 0.9378531073446328, + "eval_LOCATION_number": 86, + "eval_LOCATION_precision": 0.9120879120879121, + "eval_LOCATION_recall": 0.9651162790697675, + "eval_ORGANIZATION_f1": 0.9157303370786517, + "eval_ORGANIZATION_number": 178, + "eval_ORGANIZATION_precision": 0.9157303370786517, + "eval_ORGANIZATION_recall": 0.9157303370786517, + "eval_PERSON_f1": 0.9765625, + "eval_PERSON_number": 128, + "eval_PERSON_precision": 0.9765625, + "eval_PERSON_recall": 0.9765625, + "eval_loss": 0.056410305202007294, + "eval_overall_accuracy": 0.9846153846153847, + "eval_overall_f1": 0.9404309252217997, + "eval_overall_precision": 0.9345088161209067, + "eval_overall_recall": 0.9464285714285714, + "eval_runtime": 0.2478, + "eval_samples_per_second": 686.043, + "eval_steps_per_second": 12.107, + "step": 7392 + }, + { + "epoch": 78.0, + "grad_norm": 0.7005665898323059, + "learning_rate": 1.1000000000000001e-05, + "loss": 0.0114, + "step": 7488 + }, + { + "epoch": 78.0, + "eval_LOCATION_f1": 0.9171270718232045, + "eval_LOCATION_number": 86, + "eval_LOCATION_precision": 0.8736842105263158, + "eval_LOCATION_recall": 0.9651162790697675, + "eval_ORGANIZATION_f1": 0.902506963788301, + "eval_ORGANIZATION_number": 178, + "eval_ORGANIZATION_precision": 0.8950276243093923, + "eval_ORGANIZATION_recall": 0.9101123595505618, + "eval_PERSON_f1": 0.9765625, + "eval_PERSON_number": 128, + "eval_PERSON_precision": 0.9765625, + "eval_PERSON_recall": 0.9765625, + "eval_loss": 0.05650125443935394, + "eval_overall_accuracy": 0.9832658569500675, + "eval_overall_f1": 0.9296482412060302, + "eval_overall_precision": 0.9158415841584159, + "eval_overall_recall": 0.9438775510204082, + "eval_runtime": 0.2411, + "eval_samples_per_second": 705.234, + "eval_steps_per_second": 12.445, + "step": 7488 + }, + { + "epoch": 79.0, + "grad_norm": 0.07694126665592194, + "learning_rate": 1.05e-05, + "loss": 0.0101, + "step": 7584 + }, + { + "epoch": 79.0, + "eval_LOCATION_f1": 0.9325842696629213, + "eval_LOCATION_number": 86, + "eval_LOCATION_precision": 0.9021739130434783, + "eval_LOCATION_recall": 0.9651162790697675, + "eval_ORGANIZATION_f1": 0.9162011173184358, + "eval_ORGANIZATION_number": 178, + "eval_ORGANIZATION_precision": 0.9111111111111111, + "eval_ORGANIZATION_recall": 0.9213483146067416, + "eval_PERSON_f1": 0.9727626459143969, + "eval_PERSON_number": 128, + "eval_PERSON_precision": 0.9689922480620154, + "eval_PERSON_recall": 0.9765625, + "eval_loss": 0.05472679063677788, + "eval_overall_accuracy": 0.9846153846153847, + "eval_overall_f1": 0.9382093316519546, + "eval_overall_precision": 0.9276807980049875, + "eval_overall_recall": 0.9489795918367347, + "eval_runtime": 0.2491, + "eval_samples_per_second": 682.486, + "eval_steps_per_second": 12.044, + "step": 7584 + }, + { + "epoch": 80.0, + "grad_norm": 0.6124340295791626, + "learning_rate": 1e-05, + "loss": 0.0108, + "step": 7680 + }, + { + "epoch": 80.0, + "eval_LOCATION_f1": 0.9325842696629213, + "eval_LOCATION_number": 86, + "eval_LOCATION_precision": 0.9021739130434783, + "eval_LOCATION_recall": 0.9651162790697675, + "eval_ORGANIZATION_f1": 0.9035812672176309, + "eval_ORGANIZATION_number": 178, + "eval_ORGANIZATION_precision": 0.8864864864864865, + "eval_ORGANIZATION_recall": 0.9213483146067416, + "eval_PERSON_f1": 0.9727626459143969, + "eval_PERSON_number": 128, + "eval_PERSON_precision": 0.9689922480620154, + "eval_PERSON_recall": 0.9765625, + "eval_loss": 0.05269638076424599, + "eval_overall_accuracy": 0.9854251012145749, + "eval_overall_f1": 0.9323308270676692, + "eval_overall_precision": 0.916256157635468, + "eval_overall_recall": 0.9489795918367347, + "eval_runtime": 0.2488, + "eval_samples_per_second": 683.407, + "eval_steps_per_second": 12.06, + "step": 7680 + }, + { + "epoch": 81.0, + "grad_norm": 0.09941994398832321, + "learning_rate": 9.5e-06, + "loss": 0.0108, + "step": 7776 + }, + { + "epoch": 81.0, + "eval_LOCATION_f1": 0.9273743016759777, + "eval_LOCATION_number": 86, + "eval_LOCATION_precision": 0.8924731182795699, + "eval_LOCATION_recall": 0.9651162790697675, + "eval_ORGANIZATION_f1": 0.9095890410958904, + "eval_ORGANIZATION_number": 178, + "eval_ORGANIZATION_precision": 0.8877005347593583, + "eval_ORGANIZATION_recall": 0.9325842696629213, + "eval_PERSON_f1": 0.9727626459143969, + "eval_PERSON_number": 128, + "eval_PERSON_precision": 0.9689922480620154, + "eval_PERSON_recall": 0.9765625, + "eval_loss": 0.05239494517445564, + "eval_overall_accuracy": 0.9859649122807017, + "eval_overall_f1": 0.933832709113608, + "eval_overall_precision": 0.9144254278728606, + "eval_overall_recall": 0.9540816326530612, + "eval_runtime": 0.2481, + "eval_samples_per_second": 685.107, + "eval_steps_per_second": 12.09, + "step": 7776 + }, + { + "epoch": 82.0, + "grad_norm": 0.25073766708374023, + "learning_rate": 9e-06, + "loss": 0.0108, + "step": 7872 + }, + { + "epoch": 82.0, + "eval_LOCATION_f1": 0.9222222222222223, + "eval_LOCATION_number": 86, + "eval_LOCATION_precision": 0.8829787234042553, + "eval_LOCATION_recall": 0.9651162790697675, + "eval_ORGANIZATION_f1": 0.8999999999999999, + "eval_ORGANIZATION_number": 178, + "eval_ORGANIZATION_precision": 0.8901098901098901, + "eval_ORGANIZATION_recall": 0.9101123595505618, + "eval_PERSON_f1": 0.9765625, + "eval_PERSON_number": 128, + "eval_PERSON_precision": 0.9765625, + "eval_PERSON_recall": 0.9765625, + "eval_loss": 0.05534841865301132, + "eval_overall_accuracy": 0.9846153846153847, + "eval_overall_f1": 0.9296482412060302, + "eval_overall_precision": 0.9158415841584159, + "eval_overall_recall": 0.9438775510204082, + "eval_runtime": 0.2416, + "eval_samples_per_second": 703.691, + "eval_steps_per_second": 12.418, + "step": 7872 + }, + { + "epoch": 83.0, + "grad_norm": 0.9179547429084778, + "learning_rate": 8.500000000000002e-06, + "loss": 0.0109, + "step": 7968 + }, + { + "epoch": 83.0, + "eval_LOCATION_f1": 0.9273743016759777, + "eval_LOCATION_number": 86, + "eval_LOCATION_precision": 0.8924731182795699, + "eval_LOCATION_recall": 0.9651162790697675, + "eval_ORGANIZATION_f1": 0.9030470914127424, + "eval_ORGANIZATION_number": 178, + "eval_ORGANIZATION_precision": 0.8907103825136612, + "eval_ORGANIZATION_recall": 0.9157303370786517, + "eval_PERSON_f1": 0.9727626459143969, + "eval_PERSON_number": 128, + "eval_PERSON_precision": 0.9689922480620154, + "eval_PERSON_recall": 0.9765625, + "eval_loss": 0.0529344268143177, + "eval_overall_accuracy": 0.9851551956815114, + "eval_overall_f1": 0.9309912170639899, + "eval_overall_precision": 0.9160493827160494, + "eval_overall_recall": 0.9464285714285714, + "eval_runtime": 0.251, + "eval_samples_per_second": 677.348, + "eval_steps_per_second": 11.953, + "step": 7968 + }, + { + "epoch": 84.0, + "grad_norm": 0.5180923938751221, + "learning_rate": 8.000000000000001e-06, + "loss": 0.0095, + "step": 8064 + }, + { + "epoch": 84.0, + "eval_LOCATION_f1": 0.9273743016759777, + "eval_LOCATION_number": 86, + "eval_LOCATION_precision": 0.8924731182795699, + "eval_LOCATION_recall": 0.9651162790697675, + "eval_ORGANIZATION_f1": 0.9050279329608939, + "eval_ORGANIZATION_number": 178, + "eval_ORGANIZATION_precision": 0.9, + "eval_ORGANIZATION_recall": 0.9101123595505618, + "eval_PERSON_f1": 0.9765625, + "eval_PERSON_number": 128, + "eval_PERSON_precision": 0.9765625, + "eval_PERSON_recall": 0.9765625, + "eval_loss": 0.05454019829630852, + "eval_overall_accuracy": 0.9846153846153847, + "eval_overall_f1": 0.9331651954602775, + "eval_overall_precision": 0.9226932668329177, + "eval_overall_recall": 0.9438775510204082, + "eval_runtime": 0.249, + "eval_samples_per_second": 682.731, + "eval_steps_per_second": 12.048, + "step": 8064 + }, + { + "epoch": 85.0, + "grad_norm": 0.7953592538833618, + "learning_rate": 7.5e-06, + "loss": 0.0096, + "step": 8160 + }, + { + "epoch": 85.0, + "eval_LOCATION_f1": 0.9325842696629213, + "eval_LOCATION_number": 86, + "eval_LOCATION_precision": 0.9021739130434783, + "eval_LOCATION_recall": 0.9651162790697675, + "eval_ORGANIZATION_f1": 0.9101123595505618, + "eval_ORGANIZATION_number": 178, + "eval_ORGANIZATION_precision": 0.9101123595505618, + "eval_ORGANIZATION_recall": 0.9101123595505618, + "eval_PERSON_f1": 0.9765625, + "eval_PERSON_number": 128, + "eval_PERSON_precision": 0.9765625, + "eval_PERSON_recall": 0.9765625, + "eval_loss": 0.05559179559350014, + "eval_overall_accuracy": 0.9840755735492578, + "eval_overall_f1": 0.9367088607594937, + "eval_overall_precision": 0.9296482412060302, + "eval_overall_recall": 0.9438775510204082, + "eval_runtime": 0.2483, + "eval_samples_per_second": 684.666, + "eval_steps_per_second": 12.082, + "step": 8160 + }, + { + "epoch": 86.0, + "grad_norm": 0.07799229770898819, + "learning_rate": 7.000000000000001e-06, + "loss": 0.0098, + "step": 8256 + }, + { + "epoch": 86.0, + "eval_LOCATION_f1": 0.9325842696629213, + "eval_LOCATION_number": 86, + "eval_LOCATION_precision": 0.9021739130434783, + "eval_LOCATION_recall": 0.9651162790697675, + "eval_ORGANIZATION_f1": 0.9171270718232043, + "eval_ORGANIZATION_number": 178, + "eval_ORGANIZATION_precision": 0.9021739130434783, + "eval_ORGANIZATION_recall": 0.9325842696629213, + "eval_PERSON_f1": 0.9727626459143969, + "eval_PERSON_number": 128, + "eval_PERSON_precision": 0.9689922480620154, + "eval_PERSON_recall": 0.9765625, + "eval_loss": 0.05340668186545372, + "eval_overall_accuracy": 0.9854251012145749, + "eval_overall_f1": 0.9385194479297364, + "eval_overall_precision": 0.9234567901234568, + "eval_overall_recall": 0.9540816326530612, + "eval_runtime": 0.2486, + "eval_samples_per_second": 683.761, + "eval_steps_per_second": 12.066, + "step": 8256 + }, + { + "epoch": 87.0, + "grad_norm": 0.04336532950401306, + "learning_rate": 6.5000000000000004e-06, + "loss": 0.0103, + "step": 8352 + }, + { + "epoch": 87.0, + "eval_LOCATION_f1": 0.9378531073446328, + "eval_LOCATION_number": 86, + "eval_LOCATION_precision": 0.9120879120879121, + "eval_LOCATION_recall": 0.9651162790697675, + "eval_ORGANIZATION_f1": 0.9106145251396648, + "eval_ORGANIZATION_number": 178, + "eval_ORGANIZATION_precision": 0.9055555555555556, + "eval_ORGANIZATION_recall": 0.9157303370786517, + "eval_PERSON_f1": 0.9727626459143969, + "eval_PERSON_number": 128, + "eval_PERSON_precision": 0.9689922480620154, + "eval_PERSON_recall": 0.9765625, + "eval_loss": 0.05287463217973709, + "eval_overall_accuracy": 0.9854251012145749, + "eval_overall_f1": 0.9368686868686869, + "eval_overall_precision": 0.9275, + "eval_overall_recall": 0.9464285714285714, + "eval_runtime": 0.2473, + "eval_samples_per_second": 687.307, + "eval_steps_per_second": 12.129, + "step": 8352 + }, + { + "epoch": 88.0, + "grad_norm": 1.0946106910705566, + "learning_rate": 6e-06, + "loss": 0.0113, + "step": 8448 + }, + { + "epoch": 88.0, + "eval_LOCATION_f1": 0.9273743016759777, + "eval_LOCATION_number": 86, + "eval_LOCATION_precision": 0.8924731182795699, + "eval_LOCATION_recall": 0.9651162790697675, + "eval_ORGANIZATION_f1": 0.9141274238227147, + "eval_ORGANIZATION_number": 178, + "eval_ORGANIZATION_precision": 0.9016393442622951, + "eval_ORGANIZATION_recall": 0.9269662921348315, + "eval_PERSON_f1": 0.9765625, + "eval_PERSON_number": 128, + "eval_PERSON_precision": 0.9765625, + "eval_PERSON_recall": 0.9765625, + "eval_loss": 0.05374481528997421, + "eval_overall_accuracy": 0.9856950067476383, + "eval_overall_f1": 0.9371859296482412, + "eval_overall_precision": 0.9232673267326733, + "eval_overall_recall": 0.951530612244898, + "eval_runtime": 0.2475, + "eval_samples_per_second": 686.787, + "eval_steps_per_second": 12.12, + "step": 8448 + }, + { + "epoch": 89.0, + "grad_norm": 0.055148638784885406, + "learning_rate": 5.500000000000001e-06, + "loss": 0.0103, + "step": 8544 + }, + { + "epoch": 89.0, + "eval_LOCATION_f1": 0.9213483146067417, + "eval_LOCATION_number": 86, + "eval_LOCATION_precision": 0.8913043478260869, + "eval_LOCATION_recall": 0.9534883720930233, + "eval_ORGANIZATION_f1": 0.9060773480662982, + "eval_ORGANIZATION_number": 178, + "eval_ORGANIZATION_precision": 0.8913043478260869, + "eval_ORGANIZATION_recall": 0.9213483146067416, + "eval_PERSON_f1": 0.9727626459143969, + "eval_PERSON_number": 128, + "eval_PERSON_precision": 0.9689922480620154, + "eval_PERSON_recall": 0.9765625, + "eval_loss": 0.05380658432841301, + "eval_overall_accuracy": 0.9851551956815114, + "eval_overall_f1": 0.9309912170639899, + "eval_overall_precision": 0.9160493827160494, + "eval_overall_recall": 0.9464285714285714, + "eval_runtime": 0.2484, + "eval_samples_per_second": 684.349, + "eval_steps_per_second": 12.077, + "step": 8544 + }, + { + "epoch": 90.0, + "grad_norm": 0.522067666053772, + "learning_rate": 5e-06, + "loss": 0.0101, + "step": 8640 + }, + { + "epoch": 90.0, + "eval_LOCATION_f1": 0.9213483146067417, + "eval_LOCATION_number": 86, + "eval_LOCATION_precision": 0.8913043478260869, + "eval_LOCATION_recall": 0.9534883720930233, + "eval_ORGANIZATION_f1": 0.9035812672176309, + "eval_ORGANIZATION_number": 178, + "eval_ORGANIZATION_precision": 0.8864864864864865, + "eval_ORGANIZATION_recall": 0.9213483146067416, + "eval_PERSON_f1": 0.9727626459143969, + "eval_PERSON_number": 128, + "eval_PERSON_precision": 0.9689922480620154, + "eval_PERSON_recall": 0.9765625, + "eval_loss": 0.05221777409315109, + "eval_overall_accuracy": 0.9848852901484481, + "eval_overall_f1": 0.9298245614035087, + "eval_overall_precision": 0.9137931034482759, + "eval_overall_recall": 0.9464285714285714, + "eval_runtime": 0.2491, + "eval_samples_per_second": 682.544, + "eval_steps_per_second": 12.045, + "step": 8640 + }, + { + "epoch": 91.0, + "grad_norm": 0.78941410779953, + "learning_rate": 4.5e-06, + "loss": 0.0089, + "step": 8736 + }, + { + "epoch": 91.0, + "eval_LOCATION_f1": 0.9325842696629213, + "eval_LOCATION_number": 86, + "eval_LOCATION_precision": 0.9021739130434783, + "eval_LOCATION_recall": 0.9651162790697675, + "eval_ORGANIZATION_f1": 0.9005524861878453, + "eval_ORGANIZATION_number": 178, + "eval_ORGANIZATION_precision": 0.8858695652173914, + "eval_ORGANIZATION_recall": 0.9157303370786517, + "eval_PERSON_f1": 0.9727626459143969, + "eval_PERSON_number": 128, + "eval_PERSON_precision": 0.9689922480620154, + "eval_PERSON_recall": 0.9765625, + "eval_loss": 0.052769750356674194, + "eval_overall_accuracy": 0.9851551956815114, + "eval_overall_f1": 0.9309912170639899, + "eval_overall_precision": 0.9160493827160494, + "eval_overall_recall": 0.9464285714285714, + "eval_runtime": 0.2473, + "eval_samples_per_second": 687.479, + "eval_steps_per_second": 12.132, + "step": 8736 + }, + { + "epoch": 92.0, + "grad_norm": 0.3154488503932953, + "learning_rate": 4.000000000000001e-06, + "loss": 0.0088, + "step": 8832 + }, + { + "epoch": 92.0, + "eval_LOCATION_f1": 0.9325842696629213, + "eval_LOCATION_number": 86, + "eval_LOCATION_precision": 0.9021739130434783, + "eval_LOCATION_recall": 0.9651162790697675, + "eval_ORGANIZATION_f1": 0.9055555555555556, + "eval_ORGANIZATION_number": 178, + "eval_ORGANIZATION_precision": 0.8956043956043956, + "eval_ORGANIZATION_recall": 0.9157303370786517, + "eval_PERSON_f1": 0.9727626459143969, + "eval_PERSON_number": 128, + "eval_PERSON_precision": 0.9689922480620154, + "eval_PERSON_recall": 0.9765625, + "eval_loss": 0.05468715354800224, + "eval_overall_accuracy": 0.9848852901484481, + "eval_overall_f1": 0.9333333333333335, + "eval_overall_precision": 0.9205955334987593, + "eval_overall_recall": 0.9464285714285714, + "eval_runtime": 0.2486, + "eval_samples_per_second": 683.728, + "eval_steps_per_second": 12.066, + "step": 8832 + }, + { + "epoch": 93.0, + "grad_norm": 0.6052812933921814, + "learning_rate": 3.5000000000000004e-06, + "loss": 0.0099, + "step": 8928 + }, + { + "epoch": 93.0, + "eval_LOCATION_f1": 0.9325842696629213, + "eval_LOCATION_number": 86, + "eval_LOCATION_precision": 0.9021739130434783, + "eval_LOCATION_recall": 0.9651162790697675, + "eval_ORGANIZATION_f1": 0.9030470914127424, + "eval_ORGANIZATION_number": 178, + "eval_ORGANIZATION_precision": 0.8907103825136612, + "eval_ORGANIZATION_recall": 0.9157303370786517, + "eval_PERSON_f1": 0.9727626459143969, + "eval_PERSON_number": 128, + "eval_PERSON_precision": 0.9689922480620154, + "eval_PERSON_recall": 0.9765625, + "eval_loss": 0.054242152720689774, + "eval_overall_accuracy": 0.9848852901484481, + "eval_overall_f1": 0.9321608040201005, + "eval_overall_precision": 0.9183168316831684, + "eval_overall_recall": 0.9464285714285714, + "eval_runtime": 0.2487, + "eval_samples_per_second": 683.45, + "eval_steps_per_second": 12.061, + "step": 8928 + }, + { + "epoch": 94.0, + "grad_norm": 0.10974699258804321, + "learning_rate": 3e-06, + "loss": 0.0095, + "step": 9024 + }, + { + "epoch": 94.0, + "eval_LOCATION_f1": 0.9325842696629213, + "eval_LOCATION_number": 86, + "eval_LOCATION_precision": 0.9021739130434783, + "eval_LOCATION_recall": 0.9651162790697675, + "eval_ORGANIZATION_f1": 0.9030470914127424, + "eval_ORGANIZATION_number": 178, + "eval_ORGANIZATION_precision": 0.8907103825136612, + "eval_ORGANIZATION_recall": 0.9157303370786517, + "eval_PERSON_f1": 0.9727626459143969, + "eval_PERSON_number": 128, + "eval_PERSON_precision": 0.9689922480620154, + "eval_PERSON_recall": 0.9765625, + "eval_loss": 0.05423642322421074, + "eval_overall_accuracy": 0.9848852901484481, + "eval_overall_f1": 0.9321608040201005, + "eval_overall_precision": 0.9183168316831684, + "eval_overall_recall": 0.9464285714285714, + "eval_runtime": 0.2486, + "eval_samples_per_second": 683.781, + "eval_steps_per_second": 12.067, + "step": 9024 + }, + { + "epoch": 95.0, + "grad_norm": 0.3483632206916809, + "learning_rate": 2.5e-06, + "loss": 0.0094, + "step": 9120 + }, + { + "epoch": 95.0, + "eval_LOCATION_f1": 0.9325842696629213, + "eval_LOCATION_number": 86, + "eval_LOCATION_precision": 0.9021739130434783, + "eval_LOCATION_recall": 0.9651162790697675, + "eval_ORGANIZATION_f1": 0.9030470914127424, + "eval_ORGANIZATION_number": 178, + "eval_ORGANIZATION_precision": 0.8907103825136612, + "eval_ORGANIZATION_recall": 0.9157303370786517, + "eval_PERSON_f1": 0.9765625, + "eval_PERSON_number": 128, + "eval_PERSON_precision": 0.9765625, + "eval_PERSON_recall": 0.9765625, + "eval_loss": 0.05356050282716751, + "eval_overall_accuracy": 0.9856950067476383, + "eval_overall_f1": 0.9333333333333335, + "eval_overall_precision": 0.9205955334987593, + "eval_overall_recall": 0.9464285714285714, + "eval_runtime": 0.2433, + "eval_samples_per_second": 698.606, + "eval_steps_per_second": 12.328, + "step": 9120 + }, + { + "epoch": 96.0, + "grad_norm": 0.06140005588531494, + "learning_rate": 2.0000000000000003e-06, + "loss": 0.0093, + "step": 9216 + }, + { + "epoch": 96.0, + "eval_LOCATION_f1": 0.9325842696629213, + "eval_LOCATION_number": 86, + "eval_LOCATION_precision": 0.9021739130434783, + "eval_LOCATION_recall": 0.9651162790697675, + "eval_ORGANIZATION_f1": 0.9060773480662982, + "eval_ORGANIZATION_number": 178, + "eval_ORGANIZATION_precision": 0.8913043478260869, + "eval_ORGANIZATION_recall": 0.9213483146067416, + "eval_PERSON_f1": 0.9765625, + "eval_PERSON_number": 128, + "eval_PERSON_precision": 0.9765625, + "eval_PERSON_recall": 0.9765625, + "eval_loss": 0.05313665792346001, + "eval_overall_accuracy": 0.9862348178137652, + "eval_overall_f1": 0.934673366834171, + "eval_overall_precision": 0.9207920792079208, + "eval_overall_recall": 0.9489795918367347, + "eval_runtime": 0.2485, + "eval_samples_per_second": 684.112, + "eval_steps_per_second": 12.073, + "step": 9216 + }, + { + "epoch": 97.0, + "grad_norm": 0.0674736350774765, + "learning_rate": 1.5e-06, + "loss": 0.009, + "step": 9312 + }, + { + "epoch": 97.0, + "eval_LOCATION_f1": 0.9325842696629213, + "eval_LOCATION_number": 86, + "eval_LOCATION_precision": 0.9021739130434783, + "eval_LOCATION_recall": 0.9651162790697675, + "eval_ORGANIZATION_f1": 0.9035812672176309, + "eval_ORGANIZATION_number": 178, + "eval_ORGANIZATION_precision": 0.8864864864864865, + "eval_ORGANIZATION_recall": 0.9213483146067416, + "eval_PERSON_f1": 0.9727626459143969, + "eval_PERSON_number": 128, + "eval_PERSON_precision": 0.9689922480620154, + "eval_PERSON_recall": 0.9765625, + "eval_loss": 0.053574338555336, + "eval_overall_accuracy": 0.9856950067476383, + "eval_overall_f1": 0.9323308270676692, + "eval_overall_precision": 0.916256157635468, + "eval_overall_recall": 0.9489795918367347, + "eval_runtime": 0.248, + "eval_samples_per_second": 685.414, + "eval_steps_per_second": 12.096, + "step": 9312 + }, + { + "epoch": 98.0, + "grad_norm": 1.0472229719161987, + "learning_rate": 1.0000000000000002e-06, + "loss": 0.0099, + "step": 9408 + }, + { + "epoch": 98.0, + "eval_LOCATION_f1": 0.9325842696629213, + "eval_LOCATION_number": 86, + "eval_LOCATION_precision": 0.9021739130434783, + "eval_LOCATION_recall": 0.9651162790697675, + "eval_ORGANIZATION_f1": 0.9055555555555556, + "eval_ORGANIZATION_number": 178, + "eval_ORGANIZATION_precision": 0.8956043956043956, + "eval_ORGANIZATION_recall": 0.9157303370786517, + "eval_PERSON_f1": 0.9765625, + "eval_PERSON_number": 128, + "eval_PERSON_precision": 0.9765625, + "eval_PERSON_recall": 0.9765625, + "eval_loss": 0.05379428341984749, + "eval_overall_accuracy": 0.9854251012145749, + "eval_overall_f1": 0.9345088161209069, + "eval_overall_precision": 0.9228855721393034, + "eval_overall_recall": 0.9464285714285714, + "eval_runtime": 0.2472, + "eval_samples_per_second": 687.708, + "eval_steps_per_second": 12.136, + "step": 9408 + }, + { + "epoch": 99.0, + "grad_norm": 0.5022507905960083, + "learning_rate": 5.000000000000001e-07, + "loss": 0.0089, + "step": 9504 + }, + { + "epoch": 99.0, + "eval_LOCATION_f1": 0.9378531073446328, + "eval_LOCATION_number": 86, + "eval_LOCATION_precision": 0.9120879120879121, + "eval_LOCATION_recall": 0.9651162790697675, + "eval_ORGANIZATION_f1": 0.9080779944289694, + "eval_ORGANIZATION_number": 178, + "eval_ORGANIZATION_precision": 0.9005524861878453, + "eval_ORGANIZATION_recall": 0.9157303370786517, + "eval_PERSON_f1": 0.9765625, + "eval_PERSON_number": 128, + "eval_PERSON_precision": 0.9765625, + "eval_PERSON_recall": 0.9765625, + "eval_loss": 0.054100390523672104, + "eval_overall_accuracy": 0.9859649122807017, + "eval_overall_f1": 0.9368686868686869, + "eval_overall_precision": 0.9275, + "eval_overall_recall": 0.9464285714285714, + "eval_runtime": 0.2473, + "eval_samples_per_second": 687.373, + "eval_steps_per_second": 12.13, + "step": 9504 + }, + { + "epoch": 100.0, + "grad_norm": 0.782418429851532, + "learning_rate": 0.0, + "loss": 0.0107, + "step": 9600 + }, + { + "epoch": 100.0, + "eval_LOCATION_f1": 0.9325842696629213, + "eval_LOCATION_number": 86, + "eval_LOCATION_precision": 0.9021739130434783, + "eval_LOCATION_recall": 0.9651162790697675, + "eval_ORGANIZATION_f1": 0.9080779944289694, + "eval_ORGANIZATION_number": 178, + "eval_ORGANIZATION_precision": 0.9005524861878453, + "eval_ORGANIZATION_recall": 0.9157303370786517, + "eval_PERSON_f1": 0.9765625, + "eval_PERSON_number": 128, + "eval_PERSON_precision": 0.9765625, + "eval_PERSON_recall": 0.9765625, + "eval_loss": 0.05406568571925163, + "eval_overall_accuracy": 0.9856950067476383, + "eval_overall_f1": 0.935687263556116, + "eval_overall_precision": 0.9251870324189526, + "eval_overall_recall": 0.9464285714285714, + "eval_runtime": 0.248, + "eval_samples_per_second": 685.425, + "eval_steps_per_second": 12.096, + "step": 9600 + }, + { + "epoch": 100.0, + "step": 9600, + "total_flos": 3908603344703268.0, + "train_loss": 0.0451166748503844, + "train_runtime": 562.8472, + "train_samples_per_second": 272.01, + "train_steps_per_second": 17.056 + } + ], + "logging_steps": 500, + "max_steps": 9600, + "num_input_tokens_seen": 0, + "num_train_epochs": 100, + "save_steps": 500, + "total_flos": 3908603344703268.0, + "train_batch_size": 16, + "trial_name": null, + "trial_params": null +}