diff --git "a/trainer_state.json" "b/trainer_state.json" new file mode 100644--- /dev/null +++ "b/trainer_state.json" @@ -0,0 +1,21274 @@ +{ + "best_metric": 0.18502268195152283, + "best_model_checkpoint": "../awesome/Datasets/en-multi/Outputs_mined/checkpoint-506000", + "epoch": 2.8721767807779854, + "global_step": 506000, + "is_hyper_param_search": false, + "is_local_process_zero": true, + "is_world_process_zero": true, + "log_history": [ + { + "epoch": 0.0, + "learning_rate": 4.999990539602172e-05, + "loss": 1.9917, + "step": 1 + }, + { + "epoch": 0.0, + "learning_rate": 4.99810792043427e-05, + "loss": 0.3956, + "step": 200 + }, + { + "epoch": 0.0, + "learning_rate": 4.996215840868541e-05, + "loss": 0.3231, + "step": 400 + }, + { + "epoch": 0.0, + "learning_rate": 4.99432376130281e-05, + "loss": 0.3199, + "step": 600 + }, + { + "epoch": 0.0, + "learning_rate": 4.992431681737081e-05, + "loss": 0.3102, + "step": 800 + }, + { + "epoch": 0.01, + "learning_rate": 4.990539602171351e-05, + "loss": 0.3062, + "step": 1000 + }, + { + "epoch": 0.01, + "learning_rate": 4.98864752260562e-05, + "loss": 0.3029, + "step": 1200 + }, + { + "epoch": 0.01, + "learning_rate": 4.986755443039891e-05, + "loss": 0.3145, + "step": 1400 + }, + { + "epoch": 0.01, + "learning_rate": 4.984863363474161e-05, + "loss": 0.3028, + "step": 1600 + }, + { + "epoch": 0.01, + "learning_rate": 4.982971283908432e-05, + "loss": 0.2967, + "step": 1800 + }, + { + "epoch": 0.01, + "learning_rate": 4.981079204342701e-05, + "loss": 0.3027, + "step": 2000 + }, + { + "epoch": 0.01, + "eval_LOC_f1": 0.7291804435039456, + "eval_LOC_number": 22273, + "eval_LOC_precision": 0.7162343567314771, + "eval_LOC_recall": 0.7426031517981413, + "eval_ORG_f1": 0.5086950210480476, + "eval_ORG_number": 19008, + "eval_ORG_precision": 0.5675325516615923, + "eval_ORG_recall": 0.4609111952861953, + "eval_PER_f1": 0.7408791945125284, + "eval_PER_number": 27575, + "eval_PER_precision": 0.7210323997803405, + "eval_PER_recall": 0.7618495013599275, + "eval_loss": 0.2752716541290283, + "eval_overall_accuracy": 0.9174496871956154, + "eval_overall_f1": 0.6784108055844479, + "eval_overall_precision": 0.6843762007507463, + "eval_overall_recall": 0.6725485070291624, + "eval_runtime": 139.0881, + "eval_samples_per_second": 272.396, + "eval_steps_per_second": 34.05, + "step": 2000 + }, + { + "epoch": 0.01, + "learning_rate": 4.9791871247769715e-05, + "loss": 0.2976, + "step": 2200 + }, + { + "epoch": 0.01, + "learning_rate": 4.977295045211242e-05, + "loss": 0.29, + "step": 2400 + }, + { + "epoch": 0.01, + "learning_rate": 4.975402965645511e-05, + "loss": 0.2852, + "step": 2600 + }, + { + "epoch": 0.02, + "learning_rate": 4.973510886079782e-05, + "loss": 0.2827, + "step": 2800 + }, + { + "epoch": 0.02, + "learning_rate": 4.971618806514052e-05, + "loss": 0.2741, + "step": 3000 + }, + { + "epoch": 0.02, + "learning_rate": 4.969726726948322e-05, + "loss": 0.2766, + "step": 3200 + }, + { + "epoch": 0.02, + "learning_rate": 4.967834647382592e-05, + "loss": 0.2933, + "step": 3400 + }, + { + "epoch": 0.02, + "learning_rate": 4.965942567816862e-05, + "loss": 0.2791, + "step": 3600 + }, + { + "epoch": 0.02, + "learning_rate": 4.9640504882511327e-05, + "loss": 0.2848, + "step": 3800 + }, + { + "epoch": 0.02, + "learning_rate": 4.962158408685402e-05, + "loss": 0.2851, + "step": 4000 + }, + { + "epoch": 0.02, + "eval_LOC_f1": 0.7378174486305249, + "eval_LOC_number": 22273, + "eval_LOC_precision": 0.7446497164807024, + "eval_LOC_recall": 0.7311094149867553, + "eval_ORG_f1": 0.5609226230387555, + "eval_ORG_number": 19008, + "eval_ORG_precision": 0.572139847896263, + "eval_ORG_recall": 0.5501367845117845, + "eval_PER_f1": 0.7378623221163408, + "eval_PER_number": 27575, + "eval_PER_precision": 0.7358266012694749, + "eval_PER_recall": 0.739909338168631, + "eval_loss": 0.2609984874725342, + "eval_overall_accuracy": 0.9208337573516703, + "eval_overall_f1": 0.6895976713060141, + "eval_overall_precision": 0.6945913691747824, + "eval_overall_recall": 0.6846752643197398, + "eval_runtime": 188.2938, + "eval_samples_per_second": 201.212, + "eval_steps_per_second": 25.152, + "step": 4000 + }, + { + "epoch": 0.02, + "learning_rate": 4.9602663291196724e-05, + "loss": 0.2791, + "step": 4200 + }, + { + "epoch": 0.02, + "learning_rate": 4.958374249553943e-05, + "loss": 0.275, + "step": 4400 + }, + { + "epoch": 0.03, + "learning_rate": 4.956482169988212e-05, + "loss": 0.2751, + "step": 4600 + }, + { + "epoch": 0.03, + "learning_rate": 4.954590090422483e-05, + "loss": 0.2752, + "step": 4800 + }, + { + "epoch": 0.03, + "learning_rate": 4.952698010856753e-05, + "loss": 0.2823, + "step": 5000 + }, + { + "epoch": 0.03, + "learning_rate": 4.950805931291023e-05, + "loss": 0.2659, + "step": 5200 + }, + { + "epoch": 0.03, + "learning_rate": 4.948913851725293e-05, + "loss": 0.2753, + "step": 5400 + }, + { + "epoch": 0.03, + "learning_rate": 4.947021772159563e-05, + "loss": 0.2756, + "step": 5600 + }, + { + "epoch": 0.03, + "learning_rate": 4.9451296925938336e-05, + "loss": 0.2697, + "step": 5800 + }, + { + "epoch": 0.03, + "learning_rate": 4.943237613028103e-05, + "loss": 0.2678, + "step": 6000 + }, + { + "epoch": 0.03, + "eval_LOC_f1": 0.7411618404689998, + "eval_LOC_number": 22273, + "eval_LOC_precision": 0.7332483852541851, + "eval_LOC_recall": 0.7492479683922237, + "eval_ORG_f1": 0.5731545371696395, + "eval_ORG_number": 19008, + "eval_ORG_precision": 0.5674073310305717, + "eval_ORG_recall": 0.5790193602693603, + "eval_PER_f1": 0.7520376230971871, + "eval_PER_number": 27575, + "eval_PER_precision": 0.733733526530049, + "eval_PER_recall": 0.7712783318223028, + "eval_loss": 0.2554367482662201, + "eval_overall_accuracy": 0.922518523543921, + "eval_overall_f1": 0.6994671352447892, + "eval_overall_precision": 0.6882291754519131, + "eval_overall_recall": 0.7110781921691647, + "eval_runtime": 436.1324, + "eval_samples_per_second": 86.87, + "eval_steps_per_second": 10.859, + "step": 6000 + }, + { + "epoch": 0.04, + "learning_rate": 4.9413455334623734e-05, + "loss": 0.282, + "step": 6200 + }, + { + "epoch": 0.04, + "learning_rate": 4.939453453896644e-05, + "loss": 0.2671, + "step": 6400 + }, + { + "epoch": 0.04, + "learning_rate": 4.937561374330913e-05, + "loss": 0.265, + "step": 6600 + }, + { + "epoch": 0.04, + "learning_rate": 4.935669294765184e-05, + "loss": 0.2782, + "step": 6800 + }, + { + "epoch": 0.04, + "learning_rate": 4.933777215199454e-05, + "loss": 0.2623, + "step": 7000 + }, + { + "epoch": 0.04, + "learning_rate": 4.931885135633724e-05, + "loss": 0.2678, + "step": 7200 + }, + { + "epoch": 0.04, + "learning_rate": 4.929993056067994e-05, + "loss": 0.2704, + "step": 7400 + }, + { + "epoch": 0.04, + "learning_rate": 4.928100976502264e-05, + "loss": 0.2732, + "step": 7600 + }, + { + "epoch": 0.04, + "learning_rate": 4.9262088969365346e-05, + "loss": 0.2641, + "step": 7800 + }, + { + "epoch": 0.05, + "learning_rate": 4.924316817370804e-05, + "loss": 0.2575, + "step": 8000 + }, + { + "epoch": 0.05, + "eval_LOC_f1": 0.7363726036797187, + "eval_LOC_number": 22273, + "eval_LOC_precision": 0.7048150732728542, + "eval_LOC_recall": 0.7708885197324115, + "eval_ORG_f1": 0.5879352892788594, + "eval_ORG_number": 19008, + "eval_ORG_precision": 0.6139617455045241, + "eval_ORG_recall": 0.5640256734006734, + "eval_PER_f1": 0.7572004608294931, + "eval_PER_number": 27575, + "eval_PER_precision": 0.751760374593416, + "eval_PER_recall": 0.7627198549410699, + "eval_loss": 0.2507953643798828, + "eval_overall_accuracy": 0.9241984438123206, + "eval_overall_f1": 0.7056744749596123, + "eval_overall_precision": 0.7009025787965616, + "eval_overall_recall": 0.7105117927268503, + "eval_runtime": 195.0741, + "eval_samples_per_second": 194.218, + "eval_steps_per_second": 24.278, + "step": 8000 + }, + { + "epoch": 0.05, + "learning_rate": 4.9224247378050744e-05, + "loss": 0.2662, + "step": 8200 + }, + { + "epoch": 0.05, + "learning_rate": 4.9205326582393447e-05, + "loss": 0.2688, + "step": 8400 + }, + { + "epoch": 0.05, + "learning_rate": 4.918640578673614e-05, + "loss": 0.2603, + "step": 8600 + }, + { + "epoch": 0.05, + "learning_rate": 4.916748499107885e-05, + "loss": 0.2663, + "step": 8800 + }, + { + "epoch": 0.05, + "learning_rate": 4.914856419542155e-05, + "loss": 0.2556, + "step": 9000 + }, + { + "epoch": 0.05, + "learning_rate": 4.912964339976425e-05, + "loss": 0.2558, + "step": 9200 + }, + { + "epoch": 0.05, + "learning_rate": 4.911072260410695e-05, + "loss": 0.2589, + "step": 9400 + }, + { + "epoch": 0.05, + "learning_rate": 4.909180180844965e-05, + "loss": 0.2634, + "step": 9600 + }, + { + "epoch": 0.06, + "learning_rate": 4.9072881012792356e-05, + "loss": 0.2589, + "step": 9800 + }, + { + "epoch": 0.06, + "learning_rate": 4.905396021713505e-05, + "loss": 0.2594, + "step": 10000 + }, + { + "epoch": 0.06, + "eval_LOC_f1": 0.7529221745228886, + "eval_LOC_number": 22273, + "eval_LOC_precision": 0.7148218393123764, + "eval_LOC_recall": 0.7953127104566067, + "eval_ORG_f1": 0.5893841034313457, + "eval_ORG_number": 19008, + "eval_ORG_precision": 0.6119859521921378, + "eval_ORG_recall": 0.5683922558922558, + "eval_PER_f1": 0.761394484926418, + "eval_PER_number": 27575, + "eval_PER_precision": 0.7501143681599043, + "eval_PER_recall": 0.7730190389845875, + "eval_loss": 0.24650631844997406, + "eval_overall_accuracy": 0.9254713064772233, + "eval_overall_f1": 0.7134022389555359, + "eval_overall_precision": 0.7033534692034099, + "eval_overall_recall": 0.7237423027768096, + "eval_runtime": 178.5555, + "eval_samples_per_second": 212.186, + "eval_steps_per_second": 26.524, + "step": 10000 + }, + { + "epoch": 0.06, + "learning_rate": 4.9035039421477754e-05, + "loss": 0.2659, + "step": 10200 + }, + { + "epoch": 0.06, + "learning_rate": 4.9016118625820456e-05, + "loss": 0.2561, + "step": 10400 + }, + { + "epoch": 0.06, + "learning_rate": 4.899719783016315e-05, + "loss": 0.2576, + "step": 10600 + }, + { + "epoch": 0.06, + "learning_rate": 4.897827703450586e-05, + "loss": 0.2529, + "step": 10800 + }, + { + "epoch": 0.06, + "learning_rate": 4.895935623884856e-05, + "loss": 0.255, + "step": 11000 + }, + { + "epoch": 0.06, + "learning_rate": 4.894043544319126e-05, + "loss": 0.2616, + "step": 11200 + }, + { + "epoch": 0.06, + "learning_rate": 4.892151464753396e-05, + "loss": 0.2696, + "step": 11400 + }, + { + "epoch": 0.07, + "learning_rate": 4.890259385187666e-05, + "loss": 0.2631, + "step": 11600 + }, + { + "epoch": 0.07, + "learning_rate": 4.8883673056219366e-05, + "loss": 0.2522, + "step": 11800 + }, + { + "epoch": 0.07, + "learning_rate": 4.886475226056206e-05, + "loss": 0.256, + "step": 12000 + }, + { + "epoch": 0.07, + "eval_LOC_f1": 0.7575173391832303, + "eval_LOC_number": 22273, + "eval_LOC_precision": 0.7478561428071404, + "eval_LOC_recall": 0.7674314192071118, + "eval_ORG_f1": 0.5817149852540203, + "eval_ORG_number": 19008, + "eval_ORG_precision": 0.6173379000826739, + "eval_ORG_recall": 0.5499789562289562, + "eval_PER_f1": 0.7680199875078075, + "eval_PER_number": 27575, + "eval_PER_precision": 0.7560787069571329, + "eval_PER_recall": 0.7803445149592022, + "eval_loss": 0.24416784942150116, + "eval_overall_accuracy": 0.9262498849093085, + "eval_overall_f1": 0.7157236007176928, + "eval_overall_precision": 0.7189010989010989, + "eval_overall_recall": 0.7125740676193796, + "eval_runtime": 161.8057, + "eval_samples_per_second": 234.151, + "eval_steps_per_second": 29.27, + "step": 12000 + }, + { + "epoch": 0.07, + "learning_rate": 4.8845831464904764e-05, + "loss": 0.255, + "step": 12200 + }, + { + "epoch": 0.07, + "learning_rate": 4.8826910669247466e-05, + "loss": 0.2526, + "step": 12400 + }, + { + "epoch": 0.07, + "learning_rate": 4.880798987359017e-05, + "loss": 0.2509, + "step": 12600 + }, + { + "epoch": 0.07, + "learning_rate": 4.878906907793287e-05, + "loss": 0.2566, + "step": 12800 + }, + { + "epoch": 0.07, + "learning_rate": 4.8770148282275567e-05, + "loss": 0.252, + "step": 13000 + }, + { + "epoch": 0.07, + "learning_rate": 4.875122748661827e-05, + "loss": 0.2566, + "step": 13200 + }, + { + "epoch": 0.08, + "learning_rate": 4.873230669096097e-05, + "loss": 0.255, + "step": 13400 + }, + { + "epoch": 0.08, + "learning_rate": 4.8713385895303674e-05, + "loss": 0.252, + "step": 13600 + }, + { + "epoch": 0.08, + "learning_rate": 4.8694465099646376e-05, + "loss": 0.2542, + "step": 13800 + }, + { + "epoch": 0.08, + "learning_rate": 4.867554430398907e-05, + "loss": 0.2479, + "step": 14000 + }, + { + "epoch": 0.08, + "eval_LOC_f1": 0.7544113029827316, + "eval_LOC_number": 22273, + "eval_LOC_precision": 0.7066504587875461, + "eval_LOC_recall": 0.8090962151483859, + "eval_ORG_f1": 0.5945904053320933, + "eval_ORG_number": 19008, + "eval_ORG_precision": 0.5841327851378103, + "eval_ORG_recall": 0.6054292929292929, + "eval_PER_f1": 0.7661684443334642, + "eval_PER_number": 27575, + "eval_PER_precision": 0.7538520936436067, + "eval_PER_recall": 0.7788939256572983, + "eval_loss": 0.2448464334011078, + "eval_overall_accuracy": 0.9249754069364554, + "eval_overall_f1": 0.71563661873027, + "eval_overall_precision": 0.692145900616061, + "eval_overall_recall": 0.7407778552341118, + "eval_runtime": 131.2012, + "eval_samples_per_second": 288.77, + "eval_steps_per_second": 36.097, + "step": 14000 + }, + { + "epoch": 0.08, + "learning_rate": 4.8656623508331774e-05, + "loss": 0.2498, + "step": 14200 + }, + { + "epoch": 0.08, + "learning_rate": 4.8637702712674476e-05, + "loss": 0.2647, + "step": 14400 + }, + { + "epoch": 0.08, + "learning_rate": 4.861878191701718e-05, + "loss": 0.2533, + "step": 14600 + }, + { + "epoch": 0.08, + "learning_rate": 4.859986112135988e-05, + "loss": 0.2424, + "step": 14800 + }, + { + "epoch": 0.09, + "learning_rate": 4.8580940325702576e-05, + "loss": 0.2655, + "step": 15000 + }, + { + "epoch": 0.09, + "learning_rate": 4.856201953004528e-05, + "loss": 0.2571, + "step": 15200 + }, + { + "epoch": 0.09, + "learning_rate": 4.854309873438798e-05, + "loss": 0.2516, + "step": 15400 + }, + { + "epoch": 0.09, + "learning_rate": 4.8524177938730683e-05, + "loss": 0.2635, + "step": 15600 + }, + { + "epoch": 0.09, + "learning_rate": 4.8505257143073386e-05, + "loss": 0.2511, + "step": 15800 + }, + { + "epoch": 0.09, + "learning_rate": 4.848633634741608e-05, + "loss": 0.2445, + "step": 16000 + }, + { + "epoch": 0.09, + "eval_LOC_f1": 0.7596549610298625, + "eval_LOC_number": 22273, + "eval_LOC_precision": 0.741315215997949, + "eval_LOC_recall": 0.7789251560184978, + "eval_ORG_f1": 0.6013540005750581, + "eval_ORG_number": 19008, + "eval_ORG_precision": 0.5975894851680607, + "eval_ORG_recall": 0.6051662457912458, + "eval_PER_f1": 0.7720506513619386, + "eval_PER_number": 27575, + "eval_PER_precision": 0.7757964115708532, + "eval_PER_recall": 0.7683408884859474, + "eval_loss": 0.2392381727695465, + "eval_overall_accuracy": 0.9269638510233783, + "eval_overall_f1": 0.7209295624486739, + "eval_overall_precision": 0.7152311254681113, + "eval_overall_recall": 0.726719530614616, + "eval_runtime": 198.6724, + "eval_samples_per_second": 190.701, + "eval_steps_per_second": 23.838, + "step": 16000 + }, + { + "epoch": 0.09, + "learning_rate": 4.8467415551758784e-05, + "loss": 0.2484, + "step": 16200 + }, + { + "epoch": 0.09, + "learning_rate": 4.8448494756101486e-05, + "loss": 0.2629, + "step": 16400 + }, + { + "epoch": 0.09, + "learning_rate": 4.842957396044419e-05, + "loss": 0.2533, + "step": 16600 + }, + { + "epoch": 0.1, + "learning_rate": 4.841065316478689e-05, + "loss": 0.2532, + "step": 16800 + }, + { + "epoch": 0.1, + "learning_rate": 4.8391732369129586e-05, + "loss": 0.2558, + "step": 17000 + }, + { + "epoch": 0.1, + "learning_rate": 4.837281157347229e-05, + "loss": 0.2568, + "step": 17200 + }, + { + "epoch": 0.1, + "learning_rate": 4.835389077781499e-05, + "loss": 0.2572, + "step": 17400 + }, + { + "epoch": 0.1, + "learning_rate": 4.833496998215769e-05, + "loss": 0.2453, + "step": 17600 + }, + { + "epoch": 0.1, + "learning_rate": 4.8316049186500396e-05, + "loss": 0.2563, + "step": 17800 + }, + { + "epoch": 0.1, + "learning_rate": 4.829712839084309e-05, + "loss": 0.2488, + "step": 18000 + }, + { + "epoch": 0.1, + "eval_LOC_f1": 0.7634721568291742, + "eval_LOC_number": 22273, + "eval_LOC_precision": 0.7294401504927821, + "eval_LOC_recall": 0.8008350918152023, + "eval_ORG_f1": 0.5995581858346741, + "eval_ORG_number": 19008, + "eval_ORG_precision": 0.6224588028767201, + "eval_ORG_recall": 0.5782828282828283, + "eval_PER_f1": 0.769211550130278, + "eval_PER_number": 27575, + "eval_PER_precision": 0.7572648371341227, + "eval_PER_recall": 0.7815412511332729, + "eval_loss": 0.2373371720314026, + "eval_overall_accuracy": 0.9275695915047725, + "eval_overall_f1": 0.7226720792959758, + "eval_overall_precision": 0.7138909750464072, + "eval_overall_recall": 0.7316718949692111, + "eval_runtime": 274.4929, + "eval_samples_per_second": 138.025, + "eval_steps_per_second": 17.254, + "step": 18000 + }, + { + "epoch": 0.1, + "learning_rate": 4.8278207595185794e-05, + "loss": 0.2557, + "step": 18200 + }, + { + "epoch": 0.1, + "learning_rate": 4.8259286799528496e-05, + "loss": 0.2459, + "step": 18400 + }, + { + "epoch": 0.11, + "learning_rate": 4.82403660038712e-05, + "loss": 0.2504, + "step": 18600 + }, + { + "epoch": 0.11, + "learning_rate": 4.82214452082139e-05, + "loss": 0.2526, + "step": 18800 + }, + { + "epoch": 0.11, + "learning_rate": 4.8202524412556596e-05, + "loss": 0.2533, + "step": 19000 + }, + { + "epoch": 0.11, + "learning_rate": 4.81836036168993e-05, + "loss": 0.2398, + "step": 19200 + }, + { + "epoch": 0.11, + "learning_rate": 4.8164682821242e-05, + "loss": 0.2439, + "step": 19400 + }, + { + "epoch": 0.11, + "learning_rate": 4.81457620255847e-05, + "loss": 0.252, + "step": 19600 + }, + { + "epoch": 0.11, + "learning_rate": 4.8126841229927405e-05, + "loss": 0.2557, + "step": 19800 + }, + { + "epoch": 0.11, + "learning_rate": 4.81079204342701e-05, + "loss": 0.2454, + "step": 20000 + }, + { + "epoch": 0.11, + "eval_LOC_f1": 0.7657085488552784, + "eval_LOC_number": 22273, + "eval_LOC_precision": 0.74500976811348, + "eval_LOC_recall": 0.7875903560364567, + "eval_ORG_f1": 0.6059279573862139, + "eval_ORG_number": 19008, + "eval_ORG_precision": 0.6074336470339431, + "eval_ORG_recall": 0.6044297138047138, + "eval_PER_f1": 0.7715000798764577, + "eval_PER_number": 27575, + "eval_PER_precision": 0.7555802795354982, + "eval_PER_recall": 0.788105167724388, + "eval_loss": 0.23656196892261505, + "eval_overall_accuracy": 0.9271819175966802, + "eval_overall_f1": 0.7247819072231184, + "eval_overall_precision": 0.7127432534890904, + "eval_overall_recall": 0.737234227953991, + "eval_runtime": 296.8237, + "eval_samples_per_second": 127.641, + "eval_steps_per_second": 15.956, + "step": 20000 + }, + { + "epoch": 0.11, + "learning_rate": 4.80889996386128e-05, + "loss": 0.2426, + "step": 20200 + }, + { + "epoch": 0.12, + "learning_rate": 4.8070078842955506e-05, + "loss": 0.2581, + "step": 20400 + }, + { + "epoch": 0.12, + "learning_rate": 4.805115804729821e-05, + "loss": 0.2461, + "step": 20600 + }, + { + "epoch": 0.12, + "learning_rate": 4.803223725164091e-05, + "loss": 0.2432, + "step": 20800 + }, + { + "epoch": 0.12, + "learning_rate": 4.8013316455983606e-05, + "loss": 0.2519, + "step": 21000 + }, + { + "epoch": 0.12, + "learning_rate": 4.799439566032631e-05, + "loss": 0.2556, + "step": 21200 + }, + { + "epoch": 0.12, + "learning_rate": 4.797547486466901e-05, + "loss": 0.251, + "step": 21400 + }, + { + "epoch": 0.12, + "learning_rate": 4.795655406901171e-05, + "loss": 0.2482, + "step": 21600 + }, + { + "epoch": 0.12, + "learning_rate": 4.7937633273354415e-05, + "loss": 0.2377, + "step": 21800 + }, + { + "epoch": 0.12, + "learning_rate": 4.791871247769711e-05, + "loss": 0.2465, + "step": 22000 + }, + { + "epoch": 0.12, + "eval_LOC_f1": 0.7664768463338189, + "eval_LOC_number": 22273, + "eval_LOC_precision": 0.7538314592106977, + "eval_LOC_recall": 0.7795537197503704, + "eval_ORG_f1": 0.6054799667230228, + "eval_ORG_number": 19008, + "eval_ORG_precision": 0.6179676800876472, + "eval_ORG_recall": 0.5934869528619529, + "eval_PER_f1": 0.7700069035101694, + "eval_PER_number": 27575, + "eval_PER_precision": 0.7521267030915001, + "eval_PER_recall": 0.7887579329102448, + "eval_loss": 0.23523415625095367, + "eval_overall_accuracy": 0.9289700634977556, + "eval_overall_f1": 0.7247702463649307, + "eval_overall_precision": 0.7178018972737373, + "eval_overall_recall": 0.7318752178459393, + "eval_runtime": 167.8378, + "eval_samples_per_second": 225.736, + "eval_steps_per_second": 28.218, + "step": 22000 + }, + { + "epoch": 0.13, + "learning_rate": 4.789979168203981e-05, + "loss": 0.251, + "step": 22200 + }, + { + "epoch": 0.13, + "learning_rate": 4.7880870886382516e-05, + "loss": 0.2406, + "step": 22400 + }, + { + "epoch": 0.13, + "learning_rate": 4.786195009072522e-05, + "loss": 0.2454, + "step": 22600 + }, + { + "epoch": 0.13, + "learning_rate": 4.784302929506792e-05, + "loss": 0.2558, + "step": 22800 + }, + { + "epoch": 0.13, + "learning_rate": 4.7824108499410616e-05, + "loss": 0.2403, + "step": 23000 + }, + { + "epoch": 0.13, + "learning_rate": 4.780518770375332e-05, + "loss": 0.2399, + "step": 23200 + }, + { + "epoch": 0.13, + "learning_rate": 4.778626690809602e-05, + "loss": 0.2464, + "step": 23400 + }, + { + "epoch": 0.13, + "learning_rate": 4.776734611243872e-05, + "loss": 0.2364, + "step": 23600 + }, + { + "epoch": 0.14, + "learning_rate": 4.7748425316781425e-05, + "loss": 0.251, + "step": 23800 + }, + { + "epoch": 0.14, + "learning_rate": 4.772950452112413e-05, + "loss": 0.2426, + "step": 24000 + }, + { + "epoch": 0.14, + "eval_LOC_f1": 0.7657504256871808, + "eval_LOC_number": 22273, + "eval_LOC_precision": 0.7544884085759107, + "eval_LOC_recall": 0.7773537466888161, + "eval_ORG_f1": 0.60576382756683, + "eval_ORG_number": 19008, + "eval_ORG_precision": 0.6164025486031695, + "eval_ORG_recall": 0.5954861111111112, + "eval_PER_f1": 0.7737114865824223, + "eval_PER_number": 27575, + "eval_PER_precision": 0.7576558100733428, + "eval_PER_recall": 0.7904623753399819, + "eval_loss": 0.2315492182970047, + "eval_overall_accuracy": 0.9291396708325459, + "eval_overall_f1": 0.7259457591984799, + "eval_overall_precision": 0.7196061643835616, + "eval_overall_recall": 0.7323980481003834, + "eval_runtime": 150.589, + "eval_samples_per_second": 251.592, + "eval_steps_per_second": 31.45, + "step": 24000 + }, + { + "epoch": 0.14, + "learning_rate": 4.771058372546682e-05, + "loss": 0.238, + "step": 24200 + }, + { + "epoch": 0.14, + "learning_rate": 4.7691662929809525e-05, + "loss": 0.2398, + "step": 24400 + }, + { + "epoch": 0.14, + "learning_rate": 4.767274213415223e-05, + "loss": 0.2386, + "step": 24600 + }, + { + "epoch": 0.14, + "learning_rate": 4.765382133849493e-05, + "loss": 0.2447, + "step": 24800 + }, + { + "epoch": 0.14, + "learning_rate": 4.763490054283763e-05, + "loss": 0.2518, + "step": 25000 + }, + { + "epoch": 0.14, + "learning_rate": 4.761597974718033e-05, + "loss": 0.2365, + "step": 25200 + }, + { + "epoch": 0.14, + "learning_rate": 4.759705895152303e-05, + "loss": 0.2388, + "step": 25400 + }, + { + "epoch": 0.15, + "learning_rate": 4.757813815586573e-05, + "loss": 0.2442, + "step": 25600 + }, + { + "epoch": 0.15, + "learning_rate": 4.7559217360208435e-05, + "loss": 0.244, + "step": 25800 + }, + { + "epoch": 0.15, + "learning_rate": 4.754029656455114e-05, + "loss": 0.2472, + "step": 26000 + }, + { + "epoch": 0.15, + "eval_LOC_f1": 0.7681960975178089, + "eval_LOC_number": 22273, + "eval_LOC_precision": 0.7572070303981857, + "eval_LOC_recall": 0.779508822340951, + "eval_ORG_f1": 0.6061151079136691, + "eval_ORG_number": 19008, + "eval_ORG_precision": 0.628731343283582, + "eval_ORG_recall": 0.5850694444444444, + "eval_PER_f1": 0.7769911504424779, + "eval_PER_number": 27575, + "eval_PER_precision": 0.7902493905869117, + "eval_PER_recall": 0.7641704442429738, + "eval_loss": 0.23174811899662018, + "eval_overall_accuracy": 0.929897250261276, + "eval_overall_f1": 0.7280112826690563, + "eval_overall_precision": 0.736526857108885, + "eval_overall_recall": 0.7196903683048681, + "eval_runtime": 126.6126, + "eval_samples_per_second": 299.236, + "eval_steps_per_second": 37.405, + "step": 26000 + }, + { + "epoch": 0.15, + "learning_rate": 4.752137576889383e-05, + "loss": 0.2497, + "step": 26200 + }, + { + "epoch": 0.15, + "learning_rate": 4.7502454973236535e-05, + "loss": 0.2417, + "step": 26400 + }, + { + "epoch": 0.15, + "learning_rate": 4.748353417757924e-05, + "loss": 0.2375, + "step": 26600 + }, + { + "epoch": 0.15, + "learning_rate": 4.746461338192194e-05, + "loss": 0.2413, + "step": 26800 + }, + { + "epoch": 0.15, + "learning_rate": 4.744569258626464e-05, + "loss": 0.2489, + "step": 27000 + }, + { + "epoch": 0.15, + "learning_rate": 4.742677179060734e-05, + "loss": 0.2468, + "step": 27200 + }, + { + "epoch": 0.16, + "learning_rate": 4.740785099495004e-05, + "loss": 0.2482, + "step": 27400 + }, + { + "epoch": 0.16, + "learning_rate": 4.738893019929274e-05, + "loss": 0.2513, + "step": 27600 + }, + { + "epoch": 0.16, + "learning_rate": 4.7370009403635445e-05, + "loss": 0.2453, + "step": 27800 + }, + { + "epoch": 0.16, + "learning_rate": 4.735108860797815e-05, + "loss": 0.2474, + "step": 28000 + }, + { + "epoch": 0.16, + "eval_LOC_f1": 0.7756032765109586, + "eval_LOC_number": 22273, + "eval_LOC_precision": 0.7650347207057693, + "eval_LOC_recall": 0.7864679208009698, + "eval_ORG_f1": 0.6146051150072381, + "eval_ORG_number": 19008, + "eval_ORG_precision": 0.6265988848802886, + "eval_ORG_recall": 0.6030618686868687, + "eval_PER_f1": 0.7779559334858667, + "eval_PER_number": 27575, + "eval_PER_precision": 0.7729843906630388, + "eval_PER_recall": 0.7829918404351768, + "eval_loss": 0.2310045212507248, + "eval_overall_accuracy": 0.9300216289734556, + "eval_overall_f1": 0.7330245906985845, + "eval_overall_precision": 0.7316088711427455, + "eval_overall_recall": 0.7344457999302892, + "eval_runtime": 162.0502, + "eval_samples_per_second": 233.798, + "eval_steps_per_second": 29.226, + "step": 28000 + }, + { + "epoch": 0.16, + "learning_rate": 4.733216781232084e-05, + "loss": 0.2467, + "step": 28200 + }, + { + "epoch": 0.16, + "learning_rate": 4.7313247016663545e-05, + "loss": 0.2358, + "step": 28400 + }, + { + "epoch": 0.16, + "learning_rate": 4.729432622100625e-05, + "loss": 0.2483, + "step": 28600 + }, + { + "epoch": 0.16, + "learning_rate": 4.727540542534895e-05, + "loss": 0.2403, + "step": 28800 + }, + { + "epoch": 0.16, + "learning_rate": 4.725648462969165e-05, + "loss": 0.2357, + "step": 29000 + }, + { + "epoch": 0.17, + "learning_rate": 4.723756383403435e-05, + "loss": 0.2263, + "step": 29200 + }, + { + "epoch": 0.17, + "learning_rate": 4.721864303837705e-05, + "loss": 0.2394, + "step": 29400 + }, + { + "epoch": 0.17, + "learning_rate": 4.719972224271975e-05, + "loss": 0.2381, + "step": 29600 + }, + { + "epoch": 0.17, + "learning_rate": 4.7180801447062455e-05, + "loss": 0.2379, + "step": 29800 + }, + { + "epoch": 0.17, + "learning_rate": 4.716188065140516e-05, + "loss": 0.2431, + "step": 30000 + }, + { + "epoch": 0.17, + "eval_LOC_f1": 0.7725431677680665, + "eval_LOC_number": 22273, + "eval_LOC_precision": 0.735479157364939, + "eval_LOC_recall": 0.8135410586809141, + "eval_ORG_f1": 0.6201785145043035, + "eval_ORG_number": 19008, + "eval_ORG_precision": 0.6263683193818416, + "eval_ORG_recall": 0.6141098484848485, + "eval_PER_f1": 0.7792137953286609, + "eval_PER_number": 27575, + "eval_PER_precision": 0.7704168438962143, + "eval_PER_recall": 0.7882139619220309, + "eval_loss": 0.232611745595932, + "eval_overall_accuracy": 0.9300167830496044, + "eval_overall_f1": 0.7343256781696013, + "eval_overall_precision": 0.7208225501853536, + "eval_overall_recall": 0.7483443708609272, + "eval_runtime": 160.3069, + "eval_samples_per_second": 236.34, + "eval_steps_per_second": 29.543, + "step": 30000 + }, + { + "epoch": 0.17, + "learning_rate": 4.714295985574785e-05, + "loss": 0.2516, + "step": 30200 + }, + { + "epoch": 0.17, + "learning_rate": 4.7124039060090555e-05, + "loss": 0.238, + "step": 30400 + }, + { + "epoch": 0.17, + "learning_rate": 4.710511826443326e-05, + "loss": 0.2497, + "step": 30600 + }, + { + "epoch": 0.17, + "learning_rate": 4.708619746877596e-05, + "loss": 0.2443, + "step": 30800 + }, + { + "epoch": 0.18, + "learning_rate": 4.706727667311866e-05, + "loss": 0.2463, + "step": 31000 + }, + { + "epoch": 0.18, + "learning_rate": 4.704835587746136e-05, + "loss": 0.2485, + "step": 31200 + }, + { + "epoch": 0.18, + "learning_rate": 4.702943508180406e-05, + "loss": 0.2446, + "step": 31400 + }, + { + "epoch": 0.18, + "learning_rate": 4.701051428614676e-05, + "loss": 0.2505, + "step": 31600 + }, + { + "epoch": 0.18, + "learning_rate": 4.6991593490489465e-05, + "loss": 0.2499, + "step": 31800 + }, + { + "epoch": 0.18, + "learning_rate": 4.697267269483217e-05, + "loss": 0.2523, + "step": 32000 + }, + { + "epoch": 0.18, + "eval_LOC_f1": 0.7783391291035148, + "eval_LOC_number": 22273, + "eval_LOC_precision": 0.7588059701492538, + "eval_LOC_recall": 0.7989045032101648, + "eval_ORG_f1": 0.6195719477722439, + "eval_ORG_number": 19008, + "eval_ORG_precision": 0.6344489165793682, + "eval_ORG_recall": 0.6053766835016835, + "eval_PER_f1": 0.7759411816812826, + "eval_PER_number": 27575, + "eval_PER_precision": 0.7585382750259785, + "eval_PER_recall": 0.7941613780598368, + "eval_loss": 0.22620725631713867, + "eval_overall_accuracy": 0.9310974240684116, + "eval_overall_f1": 0.7350354956106034, + "eval_overall_precision": 0.7266843606738862, + "eval_overall_recall": 0.7435808063204369, + "eval_runtime": 119.6945, + "eval_samples_per_second": 316.531, + "eval_steps_per_second": 39.567, + "step": 32000 + }, + { + "epoch": 0.18, + "learning_rate": 4.695375189917486e-05, + "loss": 0.2411, + "step": 32200 + }, + { + "epoch": 0.18, + "learning_rate": 4.6934831103517565e-05, + "loss": 0.2348, + "step": 32400 + }, + { + "epoch": 0.19, + "learning_rate": 4.691591030786027e-05, + "loss": 0.2311, + "step": 32600 + }, + { + "epoch": 0.19, + "learning_rate": 4.689698951220297e-05, + "loss": 0.2398, + "step": 32800 + }, + { + "epoch": 0.19, + "learning_rate": 4.687806871654567e-05, + "loss": 0.2368, + "step": 33000 + }, + { + "epoch": 0.19, + "learning_rate": 4.685914792088837e-05, + "loss": 0.2517, + "step": 33200 + }, + { + "epoch": 0.19, + "learning_rate": 4.684022712523107e-05, + "loss": 0.2372, + "step": 33400 + }, + { + "epoch": 0.19, + "learning_rate": 4.682130632957377e-05, + "loss": 0.2338, + "step": 33600 + }, + { + "epoch": 0.19, + "learning_rate": 4.6802385533916475e-05, + "loss": 0.2489, + "step": 33800 + }, + { + "epoch": 0.19, + "learning_rate": 4.678346473825918e-05, + "loss": 0.2405, + "step": 34000 + }, + { + "epoch": 0.19, + "eval_LOC_f1": 0.7715908319569201, + "eval_LOC_number": 22273, + "eval_LOC_precision": 0.7777321656632001, + "eval_LOC_recall": 0.7655457280114938, + "eval_ORG_f1": 0.6044302913489203, + "eval_ORG_number": 19008, + "eval_ORG_precision": 0.644333273777619, + "eval_ORG_recall": 0.5691813973063973, + "eval_PER_f1": 0.7785559476257858, + "eval_PER_number": 27575, + "eval_PER_precision": 0.7692008211226729, + "eval_PER_recall": 0.7881414324569356, + "eval_loss": 0.2288392335176468, + "eval_overall_accuracy": 0.9304254559610516, + "eval_overall_f1": 0.7303957297993743, + "eval_overall_precision": 0.740685989039705, + "eval_overall_recall": 0.7203874753107935, + "eval_runtime": 110.0146, + "eval_samples_per_second": 344.381, + "eval_steps_per_second": 43.049, + "step": 34000 + }, + { + "epoch": 0.19, + "learning_rate": 4.676454394260187e-05, + "loss": 0.239, + "step": 34200 + }, + { + "epoch": 0.2, + "learning_rate": 4.674562314694458e-05, + "loss": 0.2468, + "step": 34400 + }, + { + "epoch": 0.2, + "learning_rate": 4.672670235128728e-05, + "loss": 0.2419, + "step": 34600 + }, + { + "epoch": 0.2, + "learning_rate": 4.670778155562998e-05, + "loss": 0.2428, + "step": 34800 + }, + { + "epoch": 0.2, + "learning_rate": 4.668886075997268e-05, + "loss": 0.2431, + "step": 35000 + }, + { + "epoch": 0.2, + "learning_rate": 4.666993996431538e-05, + "loss": 0.2312, + "step": 35200 + }, + { + "epoch": 0.2, + "learning_rate": 4.6651019168658086e-05, + "loss": 0.2338, + "step": 35400 + }, + { + "epoch": 0.2, + "learning_rate": 4.663209837300078e-05, + "loss": 0.2416, + "step": 35600 + }, + { + "epoch": 0.2, + "learning_rate": 4.6613177577343484e-05, + "loss": 0.2368, + "step": 35800 + }, + { + "epoch": 0.2, + "learning_rate": 4.659425678168619e-05, + "loss": 0.2364, + "step": 36000 + }, + { + "epoch": 0.2, + "eval_LOC_f1": 0.7726419998643853, + "eval_LOC_number": 22273, + "eval_LOC_precision": 0.7779699590350478, + "eval_LOC_recall": 0.7673865217976923, + "eval_ORG_f1": 0.6173080045461925, + "eval_ORG_number": 19008, + "eval_ORG_precision": 0.6355733868271481, + "eval_ORG_recall": 0.6000631313131313, + "eval_PER_f1": 0.7797609240851594, + "eval_PER_number": 27575, + "eval_PER_precision": 0.787298536152595, + "eval_PER_recall": 0.7723662737987307, + "eval_loss": 0.22748537361621857, + "eval_overall_accuracy": 0.9306839052331132, + "eval_overall_f1": 0.7332430203793143, + "eval_overall_precision": 0.7435790228168677, + "eval_overall_recall": 0.7231904263971186, + "eval_runtime": 438.1761, + "eval_samples_per_second": 86.465, + "eval_steps_per_second": 10.808, + "step": 36000 + }, + { + "epoch": 0.21, + "learning_rate": 4.657533598602888e-05, + "loss": 0.252, + "step": 36200 + }, + { + "epoch": 0.21, + "learning_rate": 4.655641519037159e-05, + "loss": 0.2409, + "step": 36400 + }, + { + "epoch": 0.21, + "learning_rate": 4.653749439471429e-05, + "loss": 0.2331, + "step": 36600 + }, + { + "epoch": 0.21, + "learning_rate": 4.651857359905699e-05, + "loss": 0.2409, + "step": 36800 + }, + { + "epoch": 0.21, + "learning_rate": 4.649965280339969e-05, + "loss": 0.2398, + "step": 37000 + }, + { + "epoch": 0.21, + "learning_rate": 4.648073200774239e-05, + "loss": 0.2478, + "step": 37200 + }, + { + "epoch": 0.21, + "learning_rate": 4.6461811212085096e-05, + "loss": 0.2377, + "step": 37400 + }, + { + "epoch": 0.21, + "learning_rate": 4.644289041642779e-05, + "loss": 0.2415, + "step": 37600 + }, + { + "epoch": 0.21, + "learning_rate": 4.6423969620770494e-05, + "loss": 0.2384, + "step": 37800 + }, + { + "epoch": 0.22, + "learning_rate": 4.6405048825113197e-05, + "loss": 0.2408, + "step": 38000 + }, + { + "epoch": 0.22, + "eval_LOC_f1": 0.7764915405164738, + "eval_LOC_number": 22273, + "eval_LOC_precision": 0.7520403870424905, + "eval_LOC_recall": 0.8025860907825618, + "eval_ORG_f1": 0.6271813275621182, + "eval_ORG_number": 19008, + "eval_ORG_precision": 0.6543012289225493, + "eval_ORG_recall": 0.6022201178451179, + "eval_PER_f1": 0.77649141440963, + "eval_PER_number": 27575, + "eval_PER_precision": 0.7584990489365381, + "eval_PER_recall": 0.7953581142339076, + "eval_loss": 0.2254871129989624, + "eval_overall_accuracy": 0.931431792814141, + "eval_overall_f1": 0.7372910613078627, + "eval_overall_precision": 0.7303362781419208, + "eval_overall_recall": 0.7443795747647264, + "eval_runtime": 531.2441, + "eval_samples_per_second": 71.317, + "eval_steps_per_second": 8.915, + "step": 38000 + }, + { + "epoch": 0.22, + "learning_rate": 4.638612802945589e-05, + "loss": 0.229, + "step": 38200 + }, + { + "epoch": 0.22, + "learning_rate": 4.63672072337986e-05, + "loss": 0.236, + "step": 38400 + }, + { + "epoch": 0.22, + "learning_rate": 4.63482864381413e-05, + "loss": 0.24, + "step": 38600 + }, + { + "epoch": 0.22, + "learning_rate": 4.6329365642484e-05, + "loss": 0.2404, + "step": 38800 + }, + { + "epoch": 0.22, + "learning_rate": 4.63104448468267e-05, + "loss": 0.2259, + "step": 39000 + }, + { + "epoch": 0.22, + "learning_rate": 4.62915240511694e-05, + "loss": 0.2463, + "step": 39200 + }, + { + "epoch": 0.22, + "learning_rate": 4.6272603255512106e-05, + "loss": 0.2425, + "step": 39400 + }, + { + "epoch": 0.22, + "learning_rate": 4.62536824598548e-05, + "loss": 0.2468, + "step": 39600 + }, + { + "epoch": 0.23, + "learning_rate": 4.6234761664197504e-05, + "loss": 0.2309, + "step": 39800 + }, + { + "epoch": 0.23, + "learning_rate": 4.6215840868540206e-05, + "loss": 0.233, + "step": 40000 + }, + { + "epoch": 0.23, + "eval_LOC_f1": 0.7791133514282021, + "eval_LOC_number": 22273, + "eval_LOC_precision": 0.7488508840945794, + "eval_LOC_recall": 0.8119247519418129, + "eval_ORG_f1": 0.6227407677483202, + "eval_ORG_number": 19008, + "eval_ORG_precision": 0.6166064981949458, + "eval_ORG_recall": 0.6289983164983165, + "eval_PER_f1": 0.7827357577307348, + "eval_PER_number": 27575, + "eval_PER_precision": 0.7885110767645543, + "eval_PER_recall": 0.7770444242973709, + "eval_loss": 0.2256794571876526, + "eval_overall_accuracy": 0.9313946407312822, + "eval_overall_f1": 0.7375133446538988, + "eval_overall_precision": 0.7278293948778867, + "eval_overall_recall": 0.747458464040897, + "eval_runtime": 309.4197, + "eval_samples_per_second": 122.445, + "eval_steps_per_second": 15.306, + "step": 40000 + }, + { + "epoch": 0.23, + "learning_rate": 4.61969200728829e-05, + "loss": 0.2465, + "step": 40200 + }, + { + "epoch": 0.23, + "learning_rate": 4.617799927722561e-05, + "loss": 0.2497, + "step": 40400 + }, + { + "epoch": 0.23, + "learning_rate": 4.615907848156831e-05, + "loss": 0.2331, + "step": 40600 + }, + { + "epoch": 0.23, + "learning_rate": 4.614015768591101e-05, + "loss": 0.2469, + "step": 40800 + }, + { + "epoch": 0.23, + "learning_rate": 4.612123689025371e-05, + "loss": 0.2439, + "step": 41000 + }, + { + "epoch": 0.23, + "learning_rate": 4.610231609459641e-05, + "loss": 0.2366, + "step": 41200 + }, + { + "epoch": 0.23, + "learning_rate": 4.6083395298939116e-05, + "loss": 0.234, + "step": 41400 + }, + { + "epoch": 0.24, + "learning_rate": 4.606447450328181e-05, + "loss": 0.2377, + "step": 41600 + }, + { + "epoch": 0.24, + "learning_rate": 4.6045553707624514e-05, + "loss": 0.246, + "step": 41800 + }, + { + "epoch": 0.24, + "learning_rate": 4.6026632911967216e-05, + "loss": 0.2322, + "step": 42000 + }, + { + "epoch": 0.24, + "eval_LOC_f1": 0.7771173520118434, + "eval_LOC_number": 22273, + "eval_LOC_precision": 0.765097459101984, + "eval_LOC_recall": 0.7895209446414941, + "eval_ORG_f1": 0.6246006811521504, + "eval_ORG_number": 19008, + "eval_ORG_precision": 0.6269012666278022, + "eval_ORG_recall": 0.6223169191919192, + "eval_PER_f1": 0.7819866392413127, + "eval_PER_number": 27575, + "eval_PER_precision": 0.7850303340399094, + "eval_PER_recall": 0.7789664551223935, + "eval_loss": 0.22573354840278625, + "eval_overall_accuracy": 0.9309488157369762, + "eval_overall_f1": 0.7372149111688914, + "eval_overall_precision": 0.7353030412482844, + "eval_overall_recall": 0.7391367491576624, + "eval_runtime": 440.2453, + "eval_samples_per_second": 86.059, + "eval_steps_per_second": 10.758, + "step": 42000 + }, + { + "epoch": 0.24, + "learning_rate": 4.600771211630991e-05, + "loss": 0.2403, + "step": 42200 + }, + { + "epoch": 0.24, + "learning_rate": 4.598879132065262e-05, + "loss": 0.2526, + "step": 42400 + }, + { + "epoch": 0.24, + "learning_rate": 4.5969870524995317e-05, + "loss": 0.2394, + "step": 42600 + }, + { + "epoch": 0.24, + "learning_rate": 4.595094972933802e-05, + "loss": 0.2319, + "step": 42800 + }, + { + "epoch": 0.24, + "learning_rate": 4.593202893368072e-05, + "loss": 0.2274, + "step": 43000 + }, + { + "epoch": 0.25, + "learning_rate": 4.591310813802342e-05, + "loss": 0.2353, + "step": 43200 + }, + { + "epoch": 0.25, + "learning_rate": 4.5894187342366126e-05, + "loss": 0.2356, + "step": 43400 + }, + { + "epoch": 0.25, + "learning_rate": 4.587526654670882e-05, + "loss": 0.2281, + "step": 43600 + }, + { + "epoch": 0.25, + "learning_rate": 4.5856345751051524e-05, + "loss": 0.2322, + "step": 43800 + }, + { + "epoch": 0.25, + "learning_rate": 4.5837424955394226e-05, + "loss": 0.2309, + "step": 44000 + }, + { + "epoch": 0.25, + "eval_LOC_f1": 0.7802323033092263, + "eval_LOC_number": 22273, + "eval_LOC_precision": 0.7621269854861498, + "eval_LOC_recall": 0.7992187850761011, + "eval_ORG_f1": 0.6253208096309035, + "eval_ORG_number": 19008, + "eval_ORG_precision": 0.6289987757491883, + "eval_ORG_recall": 0.6216856060606061, + "eval_PER_f1": 0.7842046718576196, + "eval_PER_number": 27575, + "eval_PER_precision": 0.7760181798814046, + "eval_PER_recall": 0.7925657298277425, + "eval_loss": 0.22290264070034027, + "eval_overall_accuracy": 0.9323282887266043, + "eval_overall_f1": 0.7397512269784354, + "eval_overall_precision": 0.7321177123188303, + "eval_overall_recall": 0.7475456024166376, + "eval_runtime": 544.863, + "eval_samples_per_second": 69.535, + "eval_steps_per_second": 8.692, + "step": 44000 + }, + { + "epoch": 0.25, + "learning_rate": 4.581850415973693e-05, + "loss": 0.2322, + "step": 44200 + }, + { + "epoch": 0.25, + "learning_rate": 4.579958336407963e-05, + "loss": 0.2331, + "step": 44400 + }, + { + "epoch": 0.25, + "learning_rate": 4.5780662568422326e-05, + "loss": 0.2385, + "step": 44600 + }, + { + "epoch": 0.25, + "learning_rate": 4.5761741772765036e-05, + "loss": 0.2333, + "step": 44800 + }, + { + "epoch": 0.26, + "learning_rate": 4.574282097710773e-05, + "loss": 0.2361, + "step": 45000 + }, + { + "epoch": 0.26, + "learning_rate": 4.5723900181450433e-05, + "loss": 0.2295, + "step": 45200 + }, + { + "epoch": 0.26, + "learning_rate": 4.5704979385793136e-05, + "loss": 0.2348, + "step": 45400 + }, + { + "epoch": 0.26, + "learning_rate": 4.568605859013583e-05, + "loss": 0.2321, + "step": 45600 + }, + { + "epoch": 0.26, + "learning_rate": 4.566713779447854e-05, + "loss": 0.2433, + "step": 45800 + }, + { + "epoch": 0.26, + "learning_rate": 4.5648216998821236e-05, + "loss": 0.2354, + "step": 46000 + }, + { + "epoch": 0.26, + "eval_LOC_f1": 0.7749009133206961, + "eval_LOC_number": 22273, + "eval_LOC_precision": 0.744772473189516, + "eval_LOC_recall": 0.8075697032281237, + "eval_ORG_f1": 0.6286089238845145, + "eval_ORG_number": 19008, + "eval_ORG_precision": 0.6272260632725749, + "eval_ORG_recall": 0.6299978956228957, + "eval_PER_f1": 0.7853058582371346, + "eval_PER_number": 27575, + "eval_PER_precision": 0.7788080887335497, + "eval_PER_recall": 0.7919129646418858, + "eval_loss": 0.22265663743019104, + "eval_overall_accuracy": 0.9320278414478328, + "eval_overall_f1": 0.7392570180821761, + "eval_overall_precision": 0.7266771414943464, + "eval_overall_recall": 0.752280120831881, + "eval_runtime": 162.1959, + "eval_samples_per_second": 233.588, + "eval_steps_per_second": 29.199, + "step": 46000 + }, + { + "epoch": 0.26, + "learning_rate": 4.562929620316394e-05, + "loss": 0.2308, + "step": 46200 + }, + { + "epoch": 0.26, + "learning_rate": 4.561037540750664e-05, + "loss": 0.2306, + "step": 46400 + }, + { + "epoch": 0.26, + "learning_rate": 4.5591454611849336e-05, + "loss": 0.2303, + "step": 46600 + }, + { + "epoch": 0.27, + "learning_rate": 4.5572533816192045e-05, + "loss": 0.2372, + "step": 46800 + }, + { + "epoch": 0.27, + "learning_rate": 4.555361302053474e-05, + "loss": 0.2418, + "step": 47000 + }, + { + "epoch": 0.27, + "learning_rate": 4.553469222487744e-05, + "loss": 0.2296, + "step": 47200 + }, + { + "epoch": 0.27, + "learning_rate": 4.5515771429220146e-05, + "loss": 0.2326, + "step": 47400 + }, + { + "epoch": 0.27, + "learning_rate": 4.549685063356284e-05, + "loss": 0.2246, + "step": 47600 + }, + { + "epoch": 0.27, + "learning_rate": 4.547792983790555e-05, + "loss": 0.2283, + "step": 47800 + }, + { + "epoch": 0.27, + "learning_rate": 4.5459009042248246e-05, + "loss": 0.2357, + "step": 48000 + }, + { + "epoch": 0.27, + "eval_LOC_f1": 0.779420409762895, + "eval_LOC_number": 22273, + "eval_LOC_precision": 0.7566367940480216, + "eval_LOC_recall": 0.8036187311992098, + "eval_ORG_f1": 0.6270356149874222, + "eval_ORG_number": 19008, + "eval_ORG_precision": 0.6312310070906861, + "eval_ORG_recall": 0.622895622895623, + "eval_PER_f1": 0.7834441709225851, + "eval_PER_number": 27575, + "eval_PER_precision": 0.79388659294836, + "eval_PER_recall": 0.7732728921124207, + "eval_loss": 0.22294096648693085, + "eval_overall_accuracy": 0.932032687371684, + "eval_overall_f1": 0.7393432178848603, + "eval_overall_precision": 0.7371232243908072, + "eval_overall_recall": 0.7415766236784013, + "eval_runtime": 375.1458, + "eval_samples_per_second": 100.993, + "eval_steps_per_second": 12.624, + "step": 48000 + }, + { + "epoch": 0.27, + "learning_rate": 4.544008824659095e-05, + "loss": 0.2398, + "step": 48200 + }, + { + "epoch": 0.27, + "learning_rate": 4.542116745093365e-05, + "loss": 0.228, + "step": 48400 + }, + { + "epoch": 0.28, + "learning_rate": 4.5402246655276346e-05, + "loss": 0.2366, + "step": 48600 + }, + { + "epoch": 0.28, + "learning_rate": 4.5383325859619055e-05, + "loss": 0.2316, + "step": 48800 + }, + { + "epoch": 0.28, + "learning_rate": 4.536440506396175e-05, + "loss": 0.2375, + "step": 49000 + }, + { + "epoch": 0.28, + "learning_rate": 4.534548426830445e-05, + "loss": 0.229, + "step": 49200 + }, + { + "epoch": 0.28, + "learning_rate": 4.5326563472647156e-05, + "loss": 0.2358, + "step": 49400 + }, + { + "epoch": 0.28, + "learning_rate": 4.530764267698985e-05, + "loss": 0.2399, + "step": 49600 + }, + { + "epoch": 0.28, + "learning_rate": 4.528872188133256e-05, + "loss": 0.2353, + "step": 49800 + }, + { + "epoch": 0.28, + "learning_rate": 4.5269801085675256e-05, + "loss": 0.2445, + "step": 50000 + }, + { + "epoch": 0.28, + "eval_LOC_f1": 0.7807140972649723, + "eval_LOC_number": 22273, + "eval_LOC_precision": 0.7649390375253112, + "eval_LOC_recall": 0.7971535042428052, + "eval_ORG_f1": 0.6239648649363707, + "eval_ORG_number": 19008, + "eval_ORG_precision": 0.6276012560540742, + "eval_ORG_recall": 0.6203703703703703, + "eval_PER_f1": 0.7826744982162804, + "eval_PER_number": 27575, + "eval_PER_precision": 0.7777419701364271, + "eval_PER_recall": 0.7876699909338168, + "eval_loss": 0.22286711633205414, + "eval_overall_accuracy": 0.9314964051321564, + "eval_overall_f1": 0.7388080672704871, + "eval_overall_precision": 0.7331502852975246, + "eval_overall_recall": 0.7445538515162078, + "eval_runtime": 222.4629, + "eval_samples_per_second": 170.307, + "eval_steps_per_second": 21.289, + "step": 50000 + }, + { + "epoch": 0.28, + "learning_rate": 4.525088029001796e-05, + "loss": 0.2255, + "step": 50200 + }, + { + "epoch": 0.29, + "learning_rate": 4.523195949436066e-05, + "loss": 0.2439, + "step": 50400 + }, + { + "epoch": 0.29, + "learning_rate": 4.5213038698703356e-05, + "loss": 0.2335, + "step": 50600 + }, + { + "epoch": 0.29, + "learning_rate": 4.5194117903046065e-05, + "loss": 0.2429, + "step": 50800 + }, + { + "epoch": 0.29, + "learning_rate": 4.517519710738876e-05, + "loss": 0.233, + "step": 51000 + }, + { + "epoch": 0.29, + "learning_rate": 4.515627631173146e-05, + "loss": 0.2383, + "step": 51200 + }, + { + "epoch": 0.29, + "learning_rate": 4.5137355516074165e-05, + "loss": 0.2441, + "step": 51400 + }, + { + "epoch": 0.29, + "learning_rate": 4.511843472041686e-05, + "loss": 0.2304, + "step": 51600 + }, + { + "epoch": 0.29, + "learning_rate": 4.509951392475957e-05, + "loss": 0.2387, + "step": 51800 + }, + { + "epoch": 0.3, + "learning_rate": 4.5080593129102266e-05, + "loss": 0.2293, + "step": 52000 + }, + { + "epoch": 0.3, + "eval_LOC_f1": 0.7795287472783323, + "eval_LOC_number": 22273, + "eval_LOC_precision": 0.7497719167288712, + "eval_LOC_recall": 0.8117451623041351, + "eval_ORG_f1": 0.6240258142142252, + "eval_ORG_number": 19008, + "eval_ORG_precision": 0.649735208701099, + "eval_ORG_recall": 0.600273569023569, + "eval_PER_f1": 0.7842404838135896, + "eval_PER_number": 27575, + "eval_PER_precision": 0.7695932972595566, + "eval_PER_recall": 0.799456029011786, + "eval_loss": 0.22172412276268005, + "eval_overall_accuracy": 0.9322701376403905, + "eval_overall_f1": 0.7405730873139047, + "eval_overall_precision": 0.7328640500568828, + "eval_overall_recall": 0.7484460322992913, + "eval_runtime": 211.5206, + "eval_samples_per_second": 179.117, + "eval_steps_per_second": 22.39, + "step": 52000 + }, + { + "epoch": 0.3, + "learning_rate": 4.506167233344497e-05, + "loss": 0.2247, + "step": 52200 + }, + { + "epoch": 0.3, + "learning_rate": 4.504275153778767e-05, + "loss": 0.2408, + "step": 52400 + }, + { + "epoch": 0.3, + "learning_rate": 4.5023830742130366e-05, + "loss": 0.2315, + "step": 52600 + }, + { + "epoch": 0.3, + "learning_rate": 4.5004909946473075e-05, + "loss": 0.2241, + "step": 52800 + }, + { + "epoch": 0.3, + "learning_rate": 4.498598915081577e-05, + "loss": 0.2393, + "step": 53000 + }, + { + "epoch": 0.3, + "learning_rate": 4.496706835515847e-05, + "loss": 0.2225, + "step": 53200 + }, + { + "epoch": 0.3, + "learning_rate": 4.4948147559501175e-05, + "loss": 0.232, + "step": 53400 + }, + { + "epoch": 0.3, + "learning_rate": 4.492922676384387e-05, + "loss": 0.2358, + "step": 53600 + }, + { + "epoch": 0.31, + "learning_rate": 4.491030596818658e-05, + "loss": 0.222, + "step": 53800 + }, + { + "epoch": 0.31, + "learning_rate": 4.4891385172529275e-05, + "loss": 0.2356, + "step": 54000 + }, + { + "epoch": 0.31, + "eval_LOC_f1": 0.7775278955107688, + "eval_LOC_number": 22273, + "eval_LOC_precision": 0.7499895707313003, + "eval_LOC_recall": 0.8071656265433484, + "eval_ORG_f1": 0.6236524963580983, + "eval_ORG_number": 19008, + "eval_ORG_precision": 0.6279938123433083, + "eval_ORG_recall": 0.6193707912457912, + "eval_PER_f1": 0.7870678431939823, + "eval_PER_number": 27575, + "eval_PER_precision": 0.7848822532366836, + "eval_PER_recall": 0.7892656391659112, + "eval_loss": 0.2190094292163849, + "eval_overall_accuracy": 0.9324510521308335, + "eval_overall_f1": 0.7396107765087614, + "eval_overall_precision": 0.7312589606370746, + "eval_overall_recall": 0.7481555710468224, + "eval_runtime": 269.0042, + "eval_samples_per_second": 140.842, + "eval_steps_per_second": 17.606, + "step": 54000 + }, + { + "epoch": 0.31, + "learning_rate": 4.487246437687198e-05, + "loss": 0.245, + "step": 54200 + }, + { + "epoch": 0.31, + "learning_rate": 4.485354358121468e-05, + "loss": 0.2269, + "step": 54400 + }, + { + "epoch": 0.31, + "learning_rate": 4.4834622785557376e-05, + "loss": 0.2404, + "step": 54600 + }, + { + "epoch": 0.31, + "learning_rate": 4.4815701989900085e-05, + "loss": 0.2307, + "step": 54800 + }, + { + "epoch": 0.31, + "learning_rate": 4.479678119424278e-05, + "loss": 0.2429, + "step": 55000 + }, + { + "epoch": 0.31, + "learning_rate": 4.477786039858548e-05, + "loss": 0.2257, + "step": 55200 + }, + { + "epoch": 0.31, + "learning_rate": 4.4758939602928185e-05, + "loss": 0.2406, + "step": 55400 + }, + { + "epoch": 0.32, + "learning_rate": 4.474001880727088e-05, + "loss": 0.2294, + "step": 55600 + }, + { + "epoch": 0.32, + "learning_rate": 4.472109801161359e-05, + "loss": 0.2359, + "step": 55800 + }, + { + "epoch": 0.32, + "learning_rate": 4.4702177215956285e-05, + "loss": 0.2349, + "step": 56000 + }, + { + "epoch": 0.32, + "eval_LOC_f1": 0.7808558706762301, + "eval_LOC_number": 22273, + "eval_LOC_precision": 0.7633587786259542, + "eval_LOC_recall": 0.7991738876666816, + "eval_ORG_f1": 0.6248509161877914, + "eval_ORG_number": 19008, + "eval_ORG_precision": 0.6444866920152091, + "eval_ORG_recall": 0.6063762626262627, + "eval_PER_f1": 0.7867336247696809, + "eval_PER_number": 27575, + "eval_PER_precision": 0.783788647734226, + "eval_PER_recall": 0.7897008159564823, + "eval_loss": 0.22081419825553894, + "eval_overall_accuracy": 0.9328807240456357, + "eval_overall_f1": 0.7414629899667008, + "eval_overall_precision": 0.7407697325505544, + "eval_overall_recall": 0.7421575461833392, + "eval_runtime": 123.1958, + "eval_samples_per_second": 307.535, + "eval_steps_per_second": 38.443, + "step": 56000 + }, + { + "epoch": 0.32, + "learning_rate": 4.4683256420298994e-05, + "loss": 0.2302, + "step": 56200 + }, + { + "epoch": 0.32, + "learning_rate": 4.466433562464169e-05, + "loss": 0.2384, + "step": 56400 + }, + { + "epoch": 0.32, + "learning_rate": 4.4645414828984386e-05, + "loss": 0.2337, + "step": 56600 + }, + { + "epoch": 0.32, + "learning_rate": 4.4626494033327095e-05, + "loss": 0.233, + "step": 56800 + }, + { + "epoch": 0.32, + "learning_rate": 4.460757323766979e-05, + "loss": 0.22, + "step": 57000 + }, + { + "epoch": 0.32, + "learning_rate": 4.45886524420125e-05, + "loss": 0.2318, + "step": 57200 + }, + { + "epoch": 0.33, + "learning_rate": 4.4569731646355195e-05, + "loss": 0.2347, + "step": 57400 + }, + { + "epoch": 0.33, + "learning_rate": 4.455081085069789e-05, + "loss": 0.2312, + "step": 57600 + }, + { + "epoch": 0.33, + "learning_rate": 4.45318900550406e-05, + "loss": 0.2289, + "step": 57800 + }, + { + "epoch": 0.33, + "learning_rate": 4.4512969259383295e-05, + "loss": 0.2307, + "step": 58000 + }, + { + "epoch": 0.33, + "eval_LOC_f1": 0.7813921698455755, + "eval_LOC_number": 22273, + "eval_LOC_precision": 0.7452130693391998, + "eval_LOC_recall": 0.8212634131010641, + "eval_ORG_f1": 0.6310557003519898, + "eval_ORG_number": 19008, + "eval_ORG_precision": 0.6449008731945741, + "eval_ORG_recall": 0.6177925084175084, + "eval_PER_f1": 0.7863532870793667, + "eval_PER_number": 27575, + "eval_PER_precision": 0.7775847397532265, + "eval_PER_recall": 0.79532184950136, + "eval_loss": 0.2236645668745041, + "eval_overall_accuracy": 0.932591583922517, + "eval_overall_f1": 0.7433537174122948, + "eval_overall_precision": 0.7323383925929058, + "eval_overall_recall": 0.7547054722899965, + "eval_runtime": 404.9137, + "eval_samples_per_second": 93.568, + "eval_steps_per_second": 11.696, + "step": 58000 + }, + { + "epoch": 0.33, + "learning_rate": 4.4494048463726004e-05, + "loss": 0.2389, + "step": 58200 + }, + { + "epoch": 0.33, + "learning_rate": 4.44751276680687e-05, + "loss": 0.2328, + "step": 58400 + }, + { + "epoch": 0.33, + "learning_rate": 4.44562068724114e-05, + "loss": 0.2342, + "step": 58600 + }, + { + "epoch": 0.33, + "learning_rate": 4.4437286076754105e-05, + "loss": 0.2289, + "step": 58800 + }, + { + "epoch": 0.33, + "learning_rate": 4.44183652810968e-05, + "loss": 0.227, + "step": 59000 + }, + { + "epoch": 0.34, + "learning_rate": 4.439944448543951e-05, + "loss": 0.2314, + "step": 59200 + }, + { + "epoch": 0.34, + "learning_rate": 4.4380523689782205e-05, + "loss": 0.2412, + "step": 59400 + }, + { + "epoch": 0.34, + "learning_rate": 4.436160289412491e-05, + "loss": 0.2447, + "step": 59600 + }, + { + "epoch": 0.34, + "learning_rate": 4.434268209846761e-05, + "loss": 0.2336, + "step": 59800 + }, + { + "epoch": 0.34, + "learning_rate": 4.4323761302810305e-05, + "loss": 0.2258, + "step": 60000 + }, + { + "epoch": 0.34, + "eval_LOC_f1": 0.783042830866252, + "eval_LOC_number": 22273, + "eval_LOC_precision": 0.7642658687753794, + "eval_LOC_recall": 0.8027656804202398, + "eval_ORG_f1": 0.6313017306245297, + "eval_ORG_number": 19008, + "eval_ORG_precision": 0.6452428037793891, + "eval_ORG_recall": 0.6179503367003367, + "eval_PER_f1": 0.7859255629956267, + "eval_PER_number": 27575, + "eval_PER_precision": 0.7865392997239575, + "eval_PER_recall": 0.7853127833182231, + "eval_loss": 0.2206861674785614, + "eval_overall_accuracy": 0.9328354954230249, + "eval_overall_f1": 0.7432729170139216, + "eval_overall_precision": 0.7417945639438168, + "eval_overall_recall": 0.744757174392936, + "eval_runtime": 161.0845, + "eval_samples_per_second": 235.2, + "eval_steps_per_second": 29.401, + "step": 60000 + }, + { + "epoch": 0.34, + "learning_rate": 4.4304840507153014e-05, + "loss": 0.2304, + "step": 60200 + }, + { + "epoch": 0.34, + "learning_rate": 4.428591971149571e-05, + "loss": 0.2331, + "step": 60400 + }, + { + "epoch": 0.34, + "learning_rate": 4.426699891583841e-05, + "loss": 0.2382, + "step": 60600 + }, + { + "epoch": 0.35, + "learning_rate": 4.4248078120181114e-05, + "loss": 0.221, + "step": 60800 + }, + { + "epoch": 0.35, + "learning_rate": 4.422915732452381e-05, + "loss": 0.2256, + "step": 61000 + }, + { + "epoch": 0.35, + "learning_rate": 4.421023652886652e-05, + "loss": 0.2295, + "step": 61200 + }, + { + "epoch": 0.35, + "learning_rate": 4.4191315733209215e-05, + "loss": 0.2476, + "step": 61400 + }, + { + "epoch": 0.35, + "learning_rate": 4.417239493755192e-05, + "loss": 0.2383, + "step": 61600 + }, + { + "epoch": 0.35, + "learning_rate": 4.415347414189462e-05, + "loss": 0.2293, + "step": 61800 + }, + { + "epoch": 0.35, + "learning_rate": 4.4134553346237315e-05, + "loss": 0.2342, + "step": 62000 + }, + { + "epoch": 0.35, + "eval_LOC_f1": 0.7818991423369672, + "eval_LOC_number": 22273, + "eval_LOC_precision": 0.7644107050952136, + "eval_LOC_recall": 0.8002065280833296, + "eval_ORG_f1": 0.6305777177161568, + "eval_ORG_number": 19008, + "eval_ORG_precision": 0.6230804786605721, + "eval_ORG_recall": 0.6382575757575758, + "eval_PER_f1": 0.786322314796407, + "eval_PER_number": 27575, + "eval_PER_precision": 0.7837584666378441, + "eval_PER_recall": 0.7889029918404352, + "eval_loss": 0.22046972811222076, + "eval_overall_accuracy": 0.9321861416269704, + "eval_overall_f1": 0.7418848054864094, + "eval_overall_precision": 0.7330139064116922, + "eval_overall_recall": 0.7509730451957709, + "eval_runtime": 420.2506, + "eval_samples_per_second": 90.153, + "eval_steps_per_second": 11.269, + "step": 62000 + }, + { + "epoch": 0.35, + "learning_rate": 4.4115632550580024e-05, + "loss": 0.2294, + "step": 62200 + }, + { + "epoch": 0.35, + "learning_rate": 4.409671175492272e-05, + "loss": 0.2318, + "step": 62400 + }, + { + "epoch": 0.36, + "learning_rate": 4.407779095926542e-05, + "loss": 0.2392, + "step": 62600 + }, + { + "epoch": 0.36, + "learning_rate": 4.4058870163608124e-05, + "loss": 0.2299, + "step": 62800 + }, + { + "epoch": 0.36, + "learning_rate": 4.403994936795082e-05, + "loss": 0.2247, + "step": 63000 + }, + { + "epoch": 0.36, + "learning_rate": 4.402102857229353e-05, + "loss": 0.2367, + "step": 63200 + }, + { + "epoch": 0.36, + "learning_rate": 4.4002107776636225e-05, + "loss": 0.2274, + "step": 63400 + }, + { + "epoch": 0.36, + "learning_rate": 4.398318698097893e-05, + "loss": 0.2239, + "step": 63600 + }, + { + "epoch": 0.36, + "learning_rate": 4.396426618532163e-05, + "loss": 0.2133, + "step": 63800 + }, + { + "epoch": 0.36, + "learning_rate": 4.3945345389664325e-05, + "loss": 0.2295, + "step": 64000 + }, + { + "epoch": 0.36, + "eval_LOC_f1": 0.7854959649932933, + "eval_LOC_number": 22273, + "eval_LOC_precision": 0.7697379762109982, + "eval_LOC_recall": 0.8019126296412697, + "eval_ORG_f1": 0.6308750373773344, + "eval_ORG_number": 19008, + "eval_ORG_precision": 0.6526801282411834, + "eval_ORG_recall": 0.610479797979798, + "eval_PER_f1": 0.7876801152737752, + "eval_PER_number": 27575, + "eval_PER_precision": 0.7824655573447844, + "eval_PER_recall": 0.7929646418857661, + "eval_loss": 0.2175518125295639, + "eval_overall_accuracy": 0.9331650182449033, + "eval_overall_f1": 0.7450937699587761, + "eval_overall_precision": 0.7447046193129062, + "eval_overall_recall": 0.7454833275241083, + "eval_runtime": 250.0041, + "eval_samples_per_second": 151.546, + "eval_steps_per_second": 18.944, + "step": 64000 + }, + { + "epoch": 0.36, + "learning_rate": 4.3926424594007034e-05, + "loss": 0.2274, + "step": 64200 + }, + { + "epoch": 0.37, + "learning_rate": 4.390750379834973e-05, + "loss": 0.2243, + "step": 64400 + }, + { + "epoch": 0.37, + "learning_rate": 4.388858300269243e-05, + "loss": 0.2314, + "step": 64600 + }, + { + "epoch": 0.37, + "learning_rate": 4.3869662207035134e-05, + "loss": 0.2317, + "step": 64800 + }, + { + "epoch": 0.37, + "learning_rate": 4.385074141137783e-05, + "loss": 0.2285, + "step": 65000 + }, + { + "epoch": 0.37, + "learning_rate": 4.383182061572054e-05, + "loss": 0.2333, + "step": 65200 + }, + { + "epoch": 0.37, + "learning_rate": 4.3812899820063234e-05, + "loss": 0.2297, + "step": 65400 + }, + { + "epoch": 0.37, + "learning_rate": 4.379397902440594e-05, + "loss": 0.2304, + "step": 65600 + }, + { + "epoch": 0.37, + "learning_rate": 4.377505822874864e-05, + "loss": 0.2235, + "step": 65800 + }, + { + "epoch": 0.37, + "learning_rate": 4.3756137433091335e-05, + "loss": 0.2218, + "step": 66000 + }, + { + "epoch": 0.37, + "eval_LOC_f1": 0.7838850042214188, + "eval_LOC_number": 22273, + "eval_LOC_precision": 0.7568979933110368, + "eval_LOC_recall": 0.812867597539622, + "eval_ORG_f1": 0.635417217143159, + "eval_ORG_number": 19008, + "eval_ORG_precision": 0.6382843189298227, + "eval_ORG_recall": 0.6325757575757576, + "eval_PER_f1": 0.7905782660993518, + "eval_PER_number": 27575, + "eval_PER_precision": 0.782834388110645, + "eval_PER_recall": 0.7984768812330009, + "eval_loss": 0.21815037727355957, + "eval_overall_accuracy": 0.9333233184240409, + "eval_overall_f1": 0.7463432088163733, + "eval_overall_precision": 0.7356667230968907, + "eval_overall_recall": 0.7573341466248402, + "eval_runtime": 117.433, + "eval_samples_per_second": 322.626, + "eval_steps_per_second": 40.329, + "step": 66000 + }, + { + "epoch": 0.38, + "learning_rate": 4.3737216637434044e-05, + "loss": 0.2255, + "step": 66200 + }, + { + "epoch": 0.38, + "learning_rate": 4.371829584177674e-05, + "loss": 0.2268, + "step": 66400 + }, + { + "epoch": 0.38, + "learning_rate": 4.369937504611944e-05, + "loss": 0.2279, + "step": 66600 + }, + { + "epoch": 0.38, + "learning_rate": 4.3680454250462144e-05, + "loss": 0.2248, + "step": 66800 + }, + { + "epoch": 0.38, + "learning_rate": 4.366153345480484e-05, + "loss": 0.2366, + "step": 67000 + }, + { + "epoch": 0.38, + "learning_rate": 4.364261265914755e-05, + "loss": 0.2275, + "step": 67200 + }, + { + "epoch": 0.38, + "learning_rate": 4.3623691863490244e-05, + "loss": 0.2277, + "step": 67400 + }, + { + "epoch": 0.38, + "learning_rate": 4.360477106783295e-05, + "loss": 0.2198, + "step": 67600 + }, + { + "epoch": 0.38, + "learning_rate": 4.358585027217565e-05, + "loss": 0.2212, + "step": 67800 + }, + { + "epoch": 0.39, + "learning_rate": 4.3566929476518345e-05, + "loss": 0.2364, + "step": 68000 + }, + { + "epoch": 0.39, + "eval_LOC_f1": 0.7828015329630108, + "eval_LOC_number": 22273, + "eval_LOC_precision": 0.7520373970959334, + "eval_LOC_recall": 0.8161900058366632, + "eval_ORG_f1": 0.6379932061667102, + "eval_ORG_number": 19008, + "eval_ORG_precision": 0.6337867303499117, + "eval_ORG_recall": 0.6422558922558923, + "eval_PER_f1": 0.7888646288209606, + "eval_PER_number": 27575, + "eval_PER_precision": 0.7916012415555961, + "eval_PER_recall": 0.7861468721668178, + "eval_loss": 0.21784912049770355, + "eval_overall_accuracy": 0.9334137756692624, + "eval_overall_f1": 0.7455110398350469, + "eval_overall_precision": 0.7351736797514826, + "eval_overall_recall": 0.7561432554897177, + "eval_runtime": 289.5236, + "eval_samples_per_second": 130.86, + "eval_steps_per_second": 16.358, + "step": 68000 + }, + { + "epoch": 0.39, + "learning_rate": 4.3548008680861054e-05, + "loss": 0.2394, + "step": 68200 + }, + { + "epoch": 0.39, + "learning_rate": 4.352908788520375e-05, + "loss": 0.2267, + "step": 68400 + }, + { + "epoch": 0.39, + "learning_rate": 4.351016708954645e-05, + "loss": 0.2304, + "step": 68600 + }, + { + "epoch": 0.39, + "learning_rate": 4.3491246293889154e-05, + "loss": 0.2321, + "step": 68800 + }, + { + "epoch": 0.39, + "learning_rate": 4.347232549823185e-05, + "loss": 0.2282, + "step": 69000 + }, + { + "epoch": 0.39, + "learning_rate": 4.345340470257456e-05, + "loss": 0.2259, + "step": 69200 + }, + { + "epoch": 0.39, + "learning_rate": 4.3434483906917254e-05, + "loss": 0.2263, + "step": 69400 + }, + { + "epoch": 0.4, + "learning_rate": 4.3415563111259956e-05, + "loss": 0.2278, + "step": 69600 + }, + { + "epoch": 0.4, + "learning_rate": 4.339664231560266e-05, + "loss": 0.2319, + "step": 69800 + }, + { + "epoch": 0.4, + "learning_rate": 4.337772151994536e-05, + "loss": 0.2191, + "step": 70000 + }, + { + "epoch": 0.4, + "eval_LOC_f1": 0.7873558157964251, + "eval_LOC_number": 22273, + "eval_LOC_precision": 0.7723601813863097, + "eval_LOC_recall": 0.8029452700579176, + "eval_ORG_f1": 0.636794714689115, + "eval_ORG_number": 19008, + "eval_ORG_precision": 0.6450080949811117, + "eval_ORG_recall": 0.6287878787878788, + "eval_PER_f1": 0.7877720467983395, + "eval_PER_number": 27575, + "eval_PER_precision": 0.7808536411571897, + "eval_PER_recall": 0.7948141432456935, + "eval_loss": 0.21838243305683136, + "eval_overall_accuracy": 0.9339145211338815, + "eval_overall_f1": 0.7467480466636365, + "eval_overall_precision": 0.7419465829426691, + "eval_overall_recall": 0.7516120599512025, + "eval_runtime": 145.5965, + "eval_samples_per_second": 260.219, + "eval_steps_per_second": 32.528, + "step": 70000 + }, + { + "epoch": 0.4, + "learning_rate": 4.3358800724288064e-05, + "loss": 0.2241, + "step": 70200 + }, + { + "epoch": 0.4, + "learning_rate": 4.333987992863076e-05, + "loss": 0.2204, + "step": 70400 + }, + { + "epoch": 0.4, + "learning_rate": 4.332095913297346e-05, + "loss": 0.2322, + "step": 70600 + }, + { + "epoch": 0.4, + "learning_rate": 4.3302038337316164e-05, + "loss": 0.2249, + "step": 70800 + }, + { + "epoch": 0.4, + "learning_rate": 4.3283117541658866e-05, + "loss": 0.235, + "step": 71000 + }, + { + "epoch": 0.4, + "learning_rate": 4.326419674600157e-05, + "loss": 0.2249, + "step": 71200 + }, + { + "epoch": 0.41, + "learning_rate": 4.3245275950344264e-05, + "loss": 0.2223, + "step": 71400 + }, + { + "epoch": 0.41, + "learning_rate": 4.3226355154686966e-05, + "loss": 0.2247, + "step": 71600 + }, + { + "epoch": 0.41, + "learning_rate": 4.320743435902967e-05, + "loss": 0.2244, + "step": 71800 + }, + { + "epoch": 0.41, + "learning_rate": 4.318851356337237e-05, + "loss": 0.2262, + "step": 72000 + }, + { + "epoch": 0.41, + "eval_LOC_f1": 0.7851083468858512, + "eval_LOC_number": 22273, + "eval_LOC_precision": 0.7689526023505101, + "eval_LOC_recall": 0.8019575270506891, + "eval_ORG_f1": 0.6311321794730459, + "eval_ORG_number": 19008, + "eval_ORG_precision": 0.6627691595276685, + "eval_ORG_recall": 0.6023779461279462, + "eval_PER_f1": 0.7904896935325554, + "eval_PER_number": 27575, + "eval_PER_precision": 0.7924487043988484, + "eval_PER_recall": 0.7885403445149592, + "eval_loss": 0.21746216714382172, + "eval_overall_accuracy": 0.9337642974944959, + "eval_overall_f1": 0.7464327485380118, + "eval_overall_precision": 0.7514423643000118, + "eval_overall_recall": 0.7414894853026607, + "eval_runtime": 171.4497, + "eval_samples_per_second": 220.98, + "eval_steps_per_second": 27.623, + "step": 72000 + }, + { + "epoch": 0.41, + "learning_rate": 4.316959276771507e-05, + "loss": 0.2249, + "step": 72200 + }, + { + "epoch": 0.41, + "learning_rate": 4.315067197205777e-05, + "loss": 0.2335, + "step": 72400 + }, + { + "epoch": 0.41, + "learning_rate": 4.313175117640047e-05, + "loss": 0.2196, + "step": 72600 + }, + { + "epoch": 0.41, + "learning_rate": 4.3112830380743174e-05, + "loss": 0.2289, + "step": 72800 + }, + { + "epoch": 0.41, + "learning_rate": 4.3093909585085876e-05, + "loss": 0.2229, + "step": 73000 + }, + { + "epoch": 0.42, + "learning_rate": 4.307498878942858e-05, + "loss": 0.2277, + "step": 73200 + }, + { + "epoch": 0.42, + "learning_rate": 4.3056067993771274e-05, + "loss": 0.2216, + "step": 73400 + }, + { + "epoch": 0.42, + "learning_rate": 4.3037147198113976e-05, + "loss": 0.2297, + "step": 73600 + }, + { + "epoch": 0.42, + "learning_rate": 4.301822640245668e-05, + "loss": 0.2464, + "step": 73800 + }, + { + "epoch": 0.42, + "learning_rate": 4.299930560679938e-05, + "loss": 0.236, + "step": 74000 + }, + { + "epoch": 0.42, + "eval_LOC_f1": 0.7812648920961264, + "eval_LOC_number": 22273, + "eval_LOC_precision": 0.7898504175461136, + "eval_LOC_recall": 0.7728640057468684, + "eval_ORG_f1": 0.6338744323898089, + "eval_ORG_number": 19008, + "eval_ORG_precision": 0.655973887106759, + "eval_ORG_recall": 0.6132154882154882, + "eval_PER_f1": 0.7901759953837276, + "eval_PER_number": 27575, + "eval_PER_precision": 0.7858398192317348, + "eval_PER_recall": 0.7945602901178603, + "eval_loss": 0.21629786491394043, + "eval_overall_accuracy": 0.9337998342694043, + "eval_overall_f1": 0.7451210564930302, + "eval_overall_precision": 0.7529209418183975, + "eval_overall_recall": 0.7374811200185896, + "eval_runtime": 79.9736, + "eval_samples_per_second": 473.744, + "eval_steps_per_second": 59.22, + "step": 74000 + }, + { + "epoch": 0.42, + "learning_rate": 4.298038481114208e-05, + "loss": 0.2275, + "step": 74200 + }, + { + "epoch": 0.42, + "learning_rate": 4.296146401548478e-05, + "loss": 0.2396, + "step": 74400 + }, + { + "epoch": 0.42, + "learning_rate": 4.294254321982748e-05, + "loss": 0.2288, + "step": 74600 + }, + { + "epoch": 0.42, + "learning_rate": 4.2923622424170183e-05, + "loss": 0.2316, + "step": 74800 + }, + { + "epoch": 0.43, + "learning_rate": 4.2904701628512886e-05, + "loss": 0.2247, + "step": 75000 + }, + { + "epoch": 0.43, + "learning_rate": 4.288578083285559e-05, + "loss": 0.2201, + "step": 75200 + }, + { + "epoch": 0.43, + "learning_rate": 4.2866860037198284e-05, + "loss": 0.2235, + "step": 75400 + }, + { + "epoch": 0.43, + "learning_rate": 4.2847939241540986e-05, + "loss": 0.2458, + "step": 75600 + }, + { + "epoch": 0.43, + "learning_rate": 4.282901844588369e-05, + "loss": 0.2268, + "step": 75800 + }, + { + "epoch": 0.43, + "learning_rate": 4.281009765022639e-05, + "loss": 0.2319, + "step": 76000 + }, + { + "epoch": 0.43, + "eval_LOC_f1": 0.7863228799571754, + "eval_LOC_number": 22273, + "eval_LOC_precision": 0.7813040202118701, + "eval_LOC_recall": 0.7914066358371122, + "eval_ORG_f1": 0.6358107186690304, + "eval_ORG_number": 19008, + "eval_ORG_precision": 0.6601347907345528, + "eval_ORG_recall": 0.6132154882154882, + "eval_PER_f1": 0.7876752492137108, + "eval_PER_number": 27575, + "eval_PER_precision": 0.7722109957494251, + "eval_PER_recall": 0.8037715321849501, + "eval_loss": 0.2159438282251358, + "eval_overall_accuracy": 0.9339726722200954, + "eval_overall_f1": 0.746820926721635, + "eval_overall_precision": 0.7464741729541498, + "eval_overall_recall": 0.747168002788428, + "eval_runtime": 75.6006, + "eval_samples_per_second": 501.147, + "eval_steps_per_second": 62.645, + "step": 76000 + }, + { + "epoch": 0.43, + "learning_rate": 4.279117685456909e-05, + "loss": 0.2375, + "step": 76200 + }, + { + "epoch": 0.43, + "learning_rate": 4.277225605891179e-05, + "loss": 0.2183, + "step": 76400 + }, + { + "epoch": 0.43, + "learning_rate": 4.275333526325449e-05, + "loss": 0.2221, + "step": 76600 + }, + { + "epoch": 0.44, + "learning_rate": 4.273441446759719e-05, + "loss": 0.2245, + "step": 76800 + }, + { + "epoch": 0.44, + "learning_rate": 4.2715493671939896e-05, + "loss": 0.2294, + "step": 77000 + }, + { + "epoch": 0.44, + "learning_rate": 4.26965728762826e-05, + "loss": 0.2197, + "step": 77200 + }, + { + "epoch": 0.44, + "learning_rate": 4.2677652080625294e-05, + "loss": 0.2256, + "step": 77400 + }, + { + "epoch": 0.44, + "learning_rate": 4.2658731284967996e-05, + "loss": 0.2249, + "step": 77600 + }, + { + "epoch": 0.44, + "learning_rate": 4.26398104893107e-05, + "loss": 0.2314, + "step": 77800 + }, + { + "epoch": 0.44, + "learning_rate": 4.26208896936534e-05, + "loss": 0.239, + "step": 78000 + }, + { + "epoch": 0.44, + "eval_LOC_f1": 0.7885555702391966, + "eval_LOC_number": 22273, + "eval_LOC_precision": 0.7739634225431277, + "eval_LOC_recall": 0.8037085260180488, + "eval_ORG_f1": 0.6391056739084581, + "eval_ORG_number": 19008, + "eval_ORG_precision": 0.6405919661733616, + "eval_ORG_recall": 0.6376262626262627, + "eval_PER_f1": 0.7898908920964175, + "eval_PER_number": 27575, + "eval_PER_precision": 0.783092166227101, + "eval_PER_recall": 0.7968087035358115, + "eval_loss": 0.21655330061912537, + "eval_overall_accuracy": 0.9337126076400836, + "eval_overall_f1": 0.7482999071695343, + "eval_overall_precision": 0.7416235183362574, + "eval_overall_recall": 0.7550975949808295, + "eval_runtime": 75.1688, + "eval_samples_per_second": 504.026, + "eval_steps_per_second": 63.005, + "step": 78000 + }, + { + "epoch": 0.44, + "learning_rate": 4.26019688979961e-05, + "loss": 0.2248, + "step": 78200 + }, + { + "epoch": 0.45, + "learning_rate": 4.25830481023388e-05, + "loss": 0.2236, + "step": 78400 + }, + { + "epoch": 0.45, + "learning_rate": 4.25641273066815e-05, + "loss": 0.2304, + "step": 78600 + }, + { + "epoch": 0.45, + "learning_rate": 4.25452065110242e-05, + "loss": 0.2271, + "step": 78800 + }, + { + "epoch": 0.45, + "learning_rate": 4.2526285715366906e-05, + "loss": 0.2235, + "step": 79000 + }, + { + "epoch": 0.45, + "learning_rate": 4.250736491970961e-05, + "loss": 0.2212, + "step": 79200 + }, + { + "epoch": 0.45, + "learning_rate": 4.2488444124052303e-05, + "loss": 0.227, + "step": 79400 + }, + { + "epoch": 0.45, + "learning_rate": 4.2469523328395006e-05, + "loss": 0.2171, + "step": 79600 + }, + { + "epoch": 0.45, + "learning_rate": 4.245060253273771e-05, + "loss": 0.2258, + "step": 79800 + }, + { + "epoch": 0.45, + "learning_rate": 4.243168173708041e-05, + "loss": 0.2348, + "step": 80000 + }, + { + "epoch": 0.45, + "eval_LOC_f1": 0.7865168539325843, + "eval_LOC_number": 22273, + "eval_LOC_precision": 0.7567646082337317, + "eval_LOC_recall": 0.8187042607641539, + "eval_ORG_f1": 0.6323823757038776, + "eval_ORG_number": 19008, + "eval_ORG_precision": 0.641715864160754, + "eval_ORG_recall": 0.6233164983164983, + "eval_PER_f1": 0.7887676175845375, + "eval_PER_number": 27575, + "eval_PER_precision": 0.7772301626416954, + "eval_PER_recall": 0.8006527651858567, + "eval_loss": 0.21634286642074585, + "eval_overall_accuracy": 0.9338224485807096, + "eval_overall_f1": 0.7461110991911086, + "eval_overall_precision": 0.7350243077573452, + "eval_overall_recall": 0.7575374695015685, + "eval_runtime": 74.1084, + "eval_samples_per_second": 511.237, + "eval_steps_per_second": 63.906, + "step": 80000 + }, + { + "epoch": 0.46, + "learning_rate": 4.241276094142311e-05, + "loss": 0.2207, + "step": 80200 + }, + { + "epoch": 0.46, + "learning_rate": 4.2393840145765815e-05, + "loss": 0.2204, + "step": 80400 + }, + { + "epoch": 0.46, + "learning_rate": 4.237491935010851e-05, + "loss": 0.2328, + "step": 80600 + }, + { + "epoch": 0.46, + "learning_rate": 4.235599855445121e-05, + "loss": 0.2308, + "step": 80800 + }, + { + "epoch": 0.46, + "learning_rate": 4.2337077758793915e-05, + "loss": 0.2277, + "step": 81000 + }, + { + "epoch": 0.46, + "learning_rate": 4.231815696313662e-05, + "loss": 0.2278, + "step": 81200 + }, + { + "epoch": 0.46, + "learning_rate": 4.229923616747932e-05, + "loss": 0.2277, + "step": 81400 + }, + { + "epoch": 0.46, + "learning_rate": 4.2280315371822016e-05, + "loss": 0.2135, + "step": 81600 + }, + { + "epoch": 0.46, + "learning_rate": 4.226139457616472e-05, + "loss": 0.2275, + "step": 81800 + }, + { + "epoch": 0.47, + "learning_rate": 4.224247378050742e-05, + "loss": 0.2178, + "step": 82000 + }, + { + "epoch": 0.47, + "eval_LOC_f1": 0.7861523072239377, + "eval_LOC_number": 22273, + "eval_LOC_precision": 0.7622822924134441, + "eval_LOC_recall": 0.8115655726664571, + "eval_ORG_f1": 0.640609951845907, + "eval_ORG_number": 19008, + "eval_ORG_precision": 0.6516982364467668, + "eval_ORG_recall": 0.6298926767676768, + "eval_PER_f1": 0.7914478619654914, + "eval_PER_number": 27575, + "eval_PER_precision": 0.7798035971982683, + "eval_PER_recall": 0.8034451495920217, + "eval_loss": 0.21521690487861633, + "eval_overall_accuracy": 0.9342181990285539, + "eval_overall_f1": 0.7492393363568517, + "eval_overall_precision": 0.7405242850658194, + "eval_overall_recall": 0.7581619611943766, + "eval_runtime": 76.1237, + "eval_samples_per_second": 497.703, + "eval_steps_per_second": 62.215, + "step": 82000 + }, + { + "epoch": 0.47, + "learning_rate": 4.222355298485012e-05, + "loss": 0.227, + "step": 82200 + }, + { + "epoch": 0.47, + "learning_rate": 4.2204632189192825e-05, + "loss": 0.221, + "step": 82400 + }, + { + "epoch": 0.47, + "learning_rate": 4.218571139353552e-05, + "loss": 0.2192, + "step": 82600 + }, + { + "epoch": 0.47, + "learning_rate": 4.216679059787822e-05, + "loss": 0.2218, + "step": 82800 + }, + { + "epoch": 0.47, + "learning_rate": 4.2147869802220925e-05, + "loss": 0.2211, + "step": 83000 + }, + { + "epoch": 0.47, + "learning_rate": 4.212894900656363e-05, + "loss": 0.2286, + "step": 83200 + }, + { + "epoch": 0.47, + "learning_rate": 4.211002821090633e-05, + "loss": 0.2264, + "step": 83400 + }, + { + "epoch": 0.47, + "learning_rate": 4.2091107415249026e-05, + "loss": 0.2238, + "step": 83600 + }, + { + "epoch": 0.48, + "learning_rate": 4.207218661959173e-05, + "loss": 0.2227, + "step": 83800 + }, + { + "epoch": 0.48, + "learning_rate": 4.205326582393443e-05, + "loss": 0.2284, + "step": 84000 + }, + { + "epoch": 0.48, + "eval_LOC_f1": 0.7866859117351179, + "eval_LOC_number": 22273, + "eval_LOC_precision": 0.7668954931138874, + "eval_LOC_recall": 0.8075248058187042, + "eval_ORG_f1": 0.6384356248026108, + "eval_ORG_number": 19008, + "eval_ORG_precision": 0.6387718559089951, + "eval_ORG_recall": 0.6380997474747475, + "eval_PER_f1": 0.7947402063738471, + "eval_PER_number": 27575, + "eval_PER_precision": 0.8005150846210449, + "eval_PER_recall": 0.7890480507706256, + "eval_loss": 0.21535621583461761, + "eval_overall_accuracy": 0.9344136512905503, + "eval_overall_f1": 0.7491930067809093, + "eval_overall_precision": 0.7450769164476235, + "eval_overall_recall": 0.753354827466016, + "eval_runtime": 73.9443, + "eval_samples_per_second": 512.372, + "eval_steps_per_second": 64.048, + "step": 84000 + }, + { + "epoch": 0.48, + "learning_rate": 4.203434502827713e-05, + "loss": 0.2177, + "step": 84200 + }, + { + "epoch": 0.48, + "learning_rate": 4.2015424232619835e-05, + "loss": 0.2244, + "step": 84400 + }, + { + "epoch": 0.48, + "learning_rate": 4.199650343696253e-05, + "loss": 0.2262, + "step": 84600 + }, + { + "epoch": 0.48, + "learning_rate": 4.197758264130523e-05, + "loss": 0.2209, + "step": 84800 + }, + { + "epoch": 0.48, + "learning_rate": 4.1958661845647935e-05, + "loss": 0.2209, + "step": 85000 + }, + { + "epoch": 0.48, + "learning_rate": 4.193974104999064e-05, + "loss": 0.2278, + "step": 85200 + }, + { + "epoch": 0.48, + "learning_rate": 4.192082025433334e-05, + "loss": 0.2286, + "step": 85400 + }, + { + "epoch": 0.49, + "learning_rate": 4.1901899458676035e-05, + "loss": 0.2219, + "step": 85600 + }, + { + "epoch": 0.49, + "learning_rate": 4.188297866301874e-05, + "loss": 0.2276, + "step": 85800 + }, + { + "epoch": 0.49, + "learning_rate": 4.186405786736144e-05, + "loss": 0.2294, + "step": 86000 + }, + { + "epoch": 0.49, + "eval_LOC_f1": 0.7861328125, + "eval_LOC_number": 22273, + "eval_LOC_precision": 0.7773339770881798, + "eval_LOC_recall": 0.7951331208189287, + "eval_ORG_f1": 0.6431424067262218, + "eval_ORG_number": 19008, + "eval_ORG_precision": 0.6423997480579466, + "eval_ORG_recall": 0.6438867845117845, + "eval_PER_f1": 0.7913160790996556, + "eval_PER_number": 27575, + "eval_PER_precision": 0.7750269480858166, + "eval_PER_recall": 0.8083046237533998, + "eval_loss": 0.21463459730148315, + "eval_overall_accuracy": 0.9340179008427062, + "eval_overall_f1": 0.7492004302617425, + "eval_overall_precision": 0.7399779018046859, + "eval_overall_recall": 0.7586557453235738, + "eval_runtime": 72.9707, + "eval_samples_per_second": 519.208, + "eval_steps_per_second": 64.903, + "step": 86000 + }, + { + "epoch": 0.49, + "learning_rate": 4.184513707170414e-05, + "loss": 0.2218, + "step": 86200 + }, + { + "epoch": 0.49, + "learning_rate": 4.1826216276046845e-05, + "loss": 0.2294, + "step": 86400 + }, + { + "epoch": 0.49, + "learning_rate": 4.180729548038954e-05, + "loss": 0.2267, + "step": 86600 + }, + { + "epoch": 0.49, + "learning_rate": 4.178837468473224e-05, + "loss": 0.226, + "step": 86800 + }, + { + "epoch": 0.49, + "learning_rate": 4.1769453889074945e-05, + "loss": 0.2248, + "step": 87000 + }, + { + "epoch": 0.49, + "learning_rate": 4.175053309341765e-05, + "loss": 0.2263, + "step": 87200 + }, + { + "epoch": 0.5, + "learning_rate": 4.173161229776035e-05, + "loss": 0.2172, + "step": 87400 + }, + { + "epoch": 0.5, + "learning_rate": 4.1712691502103045e-05, + "loss": 0.2298, + "step": 87600 + }, + { + "epoch": 0.5, + "learning_rate": 4.169377070644575e-05, + "loss": 0.2162, + "step": 87800 + }, + { + "epoch": 0.5, + "learning_rate": 4.167484991078845e-05, + "loss": 0.2237, + "step": 88000 + }, + { + "epoch": 0.5, + "eval_LOC_f1": 0.7871778210959498, + "eval_LOC_number": 22273, + "eval_LOC_precision": 0.7604301795204419, + "eval_LOC_recall": 0.8158757239707269, + "eval_ORG_f1": 0.6407888643730763, + "eval_ORG_number": 19008, + "eval_ORG_precision": 0.6644071626278032, + "eval_ORG_recall": 0.6187920875420876, + "eval_PER_f1": 0.7920023118881624, + "eval_PER_number": 27575, + "eval_PER_precision": 0.7889244719513512, + "eval_PER_recall": 0.7951042611060744, + "eval_loss": 0.21553027629852295, + "eval_overall_accuracy": 0.9343716532838403, + "eval_overall_f1": 0.750236894833161, + "eval_overall_precision": 0.7473447565246214, + "eval_overall_recall": 0.7531515045892878, + "eval_runtime": 73.3362, + "eval_samples_per_second": 516.621, + "eval_steps_per_second": 64.579, + "step": 88000 + }, + { + "epoch": 0.5, + "learning_rate": 4.165592911513115e-05, + "loss": 0.2196, + "step": 88200 + }, + { + "epoch": 0.5, + "learning_rate": 4.1637008319473855e-05, + "loss": 0.2286, + "step": 88400 + }, + { + "epoch": 0.5, + "learning_rate": 4.161808752381655e-05, + "loss": 0.2237, + "step": 88600 + }, + { + "epoch": 0.5, + "learning_rate": 4.159916672815925e-05, + "loss": 0.223, + "step": 88800 + }, + { + "epoch": 0.51, + "learning_rate": 4.1580245932501955e-05, + "loss": 0.2241, + "step": 89000 + }, + { + "epoch": 0.51, + "learning_rate": 4.156132513684466e-05, + "loss": 0.2347, + "step": 89200 + }, + { + "epoch": 0.51, + "learning_rate": 4.154240434118736e-05, + "loss": 0.2213, + "step": 89400 + }, + { + "epoch": 0.51, + "learning_rate": 4.1523483545530055e-05, + "loss": 0.2194, + "step": 89600 + }, + { + "epoch": 0.51, + "learning_rate": 4.150456274987276e-05, + "loss": 0.2301, + "step": 89800 + }, + { + "epoch": 0.51, + "learning_rate": 4.148564195421546e-05, + "loss": 0.2244, + "step": 90000 + }, + { + "epoch": 0.51, + "eval_LOC_f1": 0.7874339655321728, + "eval_LOC_number": 22273, + "eval_LOC_precision": 0.7604014217018608, + "eval_LOC_recall": 0.81645939029318, + "eval_ORG_f1": 0.6409416526856808, + "eval_ORG_number": 19008, + "eval_ORG_precision": 0.6489780510165561, + "eval_ORG_recall": 0.6331018518518519, + "eval_PER_f1": 0.7937186697091112, + "eval_PER_number": 27575, + "eval_PER_precision": 0.7865536642689267, + "eval_PER_recall": 0.8010154125113327, + "eval_loss": 0.21349573135375977, + "eval_overall_accuracy": 0.934563874929936, + "eval_overall_f1": 0.7504806450687251, + "eval_overall_precision": 0.7415225404026085, + "eval_overall_recall": 0.7596578366445916, + "eval_runtime": 74.0333, + "eval_samples_per_second": 511.756, + "eval_steps_per_second": 63.971, + "step": 90000 + }, + { + "epoch": 0.51, + "learning_rate": 4.146672115855816e-05, + "loss": 0.2273, + "step": 90200 + }, + { + "epoch": 0.51, + "learning_rate": 4.1447800362900864e-05, + "loss": 0.2238, + "step": 90400 + }, + { + "epoch": 0.51, + "learning_rate": 4.142887956724356e-05, + "loss": 0.2204, + "step": 90600 + }, + { + "epoch": 0.52, + "learning_rate": 4.140995877158627e-05, + "loss": 0.2205, + "step": 90800 + }, + { + "epoch": 0.52, + "learning_rate": 4.1391037975928965e-05, + "loss": 0.2272, + "step": 91000 + }, + { + "epoch": 0.52, + "learning_rate": 4.137211718027167e-05, + "loss": 0.2141, + "step": 91200 + }, + { + "epoch": 0.52, + "learning_rate": 4.135319638461437e-05, + "loss": 0.2203, + "step": 91400 + }, + { + "epoch": 0.52, + "learning_rate": 4.1334275588957065e-05, + "loss": 0.2135, + "step": 91600 + }, + { + "epoch": 0.52, + "learning_rate": 4.1315354793299774e-05, + "loss": 0.2223, + "step": 91800 + }, + { + "epoch": 0.52, + "learning_rate": 4.129643399764247e-05, + "loss": 0.2208, + "step": 92000 + }, + { + "epoch": 0.52, + "eval_LOC_f1": 0.7893500053896734, + "eval_LOC_number": 22273, + "eval_LOC_precision": 0.759248506967485, + "eval_LOC_recall": 0.8219368742423562, + "eval_ORG_f1": 0.6397469688982604, + "eval_ORG_number": 19008, + "eval_ORG_precision": 0.6410310585252482, + "eval_ORG_recall": 0.6384680134680135, + "eval_PER_f1": 0.7907061238577943, + "eval_PER_number": 27575, + "eval_PER_precision": 0.7921884100174723, + "eval_PER_recall": 0.7892293744333636, + "eval_loss": 0.21358641982078552, + "eval_overall_accuracy": 0.9343021950419738, + "eval_overall_f1": 0.7491605200470682, + "eval_overall_precision": 0.7403426172783482, + "eval_overall_recall": 0.7581910073196235, + "eval_runtime": 72.5683, + "eval_samples_per_second": 522.087, + "eval_steps_per_second": 65.263, + "step": 92000 + }, + { + "epoch": 0.52, + "learning_rate": 4.127751320198517e-05, + "loss": 0.2222, + "step": 92200 + }, + { + "epoch": 0.52, + "learning_rate": 4.1258592406327874e-05, + "loss": 0.225, + "step": 92400 + }, + { + "epoch": 0.53, + "learning_rate": 4.123967161067057e-05, + "loss": 0.2211, + "step": 92600 + }, + { + "epoch": 0.53, + "learning_rate": 4.122075081501328e-05, + "loss": 0.2241, + "step": 92800 + }, + { + "epoch": 0.53, + "learning_rate": 4.1201830019355975e-05, + "loss": 0.223, + "step": 93000 + }, + { + "epoch": 0.53, + "learning_rate": 4.118290922369868e-05, + "loss": 0.2265, + "step": 93200 + }, + { + "epoch": 0.53, + "learning_rate": 4.116398842804138e-05, + "loss": 0.2181, + "step": 93400 + }, + { + "epoch": 0.53, + "learning_rate": 4.1145067632384075e-05, + "loss": 0.2231, + "step": 93600 + }, + { + "epoch": 0.53, + "learning_rate": 4.1126146836726784e-05, + "loss": 0.2171, + "step": 93800 + }, + { + "epoch": 0.53, + "learning_rate": 4.110722604106948e-05, + "loss": 0.2166, + "step": 94000 + }, + { + "epoch": 0.53, + "eval_LOC_f1": 0.7889070290782332, + "eval_LOC_number": 22273, + "eval_LOC_precision": 0.7755606645555893, + "eval_LOC_recall": 0.8027207830108203, + "eval_ORG_f1": 0.6452999489562905, + "eval_ORG_number": 19008, + "eval_ORG_precision": 0.659346692286577, + "eval_ORG_recall": 0.6318392255892256, + "eval_PER_f1": 0.7951609969074039, + "eval_PER_number": 27575, + "eval_PER_precision": 0.7977733162985946, + "eval_PER_recall": 0.7925657298277425, + "eval_loss": 0.21250613033771515, + "eval_overall_accuracy": 0.9352972247394105, + "eval_overall_f1": 0.752536013205448, + "eval_overall_precision": 0.7535936384952595, + "eval_overall_recall": 0.7514813523875915, + "eval_runtime": 73.1644, + "eval_samples_per_second": 517.834, + "eval_steps_per_second": 64.731, + "step": 94000 + }, + { + "epoch": 0.53, + "learning_rate": 4.108830524541218e-05, + "loss": 0.2161, + "step": 94200 + }, + { + "epoch": 0.54, + "learning_rate": 4.1069384449754884e-05, + "loss": 0.2222, + "step": 94400 + }, + { + "epoch": 0.54, + "learning_rate": 4.105046365409758e-05, + "loss": 0.235, + "step": 94600 + }, + { + "epoch": 0.54, + "learning_rate": 4.103154285844029e-05, + "loss": 0.2126, + "step": 94800 + }, + { + "epoch": 0.54, + "learning_rate": 4.1012622062782984e-05, + "loss": 0.2203, + "step": 95000 + }, + { + "epoch": 0.54, + "learning_rate": 4.099370126712569e-05, + "loss": 0.23, + "step": 95200 + }, + { + "epoch": 0.54, + "learning_rate": 4.097478047146839e-05, + "loss": 0.2219, + "step": 95400 + }, + { + "epoch": 0.54, + "learning_rate": 4.0955859675811085e-05, + "loss": 0.2239, + "step": 95600 + }, + { + "epoch": 0.54, + "learning_rate": 4.0936938880153794e-05, + "loss": 0.216, + "step": 95800 + }, + { + "epoch": 0.54, + "learning_rate": 4.091801808449649e-05, + "loss": 0.2193, + "step": 96000 + }, + { + "epoch": 0.54, + "eval_LOC_f1": 0.7911759612054035, + "eval_LOC_number": 22273, + "eval_LOC_precision": 0.7639533425310423, + "eval_LOC_recall": 0.820410362322094, + "eval_ORG_f1": 0.6421596168766889, + "eval_ORG_number": 19008, + "eval_ORG_precision": 0.6533289781697424, + "eval_ORG_recall": 0.6313657407407407, + "eval_PER_f1": 0.7945289855072464, + "eval_PER_number": 27575, + "eval_PER_precision": 0.7938099547511313, + "eval_PER_recall": 0.7952493200362647, + "eval_loss": 0.21367917954921722, + "eval_overall_accuracy": 0.9346656393308103, + "eval_overall_f1": 0.7523726480698139, + "eval_overall_precision": 0.7466851658489837, + "eval_overall_recall": 0.7581474381317532, + "eval_runtime": 72.1232, + "eval_samples_per_second": 525.309, + "eval_steps_per_second": 65.665, + "step": 96000 + }, + { + "epoch": 0.55, + "learning_rate": 4.089909728883919e-05, + "loss": 0.2174, + "step": 96200 + }, + { + "epoch": 0.55, + "learning_rate": 4.0880176493181894e-05, + "loss": 0.2287, + "step": 96400 + }, + { + "epoch": 0.55, + "learning_rate": 4.086125569752459e-05, + "loss": 0.2251, + "step": 96600 + }, + { + "epoch": 0.55, + "learning_rate": 4.08423349018673e-05, + "loss": 0.224, + "step": 96800 + }, + { + "epoch": 0.55, + "learning_rate": 4.0823414106209994e-05, + "loss": 0.2196, + "step": 97000 + }, + { + "epoch": 0.55, + "learning_rate": 4.08044933105527e-05, + "loss": 0.2297, + "step": 97200 + }, + { + "epoch": 0.55, + "learning_rate": 4.07855725148954e-05, + "loss": 0.2176, + "step": 97400 + }, + { + "epoch": 0.55, + "learning_rate": 4.0766651719238095e-05, + "loss": 0.2194, + "step": 97600 + }, + { + "epoch": 0.56, + "learning_rate": 4.0747730923580804e-05, + "loss": 0.218, + "step": 97800 + }, + { + "epoch": 0.56, + "learning_rate": 4.07288101279235e-05, + "loss": 0.2231, + "step": 98000 + }, + { + "epoch": 0.56, + "eval_LOC_f1": 0.7923224230151219, + "eval_LOC_number": 22273, + "eval_LOC_precision": 0.7771828309871319, + "eval_LOC_recall": 0.808063574731738, + "eval_ORG_f1": 0.6451822740208701, + "eval_ORG_number": 19008, + "eval_ORG_precision": 0.665399452116062, + "eval_ORG_recall": 0.6261574074074074, + "eval_PER_f1": 0.7933723897494879, + "eval_PER_number": 27575, + "eval_PER_precision": 0.7863631505824517, + "eval_PER_recall": 0.8005077062556664, + "eval_loss": 0.21320602297782898, + "eval_overall_accuracy": 0.9350630050866048, + "eval_overall_f1": 0.7533992404255936, + "eval_overall_precision": 0.7519821748943805, + "eval_overall_recall": 0.7548216567909841, + "eval_runtime": 74.0163, + "eval_samples_per_second": 511.874, + "eval_steps_per_second": 63.986, + "step": 98000 + }, + { + "epoch": 0.56, + "learning_rate": 4.07098893322662e-05, + "loss": 0.2295, + "step": 98200 + }, + { + "epoch": 0.56, + "learning_rate": 4.0690968536608904e-05, + "loss": 0.2227, + "step": 98400 + }, + { + "epoch": 0.56, + "learning_rate": 4.06720477409516e-05, + "loss": 0.222, + "step": 98600 + }, + { + "epoch": 0.56, + "learning_rate": 4.065312694529431e-05, + "loss": 0.2154, + "step": 98800 + }, + { + "epoch": 0.56, + "learning_rate": 4.0634206149637004e-05, + "loss": 0.2238, + "step": 99000 + }, + { + "epoch": 0.56, + "learning_rate": 4.0615285353979707e-05, + "loss": 0.2218, + "step": 99200 + }, + { + "epoch": 0.56, + "learning_rate": 4.059636455832241e-05, + "loss": 0.2214, + "step": 99400 + }, + { + "epoch": 0.57, + "learning_rate": 4.0577443762665104e-05, + "loss": 0.2269, + "step": 99600 + }, + { + "epoch": 0.57, + "learning_rate": 4.0558522967007814e-05, + "loss": 0.2226, + "step": 99800 + }, + { + "epoch": 0.57, + "learning_rate": 4.053960217135051e-05, + "loss": 0.2201, + "step": 100000 + }, + { + "epoch": 0.57, + "eval_LOC_f1": 0.7882217741061168, + "eval_LOC_number": 22273, + "eval_LOC_precision": 0.7622242724146607, + "eval_LOC_recall": 0.8160553136084048, + "eval_ORG_f1": 0.6444801596098426, + "eval_ORG_number": 19008, + "eval_ORG_precision": 0.6808548009367682, + "eval_ORG_recall": 0.6117950336700336, + "eval_PER_f1": 0.7956606276462359, + "eval_PER_number": 27575, + "eval_PER_precision": 0.7973849067599068, + "eval_PER_recall": 0.7939437896645513, + "eval_loss": 0.21130937337875366, + "eval_overall_accuracy": 0.9351857684908339, + "eval_overall_f1": 0.7534064909135952, + "eval_overall_precision": 0.7560176654675207, + "eval_overall_recall": 0.750813291506913, + "eval_runtime": 73.4961, + "eval_samples_per_second": 515.497, + "eval_steps_per_second": 64.439, + "step": 100000 + }, + { + "epoch": 0.57, + "learning_rate": 4.052068137569321e-05, + "loss": 0.2223, + "step": 100200 + }, + { + "epoch": 0.57, + "learning_rate": 4.0501760580035914e-05, + "loss": 0.2208, + "step": 100400 + }, + { + "epoch": 0.57, + "learning_rate": 4.048283978437861e-05, + "loss": 0.2259, + "step": 100600 + }, + { + "epoch": 0.57, + "learning_rate": 4.046391898872132e-05, + "loss": 0.2198, + "step": 100800 + }, + { + "epoch": 0.57, + "learning_rate": 4.0444998193064014e-05, + "loss": 0.2143, + "step": 101000 + }, + { + "epoch": 0.57, + "learning_rate": 4.0426077397406716e-05, + "loss": 0.2159, + "step": 101200 + }, + { + "epoch": 0.58, + "learning_rate": 4.040715660174942e-05, + "loss": 0.2264, + "step": 101400 + }, + { + "epoch": 0.58, + "learning_rate": 4.0388235806092114e-05, + "loss": 0.2129, + "step": 101600 + }, + { + "epoch": 0.58, + "learning_rate": 4.0369315010434823e-05, + "loss": 0.2245, + "step": 101800 + }, + { + "epoch": 0.58, + "learning_rate": 4.035039421477752e-05, + "loss": 0.219, + "step": 102000 + }, + { + "epoch": 0.58, + "eval_LOC_f1": 0.7869046315558962, + "eval_LOC_number": 22273, + "eval_LOC_precision": 0.7680831053351573, + "eval_LOC_recall": 0.8066717550397342, + "eval_ORG_f1": 0.637086328194448, + "eval_ORG_number": 19008, + "eval_ORG_precision": 0.6158605450527866, + "eval_ORG_recall": 0.6598274410774411, + "eval_PER_f1": 0.7916057483150876, + "eval_PER_number": 27575, + "eval_PER_precision": 0.7786117638806075, + "eval_PER_recall": 0.8050407978241161, + "eval_loss": 0.21635876595973969, + "eval_overall_accuracy": 0.933677070865175, + "eval_overall_f1": 0.7469742921119016, + "eval_overall_precision": 0.7293407870703492, + "eval_overall_recall": 0.7654815847565934, + "eval_runtime": 72.0546, + "eval_samples_per_second": 525.81, + "eval_steps_per_second": 65.728, + "step": 102000 + }, + { + "epoch": 0.58, + "learning_rate": 4.033147341912023e-05, + "loss": 0.2258, + "step": 102200 + }, + { + "epoch": 0.58, + "learning_rate": 4.0312552623462924e-05, + "loss": 0.2253, + "step": 102400 + }, + { + "epoch": 0.58, + "learning_rate": 4.029363182780562e-05, + "loss": 0.2177, + "step": 102600 + }, + { + "epoch": 0.58, + "learning_rate": 4.027471103214833e-05, + "loss": 0.2197, + "step": 102800 + }, + { + "epoch": 0.58, + "learning_rate": 4.0255790236491024e-05, + "loss": 0.2267, + "step": 103000 + }, + { + "epoch": 0.59, + "learning_rate": 4.023686944083373e-05, + "loss": 0.2101, + "step": 103200 + }, + { + "epoch": 0.59, + "learning_rate": 4.021794864517643e-05, + "loss": 0.2251, + "step": 103400 + }, + { + "epoch": 0.59, + "learning_rate": 4.0199027849519124e-05, + "loss": 0.2219, + "step": 103600 + }, + { + "epoch": 0.59, + "learning_rate": 4.018010705386183e-05, + "loss": 0.2195, + "step": 103800 + }, + { + "epoch": 0.59, + "learning_rate": 4.016118625820453e-05, + "loss": 0.2207, + "step": 104000 + }, + { + "epoch": 0.59, + "eval_LOC_f1": 0.7910983805221989, + "eval_LOC_number": 22273, + "eval_LOC_precision": 0.7767393561786086, + "eval_LOC_recall": 0.805998293898442, + "eval_ORG_f1": 0.6457971329278889, + "eval_ORG_number": 19008, + "eval_ORG_precision": 0.6672464093357271, + "eval_ORG_recall": 0.6256839225589226, + "eval_PER_f1": 0.793726943974995, + "eval_PER_number": 27575, + "eval_PER_precision": 0.7954760690609747, + "eval_PER_recall": 0.7919854941069809, + "eval_loss": 0.21242733299732208, + "eval_overall_accuracy": 0.9352358430372959, + "eval_overall_f1": 0.7531585619981639, + "eval_overall_precision": 0.75572452112882, + "eval_overall_recall": 0.7506099686301847, + "eval_runtime": 74.374, + "eval_samples_per_second": 509.412, + "eval_steps_per_second": 63.678, + "step": 104000 + }, + { + "epoch": 0.59, + "learning_rate": 4.014226546254724e-05, + "loss": 0.2305, + "step": 104200 + }, + { + "epoch": 0.59, + "learning_rate": 4.0123344666889934e-05, + "loss": 0.2214, + "step": 104400 + }, + { + "epoch": 0.59, + "learning_rate": 4.0104423871232636e-05, + "loss": 0.2329, + "step": 104600 + }, + { + "epoch": 0.59, + "learning_rate": 4.008550307557534e-05, + "loss": 0.224, + "step": 104800 + }, + { + "epoch": 0.6, + "learning_rate": 4.0066582279918034e-05, + "loss": 0.2224, + "step": 105000 + }, + { + "epoch": 0.6, + "learning_rate": 4.004766148426074e-05, + "loss": 0.2202, + "step": 105200 + }, + { + "epoch": 0.6, + "learning_rate": 4.002874068860344e-05, + "loss": 0.2225, + "step": 105400 + }, + { + "epoch": 0.6, + "learning_rate": 4.000981989294614e-05, + "loss": 0.2269, + "step": 105600 + }, + { + "epoch": 0.6, + "learning_rate": 3.999089909728884e-05, + "loss": 0.2208, + "step": 105800 + }, + { + "epoch": 0.6, + "learning_rate": 3.997197830163154e-05, + "loss": 0.224, + "step": 106000 + }, + { + "epoch": 0.6, + "eval_LOC_f1": 0.7874562937062937, + "eval_LOC_number": 22273, + "eval_LOC_precision": 0.7671052071358624, + "eval_LOC_recall": 0.8089166255107081, + "eval_ORG_f1": 0.6415584415584416, + "eval_ORG_number": 19008, + "eval_ORG_precision": 0.646458711676103, + "eval_ORG_recall": 0.6367319023569024, + "eval_PER_f1": 0.7920689593632324, + "eval_PER_number": 27575, + "eval_PER_precision": 0.7797807294960996, + "eval_PER_recall": 0.8047506799637353, + "eval_loss": 0.211883082985878, + "eval_overall_accuracy": 0.9343248093532791, + "eval_overall_f1": 0.7498548626391347, + "eval_overall_precision": 0.7402465082711874, + "eval_overall_recall": 0.7597159288950854, + "eval_runtime": 74.7405, + "eval_samples_per_second": 506.914, + "eval_steps_per_second": 63.366, + "step": 106000 + }, + { + "epoch": 0.6, + "learning_rate": 3.995305750597425e-05, + "loss": 0.2243, + "step": 106200 + }, + { + "epoch": 0.6, + "learning_rate": 3.9934136710316943e-05, + "loss": 0.2183, + "step": 106400 + }, + { + "epoch": 0.61, + "learning_rate": 3.9915215914659646e-05, + "loss": 0.2096, + "step": 106600 + }, + { + "epoch": 0.61, + "learning_rate": 3.989629511900235e-05, + "loss": 0.2203, + "step": 106800 + }, + { + "epoch": 0.61, + "learning_rate": 3.9877374323345044e-05, + "loss": 0.2234, + "step": 107000 + }, + { + "epoch": 0.61, + "learning_rate": 3.985845352768775e-05, + "loss": 0.2159, + "step": 107200 + }, + { + "epoch": 0.61, + "learning_rate": 3.983953273203045e-05, + "loss": 0.2227, + "step": 107400 + }, + { + "epoch": 0.61, + "learning_rate": 3.982061193637315e-05, + "loss": 0.222, + "step": 107600 + }, + { + "epoch": 0.61, + "learning_rate": 3.980169114071585e-05, + "loss": 0.2144, + "step": 107800 + }, + { + "epoch": 0.61, + "learning_rate": 3.978277034505855e-05, + "loss": 0.2239, + "step": 108000 + }, + { + "epoch": 0.61, + "eval_LOC_f1": 0.7874395086912176, + "eval_LOC_number": 22273, + "eval_LOC_precision": 0.7620547715053764, + "eval_LOC_recall": 0.814573699097562, + "eval_ORG_f1": 0.6428840462738767, + "eval_ORG_number": 19008, + "eval_ORG_precision": 0.657857064199978, + "eval_ORG_recall": 0.6285774410774411, + "eval_PER_f1": 0.7935873926541875, + "eval_PER_number": 27575, + "eval_PER_precision": 0.7946116928446771, + "eval_PER_recall": 0.7925657298277425, + "eval_loss": 0.21169796586036682, + "eval_overall_accuracy": 0.9347900180429898, + "eval_overall_f1": 0.7510446034844214, + "eval_overall_precision": 0.7477041771022253, + "eval_overall_recall": 0.7544150110375276, + "eval_runtime": 73.1416, + "eval_samples_per_second": 517.995, + "eval_steps_per_second": 64.751, + "step": 108000 + }, + { + "epoch": 0.61, + "learning_rate": 3.976384954940126e-05, + "loss": 0.2244, + "step": 108200 + }, + { + "epoch": 0.62, + "learning_rate": 3.974492875374395e-05, + "loss": 0.2276, + "step": 108400 + }, + { + "epoch": 0.62, + "learning_rate": 3.9726007958086656e-05, + "loss": 0.2238, + "step": 108600 + }, + { + "epoch": 0.62, + "learning_rate": 3.970708716242936e-05, + "loss": 0.2264, + "step": 108800 + }, + { + "epoch": 0.62, + "learning_rate": 3.9688166366772054e-05, + "loss": 0.2273, + "step": 109000 + }, + { + "epoch": 0.62, + "learning_rate": 3.966924557111476e-05, + "loss": 0.2235, + "step": 109200 + }, + { + "epoch": 0.62, + "learning_rate": 3.965032477545746e-05, + "loss": 0.2127, + "step": 109400 + }, + { + "epoch": 0.62, + "learning_rate": 3.963140397980016e-05, + "loss": 0.2193, + "step": 109600 + }, + { + "epoch": 0.62, + "learning_rate": 3.961248318414286e-05, + "loss": 0.2278, + "step": 109800 + }, + { + "epoch": 0.62, + "learning_rate": 3.959356238848556e-05, + "loss": 0.2164, + "step": 110000 + }, + { + "epoch": 0.62, + "eval_LOC_f1": 0.7912549191080018, + "eval_LOC_number": 22273, + "eval_LOC_precision": 0.771125410150424, + "eval_LOC_recall": 0.8124635208548466, + "eval_ORG_f1": 0.6418721047867526, + "eval_ORG_number": 19008, + "eval_ORG_precision": 0.6497251266573245, + "eval_ORG_recall": 0.6342066498316499, + "eval_PER_f1": 0.7939134023578389, + "eval_PER_number": 27575, + "eval_PER_precision": 0.7857498046458763, + "eval_PER_recall": 0.8022484134179511, + "eval_loss": 0.2099016308784485, + "eval_overall_accuracy": 0.9353537605176739, + "eval_overall_f1": 0.7519617926937158, + "eval_overall_precision": 0.74489490559316, + "eval_overall_recall": 0.7591640525153944, + "eval_runtime": 75.2188, + "eval_samples_per_second": 503.691, + "eval_steps_per_second": 62.963, + "step": 110000 + }, + { + "epoch": 0.63, + "learning_rate": 3.957464159282827e-05, + "loss": 0.2188, + "step": 110200 + }, + { + "epoch": 0.63, + "learning_rate": 3.955572079717096e-05, + "loss": 0.2171, + "step": 110400 + }, + { + "epoch": 0.63, + "learning_rate": 3.9536800001513665e-05, + "loss": 0.2126, + "step": 110600 + }, + { + "epoch": 0.63, + "learning_rate": 3.951787920585637e-05, + "loss": 0.2241, + "step": 110800 + }, + { + "epoch": 0.63, + "learning_rate": 3.949895841019906e-05, + "loss": 0.2121, + "step": 111000 + }, + { + "epoch": 0.63, + "learning_rate": 3.948003761454177e-05, + "loss": 0.2179, + "step": 111200 + }, + { + "epoch": 0.63, + "learning_rate": 3.946111681888447e-05, + "loss": 0.2167, + "step": 111400 + }, + { + "epoch": 0.63, + "learning_rate": 3.944219602322717e-05, + "loss": 0.2231, + "step": 111600 + }, + { + "epoch": 0.63, + "learning_rate": 3.942327522756987e-05, + "loss": 0.224, + "step": 111800 + }, + { + "epoch": 0.64, + "learning_rate": 3.940435443191257e-05, + "loss": 0.206, + "step": 112000 + }, + { + "epoch": 0.64, + "eval_LOC_f1": 0.7951213068557087, + "eval_LOC_number": 22273, + "eval_LOC_precision": 0.7800768931703314, + "eval_LOC_recall": 0.8107574192969066, + "eval_ORG_f1": 0.6505921035045044, + "eval_ORG_number": 19008, + "eval_ORG_precision": 0.6613227541981415, + "eval_ORG_recall": 0.6402041245791246, + "eval_PER_f1": 0.7962397011777741, + "eval_PER_number": 27575, + "eval_PER_precision": 0.7849742758474875, + "eval_PER_recall": 0.807833182230281, + "eval_loss": 0.2094828188419342, + "eval_overall_accuracy": 0.9358480447504914, + "eval_overall_f1": 0.7566145953424026, + "eval_overall_precision": 0.7508151241276741, + "eval_overall_recall": 0.762504356918787, + "eval_runtime": 72.3593, + "eval_samples_per_second": 523.596, + "eval_steps_per_second": 65.451, + "step": 112000 + }, + { + "epoch": 0.64, + "learning_rate": 3.938543363625528e-05, + "loss": 0.2188, + "step": 112200 + }, + { + "epoch": 0.64, + "learning_rate": 3.936651284059797e-05, + "loss": 0.2209, + "step": 112400 + }, + { + "epoch": 0.64, + "learning_rate": 3.9347592044940675e-05, + "loss": 0.2213, + "step": 112600 + }, + { + "epoch": 0.64, + "learning_rate": 3.932867124928338e-05, + "loss": 0.2215, + "step": 112800 + }, + { + "epoch": 0.64, + "learning_rate": 3.930975045362607e-05, + "loss": 0.2157, + "step": 113000 + }, + { + "epoch": 0.64, + "learning_rate": 3.929082965796878e-05, + "loss": 0.2139, + "step": 113200 + }, + { + "epoch": 0.64, + "learning_rate": 3.927190886231148e-05, + "loss": 0.2122, + "step": 113400 + }, + { + "epoch": 0.64, + "learning_rate": 3.925298806665418e-05, + "loss": 0.219, + "step": 113600 + }, + { + "epoch": 0.65, + "learning_rate": 3.923406727099688e-05, + "loss": 0.2175, + "step": 113800 + }, + { + "epoch": 0.65, + "learning_rate": 3.921514647533958e-05, + "loss": 0.2163, + "step": 114000 + }, + { + "epoch": 0.65, + "eval_LOC_f1": 0.7919884801675613, + "eval_LOC_number": 22273, + "eval_LOC_precision": 0.7703408174525699, + "eval_LOC_recall": 0.8148879809634985, + "eval_ORG_f1": 0.6453399284927013, + "eval_ORG_number": 19008, + "eval_ORG_precision": 0.659831784948601, + "eval_ORG_recall": 0.6314709595959596, + "eval_PER_f1": 0.7953349858230647, + "eval_PER_number": 27575, + "eval_PER_precision": 0.7824012350010525, + "eval_PER_recall": 0.8087035358114234, + "eval_loss": 0.2122264951467514, + "eval_overall_accuracy": 0.9356138250976858, + "eval_overall_f1": 0.7541226367622745, + "eval_overall_precision": 0.7466194095709853, + "eval_overall_recall": 0.7617782037876147, + "eval_runtime": 72.8517, + "eval_samples_per_second": 520.057, + "eval_steps_per_second": 65.009, + "step": 114000 + }, + { + "epoch": 0.65, + "learning_rate": 3.919622567968229e-05, + "loss": 0.218, + "step": 114200 + }, + { + "epoch": 0.65, + "learning_rate": 3.917730488402498e-05, + "loss": 0.2211, + "step": 114400 + }, + { + "epoch": 0.65, + "learning_rate": 3.9158384088367685e-05, + "loss": 0.2165, + "step": 114600 + }, + { + "epoch": 0.65, + "learning_rate": 3.913946329271039e-05, + "loss": 0.2242, + "step": 114800 + }, + { + "epoch": 0.65, + "learning_rate": 3.912054249705308e-05, + "loss": 0.2107, + "step": 115000 + }, + { + "epoch": 0.65, + "learning_rate": 3.910162170139579e-05, + "loss": 0.2167, + "step": 115200 + }, + { + "epoch": 0.66, + "learning_rate": 3.908270090573849e-05, + "loss": 0.2289, + "step": 115400 + }, + { + "epoch": 0.66, + "learning_rate": 3.906378011008119e-05, + "loss": 0.2332, + "step": 115600 + }, + { + "epoch": 0.66, + "learning_rate": 3.904485931442389e-05, + "loss": 0.2187, + "step": 115800 + }, + { + "epoch": 0.66, + "learning_rate": 3.9025938518766595e-05, + "loss": 0.233, + "step": 116000 + }, + { + "epoch": 0.66, + "eval_LOC_f1": 0.7929061037971927, + "eval_LOC_number": 22273, + "eval_LOC_precision": 0.7693184697238409, + "eval_LOC_recall": 0.8179859022134423, + "eval_ORG_f1": 0.6453279898755536, + "eval_ORG_number": 19008, + "eval_ORG_precision": 0.6468287526427061, + "eval_ORG_recall": 0.6438341750841751, + "eval_PER_f1": 0.7972594320010123, + "eval_PER_number": 27575, + "eval_PER_precision": 0.7948597793958618, + "eval_PER_recall": 0.7996736174070717, + "eval_loss": 0.20959524810314178, + "eval_overall_accuracy": 0.9355766730148269, + "eval_overall_f1": 0.7544252873563217, + "eval_overall_precision": 0.7464460366200387, + "eval_overall_recall": 0.7625769722319042, + "eval_runtime": 72.0749, + "eval_samples_per_second": 525.661, + "eval_steps_per_second": 65.709, + "step": 116000 + }, + { + "epoch": 0.66, + "learning_rate": 3.90070177231093e-05, + "loss": 0.2222, + "step": 116200 + }, + { + "epoch": 0.66, + "learning_rate": 3.898809692745199e-05, + "loss": 0.2284, + "step": 116400 + }, + { + "epoch": 0.66, + "learning_rate": 3.8969176131794695e-05, + "loss": 0.2219, + "step": 116600 + }, + { + "epoch": 0.66, + "learning_rate": 3.89502553361374e-05, + "loss": 0.2145, + "step": 116800 + }, + { + "epoch": 0.66, + "learning_rate": 3.89313345404801e-05, + "loss": 0.2267, + "step": 117000 + }, + { + "epoch": 0.67, + "learning_rate": 3.89124137448228e-05, + "loss": 0.2298, + "step": 117200 + }, + { + "epoch": 0.67, + "learning_rate": 3.88934929491655e-05, + "loss": 0.2189, + "step": 117400 + }, + { + "epoch": 0.67, + "learning_rate": 3.88745721535082e-05, + "loss": 0.2266, + "step": 117600 + }, + { + "epoch": 0.67, + "learning_rate": 3.88556513578509e-05, + "loss": 0.2172, + "step": 117800 + }, + { + "epoch": 0.67, + "learning_rate": 3.8836730562193605e-05, + "loss": 0.2125, + "step": 118000 + }, + { + "epoch": 0.67, + "eval_LOC_f1": 0.7924553015914598, + "eval_LOC_number": 22273, + "eval_LOC_precision": 0.7712246112008159, + "eval_LOC_recall": 0.8148879809634985, + "eval_ORG_f1": 0.6456168613236676, + "eval_ORG_number": 19008, + "eval_ORG_precision": 0.6592641333552668, + "eval_ORG_recall": 0.6325231481481481, + "eval_PER_f1": 0.7941086443760239, + "eval_PER_number": 27575, + "eval_PER_precision": 0.7885853239879845, + "eval_PER_recall": 0.7997098821396192, + "eval_loss": 0.21121831238269806, + "eval_overall_accuracy": 0.9356025179420331, + "eval_overall_f1": 0.7536564423375256, + "eval_overall_precision": 0.7489065748906575, + "eval_overall_recall": 0.758466945509469, + "eval_runtime": 73.1539, + "eval_samples_per_second": 517.908, + "eval_steps_per_second": 64.74, + "step": 118000 + }, + { + "epoch": 0.67, + "learning_rate": 3.881780976653631e-05, + "loss": 0.221, + "step": 118200 + }, + { + "epoch": 0.67, + "learning_rate": 3.8798888970879e-05, + "loss": 0.2152, + "step": 118400 + }, + { + "epoch": 0.67, + "learning_rate": 3.8779968175221705e-05, + "loss": 0.2129, + "step": 118600 + }, + { + "epoch": 0.67, + "learning_rate": 3.876104737956441e-05, + "loss": 0.2192, + "step": 118800 + }, + { + "epoch": 0.68, + "learning_rate": 3.874212658390711e-05, + "loss": 0.221, + "step": 119000 + }, + { + "epoch": 0.68, + "learning_rate": 3.872320578824981e-05, + "loss": 0.2139, + "step": 119200 + }, + { + "epoch": 0.68, + "learning_rate": 3.870428499259251e-05, + "loss": 0.2142, + "step": 119400 + }, + { + "epoch": 0.68, + "learning_rate": 3.868536419693521e-05, + "loss": 0.2198, + "step": 119600 + }, + { + "epoch": 0.68, + "learning_rate": 3.866644340127791e-05, + "loss": 0.2132, + "step": 119800 + }, + { + "epoch": 0.68, + "learning_rate": 3.8647522605620615e-05, + "loss": 0.2243, + "step": 120000 + }, + { + "epoch": 0.68, + "eval_LOC_f1": 0.7922724645413721, + "eval_LOC_number": 22273, + "eval_LOC_precision": 0.7846668721652208, + "eval_LOC_recall": 0.8000269384456516, + "eval_ORG_f1": 0.6461916461916462, + "eval_ORG_number": 19008, + "eval_ORG_precision": 0.6562160989368626, + "eval_ORG_recall": 0.6364688552188552, + "eval_PER_f1": 0.7950493802730755, + "eval_PER_number": 27575, + "eval_PER_precision": 0.7888198757763976, + "eval_PER_recall": 0.8013780598368087, + "eval_loss": 0.2115221470594406, + "eval_overall_accuracy": 0.9354264493754412, + "eval_overall_f1": 0.7537586494221643, + "eval_overall_precision": 0.7521074625139172, + "eval_overall_recall": 0.7554171023585454, + "eval_runtime": 72.2222, + "eval_samples_per_second": 524.59, + "eval_steps_per_second": 65.575, + "step": 120000 + }, + { + "epoch": 0.68, + "learning_rate": 3.862860180996332e-05, + "loss": 0.2196, + "step": 120200 + }, + { + "epoch": 0.68, + "learning_rate": 3.860968101430601e-05, + "loss": 0.2226, + "step": 120400 + }, + { + "epoch": 0.68, + "learning_rate": 3.8590760218648715e-05, + "loss": 0.2133, + "step": 120600 + }, + { + "epoch": 0.69, + "learning_rate": 3.857183942299142e-05, + "loss": 0.2257, + "step": 120800 + }, + { + "epoch": 0.69, + "learning_rate": 3.855291862733412e-05, + "loss": 0.2277, + "step": 121000 + }, + { + "epoch": 0.69, + "learning_rate": 3.853399783167682e-05, + "loss": 0.214, + "step": 121200 + }, + { + "epoch": 0.69, + "learning_rate": 3.851507703601952e-05, + "loss": 0.2106, + "step": 121400 + }, + { + "epoch": 0.69, + "learning_rate": 3.849615624036222e-05, + "loss": 0.2169, + "step": 121600 + }, + { + "epoch": 0.69, + "learning_rate": 3.847723544470492e-05, + "loss": 0.2249, + "step": 121800 + }, + { + "epoch": 0.69, + "learning_rate": 3.8458314649047624e-05, + "loss": 0.2122, + "step": 122000 + }, + { + "epoch": 0.69, + "eval_LOC_f1": 0.7960111966410076, + "eval_LOC_number": 22273, + "eval_LOC_precision": 0.775953954380729, + "eval_LOC_recall": 0.8171328514344722, + "eval_ORG_f1": 0.6522518721653834, + "eval_ORG_number": 19008, + "eval_ORG_precision": 0.6538380207231973, + "eval_ORG_recall": 0.6506734006734006, + "eval_PER_f1": 0.7968987204510952, + "eval_PER_number": 27575, + "eval_PER_precision": 0.7942861260222647, + "eval_PER_recall": 0.7995285584768812, + "eval_loss": 0.20907320082187653, + "eval_overall_accuracy": 0.9363100228243013, + "eval_overall_f1": 0.7571375122316238, + "eval_overall_precision": 0.75027093315081, + "eval_overall_recall": 0.7641309399326129, + "eval_runtime": 73.0361, + "eval_samples_per_second": 518.743, + "eval_steps_per_second": 64.845, + "step": 122000 + }, + { + "epoch": 0.69, + "learning_rate": 3.843939385339033e-05, + "loss": 0.2203, + "step": 122200 + }, + { + "epoch": 0.69, + "learning_rate": 3.842047305773302e-05, + "loss": 0.2207, + "step": 122400 + }, + { + "epoch": 0.7, + "learning_rate": 3.8401552262075725e-05, + "loss": 0.2251, + "step": 122600 + }, + { + "epoch": 0.7, + "learning_rate": 3.838263146641843e-05, + "loss": 0.2227, + "step": 122800 + }, + { + "epoch": 0.7, + "learning_rate": 3.836371067076113e-05, + "loss": 0.2165, + "step": 123000 + }, + { + "epoch": 0.7, + "learning_rate": 3.834478987510383e-05, + "loss": 0.2241, + "step": 123200 + }, + { + "epoch": 0.7, + "learning_rate": 3.832586907944653e-05, + "loss": 0.2231, + "step": 123400 + }, + { + "epoch": 0.7, + "learning_rate": 3.830694828378923e-05, + "loss": 0.22, + "step": 123600 + }, + { + "epoch": 0.7, + "learning_rate": 3.828802748813193e-05, + "loss": 0.2277, + "step": 123800 + }, + { + "epoch": 0.7, + "learning_rate": 3.8269106692474634e-05, + "loss": 0.2208, + "step": 124000 + }, + { + "epoch": 0.7, + "eval_LOC_f1": 0.7915342514124293, + "eval_LOC_number": 22273, + "eval_LOC_precision": 0.778375797560658, + "eval_LOC_recall": 0.805145243119472, + "eval_ORG_f1": 0.6526839288610194, + "eval_ORG_number": 19008, + "eval_ORG_precision": 0.6711506392440244, + "eval_ORG_recall": 0.6352062289562289, + "eval_PER_f1": 0.796792774854217, + "eval_PER_number": 27575, + "eval_PER_precision": 0.78153663725456, + "eval_PER_recall": 0.8126563916591115, + "eval_loss": 0.21086519956588745, + "eval_overall_accuracy": 0.9357947395881288, + "eval_overall_f1": 0.7565929069414973, + "eval_overall_precision": 0.7520013772918998, + "eval_overall_recall": 0.7612408504705472, + "eval_runtime": 72.5914, + "eval_samples_per_second": 521.921, + "eval_steps_per_second": 65.242, + "step": 124000 + }, + { + "epoch": 0.7, + "learning_rate": 3.825018589681734e-05, + "loss": 0.2179, + "step": 124200 + }, + { + "epoch": 0.71, + "learning_rate": 3.823126510116003e-05, + "loss": 0.2228, + "step": 124400 + }, + { + "epoch": 0.71, + "learning_rate": 3.8212344305502735e-05, + "loss": 0.2174, + "step": 124600 + }, + { + "epoch": 0.71, + "learning_rate": 3.819342350984544e-05, + "loss": 0.2208, + "step": 124800 + }, + { + "epoch": 0.71, + "learning_rate": 3.817450271418814e-05, + "loss": 0.2184, + "step": 125000 + }, + { + "epoch": 0.71, + "learning_rate": 3.815558191853084e-05, + "loss": 0.2244, + "step": 125200 + }, + { + "epoch": 0.71, + "learning_rate": 3.813666112287354e-05, + "loss": 0.2162, + "step": 125400 + }, + { + "epoch": 0.71, + "learning_rate": 3.811774032721624e-05, + "loss": 0.2126, + "step": 125600 + }, + { + "epoch": 0.71, + "learning_rate": 3.809881953155894e-05, + "loss": 0.2104, + "step": 125800 + }, + { + "epoch": 0.72, + "learning_rate": 3.8079898735901644e-05, + "loss": 0.2112, + "step": 126000 + }, + { + "epoch": 0.72, + "eval_LOC_f1": 0.7951321127069401, + "eval_LOC_number": 22273, + "eval_LOC_precision": 0.7770872621292645, + "eval_LOC_recall": 0.8140349301845283, + "eval_ORG_f1": 0.6479030457662255, + "eval_ORG_number": 19008, + "eval_ORG_precision": 0.6546890106348695, + "eval_ORG_recall": 0.6412563131313131, + "eval_PER_f1": 0.7959102712060012, + "eval_PER_number": 27575, + "eval_PER_precision": 0.7915426275958538, + "eval_PER_recall": 0.8003263825929284, + "eval_loss": 0.2088223546743393, + "eval_overall_accuracy": 0.9359918071580756, + "eval_overall_f1": 0.7554997945012871, + "eval_overall_precision": 0.7502255445289342, + "eval_overall_recall": 0.7608487277797142, + "eval_runtime": 72.1078, + "eval_samples_per_second": 525.421, + "eval_steps_per_second": 65.679, + "step": 126000 + }, + { + "epoch": 0.72, + "learning_rate": 3.8060977940244346e-05, + "loss": 0.2139, + "step": 126200 + }, + { + "epoch": 0.72, + "learning_rate": 3.804205714458705e-05, + "loss": 0.2175, + "step": 126400 + }, + { + "epoch": 0.72, + "learning_rate": 3.8023136348929744e-05, + "loss": 0.213, + "step": 126600 + }, + { + "epoch": 0.72, + "learning_rate": 3.800421555327245e-05, + "loss": 0.2202, + "step": 126800 + }, + { + "epoch": 0.72, + "learning_rate": 3.798529475761515e-05, + "loss": 0.2076, + "step": 127000 + }, + { + "epoch": 0.72, + "learning_rate": 3.796637396195785e-05, + "loss": 0.2189, + "step": 127200 + }, + { + "epoch": 0.72, + "learning_rate": 3.7947453166300554e-05, + "loss": 0.2247, + "step": 127400 + }, + { + "epoch": 0.72, + "learning_rate": 3.792853237064325e-05, + "loss": 0.2135, + "step": 127600 + }, + { + "epoch": 0.73, + "learning_rate": 3.790961157498595e-05, + "loss": 0.2212, + "step": 127800 + }, + { + "epoch": 0.73, + "learning_rate": 3.7890690779328654e-05, + "loss": 0.2192, + "step": 128000 + }, + { + "epoch": 0.73, + "eval_LOC_f1": 0.7949297133709072, + "eval_LOC_number": 22273, + "eval_LOC_precision": 0.7769870530738232, + "eval_LOC_recall": 0.813720648318592, + "eval_ORG_f1": 0.6530171832429661, + "eval_ORG_number": 19008, + "eval_ORG_precision": 0.6700990976028345, + "eval_ORG_recall": 0.6367845117845118, + "eval_PER_f1": 0.7973942295712226, + "eval_PER_number": 27575, + "eval_PER_precision": 0.7829977628635347, + "eval_PER_recall": 0.8123300090661831, + "eval_loss": 0.20989789068698883, + "eval_overall_accuracy": 0.9361161858702552, + "eval_overall_f1": 0.7580390772975736, + "eval_overall_precision": 0.75186079403403, + "eval_overall_recall": 0.7643197397467177, + "eval_runtime": 72.6214, + "eval_samples_per_second": 521.706, + "eval_steps_per_second": 65.215, + "step": 128000 + }, + { + "epoch": 0.73, + "learning_rate": 3.7871769983671356e-05, + "loss": 0.2231, + "step": 128200 + }, + { + "epoch": 0.73, + "learning_rate": 3.785284918801406e-05, + "loss": 0.21, + "step": 128400 + }, + { + "epoch": 0.73, + "learning_rate": 3.7833928392356754e-05, + "loss": 0.2307, + "step": 128600 + }, + { + "epoch": 0.73, + "learning_rate": 3.7815007596699457e-05, + "loss": 0.2108, + "step": 128800 + }, + { + "epoch": 0.73, + "learning_rate": 3.779608680104216e-05, + "loss": 0.229, + "step": 129000 + }, + { + "epoch": 0.73, + "learning_rate": 3.777716600538486e-05, + "loss": 0.2252, + "step": 129200 + }, + { + "epoch": 0.73, + "learning_rate": 3.7758245209727564e-05, + "loss": 0.2151, + "step": 129400 + }, + { + "epoch": 0.74, + "learning_rate": 3.773932441407026e-05, + "loss": 0.2177, + "step": 129600 + }, + { + "epoch": 0.74, + "learning_rate": 3.772040361841296e-05, + "loss": 0.2164, + "step": 129800 + }, + { + "epoch": 0.74, + "learning_rate": 3.7701482822755664e-05, + "loss": 0.2247, + "step": 130000 + }, + { + "epoch": 0.74, + "eval_LOC_f1": 0.7919954658412174, + "eval_LOC_number": 22273, + "eval_LOC_precision": 0.7697131477479768, + "eval_LOC_recall": 0.81560633951421, + "eval_ORG_f1": 0.6515643564356435, + "eval_ORG_number": 19008, + "eval_ORG_precision": 0.6539990459532518, + "eval_ORG_recall": 0.6491477272727273, + "eval_PER_f1": 0.7982624434389141, + "eval_PER_number": 27575, + "eval_PER_precision": 0.7968202348690153, + "eval_PER_recall": 0.7997098821396192, + "eval_loss": 0.20724231004714966, + "eval_overall_accuracy": 0.9361969512677745, + "eval_overall_f1": 0.7562212677789049, + "eval_overall_precision": 0.749283606347034, + "eval_overall_recall": 0.7632886023004531, + "eval_runtime": 73.2952, + "eval_samples_per_second": 516.91, + "eval_steps_per_second": 64.615, + "step": 130000 + }, + { + "epoch": 0.74, + "learning_rate": 3.7682562027098366e-05, + "loss": 0.2207, + "step": 130200 + }, + { + "epoch": 0.74, + "learning_rate": 3.766364123144107e-05, + "loss": 0.2122, + "step": 130400 + }, + { + "epoch": 0.74, + "learning_rate": 3.7644720435783764e-05, + "loss": 0.2177, + "step": 130600 + }, + { + "epoch": 0.74, + "learning_rate": 3.7625799640126466e-05, + "loss": 0.2073, + "step": 130800 + }, + { + "epoch": 0.74, + "learning_rate": 3.760687884446917e-05, + "loss": 0.2175, + "step": 131000 + }, + { + "epoch": 0.74, + "learning_rate": 3.758795804881187e-05, + "loss": 0.2211, + "step": 131200 + }, + { + "epoch": 0.75, + "learning_rate": 3.7569037253154573e-05, + "loss": 0.2173, + "step": 131400 + }, + { + "epoch": 0.75, + "learning_rate": 3.7550116457497276e-05, + "loss": 0.2216, + "step": 131600 + }, + { + "epoch": 0.75, + "learning_rate": 3.753119566183997e-05, + "loss": 0.2147, + "step": 131800 + }, + { + "epoch": 0.75, + "learning_rate": 3.7512274866182674e-05, + "loss": 0.2053, + "step": 132000 + }, + { + "epoch": 0.75, + "eval_LOC_f1": 0.7929907153131946, + "eval_LOC_number": 22273, + "eval_LOC_precision": 0.7705536024397476, + "eval_LOC_recall": 0.8167736721591164, + "eval_ORG_f1": 0.6493764248357249, + "eval_ORG_number": 19008, + "eval_ORG_precision": 0.662362532144225, + "eval_ORG_recall": 0.6368897306397306, + "eval_PER_f1": 0.7977712693763841, + "eval_PER_number": 27575, + "eval_PER_precision": 0.7858977516625031, + "eval_PER_recall": 0.810009066183137, + "eval_loss": 0.20749878883361816, + "eval_overall_accuracy": 0.9358286610550869, + "eval_overall_f1": 0.7564366965357171, + "eval_overall_precision": 0.7486310040252038, + "eval_overall_recall": 0.7644068781224584, + "eval_runtime": 72.7757, + "eval_samples_per_second": 520.6, + "eval_steps_per_second": 65.077, + "step": 132000 + }, + { + "epoch": 0.75, + "learning_rate": 3.7493354070525376e-05, + "loss": 0.2227, + "step": 132200 + }, + { + "epoch": 0.75, + "learning_rate": 3.747443327486808e-05, + "loss": 0.219, + "step": 132400 + }, + { + "epoch": 0.75, + "learning_rate": 3.745551247921078e-05, + "loss": 0.2171, + "step": 132600 + }, + { + "epoch": 0.75, + "learning_rate": 3.7436591683553476e-05, + "loss": 0.2099, + "step": 132800 + }, + { + "epoch": 0.75, + "learning_rate": 3.741767088789618e-05, + "loss": 0.207, + "step": 133000 + }, + { + "epoch": 0.76, + "learning_rate": 3.739875009223888e-05, + "loss": 0.2193, + "step": 133200 + }, + { + "epoch": 0.76, + "learning_rate": 3.737982929658158e-05, + "loss": 0.2178, + "step": 133400 + }, + { + "epoch": 0.76, + "learning_rate": 3.7360908500924286e-05, + "loss": 0.2147, + "step": 133600 + }, + { + "epoch": 0.76, + "learning_rate": 3.734198770526698e-05, + "loss": 0.2189, + "step": 133800 + }, + { + "epoch": 0.76, + "learning_rate": 3.7323066909609684e-05, + "loss": 0.2268, + "step": 134000 + }, + { + "epoch": 0.76, + "eval_LOC_f1": 0.7908817128662813, + "eval_LOC_number": 22273, + "eval_LOC_precision": 0.7747868033422345, + "eval_LOC_recall": 0.8076594980469627, + "eval_ORG_f1": 0.6464787597065857, + "eval_ORG_number": 19008, + "eval_ORG_precision": 0.6606623098467791, + "eval_ORG_recall": 0.6328914141414141, + "eval_PER_f1": 0.7972745838679944, + "eval_PER_number": 27575, + "eval_PER_precision": 0.7946750252197723, + "eval_PER_recall": 0.7998912058023572, + "eval_loss": 0.2063896656036377, + "eval_overall_accuracy": 0.9357898936642777, + "eval_overall_f1": 0.7545114061967996, + "eval_overall_precision": 0.7527282713961523, + "eval_overall_recall": 0.7563030091785756, + "eval_runtime": 72.9421, + "eval_samples_per_second": 519.412, + "eval_steps_per_second": 64.928, + "step": 134000 + }, + { + "epoch": 0.76, + "learning_rate": 3.7304146113952386e-05, + "loss": 0.217, + "step": 134200 + }, + { + "epoch": 0.76, + "learning_rate": 3.728522531829509e-05, + "loss": 0.2153, + "step": 134400 + }, + { + "epoch": 0.76, + "learning_rate": 3.726630452263779e-05, + "loss": 0.2107, + "step": 134600 + }, + { + "epoch": 0.77, + "learning_rate": 3.7247383726980486e-05, + "loss": 0.2258, + "step": 134800 + }, + { + "epoch": 0.77, + "learning_rate": 3.722846293132319e-05, + "loss": 0.2168, + "step": 135000 + }, + { + "epoch": 0.77, + "learning_rate": 3.720954213566589e-05, + "loss": 0.2145, + "step": 135200 + }, + { + "epoch": 0.77, + "learning_rate": 3.719062134000859e-05, + "loss": 0.212, + "step": 135400 + }, + { + "epoch": 0.77, + "learning_rate": 3.7171700544351296e-05, + "loss": 0.2193, + "step": 135600 + }, + { + "epoch": 0.77, + "learning_rate": 3.715277974869399e-05, + "loss": 0.2149, + "step": 135800 + }, + { + "epoch": 0.77, + "learning_rate": 3.7133858953036693e-05, + "loss": 0.2184, + "step": 136000 + }, + { + "epoch": 0.77, + "eval_LOC_f1": 0.7926615168539326, + "eval_LOC_number": 22273, + "eval_LOC_precision": 0.7752736638763683, + "eval_LOC_recall": 0.8108472141157456, + "eval_ORG_f1": 0.6474616166479431, + "eval_ORG_number": 19008, + "eval_ORG_precision": 0.6585591468059636, + "eval_ORG_recall": 0.6367319023569024, + "eval_PER_f1": 0.7980052588629977, + "eval_PER_number": 27575, + "eval_PER_precision": 0.7980776206021037, + "eval_PER_recall": 0.797932910244787, + "eval_loss": 0.20675942301750183, + "eval_overall_accuracy": 0.9362954850527478, + "eval_overall_f1": 0.7554870056988103, + "eval_overall_precision": 0.7533757924988808, + "eval_overall_recall": 0.7576100848146857, + "eval_runtime": 73.359, + "eval_samples_per_second": 516.46, + "eval_steps_per_second": 64.559, + "step": 136000 + }, + { + "epoch": 0.77, + "learning_rate": 3.7114938157379396e-05, + "loss": 0.2167, + "step": 136200 + }, + { + "epoch": 0.77, + "learning_rate": 3.70960173617221e-05, + "loss": 0.2101, + "step": 136400 + }, + { + "epoch": 0.78, + "learning_rate": 3.70770965660648e-05, + "loss": 0.2306, + "step": 136600 + }, + { + "epoch": 0.78, + "learning_rate": 3.70581757704075e-05, + "loss": 0.2284, + "step": 136800 + }, + { + "epoch": 0.78, + "learning_rate": 3.70392549747502e-05, + "loss": 0.2239, + "step": 137000 + }, + { + "epoch": 0.78, + "learning_rate": 3.70203341790929e-05, + "loss": 0.2152, + "step": 137200 + }, + { + "epoch": 0.78, + "learning_rate": 3.70014133834356e-05, + "loss": 0.2114, + "step": 137400 + }, + { + "epoch": 0.78, + "learning_rate": 3.6982492587778305e-05, + "loss": 0.2121, + "step": 137600 + }, + { + "epoch": 0.78, + "learning_rate": 3.696357179212101e-05, + "loss": 0.2191, + "step": 137800 + }, + { + "epoch": 0.78, + "learning_rate": 3.69446509964637e-05, + "loss": 0.2165, + "step": 138000 + }, + { + "epoch": 0.78, + "eval_LOC_f1": 0.7930782772982953, + "eval_LOC_number": 22273, + "eval_LOC_precision": 0.7599259411643695, + "eval_LOC_recall": 0.8292551519777309, + "eval_ORG_f1": 0.6492922698204586, + "eval_ORG_number": 19008, + "eval_ORG_precision": 0.6606588619656956, + "eval_ORG_recall": 0.6383101851851852, + "eval_PER_f1": 0.7973900975108595, + "eval_PER_number": 27575, + "eval_PER_precision": 0.7926610764710098, + "eval_PER_recall": 0.8021758839528559, + "eval_loss": 0.20885895192623138, + "eval_overall_accuracy": 0.9360725725555948, + "eval_overall_f1": 0.7562539445751335, + "eval_overall_precision": 0.7470386533665836, + "eval_overall_recall": 0.7656994306959451, + "eval_runtime": 72.915, + "eval_samples_per_second": 519.605, + "eval_steps_per_second": 64.952, + "step": 138000 + }, + { + "epoch": 0.78, + "learning_rate": 3.6925730200806406e-05, + "loss": 0.2203, + "step": 138200 + }, + { + "epoch": 0.79, + "learning_rate": 3.690680940514911e-05, + "loss": 0.2036, + "step": 138400 + }, + { + "epoch": 0.79, + "learning_rate": 3.688788860949181e-05, + "loss": 0.2143, + "step": 138600 + }, + { + "epoch": 0.79, + "learning_rate": 3.686896781383451e-05, + "loss": 0.2178, + "step": 138800 + }, + { + "epoch": 0.79, + "learning_rate": 3.685004701817721e-05, + "loss": 0.2181, + "step": 139000 + }, + { + "epoch": 0.79, + "learning_rate": 3.683112622251991e-05, + "loss": 0.2142, + "step": 139200 + }, + { + "epoch": 0.79, + "learning_rate": 3.681220542686261e-05, + "loss": 0.2212, + "step": 139400 + }, + { + "epoch": 0.79, + "learning_rate": 3.6793284631205315e-05, + "loss": 0.2128, + "step": 139600 + }, + { + "epoch": 0.79, + "learning_rate": 3.677436383554802e-05, + "loss": 0.22, + "step": 139800 + }, + { + "epoch": 0.79, + "learning_rate": 3.675544303989071e-05, + "loss": 0.2308, + "step": 140000 + }, + { + "epoch": 0.79, + "eval_LOC_f1": 0.7954807266282676, + "eval_LOC_number": 22273, + "eval_LOC_precision": 0.7851489045349194, + "eval_LOC_recall": 0.806088088717281, + "eval_ORG_f1": 0.6516998075689545, + "eval_ORG_number": 19008, + "eval_ORG_precision": 0.6623207301173403, + "eval_ORG_recall": 0.6414141414141414, + "eval_PER_f1": 0.799718650242574, + "eval_PER_number": 27575, + "eval_PER_precision": 0.7954578071182549, + "eval_PER_recall": 0.8040253853127833, + "eval_loss": 0.20637771487236023, + "eval_overall_accuracy": 0.9367073885800958, + "eval_overall_f1": 0.7582009086758983, + "eval_overall_precision": 0.7566054926461018, + "eval_overall_recall": 0.759803067270826, + "eval_runtime": 74.5563, + "eval_samples_per_second": 508.166, + "eval_steps_per_second": 63.522, + "step": 140000 + }, + { + "epoch": 0.8, + "learning_rate": 3.6736522244233416e-05, + "loss": 0.2146, + "step": 140200 + }, + { + "epoch": 0.8, + "learning_rate": 3.671760144857612e-05, + "loss": 0.2273, + "step": 140400 + }, + { + "epoch": 0.8, + "learning_rate": 3.669868065291882e-05, + "loss": 0.2134, + "step": 140600 + }, + { + "epoch": 0.8, + "learning_rate": 3.667975985726152e-05, + "loss": 0.2137, + "step": 140800 + }, + { + "epoch": 0.8, + "learning_rate": 3.666083906160422e-05, + "loss": 0.216, + "step": 141000 + }, + { + "epoch": 0.8, + "learning_rate": 3.664191826594692e-05, + "loss": 0.2164, + "step": 141200 + }, + { + "epoch": 0.8, + "learning_rate": 3.662299747028962e-05, + "loss": 0.2203, + "step": 141400 + }, + { + "epoch": 0.8, + "learning_rate": 3.6604076674632325e-05, + "loss": 0.2143, + "step": 141600 + }, + { + "epoch": 0.8, + "learning_rate": 3.658515587897503e-05, + "loss": 0.2149, + "step": 141800 + }, + { + "epoch": 0.81, + "learning_rate": 3.656623508331772e-05, + "loss": 0.2078, + "step": 142000 + }, + { + "epoch": 0.81, + "eval_LOC_f1": 0.7933356353591161, + "eval_LOC_number": 22273, + "eval_LOC_precision": 0.7638282840876034, + "eval_LOC_recall": 0.825214385129978, + "eval_ORG_f1": 0.6465153043146447, + "eval_ORG_number": 19008, + "eval_ORG_precision": 0.6707765397884735, + "eval_ORG_recall": 0.6239478114478114, + "eval_PER_f1": 0.7958876538956683, + "eval_PER_number": 27575, + "eval_PER_precision": 0.7958732230925443, + "eval_PER_recall": 0.7959020852221215, + "eval_loss": 0.20695120096206665, + "eval_overall_accuracy": 0.9361371848736102, + "eval_overall_f1": 0.7553699629458083, + "eval_overall_precision": 0.7528418926716677, + "eval_overall_recall": 0.7579150691297781, + "eval_runtime": 71.9634, + "eval_samples_per_second": 526.476, + "eval_steps_per_second": 65.811, + "step": 142000 + }, + { + "epoch": 0.81, + "learning_rate": 3.6547314287660425e-05, + "loss": 0.2134, + "step": 142200 + }, + { + "epoch": 0.81, + "learning_rate": 3.652839349200313e-05, + "loss": 0.2154, + "step": 142400 + }, + { + "epoch": 0.81, + "learning_rate": 3.650947269634583e-05, + "loss": 0.2172, + "step": 142600 + }, + { + "epoch": 0.81, + "learning_rate": 3.649055190068853e-05, + "loss": 0.2228, + "step": 142800 + }, + { + "epoch": 0.81, + "learning_rate": 3.647163110503123e-05, + "loss": 0.2128, + "step": 143000 + }, + { + "epoch": 0.81, + "learning_rate": 3.645271030937393e-05, + "loss": 0.2129, + "step": 143200 + }, + { + "epoch": 0.81, + "learning_rate": 3.643378951371663e-05, + "loss": 0.2195, + "step": 143400 + }, + { + "epoch": 0.82, + "learning_rate": 3.6414868718059335e-05, + "loss": 0.2159, + "step": 143600 + }, + { + "epoch": 0.82, + "learning_rate": 3.639594792240204e-05, + "loss": 0.2163, + "step": 143800 + }, + { + "epoch": 0.82, + "learning_rate": 3.637702712674473e-05, + "loss": 0.2044, + "step": 144000 + }, + { + "epoch": 0.82, + "eval_LOC_f1": 0.7944116113332174, + "eval_LOC_number": 22273, + "eval_LOC_precision": 0.769693907624942, + "eval_LOC_recall": 0.8207695415974499, + "eval_ORG_f1": 0.6488801924958354, + "eval_ORG_number": 19008, + "eval_ORG_precision": 0.6522779224921589, + "eval_ORG_recall": 0.6455176767676768, + "eval_PER_f1": 0.793907477348086, + "eval_PER_number": 27575, + "eval_PER_precision": 0.7767330511414892, + "eval_PER_recall": 0.8118585675430644, + "eval_loss": 0.20746921002864838, + "eval_overall_accuracy": 0.9358916580651518, + "eval_overall_f1": 0.7549629207073588, + "eval_overall_precision": 0.7415947551271994, + "eval_overall_recall": 0.7688218891599861, + "eval_runtime": 71.7428, + "eval_samples_per_second": 528.095, + "eval_steps_per_second": 66.014, + "step": 144000 + }, + { + "epoch": 0.82, + "learning_rate": 3.6358106331087435e-05, + "loss": 0.2084, + "step": 144200 + }, + { + "epoch": 0.82, + "learning_rate": 3.633918553543014e-05, + "loss": 0.2207, + "step": 144400 + }, + { + "epoch": 0.82, + "learning_rate": 3.632026473977284e-05, + "loss": 0.2122, + "step": 144600 + }, + { + "epoch": 0.82, + "learning_rate": 3.630134394411554e-05, + "loss": 0.22, + "step": 144800 + }, + { + "epoch": 0.82, + "learning_rate": 3.628242314845824e-05, + "loss": 0.2144, + "step": 145000 + }, + { + "epoch": 0.82, + "learning_rate": 3.626350235280094e-05, + "loss": 0.2113, + "step": 145200 + }, + { + "epoch": 0.83, + "learning_rate": 3.624458155714364e-05, + "loss": 0.2113, + "step": 145400 + }, + { + "epoch": 0.83, + "learning_rate": 3.6225660761486345e-05, + "loss": 0.2128, + "step": 145600 + }, + { + "epoch": 0.83, + "learning_rate": 3.620673996582905e-05, + "loss": 0.2108, + "step": 145800 + }, + { + "epoch": 0.83, + "learning_rate": 3.618781917017174e-05, + "loss": 0.2098, + "step": 146000 + }, + { + "epoch": 0.83, + "eval_LOC_f1": 0.7963572191115275, + "eval_LOC_number": 22273, + "eval_LOC_precision": 0.7735545585372048, + "eval_LOC_recall": 0.8205450545503524, + "eval_ORG_f1": 0.6512293201630098, + "eval_ORG_number": 19008, + "eval_ORG_precision": 0.6686062621224717, + "eval_ORG_recall": 0.6347327441077442, + "eval_PER_f1": 0.7988889884418958, + "eval_PER_number": 27575, + "eval_PER_precision": 0.7896209705986539, + "eval_PER_recall": 0.808377153218495, + "eval_loss": 0.20629006624221802, + "eval_overall_accuracy": 0.9366153160269239, + "eval_overall_f1": 0.7586211866788704, + "eval_overall_precision": 0.7529506015650707, + "eval_overall_recall": 0.7643778319972115, + "eval_runtime": 74.1655, + "eval_samples_per_second": 510.844, + "eval_steps_per_second": 63.857, + "step": 146000 + }, + { + "epoch": 0.83, + "learning_rate": 3.6168898374514445e-05, + "loss": 0.2204, + "step": 146200 + }, + { + "epoch": 0.83, + "learning_rate": 3.614997757885715e-05, + "loss": 0.2069, + "step": 146400 + }, + { + "epoch": 0.83, + "learning_rate": 3.613105678319985e-05, + "loss": 0.2186, + "step": 146600 + }, + { + "epoch": 0.83, + "learning_rate": 3.611213598754255e-05, + "loss": 0.2174, + "step": 146800 + }, + { + "epoch": 0.83, + "learning_rate": 3.609321519188525e-05, + "loss": 0.2073, + "step": 147000 + }, + { + "epoch": 0.84, + "learning_rate": 3.607429439622795e-05, + "loss": 0.2142, + "step": 147200 + }, + { + "epoch": 0.84, + "learning_rate": 3.605537360057065e-05, + "loss": 0.2183, + "step": 147400 + }, + { + "epoch": 0.84, + "learning_rate": 3.6036452804913355e-05, + "loss": 0.2149, + "step": 147600 + }, + { + "epoch": 0.84, + "learning_rate": 3.601753200925606e-05, + "loss": 0.2185, + "step": 147800 + }, + { + "epoch": 0.84, + "learning_rate": 3.599861121359875e-05, + "loss": 0.2239, + "step": 148000 + }, + { + "epoch": 0.84, + "eval_LOC_f1": 0.7964371541327689, + "eval_LOC_number": 22273, + "eval_LOC_precision": 0.7824467163403223, + "eval_LOC_recall": 0.8109370089345844, + "eval_ORG_f1": 0.655743769984937, + "eval_ORG_number": 19008, + "eval_ORG_precision": 0.6587904210693994, + "eval_ORG_recall": 0.6527251683501684, + "eval_PER_f1": 0.7984921917070543, + "eval_PER_number": 27575, + "eval_PER_precision": 0.7905455837924293, + "eval_PER_recall": 0.8066001813236627, + "eval_loss": 0.20535476505756378, + "eval_overall_accuracy": 0.9367445406629547, + "eval_overall_f1": 0.7589339706856337, + "eval_overall_precision": 0.7524553189059556, + "eval_overall_recall": 0.7655251539444639, + "eval_runtime": 73.1782, + "eval_samples_per_second": 517.736, + "eval_steps_per_second": 64.719, + "step": 148000 + }, + { + "epoch": 0.84, + "learning_rate": 3.597969041794146e-05, + "loss": 0.2246, + "step": 148200 + }, + { + "epoch": 0.84, + "learning_rate": 3.596076962228416e-05, + "loss": 0.2138, + "step": 148400 + }, + { + "epoch": 0.84, + "learning_rate": 3.594184882662686e-05, + "loss": 0.2134, + "step": 148600 + }, + { + "epoch": 0.84, + "learning_rate": 3.592292803096956e-05, + "loss": 0.2081, + "step": 148800 + }, + { + "epoch": 0.85, + "learning_rate": 3.590400723531226e-05, + "loss": 0.2113, + "step": 149000 + }, + { + "epoch": 0.85, + "learning_rate": 3.588508643965497e-05, + "loss": 0.2138, + "step": 149200 + }, + { + "epoch": 0.85, + "learning_rate": 3.586616564399766e-05, + "loss": 0.2078, + "step": 149400 + }, + { + "epoch": 0.85, + "learning_rate": 3.5847244848340365e-05, + "loss": 0.2134, + "step": 149600 + }, + { + "epoch": 0.85, + "learning_rate": 3.582832405268307e-05, + "loss": 0.21, + "step": 149800 + }, + { + "epoch": 0.85, + "learning_rate": 3.580940325702576e-05, + "loss": 0.2156, + "step": 150000 + }, + { + "epoch": 0.85, + "eval_LOC_f1": 0.7981846358202348, + "eval_LOC_number": 22273, + "eval_LOC_precision": 0.7873083810106127, + "eval_LOC_recall": 0.8093655996049028, + "eval_ORG_f1": 0.6507278154967902, + "eval_ORG_number": 19008, + "eval_ORG_precision": 0.6535420535951181, + "eval_ORG_recall": 0.6479377104377104, + "eval_PER_f1": 0.7956840895020783, + "eval_PER_number": 27575, + "eval_PER_precision": 0.776657458563536, + "eval_PER_recall": 0.8156663644605621, + "eval_loss": 0.20581567287445068, + "eval_overall_accuracy": 0.9363859422979693, + "eval_overall_f1": 0.757176227804927, + "eval_overall_precision": 0.7472914486153149, + "eval_overall_recall": 0.7673260137097712, + "eval_runtime": 74.869, + "eval_samples_per_second": 506.044, + "eval_steps_per_second": 63.257, + "step": 150000 + }, + { + "epoch": 0.85, + "learning_rate": 3.579048246136847e-05, + "loss": 0.2172, + "step": 150200 + }, + { + "epoch": 0.85, + "learning_rate": 3.577156166571117e-05, + "loss": 0.2161, + "step": 150400 + }, + { + "epoch": 0.85, + "learning_rate": 3.575264087005387e-05, + "loss": 0.2126, + "step": 150600 + }, + { + "epoch": 0.86, + "learning_rate": 3.573372007439657e-05, + "loss": 0.2103, + "step": 150800 + }, + { + "epoch": 0.86, + "learning_rate": 3.571479927873927e-05, + "loss": 0.2158, + "step": 151000 + }, + { + "epoch": 0.86, + "learning_rate": 3.5695878483081977e-05, + "loss": 0.2131, + "step": 151200 + }, + { + "epoch": 0.86, + "learning_rate": 3.567695768742467e-05, + "loss": 0.2185, + "step": 151400 + }, + { + "epoch": 0.86, + "learning_rate": 3.5658036891767374e-05, + "loss": 0.2152, + "step": 151600 + }, + { + "epoch": 0.86, + "learning_rate": 3.563911609611008e-05, + "loss": 0.2103, + "step": 151800 + }, + { + "epoch": 0.86, + "learning_rate": 3.562019530045277e-05, + "loss": 0.2014, + "step": 152000 + }, + { + "epoch": 0.86, + "eval_LOC_f1": 0.7975741181090339, + "eval_LOC_number": 22273, + "eval_LOC_precision": 0.7756937961469914, + "eval_LOC_recall": 0.8207246441880304, + "eval_ORG_f1": 0.6494420056063892, + "eval_ORG_number": 19008, + "eval_ORG_precision": 0.6529299159842603, + "eval_ORG_recall": 0.6459911616161617, + "eval_PER_f1": 0.798974414925536, + "eval_PER_number": 27575, + "eval_PER_precision": 0.8012619447078562, + "eval_PER_recall": 0.7966999093381686, + "eval_loss": 0.20582552254199982, + "eval_overall_accuracy": 0.9368495356797296, + "eval_overall_f1": 0.7577283152777577, + "eval_overall_precision": 0.7526579739217653, + "eval_overall_recall": 0.7628674334843731, + "eval_runtime": 71.9536, + "eval_samples_per_second": 526.548, + "eval_steps_per_second": 65.82, + "step": 152000 + }, + { + "epoch": 0.86, + "learning_rate": 3.560127450479548e-05, + "loss": 0.218, + "step": 152200 + }, + { + "epoch": 0.87, + "learning_rate": 3.558235370913818e-05, + "loss": 0.2134, + "step": 152400 + }, + { + "epoch": 0.87, + "learning_rate": 3.556343291348088e-05, + "loss": 0.2147, + "step": 152600 + }, + { + "epoch": 0.87, + "learning_rate": 3.554451211782358e-05, + "loss": 0.207, + "step": 152800 + }, + { + "epoch": 0.87, + "learning_rate": 3.552559132216628e-05, + "loss": 0.2233, + "step": 153000 + }, + { + "epoch": 0.87, + "learning_rate": 3.5506670526508986e-05, + "loss": 0.2125, + "step": 153200 + }, + { + "epoch": 0.87, + "learning_rate": 3.548774973085168e-05, + "loss": 0.2162, + "step": 153400 + }, + { + "epoch": 0.87, + "learning_rate": 3.5468828935194384e-05, + "loss": 0.2129, + "step": 153600 + }, + { + "epoch": 0.87, + "learning_rate": 3.544990813953709e-05, + "loss": 0.206, + "step": 153800 + }, + { + "epoch": 0.87, + "learning_rate": 3.543098734387978e-05, + "loss": 0.2148, + "step": 154000 + }, + { + "epoch": 0.87, + "eval_LOC_f1": 0.7984524928416865, + "eval_LOC_number": 22273, + "eval_LOC_precision": 0.7779623477297896, + "eval_LOC_recall": 0.8200511830467382, + "eval_ORG_f1": 0.6529668083275055, + "eval_ORG_number": 19008, + "eval_ORG_precision": 0.6680242157402312, + "eval_ORG_recall": 0.6385732323232324, + "eval_PER_f1": 0.7981014706411133, + "eval_PER_number": 27575, + "eval_PER_precision": 0.7913858879737583, + "eval_PER_recall": 0.8049320036264732, + "eval_loss": 0.2045644223690033, + "eval_overall_accuracy": 0.9366880048846913, + "eval_overall_f1": 0.7592727587675296, + "eval_overall_precision": 0.754702632900495, + "eval_overall_recall": 0.7638985709306378, + "eval_runtime": 71.9576, + "eval_samples_per_second": 526.519, + "eval_steps_per_second": 65.817, + "step": 154000 + }, + { + "epoch": 0.88, + "learning_rate": 3.541206654822249e-05, + "loss": 0.2203, + "step": 154200 + }, + { + "epoch": 0.88, + "learning_rate": 3.539314575256519e-05, + "loss": 0.2082, + "step": 154400 + }, + { + "epoch": 0.88, + "learning_rate": 3.537422495690789e-05, + "loss": 0.2139, + "step": 154600 + }, + { + "epoch": 0.88, + "learning_rate": 3.535530416125059e-05, + "loss": 0.2204, + "step": 154800 + }, + { + "epoch": 0.88, + "learning_rate": 3.533638336559329e-05, + "loss": 0.2133, + "step": 155000 + }, + { + "epoch": 0.88, + "learning_rate": 3.5317462569935996e-05, + "loss": 0.2139, + "step": 155200 + }, + { + "epoch": 0.88, + "learning_rate": 3.529854177427869e-05, + "loss": 0.2089, + "step": 155400 + }, + { + "epoch": 0.88, + "learning_rate": 3.5279620978621394e-05, + "loss": 0.2184, + "step": 155600 + }, + { + "epoch": 0.88, + "learning_rate": 3.5260700182964097e-05, + "loss": 0.22, + "step": 155800 + }, + { + "epoch": 0.89, + "learning_rate": 3.524177938730679e-05, + "loss": 0.207, + "step": 156000 + }, + { + "epoch": 0.89, + "eval_LOC_f1": 0.7976265240271131, + "eval_LOC_number": 22273, + "eval_LOC_precision": 0.7704149933065596, + "eval_LOC_recall": 0.8268306918690792, + "eval_ORG_f1": 0.6557403304593866, + "eval_ORG_number": 19008, + "eval_ORG_precision": 0.6655648875643457, + "eval_ORG_recall": 0.6462015993265994, + "eval_PER_f1": 0.7976719690071947, + "eval_PER_number": 27575, + "eval_PER_precision": 0.785568605387158, + "eval_PER_recall": 0.8101541251133273, + "eval_loss": 0.20554280281066895, + "eval_overall_accuracy": 0.9368495356797296, + "eval_overall_f1": 0.7595826799281075, + "eval_overall_precision": 0.7491701625775103, + "eval_overall_recall": 0.7702887184849541, + "eval_runtime": 71.8902, + "eval_samples_per_second": 527.012, + "eval_steps_per_second": 65.878, + "step": 156000 + }, + { + "epoch": 0.89, + "learning_rate": 3.52228585916495e-05, + "loss": 0.2229, + "step": 156200 + }, + { + "epoch": 0.89, + "learning_rate": 3.52039377959922e-05, + "loss": 0.2151, + "step": 156400 + }, + { + "epoch": 0.89, + "learning_rate": 3.51850170003349e-05, + "loss": 0.2118, + "step": 156600 + }, + { + "epoch": 0.89, + "learning_rate": 3.51660962046776e-05, + "loss": 0.2191, + "step": 156800 + }, + { + "epoch": 0.89, + "learning_rate": 3.51471754090203e-05, + "loss": 0.2118, + "step": 157000 + }, + { + "epoch": 0.89, + "learning_rate": 3.5128254613363006e-05, + "loss": 0.2131, + "step": 157200 + }, + { + "epoch": 0.89, + "learning_rate": 3.51093338177057e-05, + "loss": 0.2162, + "step": 157400 + }, + { + "epoch": 0.89, + "learning_rate": 3.5090413022048404e-05, + "loss": 0.2186, + "step": 157600 + }, + { + "epoch": 0.9, + "learning_rate": 3.5071492226391106e-05, + "loss": 0.216, + "step": 157800 + }, + { + "epoch": 0.9, + "learning_rate": 3.50525714307338e-05, + "loss": 0.2168, + "step": 158000 + }, + { + "epoch": 0.9, + "eval_LOC_f1": 0.7968662562954674, + "eval_LOC_number": 22273, + "eval_LOC_precision": 0.7653065442969945, + "eval_LOC_recall": 0.8311408431733489, + "eval_ORG_f1": 0.6521552187466408, + "eval_ORG_number": 19008, + "eval_ORG_precision": 0.6665568007031422, + "eval_ORG_recall": 0.6383627946127947, + "eval_PER_f1": 0.7975602775989126, + "eval_PER_number": 27575, + "eval_PER_precision": 0.7868916104895352, + "eval_PER_recall": 0.8085222121486854, + "eval_loss": 0.20402191579341888, + "eval_overall_accuracy": 0.9366783130369889, + "eval_overall_f1": 0.7585648579329713, + "eval_overall_precision": 0.7485366060571784, + "eval_overall_recall": 0.7688654583478564, + "eval_runtime": 73.238, + "eval_samples_per_second": 517.313, + "eval_steps_per_second": 64.666, + "step": 158000 + }, + { + "epoch": 0.9, + "learning_rate": 3.503365063507651e-05, + "loss": 0.2147, + "step": 158200 + }, + { + "epoch": 0.9, + "learning_rate": 3.501472983941921e-05, + "loss": 0.209, + "step": 158400 + }, + { + "epoch": 0.9, + "learning_rate": 3.4995809043761916e-05, + "loss": 0.2219, + "step": 158600 + }, + { + "epoch": 0.9, + "learning_rate": 3.497688824810461e-05, + "loss": 0.2113, + "step": 158800 + }, + { + "epoch": 0.9, + "learning_rate": 3.495796745244731e-05, + "loss": 0.2135, + "step": 159000 + }, + { + "epoch": 0.9, + "learning_rate": 3.4939046656790016e-05, + "loss": 0.213, + "step": 159200 + }, + { + "epoch": 0.9, + "learning_rate": 3.492012586113271e-05, + "loss": 0.2236, + "step": 159400 + }, + { + "epoch": 0.91, + "learning_rate": 3.490120506547542e-05, + "loss": 0.2138, + "step": 159600 + }, + { + "epoch": 0.91, + "learning_rate": 3.4882284269818116e-05, + "loss": 0.2105, + "step": 159800 + }, + { + "epoch": 0.91, + "learning_rate": 3.486336347416081e-05, + "loss": 0.2057, + "step": 160000 + }, + { + "epoch": 0.91, + "eval_LOC_f1": 0.7943094769096082, + "eval_LOC_number": 22273, + "eval_LOC_precision": 0.7749071187598753, + "eval_LOC_recall": 0.8147083913258205, + "eval_ORG_f1": 0.654139156932453, + "eval_ORG_number": 19008, + "eval_ORG_precision": 0.6648915937618867, + "eval_ORG_recall": 0.6437289562289562, + "eval_PER_f1": 0.797669664019106, + "eval_PER_number": 27575, + "eval_PER_precision": 0.7959272097053726, + "eval_PER_recall": 0.7994197642792384, + "eval_loss": 0.2042815089225769, + "eval_overall_accuracy": 0.9368139989048212, + "eval_overall_f1": 0.7577544589945944, + "eval_overall_precision": 0.7541573163012831, + "eval_overall_recall": 0.7613860810967817, + "eval_runtime": 72.7996, + "eval_samples_per_second": 520.429, + "eval_steps_per_second": 65.055, + "step": 160000 + }, + { + "epoch": 0.91, + "learning_rate": 3.484444267850352e-05, + "loss": 0.2131, + "step": 160200 + }, + { + "epoch": 0.91, + "learning_rate": 3.4825521882846217e-05, + "loss": 0.213, + "step": 160400 + }, + { + "epoch": 0.91, + "learning_rate": 3.4806601087188926e-05, + "loss": 0.2181, + "step": 160600 + }, + { + "epoch": 0.91, + "learning_rate": 3.478768029153162e-05, + "loss": 0.2148, + "step": 160800 + }, + { + "epoch": 0.91, + "learning_rate": 3.476875949587432e-05, + "loss": 0.2296, + "step": 161000 + }, + { + "epoch": 0.92, + "learning_rate": 3.4749838700217026e-05, + "loss": 0.2137, + "step": 161200 + }, + { + "epoch": 0.92, + "learning_rate": 3.473091790455972e-05, + "loss": 0.2052, + "step": 161400 + }, + { + "epoch": 0.92, + "learning_rate": 3.471199710890243e-05, + "loss": 0.2189, + "step": 161600 + }, + { + "epoch": 0.92, + "learning_rate": 3.4693076313245126e-05, + "loss": 0.2078, + "step": 161800 + }, + { + "epoch": 0.92, + "learning_rate": 3.467415551758783e-05, + "loss": 0.2066, + "step": 162000 + }, + { + "epoch": 0.92, + "eval_LOC_f1": 0.7997034258673703, + "eval_LOC_number": 22273, + "eval_LOC_precision": 0.7774762550881954, + "eval_LOC_recall": 0.8232388991155211, + "eval_ORG_f1": 0.6586961790128444, + "eval_ORG_number": 19008, + "eval_ORG_precision": 0.6746649015389707, + "eval_ORG_recall": 0.6434659090909091, + "eval_PER_f1": 0.7984223582531723, + "eval_PER_number": 27575, + "eval_PER_precision": 0.7860355611778762, + "eval_PER_recall": 0.8112058023572076, + "eval_loss": 0.2041463404893875, + "eval_overall_accuracy": 0.9373212056012419, + "eval_overall_f1": 0.761521143375028, + "eval_overall_precision": 0.7543857148964672, + "eval_overall_recall": 0.7687928430347392, + "eval_runtime": 74.1431, + "eval_samples_per_second": 510.999, + "eval_steps_per_second": 63.877, + "step": 162000 + }, + { + "epoch": 0.92, + "learning_rate": 3.465523472193053e-05, + "loss": 0.2099, + "step": 162200 + }, + { + "epoch": 0.92, + "learning_rate": 3.4636313926273226e-05, + "loss": 0.2157, + "step": 162400 + }, + { + "epoch": 0.92, + "learning_rate": 3.4617393130615935e-05, + "loss": 0.2179, + "step": 162600 + }, + { + "epoch": 0.92, + "learning_rate": 3.459847233495863e-05, + "loss": 0.2091, + "step": 162800 + }, + { + "epoch": 0.93, + "learning_rate": 3.4579551539301333e-05, + "loss": 0.224, + "step": 163000 + }, + { + "epoch": 0.93, + "learning_rate": 3.4560630743644036e-05, + "loss": 0.2061, + "step": 163200 + }, + { + "epoch": 0.93, + "learning_rate": 3.454170994798673e-05, + "loss": 0.2266, + "step": 163400 + }, + { + "epoch": 0.93, + "learning_rate": 3.452278915232944e-05, + "loss": 0.2169, + "step": 163600 + }, + { + "epoch": 0.93, + "learning_rate": 3.4503868356672136e-05, + "loss": 0.215, + "step": 163800 + }, + { + "epoch": 0.93, + "learning_rate": 3.448494756101484e-05, + "loss": 0.2177, + "step": 164000 + }, + { + "epoch": 0.93, + "eval_LOC_f1": 0.7995789381346081, + "eval_LOC_number": 22273, + "eval_LOC_precision": 0.7815313384206465, + "eval_LOC_recall": 0.8184797737170565, + "eval_ORG_f1": 0.6569172748731584, + "eval_ORG_number": 19008, + "eval_ORG_precision": 0.6782452798476105, + "eval_ORG_recall": 0.6368897306397306, + "eval_PER_f1": 0.8027186356343182, + "eval_PER_number": 27575, + "eval_PER_precision": 0.798128629812863, + "eval_PER_recall": 0.8073617407071623, + "eval_loss": 0.20388370752334595, + "eval_overall_accuracy": 0.9374148934623642, + "eval_overall_f1": 0.7627188689505165, + "eval_overall_precision": 0.7615428050210659, + "eval_overall_recall": 0.7638985709306378, + "eval_runtime": 71.7461, + "eval_samples_per_second": 528.071, + "eval_steps_per_second": 66.011, + "step": 164000 + }, + { + "epoch": 0.93, + "learning_rate": 3.446602676535754e-05, + "loss": 0.2188, + "step": 164200 + }, + { + "epoch": 0.93, + "learning_rate": 3.4447105969700236e-05, + "loss": 0.2071, + "step": 164400 + }, + { + "epoch": 0.93, + "learning_rate": 3.4428185174042945e-05, + "loss": 0.2116, + "step": 164600 + }, + { + "epoch": 0.94, + "learning_rate": 3.440926437838564e-05, + "loss": 0.2144, + "step": 164800 + }, + { + "epoch": 0.94, + "learning_rate": 3.439034358272834e-05, + "loss": 0.2131, + "step": 165000 + }, + { + "epoch": 0.94, + "learning_rate": 3.4371422787071046e-05, + "loss": 0.2109, + "step": 165200 + }, + { + "epoch": 0.94, + "learning_rate": 3.435250199141374e-05, + "loss": 0.2091, + "step": 165400 + }, + { + "epoch": 0.94, + "learning_rate": 3.433358119575645e-05, + "loss": 0.2205, + "step": 165600 + }, + { + "epoch": 0.94, + "learning_rate": 3.4314660400099146e-05, + "loss": 0.2185, + "step": 165800 + }, + { + "epoch": 0.94, + "learning_rate": 3.429573960444185e-05, + "loss": 0.2073, + "step": 166000 + }, + { + "epoch": 0.94, + "eval_LOC_f1": 0.798373193050705, + "eval_LOC_number": 22273, + "eval_LOC_precision": 0.7862771561670077, + "eval_LOC_recall": 0.8108472141157456, + "eval_ORG_f1": 0.6551742889771058, + "eval_ORG_number": 19008, + "eval_ORG_precision": 0.6779966235227912, + "eval_ORG_recall": 0.6338383838383839, + "eval_PER_f1": 0.7988767240153563, + "eval_PER_number": 27575, + "eval_PER_precision": 0.7833664470703057, + "eval_PER_recall": 0.8150135992747054, + "eval_loss": 0.20429351925849915, + "eval_overall_accuracy": 0.9371483676505508, + "eval_overall_f1": 0.7604928986722975, + "eval_overall_precision": 0.7573601428818344, + "eval_overall_recall": 0.7636516788660392, + "eval_runtime": 71.3779, + "eval_samples_per_second": 530.794, + "eval_steps_per_second": 66.351, + "step": 166000 + }, + { + "epoch": 0.94, + "learning_rate": 3.427681880878455e-05, + "loss": 0.2148, + "step": 166200 + }, + { + "epoch": 0.94, + "learning_rate": 3.4257898013127246e-05, + "loss": 0.218, + "step": 166400 + }, + { + "epoch": 0.95, + "learning_rate": 3.4238977217469955e-05, + "loss": 0.2225, + "step": 166600 + }, + { + "epoch": 0.95, + "learning_rate": 3.422005642181265e-05, + "loss": 0.2065, + "step": 166800 + }, + { + "epoch": 0.95, + "learning_rate": 3.420113562615535e-05, + "loss": 0.2134, + "step": 167000 + }, + { + "epoch": 0.95, + "learning_rate": 3.4182214830498055e-05, + "loss": 0.2158, + "step": 167200 + }, + { + "epoch": 0.95, + "learning_rate": 3.416329403484075e-05, + "loss": 0.2222, + "step": 167400 + }, + { + "epoch": 0.95, + "learning_rate": 3.414437323918346e-05, + "loss": 0.2115, + "step": 167600 + }, + { + "epoch": 0.95, + "learning_rate": 3.4125452443526156e-05, + "loss": 0.2165, + "step": 167800 + }, + { + "epoch": 0.95, + "learning_rate": 3.410653164786886e-05, + "loss": 0.2168, + "step": 168000 + }, + { + "epoch": 0.95, + "eval_LOC_f1": 0.7983320531109404, + "eval_LOC_number": 22273, + "eval_LOC_precision": 0.7808689678859694, + "eval_LOC_recall": 0.8165940825214385, + "eval_ORG_f1": 0.651815537118845, + "eval_ORG_number": 19008, + "eval_ORG_precision": 0.6678626628394789, + "eval_ORG_recall": 0.6365214646464646, + "eval_PER_f1": 0.7990254912849543, + "eval_PER_number": 27575, + "eval_PER_precision": 0.7895072217502124, + "eval_PER_recall": 0.8087760652765186, + "eval_loss": 0.20347106456756592, + "eval_overall_accuracy": 0.9372259024321692, + "eval_overall_f1": 0.7593421508605753, + "eval_overall_precision": 0.7549816239807052, + "eval_overall_recall": 0.7637533403044033, + "eval_runtime": 73.1912, + "eval_samples_per_second": 517.644, + "eval_steps_per_second": 64.707, + "step": 168000 + }, + { + "epoch": 0.95, + "learning_rate": 3.408761085221156e-05, + "loss": 0.2116, + "step": 168200 + }, + { + "epoch": 0.96, + "learning_rate": 3.4068690056554256e-05, + "loss": 0.2076, + "step": 168400 + }, + { + "epoch": 0.96, + "learning_rate": 3.4049769260896965e-05, + "loss": 0.208, + "step": 168600 + }, + { + "epoch": 0.96, + "learning_rate": 3.403084846523966e-05, + "loss": 0.2177, + "step": 168800 + }, + { + "epoch": 0.96, + "learning_rate": 3.401192766958236e-05, + "loss": 0.2139, + "step": 169000 + }, + { + "epoch": 0.96, + "learning_rate": 3.3993006873925065e-05, + "loss": 0.2018, + "step": 169200 + }, + { + "epoch": 0.96, + "learning_rate": 3.397408607826776e-05, + "loss": 0.2154, + "step": 169400 + }, + { + "epoch": 0.96, + "learning_rate": 3.395516528261047e-05, + "loss": 0.2201, + "step": 169600 + }, + { + "epoch": 0.96, + "learning_rate": 3.3936244486953166e-05, + "loss": 0.2234, + "step": 169800 + }, + { + "epoch": 0.96, + "learning_rate": 3.391732369129587e-05, + "loss": 0.2101, + "step": 170000 + }, + { + "epoch": 0.96, + "eval_LOC_f1": 0.7984135424975627, + "eval_LOC_number": 22273, + "eval_LOC_precision": 0.7881797103985301, + "eval_LOC_recall": 0.8089166255107081, + "eval_ORG_f1": 0.653136041229445, + "eval_ORG_number": 19008, + "eval_ORG_precision": 0.6596726589750469, + "eval_ORG_recall": 0.6467276936026936, + "eval_PER_f1": 0.79729320630398, + "eval_PER_number": 27575, + "eval_PER_precision": 0.7832750174947516, + "eval_PER_recall": 0.8118223028105168, + "eval_loss": 0.20370061695575714, + "eval_overall_accuracy": 0.9372162105844669, + "eval_overall_f1": 0.7585978550349096, + "eval_overall_precision": 0.752005023261124, + "eval_overall_recall": 0.7653073080051122, + "eval_runtime": 72.1942, + "eval_samples_per_second": 524.793, + "eval_steps_per_second": 65.601, + "step": 170000 + }, + { + "epoch": 0.97, + "learning_rate": 3.389840289563857e-05, + "loss": 0.2163, + "step": 170200 + }, + { + "epoch": 0.97, + "learning_rate": 3.3879482099981266e-05, + "loss": 0.2066, + "step": 170400 + }, + { + "epoch": 0.97, + "learning_rate": 3.3860561304323975e-05, + "loss": 0.2101, + "step": 170600 + }, + { + "epoch": 0.97, + "learning_rate": 3.384164050866667e-05, + "loss": 0.2157, + "step": 170800 + }, + { + "epoch": 0.97, + "learning_rate": 3.382271971300937e-05, + "loss": 0.2205, + "step": 171000 + }, + { + "epoch": 0.97, + "learning_rate": 3.3803798917352075e-05, + "loss": 0.2107, + "step": 171200 + }, + { + "epoch": 0.97, + "learning_rate": 3.378487812169477e-05, + "loss": 0.1989, + "step": 171400 + }, + { + "epoch": 0.97, + "learning_rate": 3.376595732603748e-05, + "loss": 0.2121, + "step": 171600 + }, + { + "epoch": 0.98, + "learning_rate": 3.3747036530380175e-05, + "loss": 0.2088, + "step": 171800 + }, + { + "epoch": 0.98, + "learning_rate": 3.372811573472288e-05, + "loss": 0.2065, + "step": 172000 + }, + { + "epoch": 0.98, + "eval_LOC_f1": 0.7957999086936672, + "eval_LOC_number": 22273, + "eval_LOC_precision": 0.7714321841018292, + "eval_LOC_recall": 0.8217572846046783, + "eval_ORG_f1": 0.6584528923833455, + "eval_ORG_number": 19008, + "eval_ORG_precision": 0.6768316391712492, + "eval_ORG_recall": 0.6410458754208754, + "eval_PER_f1": 0.7996515995572412, + "eval_PER_number": 27575, + "eval_PER_precision": 0.8002469673857776, + "eval_PER_recall": 0.7990571169537625, + "eval_loss": 0.20420855283737183, + "eval_overall_accuracy": 0.9370627563291803, + "eval_overall_f1": 0.7605325842208531, + "eval_overall_precision": 0.7582980812266289, + "eval_overall_recall": 0.7627802951086325, + "eval_runtime": 75.1352, + "eval_samples_per_second": 504.251, + "eval_steps_per_second": 63.033, + "step": 172000 + }, + { + "epoch": 0.98, + "learning_rate": 3.370919493906558e-05, + "loss": 0.2147, + "step": 172200 + }, + { + "epoch": 0.98, + "learning_rate": 3.369027414340828e-05, + "loss": 0.2078, + "step": 172400 + }, + { + "epoch": 0.98, + "learning_rate": 3.3671353347750985e-05, + "loss": 0.2092, + "step": 172600 + }, + { + "epoch": 0.98, + "learning_rate": 3.365243255209368e-05, + "loss": 0.211, + "step": 172800 + }, + { + "epoch": 0.98, + "learning_rate": 3.363351175643638e-05, + "loss": 0.2159, + "step": 173000 + }, + { + "epoch": 0.98, + "learning_rate": 3.3614590960779085e-05, + "loss": 0.2111, + "step": 173200 + }, + { + "epoch": 0.98, + "learning_rate": 3.359567016512179e-05, + "loss": 0.212, + "step": 173400 + }, + { + "epoch": 0.99, + "learning_rate": 3.357674936946449e-05, + "loss": 0.2129, + "step": 173600 + }, + { + "epoch": 0.99, + "learning_rate": 3.3557828573807185e-05, + "loss": 0.2121, + "step": 173800 + }, + { + "epoch": 0.99, + "learning_rate": 3.353890777814989e-05, + "loss": 0.215, + "step": 174000 + }, + { + "epoch": 0.99, + "eval_LOC_f1": 0.8006581834137779, + "eval_LOC_number": 22273, + "eval_LOC_precision": 0.7828978418500879, + "eval_LOC_recall": 0.8192430296771877, + "eval_ORG_f1": 0.6553704767375073, + "eval_ORG_number": 19008, + "eval_ORG_precision": 0.6505131128848347, + "eval_ORG_recall": 0.6603009259259259, + "eval_PER_f1": 0.8014523081325678, + "eval_PER_number": 27575, + "eval_PER_precision": 0.7906902879728966, + "eval_PER_recall": 0.8125113327289212, + "eval_loss": 0.20291385054588318, + "eval_overall_accuracy": 0.937074063484833, + "eval_overall_f1": 0.7611682988418591, + "eval_overall_precision": 0.7500035242539155, + "eval_overall_recall": 0.7726705007551993, + "eval_runtime": 72.0085, + "eval_samples_per_second": 526.147, + "eval_steps_per_second": 65.77, + "step": 174000 + }, + { + "epoch": 0.99, + "learning_rate": 3.351998698249259e-05, + "loss": 0.2111, + "step": 174200 + }, + { + "epoch": 0.99, + "learning_rate": 3.350106618683529e-05, + "loss": 0.2209, + "step": 174400 + }, + { + "epoch": 0.99, + "learning_rate": 3.3482145391177995e-05, + "loss": 0.2124, + "step": 174600 + }, + { + "epoch": 0.99, + "learning_rate": 3.346322459552069e-05, + "loss": 0.2142, + "step": 174800 + }, + { + "epoch": 0.99, + "learning_rate": 3.344430379986339e-05, + "loss": 0.227, + "step": 175000 + }, + { + "epoch": 0.99, + "learning_rate": 3.3425383004206095e-05, + "loss": 0.2181, + "step": 175200 + }, + { + "epoch": 1.0, + "learning_rate": 3.34064622085488e-05, + "loss": 0.2129, + "step": 175400 + }, + { + "epoch": 1.0, + "learning_rate": 3.33875414128915e-05, + "loss": 0.2102, + "step": 175600 + }, + { + "epoch": 1.0, + "learning_rate": 3.3368620617234195e-05, + "loss": 0.2123, + "step": 175800 + }, + { + "epoch": 1.0, + "learning_rate": 3.33496998215769e-05, + "loss": 0.2133, + "step": 176000 + }, + { + "epoch": 1.0, + "eval_LOC_f1": 0.8005965435564524, + "eval_LOC_number": 22273, + "eval_LOC_precision": 0.7825751404193285, + "eval_LOC_recall": 0.819467516724285, + "eval_ORG_f1": 0.662128712871287, + "eval_ORG_number": 19008, + "eval_ORG_precision": 0.6775881057268722, + "eval_ORG_recall": 0.6473590067340067, + "eval_PER_f1": 0.801155756207675, + "eval_PER_number": 27575, + "eval_PER_precision": 0.7979136690647483, + "eval_PER_recall": 0.8044242973708069, + "eval_loss": 0.2020001858472824, + "eval_overall_accuracy": 0.9377201866649868, + "eval_overall_f1": 0.7635642360231362, + "eval_overall_precision": 0.7612112639464227, + "eval_overall_recall": 0.7659317996979202, + "eval_runtime": 72.8012, + "eval_samples_per_second": 520.417, + "eval_steps_per_second": 65.054, + "step": 176000 + }, + { + "epoch": 1.0, + "learning_rate": 3.33307790259196e-05, + "loss": 0.2145, + "step": 176200 + }, + { + "epoch": 1.0, + "learning_rate": 3.33118582302623e-05, + "loss": 0.1955, + "step": 176400 + }, + { + "epoch": 1.0, + "learning_rate": 3.3292937434605005e-05, + "loss": 0.205, + "step": 176600 + }, + { + "epoch": 1.0, + "learning_rate": 3.32740166389477e-05, + "loss": 0.2052, + "step": 176800 + }, + { + "epoch": 1.0, + "learning_rate": 3.32550958432904e-05, + "loss": 0.2032, + "step": 177000 + }, + { + "epoch": 1.01, + "learning_rate": 3.3236175047633105e-05, + "loss": 0.2121, + "step": 177200 + }, + { + "epoch": 1.01, + "learning_rate": 3.321725425197581e-05, + "loss": 0.2008, + "step": 177400 + }, + { + "epoch": 1.01, + "learning_rate": 3.319833345631851e-05, + "loss": 0.2061, + "step": 177600 + }, + { + "epoch": 1.01, + "learning_rate": 3.3179412660661205e-05, + "loss": 0.2007, + "step": 177800 + }, + { + "epoch": 1.01, + "learning_rate": 3.316049186500391e-05, + "loss": 0.1993, + "step": 178000 + }, + { + "epoch": 1.01, + "eval_LOC_f1": 0.7991622574955908, + "eval_LOC_number": 22273, + "eval_LOC_precision": 0.7850738510850263, + "eval_LOC_recall": 0.8137655457280115, + "eval_ORG_f1": 0.6551658870291756, + "eval_ORG_number": 19008, + "eval_ORG_precision": 0.6739916550764952, + "eval_ORG_recall": 0.6373632154882155, + "eval_PER_f1": 0.802734058818239, + "eval_PER_number": 27575, + "eval_PER_precision": 0.7963597430406852, + "eval_PER_recall": 0.8092112420670897, + "eval_loss": 0.20281319320201874, + "eval_overall_accuracy": 0.9375328109427422, + "eval_overall_f1": 0.7619945192767765, + "eval_overall_precision": 0.7607480964650705, + "eval_overall_recall": 0.7632450331125827, + "eval_runtime": 72.4742, + "eval_samples_per_second": 522.765, + "eval_steps_per_second": 65.347, + "step": 178000 + }, + { + "epoch": 1.01, + "learning_rate": 3.314157106934661e-05, + "loss": 0.2126, + "step": 178200 + }, + { + "epoch": 1.01, + "learning_rate": 3.312265027368931e-05, + "loss": 0.212, + "step": 178400 + }, + { + "epoch": 1.01, + "learning_rate": 3.3103729478032014e-05, + "loss": 0.2013, + "step": 178600 + }, + { + "epoch": 1.01, + "learning_rate": 3.308480868237471e-05, + "loss": 0.2068, + "step": 178800 + }, + { + "epoch": 1.02, + "learning_rate": 3.306588788671741e-05, + "loss": 0.2016, + "step": 179000 + }, + { + "epoch": 1.02, + "learning_rate": 3.3046967091060115e-05, + "loss": 0.1926, + "step": 179200 + }, + { + "epoch": 1.02, + "learning_rate": 3.302804629540282e-05, + "loss": 0.2043, + "step": 179400 + }, + { + "epoch": 1.02, + "learning_rate": 3.300912549974552e-05, + "loss": 0.1992, + "step": 179600 + }, + { + "epoch": 1.02, + "learning_rate": 3.2990204704088215e-05, + "loss": 0.2049, + "step": 179800 + }, + { + "epoch": 1.02, + "learning_rate": 3.297128390843092e-05, + "loss": 0.2204, + "step": 180000 + }, + { + "epoch": 1.02, + "eval_LOC_f1": 0.8000087523521946, + "eval_LOC_number": 22273, + "eval_LOC_precision": 0.780272312091852, + "eval_LOC_recall": 0.8207695415974499, + "eval_ORG_f1": 0.6567659458724329, + "eval_ORG_number": 19008, + "eval_ORG_precision": 0.67990088415836, + "eval_ORG_recall": 0.6351536195286195, + "eval_PER_f1": 0.8008054026122129, + "eval_PER_number": 27575, + "eval_PER_precision": 0.78718559517971, + "eval_PER_recall": 0.8149048050770625, + "eval_loss": 0.20229148864746094, + "eval_overall_accuracy": 0.9376959570457309, + "eval_overall_f1": 0.7623315149940832, + "eval_overall_precision": 0.7575431652612861, + "eval_overall_recall": 0.7671807830835367, + "eval_runtime": 74.6353, + "eval_samples_per_second": 507.628, + "eval_steps_per_second": 63.455, + "step": 180000 + }, + { + "epoch": 1.02, + "learning_rate": 3.295236311277362e-05, + "loss": 0.2066, + "step": 180200 + }, + { + "epoch": 1.02, + "learning_rate": 3.293344231711632e-05, + "loss": 0.2024, + "step": 180400 + }, + { + "epoch": 1.03, + "learning_rate": 3.2914521521459024e-05, + "loss": 0.2052, + "step": 180600 + }, + { + "epoch": 1.03, + "learning_rate": 3.289560072580172e-05, + "loss": 0.202, + "step": 180800 + }, + { + "epoch": 1.03, + "learning_rate": 3.287667993014442e-05, + "loss": 0.2004, + "step": 181000 + }, + { + "epoch": 1.03, + "learning_rate": 3.2857759134487125e-05, + "loss": 0.2034, + "step": 181200 + }, + { + "epoch": 1.03, + "learning_rate": 3.283883833882983e-05, + "loss": 0.19, + "step": 181400 + }, + { + "epoch": 1.03, + "learning_rate": 3.281991754317253e-05, + "loss": 0.1954, + "step": 181600 + }, + { + "epoch": 1.03, + "learning_rate": 3.2800996747515225e-05, + "loss": 0.2005, + "step": 181800 + }, + { + "epoch": 1.03, + "learning_rate": 3.278207595185793e-05, + "loss": 0.2019, + "step": 182000 + }, + { + "epoch": 1.03, + "eval_LOC_f1": 0.798909161742744, + "eval_LOC_number": 22273, + "eval_LOC_precision": 0.7712494776431258, + "eval_LOC_recall": 0.8286265882458582, + "eval_ORG_f1": 0.6528567560249119, + "eval_ORG_number": 19008, + "eval_ORG_precision": 0.672636982479634, + "eval_ORG_recall": 0.6342066498316499, + "eval_PER_f1": 0.8013380813639797, + "eval_PER_number": 27575, + "eval_PER_precision": 0.7948763692153994, + "eval_PER_recall": 0.8079057116953763, + "eval_loss": 0.2030937820672989, + "eval_overall_accuracy": 0.9373955097669595, + "eval_overall_f1": 0.7610047933109886, + "eval_overall_precision": 0.7554343937377467, + "eval_overall_recall": 0.7666579528290925, + "eval_runtime": 72.4146, + "eval_samples_per_second": 523.195, + "eval_steps_per_second": 65.401, + "step": 182000 + }, + { + "epoch": 1.03, + "learning_rate": 3.276315515620063e-05, + "loss": 0.2039, + "step": 182200 + }, + { + "epoch": 1.04, + "learning_rate": 3.274423436054333e-05, + "loss": 0.2094, + "step": 182400 + }, + { + "epoch": 1.04, + "learning_rate": 3.2725313564886034e-05, + "loss": 0.2003, + "step": 182600 + }, + { + "epoch": 1.04, + "learning_rate": 3.2706392769228736e-05, + "loss": 0.2111, + "step": 182800 + }, + { + "epoch": 1.04, + "learning_rate": 3.268747197357143e-05, + "loss": 0.1909, + "step": 183000 + }, + { + "epoch": 1.04, + "learning_rate": 3.2668551177914134e-05, + "loss": 0.2088, + "step": 183200 + }, + { + "epoch": 1.04, + "learning_rate": 3.264963038225684e-05, + "loss": 0.1964, + "step": 183400 + }, + { + "epoch": 1.04, + "learning_rate": 3.263070958659954e-05, + "loss": 0.1987, + "step": 183600 + }, + { + "epoch": 1.04, + "learning_rate": 3.261178879094224e-05, + "loss": 0.2061, + "step": 183800 + }, + { + "epoch": 1.04, + "learning_rate": 3.259286799528494e-05, + "loss": 0.2146, + "step": 184000 + }, + { + "epoch": 1.04, + "eval_LOC_f1": 0.8025697066030862, + "eval_LOC_number": 22273, + "eval_LOC_precision": 0.7894805420430855, + "eval_LOC_recall": 0.8161002110178243, + "eval_ORG_f1": 0.657666649156905, + "eval_ORG_number": 19008, + "eval_ORG_precision": 0.6566663170040911, + "eval_ORG_recall": 0.6586700336700336, + "eval_PER_f1": 0.8005301524188204, + "eval_PER_number": 27575, + "eval_PER_precision": 0.7908556868851299, + "eval_PER_recall": 0.8104442429737081, + "eval_loss": 0.20245549082756042, + "eval_overall_accuracy": 0.9372339789719211, + "eval_overall_f1": 0.7621189046291773, + "eval_overall_precision": 0.7540370728378917, + "eval_overall_recall": 0.7703758568606948, + "eval_runtime": 72.3873, + "eval_samples_per_second": 523.393, + "eval_steps_per_second": 65.426, + "step": 184000 + }, + { + "epoch": 1.05, + "learning_rate": 3.257394719962764e-05, + "loss": 0.2129, + "step": 184200 + }, + { + "epoch": 1.05, + "learning_rate": 3.255502640397034e-05, + "loss": 0.2047, + "step": 184400 + }, + { + "epoch": 1.05, + "learning_rate": 3.2536105608313044e-05, + "loss": 0.214, + "step": 184600 + }, + { + "epoch": 1.05, + "learning_rate": 3.2517184812655746e-05, + "loss": 0.2049, + "step": 184800 + }, + { + "epoch": 1.05, + "learning_rate": 3.249826401699844e-05, + "loss": 0.2117, + "step": 185000 + }, + { + "epoch": 1.05, + "learning_rate": 3.2479343221341144e-05, + "loss": 0.1996, + "step": 185200 + }, + { + "epoch": 1.05, + "learning_rate": 3.2460422425683847e-05, + "loss": 0.2019, + "step": 185400 + }, + { + "epoch": 1.05, + "learning_rate": 3.244150163002655e-05, + "loss": 0.2061, + "step": 185600 + }, + { + "epoch": 1.05, + "learning_rate": 3.242258083436925e-05, + "loss": 0.2141, + "step": 185800 + }, + { + "epoch": 1.06, + "learning_rate": 3.240366003871195e-05, + "loss": 0.2067, + "step": 186000 + }, + { + "epoch": 1.06, + "eval_LOC_f1": 0.8002538404306534, + "eval_LOC_number": 22273, + "eval_LOC_precision": 0.7805763073639275, + "eval_LOC_recall": 0.8209491312351277, + "eval_ORG_f1": 0.6605775688381463, + "eval_ORG_number": 19008, + "eval_ORG_precision": 0.6749190316737114, + "eval_ORG_recall": 0.6468329124579124, + "eval_PER_f1": 0.802926759454721, + "eval_PER_number": 27575, + "eval_PER_precision": 0.7923242594357942, + "eval_PER_recall": 0.8138168631006346, + "eval_loss": 0.20096023380756378, + "eval_overall_accuracy": 0.938057786026617, + "eval_overall_f1": 0.7638757824824775, + "eval_overall_precision": 0.7578217680268706, + "eval_overall_recall": 0.7700273033577321, + "eval_runtime": 72.1632, + "eval_samples_per_second": 525.018, + "eval_steps_per_second": 65.629, + "step": 186000 + }, + { + "epoch": 1.06, + "learning_rate": 3.238473924305465e-05, + "loss": 0.2123, + "step": 186200 + }, + { + "epoch": 1.06, + "learning_rate": 3.236581844739735e-05, + "loss": 0.1924, + "step": 186400 + }, + { + "epoch": 1.06, + "learning_rate": 3.2346897651740054e-05, + "loss": 0.2061, + "step": 186600 + }, + { + "epoch": 1.06, + "learning_rate": 3.2327976856082756e-05, + "loss": 0.2002, + "step": 186800 + }, + { + "epoch": 1.06, + "learning_rate": 3.230905606042545e-05, + "loss": 0.2028, + "step": 187000 + }, + { + "epoch": 1.06, + "learning_rate": 3.2290135264768154e-05, + "loss": 0.2189, + "step": 187200 + }, + { + "epoch": 1.06, + "learning_rate": 3.2271214469110856e-05, + "loss": 0.2097, + "step": 187400 + }, + { + "epoch": 1.06, + "learning_rate": 3.225229367345356e-05, + "loss": 0.1981, + "step": 187600 + }, + { + "epoch": 1.07, + "learning_rate": 3.223337287779626e-05, + "loss": 0.2158, + "step": 187800 + }, + { + "epoch": 1.07, + "learning_rate": 3.221445208213896e-05, + "loss": 0.1901, + "step": 188000 + }, + { + "epoch": 1.07, + "eval_LOC_f1": 0.8014807758919293, + "eval_LOC_number": 22273, + "eval_LOC_precision": 0.7739035912872612, + "eval_LOC_recall": 0.8310959457639294, + "eval_ORG_f1": 0.6578533543004877, + "eval_ORG_number": 19008, + "eval_ORG_precision": 0.6666846739776349, + "eval_ORG_recall": 0.6492529461279462, + "eval_PER_f1": 0.8023383004285689, + "eval_PER_number": 27575, + "eval_PER_precision": 0.7935584562996595, + "eval_PER_recall": 0.8113145965548504, + "eval_loss": 0.20507758855819702, + "eval_overall_accuracy": 0.9379334073144374, + "eval_overall_f1": 0.7631884598287901, + "eval_overall_precision": 0.7536461725807821, + "eval_overall_recall": 0.7729754850702916, + "eval_runtime": 75.1876, + "eval_samples_per_second": 503.9, + "eval_steps_per_second": 62.989, + "step": 188000 + }, + { + "epoch": 1.07, + "learning_rate": 3.219553128648166e-05, + "loss": 0.1985, + "step": 188200 + }, + { + "epoch": 1.07, + "learning_rate": 3.217661049082436e-05, + "loss": 0.2041, + "step": 188400 + }, + { + "epoch": 1.07, + "learning_rate": 3.2157689695167064e-05, + "loss": 0.2104, + "step": 188600 + }, + { + "epoch": 1.07, + "learning_rate": 3.2138768899509766e-05, + "loss": 0.2058, + "step": 188800 + }, + { + "epoch": 1.07, + "learning_rate": 3.211984810385246e-05, + "loss": 0.1967, + "step": 189000 + }, + { + "epoch": 1.07, + "learning_rate": 3.2100927308195164e-05, + "loss": 0.2094, + "step": 189200 + }, + { + "epoch": 1.08, + "learning_rate": 3.2082006512537866e-05, + "loss": 0.2101, + "step": 189400 + }, + { + "epoch": 1.08, + "learning_rate": 3.206308571688057e-05, + "loss": 0.2154, + "step": 189600 + }, + { + "epoch": 1.08, + "learning_rate": 3.204416492122327e-05, + "loss": 0.1933, + "step": 189800 + }, + { + "epoch": 1.08, + "learning_rate": 3.2025244125565967e-05, + "loss": 0.2064, + "step": 190000 + }, + { + "epoch": 1.08, + "eval_LOC_f1": 0.8010824844337859, + "eval_LOC_number": 22273, + "eval_LOC_precision": 0.7854430925877988, + "eval_LOC_recall": 0.8173573384815697, + "eval_ORG_f1": 0.6582857142857144, + "eval_ORG_number": 19008, + "eval_ORG_precision": 0.6651984745125423, + "eval_ORG_recall": 0.6515151515151515, + "eval_PER_f1": 0.802683940967575, + "eval_PER_number": 27575, + "eval_PER_precision": 0.7901763755182348, + "eval_PER_recall": 0.8155938349954669, + "eval_loss": 0.2031039297580719, + "eval_overall_accuracy": 0.9376087304164102, + "eval_overall_f1": 0.7631062517521726, + "eval_overall_precision": 0.7554976728297537, + "eval_overall_recall": 0.770869640989892, + "eval_runtime": 75.2415, + "eval_samples_per_second": 503.539, + "eval_steps_per_second": 62.944, + "step": 190000 + }, + { + "epoch": 1.08, + "learning_rate": 3.200632332990867e-05, + "loss": 0.2153, + "step": 190200 + }, + { + "epoch": 1.08, + "learning_rate": 3.198740253425137e-05, + "loss": 0.2041, + "step": 190400 + }, + { + "epoch": 1.08, + "learning_rate": 3.1968481738594074e-05, + "loss": 0.201, + "step": 190600 + }, + { + "epoch": 1.08, + "learning_rate": 3.1949560942936776e-05, + "loss": 0.2054, + "step": 190800 + }, + { + "epoch": 1.08, + "learning_rate": 3.193064014727947e-05, + "loss": 0.2049, + "step": 191000 + }, + { + "epoch": 1.09, + "learning_rate": 3.1911719351622174e-05, + "loss": 0.2035, + "step": 191200 + }, + { + "epoch": 1.09, + "learning_rate": 3.1892798555964876e-05, + "loss": 0.2099, + "step": 191400 + }, + { + "epoch": 1.09, + "learning_rate": 3.187387776030758e-05, + "loss": 0.214, + "step": 191600 + }, + { + "epoch": 1.09, + "learning_rate": 3.185495696465028e-05, + "loss": 0.202, + "step": 191800 + }, + { + "epoch": 1.09, + "learning_rate": 3.1836036168992976e-05, + "loss": 0.2021, + "step": 192000 + }, + { + "epoch": 1.09, + "eval_LOC_f1": 0.8023245614035087, + "eval_LOC_number": 22273, + "eval_LOC_precision": 0.7841985681827924, + "eval_LOC_recall": 0.8213083105104836, + "eval_ORG_f1": 0.6608044292460511, + "eval_ORG_number": 19008, + "eval_ORG_precision": 0.6824756138580559, + "eval_ORG_recall": 0.6404671717171717, + "eval_PER_f1": 0.8027082119976705, + "eval_PER_number": 27575, + "eval_PER_precision": 0.8057291095765282, + "eval_PER_recall": 0.7997098821396192, + "eval_loss": 0.20331473648548126, + "eval_overall_accuracy": 0.9381983178183004, + "eval_overall_f1": 0.7645243467501274, + "eval_overall_precision": 0.7663203665333995, + "eval_overall_recall": 0.7627367259207621, + "eval_runtime": 94.3304, + "eval_samples_per_second": 401.641, + "eval_steps_per_second": 50.206, + "step": 192000 + }, + { + "epoch": 1.09, + "learning_rate": 3.181711537333568e-05, + "loss": 0.2011, + "step": 192200 + }, + { + "epoch": 1.09, + "learning_rate": 3.179819457767838e-05, + "loss": 0.1992, + "step": 192400 + }, + { + "epoch": 1.09, + "learning_rate": 3.1779273782021083e-05, + "loss": 0.2076, + "step": 192600 + }, + { + "epoch": 1.09, + "learning_rate": 3.1760352986363786e-05, + "loss": 0.2053, + "step": 192800 + }, + { + "epoch": 1.1, + "learning_rate": 3.174143219070648e-05, + "loss": 0.1925, + "step": 193000 + }, + { + "epoch": 1.1, + "learning_rate": 3.1722511395049184e-05, + "loss": 0.2019, + "step": 193200 + }, + { + "epoch": 1.1, + "learning_rate": 3.1703590599391886e-05, + "loss": 0.2093, + "step": 193400 + }, + { + "epoch": 1.1, + "learning_rate": 3.168466980373459e-05, + "loss": 0.2033, + "step": 193600 + }, + { + "epoch": 1.1, + "learning_rate": 3.166574900807729e-05, + "loss": 0.2126, + "step": 193800 + }, + { + "epoch": 1.1, + "learning_rate": 3.1646828212419986e-05, + "loss": 0.1974, + "step": 194000 + }, + { + "epoch": 1.1, + "eval_LOC_f1": 0.8001476687875959, + "eval_LOC_number": 22273, + "eval_LOC_precision": 0.774856998654105, + "eval_LOC_recall": 0.8271449737350155, + "eval_ORG_f1": 0.6592502622425882, + "eval_ORG_number": 19008, + "eval_ORG_precision": 0.6935184109652689, + "eval_ORG_recall": 0.6282091750841751, + "eval_PER_f1": 0.8031629077185731, + "eval_PER_number": 27575, + "eval_PER_precision": 0.7960812255076595, + "eval_PER_recall": 0.8103717135086129, + "eval_loss": 0.20409493148326874, + "eval_overall_accuracy": 0.9377444162842425, + "eval_overall_f1": 0.7643561484918794, + "eval_overall_precision": 0.7632051430557164, + "eval_overall_recall": 0.7655106308818403, + "eval_runtime": 73.8049, + "eval_samples_per_second": 513.34, + "eval_steps_per_second": 64.169, + "step": 194000 + }, + { + "epoch": 1.1, + "learning_rate": 3.1627907416762695e-05, + "loss": 0.2076, + "step": 194200 + }, + { + "epoch": 1.1, + "learning_rate": 3.160898662110539e-05, + "loss": 0.2013, + "step": 194400 + }, + { + "epoch": 1.1, + "learning_rate": 3.159006582544809e-05, + "loss": 0.2082, + "step": 194600 + }, + { + "epoch": 1.11, + "learning_rate": 3.1571145029790796e-05, + "loss": 0.1932, + "step": 194800 + }, + { + "epoch": 1.11, + "learning_rate": 3.155222423413349e-05, + "loss": 0.2012, + "step": 195000 + }, + { + "epoch": 1.11, + "learning_rate": 3.15333034384762e-05, + "loss": 0.2007, + "step": 195200 + }, + { + "epoch": 1.11, + "learning_rate": 3.1514382642818896e-05, + "loss": 0.195, + "step": 195400 + }, + { + "epoch": 1.11, + "learning_rate": 3.14954618471616e-05, + "loss": 0.2024, + "step": 195600 + }, + { + "epoch": 1.11, + "learning_rate": 3.14765410515043e-05, + "loss": 0.2111, + "step": 195800 + }, + { + "epoch": 1.11, + "learning_rate": 3.1457620255846996e-05, + "loss": 0.1971, + "step": 196000 + }, + { + "epoch": 1.11, + "eval_LOC_f1": 0.8016777565496189, + "eval_LOC_number": 22273, + "eval_LOC_precision": 0.7846028198074277, + "eval_LOC_recall": 0.8195124141337045, + "eval_ORG_f1": 0.6574344413789409, + "eval_ORG_number": 19008, + "eval_ORG_precision": 0.6696862210095498, + "eval_ORG_recall": 0.6456228956228957, + "eval_PER_f1": 0.801704993194355, + "eval_PER_number": 27575, + "eval_PER_precision": 0.791974806270125, + "eval_PER_recall": 0.8116772438803264, + "eval_loss": 0.20301629602909088, + "eval_overall_accuracy": 0.9377831836750518, + "eval_overall_f1": 0.7628653410811357, + "eval_overall_precision": 0.7574373657838225, + "eval_overall_recall": 0.7683716742186593, + "eval_runtime": 72.5166, + "eval_samples_per_second": 522.46, + "eval_steps_per_second": 65.309, + "step": 196000 + }, + { + "epoch": 1.11, + "learning_rate": 3.1438699460189705e-05, + "loss": 0.2088, + "step": 196200 + }, + { + "epoch": 1.11, + "learning_rate": 3.14197786645324e-05, + "loss": 0.2049, + "step": 196400 + }, + { + "epoch": 1.12, + "learning_rate": 3.14008578688751e-05, + "loss": 0.2074, + "step": 196600 + }, + { + "epoch": 1.12, + "learning_rate": 3.1381937073217806e-05, + "loss": 0.2073, + "step": 196800 + }, + { + "epoch": 1.12, + "learning_rate": 3.13630162775605e-05, + "loss": 0.2059, + "step": 197000 + }, + { + "epoch": 1.12, + "learning_rate": 3.134409548190321e-05, + "loss": 0.2018, + "step": 197200 + }, + { + "epoch": 1.12, + "learning_rate": 3.1325174686245906e-05, + "loss": 0.1931, + "step": 197400 + }, + { + "epoch": 1.12, + "learning_rate": 3.130625389058861e-05, + "loss": 0.2082, + "step": 197600 + }, + { + "epoch": 1.12, + "learning_rate": 3.128733309493131e-05, + "loss": 0.1954, + "step": 197800 + }, + { + "epoch": 1.12, + "learning_rate": 3.1268412299274006e-05, + "loss": 0.2006, + "step": 198000 + }, + { + "epoch": 1.12, + "eval_LOC_f1": 0.802310295036745, + "eval_LOC_number": 22273, + "eval_LOC_precision": 0.7714688328912467, + "eval_LOC_recall": 0.8357203789341355, + "eval_ORG_f1": 0.660884326675685, + "eval_ORG_number": 19008, + "eval_ORG_precision": 0.6664526828224469, + "eval_ORG_recall": 0.6554082491582491, + "eval_PER_f1": 0.8000861310987097, + "eval_PER_number": 27575, + "eval_PER_precision": 0.7918590608794488, + "eval_PER_recall": 0.8084859474161378, + "eval_loss": 0.2019522786140442, + "eval_overall_accuracy": 0.9379253307746855, + "eval_overall_f1": 0.76329283206156, + "eval_overall_precision": 0.7518985558295175, + "eval_overall_recall": 0.775037759962821, + "eval_runtime": 72.8837, + "eval_samples_per_second": 519.828, + "eval_steps_per_second": 64.98, + "step": 198000 + }, + { + "epoch": 1.13, + "learning_rate": 3.1249491503616715e-05, + "loss": 0.2003, + "step": 198200 + }, + { + "epoch": 1.13, + "learning_rate": 3.123057070795941e-05, + "loss": 0.2025, + "step": 198400 + }, + { + "epoch": 1.13, + "learning_rate": 3.121164991230211e-05, + "loss": 0.2038, + "step": 198600 + }, + { + "epoch": 1.13, + "learning_rate": 3.1192729116644815e-05, + "loss": 0.205, + "step": 198800 + }, + { + "epoch": 1.13, + "learning_rate": 3.117380832098751e-05, + "loss": 0.2047, + "step": 199000 + }, + { + "epoch": 1.13, + "learning_rate": 3.115488752533022e-05, + "loss": 0.2103, + "step": 199200 + }, + { + "epoch": 1.13, + "learning_rate": 3.1135966729672916e-05, + "loss": 0.2018, + "step": 199400 + }, + { + "epoch": 1.13, + "learning_rate": 3.111704593401562e-05, + "loss": 0.199, + "step": 199600 + }, + { + "epoch": 1.13, + "learning_rate": 3.109812513835832e-05, + "loss": 0.212, + "step": 199800 + }, + { + "epoch": 1.14, + "learning_rate": 3.1079204342701016e-05, + "loss": 0.2094, + "step": 200000 + }, + { + "epoch": 1.14, + "eval_LOC_f1": 0.8006940783693551, + "eval_LOC_number": 22273, + "eval_LOC_precision": 0.7837884325951409, + "eval_LOC_recall": 0.8183450814887981, + "eval_ORG_f1": 0.6638983878847093, + "eval_ORG_number": 19008, + "eval_ORG_precision": 0.6856710393541877, + "eval_ORG_recall": 0.6434659090909091, + "eval_PER_f1": 0.8014340772609125, + "eval_PER_number": 27575, + "eval_PER_precision": 0.7924140375753279, + "eval_PER_recall": 0.8106618313689936, + "eval_loss": 0.20304004848003387, + "eval_overall_accuracy": 0.9379592522416436, + "eval_overall_f1": 0.7645104553449286, + "eval_overall_precision": 0.7620449331197784, + "eval_overall_recall": 0.7669919832694319, + "eval_runtime": 73.6386, + "eval_samples_per_second": 514.499, + "eval_steps_per_second": 64.314, + "step": 200000 + }, + { + "epoch": 1.14, + "learning_rate": 3.1060283547043725e-05, + "loss": 0.2033, + "step": 200200 + }, + { + "epoch": 1.14, + "learning_rate": 3.104136275138642e-05, + "loss": 0.197, + "step": 200400 + }, + { + "epoch": 1.14, + "learning_rate": 3.102244195572912e-05, + "loss": 0.2087, + "step": 200600 + }, + { + "epoch": 1.14, + "learning_rate": 3.1003521160071825e-05, + "loss": 0.2134, + "step": 200800 + }, + { + "epoch": 1.14, + "learning_rate": 3.098460036441452e-05, + "loss": 0.2056, + "step": 201000 + }, + { + "epoch": 1.14, + "learning_rate": 3.096567956875723e-05, + "loss": 0.2033, + "step": 201200 + }, + { + "epoch": 1.14, + "learning_rate": 3.0946758773099925e-05, + "loss": 0.2044, + "step": 201400 + }, + { + "epoch": 1.14, + "learning_rate": 3.092783797744263e-05, + "loss": 0.2069, + "step": 201600 + }, + { + "epoch": 1.15, + "learning_rate": 3.090891718178533e-05, + "loss": 0.1999, + "step": 201800 + }, + { + "epoch": 1.15, + "learning_rate": 3.0889996386128026e-05, + "loss": 0.1983, + "step": 202000 + }, + { + "epoch": 1.15, + "eval_LOC_f1": 0.804486687007485, + "eval_LOC_number": 22273, + "eval_LOC_precision": 0.7870211303899673, + "eval_LOC_recall": 0.8227450276119068, + "eval_ORG_f1": 0.6619026113752889, + "eval_ORG_number": 19008, + "eval_ORG_precision": 0.6685269653877113, + "eval_ORG_recall": 0.6554082491582491, + "eval_PER_f1": 0.8027878069371847, + "eval_PER_number": 27575, + "eval_PER_precision": 0.7953445330296127, + "eval_PER_recall": 0.8103717135086129, + "eval_loss": 0.20248880982398987, + "eval_overall_accuracy": 0.9381094758810293, + "eval_overall_f1": 0.765156152112392, + "eval_overall_precision": 0.7588231093337142, + "eval_overall_recall": 0.7715957941210643, + "eval_runtime": 73.9045, + "eval_samples_per_second": 512.648, + "eval_steps_per_second": 64.083, + "step": 202000 + }, + { + "epoch": 1.15, + "learning_rate": 3.0871075590470735e-05, + "loss": 0.201, + "step": 202200 + }, + { + "epoch": 1.15, + "learning_rate": 3.085215479481343e-05, + "loss": 0.2044, + "step": 202400 + }, + { + "epoch": 1.15, + "learning_rate": 3.083323399915613e-05, + "loss": 0.2089, + "step": 202600 + }, + { + "epoch": 1.15, + "learning_rate": 3.0814313203498835e-05, + "loss": 0.205, + "step": 202800 + }, + { + "epoch": 1.15, + "learning_rate": 3.079539240784153e-05, + "loss": 0.2067, + "step": 203000 + }, + { + "epoch": 1.15, + "learning_rate": 3.077647161218424e-05, + "loss": 0.2124, + "step": 203200 + }, + { + "epoch": 1.15, + "learning_rate": 3.0757550816526935e-05, + "loss": 0.2105, + "step": 203400 + }, + { + "epoch": 1.16, + "learning_rate": 3.073863002086964e-05, + "loss": 0.2052, + "step": 203600 + }, + { + "epoch": 1.16, + "learning_rate": 3.071970922521234e-05, + "loss": 0.2077, + "step": 203800 + }, + { + "epoch": 1.16, + "learning_rate": 3.0700788429555036e-05, + "loss": 0.2071, + "step": 204000 + }, + { + "epoch": 1.16, + "eval_LOC_f1": 0.8012027038890351, + "eval_LOC_number": 22273, + "eval_LOC_precision": 0.7836932720793439, + "eval_LOC_recall": 0.8195124141337045, + "eval_ORG_f1": 0.6594456660422944, + "eval_ORG_number": 19008, + "eval_ORG_precision": 0.6812496494475293, + "eval_ORG_recall": 0.6389941077441077, + "eval_PER_f1": 0.8005876271831982, + "eval_PER_number": 27575, + "eval_PER_precision": 0.800776431318482, + "eval_PER_recall": 0.8003989120580236, + "eval_loss": 0.20454740524291992, + "eval_overall_accuracy": 0.9378784868441244, + "eval_overall_f1": 0.7629891375474415, + "eval_overall_precision": 0.7639556215602341, + "eval_overall_recall": 0.7620250958522133, + "eval_runtime": 71.9541, + "eval_samples_per_second": 526.544, + "eval_steps_per_second": 65.82, + "step": 204000 + }, + { + "epoch": 1.16, + "learning_rate": 3.0681867633897745e-05, + "loss": 0.2009, + "step": 204200 + }, + { + "epoch": 1.16, + "learning_rate": 3.066294683824044e-05, + "loss": 0.2037, + "step": 204400 + }, + { + "epoch": 1.16, + "learning_rate": 3.064402604258315e-05, + "loss": 0.1999, + "step": 204600 + }, + { + "epoch": 1.16, + "learning_rate": 3.0625105246925845e-05, + "loss": 0.1974, + "step": 204800 + }, + { + "epoch": 1.16, + "learning_rate": 3.060618445126854e-05, + "loss": 0.2076, + "step": 205000 + }, + { + "epoch": 1.16, + "learning_rate": 3.058726365561125e-05, + "loss": 0.1934, + "step": 205200 + }, + { + "epoch": 1.17, + "learning_rate": 3.0568342859953945e-05, + "loss": 0.2113, + "step": 205400 + }, + { + "epoch": 1.17, + "learning_rate": 3.0549422064296654e-05, + "loss": 0.2003, + "step": 205600 + }, + { + "epoch": 1.17, + "learning_rate": 3.053050126863935e-05, + "loss": 0.1993, + "step": 205800 + }, + { + "epoch": 1.17, + "learning_rate": 3.051158047298205e-05, + "loss": 0.2095, + "step": 206000 + }, + { + "epoch": 1.17, + "eval_LOC_f1": 0.8047283176593523, + "eval_LOC_number": 22273, + "eval_LOC_precision": 0.7810928453704095, + "eval_LOC_recall": 0.8298388183001841, + "eval_ORG_f1": 0.6579969867575926, + "eval_ORG_number": 19008, + "eval_ORG_precision": 0.661195219123506, + "eval_ORG_recall": 0.6548295454545454, + "eval_PER_f1": 0.80428299378391, + "eval_PER_number": 27575, + "eval_PER_precision": 0.7969523266991847, + "eval_PER_recall": 0.8117497733454215, + "eval_loss": 0.20007087290287018, + "eval_overall_accuracy": 0.9382225474375563, + "eval_overall_f1": 0.764736679791438, + "eval_overall_precision": 0.7554233085370173, + "eval_overall_recall": 0.7742825607064018, + "eval_runtime": 72.3399, + "eval_samples_per_second": 523.736, + "eval_steps_per_second": 65.469, + "step": 206000 + }, + { + "epoch": 1.17, + "learning_rate": 3.0492659677324755e-05, + "loss": 0.1982, + "step": 206200 + }, + { + "epoch": 1.17, + "learning_rate": 3.047373888166745e-05, + "loss": 0.2071, + "step": 206400 + }, + { + "epoch": 1.17, + "learning_rate": 3.0454818086010156e-05, + "loss": 0.2051, + "step": 206600 + }, + { + "epoch": 1.17, + "learning_rate": 3.0435897290352855e-05, + "loss": 0.1991, + "step": 206800 + }, + { + "epoch": 1.17, + "learning_rate": 3.0416976494695554e-05, + "loss": 0.2038, + "step": 207000 + }, + { + "epoch": 1.18, + "learning_rate": 3.039805569903826e-05, + "loss": 0.1903, + "step": 207200 + }, + { + "epoch": 1.18, + "learning_rate": 3.037913490338096e-05, + "loss": 0.2116, + "step": 207400 + }, + { + "epoch": 1.18, + "learning_rate": 3.036021410772366e-05, + "loss": 0.2, + "step": 207600 + }, + { + "epoch": 1.18, + "learning_rate": 3.034129331206636e-05, + "loss": 0.2049, + "step": 207800 + }, + { + "epoch": 1.18, + "learning_rate": 3.032237251640906e-05, + "loss": 0.1952, + "step": 208000 + }, + { + "epoch": 1.18, + "eval_LOC_f1": 0.8012256684963517, + "eval_LOC_number": 22273, + "eval_LOC_precision": 0.7870506712862712, + "eval_LOC_recall": 0.8159206213801463, + "eval_ORG_f1": 0.663743240698394, + "eval_ORG_number": 19008, + "eval_ORG_precision": 0.6791829543577603, + "eval_ORG_recall": 0.648989898989899, + "eval_PER_f1": 0.8016616613016347, + "eval_PER_number": 27575, + "eval_PER_precision": 0.795126997716895, + "eval_PER_recall": 0.8083046237533998, + "eval_loss": 0.19951093196868896, + "eval_overall_accuracy": 0.938288775063522, + "eval_overall_f1": 0.7644073808644791, + "eval_overall_precision": 0.7620408457819153, + "eval_overall_recall": 0.7667886603927037, + "eval_runtime": 73.4978, + "eval_samples_per_second": 515.485, + "eval_steps_per_second": 64.437, + "step": 208000 + }, + { + "epoch": 1.18, + "learning_rate": 3.0303451720751764e-05, + "loss": 0.2009, + "step": 208200 + }, + { + "epoch": 1.18, + "learning_rate": 3.0284530925094463e-05, + "loss": 0.2048, + "step": 208400 + }, + { + "epoch": 1.18, + "learning_rate": 3.0265610129437166e-05, + "loss": 0.2154, + "step": 208600 + }, + { + "epoch": 1.19, + "learning_rate": 3.0246689333779865e-05, + "loss": 0.1992, + "step": 208800 + }, + { + "epoch": 1.19, + "learning_rate": 3.0227768538122564e-05, + "loss": 0.2026, + "step": 209000 + }, + { + "epoch": 1.19, + "learning_rate": 3.020884774246527e-05, + "loss": 0.2031, + "step": 209200 + }, + { + "epoch": 1.19, + "learning_rate": 3.018992694680797e-05, + "loss": 0.1972, + "step": 209400 + }, + { + "epoch": 1.19, + "learning_rate": 3.017100615115067e-05, + "loss": 0.208, + "step": 209600 + }, + { + "epoch": 1.19, + "learning_rate": 3.015208535549337e-05, + "loss": 0.2088, + "step": 209800 + }, + { + "epoch": 1.19, + "learning_rate": 3.013316455983607e-05, + "loss": 0.1953, + "step": 210000 + }, + { + "epoch": 1.19, + "eval_LOC_f1": 0.8021865092380015, + "eval_LOC_number": 22273, + "eval_LOC_precision": 0.7818600289830364, + "eval_LOC_recall": 0.8235980783908768, + "eval_ORG_f1": 0.6639947609692206, + "eval_ORG_number": 19008, + "eval_ORG_precision": 0.6612406740752335, + "eval_ORG_recall": 0.6667718855218855, + "eval_PER_f1": 0.8036488439306357, + "eval_PER_number": 27575, + "eval_PER_precision": 0.8006118409213604, + "eval_PER_recall": 0.8067089755213055, + "eval_loss": 0.20208784937858582, + "eval_overall_accuracy": 0.9378849480759259, + "eval_overall_f1": 0.76488834637754, + "eval_overall_precision": 0.7564262788649985, + "eval_overall_recall": 0.773541884512606, + "eval_runtime": 73.7992, + "eval_samples_per_second": 513.379, + "eval_steps_per_second": 64.174, + "step": 210000 + }, + { + "epoch": 1.19, + "learning_rate": 3.0114243764178774e-05, + "loss": 0.1984, + "step": 210200 + }, + { + "epoch": 1.19, + "learning_rate": 3.0095322968521473e-05, + "loss": 0.2124, + "step": 210400 + }, + { + "epoch": 1.2, + "learning_rate": 3.0076402172864176e-05, + "loss": 0.197, + "step": 210600 + }, + { + "epoch": 1.2, + "learning_rate": 3.0057481377206875e-05, + "loss": 0.2031, + "step": 210800 + }, + { + "epoch": 1.2, + "learning_rate": 3.0038560581549574e-05, + "loss": 0.2029, + "step": 211000 + }, + { + "epoch": 1.2, + "learning_rate": 3.001963978589228e-05, + "loss": 0.2011, + "step": 211200 + }, + { + "epoch": 1.2, + "learning_rate": 3.0000718990234978e-05, + "loss": 0.2011, + "step": 211400 + }, + { + "epoch": 1.2, + "learning_rate": 2.998179819457768e-05, + "loss": 0.1977, + "step": 211600 + }, + { + "epoch": 1.2, + "learning_rate": 2.996287739892038e-05, + "loss": 0.2112, + "step": 211800 + }, + { + "epoch": 1.2, + "learning_rate": 2.994395660326308e-05, + "loss": 0.1969, + "step": 212000 + }, + { + "epoch": 1.2, + "eval_LOC_f1": 0.8011895390536167, + "eval_LOC_number": 22273, + "eval_LOC_precision": 0.7809369538343492, + "eval_LOC_recall": 0.8225205405648094, + "eval_ORG_f1": 0.6592677221043469, + "eval_ORG_number": 19008, + "eval_ORG_precision": 0.6825504111749465, + "eval_ORG_recall": 0.6375210437710438, + "eval_PER_f1": 0.8039523835329395, + "eval_PER_number": 27575, + "eval_PER_precision": 0.800928591995393, + "eval_PER_recall": 0.8069990933816863, + "eval_loss": 0.20179606974124908, + "eval_overall_accuracy": 0.9381789341228959, + "eval_overall_f1": 0.7644519887126141, + "eval_overall_precision": 0.7636708842413438, + "eval_overall_recall": 0.7652346926919948, + "eval_runtime": 72.786, + "eval_samples_per_second": 520.526, + "eval_steps_per_second": 65.067, + "step": 212000 + }, + { + "epoch": 1.2, + "learning_rate": 2.9925035807605784e-05, + "loss": 0.202, + "step": 212200 + }, + { + "epoch": 1.21, + "learning_rate": 2.9906115011948483e-05, + "loss": 0.2053, + "step": 212400 + }, + { + "epoch": 1.21, + "learning_rate": 2.9887194216291185e-05, + "loss": 0.2014, + "step": 212600 + }, + { + "epoch": 1.21, + "learning_rate": 2.9868273420633884e-05, + "loss": 0.1994, + "step": 212800 + }, + { + "epoch": 1.21, + "learning_rate": 2.9849352624976583e-05, + "loss": 0.2006, + "step": 213000 + }, + { + "epoch": 1.21, + "learning_rate": 2.983043182931929e-05, + "loss": 0.1975, + "step": 213200 + }, + { + "epoch": 1.21, + "learning_rate": 2.9811511033661988e-05, + "loss": 0.1982, + "step": 213400 + }, + { + "epoch": 1.21, + "learning_rate": 2.979259023800469e-05, + "loss": 0.2016, + "step": 213600 + }, + { + "epoch": 1.21, + "learning_rate": 2.977366944234739e-05, + "loss": 0.206, + "step": 213800 + }, + { + "epoch": 1.21, + "learning_rate": 2.9754748646690088e-05, + "loss": 0.2108, + "step": 214000 + }, + { + "epoch": 1.21, + "eval_LOC_f1": 0.8025583738555219, + "eval_LOC_number": 22273, + "eval_LOC_precision": 0.7835421923784269, + "eval_LOC_recall": 0.8225205405648094, + "eval_ORG_f1": 0.6581778265642152, + "eval_ORG_number": 19008, + "eval_ORG_precision": 0.6879302432308398, + "eval_ORG_recall": 0.6308922558922558, + "eval_PER_f1": 0.8009885209792088, + "eval_PER_number": 27575, + "eval_PER_precision": 0.7911979056109814, + "eval_PER_recall": 0.8110244786944696, + "eval_loss": 0.20085571706295013, + "eval_overall_accuracy": 0.9380480941789148, + "eval_overall_f1": 0.7637800413238117, + "eval_overall_precision": 0.7625472285354449, + "eval_overall_recall": 0.7650168467526431, + "eval_runtime": 73.0132, + "eval_samples_per_second": 518.906, + "eval_steps_per_second": 64.865, + "step": 214000 + }, + { + "epoch": 1.22, + "learning_rate": 2.9735827851032794e-05, + "loss": 0.202, + "step": 214200 + }, + { + "epoch": 1.22, + "learning_rate": 2.9716907055375493e-05, + "loss": 0.2051, + "step": 214400 + }, + { + "epoch": 1.22, + "learning_rate": 2.9697986259718195e-05, + "loss": 0.2065, + "step": 214600 + }, + { + "epoch": 1.22, + "learning_rate": 2.9679065464060894e-05, + "loss": 0.2036, + "step": 214800 + }, + { + "epoch": 1.22, + "learning_rate": 2.9660144668403593e-05, + "loss": 0.1926, + "step": 215000 + }, + { + "epoch": 1.22, + "learning_rate": 2.96412238727463e-05, + "loss": 0.2051, + "step": 215200 + }, + { + "epoch": 1.22, + "learning_rate": 2.9622303077088998e-05, + "loss": 0.2002, + "step": 215400 + }, + { + "epoch": 1.22, + "learning_rate": 2.96033822814317e-05, + "loss": 0.2055, + "step": 215600 + }, + { + "epoch": 1.22, + "learning_rate": 2.95844614857744e-05, + "loss": 0.1987, + "step": 215800 + }, + { + "epoch": 1.23, + "learning_rate": 2.9565540690117098e-05, + "loss": 0.2065, + "step": 216000 + }, + { + "epoch": 1.23, + "eval_LOC_f1": 0.8038981096778404, + "eval_LOC_number": 22273, + "eval_LOC_precision": 0.7797282471094608, + "eval_LOC_recall": 0.8296143312530867, + "eval_ORG_f1": 0.6647867465632711, + "eval_ORG_number": 19008, + "eval_ORG_precision": 0.685894925306328, + "eval_ORG_recall": 0.6449389730639731, + "eval_PER_f1": 0.8022857961772029, + "eval_PER_number": 27575, + "eval_PER_precision": 0.7927286887567261, + "eval_PER_recall": 0.8120761559383499, + "eval_loss": 0.1996271312236786, + "eval_overall_accuracy": 0.93849068855732, + "eval_overall_f1": 0.7662520281233098, + "eval_overall_precision": 0.7609676449104112, + "eval_overall_recall": 0.7716103171836877, + "eval_runtime": 72.2203, + "eval_samples_per_second": 524.603, + "eval_steps_per_second": 65.577, + "step": 216000 + }, + { + "epoch": 1.23, + "learning_rate": 2.9546619894459804e-05, + "loss": 0.2012, + "step": 216200 + }, + { + "epoch": 1.23, + "learning_rate": 2.9527699098802503e-05, + "loss": 0.1931, + "step": 216400 + }, + { + "epoch": 1.23, + "learning_rate": 2.950877830314521e-05, + "loss": 0.1945, + "step": 216600 + }, + { + "epoch": 1.23, + "learning_rate": 2.9489857507487904e-05, + "loss": 0.2068, + "step": 216800 + }, + { + "epoch": 1.23, + "learning_rate": 2.9470936711830603e-05, + "loss": 0.1944, + "step": 217000 + }, + { + "epoch": 1.23, + "learning_rate": 2.945201591617331e-05, + "loss": 0.1986, + "step": 217200 + }, + { + "epoch": 1.23, + "learning_rate": 2.9433095120516008e-05, + "loss": 0.2009, + "step": 217400 + }, + { + "epoch": 1.24, + "learning_rate": 2.9414174324858714e-05, + "loss": 0.1962, + "step": 217600 + }, + { + "epoch": 1.24, + "learning_rate": 2.9395253529201412e-05, + "loss": 0.2043, + "step": 217800 + }, + { + "epoch": 1.24, + "learning_rate": 2.9376332733544108e-05, + "loss": 0.1937, + "step": 218000 + }, + { + "epoch": 1.24, + "eval_LOC_f1": 0.801826760746288, + "eval_LOC_number": 22273, + "eval_LOC_precision": 0.7740816582389569, + "eval_LOC_recall": 0.8316347146769631, + "eval_ORG_f1": 0.6627225242511459, + "eval_ORG_number": 19008, + "eval_ORG_precision": 0.6715273277165694, + "eval_ORG_recall": 0.654145622895623, + "eval_PER_f1": 0.8050266839751912, + "eval_PER_number": 27575, + "eval_PER_precision": 0.800494818745742, + "eval_PER_recall": 0.8096101541251133, + "eval_loss": 0.20139047503471375, + "eval_overall_accuracy": 0.9386312203490035, + "eval_overall_f1": 0.765600977081687, + "eval_overall_precision": 0.7575568004094748, + "eval_overall_recall": 0.7738178227024515, + "eval_runtime": 73.1341, + "eval_samples_per_second": 518.048, + "eval_steps_per_second": 64.758, + "step": 218000 + }, + { + "epoch": 1.24, + "learning_rate": 2.9357411937886814e-05, + "loss": 0.2066, + "step": 218200 + }, + { + "epoch": 1.24, + "learning_rate": 2.9338491142229513e-05, + "loss": 0.1987, + "step": 218400 + }, + { + "epoch": 1.24, + "learning_rate": 2.931957034657222e-05, + "loss": 0.2051, + "step": 218600 + }, + { + "epoch": 1.24, + "learning_rate": 2.9300649550914917e-05, + "loss": 0.2096, + "step": 218800 + }, + { + "epoch": 1.24, + "learning_rate": 2.928172875525762e-05, + "loss": 0.2016, + "step": 219000 + }, + { + "epoch": 1.24, + "learning_rate": 2.926280795960032e-05, + "loss": 0.206, + "step": 219200 + }, + { + "epoch": 1.25, + "learning_rate": 2.9243887163943018e-05, + "loss": 0.2086, + "step": 219400 + }, + { + "epoch": 1.25, + "learning_rate": 2.9224966368285723e-05, + "loss": 0.2079, + "step": 219600 + }, + { + "epoch": 1.25, + "learning_rate": 2.9206045572628422e-05, + "loss": 0.1935, + "step": 219800 + }, + { + "epoch": 1.25, + "learning_rate": 2.9187124776971125e-05, + "loss": 0.2062, + "step": 220000 + }, + { + "epoch": 1.25, + "eval_LOC_f1": 0.8020759717314487, + "eval_LOC_number": 22273, + "eval_LOC_precision": 0.7892815230147346, + "eval_LOC_recall": 0.8152920576482737, + "eval_ORG_f1": 0.6618652279810388, + "eval_ORG_number": 19008, + "eval_ORG_precision": 0.6589830508474577, + "eval_ORG_recall": 0.6647727272727273, + "eval_PER_f1": 0.802739136670013, + "eval_PER_number": 27575, + "eval_PER_precision": 0.7937183168492326, + "eval_PER_recall": 0.8119673617407072, + "eval_loss": 0.20022524893283844, + "eval_overall_accuracy": 0.9380852462617736, + "eval_overall_f1": 0.7638943747441597, + "eval_overall_precision": 0.7555653421602193, + "eval_overall_recall": 0.7724090856279773, + "eval_runtime": 72.3024, + "eval_samples_per_second": 524.007, + "eval_steps_per_second": 65.503, + "step": 220000 + }, + { + "epoch": 1.25, + "learning_rate": 2.9168203981313824e-05, + "loss": 0.2047, + "step": 220200 + }, + { + "epoch": 1.25, + "learning_rate": 2.9149283185656523e-05, + "loss": 0.2011, + "step": 220400 + }, + { + "epoch": 1.25, + "learning_rate": 2.913036238999923e-05, + "loss": 0.2059, + "step": 220600 + }, + { + "epoch": 1.25, + "learning_rate": 2.9111441594341927e-05, + "loss": 0.2135, + "step": 220800 + }, + { + "epoch": 1.25, + "learning_rate": 2.909252079868463e-05, + "loss": 0.2049, + "step": 221000 + }, + { + "epoch": 1.26, + "learning_rate": 2.907360000302733e-05, + "loss": 0.2052, + "step": 221200 + }, + { + "epoch": 1.26, + "learning_rate": 2.9054679207370028e-05, + "loss": 0.2038, + "step": 221400 + }, + { + "epoch": 1.26, + "learning_rate": 2.9035758411712733e-05, + "loss": 0.1977, + "step": 221600 + }, + { + "epoch": 1.26, + "learning_rate": 2.9016837616055432e-05, + "loss": 0.2091, + "step": 221800 + }, + { + "epoch": 1.26, + "learning_rate": 2.8997916820398135e-05, + "loss": 0.2085, + "step": 222000 + }, + { + "epoch": 1.26, + "eval_LOC_f1": 0.8026663752595344, + "eval_LOC_number": 22273, + "eval_LOC_precision": 0.782003236521591, + "eval_LOC_recall": 0.8244511291698469, + "eval_ORG_f1": 0.6624083769633509, + "eval_ORG_number": 19008, + "eval_ORG_precision": 0.6592330137557315, + "eval_ORG_recall": 0.6656144781144782, + "eval_PER_f1": 0.8033339308119735, + "eval_PER_number": 27575, + "eval_PER_precision": 0.7942229310650363, + "eval_PER_recall": 0.8126563916591115, + "eval_loss": 0.20085027813911438, + "eval_overall_accuracy": 0.937960867549594, + "eval_overall_f1": 0.7645926508998533, + "eval_overall_precision": 0.7536289128073467, + "eval_overall_recall": 0.7758800975949808, + "eval_runtime": 73.1192, + "eval_samples_per_second": 518.154, + "eval_steps_per_second": 64.771, + "step": 222000 + }, + { + "epoch": 1.26, + "learning_rate": 2.8978996024740833e-05, + "loss": 0.1989, + "step": 222200 + }, + { + "epoch": 1.26, + "learning_rate": 2.8960075229083532e-05, + "loss": 0.2043, + "step": 222400 + }, + { + "epoch": 1.26, + "learning_rate": 2.8941154433426238e-05, + "loss": 0.2001, + "step": 222600 + }, + { + "epoch": 1.26, + "learning_rate": 2.8922233637768937e-05, + "loss": 0.1945, + "step": 222800 + }, + { + "epoch": 1.27, + "learning_rate": 2.890331284211164e-05, + "loss": 0.2004, + "step": 223000 + }, + { + "epoch": 1.27, + "learning_rate": 2.888439204645434e-05, + "loss": 0.2045, + "step": 223200 + }, + { + "epoch": 1.27, + "learning_rate": 2.8865471250797037e-05, + "loss": 0.1922, + "step": 223400 + }, + { + "epoch": 1.27, + "learning_rate": 2.8846550455139743e-05, + "loss": 0.2069, + "step": 223600 + }, + { + "epoch": 1.27, + "learning_rate": 2.8827629659482442e-05, + "loss": 0.2065, + "step": 223800 + }, + { + "epoch": 1.27, + "learning_rate": 2.8808708863825144e-05, + "loss": 0.2077, + "step": 224000 + }, + { + "epoch": 1.27, + "eval_LOC_f1": 0.8050212504929238, + "eval_LOC_number": 22273, + "eval_LOC_precision": 0.7860779531938562, + "eval_LOC_recall": 0.8249001032640416, + "eval_ORG_f1": 0.6587911942704617, + "eval_ORG_number": 19008, + "eval_ORG_precision": 0.6618873134724657, + "eval_ORG_recall": 0.6557239057239057, + "eval_PER_f1": 0.8033784397420761, + "eval_PER_number": 27575, + "eval_PER_precision": 0.8047671033478894, + "eval_PER_recall": 0.8019945602901178, + "eval_loss": 0.20135800540447235, + "eval_overall_accuracy": 0.9384535364744612, + "eval_overall_f1": 0.7644290457629564, + "eval_overall_precision": 0.759887492107227, + "eval_overall_recall": 0.7690252120367143, + "eval_runtime": 74.4241, + "eval_samples_per_second": 509.069, + "eval_steps_per_second": 63.635, + "step": 224000 + }, + { + "epoch": 1.27, + "learning_rate": 2.8789788068167843e-05, + "loss": 0.2016, + "step": 224200 + }, + { + "epoch": 1.27, + "learning_rate": 2.8770867272510542e-05, + "loss": 0.1966, + "step": 224400 + }, + { + "epoch": 1.27, + "learning_rate": 2.8751946476853248e-05, + "loss": 0.2028, + "step": 224600 + }, + { + "epoch": 1.28, + "learning_rate": 2.8733025681195947e-05, + "loss": 0.2045, + "step": 224800 + }, + { + "epoch": 1.28, + "learning_rate": 2.871410488553865e-05, + "loss": 0.2007, + "step": 225000 + }, + { + "epoch": 1.28, + "learning_rate": 2.8695184089881348e-05, + "loss": 0.2021, + "step": 225200 + }, + { + "epoch": 1.28, + "learning_rate": 2.8676263294224047e-05, + "loss": 0.2062, + "step": 225400 + }, + { + "epoch": 1.28, + "learning_rate": 2.8657342498566753e-05, + "loss": 0.2073, + "step": 225600 + }, + { + "epoch": 1.28, + "learning_rate": 2.8638421702909452e-05, + "loss": 0.2115, + "step": 225800 + }, + { + "epoch": 1.28, + "learning_rate": 2.8619500907252154e-05, + "loss": 0.2058, + "step": 226000 + }, + { + "epoch": 1.28, + "eval_LOC_f1": 0.8044207617919874, + "eval_LOC_number": 22273, + "eval_LOC_precision": 0.7861980282897557, + "eval_LOC_recall": 0.8235082835720379, + "eval_ORG_f1": 0.665287456445993, + "eval_ORG_number": 19008, + "eval_ORG_precision": 0.6893050541516246, + "eval_ORG_recall": 0.6428872053872053, + "eval_PER_f1": 0.8048363590564782, + "eval_PER_number": 27575, + "eval_PER_precision": 0.7986716183402371, + "eval_PER_recall": 0.8110970081595649, + "eval_loss": 0.2011905163526535, + "eval_overall_accuracy": 0.9388105195314961, + "eval_overall_f1": 0.7675285314462216, + "eval_overall_precision": 0.7663838290231966, + "eval_overall_recall": 0.7686766585337516, + "eval_runtime": 74.0431, + "eval_samples_per_second": 511.688, + "eval_steps_per_second": 63.963, + "step": 226000 + }, + { + "epoch": 1.28, + "learning_rate": 2.8600580111594853e-05, + "loss": 0.2139, + "step": 226200 + }, + { + "epoch": 1.29, + "learning_rate": 2.8581659315937552e-05, + "loss": 0.2142, + "step": 226400 + }, + { + "epoch": 1.29, + "learning_rate": 2.8562738520280258e-05, + "loss": 0.2017, + "step": 226600 + }, + { + "epoch": 1.29, + "learning_rate": 2.8543817724622957e-05, + "loss": 0.1932, + "step": 226800 + }, + { + "epoch": 1.29, + "learning_rate": 2.8524896928965663e-05, + "loss": 0.2029, + "step": 227000 + }, + { + "epoch": 1.29, + "learning_rate": 2.8505976133308358e-05, + "loss": 0.2093, + "step": 227200 + }, + { + "epoch": 1.29, + "learning_rate": 2.8487055337651057e-05, + "loss": 0.1995, + "step": 227400 + }, + { + "epoch": 1.29, + "learning_rate": 2.8468134541993763e-05, + "loss": 0.2058, + "step": 227600 + }, + { + "epoch": 1.29, + "learning_rate": 2.8449213746336462e-05, + "loss": 0.2068, + "step": 227800 + }, + { + "epoch": 1.29, + "learning_rate": 2.8430292950679168e-05, + "loss": 0.2077, + "step": 228000 + }, + { + "epoch": 1.29, + "eval_LOC_f1": 0.8032318156218862, + "eval_LOC_number": 22273, + "eval_LOC_precision": 0.7760013393043987, + "eval_LOC_recall": 0.8324428680465137, + "eval_ORG_f1": 0.6622304177022369, + "eval_ORG_number": 19008, + "eval_ORG_precision": 0.6729672478409646, + "eval_ORG_recall": 0.6518308080808081, + "eval_PER_f1": 0.8016078606520768, + "eval_PER_number": 27575, + "eval_PER_precision": 0.7899647887323944, + "eval_PER_recall": 0.8135992747053491, + "eval_loss": 0.19863562285900116, + "eval_overall_accuracy": 0.9383000822191747, + "eval_overall_f1": 0.7647750071653769, + "eval_overall_precision": 0.7547804933242815, + "eval_overall_recall": 0.775037759962821, + "eval_runtime": 70.7772, + "eval_samples_per_second": 535.299, + "eval_steps_per_second": 66.914, + "step": 228000 + }, + { + "epoch": 1.3, + "learning_rate": 2.8411372155021866e-05, + "loss": 0.204, + "step": 228200 + }, + { + "epoch": 1.3, + "learning_rate": 2.8392451359364562e-05, + "loss": 0.201, + "step": 228400 + }, + { + "epoch": 1.3, + "learning_rate": 2.8373530563707268e-05, + "loss": 0.2087, + "step": 228600 + }, + { + "epoch": 1.3, + "learning_rate": 2.8354609768049967e-05, + "loss": 0.1999, + "step": 228800 + }, + { + "epoch": 1.3, + "learning_rate": 2.8335688972392672e-05, + "loss": 0.207, + "step": 229000 + }, + { + "epoch": 1.3, + "learning_rate": 2.831676817673537e-05, + "loss": 0.2037, + "step": 229200 + }, + { + "epoch": 1.3, + "learning_rate": 2.8297847381078067e-05, + "loss": 0.2066, + "step": 229400 + }, + { + "epoch": 1.3, + "learning_rate": 2.8278926585420773e-05, + "loss": 0.1985, + "step": 229600 + }, + { + "epoch": 1.3, + "learning_rate": 2.826000578976347e-05, + "loss": 0.2044, + "step": 229800 + }, + { + "epoch": 1.31, + "learning_rate": 2.8241084994106177e-05, + "loss": 0.2054, + "step": 230000 + }, + { + "epoch": 1.31, + "eval_LOC_f1": 0.7992406886168751, + "eval_LOC_number": 22273, + "eval_LOC_precision": 0.7774429068681552, + "eval_LOC_recall": 0.8222960535177121, + "eval_ORG_f1": 0.664688898895452, + "eval_ORG_number": 19008, + "eval_ORG_precision": 0.6693518271539077, + "eval_ORG_recall": 0.6600904882154882, + "eval_PER_f1": 0.803645401822701, + "eval_PER_number": 27575, + "eval_PER_precision": 0.7982967760403621, + "eval_PER_recall": 0.8090661831368994, + "eval_loss": 0.19904367625713348, + "eval_overall_accuracy": 0.9381821647387967, + "eval_overall_f1": 0.7644817621094705, + "eval_overall_precision": 0.7568967971530249, + "eval_overall_recall": 0.7722202858138725, + "eval_runtime": 72.0549, + "eval_samples_per_second": 525.807, + "eval_steps_per_second": 65.728, + "step": 230000 + }, + { + "epoch": 1.31, + "learning_rate": 2.8222164198448876e-05, + "loss": 0.2025, + "step": 230200 + }, + { + "epoch": 1.31, + "learning_rate": 2.8203243402791575e-05, + "loss": 0.2109, + "step": 230400 + }, + { + "epoch": 1.31, + "learning_rate": 2.8184322607134278e-05, + "loss": 0.1973, + "step": 230600 + }, + { + "epoch": 1.31, + "learning_rate": 2.8165401811476977e-05, + "loss": 0.1931, + "step": 230800 + }, + { + "epoch": 1.31, + "learning_rate": 2.8146481015819682e-05, + "loss": 0.2117, + "step": 231000 + }, + { + "epoch": 1.31, + "learning_rate": 2.812756022016238e-05, + "loss": 0.207, + "step": 231200 + }, + { + "epoch": 1.31, + "learning_rate": 2.810863942450508e-05, + "loss": 0.2034, + "step": 231400 + }, + { + "epoch": 1.31, + "learning_rate": 2.8089718628847783e-05, + "loss": 0.1972, + "step": 231600 + }, + { + "epoch": 1.32, + "learning_rate": 2.807079783319048e-05, + "loss": 0.204, + "step": 231800 + }, + { + "epoch": 1.32, + "learning_rate": 2.8051877037533187e-05, + "loss": 0.205, + "step": 232000 + }, + { + "epoch": 1.32, + "eval_LOC_f1": 0.8057541335906319, + "eval_LOC_number": 22273, + "eval_LOC_precision": 0.7875176818552017, + "eval_LOC_recall": 0.8248552058546222, + "eval_ORG_f1": 0.6639531097362423, + "eval_ORG_number": 19008, + "eval_ORG_precision": 0.6856086079354405, + "eval_ORG_recall": 0.6436237373737373, + "eval_PER_f1": 0.804186105156979, + "eval_PER_number": 27575, + "eval_PER_precision": 0.7989048743826498, + "eval_PER_recall": 0.8095376246600181, + "eval_loss": 0.19997338950634003, + "eval_overall_accuracy": 0.9389898187139887, + "eval_overall_f1": 0.7672481898369947, + "eval_overall_precision": 0.7658106055125515, + "eval_overall_recall": 0.7686911815963751, + "eval_runtime": 74.5847, + "eval_samples_per_second": 507.973, + "eval_steps_per_second": 63.498, + "step": 232000 + }, + { + "epoch": 1.32, + "learning_rate": 2.8032956241875886e-05, + "loss": 0.2088, + "step": 232200 + }, + { + "epoch": 1.32, + "learning_rate": 2.8014035446218585e-05, + "loss": 0.201, + "step": 232400 + }, + { + "epoch": 1.32, + "learning_rate": 2.7995114650561287e-05, + "loss": 0.1949, + "step": 232600 + }, + { + "epoch": 1.32, + "learning_rate": 2.7976193854903986e-05, + "loss": 0.2048, + "step": 232800 + }, + { + "epoch": 1.32, + "learning_rate": 2.7957273059246692e-05, + "loss": 0.1942, + "step": 233000 + }, + { + "epoch": 1.32, + "learning_rate": 2.793835226358939e-05, + "loss": 0.201, + "step": 233200 + }, + { + "epoch": 1.32, + "learning_rate": 2.791943146793209e-05, + "loss": 0.2037, + "step": 233400 + }, + { + "epoch": 1.33, + "learning_rate": 2.7900510672274792e-05, + "loss": 0.1968, + "step": 233600 + }, + { + "epoch": 1.33, + "learning_rate": 2.788158987661749e-05, + "loss": 0.206, + "step": 233800 + }, + { + "epoch": 1.33, + "learning_rate": 2.7862669080960197e-05, + "loss": 0.2006, + "step": 234000 + }, + { + "epoch": 1.33, + "eval_LOC_f1": 0.8038841761469498, + "eval_LOC_number": 22273, + "eval_LOC_precision": 0.7853777625492547, + "eval_LOC_recall": 0.8232837965249405, + "eval_ORG_f1": 0.6629159556313994, + "eval_ORG_number": 19008, + "eval_ORG_precision": 0.6720912629757786, + "eval_ORG_recall": 0.6539877946127947, + "eval_PER_f1": 0.8033891568659332, + "eval_PER_number": 27575, + "eval_PER_precision": 0.800453596371229, + "eval_PER_recall": 0.8063463281958295, + "eval_loss": 0.20030532777309418, + "eval_overall_accuracy": 0.9386570652762096, + "eval_overall_f1": 0.7655078785077774, + "eval_overall_precision": 0.7612967165551119, + "eval_overall_recall": 0.7697658882305101, + "eval_runtime": 72.9214, + "eval_samples_per_second": 519.559, + "eval_steps_per_second": 64.947, + "step": 234000 + }, + { + "epoch": 1.33, + "learning_rate": 2.7843748285302896e-05, + "loss": 0.1956, + "step": 234200 + }, + { + "epoch": 1.33, + "learning_rate": 2.7824827489645595e-05, + "loss": 0.2005, + "step": 234400 + }, + { + "epoch": 1.33, + "learning_rate": 2.7805906693988297e-05, + "loss": 0.2023, + "step": 234600 + }, + { + "epoch": 1.33, + "learning_rate": 2.7786985898330996e-05, + "loss": 0.1981, + "step": 234800 + }, + { + "epoch": 1.33, + "learning_rate": 2.7768065102673702e-05, + "loss": 0.2058, + "step": 235000 + }, + { + "epoch": 1.34, + "learning_rate": 2.77491443070164e-05, + "loss": 0.1996, + "step": 235200 + }, + { + "epoch": 1.34, + "learning_rate": 2.77302235113591e-05, + "loss": 0.204, + "step": 235400 + }, + { + "epoch": 1.34, + "learning_rate": 2.7711302715701802e-05, + "loss": 0.2055, + "step": 235600 + }, + { + "epoch": 1.34, + "learning_rate": 2.76923819200445e-05, + "loss": 0.2019, + "step": 235800 + }, + { + "epoch": 1.34, + "learning_rate": 2.7673461124387207e-05, + "loss": 0.2034, + "step": 236000 + }, + { + "epoch": 1.34, + "eval_LOC_f1": 0.8045770402683092, + "eval_LOC_number": 22273, + "eval_LOC_precision": 0.7860875524715155, + "eval_LOC_recall": 0.8239572576662326, + "eval_ORG_f1": 0.6641657922350473, + "eval_ORG_number": 19008, + "eval_ORG_precision": 0.6623587275010465, + "eval_ORG_recall": 0.6659827441077442, + "eval_PER_f1": 0.8055365563915188, + "eval_PER_number": 27575, + "eval_PER_precision": 0.7996640817639281, + "eval_PER_recall": 0.8114959202175884, + "eval_loss": 0.1983485370874405, + "eval_overall_accuracy": 0.938831518534851, + "eval_overall_f1": 0.766534814102242, + "eval_overall_precision": 0.7579108757683736, + "eval_overall_recall": 0.7753572673405368, + "eval_runtime": 72.3361, + "eval_samples_per_second": 523.763, + "eval_steps_per_second": 65.472, + "step": 236000 + }, + { + "epoch": 1.34, + "learning_rate": 2.7654540328729906e-05, + "loss": 0.2025, + "step": 236200 + }, + { + "epoch": 1.34, + "learning_rate": 2.7635619533072605e-05, + "loss": 0.2, + "step": 236400 + }, + { + "epoch": 1.34, + "learning_rate": 2.7616698737415307e-05, + "loss": 0.1996, + "step": 236600 + }, + { + "epoch": 1.34, + "learning_rate": 2.7597777941758006e-05, + "loss": 0.1983, + "step": 236800 + }, + { + "epoch": 1.35, + "learning_rate": 2.7578857146100712e-05, + "loss": 0.2086, + "step": 237000 + }, + { + "epoch": 1.35, + "learning_rate": 2.755993635044341e-05, + "loss": 0.2033, + "step": 237200 + }, + { + "epoch": 1.35, + "learning_rate": 2.754101555478611e-05, + "loss": 0.1969, + "step": 237400 + }, + { + "epoch": 1.35, + "learning_rate": 2.7522094759128812e-05, + "loss": 0.2066, + "step": 237600 + }, + { + "epoch": 1.35, + "learning_rate": 2.750317396347151e-05, + "loss": 0.1961, + "step": 237800 + }, + { + "epoch": 1.35, + "learning_rate": 2.7484253167814217e-05, + "loss": 0.1998, + "step": 238000 + }, + { + "epoch": 1.35, + "eval_LOC_f1": 0.8047859217047664, + "eval_LOC_number": 22273, + "eval_LOC_precision": 0.7833219993199592, + "eval_LOC_recall": 0.8274592556009518, + "eval_ORG_f1": 0.6648719247274568, + "eval_ORG_number": 19008, + "eval_ORG_precision": 0.6738343508563401, + "eval_ORG_recall": 0.6561447811447811, + "eval_PER_f1": 0.8048154093097913, + "eval_PER_number": 27575, + "eval_PER_precision": 0.7918231268643622, + "eval_PER_recall": 0.8182411604714416, + "eval_loss": 0.1994849294424057, + "eval_overall_accuracy": 0.9388751318495114, + "eval_overall_f1": 0.7671392085401899, + "eval_overall_precision": 0.7580247263653377, + "eval_overall_recall": 0.7764755431625421, + "eval_runtime": 74.1272, + "eval_samples_per_second": 511.108, + "eval_steps_per_second": 63.89, + "step": 238000 + }, + { + "epoch": 1.35, + "learning_rate": 2.7465332372156916e-05, + "loss": 0.2025, + "step": 238200 + }, + { + "epoch": 1.35, + "learning_rate": 2.7446411576499615e-05, + "loss": 0.1968, + "step": 238400 + }, + { + "epoch": 1.35, + "learning_rate": 2.7427490780842317e-05, + "loss": 0.1991, + "step": 238600 + }, + { + "epoch": 1.36, + "learning_rate": 2.7408569985185016e-05, + "loss": 0.199, + "step": 238800 + }, + { + "epoch": 1.36, + "learning_rate": 2.7389649189527722e-05, + "loss": 0.1954, + "step": 239000 + }, + { + "epoch": 1.36, + "learning_rate": 2.737072839387042e-05, + "loss": 0.1895, + "step": 239200 + }, + { + "epoch": 1.36, + "learning_rate": 2.735180759821312e-05, + "loss": 0.2058, + "step": 239400 + }, + { + "epoch": 1.36, + "learning_rate": 2.7332886802555825e-05, + "loss": 0.1999, + "step": 239600 + }, + { + "epoch": 1.36, + "learning_rate": 2.731396600689852e-05, + "loss": 0.1995, + "step": 239800 + }, + { + "epoch": 1.36, + "learning_rate": 2.7295045211241227e-05, + "loss": 0.2013, + "step": 240000 + }, + { + "epoch": 1.36, + "eval_LOC_f1": 0.8021767522855899, + "eval_LOC_number": 22273, + "eval_LOC_precision": 0.7785524147547218, + "eval_LOC_recall": 0.827279665963274, + "eval_ORG_f1": 0.6660906660906661, + "eval_ORG_number": 19008, + "eval_ORG_precision": 0.6841754950357757, + "eval_ORG_recall": 0.6489372895622896, + "eval_PER_f1": 0.805549049635168, + "eval_PER_number": 27575, + "eval_PER_precision": 0.8004296455424275, + "eval_PER_recall": 0.8107343608340889, + "eval_loss": 0.19804155826568604, + "eval_overall_accuracy": 0.9388977461608168, + "eval_overall_f1": 0.7671321904651868, + "eval_overall_precision": 0.7628902996007239, + "eval_overall_recall": 0.7714215173695829, + "eval_runtime": 79.3981, + "eval_samples_per_second": 477.178, + "eval_steps_per_second": 59.649, + "step": 240000 + }, + { + "epoch": 1.36, + "learning_rate": 2.7276124415583926e-05, + "loss": 0.2045, + "step": 240200 + }, + { + "epoch": 1.36, + "learning_rate": 2.7257203619926625e-05, + "loss": 0.1976, + "step": 240400 + }, + { + "epoch": 1.37, + "learning_rate": 2.723828282426933e-05, + "loss": 0.1982, + "step": 240600 + }, + { + "epoch": 1.37, + "learning_rate": 2.721936202861203e-05, + "loss": 0.196, + "step": 240800 + }, + { + "epoch": 1.37, + "learning_rate": 2.720044123295473e-05, + "loss": 0.2062, + "step": 241000 + }, + { + "epoch": 1.37, + "learning_rate": 2.718152043729743e-05, + "loss": 0.2094, + "step": 241200 + }, + { + "epoch": 1.37, + "learning_rate": 2.716259964164013e-05, + "loss": 0.2022, + "step": 241400 + }, + { + "epoch": 1.37, + "learning_rate": 2.7143678845982835e-05, + "loss": 0.2052, + "step": 241600 + }, + { + "epoch": 1.37, + "learning_rate": 2.7124758050325534e-05, + "loss": 0.191, + "step": 241800 + }, + { + "epoch": 1.37, + "learning_rate": 2.7105837254668237e-05, + "loss": 0.1895, + "step": 242000 + }, + { + "epoch": 1.37, + "eval_LOC_f1": 0.8073145656976618, + "eval_LOC_number": 22273, + "eval_LOC_precision": 0.7871523630779731, + "eval_LOC_recall": 0.8285367934270192, + "eval_ORG_f1": 0.6669280840718375, + "eval_ORG_number": 19008, + "eval_ORG_precision": 0.66282151208106, + "eval_ORG_recall": 0.6710858585858586, + "eval_PER_f1": 0.8052574245048842, + "eval_PER_number": 27575, + "eval_PER_precision": 0.7932587432270776, + "eval_PER_recall": 0.8176246600181324, + "eval_loss": 0.19757333397865295, + "eval_overall_accuracy": 0.938818596071248, + "eval_overall_f1": 0.7681239149228032, + "eval_overall_precision": 0.7559449311639549, + "eval_overall_recall": 0.7807017543859649, + "eval_runtime": 73.2239, + "eval_samples_per_second": 517.413, + "eval_steps_per_second": 64.678, + "step": 242000 + }, + { + "epoch": 1.37, + "learning_rate": 2.7086916459010936e-05, + "loss": 0.2007, + "step": 242200 + }, + { + "epoch": 1.38, + "learning_rate": 2.7067995663353634e-05, + "loss": 0.2055, + "step": 242400 + }, + { + "epoch": 1.38, + "learning_rate": 2.704907486769634e-05, + "loss": 0.1913, + "step": 242600 + }, + { + "epoch": 1.38, + "learning_rate": 2.703015407203904e-05, + "loss": 0.2118, + "step": 242800 + }, + { + "epoch": 1.38, + "learning_rate": 2.701123327638174e-05, + "loss": 0.2052, + "step": 243000 + }, + { + "epoch": 1.38, + "learning_rate": 2.699231248072444e-05, + "loss": 0.2087, + "step": 243200 + }, + { + "epoch": 1.38, + "learning_rate": 2.697339168506714e-05, + "loss": 0.196, + "step": 243400 + }, + { + "epoch": 1.38, + "learning_rate": 2.6954470889409845e-05, + "loss": 0.2005, + "step": 243600 + }, + { + "epoch": 1.38, + "learning_rate": 2.6935550093752544e-05, + "loss": 0.1943, + "step": 243800 + }, + { + "epoch": 1.39, + "learning_rate": 2.6916629298095246e-05, + "loss": 0.2029, + "step": 244000 + }, + { + "epoch": 1.39, + "eval_LOC_f1": 0.806196787501097, + "eval_LOC_number": 22273, + "eval_LOC_precision": 0.7884458560453238, + "eval_LOC_recall": 0.8247654110357833, + "eval_ORG_f1": 0.6651159059918184, + "eval_ORG_number": 19008, + "eval_ORG_precision": 0.6762355243842766, + "eval_ORG_recall": 0.6543560606060606, + "eval_PER_f1": 0.8050779833151976, + "eval_PER_number": 27575, + "eval_PER_precision": 0.805224015962271, + "eval_PER_recall": 0.8049320036264732, + "eval_loss": 0.1990160197019577, + "eval_overall_accuracy": 0.9388961308528665, + "eval_overall_f1": 0.7675454157103242, + "eval_overall_precision": 0.7653233608155132, + "eval_overall_recall": 0.7697804112931335, + "eval_runtime": 72.363, + "eval_samples_per_second": 523.569, + "eval_steps_per_second": 65.448, + "step": 244000 + }, + { + "epoch": 1.39, + "learning_rate": 2.6897708502437945e-05, + "loss": 0.2068, + "step": 244200 + }, + { + "epoch": 1.39, + "learning_rate": 2.6878787706780644e-05, + "loss": 0.199, + "step": 244400 + }, + { + "epoch": 1.39, + "learning_rate": 2.685986691112335e-05, + "loss": 0.1973, + "step": 244600 + }, + { + "epoch": 1.39, + "learning_rate": 2.684094611546605e-05, + "loss": 0.1947, + "step": 244800 + }, + { + "epoch": 1.39, + "learning_rate": 2.682202531980875e-05, + "loss": 0.2057, + "step": 245000 + }, + { + "epoch": 1.39, + "learning_rate": 2.680310452415145e-05, + "loss": 0.1986, + "step": 245200 + }, + { + "epoch": 1.39, + "learning_rate": 2.678418372849415e-05, + "loss": 0.1999, + "step": 245400 + }, + { + "epoch": 1.39, + "learning_rate": 2.6765262932836855e-05, + "loss": 0.1962, + "step": 245600 + }, + { + "epoch": 1.4, + "learning_rate": 2.6746342137179554e-05, + "loss": 0.2101, + "step": 245800 + }, + { + "epoch": 1.4, + "learning_rate": 2.6727421341522256e-05, + "loss": 0.1992, + "step": 246000 + }, + { + "epoch": 1.4, + "eval_LOC_f1": 0.8061561495442335, + "eval_LOC_number": 22273, + "eval_LOC_precision": 0.7909699719161806, + "eval_LOC_recall": 0.8219368742423562, + "eval_ORG_f1": 0.6679738215733552, + "eval_ORG_number": 19008, + "eval_ORG_precision": 0.6728767415790317, + "eval_ORG_recall": 0.663141835016835, + "eval_PER_f1": 0.8026616240901956, + "eval_PER_number": 27575, + "eval_PER_precision": 0.789895017731119, + "eval_PER_recall": 0.8158476881233001, + "eval_loss": 0.19751431047916412, + "eval_overall_accuracy": 0.9387572143691334, + "eval_overall_f1": 0.7672880077577847, + "eval_overall_precision": 0.7590926533918901, + "eval_overall_recall": 0.7756622516556292, + "eval_runtime": 72.8556, + "eval_samples_per_second": 520.029, + "eval_steps_per_second": 65.005, + "step": 246000 + }, + { + "epoch": 1.4, + "learning_rate": 2.6708500545864955e-05, + "loss": 0.1944, + "step": 246200 + }, + { + "epoch": 1.4, + "learning_rate": 2.6689579750207654e-05, + "loss": 0.193, + "step": 246400 + }, + { + "epoch": 1.4, + "learning_rate": 2.667065895455036e-05, + "loss": 0.2034, + "step": 246600 + }, + { + "epoch": 1.4, + "learning_rate": 2.665173815889306e-05, + "loss": 0.1955, + "step": 246800 + }, + { + "epoch": 1.4, + "learning_rate": 2.663281736323576e-05, + "loss": 0.2013, + "step": 247000 + }, + { + "epoch": 1.4, + "learning_rate": 2.661389656757846e-05, + "loss": 0.2008, + "step": 247200 + }, + { + "epoch": 1.4, + "learning_rate": 2.659497577192116e-05, + "loss": 0.2053, + "step": 247400 + }, + { + "epoch": 1.41, + "learning_rate": 2.6576054976263865e-05, + "loss": 0.2006, + "step": 247600 + }, + { + "epoch": 1.41, + "learning_rate": 2.6557134180606564e-05, + "loss": 0.205, + "step": 247800 + }, + { + "epoch": 1.41, + "learning_rate": 2.6538213384949266e-05, + "loss": 0.1979, + "step": 248000 + }, + { + "epoch": 1.41, + "eval_LOC_f1": 0.8076813966175669, + "eval_LOC_number": 22273, + "eval_LOC_precision": 0.785750679347826, + "eval_LOC_recall": 0.830871458716832, + "eval_ORG_f1": 0.6672243961678587, + "eval_ORG_number": 19008, + "eval_ORG_precision": 0.6849891948800355, + "eval_ORG_recall": 0.6503577441077442, + "eval_PER_f1": 0.8032678216314026, + "eval_PER_number": 27575, + "eval_PER_precision": 0.7937823100347001, + "eval_PER_recall": 0.8129827742520399, + "eval_loss": 0.1992349922657013, + "eval_overall_accuracy": 0.9392030393634394, + "eval_overall_f1": 0.7683799937994333, + "eval_overall_precision": 0.762961584169757, + "eval_overall_recall": 0.7738759149529453, + "eval_runtime": 74.2592, + "eval_samples_per_second": 510.2, + "eval_steps_per_second": 63.777, + "step": 248000 + }, + { + "epoch": 1.41, + "learning_rate": 2.6519292589291965e-05, + "loss": 0.1999, + "step": 248200 + }, + { + "epoch": 1.41, + "learning_rate": 2.6500371793634664e-05, + "loss": 0.2017, + "step": 248400 + }, + { + "epoch": 1.41, + "learning_rate": 2.648145099797737e-05, + "loss": 0.1951, + "step": 248600 + }, + { + "epoch": 1.41, + "learning_rate": 2.646253020232007e-05, + "loss": 0.1901, + "step": 248800 + }, + { + "epoch": 1.41, + "learning_rate": 2.644360940666277e-05, + "loss": 0.1914, + "step": 249000 + }, + { + "epoch": 1.41, + "learning_rate": 2.642468861100547e-05, + "loss": 0.2051, + "step": 249200 + }, + { + "epoch": 1.42, + "learning_rate": 2.640576781534817e-05, + "loss": 0.1976, + "step": 249400 + }, + { + "epoch": 1.42, + "learning_rate": 2.6386847019690875e-05, + "loss": 0.2106, + "step": 249600 + }, + { + "epoch": 1.42, + "learning_rate": 2.6367926224033574e-05, + "loss": 0.2019, + "step": 249800 + }, + { + "epoch": 1.42, + "learning_rate": 2.634900542837628e-05, + "loss": 0.1972, + "step": 250000 + }, + { + "epoch": 1.42, + "eval_LOC_f1": 0.8056358060358237, + "eval_LOC_number": 22273, + "eval_LOC_precision": 0.7976237623762377, + "eval_LOC_recall": 0.813810443137431, + "eval_ORG_f1": 0.668167272332809, + "eval_ORG_number": 19008, + "eval_ORG_precision": 0.6895219970894436, + "eval_ORG_recall": 0.6480955387205387, + "eval_PER_f1": 0.8043517159932722, + "eval_PER_number": 27575, + "eval_PER_precision": 0.7938963653703508, + "eval_PER_recall": 0.8150861287398006, + "eval_loss": 0.19725140929222107, + "eval_overall_accuracy": 0.9391109668102675, + "eval_overall_f1": 0.7683183553768203, + "eval_overall_precision": 0.7680618849960814, + "eval_overall_recall": 0.7685749970953875, + "eval_runtime": 73.4063, + "eval_samples_per_second": 516.127, + "eval_steps_per_second": 64.518, + "step": 250000 + }, + { + "epoch": 1.42, + "learning_rate": 2.6330084632718975e-05, + "loss": 0.1997, + "step": 250200 + }, + { + "epoch": 1.42, + "learning_rate": 2.6311163837061674e-05, + "loss": 0.1971, + "step": 250400 + }, + { + "epoch": 1.42, + "learning_rate": 2.629224304140438e-05, + "loss": 0.2133, + "step": 250600 + }, + { + "epoch": 1.42, + "learning_rate": 2.627332224574708e-05, + "loss": 0.1928, + "step": 250800 + }, + { + "epoch": 1.42, + "learning_rate": 2.6254401450089784e-05, + "loss": 0.1987, + "step": 251000 + }, + { + "epoch": 1.43, + "learning_rate": 2.6235480654432483e-05, + "loss": 0.1929, + "step": 251200 + }, + { + "epoch": 1.43, + "learning_rate": 2.621655985877518e-05, + "loss": 0.2035, + "step": 251400 + }, + { + "epoch": 1.43, + "learning_rate": 2.6197639063117885e-05, + "loss": 0.2046, + "step": 251600 + }, + { + "epoch": 1.43, + "learning_rate": 2.6178718267460584e-05, + "loss": 0.1962, + "step": 251800 + }, + { + "epoch": 1.43, + "learning_rate": 2.615979747180329e-05, + "loss": 0.1949, + "step": 252000 + }, + { + "epoch": 1.43, + "eval_LOC_f1": 0.8065794949295836, + "eval_LOC_number": 22273, + "eval_LOC_precision": 0.7874187478617858, + "eval_LOC_recall": 0.8266959996408207, + "eval_ORG_f1": 0.6663247401808612, + "eval_ORG_number": 19008, + "eval_ORG_precision": 0.6842601319509897, + "eval_ORG_recall": 0.6493055555555556, + "eval_PER_f1": 0.8064341225375024, + "eval_PER_number": 27575, + "eval_PER_precision": 0.8038191316879841, + "eval_PER_recall": 0.8090661831368994, + "eval_loss": 0.19771121442317963, + "eval_overall_accuracy": 0.939167502588531, + "eval_overall_f1": 0.7688796800741855, + "eval_overall_precision": 0.7671013068116109, + "eval_overall_recall": 0.7706663181131637, + "eval_runtime": 72.0293, + "eval_samples_per_second": 525.994, + "eval_steps_per_second": 65.751, + "step": 252000 + }, + { + "epoch": 1.43, + "learning_rate": 2.6140876676145988e-05, + "loss": 0.1901, + "step": 252200 + }, + { + "epoch": 1.43, + "learning_rate": 2.6121955880488684e-05, + "loss": 0.2036, + "step": 252400 + }, + { + "epoch": 1.43, + "learning_rate": 2.610303508483139e-05, + "loss": 0.2034, + "step": 252600 + }, + { + "epoch": 1.43, + "learning_rate": 2.608411428917409e-05, + "loss": 0.193, + "step": 252800 + }, + { + "epoch": 1.44, + "learning_rate": 2.6065193493516794e-05, + "loss": 0.2076, + "step": 253000 + }, + { + "epoch": 1.44, + "learning_rate": 2.6046272697859493e-05, + "loss": 0.2146, + "step": 253200 + }, + { + "epoch": 1.44, + "learning_rate": 2.6027351902202192e-05, + "loss": 0.2002, + "step": 253400 + }, + { + "epoch": 1.44, + "learning_rate": 2.6008431106544894e-05, + "loss": 0.1984, + "step": 253600 + }, + { + "epoch": 1.44, + "learning_rate": 2.5989510310887593e-05, + "loss": 0.2003, + "step": 253800 + }, + { + "epoch": 1.44, + "learning_rate": 2.59705895152303e-05, + "loss": 0.1967, + "step": 254000 + }, + { + "epoch": 1.44, + "eval_LOC_f1": 0.8055372445616349, + "eval_LOC_number": 22273, + "eval_LOC_precision": 0.7888281619830443, + "eval_LOC_recall": 0.8229695146590041, + "eval_ORG_f1": 0.6627292340884575, + "eval_ORG_number": 19008, + "eval_ORG_precision": 0.6798915449313856, + "eval_ORG_recall": 0.6464120370370371, + "eval_PER_f1": 0.803171320234402, + "eval_PER_number": 27575, + "eval_PER_precision": 0.8036232936392681, + "eval_PER_recall": 0.8027198549410698, + "eval_loss": 0.1980707347393036, + "eval_overall_accuracy": 0.9388153654553473, + "eval_overall_f1": 0.7661372895017754, + "eval_overall_precision": 0.7661539802187268, + "eval_overall_recall": 0.766120599512025, + "eval_runtime": 74.8855, + "eval_samples_per_second": 505.933, + "eval_steps_per_second": 63.243, + "step": 254000 + }, + { + "epoch": 1.44, + "learning_rate": 2.5951668719572998e-05, + "loss": 0.1975, + "step": 254200 + }, + { + "epoch": 1.44, + "learning_rate": 2.5932747923915697e-05, + "loss": 0.1946, + "step": 254400 + }, + { + "epoch": 1.45, + "learning_rate": 2.59138271282584e-05, + "loss": 0.2081, + "step": 254600 + }, + { + "epoch": 1.45, + "learning_rate": 2.58949063326011e-05, + "loss": 0.2066, + "step": 254800 + }, + { + "epoch": 1.45, + "learning_rate": 2.5875985536943804e-05, + "loss": 0.1936, + "step": 255000 + }, + { + "epoch": 1.45, + "learning_rate": 2.5857064741286503e-05, + "loss": 0.1977, + "step": 255200 + }, + { + "epoch": 1.45, + "learning_rate": 2.5838143945629202e-05, + "loss": 0.2038, + "step": 255400 + }, + { + "epoch": 1.45, + "learning_rate": 2.5819223149971904e-05, + "loss": 0.2026, + "step": 255600 + }, + { + "epoch": 1.45, + "learning_rate": 2.5800302354314603e-05, + "loss": 0.2046, + "step": 255800 + }, + { + "epoch": 1.45, + "learning_rate": 2.578138155865731e-05, + "loss": 0.1999, + "step": 256000 + }, + { + "epoch": 1.45, + "eval_LOC_f1": 0.8048828423558185, + "eval_LOC_number": 22273, + "eval_LOC_precision": 0.783545918367347, + "eval_LOC_recall": 0.8274143581915323, + "eval_ORG_f1": 0.6659381488361928, + "eval_ORG_number": 19008, + "eval_ORG_precision": 0.6926574221413958, + "eval_ORG_recall": 0.6412037037037037, + "eval_PER_f1": 0.8066318320707299, + "eval_PER_number": 27575, + "eval_PER_precision": 0.8043196077017379, + "eval_PER_recall": 0.8089573889392566, + "eval_loss": 0.19730117917060852, + "eval_overall_accuracy": 0.9389268217039237, + "eval_overall_f1": 0.7686520558290851, + "eval_overall_precision": 0.7686855482933914, + "eval_overall_recall": 0.7686185662832579, + "eval_runtime": 73.659, + "eval_samples_per_second": 514.357, + "eval_steps_per_second": 64.296, + "step": 256000 + }, + { + "epoch": 1.45, + "learning_rate": 2.5762460763000008e-05, + "loss": 0.2028, + "step": 256200 + }, + { + "epoch": 1.46, + "learning_rate": 2.5743539967342707e-05, + "loss": 0.203, + "step": 256400 + }, + { + "epoch": 1.46, + "learning_rate": 2.572461917168541e-05, + "loss": 0.197, + "step": 256600 + }, + { + "epoch": 1.46, + "learning_rate": 2.5705698376028108e-05, + "loss": 0.211, + "step": 256800 + }, + { + "epoch": 1.46, + "learning_rate": 2.5686777580370814e-05, + "loss": 0.1949, + "step": 257000 + }, + { + "epoch": 1.46, + "learning_rate": 2.5667856784713513e-05, + "loss": 0.202, + "step": 257200 + }, + { + "epoch": 1.46, + "learning_rate": 2.5648935989056212e-05, + "loss": 0.1977, + "step": 257400 + }, + { + "epoch": 1.46, + "learning_rate": 2.5630015193398914e-05, + "loss": 0.1972, + "step": 257600 + }, + { + "epoch": 1.46, + "learning_rate": 2.5611094397741613e-05, + "loss": 0.1928, + "step": 257800 + }, + { + "epoch": 1.46, + "learning_rate": 2.559217360208432e-05, + "loss": 0.2056, + "step": 258000 + }, + { + "epoch": 1.46, + "eval_LOC_f1": 0.8072924713620412, + "eval_LOC_number": 22273, + "eval_LOC_precision": 0.7939567595728054, + "eval_LOC_recall": 0.8210838234633862, + "eval_ORG_f1": 0.6676406523933487, + "eval_ORG_number": 19008, + "eval_ORG_precision": 0.6742676252816826, + "eval_ORG_recall": 0.6611426767676768, + "eval_PER_f1": 0.8072782356229599, + "eval_PER_number": 27575, + "eval_PER_precision": 0.8028911686634622, + "eval_PER_recall": 0.811713508612874, + "eval_loss": 0.19722256064414978, + "eval_overall_accuracy": 0.9391965781316379, + "eval_overall_f1": 0.7693019088767828, + "eval_overall_precision": 0.7654636951833214, + "eval_overall_recall": 0.7731788079470199, + "eval_runtime": 72.5111, + "eval_samples_per_second": 522.499, + "eval_steps_per_second": 65.314, + "step": 258000 + }, + { + "epoch": 1.47, + "learning_rate": 2.5573252806427018e-05, + "loss": 0.1997, + "step": 258200 + }, + { + "epoch": 1.47, + "learning_rate": 2.5554332010769717e-05, + "loss": 0.1991, + "step": 258400 + }, + { + "epoch": 1.47, + "learning_rate": 2.553541121511242e-05, + "loss": 0.2132, + "step": 258600 + }, + { + "epoch": 1.47, + "learning_rate": 2.5516490419455118e-05, + "loss": 0.1943, + "step": 258800 + }, + { + "epoch": 1.47, + "learning_rate": 2.5497569623797824e-05, + "loss": 0.1998, + "step": 259000 + }, + { + "epoch": 1.47, + "learning_rate": 2.5478648828140523e-05, + "loss": 0.2027, + "step": 259200 + }, + { + "epoch": 1.47, + "learning_rate": 2.5459728032483222e-05, + "loss": 0.1944, + "step": 259400 + }, + { + "epoch": 1.47, + "learning_rate": 2.5440807236825924e-05, + "loss": 0.2064, + "step": 259600 + }, + { + "epoch": 1.47, + "learning_rate": 2.5421886441168623e-05, + "loss": 0.1971, + "step": 259800 + }, + { + "epoch": 1.48, + "learning_rate": 2.540296564551133e-05, + "loss": 0.2058, + "step": 260000 + }, + { + "epoch": 1.48, + "eval_LOC_f1": 0.8084274353003208, + "eval_LOC_number": 22273, + "eval_LOC_precision": 0.7915250591525059, + "eval_LOC_recall": 0.826067435908948, + "eval_ORG_f1": 0.6674334010912593, + "eval_ORG_number": 19008, + "eval_ORG_precision": 0.678835690968444, + "eval_ORG_recall": 0.6564078282828283, + "eval_PER_f1": 0.8047131404795639, + "eval_PER_number": 27575, + "eval_PER_precision": 0.7960190178824865, + "eval_PER_recall": 0.8135992747053491, + "eval_loss": 0.19737641513347626, + "eval_overall_accuracy": 0.9390689688035575, + "eval_overall_f1": 0.7689178956477842, + "eval_overall_precision": 0.7636694409030355, + "eval_overall_recall": 0.7742389915185315, + "eval_runtime": 73.4669, + "eval_samples_per_second": 515.701, + "eval_steps_per_second": 64.464, + "step": 260000 + }, + { + "epoch": 1.48, + "learning_rate": 2.5384044849854028e-05, + "loss": 0.197, + "step": 260200 + }, + { + "epoch": 1.48, + "learning_rate": 2.5365124054196727e-05, + "loss": 0.1924, + "step": 260400 + }, + { + "epoch": 1.48, + "learning_rate": 2.534620325853943e-05, + "loss": 0.2054, + "step": 260600 + }, + { + "epoch": 1.48, + "learning_rate": 2.5327282462882128e-05, + "loss": 0.2013, + "step": 260800 + }, + { + "epoch": 1.48, + "learning_rate": 2.5308361667224834e-05, + "loss": 0.1981, + "step": 261000 + }, + { + "epoch": 1.48, + "learning_rate": 2.5289440871567533e-05, + "loss": 0.2116, + "step": 261200 + }, + { + "epoch": 1.48, + "learning_rate": 2.527052007591023e-05, + "loss": 0.1954, + "step": 261400 + }, + { + "epoch": 1.48, + "learning_rate": 2.5251599280252934e-05, + "loss": 0.1918, + "step": 261600 + }, + { + "epoch": 1.49, + "learning_rate": 2.5232678484595633e-05, + "loss": 0.2009, + "step": 261800 + }, + { + "epoch": 1.49, + "learning_rate": 2.521375768893834e-05, + "loss": 0.203, + "step": 262000 + }, + { + "epoch": 1.49, + "eval_LOC_f1": 0.8056425645631173, + "eval_LOC_number": 22273, + "eval_LOC_precision": 0.7865697177074422, + "eval_LOC_recall": 0.8256633592241728, + "eval_ORG_f1": 0.6660738345459445, + "eval_ORG_number": 19008, + "eval_ORG_precision": 0.6827422384266932, + "eval_ORG_recall": 0.6501999158249159, + "eval_PER_f1": 0.8062823448798169, + "eval_PER_number": 27575, + "eval_PER_precision": 0.8018434832508429, + "eval_PER_recall": 0.8107706255666365, + "eval_loss": 0.1965673416852951, + "eval_overall_accuracy": 0.9389235910880229, + "eval_overall_f1": 0.7684271451309508, + "eval_overall_precision": 0.7656132864310017, + "eval_overall_recall": 0.771261763680725, + "eval_runtime": 72.0337, + "eval_samples_per_second": 525.962, + "eval_steps_per_second": 65.747, + "step": 262000 + }, + { + "epoch": 1.49, + "learning_rate": 2.5194836893281038e-05, + "loss": 0.2047, + "step": 262200 + }, + { + "epoch": 1.49, + "learning_rate": 2.5175916097623736e-05, + "loss": 0.2015, + "step": 262400 + }, + { + "epoch": 1.49, + "learning_rate": 2.5156995301966442e-05, + "loss": 0.197, + "step": 262600 + }, + { + "epoch": 1.49, + "learning_rate": 2.5138074506309138e-05, + "loss": 0.1964, + "step": 262800 + }, + { + "epoch": 1.49, + "learning_rate": 2.5119153710651844e-05, + "loss": 0.1955, + "step": 263000 + }, + { + "epoch": 1.49, + "learning_rate": 2.5100232914994542e-05, + "loss": 0.2038, + "step": 263200 + }, + { + "epoch": 1.5, + "learning_rate": 2.508131211933724e-05, + "loss": 0.1994, + "step": 263400 + }, + { + "epoch": 1.5, + "learning_rate": 2.5062391323679947e-05, + "loss": 0.2012, + "step": 263600 + }, + { + "epoch": 1.5, + "learning_rate": 2.5043470528022646e-05, + "loss": 0.2039, + "step": 263800 + }, + { + "epoch": 1.5, + "learning_rate": 2.502454973236535e-05, + "loss": 0.1931, + "step": 264000 + }, + { + "epoch": 1.5, + "eval_LOC_f1": 0.806477093206951, + "eval_LOC_number": 22273, + "eval_LOC_precision": 0.7886538213963867, + "eval_LOC_recall": 0.825124590311139, + "eval_ORG_f1": 0.6698234349919743, + "eval_ORG_number": 19008, + "eval_ORG_precision": 0.6814173742651861, + "eval_ORG_recall": 0.6586174242424242, + "eval_PER_f1": 0.8072748184889657, + "eval_PER_number": 27575, + "eval_PER_precision": 0.8001710071609248, + "eval_PER_recall": 0.814505893019039, + "eval_loss": 0.1977505385875702, + "eval_overall_accuracy": 0.9391271198897714, + "eval_overall_f1": 0.7699422799422799, + "eval_overall_precision": 0.7650407203487039, + "eval_overall_recall": 0.77490705239921, + "eval_runtime": 71.8807, + "eval_samples_per_second": 527.081, + "eval_steps_per_second": 65.887, + "step": 264000 + }, + { + "epoch": 1.5, + "learning_rate": 2.5005628936708047e-05, + "loss": 0.2081, + "step": 264200 + }, + { + "epoch": 1.5, + "learning_rate": 2.498670814105075e-05, + "loss": 0.1963, + "step": 264400 + }, + { + "epoch": 1.5, + "learning_rate": 2.4967787345393452e-05, + "loss": 0.2015, + "step": 264600 + }, + { + "epoch": 1.5, + "learning_rate": 2.494886654973615e-05, + "loss": 0.2009, + "step": 264800 + }, + { + "epoch": 1.5, + "learning_rate": 2.492994575407885e-05, + "loss": 0.2019, + "step": 265000 + }, + { + "epoch": 1.51, + "learning_rate": 2.4911024958421552e-05, + "loss": 0.201, + "step": 265200 + }, + { + "epoch": 1.51, + "learning_rate": 2.4892104162764255e-05, + "loss": 0.2073, + "step": 265400 + }, + { + "epoch": 1.51, + "learning_rate": 2.4873183367106957e-05, + "loss": 0.1994, + "step": 265600 + }, + { + "epoch": 1.51, + "learning_rate": 2.4854262571449656e-05, + "loss": 0.2031, + "step": 265800 + }, + { + "epoch": 1.51, + "learning_rate": 2.4835341775792355e-05, + "loss": 0.1957, + "step": 266000 + }, + { + "epoch": 1.51, + "eval_LOC_f1": 0.8082080544277406, + "eval_LOC_number": 22273, + "eval_LOC_precision": 0.7905289369740683, + "eval_LOC_recall": 0.8266959996408207, + "eval_ORG_f1": 0.6658888417479847, + "eval_ORG_number": 19008, + "eval_ORG_precision": 0.6713002566295979, + "eval_ORG_recall": 0.6605639730639731, + "eval_PER_f1": 0.8049375627960385, + "eval_PER_number": 27575, + "eval_PER_precision": 0.7965626220659777, + "eval_PER_recall": 0.8134904805077062, + "eval_loss": 0.19737906754016876, + "eval_overall_accuracy": 0.9392531139099014, + "eval_overall_f1": 0.7682878579701179, + "eval_overall_precision": 0.7611642459056117, + "eval_overall_recall": 0.7755460671546416, + "eval_runtime": 73.1559, + "eval_samples_per_second": 517.894, + "eval_steps_per_second": 64.738, + "step": 266000 + }, + { + "epoch": 1.51, + "learning_rate": 2.4816420980135057e-05, + "loss": 0.1969, + "step": 266200 + }, + { + "epoch": 1.51, + "learning_rate": 2.479750018447776e-05, + "loss": 0.1937, + "step": 266400 + }, + { + "epoch": 1.51, + "learning_rate": 2.4778579388820462e-05, + "loss": 0.2018, + "step": 266600 + }, + { + "epoch": 1.51, + "learning_rate": 2.475965859316316e-05, + "loss": 0.1911, + "step": 266800 + }, + { + "epoch": 1.52, + "learning_rate": 2.474073779750586e-05, + "loss": 0.1996, + "step": 267000 + }, + { + "epoch": 1.52, + "learning_rate": 2.4721817001848562e-05, + "loss": 0.1935, + "step": 267200 + }, + { + "epoch": 1.52, + "learning_rate": 2.4702896206191265e-05, + "loss": 0.1976, + "step": 267400 + }, + { + "epoch": 1.52, + "learning_rate": 2.4683975410533967e-05, + "loss": 0.2036, + "step": 267600 + }, + { + "epoch": 1.52, + "learning_rate": 2.4665054614876666e-05, + "loss": 0.2012, + "step": 267800 + }, + { + "epoch": 1.52, + "learning_rate": 2.4646133819219365e-05, + "loss": 0.1933, + "step": 268000 + }, + { + "epoch": 1.52, + "eval_LOC_f1": 0.8071643352072165, + "eval_LOC_number": 22273, + "eval_LOC_precision": 0.7841327632191694, + "eval_LOC_recall": 0.8315898172675437, + "eval_ORG_f1": 0.666934791934792, + "eval_ORG_number": 19008, + "eval_ORG_precision": 0.6800634295713036, + "eval_ORG_recall": 0.6543034511784511, + "eval_PER_f1": 0.8070601686437046, + "eval_PER_number": 27575, + "eval_PER_precision": 0.7953241083060456, + "eval_PER_recall": 0.8191477787851315, + "eval_loss": 0.19578467309474945, + "eval_overall_accuracy": 0.9393871844697832, + "eval_overall_f1": 0.7695414109768192, + "eval_overall_precision": 0.7615844118902005, + "eval_overall_recall": 0.7776664342976647, + "eval_runtime": 74.5364, + "eval_samples_per_second": 508.302, + "eval_steps_per_second": 63.539, + "step": 268000 + }, + { + "epoch": 1.52, + "learning_rate": 2.4627213023562067e-05, + "loss": 0.2054, + "step": 268200 + }, + { + "epoch": 1.52, + "learning_rate": 2.460829222790477e-05, + "loss": 0.2006, + "step": 268400 + }, + { + "epoch": 1.52, + "learning_rate": 2.4589371432247472e-05, + "loss": 0.1996, + "step": 268600 + }, + { + "epoch": 1.53, + "learning_rate": 2.457045063659017e-05, + "loss": 0.2056, + "step": 268800 + }, + { + "epoch": 1.53, + "learning_rate": 2.4551529840932873e-05, + "loss": 0.1977, + "step": 269000 + }, + { + "epoch": 1.53, + "learning_rate": 2.4532609045275572e-05, + "loss": 0.2004, + "step": 269200 + }, + { + "epoch": 1.53, + "learning_rate": 2.4513688249618274e-05, + "loss": 0.1987, + "step": 269400 + }, + { + "epoch": 1.53, + "learning_rate": 2.4494767453960977e-05, + "loss": 0.1994, + "step": 269600 + }, + { + "epoch": 1.53, + "learning_rate": 2.4475846658303676e-05, + "loss": 0.1931, + "step": 269800 + }, + { + "epoch": 1.53, + "learning_rate": 2.4456925862646378e-05, + "loss": 0.1971, + "step": 270000 + }, + { + "epoch": 1.53, + "eval_LOC_f1": 0.8093193279161437, + "eval_LOC_number": 22273, + "eval_LOC_precision": 0.7941913734981416, + "eval_LOC_recall": 0.8250347954923001, + "eval_ORG_f1": 0.6638584667228307, + "eval_ORG_number": 19008, + "eval_ORG_precision": 0.6644182124789207, + "eval_ORG_recall": 0.6632996632996633, + "eval_PER_f1": 0.8073575949367088, + "eval_PER_number": 27575, + "eval_PER_precision": 0.8006490496059342, + "eval_PER_recall": 0.8141795104261106, + "eval_loss": 0.19648274779319763, + "eval_overall_accuracy": 0.9393290333835694, + "eval_overall_f1": 0.7687880815187289, + "eval_overall_precision": 0.7616705865583351, + "eval_overall_recall": 0.7760398512838388, + "eval_runtime": 73.5802, + "eval_samples_per_second": 514.907, + "eval_steps_per_second": 64.365, + "step": 270000 + }, + { + "epoch": 1.53, + "learning_rate": 2.4438005066989077e-05, + "loss": 0.2049, + "step": 270200 + }, + { + "epoch": 1.53, + "learning_rate": 2.441908427133178e-05, + "loss": 0.1924, + "step": 270400 + }, + { + "epoch": 1.54, + "learning_rate": 2.440016347567448e-05, + "loss": 0.1951, + "step": 270600 + }, + { + "epoch": 1.54, + "learning_rate": 2.438124268001718e-05, + "loss": 0.1988, + "step": 270800 + }, + { + "epoch": 1.54, + "learning_rate": 2.4362321884359883e-05, + "loss": 0.1974, + "step": 271000 + }, + { + "epoch": 1.54, + "learning_rate": 2.4343401088702582e-05, + "loss": 0.2055, + "step": 271200 + }, + { + "epoch": 1.54, + "learning_rate": 2.4324480293045284e-05, + "loss": 0.2012, + "step": 271400 + }, + { + "epoch": 1.54, + "learning_rate": 2.4305559497387987e-05, + "loss": 0.2017, + "step": 271600 + }, + { + "epoch": 1.54, + "learning_rate": 2.4286638701730686e-05, + "loss": 0.2015, + "step": 271800 + }, + { + "epoch": 1.54, + "learning_rate": 2.4267717906073388e-05, + "loss": 0.1996, + "step": 272000 + }, + { + "epoch": 1.54, + "eval_LOC_f1": 0.8054286718200985, + "eval_LOC_number": 22273, + "eval_LOC_precision": 0.7882904182607574, + "eval_LOC_recall": 0.82332869393436, + "eval_ORG_f1": 0.666613928222978, + "eval_ORG_number": 19008, + "eval_ORG_precision": 0.6682527094898228, + "eval_ORG_recall": 0.664983164983165, + "eval_PER_f1": 0.8063059009843973, + "eval_PER_number": 27575, + "eval_PER_precision": 0.8003000857387825, + "eval_PER_recall": 0.8124025385312783, + "eval_loss": 0.19627003371715546, + "eval_overall_accuracy": 0.939096429038714, + "eval_overall_f1": 0.7679139153827342, + "eval_overall_precision": 0.7607239561065983, + "eval_overall_recall": 0.7752410828395492, + "eval_runtime": 72.7189, + "eval_samples_per_second": 521.006, + "eval_steps_per_second": 65.127, + "step": 272000 + }, + { + "epoch": 1.55, + "learning_rate": 2.4248797110416087e-05, + "loss": 0.1919, + "step": 272200 + }, + { + "epoch": 1.55, + "learning_rate": 2.422987631475879e-05, + "loss": 0.2032, + "step": 272400 + }, + { + "epoch": 1.55, + "learning_rate": 2.421095551910149e-05, + "loss": 0.2037, + "step": 272600 + }, + { + "epoch": 1.55, + "learning_rate": 2.419203472344419e-05, + "loss": 0.1942, + "step": 272800 + }, + { + "epoch": 1.55, + "learning_rate": 2.4173113927786893e-05, + "loss": 0.2052, + "step": 273000 + }, + { + "epoch": 1.55, + "learning_rate": 2.4154193132129592e-05, + "loss": 0.2056, + "step": 273200 + }, + { + "epoch": 1.55, + "learning_rate": 2.4135272336472294e-05, + "loss": 0.1967, + "step": 273400 + }, + { + "epoch": 1.55, + "learning_rate": 2.4116351540814996e-05, + "loss": 0.2089, + "step": 273600 + }, + { + "epoch": 1.55, + "learning_rate": 2.4097430745157695e-05, + "loss": 0.1979, + "step": 273800 + }, + { + "epoch": 1.56, + "learning_rate": 2.4078509949500398e-05, + "loss": 0.2059, + "step": 274000 + }, + { + "epoch": 1.56, + "eval_LOC_f1": 0.806506061071045, + "eval_LOC_number": 22273, + "eval_LOC_precision": 0.7879722436391673, + "eval_LOC_recall": 0.8259327436806896, + "eval_ORG_f1": 0.6701924649691946, + "eval_ORG_number": 19008, + "eval_ORG_precision": 0.6954848930632568, + "eval_ORG_recall": 0.6466750841750841, + "eval_PER_f1": 0.806701952650025, + "eval_PER_number": 27575, + "eval_PER_precision": 0.7956894423083707, + "eval_PER_recall": 0.818023572076156, + "eval_loss": 0.19750934839248657, + "eval_overall_accuracy": 0.9392838047609586, + "eval_overall_f1": 0.770410562488696, + "eval_overall_precision": 0.7675618792255907, + "eval_overall_recall": 0.773280469385384, + "eval_runtime": 72.1121, + "eval_samples_per_second": 525.39, + "eval_steps_per_second": 65.676, + "step": 274000 + }, + { + "epoch": 1.56, + "learning_rate": 2.40595891538431e-05, + "loss": 0.1972, + "step": 274200 + }, + { + "epoch": 1.56, + "learning_rate": 2.40406683581858e-05, + "loss": 0.1968, + "step": 274400 + }, + { + "epoch": 1.56, + "learning_rate": 2.40217475625285e-05, + "loss": 0.1986, + "step": 274600 + }, + { + "epoch": 1.56, + "learning_rate": 2.40028267668712e-05, + "loss": 0.1912, + "step": 274800 + }, + { + "epoch": 1.56, + "learning_rate": 2.3983905971213903e-05, + "loss": 0.2039, + "step": 275000 + }, + { + "epoch": 1.56, + "learning_rate": 2.3964985175556605e-05, + "loss": 0.2069, + "step": 275200 + }, + { + "epoch": 1.56, + "learning_rate": 2.3946064379899304e-05, + "loss": 0.2009, + "step": 275400 + }, + { + "epoch": 1.56, + "learning_rate": 2.3927143584242006e-05, + "loss": 0.1975, + "step": 275600 + }, + { + "epoch": 1.57, + "learning_rate": 2.3908222788584705e-05, + "loss": 0.1917, + "step": 275800 + }, + { + "epoch": 1.57, + "learning_rate": 2.3889301992927408e-05, + "loss": 0.2002, + "step": 276000 + }, + { + "epoch": 1.57, + "eval_LOC_f1": 0.8062621454616914, + "eval_LOC_number": 22273, + "eval_LOC_precision": 0.784791294737737, + "eval_LOC_recall": 0.8289408701117945, + "eval_ORG_f1": 0.6699341980557285, + "eval_ORG_number": 19008, + "eval_ORG_precision": 0.6902516600636125, + "eval_ORG_recall": 0.6507786195286195, + "eval_PER_f1": 0.8075334841466479, + "eval_PER_number": 27575, + "eval_PER_precision": 0.799665754009174, + "eval_PER_recall": 0.8155575702629193, + "eval_loss": 0.19593320786952972, + "eval_overall_accuracy": 0.9395729448840774, + "eval_overall_f1": 0.7704044037969745, + "eval_overall_precision": 0.7664510564898663, + "eval_overall_recall": 0.7743987452073894, + "eval_runtime": 74.2387, + "eval_samples_per_second": 510.34, + "eval_steps_per_second": 63.794, + "step": 276000 + }, + { + "epoch": 1.57, + "learning_rate": 2.387038119727011e-05, + "loss": 0.1945, + "step": 276200 + }, + { + "epoch": 1.57, + "learning_rate": 2.385146040161281e-05, + "loss": 0.1995, + "step": 276400 + }, + { + "epoch": 1.57, + "learning_rate": 2.383253960595551e-05, + "loss": 0.2027, + "step": 276600 + }, + { + "epoch": 1.57, + "learning_rate": 2.381361881029821e-05, + "loss": 0.1944, + "step": 276800 + }, + { + "epoch": 1.57, + "learning_rate": 2.3794698014640913e-05, + "loss": 0.1978, + "step": 277000 + }, + { + "epoch": 1.57, + "learning_rate": 2.3775777218983615e-05, + "loss": 0.1958, + "step": 277200 + }, + { + "epoch": 1.57, + "learning_rate": 2.3756856423326314e-05, + "loss": 0.201, + "step": 277400 + }, + { + "epoch": 1.58, + "learning_rate": 2.3737935627669016e-05, + "loss": 0.1907, + "step": 277600 + }, + { + "epoch": 1.58, + "learning_rate": 2.3719014832011715e-05, + "loss": 0.2033, + "step": 277800 + }, + { + "epoch": 1.58, + "learning_rate": 2.3700094036354417e-05, + "loss": 0.1946, + "step": 278000 + }, + { + "epoch": 1.58, + "eval_LOC_f1": 0.8068746188028231, + "eval_LOC_number": 22273, + "eval_LOC_precision": 0.783625978421832, + "eval_LOC_recall": 0.8315449198581242, + "eval_ORG_f1": 0.6685622317596568, + "eval_ORG_number": 19008, + "eval_ORG_precision": 0.6820271453590193, + "eval_ORG_recall": 0.6556186868686869, + "eval_PER_f1": 0.8069046120650905, + "eval_PER_number": 27575, + "eval_PER_precision": 0.8002139800285307, + "eval_PER_recall": 0.8137080689029919, + "eval_loss": 0.19666612148284912, + "eval_overall_accuracy": 0.9393209568438174, + "eval_overall_f1": 0.769738406230413, + "eval_overall_precision": 0.7637353996597424, + "eval_overall_recall": 0.7758365284071105, + "eval_runtime": 73.2319, + "eval_samples_per_second": 517.356, + "eval_steps_per_second": 64.671, + "step": 278000 + }, + { + "epoch": 1.58, + "learning_rate": 2.368117324069712e-05, + "loss": 0.2032, + "step": 278200 + }, + { + "epoch": 1.58, + "learning_rate": 2.366225244503982e-05, + "loss": 0.1971, + "step": 278400 + }, + { + "epoch": 1.58, + "learning_rate": 2.364333164938252e-05, + "loss": 0.1935, + "step": 278600 + }, + { + "epoch": 1.58, + "learning_rate": 2.362441085372522e-05, + "loss": 0.2018, + "step": 278800 + }, + { + "epoch": 1.58, + "learning_rate": 2.3605490058067922e-05, + "loss": 0.1913, + "step": 279000 + }, + { + "epoch": 1.58, + "learning_rate": 2.3586569262410625e-05, + "loss": 0.1945, + "step": 279200 + }, + { + "epoch": 1.59, + "learning_rate": 2.3567648466753324e-05, + "loss": 0.1961, + "step": 279400 + }, + { + "epoch": 1.59, + "learning_rate": 2.3548727671096026e-05, + "loss": 0.2036, + "step": 279600 + }, + { + "epoch": 1.59, + "learning_rate": 2.3529806875438725e-05, + "loss": 0.2047, + "step": 279800 + }, + { + "epoch": 1.59, + "learning_rate": 2.3510886079781427e-05, + "loss": 0.192, + "step": 280000 + }, + { + "epoch": 1.59, + "eval_LOC_f1": 0.8096374986310372, + "eval_LOC_number": 22273, + "eval_LOC_precision": 0.7904370883585664, + "eval_LOC_recall": 0.8297939208907646, + "eval_ORG_f1": 0.6699548745933466, + "eval_ORG_number": 19008, + "eval_ORG_precision": 0.668201800293071, + "eval_ORG_recall": 0.6717171717171717, + "eval_PER_f1": 0.8081476949652715, + "eval_PER_number": 27575, + "eval_PER_precision": 0.8020286438801386, + "eval_PER_recall": 0.8143608340888486, + "eval_loss": 0.19625282287597656, + "eval_overall_accuracy": 0.9395987898112835, + "eval_overall_f1": 0.7708349779324698, + "eval_overall_precision": 0.7619061130105407, + "eval_overall_recall": 0.7799756012547926, + "eval_runtime": 71.9066, + "eval_samples_per_second": 526.892, + "eval_steps_per_second": 65.863, + "step": 280000 + }, + { + "epoch": 1.59, + "learning_rate": 2.349196528412413e-05, + "loss": 0.1963, + "step": 280200 + }, + { + "epoch": 1.59, + "learning_rate": 2.3473044488466832e-05, + "loss": 0.1979, + "step": 280400 + }, + { + "epoch": 1.59, + "learning_rate": 2.345412369280953e-05, + "loss": 0.207, + "step": 280600 + }, + { + "epoch": 1.59, + "learning_rate": 2.343520289715223e-05, + "loss": 0.2002, + "step": 280800 + }, + { + "epoch": 1.6, + "learning_rate": 2.3416282101494932e-05, + "loss": 0.1924, + "step": 281000 + }, + { + "epoch": 1.6, + "learning_rate": 2.3397361305837635e-05, + "loss": 0.1864, + "step": 281200 + }, + { + "epoch": 1.6, + "learning_rate": 2.3378440510180337e-05, + "loss": 0.1972, + "step": 281400 + }, + { + "epoch": 1.6, + "learning_rate": 2.3359519714523036e-05, + "loss": 0.1969, + "step": 281600 + }, + { + "epoch": 1.6, + "learning_rate": 2.3340598918865735e-05, + "loss": 0.1956, + "step": 281800 + }, + { + "epoch": 1.6, + "learning_rate": 2.3321678123208437e-05, + "loss": 0.2049, + "step": 282000 + }, + { + "epoch": 1.6, + "eval_LOC_f1": 0.8087823652904128, + "eval_LOC_number": 22273, + "eval_LOC_precision": 0.7884033255169474, + "eval_LOC_recall": 0.8302428949849594, + "eval_ORG_f1": 0.6711070526937745, + "eval_ORG_number": 19008, + "eval_ORG_precision": 0.6855042247338966, + "eval_ORG_recall": 0.6573021885521886, + "eval_PER_f1": 0.8062567036110119, + "eval_PER_number": 27575, + "eval_PER_precision": 0.7950290851401375, + "eval_PER_recall": 0.8178059836808703, + "eval_loss": 0.19640891253948212, + "eval_overall_accuracy": 0.9394711804832032, + "eval_overall_f1": 0.7708600308130912, + "eval_overall_precision": 0.7643120235273962, + "eval_overall_recall": 0.7775212036714302, + "eval_runtime": 74.5735, + "eval_samples_per_second": 508.049, + "eval_steps_per_second": 63.508, + "step": 282000 + }, + { + "epoch": 1.6, + "learning_rate": 2.330275732755114e-05, + "loss": 0.2027, + "step": 282200 + }, + { + "epoch": 1.6, + "learning_rate": 2.3283836531893842e-05, + "loss": 0.2046, + "step": 282400 + }, + { + "epoch": 1.6, + "learning_rate": 2.326491573623654e-05, + "loss": 0.1993, + "step": 282600 + }, + { + "epoch": 1.61, + "learning_rate": 2.324599494057924e-05, + "loss": 0.1908, + "step": 282800 + }, + { + "epoch": 1.61, + "learning_rate": 2.3227074144921942e-05, + "loss": 0.1993, + "step": 283000 + }, + { + "epoch": 1.61, + "learning_rate": 2.3208153349264644e-05, + "loss": 0.1977, + "step": 283200 + }, + { + "epoch": 1.61, + "learning_rate": 2.3189232553607347e-05, + "loss": 0.1986, + "step": 283400 + }, + { + "epoch": 1.61, + "learning_rate": 2.3170311757950046e-05, + "loss": 0.198, + "step": 283600 + }, + { + "epoch": 1.61, + "learning_rate": 2.3151390962292745e-05, + "loss": 0.187, + "step": 283800 + }, + { + "epoch": 1.61, + "learning_rate": 2.3132470166635447e-05, + "loss": 0.1996, + "step": 284000 + }, + { + "epoch": 1.61, + "eval_LOC_f1": 0.808598286478077, + "eval_LOC_number": 22273, + "eval_LOC_precision": 0.7897192261599041, + "eval_LOC_recall": 0.8284021011987608, + "eval_ORG_f1": 0.6688299696609945, + "eval_ORG_number": 19008, + "eval_ORG_precision": 0.6707943059744933, + "eval_ORG_recall": 0.6668771043771043, + "eval_PER_f1": 0.8080753797649756, + "eval_PER_number": 27575, + "eval_PER_precision": 0.7984353428439944, + "eval_PER_recall": 0.8179510426110608, + "eval_loss": 0.1954508274793625, + "eval_overall_accuracy": 0.9395018713342606, + "eval_overall_f1": 0.7703744098273609, + "eval_overall_precision": 0.7613388171890512, + "eval_overall_recall": 0.7796270477518299, + "eval_runtime": 73.6103, + "eval_samples_per_second": 514.697, + "eval_steps_per_second": 64.339, + "step": 284000 + }, + { + "epoch": 1.61, + "learning_rate": 2.311354937097815e-05, + "loss": 0.1998, + "step": 284200 + }, + { + "epoch": 1.61, + "learning_rate": 2.3094628575320852e-05, + "loss": 0.2002, + "step": 284400 + }, + { + "epoch": 1.62, + "learning_rate": 2.307570777966355e-05, + "loss": 0.2029, + "step": 284600 + }, + { + "epoch": 1.62, + "learning_rate": 2.3056786984006253e-05, + "loss": 0.1988, + "step": 284800 + }, + { + "epoch": 1.62, + "learning_rate": 2.3037866188348952e-05, + "loss": 0.2007, + "step": 285000 + }, + { + "epoch": 1.62, + "learning_rate": 2.3018945392691654e-05, + "loss": 0.1941, + "step": 285200 + }, + { + "epoch": 1.62, + "learning_rate": 2.3000024597034357e-05, + "loss": 0.1964, + "step": 285400 + }, + { + "epoch": 1.62, + "learning_rate": 2.298110380137706e-05, + "loss": 0.1968, + "step": 285600 + }, + { + "epoch": 1.62, + "learning_rate": 2.2962183005719758e-05, + "loss": 0.1968, + "step": 285800 + }, + { + "epoch": 1.62, + "learning_rate": 2.2943262210062457e-05, + "loss": 0.1926, + "step": 286000 + }, + { + "epoch": 1.62, + "eval_LOC_f1": 0.8085422307625121, + "eval_LOC_number": 22273, + "eval_LOC_precision": 0.787059473706585, + "eval_LOC_recall": 0.8312306379921879, + "eval_ORG_f1": 0.6715661293350857, + "eval_ORG_number": 19008, + "eval_ORG_precision": 0.6850341997264022, + "eval_ORG_recall": 0.6586174242424242, + "eval_PER_f1": 0.8081692456134082, + "eval_PER_number": 27575, + "eval_PER_precision": 0.7965552463808954, + "eval_PER_recall": 0.8201269265639166, + "eval_loss": 0.1961415708065033, + "eval_overall_accuracy": 0.9397926267653297, + "eval_overall_f1": 0.7716638498327879, + "eval_overall_precision": 0.7643362920115687, + "eval_overall_recall": 0.7791332636226327, + "eval_runtime": 72.7823, + "eval_samples_per_second": 520.553, + "eval_steps_per_second": 65.071, + "step": 286000 + }, + { + "epoch": 1.62, + "learning_rate": 2.292434141440516e-05, + "loss": 0.1914, + "step": 286200 + }, + { + "epoch": 1.63, + "learning_rate": 2.290542061874786e-05, + "loss": 0.1945, + "step": 286400 + }, + { + "epoch": 1.63, + "learning_rate": 2.2886499823090564e-05, + "loss": 0.2001, + "step": 286600 + }, + { + "epoch": 1.63, + "learning_rate": 2.2867579027433263e-05, + "loss": 0.1972, + "step": 286800 + }, + { + "epoch": 1.63, + "learning_rate": 2.2848658231775962e-05, + "loss": 0.2029, + "step": 287000 + }, + { + "epoch": 1.63, + "learning_rate": 2.2829737436118664e-05, + "loss": 0.1886, + "step": 287200 + }, + { + "epoch": 1.63, + "learning_rate": 2.2810816640461367e-05, + "loss": 0.2027, + "step": 287400 + }, + { + "epoch": 1.63, + "learning_rate": 2.279189584480407e-05, + "loss": 0.2016, + "step": 287600 + }, + { + "epoch": 1.63, + "learning_rate": 2.2772975049146768e-05, + "loss": 0.2125, + "step": 287800 + }, + { + "epoch": 1.63, + "learning_rate": 2.2754054253489467e-05, + "loss": 0.1984, + "step": 288000 + }, + { + "epoch": 1.63, + "eval_LOC_f1": 0.8082313826897586, + "eval_LOC_number": 22273, + "eval_LOC_precision": 0.7831634801650804, + "eval_LOC_recall": 0.8349571229740044, + "eval_ORG_f1": 0.666946528834549, + "eval_ORG_number": 19008, + "eval_ORG_precision": 0.6652360515021459, + "eval_ORG_recall": 0.6686658249158249, + "eval_PER_f1": 0.8076874988836694, + "eval_PER_number": 27575, + "eval_PER_precision": 0.7957905110516683, + "eval_PER_recall": 0.8199456029011786, + "eval_loss": 0.19612962007522583, + "eval_overall_accuracy": 0.9395890979635813, + "eval_overall_f1": 0.7695832143876676, + "eval_overall_precision": 0.7565811629995509, + "eval_overall_recall": 0.7830399674683397, + "eval_runtime": 72.9115, + "eval_samples_per_second": 519.63, + "eval_steps_per_second": 64.955, + "step": 288000 + }, + { + "epoch": 1.64, + "learning_rate": 2.273513345783217e-05, + "loss": 0.2012, + "step": 288200 + }, + { + "epoch": 1.64, + "learning_rate": 2.271621266217487e-05, + "loss": 0.2054, + "step": 288400 + }, + { + "epoch": 1.64, + "learning_rate": 2.2697291866517574e-05, + "loss": 0.2002, + "step": 288600 + }, + { + "epoch": 1.64, + "learning_rate": 2.2678371070860273e-05, + "loss": 0.19, + "step": 288800 + }, + { + "epoch": 1.64, + "learning_rate": 2.2659450275202972e-05, + "loss": 0.1981, + "step": 289000 + }, + { + "epoch": 1.64, + "learning_rate": 2.2640529479545674e-05, + "loss": 0.2013, + "step": 289200 + }, + { + "epoch": 1.64, + "learning_rate": 2.2621608683888376e-05, + "loss": 0.1962, + "step": 289400 + }, + { + "epoch": 1.64, + "learning_rate": 2.260268788823108e-05, + "loss": 0.1958, + "step": 289600 + }, + { + "epoch": 1.64, + "learning_rate": 2.2583767092573778e-05, + "loss": 0.2033, + "step": 289800 + }, + { + "epoch": 1.65, + "learning_rate": 2.2564846296916477e-05, + "loss": 0.1977, + "step": 290000 + }, + { + "epoch": 1.65, + "eval_LOC_f1": 0.8102116076916323, + "eval_LOC_number": 22273, + "eval_LOC_precision": 0.7926384057037323, + "eval_LOC_recall": 0.8285816908364387, + "eval_ORG_f1": 0.6688140870778976, + "eval_ORG_number": 19008, + "eval_ORG_precision": 0.6827797873506521, + "eval_ORG_recall": 0.6554082491582491, + "eval_PER_f1": 0.8084227910817506, + "eval_PER_number": 27575, + "eval_PER_precision": 0.800433685258256, + "eval_PER_recall": 0.816572982774252, + "eval_loss": 0.19573438167572021, + "eval_overall_accuracy": 0.9398766227787496, + "eval_overall_f1": 0.7714632244650438, + "eval_overall_precision": 0.7670111972437554, + "eval_overall_recall": 0.7759672359707215, + "eval_runtime": 72.9681, + "eval_samples_per_second": 519.227, + "eval_steps_per_second": 64.905, + "step": 290000 + }, + { + "epoch": 1.65, + "learning_rate": 2.254592550125918e-05, + "loss": 0.1924, + "step": 290200 + }, + { + "epoch": 1.65, + "learning_rate": 2.252700470560188e-05, + "loss": 0.1937, + "step": 290400 + }, + { + "epoch": 1.65, + "learning_rate": 2.2508083909944584e-05, + "loss": 0.2079, + "step": 290600 + }, + { + "epoch": 1.65, + "learning_rate": 2.2489163114287286e-05, + "loss": 0.1909, + "step": 290800 + }, + { + "epoch": 1.65, + "learning_rate": 2.247024231862998e-05, + "loss": 0.1995, + "step": 291000 + }, + { + "epoch": 1.65, + "learning_rate": 2.2451321522972684e-05, + "loss": 0.1979, + "step": 291200 + }, + { + "epoch": 1.65, + "learning_rate": 2.2432400727315386e-05, + "loss": 0.2024, + "step": 291400 + }, + { + "epoch": 1.66, + "learning_rate": 2.241347993165809e-05, + "loss": 0.2016, + "step": 291600 + }, + { + "epoch": 1.66, + "learning_rate": 2.239455913600079e-05, + "loss": 0.2097, + "step": 291800 + }, + { + "epoch": 1.66, + "learning_rate": 2.237563834034349e-05, + "loss": 0.195, + "step": 292000 + }, + { + "epoch": 1.66, + "eval_LOC_f1": 0.8076644498584317, + "eval_LOC_number": 22273, + "eval_LOC_precision": 0.7900635520439712, + "eval_LOC_recall": 0.826067435908948, + "eval_ORG_f1": 0.6676122186326268, + "eval_ORG_number": 19008, + "eval_ORG_precision": 0.6666142145292421, + "eval_ORG_recall": 0.6686132154882155, + "eval_PER_f1": 0.8068052391799544, + "eval_PER_number": 27575, + "eval_PER_precision": 0.7921165740643673, + "eval_PER_recall": 0.8220489573889392, + "eval_loss": 0.19620227813720703, + "eval_overall_accuracy": 0.9394469508639475, + "eval_overall_f1": 0.7691845579505958, + "eval_overall_precision": 0.7577286177257996, + "eval_overall_recall": 0.7809922156384338, + "eval_runtime": 73.714, + "eval_samples_per_second": 513.973, + "eval_steps_per_second": 64.248, + "step": 292000 + }, + { + "epoch": 1.66, + "learning_rate": 2.235671754468619e-05, + "loss": 0.2016, + "step": 292200 + }, + { + "epoch": 1.66, + "learning_rate": 2.233779674902889e-05, + "loss": 0.2074, + "step": 292400 + }, + { + "epoch": 1.66, + "learning_rate": 2.2318875953371594e-05, + "loss": 0.1901, + "step": 292600 + }, + { + "epoch": 1.66, + "learning_rate": 2.2299955157714296e-05, + "loss": 0.2018, + "step": 292800 + }, + { + "epoch": 1.66, + "learning_rate": 2.2281034362056995e-05, + "loss": 0.2058, + "step": 293000 + }, + { + "epoch": 1.66, + "learning_rate": 2.2262113566399694e-05, + "loss": 0.197, + "step": 293200 + }, + { + "epoch": 1.67, + "learning_rate": 2.2243192770742396e-05, + "loss": 0.204, + "step": 293400 + }, + { + "epoch": 1.67, + "learning_rate": 2.22242719750851e-05, + "loss": 0.2012, + "step": 293600 + }, + { + "epoch": 1.67, + "learning_rate": 2.22053511794278e-05, + "loss": 0.1967, + "step": 293800 + }, + { + "epoch": 1.67, + "learning_rate": 2.21864303837705e-05, + "loss": 0.2059, + "step": 294000 + }, + { + "epoch": 1.67, + "eval_LOC_f1": 0.8097265504981783, + "eval_LOC_number": 22273, + "eval_LOC_precision": 0.791997595844245, + "eval_LOC_recall": 0.8282674089705024, + "eval_ORG_f1": 0.67035729567822, + "eval_ORG_number": 19008, + "eval_ORG_precision": 0.6817341166231505, + "eval_ORG_recall": 0.6593539562289562, + "eval_PER_f1": 0.8081247863555404, + "eval_PER_number": 27575, + "eval_PER_precision": 0.8018780348471866, + "eval_PER_recall": 0.8144696282864914, + "eval_loss": 0.1955600529909134, + "eval_overall_accuracy": 0.9398960064741543, + "eval_overall_f1": 0.7714683739831529, + "eval_overall_precision": 0.7668795293104685, + "eval_overall_recall": 0.776112466596956, + "eval_runtime": 74.1584, + "eval_samples_per_second": 510.893, + "eval_steps_per_second": 63.863, + "step": 294000 + }, + { + "epoch": 1.67, + "learning_rate": 2.21675095881132e-05, + "loss": 0.1986, + "step": 294200 + }, + { + "epoch": 1.67, + "learning_rate": 2.21485887924559e-05, + "loss": 0.1862, + "step": 294400 + }, + { + "epoch": 1.67, + "learning_rate": 2.2129667996798603e-05, + "loss": 0.204, + "step": 294600 + }, + { + "epoch": 1.67, + "learning_rate": 2.2110747201141306e-05, + "loss": 0.1935, + "step": 294800 + }, + { + "epoch": 1.67, + "learning_rate": 2.2091826405484005e-05, + "loss": 0.1851, + "step": 295000 + }, + { + "epoch": 1.68, + "learning_rate": 2.2072905609826704e-05, + "loss": 0.1948, + "step": 295200 + }, + { + "epoch": 1.68, + "learning_rate": 2.2053984814169406e-05, + "loss": 0.196, + "step": 295400 + }, + { + "epoch": 1.68, + "learning_rate": 2.203506401851211e-05, + "loss": 0.2061, + "step": 295600 + }, + { + "epoch": 1.68, + "learning_rate": 2.201614322285481e-05, + "loss": 0.1941, + "step": 295800 + }, + { + "epoch": 1.68, + "learning_rate": 2.199722242719751e-05, + "loss": 0.1958, + "step": 296000 + }, + { + "epoch": 1.68, + "eval_LOC_f1": 0.8102852869366298, + "eval_LOC_number": 22273, + "eval_LOC_precision": 0.7942647692971109, + "eval_LOC_recall": 0.8269653840973376, + "eval_ORG_f1": 0.6719655932710891, + "eval_ORG_number": 19008, + "eval_ORG_precision": 0.6962039596141915, + "eval_ORG_recall": 0.649358164983165, + "eval_PER_f1": 0.8082609631809373, + "eval_PER_number": 27575, + "eval_PER_precision": 0.7999076606172533, + "eval_PER_recall": 0.8167905711695376, + "eval_loss": 0.1940823197364807, + "eval_overall_accuracy": 0.9399751565637231, + "eval_overall_f1": 0.7726270916103588, + "eval_overall_precision": 0.7713967224506341, + "eval_overall_recall": 0.7738613918903219, + "eval_runtime": 73.2505, + "eval_samples_per_second": 517.225, + "eval_steps_per_second": 64.655, + "step": 296000 + }, + { + "epoch": 1.68, + "learning_rate": 2.197830163154021e-05, + "loss": 0.1973, + "step": 296200 + }, + { + "epoch": 1.68, + "learning_rate": 2.195938083588291e-05, + "loss": 0.1951, + "step": 296400 + }, + { + "epoch": 1.68, + "learning_rate": 2.1940460040225613e-05, + "loss": 0.191, + "step": 296600 + }, + { + "epoch": 1.68, + "learning_rate": 2.1921539244568316e-05, + "loss": 0.2041, + "step": 296800 + }, + { + "epoch": 1.69, + "learning_rate": 2.1902618448911015e-05, + "loss": 0.2062, + "step": 297000 + }, + { + "epoch": 1.69, + "learning_rate": 2.1883697653253714e-05, + "loss": 0.1968, + "step": 297200 + }, + { + "epoch": 1.69, + "learning_rate": 2.1864776857596416e-05, + "loss": 0.1934, + "step": 297400 + }, + { + "epoch": 1.69, + "learning_rate": 2.1845856061939118e-05, + "loss": 0.2078, + "step": 297600 + }, + { + "epoch": 1.69, + "learning_rate": 2.182693526628182e-05, + "loss": 0.1941, + "step": 297800 + }, + { + "epoch": 1.69, + "learning_rate": 2.180801447062452e-05, + "loss": 0.1963, + "step": 298000 + }, + { + "epoch": 1.69, + "eval_LOC_f1": 0.8088914170695954, + "eval_LOC_number": 22273, + "eval_LOC_precision": 0.7888538021677904, + "eval_LOC_recall": 0.8299735105284425, + "eval_ORG_f1": 0.6718605082544982, + "eval_ORG_number": 19008, + "eval_ORG_precision": 0.6768647551924822, + "eval_ORG_recall": 0.6669297138047138, + "eval_PER_f1": 0.8083798983056938, + "eval_PER_number": 27575, + "eval_PER_precision": 0.8010825439783491, + "eval_PER_recall": 0.8158114233907525, + "eval_loss": 0.19401219487190247, + "eval_overall_accuracy": 0.9400833821963989, + "eval_overall_f1": 0.7715113478695337, + "eval_overall_precision": 0.7638835504306356, + "eval_overall_recall": 0.7792930173114907, + "eval_runtime": 72.25, + "eval_samples_per_second": 524.388, + "eval_steps_per_second": 65.55, + "step": 298000 + }, + { + "epoch": 1.69, + "learning_rate": 2.1789093674967222e-05, + "loss": 0.1936, + "step": 298200 + }, + { + "epoch": 1.69, + "learning_rate": 2.177017287930992e-05, + "loss": 0.2027, + "step": 298400 + }, + { + "epoch": 1.69, + "learning_rate": 2.1751252083652623e-05, + "loss": 0.1943, + "step": 298600 + }, + { + "epoch": 1.7, + "learning_rate": 2.1732331287995325e-05, + "loss": 0.204, + "step": 298800 + }, + { + "epoch": 1.7, + "learning_rate": 2.1713410492338024e-05, + "loss": 0.1896, + "step": 299000 + }, + { + "epoch": 1.7, + "learning_rate": 2.1694489696680727e-05, + "loss": 0.2076, + "step": 299200 + }, + { + "epoch": 1.7, + "learning_rate": 2.1675568901023426e-05, + "loss": 0.2019, + "step": 299400 + }, + { + "epoch": 1.7, + "learning_rate": 2.1656648105366128e-05, + "loss": 0.1955, + "step": 299600 + }, + { + "epoch": 1.7, + "learning_rate": 2.163772730970883e-05, + "loss": 0.2053, + "step": 299800 + }, + { + "epoch": 1.7, + "learning_rate": 2.161880651405153e-05, + "loss": 0.2077, + "step": 300000 + }, + { + "epoch": 1.7, + "eval_LOC_f1": 0.8088122015097827, + "eval_LOC_number": 22273, + "eval_LOC_precision": 0.7932567777585909, + "eval_LOC_recall": 0.8249898980828806, + "eval_ORG_f1": 0.6701526211015263, + "eval_ORG_number": 19008, + "eval_ORG_precision": 0.6762736379707506, + "eval_ORG_recall": 0.6641414141414141, + "eval_PER_f1": 0.8100168030787577, + "eval_PER_number": 27575, + "eval_PER_precision": 0.8071438859282731, + "eval_PER_recall": 0.8129102447869447, + "eval_loss": 0.19428126513957977, + "eval_overall_accuracy": 0.9400753056566469, + "eval_overall_f1": 0.7715641453426646, + "eval_overall_precision": 0.7674238179388819, + "eval_overall_recall": 0.7757493900313698, + "eval_runtime": 73.4422, + "eval_samples_per_second": 515.875, + "eval_steps_per_second": 64.486, + "step": 300000 + }, + { + "epoch": 1.7, + "learning_rate": 2.1599885718394232e-05, + "loss": 0.1996, + "step": 300200 + }, + { + "epoch": 1.71, + "learning_rate": 2.158096492273693e-05, + "loss": 0.201, + "step": 300400 + }, + { + "epoch": 1.71, + "learning_rate": 2.1562044127079633e-05, + "loss": 0.2046, + "step": 300600 + }, + { + "epoch": 1.71, + "learning_rate": 2.1543123331422335e-05, + "loss": 0.1945, + "step": 300800 + }, + { + "epoch": 1.71, + "learning_rate": 2.1524202535765034e-05, + "loss": 0.2019, + "step": 301000 + }, + { + "epoch": 1.71, + "learning_rate": 2.1505281740107737e-05, + "loss": 0.1895, + "step": 301200 + }, + { + "epoch": 1.71, + "learning_rate": 2.1486360944450436e-05, + "loss": 0.2001, + "step": 301400 + }, + { + "epoch": 1.71, + "learning_rate": 2.1467440148793138e-05, + "loss": 0.19, + "step": 301600 + }, + { + "epoch": 1.71, + "learning_rate": 2.144851935313584e-05, + "loss": 0.2018, + "step": 301800 + }, + { + "epoch": 1.71, + "learning_rate": 2.142959855747854e-05, + "loss": 0.1977, + "step": 302000 + }, + { + "epoch": 1.71, + "eval_LOC_f1": 0.808436159971996, + "eval_LOC_number": 22273, + "eval_LOC_precision": 0.7883934286323875, + "eval_LOC_recall": 0.8295245364342477, + "eval_ORG_f1": 0.665055741666209, + "eval_ORG_number": 19008, + "eval_ORG_precision": 0.6955772544514647, + "eval_ORG_recall": 0.6371001683501684, + "eval_PER_f1": 0.8088842227210193, + "eval_PER_number": 27575, + "eval_PER_precision": 0.8075102099822906, + "eval_PER_recall": 0.81026291931097, + "eval_loss": 0.1959424614906311, + "eval_overall_accuracy": 0.9396181735066882, + "eval_overall_f1": 0.7706049355754533, + "eval_overall_precision": 0.7725282424030125, + "eval_overall_recall": 0.7686911815963751, + "eval_runtime": 71.7605, + "eval_samples_per_second": 527.965, + "eval_steps_per_second": 65.997, + "step": 302000 + }, + { + "epoch": 1.72, + "learning_rate": 2.141067776182124e-05, + "loss": 0.1854, + "step": 302200 + }, + { + "epoch": 1.72, + "learning_rate": 2.139175696616394e-05, + "loss": 0.2011, + "step": 302400 + }, + { + "epoch": 1.72, + "learning_rate": 2.1372836170506643e-05, + "loss": 0.1972, + "step": 302600 + }, + { + "epoch": 1.72, + "learning_rate": 2.1353915374849345e-05, + "loss": 0.2041, + "step": 302800 + }, + { + "epoch": 1.72, + "learning_rate": 2.1334994579192044e-05, + "loss": 0.1923, + "step": 303000 + }, + { + "epoch": 1.72, + "learning_rate": 2.1316073783534747e-05, + "loss": 0.1871, + "step": 303200 + }, + { + "epoch": 1.72, + "learning_rate": 2.129715298787745e-05, + "loss": 0.1983, + "step": 303400 + }, + { + "epoch": 1.72, + "learning_rate": 2.1278232192220148e-05, + "loss": 0.1939, + "step": 303600 + }, + { + "epoch": 1.72, + "learning_rate": 2.125931139656285e-05, + "loss": 0.2013, + "step": 303800 + }, + { + "epoch": 1.73, + "learning_rate": 2.124039060090555e-05, + "loss": 0.1943, + "step": 304000 + }, + { + "epoch": 1.73, + "eval_LOC_f1": 0.8091804698495182, + "eval_LOC_number": 22273, + "eval_LOC_precision": 0.7814486450317832, + "eval_LOC_recall": 0.8389529924123378, + "eval_ORG_f1": 0.6719865374698043, + "eval_ORG_number": 19008, + "eval_ORG_precision": 0.6940846649845809, + "eval_ORG_recall": 0.6512521043771043, + "eval_PER_f1": 0.8084831289169824, + "eval_PER_number": 27575, + "eval_PER_precision": 0.8007612407512806, + "eval_PER_recall": 0.8163553943789664, + "eval_loss": 0.1950448900461197, + "eval_overall_accuracy": 0.9400494607294407, + "eval_overall_f1": 0.7724615218253252, + "eval_overall_precision": 0.7669162169512876, + "eval_overall_recall": 0.7780876031137446, + "eval_runtime": 73.7076, + "eval_samples_per_second": 514.017, + "eval_steps_per_second": 64.254, + "step": 304000 + }, + { + "epoch": 1.73, + "learning_rate": 2.122146980524825e-05, + "loss": 0.1945, + "step": 304200 + }, + { + "epoch": 1.73, + "learning_rate": 2.1202549009590954e-05, + "loss": 0.195, + "step": 304400 + }, + { + "epoch": 1.73, + "learning_rate": 2.1183628213933653e-05, + "loss": 0.1968, + "step": 304600 + }, + { + "epoch": 1.73, + "learning_rate": 2.1164707418276355e-05, + "loss": 0.203, + "step": 304800 + }, + { + "epoch": 1.73, + "learning_rate": 2.1145786622619054e-05, + "loss": 0.1963, + "step": 305000 + }, + { + "epoch": 1.73, + "learning_rate": 2.1126865826961756e-05, + "loss": 0.1975, + "step": 305200 + }, + { + "epoch": 1.73, + "learning_rate": 2.110794503130446e-05, + "loss": 0.1996, + "step": 305400 + }, + { + "epoch": 1.73, + "learning_rate": 2.1089024235647158e-05, + "loss": 0.1993, + "step": 305600 + }, + { + "epoch": 1.74, + "learning_rate": 2.107010343998986e-05, + "loss": 0.1969, + "step": 305800 + }, + { + "epoch": 1.74, + "learning_rate": 2.105118264433256e-05, + "loss": 0.1895, + "step": 306000 + }, + { + "epoch": 1.74, + "eval_LOC_f1": 0.8106248219334196, + "eval_LOC_number": 22273, + "eval_LOC_precision": 0.7918307929439973, + "eval_LOC_recall": 0.8303326898037984, + "eval_ORG_f1": 0.6712868083270412, + "eval_ORG_number": 19008, + "eval_ORG_precision": 0.6885926089842885, + "eval_ORG_recall": 0.6548295454545454, + "eval_PER_f1": 0.8088823253971112, + "eval_PER_number": 27575, + "eval_PER_precision": 0.8034417373260349, + "eval_PER_recall": 0.8143970988213962, + "eval_loss": 0.19554340839385986, + "eval_overall_accuracy": 0.9402982181537999, + "eval_overall_f1": 0.7725460977003595, + "eval_overall_precision": 0.7696121528328265, + "eval_overall_recall": 0.7755024979667713, + "eval_runtime": 74.7177, + "eval_samples_per_second": 507.068, + "eval_steps_per_second": 63.385, + "step": 306000 + }, + { + "epoch": 1.74, + "learning_rate": 2.103226184867526e-05, + "loss": 0.1979, + "step": 306200 + }, + { + "epoch": 1.74, + "learning_rate": 2.1013341053017964e-05, + "loss": 0.1905, + "step": 306400 + }, + { + "epoch": 1.74, + "learning_rate": 2.0994420257360663e-05, + "loss": 0.1984, + "step": 306600 + }, + { + "epoch": 1.74, + "learning_rate": 2.0975499461703365e-05, + "loss": 0.1945, + "step": 306800 + }, + { + "epoch": 1.74, + "learning_rate": 2.0956578666046064e-05, + "loss": 0.1977, + "step": 307000 + }, + { + "epoch": 1.74, + "learning_rate": 2.0937657870388766e-05, + "loss": 0.1998, + "step": 307200 + }, + { + "epoch": 1.74, + "learning_rate": 2.091873707473147e-05, + "loss": 0.1949, + "step": 307400 + }, + { + "epoch": 1.75, + "learning_rate": 2.0899816279074168e-05, + "loss": 0.1937, + "step": 307600 + }, + { + "epoch": 1.75, + "learning_rate": 2.088089548341687e-05, + "loss": 0.1971, + "step": 307800 + }, + { + "epoch": 1.75, + "learning_rate": 2.086197468775957e-05, + "loss": 0.1982, + "step": 308000 + }, + { + "epoch": 1.75, + "eval_LOC_f1": 0.8100895392006988, + "eval_LOC_number": 22273, + "eval_LOC_precision": 0.7886635200068036, + "eval_LOC_recall": 0.8327122525030306, + "eval_ORG_f1": 0.6704665743831936, + "eval_ORG_number": 19008, + "eval_ORG_precision": 0.6925147182506308, + "eval_ORG_recall": 0.6497790404040404, + "eval_PER_f1": 0.8087129707713562, + "eval_PER_number": 27575, + "eval_PER_precision": 0.8003409333049222, + "eval_PER_recall": 0.8172620126926564, + "eval_loss": 0.19455909729003906, + "eval_overall_accuracy": 0.9397780889937762, + "eval_overall_f1": 0.7723573710304555, + "eval_overall_precision": 0.7687239246151633, + "eval_overall_recall": 0.7760253282212153, + "eval_runtime": 75.0435, + "eval_samples_per_second": 504.867, + "eval_steps_per_second": 63.11, + "step": 308000 + }, + { + "epoch": 1.75, + "learning_rate": 2.084305389210227e-05, + "loss": 0.1914, + "step": 308200 + }, + { + "epoch": 1.75, + "learning_rate": 2.0824133096444974e-05, + "loss": 0.2002, + "step": 308400 + }, + { + "epoch": 1.75, + "learning_rate": 2.0805212300787676e-05, + "loss": 0.1905, + "step": 308600 + }, + { + "epoch": 1.75, + "learning_rate": 2.0786291505130375e-05, + "loss": 0.2046, + "step": 308800 + }, + { + "epoch": 1.75, + "learning_rate": 2.0767370709473074e-05, + "loss": 0.197, + "step": 309000 + }, + { + "epoch": 1.76, + "learning_rate": 2.0748449913815776e-05, + "loss": 0.2117, + "step": 309200 + }, + { + "epoch": 1.76, + "learning_rate": 2.072952911815848e-05, + "loss": 0.203, + "step": 309400 + }, + { + "epoch": 1.76, + "learning_rate": 2.071060832250118e-05, + "loss": 0.1919, + "step": 309600 + }, + { + "epoch": 1.76, + "learning_rate": 2.069168752684388e-05, + "loss": 0.1915, + "step": 309800 + }, + { + "epoch": 1.76, + "learning_rate": 2.067276673118658e-05, + "loss": 0.198, + "step": 310000 + }, + { + "epoch": 1.76, + "eval_LOC_f1": 0.8099195127256907, + "eval_LOC_number": 22273, + "eval_LOC_precision": 0.7855846731653796, + "eval_LOC_recall": 0.8358101737529745, + "eval_ORG_f1": 0.6737967914438503, + "eval_ORG_number": 19008, + "eval_ORG_precision": 0.6815759728725981, + "eval_ORG_recall": 0.6661931818181818, + "eval_PER_f1": 0.8103946447479407, + "eval_PER_number": 27575, + "eval_PER_precision": 0.8021743764655723, + "eval_PER_recall": 0.8187851314596555, + "eval_loss": 0.19465883076190948, + "eval_overall_accuracy": 0.9402626813788915, + "eval_overall_f1": 0.7733741150791942, + "eval_overall_precision": 0.7647752122916134, + "eval_overall_recall": 0.782168583710933, + "eval_runtime": 72.245, + "eval_samples_per_second": 524.424, + "eval_steps_per_second": 65.555, + "step": 310000 + }, + { + "epoch": 1.76, + "learning_rate": 2.065384593552928e-05, + "loss": 0.2001, + "step": 310200 + }, + { + "epoch": 1.76, + "learning_rate": 2.0634925139871983e-05, + "loss": 0.1936, + "step": 310400 + }, + { + "epoch": 1.76, + "learning_rate": 2.0616004344214686e-05, + "loss": 0.1954, + "step": 310600 + }, + { + "epoch": 1.76, + "learning_rate": 2.0597083548557385e-05, + "loss": 0.2056, + "step": 310800 + }, + { + "epoch": 1.77, + "learning_rate": 2.0578162752900084e-05, + "loss": 0.1955, + "step": 311000 + }, + { + "epoch": 1.77, + "learning_rate": 2.0559241957242786e-05, + "loss": 0.2003, + "step": 311200 + }, + { + "epoch": 1.77, + "learning_rate": 2.054032116158549e-05, + "loss": 0.1984, + "step": 311400 + }, + { + "epoch": 1.77, + "learning_rate": 2.052140036592819e-05, + "loss": 0.1936, + "step": 311600 + }, + { + "epoch": 1.77, + "learning_rate": 2.050247957027089e-05, + "loss": 0.1991, + "step": 311800 + }, + { + "epoch": 1.77, + "learning_rate": 2.048355877461359e-05, + "loss": 0.1926, + "step": 312000 + }, + { + "epoch": 1.77, + "eval_LOC_f1": 0.8101813048656756, + "eval_LOC_number": 22273, + "eval_LOC_precision": 0.7962714068935617, + "eval_LOC_recall": 0.8245858213981053, + "eval_ORG_f1": 0.6729764316945719, + "eval_ORG_number": 19008, + "eval_ORG_precision": 0.6822162162162162, + "eval_ORG_recall": 0.6639835858585859, + "eval_PER_f1": 0.810407719537834, + "eval_PER_number": 27575, + "eval_PER_precision": 0.8031770907536686, + "eval_PER_recall": 0.8177697189483227, + "eval_loss": 0.19470322132110596, + "eval_overall_accuracy": 0.9401738394416204, + "eval_overall_f1": 0.7731142190805577, + "eval_overall_precision": 0.7687569104406887, + "eval_overall_recall": 0.7775212036714302, + "eval_runtime": 73.3216, + "eval_samples_per_second": 516.724, + "eval_steps_per_second": 64.592, + "step": 312000 + }, + { + "epoch": 1.77, + "learning_rate": 2.046463797895629e-05, + "loss": 0.1976, + "step": 312200 + }, + { + "epoch": 1.77, + "learning_rate": 2.0445717183298993e-05, + "loss": 0.2038, + "step": 312400 + }, + { + "epoch": 1.77, + "learning_rate": 2.0426796387641696e-05, + "loss": 0.204, + "step": 312600 + }, + { + "epoch": 1.78, + "learning_rate": 2.0407875591984395e-05, + "loss": 0.2022, + "step": 312800 + }, + { + "epoch": 1.78, + "learning_rate": 2.0388954796327093e-05, + "loss": 0.1953, + "step": 313000 + }, + { + "epoch": 1.78, + "learning_rate": 2.0370034000669796e-05, + "loss": 0.2044, + "step": 313200 + }, + { + "epoch": 1.78, + "learning_rate": 2.0351113205012498e-05, + "loss": 0.197, + "step": 313400 + }, + { + "epoch": 1.78, + "learning_rate": 2.03321924093552e-05, + "loss": 0.1924, + "step": 313600 + }, + { + "epoch": 1.78, + "learning_rate": 2.0313271613697903e-05, + "loss": 0.1985, + "step": 313800 + }, + { + "epoch": 1.78, + "learning_rate": 2.02943508180406e-05, + "loss": 0.2005, + "step": 314000 + }, + { + "epoch": 1.78, + "eval_LOC_f1": 0.8108757895657115, + "eval_LOC_number": 22273, + "eval_LOC_precision": 0.7900340715502555, + "eval_LOC_recall": 0.832846944731289, + "eval_ORG_f1": 0.6700048719753153, + "eval_ORG_number": 19008, + "eval_ORG_precision": 0.6899877355335043, + "eval_ORG_recall": 0.6511468855218855, + "eval_PER_f1": 0.8105984138428263, + "eval_PER_number": 27575, + "eval_PER_precision": 0.8058054112166279, + "eval_PER_recall": 0.8154487760652765, + "eval_loss": 0.19280676543712616, + "eval_overall_accuracy": 0.9402707579186435, + "eval_overall_f1": 0.773098661880604, + "eval_overall_precision": 0.7704946410282302, + "eval_overall_recall": 0.7757203439061229, + "eval_runtime": 72.4808, + "eval_samples_per_second": 522.718, + "eval_steps_per_second": 65.341, + "step": 314000 + }, + { + "epoch": 1.78, + "learning_rate": 2.02754300223833e-05, + "loss": 0.2025, + "step": 314200 + }, + { + "epoch": 1.78, + "learning_rate": 2.0256509226726003e-05, + "loss": 0.1936, + "step": 314400 + }, + { + "epoch": 1.79, + "learning_rate": 2.0237588431068705e-05, + "loss": 0.1981, + "step": 314600 + }, + { + "epoch": 1.79, + "learning_rate": 2.0218667635411408e-05, + "loss": 0.1936, + "step": 314800 + }, + { + "epoch": 1.79, + "learning_rate": 2.0199746839754107e-05, + "loss": 0.204, + "step": 315000 + }, + { + "epoch": 1.79, + "learning_rate": 2.0180826044096806e-05, + "loss": 0.1953, + "step": 315200 + }, + { + "epoch": 1.79, + "learning_rate": 2.0161905248439508e-05, + "loss": 0.1958, + "step": 315400 + }, + { + "epoch": 1.79, + "learning_rate": 2.014298445278221e-05, + "loss": 0.1902, + "step": 315600 + }, + { + "epoch": 1.79, + "learning_rate": 2.0124063657124913e-05, + "loss": 0.2028, + "step": 315800 + }, + { + "epoch": 1.79, + "learning_rate": 2.010514286146761e-05, + "loss": 0.2048, + "step": 316000 + }, + { + "epoch": 1.79, + "eval_LOC_f1": 0.8104307190406913, + "eval_LOC_number": 22273, + "eval_LOC_precision": 0.7960420906768285, + "eval_LOC_recall": 0.8253490773582365, + "eval_ORG_f1": 0.675379154565989, + "eval_ORG_number": 19008, + "eval_ORG_precision": 0.6907590759075908, + "eval_ORG_recall": 0.6606691919191919, + "eval_PER_f1": 0.8099274040696636, + "eval_PER_number": 27575, + "eval_PER_precision": 0.7988430743183662, + "eval_PER_recall": 0.8213236627379873, + "eval_loss": 0.19295653700828552, + "eval_overall_accuracy": 0.9404678254885903, + "eval_overall_f1": 0.7739601386481801, + "eval_overall_precision": 0.7696914856945881, + "eval_overall_recall": 0.7782764029278494, + "eval_runtime": 72.4403, + "eval_samples_per_second": 523.01, + "eval_steps_per_second": 65.378, + "step": 316000 + }, + { + "epoch": 1.79, + "learning_rate": 2.008622206581031e-05, + "loss": 0.1928, + "step": 316200 + }, + { + "epoch": 1.8, + "learning_rate": 2.0067301270153013e-05, + "loss": 0.1915, + "step": 316400 + }, + { + "epoch": 1.8, + "learning_rate": 2.0048380474495715e-05, + "loss": 0.2029, + "step": 316600 + }, + { + "epoch": 1.8, + "learning_rate": 2.0029459678838418e-05, + "loss": 0.2, + "step": 316800 + }, + { + "epoch": 1.8, + "learning_rate": 2.0010538883181117e-05, + "loss": 0.2132, + "step": 317000 + }, + { + "epoch": 1.8, + "learning_rate": 1.9991618087523816e-05, + "loss": 0.2083, + "step": 317200 + }, + { + "epoch": 1.8, + "learning_rate": 1.9972697291866518e-05, + "loss": 0.1915, + "step": 317400 + }, + { + "epoch": 1.8, + "learning_rate": 1.995377649620922e-05, + "loss": 0.1986, + "step": 317600 + }, + { + "epoch": 1.8, + "learning_rate": 1.9934855700551923e-05, + "loss": 0.2009, + "step": 317800 + }, + { + "epoch": 1.81, + "learning_rate": 1.991593490489462e-05, + "loss": 0.1938, + "step": 318000 + }, + { + "epoch": 1.81, + "eval_LOC_f1": 0.811801598869308, + "eval_LOC_number": 22273, + "eval_LOC_precision": 0.7988178538832631, + "eval_LOC_recall": 0.825214385129978, + "eval_ORG_f1": 0.6749640211076168, + "eval_ORG_number": 19008, + "eval_ORG_precision": 0.6839688884087718, + "eval_ORG_recall": 0.6661931818181818, + "eval_PER_f1": 0.8105053549324579, + "eval_PER_number": 27575, + "eval_PER_precision": 0.8019738710593581, + "eval_PER_recall": 0.8192203082502266, + "eval_loss": 0.19275246560573578, + "eval_overall_accuracy": 0.9405712051974149, + "eval_overall_f1": 0.7742210224689096, + "eval_overall_precision": 0.769582872967815, + "eval_overall_recall": 0.7789154176832811, + "eval_runtime": 73.2188, + "eval_samples_per_second": 517.449, + "eval_steps_per_second": 64.683, + "step": 318000 + }, + { + "epoch": 1.81, + "learning_rate": 1.989701410923732e-05, + "loss": 0.199, + "step": 318200 + }, + { + "epoch": 1.81, + "learning_rate": 1.9878093313580023e-05, + "loss": 0.1884, + "step": 318400 + }, + { + "epoch": 1.81, + "learning_rate": 1.9859172517922725e-05, + "loss": 0.1933, + "step": 318600 + }, + { + "epoch": 1.81, + "learning_rate": 1.9840251722265428e-05, + "loss": 0.1998, + "step": 318800 + }, + { + "epoch": 1.81, + "learning_rate": 1.9821330926608126e-05, + "loss": 0.1939, + "step": 319000 + }, + { + "epoch": 1.81, + "learning_rate": 1.9802410130950825e-05, + "loss": 0.1859, + "step": 319200 + }, + { + "epoch": 1.81, + "learning_rate": 1.9783489335293528e-05, + "loss": 0.193, + "step": 319400 + }, + { + "epoch": 1.81, + "learning_rate": 1.976456853963623e-05, + "loss": 0.1976, + "step": 319600 + }, + { + "epoch": 1.82, + "learning_rate": 1.9745647743978932e-05, + "loss": 0.1996, + "step": 319800 + }, + { + "epoch": 1.82, + "learning_rate": 1.972672694832163e-05, + "loss": 0.1957, + "step": 320000 + }, + { + "epoch": 1.82, + "eval_LOC_f1": 0.809111144728027, + "eval_LOC_number": 22273, + "eval_LOC_precision": 0.78, + "eval_LOC_recall": 0.8404795043326, + "eval_ORG_f1": 0.675734058724698, + "eval_ORG_number": 19008, + "eval_ORG_precision": 0.6852182378711666, + "eval_ORG_recall": 0.6665088383838383, + "eval_PER_f1": 0.8092052160499363, + "eval_PER_number": 27575, + "eval_PER_precision": 0.8005749574105622, + "eval_PER_recall": 0.818023572076156, + "eval_loss": 0.19366350769996643, + "eval_overall_accuracy": 0.9401738394416204, + "eval_overall_f1": 0.7733029436428924, + "eval_overall_precision": 0.7634047972829547, + "eval_overall_recall": 0.7834611362844197, + "eval_runtime": 76.2551, + "eval_samples_per_second": 496.845, + "eval_steps_per_second": 62.107, + "step": 320000 + }, + { + "epoch": 1.82, + "learning_rate": 1.970780615266433e-05, + "loss": 0.2006, + "step": 320200 + }, + { + "epoch": 1.82, + "learning_rate": 1.9688885357007033e-05, + "loss": 0.199, + "step": 320400 + }, + { + "epoch": 1.82, + "learning_rate": 1.9669964561349735e-05, + "loss": 0.1944, + "step": 320600 + }, + { + "epoch": 1.82, + "learning_rate": 1.9651043765692437e-05, + "loss": 0.2037, + "step": 320800 + }, + { + "epoch": 1.82, + "learning_rate": 1.9632122970035136e-05, + "loss": 0.1937, + "step": 321000 + }, + { + "epoch": 1.82, + "learning_rate": 1.961320217437784e-05, + "loss": 0.2118, + "step": 321200 + }, + { + "epoch": 1.82, + "learning_rate": 1.9594281378720538e-05, + "loss": 0.1842, + "step": 321400 + }, + { + "epoch": 1.83, + "learning_rate": 1.957536058306324e-05, + "loss": 0.1899, + "step": 321600 + }, + { + "epoch": 1.83, + "learning_rate": 1.9556439787405942e-05, + "loss": 0.1997, + "step": 321800 + }, + { + "epoch": 1.83, + "learning_rate": 1.953751899174864e-05, + "loss": 0.1949, + "step": 322000 + }, + { + "epoch": 1.83, + "eval_LOC_f1": 0.8115808823529412, + "eval_LOC_number": 22273, + "eval_LOC_precision": 0.7916577722751142, + "eval_LOC_recall": 0.8325326628653527, + "eval_ORG_f1": 0.6746634679461548, + "eval_ORG_number": 19008, + "eval_ORG_precision": 0.6837953206894688, + "eval_ORG_recall": 0.6657723063973064, + "eval_PER_f1": 0.8106713806414373, + "eval_PER_number": 27575, + "eval_PER_precision": 0.8064817141011377, + "eval_PER_recall": 0.8149048050770625, + "eval_loss": 0.19242261350154877, + "eval_overall_accuracy": 0.940521130650953, + "eval_overall_f1": 0.7741707477154541, + "eval_overall_precision": 0.7689739658705028, + "eval_overall_recall": 0.7794382479377251, + "eval_runtime": 74.0396, + "eval_samples_per_second": 511.713, + "eval_steps_per_second": 63.966, + "step": 322000 + }, + { + "epoch": 1.83, + "learning_rate": 1.9518598196091344e-05, + "loss": 0.1934, + "step": 322200 + }, + { + "epoch": 1.83, + "learning_rate": 1.9499677400434043e-05, + "loss": 0.2031, + "step": 322400 + }, + { + "epoch": 1.83, + "learning_rate": 1.9480756604776745e-05, + "loss": 0.1967, + "step": 322600 + }, + { + "epoch": 1.83, + "learning_rate": 1.9461835809119447e-05, + "loss": 0.1972, + "step": 322800 + }, + { + "epoch": 1.83, + "learning_rate": 1.9442915013462146e-05, + "loss": 0.1984, + "step": 323000 + }, + { + "epoch": 1.83, + "learning_rate": 1.942399421780485e-05, + "loss": 0.1961, + "step": 323200 + }, + { + "epoch": 1.84, + "learning_rate": 1.9405073422147548e-05, + "loss": 0.2005, + "step": 323400 + }, + { + "epoch": 1.84, + "learning_rate": 1.938615262649025e-05, + "loss": 0.1975, + "step": 323600 + }, + { + "epoch": 1.84, + "learning_rate": 1.9367231830832952e-05, + "loss": 0.1947, + "step": 323800 + }, + { + "epoch": 1.84, + "learning_rate": 1.934831103517565e-05, + "loss": 0.1904, + "step": 324000 + }, + { + "epoch": 1.84, + "eval_LOC_f1": 0.8118614286643961, + "eval_LOC_number": 22273, + "eval_LOC_precision": 0.7854025014689835, + "eval_LOC_recall": 0.8401652224666637, + "eval_ORG_f1": 0.6732424485431703, + "eval_ORG_number": 19008, + "eval_ORG_precision": 0.684327790457559, + "eval_ORG_recall": 0.6625105218855218, + "eval_PER_f1": 0.8115343154868846, + "eval_PER_number": 27575, + "eval_PER_precision": 0.8041659248709275, + "eval_PER_recall": 0.8190389845874887, + "eval_loss": 0.1931518316268921, + "eval_overall_accuracy": 0.9406487399790333, + "eval_overall_f1": 0.774468452025954, + "eval_overall_precision": 0.7664443275070044, + "eval_overall_recall": 0.7826623678401301, + "eval_runtime": 75.7429, + "eval_samples_per_second": 500.205, + "eval_steps_per_second": 62.527, + "step": 324000 + }, + { + "epoch": 1.84, + "learning_rate": 1.9329390239518353e-05, + "loss": 0.1892, + "step": 324200 + }, + { + "epoch": 1.84, + "learning_rate": 1.9310469443861052e-05, + "loss": 0.2034, + "step": 324400 + }, + { + "epoch": 1.84, + "learning_rate": 1.9291548648203755e-05, + "loss": 0.1936, + "step": 324600 + }, + { + "epoch": 1.84, + "learning_rate": 1.9272627852546457e-05, + "loss": 0.2014, + "step": 324800 + }, + { + "epoch": 1.84, + "learning_rate": 1.9253707056889156e-05, + "loss": 0.1961, + "step": 325000 + }, + { + "epoch": 1.85, + "learning_rate": 1.923478626123186e-05, + "loss": 0.1968, + "step": 325200 + }, + { + "epoch": 1.85, + "learning_rate": 1.9215865465574557e-05, + "loss": 0.1962, + "step": 325400 + }, + { + "epoch": 1.85, + "learning_rate": 1.919694466991726e-05, + "loss": 0.1941, + "step": 325600 + }, + { + "epoch": 1.85, + "learning_rate": 1.9178023874259962e-05, + "loss": 0.1981, + "step": 325800 + }, + { + "epoch": 1.85, + "learning_rate": 1.915910307860266e-05, + "loss": 0.194, + "step": 326000 + }, + { + "epoch": 1.85, + "eval_LOC_f1": 0.8103212947115281, + "eval_LOC_number": 22273, + "eval_LOC_precision": 0.7829927985262101, + "eval_LOC_recall": 0.83962645355363, + "eval_ORG_f1": 0.6762650857505822, + "eval_ORG_number": 19008, + "eval_ORG_precision": 0.6804431188751597, + "eval_ORG_recall": 0.6721380471380471, + "eval_PER_f1": 0.8114796788850774, + "eval_PER_number": 27575, + "eval_PER_precision": 0.8038141322137622, + "eval_PER_recall": 0.8192928377153218, + "eval_loss": 0.19200630486011505, + "eval_overall_accuracy": 0.9404855938760445, + "eval_overall_f1": 0.774505450430448, + "eval_overall_precision": 0.7640533589576916, + "eval_overall_recall": 0.7852474729871035, + "eval_runtime": 72.1381, + "eval_samples_per_second": 525.201, + "eval_steps_per_second": 65.652, + "step": 326000 + }, + { + "epoch": 1.85, + "learning_rate": 1.9140182282945363e-05, + "loss": 0.1891, + "step": 326200 + }, + { + "epoch": 1.85, + "learning_rate": 1.9121261487288066e-05, + "loss": 0.195, + "step": 326400 + }, + { + "epoch": 1.85, + "learning_rate": 1.9102340691630765e-05, + "loss": 0.1936, + "step": 326600 + }, + { + "epoch": 1.85, + "learning_rate": 1.9083419895973467e-05, + "loss": 0.1857, + "step": 326800 + }, + { + "epoch": 1.86, + "learning_rate": 1.9064499100316166e-05, + "loss": 0.1859, + "step": 327000 + }, + { + "epoch": 1.86, + "learning_rate": 1.9045578304658868e-05, + "loss": 0.1878, + "step": 327200 + }, + { + "epoch": 1.86, + "learning_rate": 1.902665750900157e-05, + "loss": 0.1909, + "step": 327400 + }, + { + "epoch": 1.86, + "learning_rate": 1.900773671334427e-05, + "loss": 0.197, + "step": 327600 + }, + { + "epoch": 1.86, + "learning_rate": 1.8988815917686972e-05, + "loss": 0.2036, + "step": 327800 + }, + { + "epoch": 1.86, + "learning_rate": 1.896989512202967e-05, + "loss": 0.189, + "step": 328000 + }, + { + "epoch": 1.86, + "eval_LOC_f1": 0.8113617594645108, + "eval_LOC_number": 22273, + "eval_LOC_precision": 0.7862769049323954, + "eval_LOC_recall": 0.8380999416333678, + "eval_ORG_f1": 0.6722840011622072, + "eval_ORG_number": 19008, + "eval_ORG_precision": 0.6750835499443001, + "eval_ORG_recall": 0.6695075757575758, + "eval_PER_f1": 0.8110718662149499, + "eval_PER_number": 27575, + "eval_PER_precision": 0.8061038077157288, + "eval_PER_recall": 0.8161015412511333, + "eval_loss": 0.19296719133853912, + "eval_overall_accuracy": 0.9404258274818803, + "eval_overall_f1": 0.7734653607433717, + "eval_overall_precision": 0.764398871066105, + "eval_overall_recall": 0.7827495062158708, + "eval_runtime": 71.833, + "eval_samples_per_second": 527.432, + "eval_steps_per_second": 65.931, + "step": 328000 + }, + { + "epoch": 1.86, + "learning_rate": 1.8950974326372373e-05, + "loss": 0.1927, + "step": 328200 + }, + { + "epoch": 1.86, + "learning_rate": 1.8932053530715076e-05, + "loss": 0.193, + "step": 328400 + }, + { + "epoch": 1.87, + "learning_rate": 1.8913132735057775e-05, + "loss": 0.1899, + "step": 328600 + }, + { + "epoch": 1.87, + "learning_rate": 1.8894211939400477e-05, + "loss": 0.2017, + "step": 328800 + }, + { + "epoch": 1.87, + "learning_rate": 1.887529114374318e-05, + "loss": 0.2063, + "step": 329000 + }, + { + "epoch": 1.87, + "learning_rate": 1.8856370348085878e-05, + "loss": 0.197, + "step": 329200 + }, + { + "epoch": 1.87, + "learning_rate": 1.883744955242858e-05, + "loss": 0.1831, + "step": 329400 + }, + { + "epoch": 1.87, + "learning_rate": 1.881852875677128e-05, + "loss": 0.195, + "step": 329600 + }, + { + "epoch": 1.87, + "learning_rate": 1.8799607961113982e-05, + "loss": 0.2002, + "step": 329800 + }, + { + "epoch": 1.87, + "learning_rate": 1.8780687165456684e-05, + "loss": 0.1935, + "step": 330000 + }, + { + "epoch": 1.87, + "eval_LOC_f1": 0.8133379918766651, + "eval_LOC_number": 22273, + "eval_LOC_precision": 0.7917605543981974, + "eval_LOC_recall": 0.8361244556189108, + "eval_ORG_f1": 0.674191240759876, + "eval_ORG_number": 19008, + "eval_ORG_precision": 0.6791032826261009, + "eval_ORG_recall": 0.6693497474747475, + "eval_PER_f1": 0.810127490182718, + "eval_PER_number": 27575, + "eval_PER_precision": 0.801234305171313, + "eval_PER_recall": 0.8192203082502266, + "eval_loss": 0.19240093231201172, + "eval_overall_accuracy": 0.9407391972242548, + "eval_overall_f1": 0.7743528634803957, + "eval_overall_precision": 0.7655926188786373, + "eval_overall_recall": 0.7833159056581852, + "eval_runtime": 72.7584, + "eval_samples_per_second": 520.723, + "eval_steps_per_second": 65.092, + "step": 330000 + }, + { + "epoch": 1.87, + "learning_rate": 1.8761766369799383e-05, + "loss": 0.1976, + "step": 330200 + }, + { + "epoch": 1.88, + "learning_rate": 1.8742845574142085e-05, + "loss": 0.2012, + "step": 330400 + }, + { + "epoch": 1.88, + "learning_rate": 1.8723924778484784e-05, + "loss": 0.1967, + "step": 330600 + }, + { + "epoch": 1.88, + "learning_rate": 1.8705003982827487e-05, + "loss": 0.1978, + "step": 330800 + }, + { + "epoch": 1.88, + "learning_rate": 1.868608318717019e-05, + "loss": 0.1923, + "step": 331000 + }, + { + "epoch": 1.88, + "learning_rate": 1.8667162391512888e-05, + "loss": 0.1979, + "step": 331200 + }, + { + "epoch": 1.88, + "learning_rate": 1.864824159585559e-05, + "loss": 0.1945, + "step": 331400 + }, + { + "epoch": 1.88, + "learning_rate": 1.8629320800198293e-05, + "loss": 0.1961, + "step": 331600 + }, + { + "epoch": 1.88, + "learning_rate": 1.861040000454099e-05, + "loss": 0.2006, + "step": 331800 + }, + { + "epoch": 1.88, + "learning_rate": 1.8591479208883694e-05, + "loss": 0.1943, + "step": 332000 + }, + { + "epoch": 1.88, + "eval_LOC_f1": 0.8145059167721933, + "eval_LOC_number": 22273, + "eval_LOC_precision": 0.7927663734115347, + "eval_LOC_recall": 0.8374713779014951, + "eval_ORG_f1": 0.6746879150066402, + "eval_ORG_number": 19008, + "eval_ORG_precision": 0.6813110181311018, + "eval_ORG_recall": 0.6681923400673401, + "eval_PER_f1": 0.8117326795146838, + "eval_PER_number": 27575, + "eval_PER_precision": 0.8070911307091131, + "eval_PER_recall": 0.8164279238440616, + "eval_loss": 0.19237257540225983, + "eval_overall_accuracy": 0.9406212797438768, + "eval_overall_f1": 0.775505503127677, + "eval_overall_precision": 0.7688146720902019, + "eval_overall_recall": 0.7823138143371674, + "eval_runtime": 73.3694, + "eval_samples_per_second": 516.387, + "eval_steps_per_second": 64.55, + "step": 332000 + }, + { + "epoch": 1.89, + "learning_rate": 1.8572558413226393e-05, + "loss": 0.1954, + "step": 332200 + }, + { + "epoch": 1.89, + "learning_rate": 1.8553637617569095e-05, + "loss": 0.1957, + "step": 332400 + }, + { + "epoch": 1.89, + "learning_rate": 1.8534716821911798e-05, + "loss": 0.1991, + "step": 332600 + }, + { + "epoch": 1.89, + "learning_rate": 1.8515796026254497e-05, + "loss": 0.2038, + "step": 332800 + }, + { + "epoch": 1.89, + "learning_rate": 1.84968752305972e-05, + "loss": 0.1962, + "step": 333000 + }, + { + "epoch": 1.89, + "learning_rate": 1.8477954434939898e-05, + "loss": 0.1884, + "step": 333200 + }, + { + "epoch": 1.89, + "learning_rate": 1.84590336392826e-05, + "loss": 0.1978, + "step": 333400 + }, + { + "epoch": 1.89, + "learning_rate": 1.8440112843625303e-05, + "loss": 0.2023, + "step": 333600 + }, + { + "epoch": 1.89, + "learning_rate": 1.8421192047968e-05, + "loss": 0.1947, + "step": 333800 + }, + { + "epoch": 1.9, + "learning_rate": 1.8402271252310704e-05, + "loss": 0.1925, + "step": 334000 + }, + { + "epoch": 1.9, + "eval_LOC_f1": 0.8111612422795567, + "eval_LOC_number": 22273, + "eval_LOC_precision": 0.7918751336326705, + "eval_LOC_recall": 0.8314102276298657, + "eval_ORG_f1": 0.6762463810459775, + "eval_ORG_number": 19008, + "eval_ORG_precision": 0.6829032777211523, + "eval_ORG_recall": 0.6697180134680135, + "eval_PER_f1": 0.8121988631237032, + "eval_PER_number": 27575, + "eval_PER_precision": 0.8083333333333333, + "eval_PER_recall": 0.8161015412511333, + "eval_loss": 0.19252440333366394, + "eval_overall_accuracy": 0.9407214288368005, + "eval_overall_f1": 0.7749599919262986, + "eval_overall_precision": 0.7693584862450977, + "eval_overall_recall": 0.7806436621354711, + "eval_runtime": 75.0935, + "eval_samples_per_second": 504.531, + "eval_steps_per_second": 63.068, + "step": 334000 + }, + { + "epoch": 1.9, + "learning_rate": 1.8383350456653403e-05, + "loss": 0.189, + "step": 334200 + }, + { + "epoch": 1.9, + "learning_rate": 1.8364429660996105e-05, + "loss": 0.1868, + "step": 334400 + }, + { + "epoch": 1.9, + "learning_rate": 1.8345508865338807e-05, + "loss": 0.1981, + "step": 334600 + }, + { + "epoch": 1.9, + "learning_rate": 1.8326588069681506e-05, + "loss": 0.201, + "step": 334800 + }, + { + "epoch": 1.9, + "learning_rate": 1.830766727402421e-05, + "loss": 0.1832, + "step": 335000 + }, + { + "epoch": 1.9, + "learning_rate": 1.8288746478366908e-05, + "loss": 0.1937, + "step": 335200 + }, + { + "epoch": 1.9, + "learning_rate": 1.826982568270961e-05, + "loss": 0.1971, + "step": 335400 + }, + { + "epoch": 1.9, + "learning_rate": 1.8250904887052312e-05, + "loss": 0.1924, + "step": 335600 + }, + { + "epoch": 1.91, + "learning_rate": 1.823198409139501e-05, + "loss": 0.1957, + "step": 335800 + }, + { + "epoch": 1.91, + "learning_rate": 1.8213063295737714e-05, + "loss": 0.1855, + "step": 336000 + }, + { + "epoch": 1.91, + "eval_LOC_f1": 0.8114076239935412, + "eval_LOC_number": 22273, + "eval_LOC_precision": 0.7893105790456784, + "eval_LOC_recall": 0.8347775333363265, + "eval_ORG_f1": 0.6730753782300174, + "eval_ORG_number": 19008, + "eval_ORG_precision": 0.6853901946883351, + "eval_ORG_recall": 0.6611952861952862, + "eval_PER_f1": 0.8111125069542902, + "eval_PER_number": 27575, + "eval_PER_precision": 0.802884957009877, + "eval_PER_recall": 0.8195104261106074, + "eval_loss": 0.19400279223918915, + "eval_overall_accuracy": 0.9406535859028845, + "eval_overall_f1": 0.7740955397962489, + "eval_overall_precision": 0.7675580747869044, + "eval_overall_recall": 0.7807453235738353, + "eval_runtime": 72.7832, + "eval_samples_per_second": 520.546, + "eval_steps_per_second": 65.07, + "step": 336000 + }, + { + "epoch": 1.91, + "learning_rate": 1.8194142500080413e-05, + "loss": 0.1933, + "step": 336200 + }, + { + "epoch": 1.91, + "learning_rate": 1.8175221704423115e-05, + "loss": 0.1916, + "step": 336400 + }, + { + "epoch": 1.91, + "learning_rate": 1.8156300908765817e-05, + "loss": 0.205, + "step": 336600 + }, + { + "epoch": 1.91, + "learning_rate": 1.813738011310852e-05, + "loss": 0.1882, + "step": 336800 + }, + { + "epoch": 1.91, + "learning_rate": 1.811845931745122e-05, + "loss": 0.2017, + "step": 337000 + }, + { + "epoch": 1.91, + "learning_rate": 1.8099538521793918e-05, + "loss": 0.2029, + "step": 337200 + }, + { + "epoch": 1.92, + "learning_rate": 1.808061772613662e-05, + "loss": 0.1966, + "step": 337400 + }, + { + "epoch": 1.92, + "learning_rate": 1.8061696930479322e-05, + "loss": 0.1954, + "step": 337600 + }, + { + "epoch": 1.92, + "learning_rate": 1.8042776134822025e-05, + "loss": 0.1975, + "step": 337800 + }, + { + "epoch": 1.92, + "learning_rate": 1.8023855339164724e-05, + "loss": 0.1905, + "step": 338000 + }, + { + "epoch": 1.92, + "eval_LOC_f1": 0.8118394154234265, + "eval_LOC_number": 22273, + "eval_LOC_precision": 0.7872537854823063, + "eval_LOC_recall": 0.8380101468145288, + "eval_ORG_f1": 0.6748459740811558, + "eval_ORG_number": 19008, + "eval_ORG_precision": 0.6813599313599313, + "eval_ORG_recall": 0.6684553872053872, + "eval_PER_f1": 0.8112207306773052, + "eval_PER_number": 27575, + "eval_PER_precision": 0.8040396124251924, + "eval_PER_recall": 0.8185312783318223, + "eval_loss": 0.1921730637550354, + "eval_overall_accuracy": 0.9406422787472317, + "eval_overall_f1": 0.7745557669526509, + "eval_overall_precision": 0.7659060898209544, + "eval_overall_recall": 0.7834030440339259, + "eval_runtime": 71.6911, + "eval_samples_per_second": 528.476, + "eval_steps_per_second": 66.061, + "step": 338000 + }, + { + "epoch": 1.92, + "learning_rate": 1.8004934543507423e-05, + "loss": 0.1883, + "step": 338200 + }, + { + "epoch": 1.92, + "learning_rate": 1.7986013747850125e-05, + "loss": 0.1934, + "step": 338400 + }, + { + "epoch": 1.92, + "learning_rate": 1.7967092952192827e-05, + "loss": 0.1972, + "step": 338600 + }, + { + "epoch": 1.92, + "learning_rate": 1.794817215653553e-05, + "loss": 0.1999, + "step": 338800 + }, + { + "epoch": 1.92, + "learning_rate": 1.792925136087823e-05, + "loss": 0.1947, + "step": 339000 + }, + { + "epoch": 1.93, + "learning_rate": 1.7910330565220927e-05, + "loss": 0.2001, + "step": 339200 + }, + { + "epoch": 1.93, + "learning_rate": 1.789140976956363e-05, + "loss": 0.1904, + "step": 339400 + }, + { + "epoch": 1.93, + "learning_rate": 1.7872488973906332e-05, + "loss": 0.1912, + "step": 339600 + }, + { + "epoch": 1.93, + "learning_rate": 1.7853568178249034e-05, + "loss": 0.1924, + "step": 339800 + }, + { + "epoch": 1.93, + "learning_rate": 1.7834647382591733e-05, + "loss": 0.1949, + "step": 340000 + }, + { + "epoch": 1.93, + "eval_LOC_f1": 0.8140738945559497, + "eval_LOC_number": 22273, + "eval_LOC_precision": 0.7992299705831458, + "eval_LOC_recall": 0.8294796390248282, + "eval_ORG_f1": 0.6757429052879343, + "eval_ORG_number": 19008, + "eval_ORG_precision": 0.6873061646444312, + "eval_ORG_recall": 0.6645622895622896, + "eval_PER_f1": 0.8111063311106331, + "eval_PER_number": 27575, + "eval_PER_precision": 0.8019352780633041, + "eval_PER_recall": 0.8204895738893926, + "eval_loss": 0.19167812168598175, + "eval_overall_accuracy": 0.9409976464963163, + "eval_overall_f1": 0.7755549782050172, + "eval_overall_precision": 0.7708154013886498, + "eval_overall_recall": 0.7803532008830022, + "eval_runtime": 72.0843, + "eval_samples_per_second": 525.593, + "eval_steps_per_second": 65.701, + "step": 340000 + }, + { + "epoch": 1.93, + "learning_rate": 1.7815726586934432e-05, + "loss": 0.1912, + "step": 340200 + }, + { + "epoch": 1.93, + "learning_rate": 1.7796805791277135e-05, + "loss": 0.19, + "step": 340400 + }, + { + "epoch": 1.93, + "learning_rate": 1.7777884995619837e-05, + "loss": 0.1995, + "step": 340600 + }, + { + "epoch": 1.93, + "learning_rate": 1.775896419996254e-05, + "loss": 0.1906, + "step": 340800 + }, + { + "epoch": 1.94, + "learning_rate": 1.774004340430524e-05, + "loss": 0.1984, + "step": 341000 + }, + { + "epoch": 1.94, + "learning_rate": 1.7721122608647937e-05, + "loss": 0.1922, + "step": 341200 + }, + { + "epoch": 1.94, + "learning_rate": 1.770220181299064e-05, + "loss": 0.1926, + "step": 341400 + }, + { + "epoch": 1.94, + "learning_rate": 1.7683281017333342e-05, + "loss": 0.1939, + "step": 341600 + }, + { + "epoch": 1.94, + "learning_rate": 1.7664360221676044e-05, + "loss": 0.2074, + "step": 341800 + }, + { + "epoch": 1.94, + "learning_rate": 1.7645439426018747e-05, + "loss": 0.1919, + "step": 342000 + }, + { + "epoch": 1.94, + "eval_LOC_f1": 0.8108025232171017, + "eval_LOC_number": 22273, + "eval_LOC_precision": 0.7915579694649959, + "eval_LOC_recall": 0.8310061509450904, + "eval_ORG_f1": 0.6733894142849678, + "eval_ORG_number": 19008, + "eval_ORG_precision": 0.6688116242864557, + "eval_ORG_recall": 0.678030303030303, + "eval_PER_f1": 0.8109796599824067, + "eval_PER_number": 27575, + "eval_PER_precision": 0.8030076791808873, + "eval_PER_recall": 0.8191115140525839, + "eval_loss": 0.19148632884025574, + "eval_overall_accuracy": 0.9406067419723233, + "eval_overall_f1": 0.7732048096134979, + "eval_overall_precision": 0.762690552549413, + "eval_overall_recall": 0.7840130126641106, + "eval_runtime": 71.8632, + "eval_samples_per_second": 527.21, + "eval_steps_per_second": 65.903, + "step": 342000 + }, + { + "epoch": 1.94, + "learning_rate": 1.7626518630361442e-05, + "loss": 0.2001, + "step": 342200 + }, + { + "epoch": 1.94, + "learning_rate": 1.7607597834704145e-05, + "loss": 0.1904, + "step": 342400 + }, + { + "epoch": 1.94, + "learning_rate": 1.7588677039046847e-05, + "loss": 0.1862, + "step": 342600 + }, + { + "epoch": 1.95, + "learning_rate": 1.756975624338955e-05, + "loss": 0.1911, + "step": 342800 + }, + { + "epoch": 1.95, + "learning_rate": 1.755083544773225e-05, + "loss": 0.1898, + "step": 343000 + }, + { + "epoch": 1.95, + "learning_rate": 1.7531914652074947e-05, + "loss": 0.1963, + "step": 343200 + }, + { + "epoch": 1.95, + "learning_rate": 1.751299385641765e-05, + "loss": 0.1856, + "step": 343400 + }, + { + "epoch": 1.95, + "learning_rate": 1.7494073060760352e-05, + "loss": 0.1932, + "step": 343600 + }, + { + "epoch": 1.95, + "learning_rate": 1.7475152265103054e-05, + "loss": 0.1898, + "step": 343800 + }, + { + "epoch": 1.95, + "learning_rate": 1.7456231469445757e-05, + "loss": 0.1947, + "step": 344000 + }, + { + "epoch": 1.95, + "eval_LOC_f1": 0.8123064846714055, + "eval_LOC_number": 22273, + "eval_LOC_precision": 0.7896477171563017, + "eval_LOC_recall": 0.8363040452565887, + "eval_ORG_f1": 0.676284700925358, + "eval_ORG_number": 19008, + "eval_ORG_precision": 0.6857390081661349, + "eval_ORG_recall": 0.6670875420875421, + "eval_PER_f1": 0.8116203536629376, + "eval_PER_number": 27575, + "eval_PER_precision": 0.8076854013287843, + "eval_PER_recall": 0.8155938349954669, + "eval_loss": 0.1906026154756546, + "eval_overall_accuracy": 0.9409782628009117, + "eval_overall_f1": 0.7752790367557518, + "eval_overall_precision": 0.769352878083661, + "eval_overall_recall": 0.7812971999535262, + "eval_runtime": 72.0747, + "eval_samples_per_second": 525.663, + "eval_steps_per_second": 65.71, + "step": 344000 + }, + { + "epoch": 1.95, + "learning_rate": 1.7437310673788456e-05, + "loss": 0.1862, + "step": 344200 + }, + { + "epoch": 1.95, + "learning_rate": 1.7418389878131154e-05, + "loss": 0.1957, + "step": 344400 + }, + { + "epoch": 1.96, + "learning_rate": 1.7399469082473857e-05, + "loss": 0.1894, + "step": 344600 + }, + { + "epoch": 1.96, + "learning_rate": 1.738054828681656e-05, + "loss": 0.1967, + "step": 344800 + }, + { + "epoch": 1.96, + "learning_rate": 1.736162749115926e-05, + "loss": 0.1945, + "step": 345000 + }, + { + "epoch": 1.96, + "learning_rate": 1.734270669550196e-05, + "loss": 0.1982, + "step": 345200 + }, + { + "epoch": 1.96, + "learning_rate": 1.732378589984466e-05, + "loss": 0.1942, + "step": 345400 + }, + { + "epoch": 1.96, + "learning_rate": 1.7304865104187362e-05, + "loss": 0.1872, + "step": 345600 + }, + { + "epoch": 1.96, + "learning_rate": 1.7285944308530064e-05, + "loss": 0.1916, + "step": 345800 + }, + { + "epoch": 1.96, + "learning_rate": 1.7267023512872766e-05, + "loss": 0.1911, + "step": 346000 + }, + { + "epoch": 1.96, + "eval_LOC_f1": 0.8130536335992244, + "eval_LOC_number": 22273, + "eval_LOC_precision": 0.7983469643861699, + "eval_LOC_recall": 0.8283123063799219, + "eval_ORG_f1": 0.6727510474208096, + "eval_ORG_number": 19008, + "eval_ORG_precision": 0.6826428378012456, + "eval_ORG_recall": 0.663141835016835, + "eval_PER_f1": 0.8112686165440517, + "eval_PER_number": 27575, + "eval_PER_precision": 0.8029124489433493, + "eval_PER_recall": 0.8198005439709882, + "eval_loss": 0.19253480434417725, + "eval_overall_accuracy": 0.9407456584560564, + "eval_overall_f1": 0.7743983836634556, + "eval_overall_precision": 0.7695506890963587, + "eval_overall_recall": 0.7793075403741141, + "eval_runtime": 75.1501, + "eval_samples_per_second": 504.151, + "eval_steps_per_second": 63.021, + "step": 346000 + }, + { + "epoch": 1.97, + "learning_rate": 1.7248102717215465e-05, + "loss": 0.1889, + "step": 346200 + }, + { + "epoch": 1.97, + "learning_rate": 1.7229181921558164e-05, + "loss": 0.1948, + "step": 346400 + }, + { + "epoch": 1.97, + "learning_rate": 1.7210261125900867e-05, + "loss": 0.1842, + "step": 346600 + }, + { + "epoch": 1.97, + "learning_rate": 1.719134033024357e-05, + "loss": 0.1868, + "step": 346800 + }, + { + "epoch": 1.97, + "learning_rate": 1.717241953458627e-05, + "loss": 0.1863, + "step": 347000 + }, + { + "epoch": 1.97, + "learning_rate": 1.715349873892897e-05, + "loss": 0.1939, + "step": 347200 + }, + { + "epoch": 1.97, + "learning_rate": 1.713457794327167e-05, + "loss": 0.1911, + "step": 347400 + }, + { + "epoch": 1.97, + "learning_rate": 1.711565714761437e-05, + "loss": 0.1947, + "step": 347600 + }, + { + "epoch": 1.97, + "learning_rate": 1.7096736351957074e-05, + "loss": 0.1951, + "step": 347800 + }, + { + "epoch": 1.98, + "learning_rate": 1.7077815556299776e-05, + "loss": 0.1938, + "step": 348000 + }, + { + "epoch": 1.98, + "eval_LOC_f1": 0.810579981698549, + "eval_LOC_number": 22273, + "eval_LOC_precision": 0.7873862433862434, + "eval_LOC_recall": 0.8351816100211018, + "eval_ORG_f1": 0.6735963419821352, + "eval_ORG_number": 19008, + "eval_ORG_precision": 0.6808361994840929, + "eval_ORG_recall": 0.6665088383838383, + "eval_PER_f1": 0.8104082548227906, + "eval_PER_number": 27575, + "eval_PER_precision": 0.8021314387211368, + "eval_PER_recall": 0.8188576609247507, + "eval_loss": 0.19083885848522186, + "eval_overall_accuracy": 0.9407278900686021, + "eval_overall_f1": 0.773504549731038, + "eval_overall_precision": 0.7651137348507452, + "eval_overall_recall": 0.7820814453351923, + "eval_runtime": 71.9532, + "eval_samples_per_second": 526.551, + "eval_steps_per_second": 65.821, + "step": 348000 + }, + { + "epoch": 1.98, + "learning_rate": 1.7058894760642475e-05, + "loss": 0.1904, + "step": 348200 + }, + { + "epoch": 1.98, + "learning_rate": 1.7039973964985174e-05, + "loss": 0.1939, + "step": 348400 + }, + { + "epoch": 1.98, + "learning_rate": 1.7021053169327877e-05, + "loss": 0.1939, + "step": 348600 + }, + { + "epoch": 1.98, + "learning_rate": 1.700213237367058e-05, + "loss": 0.192, + "step": 348800 + }, + { + "epoch": 1.98, + "learning_rate": 1.698321157801328e-05, + "loss": 0.1957, + "step": 349000 + }, + { + "epoch": 1.98, + "learning_rate": 1.696429078235598e-05, + "loss": 0.1862, + "step": 349200 + }, + { + "epoch": 1.98, + "learning_rate": 1.6945369986698683e-05, + "loss": 0.1954, + "step": 349400 + }, + { + "epoch": 1.98, + "learning_rate": 1.692644919104138e-05, + "loss": 0.1957, + "step": 349600 + }, + { + "epoch": 1.99, + "learning_rate": 1.6907528395384084e-05, + "loss": 0.202, + "step": 349800 + }, + { + "epoch": 1.99, + "learning_rate": 1.6888607599726786e-05, + "loss": 0.1921, + "step": 350000 + }, + { + "epoch": 1.99, + "eval_LOC_f1": 0.8118726202838352, + "eval_LOC_number": 22273, + "eval_LOC_precision": 0.783432842052524, + "eval_LOC_recall": 0.842454990347057, + "eval_ORG_f1": 0.6759563420578574, + "eval_ORG_number": 19008, + "eval_ORG_precision": 0.6807897545357524, + "eval_ORG_recall": 0.6711910774410774, + "eval_PER_f1": 0.8102948327720044, + "eval_PER_number": 27575, + "eval_PER_precision": 0.804887648490053, + "eval_PER_recall": 0.8157751586582049, + "eval_loss": 0.18999794125556946, + "eval_overall_accuracy": 0.940966955645259, + "eval_overall_f1": 0.7744650345890534, + "eval_overall_precision": 0.7646908931326888, + "eval_overall_recall": 0.7844922737306843, + "eval_runtime": 74.8625, + "eval_samples_per_second": 506.088, + "eval_steps_per_second": 63.263, + "step": 350000 + }, + { + "epoch": 1.99, + "learning_rate": 1.6869686804069485e-05, + "loss": 0.1984, + "step": 350200 + }, + { + "epoch": 1.99, + "learning_rate": 1.6850766008412187e-05, + "loss": 0.1989, + "step": 350400 + }, + { + "epoch": 1.99, + "learning_rate": 1.6831845212754886e-05, + "loss": 0.2025, + "step": 350600 + }, + { + "epoch": 1.99, + "learning_rate": 1.681292441709759e-05, + "loss": 0.1854, + "step": 350800 + }, + { + "epoch": 1.99, + "learning_rate": 1.679400362144029e-05, + "loss": 0.1872, + "step": 351000 + }, + { + "epoch": 1.99, + "learning_rate": 1.677508282578299e-05, + "loss": 0.1849, + "step": 351200 + }, + { + "epoch": 1.99, + "learning_rate": 1.6756162030125692e-05, + "loss": 0.2013, + "step": 351400 + }, + { + "epoch": 2.0, + "learning_rate": 1.673724123446839e-05, + "loss": 0.1846, + "step": 351600 + }, + { + "epoch": 2.0, + "learning_rate": 1.6718320438811094e-05, + "loss": 0.186, + "step": 351800 + }, + { + "epoch": 2.0, + "learning_rate": 1.6699399643153796e-05, + "loss": 0.1963, + "step": 352000 + }, + { + "epoch": 2.0, + "eval_LOC_f1": 0.8128100443390842, + "eval_LOC_number": 22273, + "eval_LOC_precision": 0.7951470904015461, + "eval_LOC_recall": 0.8312755354016074, + "eval_ORG_f1": 0.6738273283480625, + "eval_ORG_number": 19008, + "eval_ORG_precision": 0.6973602746665166, + "eval_ORG_recall": 0.6518308080808081, + "eval_PER_f1": 0.8088841506751954, + "eval_PER_number": 27575, + "eval_PER_precision": 0.7929628984497474, + "eval_PER_recall": 0.8254578422484135, + "eval_loss": 0.1902003288269043, + "eval_overall_accuracy": 0.9409508025657551, + "eval_overall_f1": 0.7743429548455051, + "eval_overall_precision": 0.7693421448743495, + "eval_overall_recall": 0.7794092018124782, + "eval_runtime": 71.9803, + "eval_samples_per_second": 526.352, + "eval_steps_per_second": 65.796, + "step": 352000 + }, + { + "epoch": 2.0, + "learning_rate": 1.6680478847496495e-05, + "loss": 0.1942, + "step": 352200 + }, + { + "epoch": 2.0, + "learning_rate": 1.6661558051839197e-05, + "loss": 0.194, + "step": 352400 + }, + { + "epoch": 2.0, + "learning_rate": 1.6642637256181896e-05, + "loss": 0.1837, + "step": 352600 + }, + { + "epoch": 2.0, + "learning_rate": 1.66237164605246e-05, + "loss": 0.1824, + "step": 352800 + }, + { + "epoch": 2.0, + "learning_rate": 1.66047956648673e-05, + "loss": 0.1868, + "step": 353000 + }, + { + "epoch": 2.0, + "learning_rate": 1.658587486921e-05, + "loss": 0.1834, + "step": 353200 + }, + { + "epoch": 2.01, + "learning_rate": 1.6566954073552702e-05, + "loss": 0.1838, + "step": 353400 + }, + { + "epoch": 2.01, + "learning_rate": 1.65480332778954e-05, + "loss": 0.1865, + "step": 353600 + }, + { + "epoch": 2.01, + "learning_rate": 1.6529112482238104e-05, + "loss": 0.1804, + "step": 353800 + }, + { + "epoch": 2.01, + "learning_rate": 1.6510191686580806e-05, + "loss": 0.1842, + "step": 354000 + }, + { + "epoch": 2.01, + "eval_LOC_f1": 0.8120323780354408, + "eval_LOC_number": 22273, + "eval_LOC_precision": 0.7918675598412767, + "eval_LOC_recall": 0.8332510214160643, + "eval_ORG_f1": 0.6776749980121392, + "eval_ORG_number": 19008, + "eval_ORG_precision": 0.6828695048341434, + "eval_ORG_recall": 0.6725589225589226, + "eval_PER_f1": 0.8108966353873052, + "eval_PER_number": 27575, + "eval_PER_precision": 0.7980266161030936, + "eval_PER_recall": 0.8241885766092475, + "eval_loss": 0.19240854680538177, + "eval_overall_accuracy": 0.9409911852645148, + "eval_overall_f1": 0.7752360333493437, + "eval_overall_precision": 0.765462859407959, + "eval_overall_recall": 0.785261996049727, + "eval_runtime": 71.8926, + "eval_samples_per_second": 526.994, + "eval_steps_per_second": 65.876, + "step": 354000 + }, + { + "epoch": 2.01, + "learning_rate": 1.6491270890923505e-05, + "loss": 0.1822, + "step": 354200 + }, + { + "epoch": 2.01, + "learning_rate": 1.6472350095266207e-05, + "loss": 0.1808, + "step": 354400 + }, + { + "epoch": 2.01, + "learning_rate": 1.645342929960891e-05, + "loss": 0.1853, + "step": 354600 + }, + { + "epoch": 2.01, + "learning_rate": 1.643450850395161e-05, + "loss": 0.1862, + "step": 354800 + }, + { + "epoch": 2.02, + "learning_rate": 1.641558770829431e-05, + "loss": 0.1827, + "step": 355000 + }, + { + "epoch": 2.02, + "learning_rate": 1.639666691263701e-05, + "loss": 0.1779, + "step": 355200 + }, + { + "epoch": 2.02, + "learning_rate": 1.6377746116979712e-05, + "loss": 0.1778, + "step": 355400 + }, + { + "epoch": 2.02, + "learning_rate": 1.6358825321322414e-05, + "loss": 0.1905, + "step": 355600 + }, + { + "epoch": 2.02, + "learning_rate": 1.6339904525665113e-05, + "loss": 0.1825, + "step": 355800 + }, + { + "epoch": 2.02, + "learning_rate": 1.6320983730007816e-05, + "loss": 0.1797, + "step": 356000 + }, + { + "epoch": 2.02, + "eval_LOC_f1": 0.8144481840779604, + "eval_LOC_number": 22273, + "eval_LOC_precision": 0.7984128353316656, + "eval_LOC_recall": 0.8311408431733489, + "eval_ORG_f1": 0.6731438359830845, + "eval_ORG_number": 19008, + "eval_ORG_precision": 0.6851367549308053, + "eval_ORG_recall": 0.6615635521885522, + "eval_PER_f1": 0.8114039032287077, + "eval_PER_number": 27575, + "eval_PER_precision": 0.8049607423269094, + "eval_PER_recall": 0.8179510426110608, + "eval_loss": 0.1914784014225006, + "eval_overall_accuracy": 0.9411543313675036, + "eval_overall_f1": 0.775083805340423, + "eval_overall_precision": 0.7711615871190339, + "eval_overall_recall": 0.7790461252468921, + "eval_runtime": 72.2679, + "eval_samples_per_second": 524.258, + "eval_steps_per_second": 65.534, + "step": 356000 + }, + { + "epoch": 2.02, + "learning_rate": 1.6302062934350515e-05, + "loss": 0.1827, + "step": 356200 + }, + { + "epoch": 2.02, + "learning_rate": 1.6283142138693217e-05, + "loss": 0.1906, + "step": 356400 + }, + { + "epoch": 2.02, + "learning_rate": 1.626422134303592e-05, + "loss": 0.1826, + "step": 356600 + }, + { + "epoch": 2.03, + "learning_rate": 1.624530054737862e-05, + "loss": 0.1825, + "step": 356800 + }, + { + "epoch": 2.03, + "learning_rate": 1.622637975172132e-05, + "loss": 0.1748, + "step": 357000 + }, + { + "epoch": 2.03, + "learning_rate": 1.620745895606402e-05, + "loss": 0.1915, + "step": 357200 + }, + { + "epoch": 2.03, + "learning_rate": 1.6188538160406722e-05, + "loss": 0.183, + "step": 357400 + }, + { + "epoch": 2.03, + "learning_rate": 1.6169617364749424e-05, + "loss": 0.1723, + "step": 357600 + }, + { + "epoch": 2.03, + "learning_rate": 1.6150696569092123e-05, + "loss": 0.1925, + "step": 357800 + }, + { + "epoch": 2.03, + "learning_rate": 1.6131775773434826e-05, + "loss": 0.1833, + "step": 358000 + }, + { + "epoch": 2.03, + "eval_LOC_f1": 0.8131013575853434, + "eval_LOC_number": 22273, + "eval_LOC_precision": 0.7960342380317433, + "eval_LOC_recall": 0.8309163561262515, + "eval_ORG_f1": 0.6726537943421642, + "eval_ORG_number": 19008, + "eval_ORG_precision": 0.675454883030078, + "eval_ORG_recall": 0.6698758417508418, + "eval_PER_f1": 0.8109182217244432, + "eval_PER_number": 27575, + "eval_PER_precision": 0.8078971996256569, + "eval_PER_recall": 0.8139619220308251, + "eval_loss": 0.19205814599990845, + "eval_overall_accuracy": 0.9407876564627663, + "eval_overall_f1": 0.7739047701422825, + "eval_overall_precision": 0.7682235768867519, + "eval_overall_recall": 0.7796706169397002, + "eval_runtime": 74.4026, + "eval_samples_per_second": 509.216, + "eval_steps_per_second": 63.654, + "step": 358000 + }, + { + "epoch": 2.03, + "learning_rate": 1.6112854977777525e-05, + "loss": 0.1796, + "step": 358200 + }, + { + "epoch": 2.03, + "learning_rate": 1.6093934182120227e-05, + "loss": 0.18, + "step": 358400 + }, + { + "epoch": 2.04, + "learning_rate": 1.607501338646293e-05, + "loss": 0.1867, + "step": 358600 + }, + { + "epoch": 2.04, + "learning_rate": 1.6056092590805628e-05, + "loss": 0.1775, + "step": 358800 + }, + { + "epoch": 2.04, + "learning_rate": 1.603717179514833e-05, + "loss": 0.1949, + "step": 359000 + }, + { + "epoch": 2.04, + "learning_rate": 1.601825099949103e-05, + "loss": 0.1863, + "step": 359200 + }, + { + "epoch": 2.04, + "learning_rate": 1.5999330203833732e-05, + "loss": 0.1808, + "step": 359400 + }, + { + "epoch": 2.04, + "learning_rate": 1.5980409408176434e-05, + "loss": 0.1839, + "step": 359600 + }, + { + "epoch": 2.04, + "learning_rate": 1.5961488612519137e-05, + "loss": 0.1785, + "step": 359800 + }, + { + "epoch": 2.04, + "learning_rate": 1.5942567816861835e-05, + "loss": 0.184, + "step": 360000 + }, + { + "epoch": 2.04, + "eval_LOC_f1": 0.812438323720971, + "eval_LOC_number": 22273, + "eval_LOC_precision": 0.7940672153635117, + "eval_LOC_recall": 0.8316796120863826, + "eval_ORG_f1": 0.6760306163555794, + "eval_ORG_number": 19008, + "eval_ORG_precision": 0.6905140725297635, + "eval_ORG_recall": 0.6621422558922558, + "eval_PER_f1": 0.8111905613156953, + "eval_PER_number": 27575, + "eval_PER_precision": 0.7998942358540455, + "eval_PER_recall": 0.8228105167724388, + "eval_loss": 0.1923096776008606, + "eval_overall_accuracy": 0.9411091027448928, + "eval_overall_f1": 0.7753357929324954, + "eval_overall_precision": 0.7694364988558352, + "eval_overall_recall": 0.7813262460787731, + "eval_runtime": 75.1309, + "eval_samples_per_second": 504.28, + "eval_steps_per_second": 63.037, + "step": 360000 + }, + { + "epoch": 2.04, + "learning_rate": 1.5923647021204534e-05, + "loss": 0.1884, + "step": 360200 + }, + { + "epoch": 2.05, + "learning_rate": 1.5904726225547237e-05, + "loss": 0.1794, + "step": 360400 + }, + { + "epoch": 2.05, + "learning_rate": 1.588580542988994e-05, + "loss": 0.183, + "step": 360600 + }, + { + "epoch": 2.05, + "learning_rate": 1.586688463423264e-05, + "loss": 0.1747, + "step": 360800 + }, + { + "epoch": 2.05, + "learning_rate": 1.584796383857534e-05, + "loss": 0.1906, + "step": 361000 + }, + { + "epoch": 2.05, + "learning_rate": 1.582904304291804e-05, + "loss": 0.1912, + "step": 361200 + }, + { + "epoch": 2.05, + "learning_rate": 1.581012224726074e-05, + "loss": 0.1853, + "step": 361400 + }, + { + "epoch": 2.05, + "learning_rate": 1.5791201451603444e-05, + "loss": 0.1842, + "step": 361600 + }, + { + "epoch": 2.05, + "learning_rate": 1.5772280655946146e-05, + "loss": 0.1823, + "step": 361800 + }, + { + "epoch": 2.05, + "learning_rate": 1.5753359860288845e-05, + "loss": 0.1867, + "step": 362000 + }, + { + "epoch": 2.05, + "eval_LOC_f1": 0.8145790509029022, + "eval_LOC_number": 22273, + "eval_LOC_precision": 0.8022202873313017, + "eval_LOC_recall": 0.8273245633726934, + "eval_ORG_f1": 0.6754754326370128, + "eval_ORG_number": 19008, + "eval_ORG_precision": 0.6870341150225802, + "eval_ORG_recall": 0.6642992424242424, + "eval_PER_f1": 0.8104498133961894, + "eval_PER_number": 27575, + "eval_PER_precision": 0.7983184408639977, + "eval_PER_recall": 0.8229555757026292, + "eval_loss": 0.1915552020072937, + "eval_overall_accuracy": 0.9412787100796831, + "eval_overall_f1": 0.7753965563257857, + "eval_overall_precision": 0.7702902185596561, + "eval_overall_recall": 0.7805710468223539, + "eval_runtime": 73.7001, + "eval_samples_per_second": 514.07, + "eval_steps_per_second": 64.26, + "step": 362000 + }, + { + "epoch": 2.06, + "learning_rate": 1.5734439064631544e-05, + "loss": 0.1839, + "step": 362200 + }, + { + "epoch": 2.06, + "learning_rate": 1.5715518268974247e-05, + "loss": 0.176, + "step": 362400 + }, + { + "epoch": 2.06, + "learning_rate": 1.569659747331695e-05, + "loss": 0.1854, + "step": 362600 + }, + { + "epoch": 2.06, + "learning_rate": 1.567767667765965e-05, + "loss": 0.1846, + "step": 362800 + }, + { + "epoch": 2.06, + "learning_rate": 1.565875588200235e-05, + "loss": 0.1865, + "step": 363000 + }, + { + "epoch": 2.06, + "learning_rate": 1.563983508634505e-05, + "loss": 0.1857, + "step": 363200 + }, + { + "epoch": 2.06, + "learning_rate": 1.562091429068775e-05, + "loss": 0.184, + "step": 363400 + }, + { + "epoch": 2.06, + "learning_rate": 1.5601993495030454e-05, + "loss": 0.1911, + "step": 363600 + }, + { + "epoch": 2.07, + "learning_rate": 1.5583072699373156e-05, + "loss": 0.1893, + "step": 363800 + }, + { + "epoch": 2.07, + "learning_rate": 1.5564151903715855e-05, + "loss": 0.1776, + "step": 364000 + }, + { + "epoch": 2.07, + "eval_LOC_f1": 0.8136382609459786, + "eval_LOC_number": 22273, + "eval_LOC_precision": 0.7966101694915254, + "eval_LOC_recall": 0.8314102276298657, + "eval_ORG_f1": 0.67585042092207, + "eval_ORG_number": 19008, + "eval_ORG_precision": 0.678051363516018, + "eval_ORG_recall": 0.6736637205387206, + "eval_PER_f1": 0.8096475352554339, + "eval_PER_number": 27575, + "eval_PER_precision": 0.8002266931141967, + "eval_PER_recall": 0.8192928377153218, + "eval_loss": 0.19348245859146118, + "eval_overall_accuracy": 0.9410073383440186, + "eval_overall_f1": 0.7745350850099483, + "eval_overall_precision": 0.7662407799553743, + "eval_overall_recall": 0.7830109213430928, + "eval_runtime": 72.1169, + "eval_samples_per_second": 525.355, + "eval_steps_per_second": 65.671, + "step": 364000 + }, + { + "epoch": 2.07, + "learning_rate": 1.5545231108058554e-05, + "loss": 0.1824, + "step": 364200 + }, + { + "epoch": 2.07, + "learning_rate": 1.5526310312401256e-05, + "loss": 0.1861, + "step": 364400 + }, + { + "epoch": 2.07, + "learning_rate": 1.550738951674396e-05, + "loss": 0.1804, + "step": 364600 + }, + { + "epoch": 2.07, + "learning_rate": 1.548846872108666e-05, + "loss": 0.1862, + "step": 364800 + }, + { + "epoch": 2.07, + "learning_rate": 1.5469547925429364e-05, + "loss": 0.1769, + "step": 365000 + }, + { + "epoch": 2.07, + "learning_rate": 1.545062712977206e-05, + "loss": 0.1796, + "step": 365200 + }, + { + "epoch": 2.07, + "learning_rate": 1.543170633411476e-05, + "loss": 0.1891, + "step": 365400 + }, + { + "epoch": 2.08, + "learning_rate": 1.5412785538457464e-05, + "loss": 0.1835, + "step": 365600 + }, + { + "epoch": 2.08, + "learning_rate": 1.5393864742800166e-05, + "loss": 0.1909, + "step": 365800 + }, + { + "epoch": 2.08, + "learning_rate": 1.537494394714287e-05, + "loss": 0.1835, + "step": 366000 + }, + { + "epoch": 2.08, + "eval_LOC_f1": 0.8119707127734651, + "eval_LOC_number": 22273, + "eval_LOC_precision": 0.7979281348879546, + "eval_LOC_recall": 0.8265164100031428, + "eval_ORG_f1": 0.6757626532620549, + "eval_ORG_number": 19008, + "eval_ORG_precision": 0.6833414017535366, + "eval_ORG_recall": 0.6683501683501684, + "eval_PER_f1": 0.8121236161696297, + "eval_PER_number": 27575, + "eval_PER_precision": 0.8075805931078998, + "eval_PER_recall": 0.8167180417044424, + "eval_loss": 0.1908581256866455, + "eval_overall_accuracy": 0.9411317170561981, + "eval_overall_f1": 0.7750298038365665, + "eval_overall_precision": 0.7711685286632446, + "eval_overall_recall": 0.7789299407459045, + "eval_runtime": 72.5503, + "eval_samples_per_second": 522.217, + "eval_steps_per_second": 65.279, + "step": 366000 + }, + { + "epoch": 2.08, + "learning_rate": 1.5356023151485564e-05, + "loss": 0.1934, + "step": 366200 + }, + { + "epoch": 2.08, + "learning_rate": 1.5337102355828266e-05, + "loss": 0.1781, + "step": 366400 + }, + { + "epoch": 2.08, + "learning_rate": 1.531818156017097e-05, + "loss": 0.1922, + "step": 366600 + }, + { + "epoch": 2.08, + "learning_rate": 1.529926076451367e-05, + "loss": 0.1863, + "step": 366800 + }, + { + "epoch": 2.08, + "learning_rate": 1.5280339968856373e-05, + "loss": 0.1756, + "step": 367000 + }, + { + "epoch": 2.08, + "learning_rate": 1.5261419173199072e-05, + "loss": 0.1778, + "step": 367200 + }, + { + "epoch": 2.09, + "learning_rate": 1.5242498377541773e-05, + "loss": 0.1852, + "step": 367400 + }, + { + "epoch": 2.09, + "learning_rate": 1.5223577581884474e-05, + "loss": 0.1883, + "step": 367600 + }, + { + "epoch": 2.09, + "learning_rate": 1.5204656786227176e-05, + "loss": 0.1802, + "step": 367800 + }, + { + "epoch": 2.09, + "learning_rate": 1.5185735990569877e-05, + "loss": 0.1847, + "step": 368000 + }, + { + "epoch": 2.09, + "eval_LOC_f1": 0.8145645974994516, + "eval_LOC_number": 22273, + "eval_LOC_precision": 0.7963288587725693, + "eval_LOC_recall": 0.8336550981008396, + "eval_ORG_f1": 0.6789466648843737, + "eval_ORG_number": 19008, + "eval_ORG_precision": 0.6902212317225634, + "eval_ORG_recall": 0.6680345117845118, + "eval_PER_f1": 0.8115221628193896, + "eval_PER_number": 27575, + "eval_PER_precision": 0.8004020880361173, + "eval_PER_recall": 0.8229555757026292, + "eval_loss": 0.19038023054599762, + "eval_overall_accuracy": 0.9412431733047747, + "eval_overall_f1": 0.7768244050618331, + "eval_overall_precision": 0.770116747067051, + "eval_overall_recall": 0.7836499360985244, + "eval_runtime": 72.2205, + "eval_samples_per_second": 524.601, + "eval_steps_per_second": 65.577, + "step": 368000 + }, + { + "epoch": 2.09, + "learning_rate": 1.5166815194912576e-05, + "loss": 0.178, + "step": 368200 + }, + { + "epoch": 2.09, + "learning_rate": 1.5147894399255278e-05, + "loss": 0.1821, + "step": 368400 + }, + { + "epoch": 2.09, + "learning_rate": 1.5128973603597979e-05, + "loss": 0.1737, + "step": 368600 + }, + { + "epoch": 2.09, + "learning_rate": 1.5110052807940681e-05, + "loss": 0.1829, + "step": 368800 + }, + { + "epoch": 2.09, + "learning_rate": 1.5091132012283382e-05, + "loss": 0.1875, + "step": 369000 + }, + { + "epoch": 2.1, + "learning_rate": 1.507221121662608e-05, + "loss": 0.1829, + "step": 369200 + }, + { + "epoch": 2.1, + "learning_rate": 1.5053290420968783e-05, + "loss": 0.1808, + "step": 369400 + }, + { + "epoch": 2.1, + "learning_rate": 1.5034369625311483e-05, + "loss": 0.1875, + "step": 369600 + }, + { + "epoch": 2.1, + "learning_rate": 1.5015448829654186e-05, + "loss": 0.1835, + "step": 369800 + }, + { + "epoch": 2.1, + "learning_rate": 1.4996528033996886e-05, + "loss": 0.1844, + "step": 370000 + }, + { + "epoch": 2.1, + "eval_LOC_f1": 0.8128290415666234, + "eval_LOC_number": 22273, + "eval_LOC_precision": 0.7978723404255319, + "eval_LOC_recall": 0.8283572037893413, + "eval_ORG_f1": 0.6743608239160055, + "eval_ORG_number": 19008, + "eval_ORG_precision": 0.6850675785702655, + "eval_ORG_recall": 0.6639835858585859, + "eval_PER_f1": 0.8112113052749518, + "eval_PER_number": 27575, + "eval_PER_precision": 0.8084792161947987, + "eval_PER_recall": 0.8139619220308251, + "eval_loss": 0.19126270711421967, + "eval_overall_accuracy": 0.9410977955892401, + "eval_overall_f1": 0.7746679694568089, + "eval_overall_precision": 0.7721363747853814, + "eval_overall_recall": 0.7772162193563379, + "eval_runtime": 71.7337, + "eval_samples_per_second": 528.162, + "eval_steps_per_second": 66.022, + "step": 370000 + }, + { + "epoch": 2.1, + "learning_rate": 1.4977607238339585e-05, + "loss": 0.1872, + "step": 370200 + }, + { + "epoch": 2.1, + "learning_rate": 1.4958686442682288e-05, + "loss": 0.1823, + "step": 370400 + }, + { + "epoch": 2.1, + "learning_rate": 1.4939765647024988e-05, + "loss": 0.1819, + "step": 370600 + }, + { + "epoch": 2.1, + "learning_rate": 1.492084485136769e-05, + "loss": 0.1932, + "step": 370800 + }, + { + "epoch": 2.11, + "learning_rate": 1.4901924055710393e-05, + "loss": 0.1891, + "step": 371000 + }, + { + "epoch": 2.11, + "learning_rate": 1.488300326005309e-05, + "loss": 0.1882, + "step": 371200 + }, + { + "epoch": 2.11, + "learning_rate": 1.4864082464395793e-05, + "loss": 0.1874, + "step": 371400 + }, + { + "epoch": 2.11, + "learning_rate": 1.4845161668738495e-05, + "loss": 0.1888, + "step": 371600 + }, + { + "epoch": 2.11, + "learning_rate": 1.4826240873081196e-05, + "loss": 0.1733, + "step": 371800 + }, + { + "epoch": 2.11, + "learning_rate": 1.4807320077423898e-05, + "loss": 0.1851, + "step": 372000 + }, + { + "epoch": 2.11, + "eval_LOC_f1": 0.8134017453843793, + "eval_LOC_number": 22273, + "eval_LOC_precision": 0.7949256417948828, + "eval_LOC_recall": 0.8327571499124501, + "eval_ORG_f1": 0.6764956809214034, + "eval_ORG_number": 19008, + "eval_ORG_precision": 0.6857837837837838, + "eval_ORG_recall": 0.6674558080808081, + "eval_PER_f1": 0.8120281831907398, + "eval_PER_number": 27575, + "eval_PER_precision": 0.8049962581518834, + "eval_PER_recall": 0.819184043517679, + "eval_loss": 0.1906796395778656, + "eval_overall_accuracy": 0.9410864884335874, + "eval_overall_f1": 0.7758414414414415, + "eval_overall_precision": 0.7700804074741752, + "eval_overall_recall": 0.7816893226443592, + "eval_runtime": 73.5127, + "eval_samples_per_second": 515.38, + "eval_steps_per_second": 64.424, + "step": 372000 + }, + { + "epoch": 2.11, + "learning_rate": 1.4788399281766599e-05, + "loss": 0.1799, + "step": 372200 + }, + { + "epoch": 2.11, + "learning_rate": 1.4769478486109298e-05, + "loss": 0.1848, + "step": 372400 + }, + { + "epoch": 2.11, + "learning_rate": 1.4750557690452e-05, + "loss": 0.1907, + "step": 372600 + }, + { + "epoch": 2.12, + "learning_rate": 1.47316368947947e-05, + "loss": 0.184, + "step": 372800 + }, + { + "epoch": 2.12, + "learning_rate": 1.4712716099137403e-05, + "loss": 0.1749, + "step": 373000 + }, + { + "epoch": 2.12, + "learning_rate": 1.4693795303480104e-05, + "loss": 0.1778, + "step": 373200 + }, + { + "epoch": 2.12, + "learning_rate": 1.4674874507822803e-05, + "loss": 0.1826, + "step": 373400 + }, + { + "epoch": 2.12, + "learning_rate": 1.4655953712165505e-05, + "loss": 0.1871, + "step": 373600 + }, + { + "epoch": 2.12, + "learning_rate": 1.4637032916508206e-05, + "loss": 0.175, + "step": 373800 + }, + { + "epoch": 2.12, + "learning_rate": 1.4618112120850908e-05, + "loss": 0.1831, + "step": 374000 + }, + { + "epoch": 2.12, + "eval_LOC_f1": 0.8129653010579476, + "eval_LOC_number": 22273, + "eval_LOC_precision": 0.7890555672934714, + "eval_LOC_recall": 0.8383693260898846, + "eval_ORG_f1": 0.6780045950299733, + "eval_ORG_number": 19008, + "eval_ORG_precision": 0.6806829630415187, + "eval_ORG_recall": 0.6753472222222222, + "eval_PER_f1": 0.8116452511456556, + "eval_PER_number": 27575, + "eval_PER_precision": 0.8044887780548629, + "eval_PER_recall": 0.8189301903898458, + "eval_loss": 0.19094158709049225, + "eval_overall_accuracy": 0.9411817916026601, + "eval_overall_f1": 0.7757906059519541, + "eval_overall_precision": 0.7662407570048446, + "eval_overall_recall": 0.7855815034274428, + "eval_runtime": 73.231, + "eval_samples_per_second": 517.363, + "eval_steps_per_second": 64.672, + "step": 374000 + }, + { + "epoch": 2.12, + "learning_rate": 1.4599191325193609e-05, + "loss": 0.1871, + "step": 374200 + }, + { + "epoch": 2.13, + "learning_rate": 1.4580270529536307e-05, + "loss": 0.1824, + "step": 374400 + }, + { + "epoch": 2.13, + "learning_rate": 1.456134973387901e-05, + "loss": 0.1745, + "step": 374600 + }, + { + "epoch": 2.13, + "learning_rate": 1.454242893822171e-05, + "loss": 0.171, + "step": 374800 + }, + { + "epoch": 2.13, + "learning_rate": 1.4523508142564413e-05, + "loss": 0.1845, + "step": 375000 + }, + { + "epoch": 2.13, + "learning_rate": 1.4504587346907113e-05, + "loss": 0.1874, + "step": 375200 + }, + { + "epoch": 2.13, + "learning_rate": 1.4485666551249812e-05, + "loss": 0.1789, + "step": 375400 + }, + { + "epoch": 2.13, + "learning_rate": 1.4466745755592515e-05, + "loss": 0.1801, + "step": 375600 + }, + { + "epoch": 2.13, + "learning_rate": 1.4447824959935215e-05, + "loss": 0.1852, + "step": 375800 + }, + { + "epoch": 2.13, + "learning_rate": 1.4428904164277918e-05, + "loss": 0.1808, + "step": 376000 + }, + { + "epoch": 2.13, + "eval_LOC_f1": 0.8125232784873913, + "eval_LOC_number": 22273, + "eval_LOC_precision": 0.7934531450577663, + "eval_LOC_recall": 0.8325326628653527, + "eval_ORG_f1": 0.6775759183132211, + "eval_ORG_number": 19008, + "eval_ORG_precision": 0.6832112103546023, + "eval_ORG_recall": 0.6720328282828283, + "eval_PER_f1": 0.8127365120553574, + "eval_PER_number": 27575, + "eval_PER_precision": 0.8077295032057022, + "eval_PER_recall": 0.8178059836808703, + "eval_loss": 0.19113685190677643, + "eval_overall_accuracy": 0.9412092518378167, + "eval_overall_f1": 0.7759611644891314, + "eval_overall_precision": 0.7696967965021576, + "eval_overall_recall": 0.7823283373997909, + "eval_runtime": 73.5005, + "eval_samples_per_second": 515.466, + "eval_steps_per_second": 64.435, + "step": 376000 + }, + { + "epoch": 2.14, + "learning_rate": 1.440998336862062e-05, + "loss": 0.1862, + "step": 376200 + }, + { + "epoch": 2.14, + "learning_rate": 1.4391062572963317e-05, + "loss": 0.188, + "step": 376400 + }, + { + "epoch": 2.14, + "learning_rate": 1.437214177730602e-05, + "loss": 0.1853, + "step": 376600 + }, + { + "epoch": 2.14, + "learning_rate": 1.4353220981648722e-05, + "loss": 0.1929, + "step": 376800 + }, + { + "epoch": 2.14, + "learning_rate": 1.4334300185991423e-05, + "loss": 0.1754, + "step": 377000 + }, + { + "epoch": 2.14, + "learning_rate": 1.4315379390334125e-05, + "loss": 0.1754, + "step": 377200 + }, + { + "epoch": 2.14, + "learning_rate": 1.4296458594676822e-05, + "loss": 0.1915, + "step": 377400 + }, + { + "epoch": 2.14, + "learning_rate": 1.4277537799019525e-05, + "loss": 0.1778, + "step": 377600 + }, + { + "epoch": 2.14, + "learning_rate": 1.4258617003362227e-05, + "loss": 0.184, + "step": 377800 + }, + { + "epoch": 2.15, + "learning_rate": 1.4239696207704928e-05, + "loss": 0.1786, + "step": 378000 + }, + { + "epoch": 2.15, + "eval_LOC_f1": 0.8147729986755542, + "eval_LOC_number": 22273, + "eval_LOC_precision": 0.7888916918937101, + "eval_LOC_recall": 0.8424100929376375, + "eval_ORG_f1": 0.678851725046089, + "eval_ORG_number": 19008, + "eval_ORG_precision": 0.6796751397531906, + "eval_ORG_recall": 0.678030303030303, + "eval_PER_f1": 0.8101402373247033, + "eval_PER_number": 27575, + "eval_PER_precision": 0.8033517561062578, + "eval_PER_recall": 0.8170444242973708, + "eval_loss": 0.19270993769168854, + "eval_overall_accuracy": 0.9411042568210416, + "eval_overall_f1": 0.7759708407627806, + "eval_overall_precision": 0.7653656538260514, + "eval_overall_recall": 0.7868740560009295, + "eval_runtime": 73.0569, + "eval_samples_per_second": 518.596, + "eval_steps_per_second": 64.826, + "step": 378000 + }, + { + "epoch": 2.15, + "learning_rate": 1.422077541204763e-05, + "loss": 0.1837, + "step": 378200 + }, + { + "epoch": 2.15, + "learning_rate": 1.4201854616390329e-05, + "loss": 0.1853, + "step": 378400 + }, + { + "epoch": 2.15, + "learning_rate": 1.418293382073303e-05, + "loss": 0.1871, + "step": 378600 + }, + { + "epoch": 2.15, + "learning_rate": 1.4164013025075732e-05, + "loss": 0.18, + "step": 378800 + }, + { + "epoch": 2.15, + "learning_rate": 1.4145092229418433e-05, + "loss": 0.1843, + "step": 379000 + }, + { + "epoch": 2.15, + "learning_rate": 1.4126171433761135e-05, + "loss": 0.1765, + "step": 379200 + }, + { + "epoch": 2.15, + "learning_rate": 1.4107250638103834e-05, + "loss": 0.1814, + "step": 379400 + }, + { + "epoch": 2.15, + "learning_rate": 1.4088329842446534e-05, + "loss": 0.1872, + "step": 379600 + }, + { + "epoch": 2.16, + "learning_rate": 1.4069409046789237e-05, + "loss": 0.1826, + "step": 379800 + }, + { + "epoch": 2.16, + "learning_rate": 1.4050488251131937e-05, + "loss": 0.1846, + "step": 380000 + }, + { + "epoch": 2.16, + "eval_LOC_f1": 0.8148655516029961, + "eval_LOC_number": 22273, + "eval_LOC_precision": 0.8022274427912642, + "eval_LOC_recall": 0.8279082296951465, + "eval_ORG_f1": 0.6780476327416276, + "eval_ORG_number": 19008, + "eval_ORG_precision": 0.6739073948968456, + "eval_ORG_recall": 0.6822390572390572, + "eval_PER_f1": 0.8107353730542136, + "eval_PER_number": 27575, + "eval_PER_precision": 0.8001412678792159, + "eval_PER_recall": 0.8216137805983681, + "eval_loss": 0.19201837480068207, + "eval_overall_accuracy": 0.9409604944134574, + "eval_overall_f1": 0.7756671449067432, + "eval_overall_precision": 0.7663869358131096, + "eval_overall_recall": 0.7851748576739863, + "eval_runtime": 72.4608, + "eval_samples_per_second": 522.862, + "eval_steps_per_second": 65.359, + "step": 380000 + }, + { + "epoch": 2.16, + "learning_rate": 1.403156745547464e-05, + "loss": 0.1822, + "step": 380200 + }, + { + "epoch": 2.16, + "learning_rate": 1.4012646659817339e-05, + "loss": 0.1808, + "step": 380400 + }, + { + "epoch": 2.16, + "learning_rate": 1.399372586416004e-05, + "loss": 0.18, + "step": 380600 + }, + { + "epoch": 2.16, + "learning_rate": 1.3974805068502742e-05, + "loss": 0.1856, + "step": 380800 + }, + { + "epoch": 2.16, + "learning_rate": 1.3955884272845442e-05, + "loss": 0.1829, + "step": 381000 + }, + { + "epoch": 2.16, + "learning_rate": 1.3936963477188145e-05, + "loss": 0.1702, + "step": 381200 + }, + { + "epoch": 2.16, + "learning_rate": 1.3918042681530844e-05, + "loss": 0.1851, + "step": 381400 + }, + { + "epoch": 2.17, + "learning_rate": 1.3899121885873544e-05, + "loss": 0.1822, + "step": 381600 + }, + { + "epoch": 2.17, + "learning_rate": 1.3880201090216247e-05, + "loss": 0.1841, + "step": 381800 + }, + { + "epoch": 2.17, + "learning_rate": 1.3861280294558947e-05, + "loss": 0.1768, + "step": 382000 + }, + { + "epoch": 2.17, + "eval_LOC_f1": 0.8145944874082888, + "eval_LOC_number": 22273, + "eval_LOC_precision": 0.7997750281214848, + "eval_LOC_recall": 0.8299735105284425, + "eval_ORG_f1": 0.6795692129446771, + "eval_ORG_number": 19008, + "eval_ORG_precision": 0.6802678053666509, + "eval_ORG_recall": 0.6788720538720538, + "eval_PER_f1": 0.8119313891186307, + "eval_PER_number": 27575, + "eval_PER_precision": 0.8051563670078095, + "eval_PER_recall": 0.8188213961922031, + "eval_loss": 0.1915133148431778, + "eval_overall_accuracy": 0.9414693164178285, + "eval_overall_f1": 0.7766329452734887, + "eval_overall_precision": 0.7696004335054045, + "eval_overall_recall": 0.7837951667247589, + "eval_runtime": 71.9615, + "eval_samples_per_second": 526.49, + "eval_steps_per_second": 65.813, + "step": 382000 + }, + { + "epoch": 2.17, + "learning_rate": 1.384235949890165e-05, + "loss": 0.1764, + "step": 382200 + }, + { + "epoch": 2.17, + "learning_rate": 1.3823438703244349e-05, + "loss": 0.1791, + "step": 382400 + }, + { + "epoch": 2.17, + "learning_rate": 1.380451790758705e-05, + "loss": 0.1736, + "step": 382600 + }, + { + "epoch": 2.17, + "learning_rate": 1.3785597111929752e-05, + "loss": 0.1751, + "step": 382800 + }, + { + "epoch": 2.17, + "learning_rate": 1.3766676316272454e-05, + "loss": 0.1859, + "step": 383000 + }, + { + "epoch": 2.18, + "learning_rate": 1.3747755520615155e-05, + "loss": 0.1757, + "step": 383200 + }, + { + "epoch": 2.18, + "learning_rate": 1.3728834724957854e-05, + "loss": 0.1783, + "step": 383400 + }, + { + "epoch": 2.18, + "learning_rate": 1.3709913929300556e-05, + "loss": 0.1813, + "step": 383600 + }, + { + "epoch": 2.18, + "learning_rate": 1.3690993133643257e-05, + "loss": 0.1833, + "step": 383800 + }, + { + "epoch": 2.18, + "learning_rate": 1.3672072337985959e-05, + "loss": 0.1857, + "step": 384000 + }, + { + "epoch": 2.18, + "eval_LOC_f1": 0.8150068092957871, + "eval_LOC_number": 22273, + "eval_LOC_precision": 0.7978325377370662, + "eval_LOC_recall": 0.8329367395501279, + "eval_ORG_f1": 0.6772554002541296, + "eval_ORG_number": 19008, + "eval_ORG_precision": 0.6815856777493606, + "eval_ORG_recall": 0.672979797979798, + "eval_PER_f1": 0.8116566952334753, + "eval_PER_number": 27575, + "eval_PER_precision": 0.804301381569577, + "eval_PER_recall": 0.8191477787851315, + "eval_loss": 0.19086118042469025, + "eval_overall_accuracy": 0.9413352458579466, + "eval_overall_f1": 0.7762177877246134, + "eval_overall_precision": 0.7693031880750303, + "eval_overall_recall": 0.7832578134076914, + "eval_runtime": 72.172, + "eval_samples_per_second": 524.954, + "eval_steps_per_second": 65.621, + "step": 384000 + }, + { + "epoch": 2.18, + "learning_rate": 1.365315154232866e-05, + "loss": 0.1785, + "step": 384200 + }, + { + "epoch": 2.18, + "learning_rate": 1.3634230746671359e-05, + "loss": 0.1721, + "step": 384400 + }, + { + "epoch": 2.18, + "learning_rate": 1.3615309951014061e-05, + "loss": 0.1822, + "step": 384600 + }, + { + "epoch": 2.18, + "learning_rate": 1.3596389155356761e-05, + "loss": 0.1836, + "step": 384800 + }, + { + "epoch": 2.19, + "learning_rate": 1.3577468359699464e-05, + "loss": 0.1891, + "step": 385000 + }, + { + "epoch": 2.19, + "learning_rate": 1.3558547564042164e-05, + "loss": 0.1782, + "step": 385200 + }, + { + "epoch": 2.19, + "learning_rate": 1.3539626768384863e-05, + "loss": 0.1732, + "step": 385400 + }, + { + "epoch": 2.19, + "learning_rate": 1.3520705972727566e-05, + "loss": 0.1862, + "step": 385600 + }, + { + "epoch": 2.19, + "learning_rate": 1.3501785177070266e-05, + "loss": 0.1779, + "step": 385800 + }, + { + "epoch": 2.19, + "learning_rate": 1.3482864381412969e-05, + "loss": 0.1728, + "step": 386000 + }, + { + "epoch": 2.19, + "eval_LOC_f1": 0.8160270632427563, + "eval_LOC_number": 22273, + "eval_LOC_precision": 0.7988817204301075, + "eval_LOC_recall": 0.8339244825573564, + "eval_ORG_f1": 0.6774749913763366, + "eval_ORG_number": 19008, + "eval_ORG_precision": 0.6834412977140104, + "eval_ORG_recall": 0.6716119528619529, + "eval_PER_f1": 0.8108021045849888, + "eval_PER_number": 27575, + "eval_PER_precision": 0.8003745186022683, + "eval_PER_recall": 0.8215049864007253, + "eval_loss": 0.19203363358974457, + "eval_overall_accuracy": 0.941378859172607, + "eval_overall_f1": 0.7763861727826987, + "eval_overall_precision": 0.7687806128260622, + "eval_overall_recall": 0.7841437202277216, + "eval_runtime": 72.119, + "eval_samples_per_second": 525.34, + "eval_steps_per_second": 65.669, + "step": 386000 + }, + { + "epoch": 2.19, + "learning_rate": 1.346394358575567e-05, + "loss": 0.1804, + "step": 386200 + }, + { + "epoch": 2.19, + "learning_rate": 1.3445022790098368e-05, + "loss": 0.1782, + "step": 386400 + }, + { + "epoch": 2.19, + "learning_rate": 1.342610199444107e-05, + "loss": 0.181, + "step": 386600 + }, + { + "epoch": 2.2, + "learning_rate": 1.3407181198783771e-05, + "loss": 0.1849, + "step": 386800 + }, + { + "epoch": 2.2, + "learning_rate": 1.3388260403126474e-05, + "loss": 0.1814, + "step": 387000 + }, + { + "epoch": 2.2, + "learning_rate": 1.3369339607469174e-05, + "loss": 0.1828, + "step": 387200 + }, + { + "epoch": 2.2, + "learning_rate": 1.3350418811811873e-05, + "loss": 0.1808, + "step": 387400 + }, + { + "epoch": 2.2, + "learning_rate": 1.3331498016154576e-05, + "loss": 0.1799, + "step": 387600 + }, + { + "epoch": 2.2, + "learning_rate": 1.3312577220497276e-05, + "loss": 0.1739, + "step": 387800 + }, + { + "epoch": 2.2, + "learning_rate": 1.3293656424839979e-05, + "loss": 0.178, + "step": 388000 + }, + { + "epoch": 2.2, + "eval_LOC_f1": 0.8151458342416182, + "eval_LOC_number": 22273, + "eval_LOC_precision": 0.7922121943985424, + "eval_LOC_recall": 0.839446863915952, + "eval_ORG_f1": 0.67711062396407, + "eval_ORG_number": 19008, + "eval_ORG_precision": 0.688335688661811, + "eval_ORG_recall": 0.6662457912457912, + "eval_PER_f1": 0.8122330552001298, + "eval_PER_number": 27575, + "eval_PER_precision": 0.8073009959160278, + "eval_PER_recall": 0.8172257479601088, + "eval_loss": 0.1918538510799408, + "eval_overall_accuracy": 0.9416114635174623, + "eval_overall_f1": 0.7767729103762369, + "eval_overall_precision": 0.7709009769284683, + "eval_overall_recall": 0.7827349831532474, + "eval_runtime": 74.8422, + "eval_samples_per_second": 506.225, + "eval_steps_per_second": 63.28, + "step": 388000 + }, + { + "epoch": 2.2, + "learning_rate": 1.3274735629182681e-05, + "loss": 0.1754, + "step": 388200 + }, + { + "epoch": 2.2, + "learning_rate": 1.3255814833525378e-05, + "loss": 0.191, + "step": 388400 + }, + { + "epoch": 2.21, + "learning_rate": 1.323689403786808e-05, + "loss": 0.1791, + "step": 388600 + }, + { + "epoch": 2.21, + "learning_rate": 1.3217973242210783e-05, + "loss": 0.1789, + "step": 388800 + }, + { + "epoch": 2.21, + "learning_rate": 1.3199052446553484e-05, + "loss": 0.1786, + "step": 389000 + }, + { + "epoch": 2.21, + "learning_rate": 1.3180131650896186e-05, + "loss": 0.1846, + "step": 389200 + }, + { + "epoch": 2.21, + "learning_rate": 1.3161210855238885e-05, + "loss": 0.1898, + "step": 389400 + }, + { + "epoch": 2.21, + "learning_rate": 1.3142290059581586e-05, + "loss": 0.1791, + "step": 389600 + }, + { + "epoch": 2.21, + "learning_rate": 1.3123369263924288e-05, + "loss": 0.1826, + "step": 389800 + }, + { + "epoch": 2.21, + "learning_rate": 1.3104448468266988e-05, + "loss": 0.1729, + "step": 390000 + }, + { + "epoch": 2.21, + "eval_LOC_f1": 0.8139244115905152, + "eval_LOC_number": 22273, + "eval_LOC_precision": 0.7957196774746955, + "eval_LOC_recall": 0.8329816369595474, + "eval_ORG_f1": 0.6809079485602759, + "eval_ORG_number": 19008, + "eval_ORG_precision": 0.6922533297091601, + "eval_ORG_recall": 0.6699284511784511, + "eval_PER_f1": 0.8121730689723354, + "eval_PER_number": 27575, + "eval_PER_precision": 0.8052110065583119, + "eval_PER_recall": 0.8192565729827742, + "eval_loss": 0.19083505868911743, + "eval_overall_accuracy": 0.9414919307291338, + "eval_overall_f1": 0.7773313230849138, + "eval_overall_precision": 0.7722562242894205, + "eval_overall_recall": 0.7824735680260253, + "eval_runtime": 72.3192, + "eval_samples_per_second": 523.886, + "eval_steps_per_second": 65.487, + "step": 390000 + }, + { + "epoch": 2.21, + "learning_rate": 1.308552767260969e-05, + "loss": 0.1778, + "step": 390200 + }, + { + "epoch": 2.22, + "learning_rate": 1.306660687695239e-05, + "loss": 0.1926, + "step": 390400 + }, + { + "epoch": 2.22, + "learning_rate": 1.304768608129509e-05, + "loss": 0.1747, + "step": 390600 + }, + { + "epoch": 2.22, + "learning_rate": 1.3028765285637793e-05, + "loss": 0.1818, + "step": 390800 + }, + { + "epoch": 2.22, + "learning_rate": 1.3009844489980493e-05, + "loss": 0.1829, + "step": 391000 + }, + { + "epoch": 2.22, + "learning_rate": 1.2990923694323196e-05, + "loss": 0.1885, + "step": 391200 + }, + { + "epoch": 2.22, + "learning_rate": 1.2972002898665895e-05, + "loss": 0.1808, + "step": 391400 + }, + { + "epoch": 2.22, + "learning_rate": 1.2953082103008595e-05, + "loss": 0.1865, + "step": 391600 + }, + { + "epoch": 2.22, + "learning_rate": 1.2934161307351298e-05, + "loss": 0.1811, + "step": 391800 + }, + { + "epoch": 2.23, + "learning_rate": 1.2915240511693998e-05, + "loss": 0.1787, + "step": 392000 + }, + { + "epoch": 2.23, + "eval_LOC_f1": 0.8160069785192454, + "eval_LOC_number": 22273, + "eval_LOC_precision": 0.7933593418709185, + "eval_LOC_recall": 0.8399856328289858, + "eval_ORG_f1": 0.679913218329982, + "eval_ORG_number": 19008, + "eval_ORG_precision": 0.6838939748775814, + "eval_ORG_recall": 0.6759785353535354, + "eval_PER_f1": 0.8118971637563196, + "eval_PER_number": 27575, + "eval_PER_precision": 0.8028578520015601, + "eval_PER_recall": 0.8211423390752494, + "eval_loss": 0.1907501369714737, + "eval_overall_accuracy": 0.9415630042789508, + "eval_overall_f1": 0.7774709708884092, + "eval_overall_precision": 0.7680132628625679, + "eval_overall_recall": 0.7871645172533984, + "eval_runtime": 72.1661, + "eval_samples_per_second": 524.997, + "eval_steps_per_second": 65.626, + "step": 392000 + }, + { + "epoch": 2.23, + "learning_rate": 1.28963197160367e-05, + "loss": 0.1888, + "step": 392200 + }, + { + "epoch": 2.23, + "learning_rate": 1.28773989203794e-05, + "loss": 0.1868, + "step": 392400 + }, + { + "epoch": 2.23, + "learning_rate": 1.28584781247221e-05, + "loss": 0.1834, + "step": 392600 + }, + { + "epoch": 2.23, + "learning_rate": 1.2839557329064803e-05, + "loss": 0.1788, + "step": 392800 + }, + { + "epoch": 2.23, + "learning_rate": 1.2820636533407503e-05, + "loss": 0.1857, + "step": 393000 + }, + { + "epoch": 2.23, + "learning_rate": 1.2801715737750206e-05, + "loss": 0.1824, + "step": 393200 + }, + { + "epoch": 2.23, + "learning_rate": 1.2782794942092905e-05, + "loss": 0.1775, + "step": 393400 + }, + { + "epoch": 2.23, + "learning_rate": 1.2763874146435605e-05, + "loss": 0.1792, + "step": 393600 + }, + { + "epoch": 2.24, + "learning_rate": 1.2744953350778308e-05, + "loss": 0.1851, + "step": 393800 + }, + { + "epoch": 2.24, + "learning_rate": 1.272603255512101e-05, + "loss": 0.1897, + "step": 394000 + }, + { + "epoch": 2.24, + "eval_LOC_f1": 0.8152650238928894, + "eval_LOC_number": 22273, + "eval_LOC_precision": 0.8000259313683119, + "eval_LOC_recall": 0.8310959457639294, + "eval_ORG_f1": 0.6795407372375923, + "eval_ORG_number": 19008, + "eval_ORG_precision": 0.678738256442555, + "eval_ORG_recall": 0.6803451178451179, + "eval_PER_f1": 0.8122895019706198, + "eval_PER_number": 27575, + "eval_PER_precision": 0.8026553372278279, + "eval_PER_recall": 0.822157751586582, + "eval_loss": 0.18967469036579132, + "eval_overall_accuracy": 0.9415888492061569, + "eval_overall_f1": 0.7769864744565375, + "eval_overall_precision": 0.7682719064114941, + "eval_overall_recall": 0.7859010108051586, + "eval_runtime": 71.9895, + "eval_samples_per_second": 526.285, + "eval_steps_per_second": 65.787, + "step": 394000 + }, + { + "epoch": 2.24, + "learning_rate": 1.270711175946371e-05, + "loss": 0.1813, + "step": 394200 + }, + { + "epoch": 2.24, + "learning_rate": 1.268819096380641e-05, + "loss": 0.1876, + "step": 394400 + }, + { + "epoch": 2.24, + "learning_rate": 1.2669270168149112e-05, + "loss": 0.1802, + "step": 394600 + }, + { + "epoch": 2.24, + "learning_rate": 1.2650349372491813e-05, + "loss": 0.1782, + "step": 394800 + }, + { + "epoch": 2.24, + "learning_rate": 1.2631428576834515e-05, + "loss": 0.177, + "step": 395000 + }, + { + "epoch": 2.24, + "learning_rate": 1.2612507781177215e-05, + "loss": 0.1796, + "step": 395200 + }, + { + "epoch": 2.24, + "learning_rate": 1.2593586985519914e-05, + "loss": 0.1815, + "step": 395400 + }, + { + "epoch": 2.25, + "learning_rate": 1.2574666189862617e-05, + "loss": 0.1722, + "step": 395600 + }, + { + "epoch": 2.25, + "learning_rate": 1.2555745394205317e-05, + "loss": 0.1801, + "step": 395800 + }, + { + "epoch": 2.25, + "learning_rate": 1.253682459854802e-05, + "loss": 0.1869, + "step": 396000 + }, + { + "epoch": 2.25, + "eval_LOC_f1": 0.8146057582917652, + "eval_LOC_number": 22273, + "eval_LOC_precision": 0.7944100704075101, + "eval_LOC_recall": 0.8358550711623939, + "eval_ORG_f1": 0.677409949905908, + "eval_ORG_number": 19008, + "eval_ORG_precision": 0.6826024250841302, + "eval_ORG_recall": 0.6722958754208754, + "eval_PER_f1": 0.8137214509628303, + "eval_PER_number": 27575, + "eval_PER_precision": 0.804, + "eval_PER_recall": 0.8236808703535812, + "eval_loss": 0.19103264808654785, + "eval_overall_accuracy": 0.9414596245701262, + "eval_overall_f1": 0.7770820467895048, + "eval_overall_precision": 0.7685282504332017, + "eval_overall_recall": 0.7858283954920413, + "eval_runtime": 71.8229, + "eval_samples_per_second": 527.506, + "eval_steps_per_second": 65.94, + "step": 396000 + }, + { + "epoch": 2.25, + "learning_rate": 1.251790380289072e-05, + "loss": 0.1857, + "step": 396200 + }, + { + "epoch": 2.25, + "learning_rate": 1.2498983007233421e-05, + "loss": 0.1755, + "step": 396400 + }, + { + "epoch": 2.25, + "learning_rate": 1.2480062211576122e-05, + "loss": 0.1778, + "step": 396600 + }, + { + "epoch": 2.25, + "learning_rate": 1.2461141415918822e-05, + "loss": 0.1874, + "step": 396800 + }, + { + "epoch": 2.25, + "learning_rate": 1.2442220620261525e-05, + "loss": 0.1838, + "step": 397000 + }, + { + "epoch": 2.25, + "learning_rate": 1.2423299824604225e-05, + "loss": 0.1888, + "step": 397200 + }, + { + "epoch": 2.26, + "learning_rate": 1.2404379028946926e-05, + "loss": 0.1806, + "step": 397400 + }, + { + "epoch": 2.26, + "learning_rate": 1.2385458233289627e-05, + "loss": 0.1716, + "step": 397600 + }, + { + "epoch": 2.26, + "learning_rate": 1.2366537437632327e-05, + "loss": 0.1859, + "step": 397800 + }, + { + "epoch": 2.26, + "learning_rate": 1.234761664197503e-05, + "loss": 0.1831, + "step": 398000 + }, + { + "epoch": 2.26, + "eval_LOC_f1": 0.8156910764655718, + "eval_LOC_number": 22273, + "eval_LOC_precision": 0.7950554134697357, + "eval_LOC_recall": 0.8374264804920756, + "eval_ORG_f1": 0.6819181931447658, + "eval_ORG_number": 19008, + "eval_ORG_precision": 0.6872562115949773, + "eval_ORG_recall": 0.6766624579124579, + "eval_PER_f1": 0.8134087816059022, + "eval_PER_number": 27575, + "eval_PER_precision": 0.8034242456401005, + "eval_PER_recall": 0.8236446056210336, + "eval_loss": 0.1896025389432907, + "eval_overall_accuracy": 0.941790762699955, + "eval_overall_f1": 0.7785498923187366, + "eval_overall_precision": 0.769774572710238, + "eval_overall_recall": 0.7875275938189845, + "eval_runtime": 73.0441, + "eval_samples_per_second": 518.686, + "eval_steps_per_second": 64.838, + "step": 398000 + }, + { + "epoch": 2.26, + "learning_rate": 1.232869584631773e-05, + "loss": 0.1877, + "step": 398200 + }, + { + "epoch": 2.26, + "learning_rate": 1.2309775050660431e-05, + "loss": 0.1738, + "step": 398400 + }, + { + "epoch": 2.26, + "learning_rate": 1.2290854255003132e-05, + "loss": 0.1749, + "step": 398600 + }, + { + "epoch": 2.26, + "learning_rate": 1.2271933459345832e-05, + "loss": 0.1764, + "step": 398800 + }, + { + "epoch": 2.26, + "learning_rate": 1.2253012663688535e-05, + "loss": 0.1794, + "step": 399000 + }, + { + "epoch": 2.27, + "learning_rate": 1.2234091868031235e-05, + "loss": 0.1832, + "step": 399200 + }, + { + "epoch": 2.27, + "learning_rate": 1.2215171072373936e-05, + "loss": 0.1775, + "step": 399400 + }, + { + "epoch": 2.27, + "learning_rate": 1.2196250276716637e-05, + "loss": 0.1764, + "step": 399600 + }, + { + "epoch": 2.27, + "learning_rate": 1.2177329481059339e-05, + "loss": 0.1796, + "step": 399800 + }, + { + "epoch": 2.27, + "learning_rate": 1.215840868540204e-05, + "loss": 0.1846, + "step": 400000 + }, + { + "epoch": 2.27, + "eval_LOC_f1": 0.814241893076249, + "eval_LOC_number": 22273, + "eval_LOC_precision": 0.7951812384987376, + "eval_LOC_recall": 0.8342387644232928, + "eval_ORG_f1": 0.6771942710246053, + "eval_ORG_number": 19008, + "eval_ORG_precision": 0.6753165219211049, + "eval_ORG_recall": 0.6790824915824916, + "eval_PER_f1": 0.8136782270830334, + "eval_PER_number": 27575, + "eval_PER_precision": 0.807259877931256, + "eval_PER_recall": 0.8201994560290118, + "eval_loss": 0.1892879605293274, + "eval_overall_accuracy": 0.9415630042789508, + "eval_overall_f1": 0.7765259698322259, + "eval_overall_precision": 0.7674827654685239, + "eval_overall_recall": 0.785784826304171, + "eval_runtime": 72.4387, + "eval_samples_per_second": 523.021, + "eval_steps_per_second": 65.379, + "step": 400000 + }, + { + "epoch": 2.27, + "learning_rate": 1.213948788974474e-05, + "loss": 0.1904, + "step": 400200 + }, + { + "epoch": 2.27, + "learning_rate": 1.212056709408744e-05, + "loss": 0.1825, + "step": 400400 + }, + { + "epoch": 2.27, + "learning_rate": 1.2101646298430141e-05, + "loss": 0.1773, + "step": 400600 + }, + { + "epoch": 2.28, + "learning_rate": 1.2082725502772844e-05, + "loss": 0.1821, + "step": 400800 + }, + { + "epoch": 2.28, + "learning_rate": 1.2063804707115544e-05, + "loss": 0.1811, + "step": 401000 + }, + { + "epoch": 2.28, + "learning_rate": 1.2044883911458245e-05, + "loss": 0.1883, + "step": 401200 + }, + { + "epoch": 2.28, + "learning_rate": 1.2025963115800946e-05, + "loss": 0.1771, + "step": 401400 + }, + { + "epoch": 2.28, + "learning_rate": 1.2007042320143646e-05, + "loss": 0.1786, + "step": 401600 + }, + { + "epoch": 2.28, + "learning_rate": 1.1988121524486349e-05, + "loss": 0.1817, + "step": 401800 + }, + { + "epoch": 2.28, + "learning_rate": 1.196920072882905e-05, + "loss": 0.1805, + "step": 402000 + }, + { + "epoch": 2.28, + "eval_LOC_f1": 0.8137100994243852, + "eval_LOC_number": 22273, + "eval_LOC_precision": 0.7909796108685516, + "eval_LOC_recall": 0.8377856597674315, + "eval_ORG_f1": 0.6796064734582877, + "eval_ORG_number": 19008, + "eval_ORG_precision": 0.6889561597924212, + "eval_ORG_recall": 0.6705071548821548, + "eval_PER_f1": 0.8135958996493121, + "eval_PER_number": 27575, + "eval_PER_precision": 0.8069925080271138, + "eval_PER_recall": 0.8203082502266545, + "eval_loss": 0.19045652449131012, + "eval_overall_accuracy": 0.9416583074480235, + "eval_overall_f1": 0.7774723693299563, + "eval_overall_precision": 0.7704649172846548, + "eval_overall_recall": 0.7846084582316719, + "eval_runtime": 75.5974, + "eval_samples_per_second": 501.168, + "eval_steps_per_second": 62.648, + "step": 402000 + }, + { + "epoch": 2.28, + "learning_rate": 1.195027993317175e-05, + "loss": 0.1796, + "step": 402200 + }, + { + "epoch": 2.28, + "learning_rate": 1.193135913751445e-05, + "loss": 0.1767, + "step": 402400 + }, + { + "epoch": 2.29, + "learning_rate": 1.1912438341857151e-05, + "loss": 0.1787, + "step": 402600 + }, + { + "epoch": 2.29, + "learning_rate": 1.1893517546199854e-05, + "loss": 0.1801, + "step": 402800 + }, + { + "epoch": 2.29, + "learning_rate": 1.1874596750542554e-05, + "loss": 0.1873, + "step": 403000 + }, + { + "epoch": 2.29, + "learning_rate": 1.1855675954885255e-05, + "loss": 0.179, + "step": 403200 + }, + { + "epoch": 2.29, + "learning_rate": 1.1836755159227957e-05, + "loss": 0.1827, + "step": 403400 + }, + { + "epoch": 2.29, + "learning_rate": 1.1817834363570656e-05, + "loss": 0.1776, + "step": 403600 + }, + { + "epoch": 2.29, + "learning_rate": 1.1798913567913359e-05, + "loss": 0.178, + "step": 403800 + }, + { + "epoch": 2.29, + "learning_rate": 1.177999277225606e-05, + "loss": 0.1869, + "step": 404000 + }, + { + "epoch": 2.29, + "eval_LOC_f1": 0.8164024056861673, + "eval_LOC_number": 22273, + "eval_LOC_precision": 0.7958809483199727, + "eval_LOC_recall": 0.8380101468145288, + "eval_ORG_f1": 0.6811150456614926, + "eval_ORG_number": 19008, + "eval_ORG_precision": 0.6895046089159614, + "eval_ORG_recall": 0.6729271885521886, + "eval_PER_f1": 0.8126663310076963, + "eval_PER_number": 27575, + "eval_PER_precision": 0.8059706815993152, + "eval_PER_recall": 0.8194741613780598, + "eval_loss": 0.1897078901529312, + "eval_overall_accuracy": 0.9417552259250465, + "eval_overall_f1": 0.778323349844488, + "eval_overall_precision": 0.7717447173043975, + "eval_overall_recall": 0.7850151039851284, + "eval_runtime": 75.8171, + "eval_samples_per_second": 499.716, + "eval_steps_per_second": 62.466, + "step": 404000 + }, + { + "epoch": 2.29, + "learning_rate": 1.176107197659876e-05, + "loss": 0.176, + "step": 404200 + }, + { + "epoch": 2.3, + "learning_rate": 1.1742151180941462e-05, + "loss": 0.1786, + "step": 404400 + }, + { + "epoch": 2.3, + "learning_rate": 1.1723230385284161e-05, + "loss": 0.1803, + "step": 404600 + }, + { + "epoch": 2.3, + "learning_rate": 1.1704309589626864e-05, + "loss": 0.1755, + "step": 404800 + }, + { + "epoch": 2.3, + "learning_rate": 1.1685388793969564e-05, + "loss": 0.1836, + "step": 405000 + }, + { + "epoch": 2.3, + "learning_rate": 1.1666467998312265e-05, + "loss": 0.189, + "step": 405200 + }, + { + "epoch": 2.3, + "learning_rate": 1.1647547202654967e-05, + "loss": 0.1826, + "step": 405400 + }, + { + "epoch": 2.3, + "learning_rate": 1.1628626406997668e-05, + "loss": 0.1773, + "step": 405600 + }, + { + "epoch": 2.3, + "learning_rate": 1.1609705611340368e-05, + "loss": 0.1767, + "step": 405800 + }, + { + "epoch": 2.3, + "learning_rate": 1.159078481568307e-05, + "loss": 0.1797, + "step": 406000 + }, + { + "epoch": 2.3, + "eval_LOC_f1": 0.8150796938708764, + "eval_LOC_number": 22273, + "eval_LOC_precision": 0.7923272573124205, + "eval_LOC_recall": 0.8391774794594352, + "eval_ORG_f1": 0.680857865991741, + "eval_ORG_number": 19008, + "eval_ORG_precision": 0.6896961191774167, + "eval_ORG_recall": 0.672243265993266, + "eval_PER_f1": 0.8130359587807162, + "eval_PER_number": 27575, + "eval_PER_precision": 0.8078258690437833, + "eval_PER_recall": 0.8183136899365367, + "eval_loss": 0.19212552905082703, + "eval_overall_accuracy": 0.9416260012890157, + "eval_overall_f1": 0.7779937511698559, + "eval_overall_precision": 0.7713633119200571, + "eval_overall_recall": 0.7847391657952829, + "eval_runtime": 72.1498, + "eval_samples_per_second": 525.116, + "eval_steps_per_second": 65.641, + "step": 406000 + }, + { + "epoch": 2.31, + "learning_rate": 1.157186402002577e-05, + "loss": 0.1806, + "step": 406200 + }, + { + "epoch": 2.31, + "learning_rate": 1.1552943224368472e-05, + "loss": 0.1801, + "step": 406400 + }, + { + "epoch": 2.31, + "learning_rate": 1.1534022428711173e-05, + "loss": 0.186, + "step": 406600 + }, + { + "epoch": 2.31, + "learning_rate": 1.1515101633053873e-05, + "loss": 0.1833, + "step": 406800 + }, + { + "epoch": 2.31, + "learning_rate": 1.1496180837396576e-05, + "loss": 0.1725, + "step": 407000 + }, + { + "epoch": 2.31, + "learning_rate": 1.1477260041739275e-05, + "loss": 0.1822, + "step": 407200 + }, + { + "epoch": 2.31, + "learning_rate": 1.1458339246081977e-05, + "loss": 0.1786, + "step": 407400 + }, + { + "epoch": 2.31, + "learning_rate": 1.1439418450424678e-05, + "loss": 0.1771, + "step": 407600 + }, + { + "epoch": 2.31, + "learning_rate": 1.1420497654767378e-05, + "loss": 0.1792, + "step": 407800 + }, + { + "epoch": 2.32, + "learning_rate": 1.140157685911008e-05, + "loss": 0.1858, + "step": 408000 + }, + { + "epoch": 2.32, + "eval_LOC_f1": 0.8161969035331481, + "eval_LOC_number": 22273, + "eval_LOC_precision": 0.8021153929515801, + "eval_LOC_recall": 0.8307816638979931, + "eval_ORG_f1": 0.6829203727234223, + "eval_ORG_number": 19008, + "eval_ORG_precision": 0.6872868712702472, + "eval_ORG_recall": 0.6786090067340067, + "eval_PER_f1": 0.8119246260069045, + "eval_PER_number": 27575, + "eval_PER_precision": 0.8051781320209693, + "eval_PER_recall": 0.8187851314596555, + "eval_loss": 0.18959172070026398, + "eval_overall_accuracy": 0.9418909117928788, + "eval_overall_f1": 0.7781942422189226, + "eval_overall_precision": 0.77250350611065, + "eval_overall_recall": 0.7839694434762403, + "eval_runtime": 71.7147, + "eval_samples_per_second": 528.302, + "eval_steps_per_second": 66.039, + "step": 408000 + }, + { + "epoch": 2.32, + "learning_rate": 1.138265606345278e-05, + "loss": 0.1795, + "step": 408200 + }, + { + "epoch": 2.32, + "learning_rate": 1.1363735267795482e-05, + "loss": 0.18, + "step": 408400 + }, + { + "epoch": 2.32, + "learning_rate": 1.1344814472138184e-05, + "loss": 0.1914, + "step": 408600 + }, + { + "epoch": 2.32, + "learning_rate": 1.1325893676480883e-05, + "loss": 0.184, + "step": 408800 + }, + { + "epoch": 2.32, + "learning_rate": 1.1306972880823586e-05, + "loss": 0.1815, + "step": 409000 + }, + { + "epoch": 2.32, + "learning_rate": 1.1288052085166286e-05, + "loss": 0.1818, + "step": 409200 + }, + { + "epoch": 2.32, + "learning_rate": 1.1269131289508987e-05, + "loss": 0.1912, + "step": 409400 + }, + { + "epoch": 2.32, + "learning_rate": 1.125021049385169e-05, + "loss": 0.1755, + "step": 409600 + }, + { + "epoch": 2.33, + "learning_rate": 1.1231289698194388e-05, + "loss": 0.1897, + "step": 409800 + }, + { + "epoch": 2.33, + "learning_rate": 1.121236890253709e-05, + "loss": 0.1789, + "step": 410000 + }, + { + "epoch": 2.33, + "eval_LOC_f1": 0.8152435029482419, + "eval_LOC_number": 22273, + "eval_LOC_precision": 0.7936811668154952, + "eval_LOC_recall": 0.8380101468145288, + "eval_ORG_f1": 0.6810049282194129, + "eval_ORG_number": 19008, + "eval_ORG_precision": 0.6936381492797905, + "eval_ORG_recall": 0.6688236531986532, + "eval_PER_f1": 0.812981613562238, + "eval_PER_number": 27575, + "eval_PER_precision": 0.8035923052396642, + "eval_PER_recall": 0.8225929283771533, + "eval_loss": 0.19078943133354187, + "eval_overall_accuracy": 0.9420702109753714, + "eval_overall_f1": 0.7782592421973972, + "eval_overall_precision": 0.7715064505080489, + "eval_overall_recall": 0.7851312884861159, + "eval_runtime": 71.7234, + "eval_samples_per_second": 528.237, + "eval_steps_per_second": 66.031, + "step": 410000 + }, + { + "epoch": 2.33, + "learning_rate": 1.1193448106879791e-05, + "loss": 0.1888, + "step": 410200 + }, + { + "epoch": 2.33, + "learning_rate": 1.1174527311222492e-05, + "loss": 0.1757, + "step": 410400 + }, + { + "epoch": 2.33, + "learning_rate": 1.1155606515565194e-05, + "loss": 0.1823, + "step": 410600 + }, + { + "epoch": 2.33, + "learning_rate": 1.1136685719907893e-05, + "loss": 0.1681, + "step": 410800 + }, + { + "epoch": 2.33, + "learning_rate": 1.1117764924250595e-05, + "loss": 0.1748, + "step": 411000 + }, + { + "epoch": 2.33, + "learning_rate": 1.1098844128593296e-05, + "loss": 0.1817, + "step": 411200 + }, + { + "epoch": 2.34, + "learning_rate": 1.1079923332935997e-05, + "loss": 0.1848, + "step": 411400 + }, + { + "epoch": 2.34, + "learning_rate": 1.1061002537278699e-05, + "loss": 0.1781, + "step": 411600 + }, + { + "epoch": 2.34, + "learning_rate": 1.10420817416214e-05, + "loss": 0.182, + "step": 411800 + }, + { + "epoch": 2.34, + "learning_rate": 1.10231609459641e-05, + "loss": 0.1784, + "step": 412000 + }, + { + "epoch": 2.34, + "eval_LOC_f1": 0.8146338270904688, + "eval_LOC_number": 22273, + "eval_LOC_precision": 0.7922854960536366, + "eval_LOC_recall": 0.8382795312710457, + "eval_ORG_f1": 0.6817082569294959, + "eval_ORG_number": 19008, + "eval_ORG_precision": 0.6887922238333065, + "eval_ORG_recall": 0.6747685185185185, + "eval_PER_f1": 0.8120026045434815, + "eval_PER_number": 27575, + "eval_PER_precision": 0.8099808754014362, + "eval_PER_recall": 0.8140344514959202, + "eval_loss": 0.19108271598815918, + "eval_overall_accuracy": 0.9418359913225657, + "eval_overall_f1": 0.7775375479258554, + "eval_overall_precision": 0.7717310443490701, + "eval_overall_recall": 0.7834320901591728, + "eval_runtime": 73.6762, + "eval_samples_per_second": 514.237, + "eval_steps_per_second": 64.281, + "step": 412000 + }, + { + "epoch": 2.34, + "learning_rate": 1.1004240150306801e-05, + "loss": 0.1749, + "step": 412200 + }, + { + "epoch": 2.34, + "learning_rate": 1.0985319354649502e-05, + "loss": 0.1723, + "step": 412400 + }, + { + "epoch": 2.34, + "learning_rate": 1.0966398558992204e-05, + "loss": 0.1884, + "step": 412600 + }, + { + "epoch": 2.34, + "learning_rate": 1.0947477763334905e-05, + "loss": 0.1762, + "step": 412800 + }, + { + "epoch": 2.34, + "learning_rate": 1.0928556967677605e-05, + "loss": 0.1857, + "step": 413000 + }, + { + "epoch": 2.35, + "learning_rate": 1.0909636172020306e-05, + "loss": 0.1775, + "step": 413200 + }, + { + "epoch": 2.35, + "learning_rate": 1.0890715376363007e-05, + "loss": 0.1848, + "step": 413400 + }, + { + "epoch": 2.35, + "learning_rate": 1.0871794580705709e-05, + "loss": 0.177, + "step": 413600 + }, + { + "epoch": 2.35, + "learning_rate": 1.085287378504841e-05, + "loss": 0.1748, + "step": 413800 + }, + { + "epoch": 2.35, + "learning_rate": 1.083395298939111e-05, + "loss": 0.1802, + "step": 414000 + }, + { + "epoch": 2.35, + "eval_LOC_f1": 0.8167596543047806, + "eval_LOC_number": 22273, + "eval_LOC_precision": 0.7965602594742233, + "eval_LOC_recall": 0.8380101468145288, + "eval_ORG_f1": 0.6829897599913995, + "eval_ORG_number": 19008, + "eval_ORG_precision": 0.6981702291334688, + "eval_ORG_recall": 0.6684553872053872, + "eval_PER_f1": 0.8134954123498465, + "eval_PER_number": 27575, + "eval_PER_precision": 0.8056405149726154, + "eval_PER_recall": 0.8215049864007253, + "eval_loss": 0.18926027417182922, + "eval_overall_accuracy": 0.9423690429461925, + "eval_overall_f1": 0.7795389776703581, + "eval_overall_precision": 0.7745487390500222, + "eval_overall_recall": 0.7845939351690484, + "eval_runtime": 73.3686, + "eval_samples_per_second": 516.392, + "eval_steps_per_second": 64.551, + "step": 414000 + }, + { + "epoch": 2.35, + "learning_rate": 1.0815032193733811e-05, + "loss": 0.18, + "step": 414200 + }, + { + "epoch": 2.35, + "learning_rate": 1.0796111398076513e-05, + "loss": 0.1801, + "step": 414400 + }, + { + "epoch": 2.35, + "learning_rate": 1.0777190602419214e-05, + "loss": 0.1791, + "step": 414600 + }, + { + "epoch": 2.35, + "learning_rate": 1.0758269806761915e-05, + "loss": 0.1833, + "step": 414800 + }, + { + "epoch": 2.36, + "learning_rate": 1.0739349011104615e-05, + "loss": 0.1797, + "step": 415000 + }, + { + "epoch": 2.36, + "learning_rate": 1.0720428215447316e-05, + "loss": 0.1913, + "step": 415200 + }, + { + "epoch": 2.36, + "learning_rate": 1.0701507419790018e-05, + "loss": 0.1781, + "step": 415400 + }, + { + "epoch": 2.36, + "learning_rate": 1.0682586624132719e-05, + "loss": 0.1805, + "step": 415600 + }, + { + "epoch": 2.36, + "learning_rate": 1.066366582847542e-05, + "loss": 0.1829, + "step": 415800 + }, + { + "epoch": 2.36, + "learning_rate": 1.064474503281812e-05, + "loss": 0.1765, + "step": 416000 + }, + { + "epoch": 2.36, + "eval_LOC_f1": 0.8169739099828413, + "eval_LOC_number": 22273, + "eval_LOC_precision": 0.8009057580332112, + "eval_LOC_recall": 0.8336999955102591, + "eval_ORG_f1": 0.6847129523758875, + "eval_ORG_number": 19008, + "eval_ORG_precision": 0.6922415183612022, + "eval_ORG_recall": 0.6773463804713805, + "eval_PER_f1": 0.8138698661077282, + "eval_PER_number": 27575, + "eval_PER_precision": 0.8110414866032843, + "eval_PER_recall": 0.8167180417044424, + "eval_loss": 0.18956801295280457, + "eval_overall_accuracy": 0.942165514144444, + "eval_overall_f1": 0.7797959655511242, + "eval_overall_precision": 0.775894294916034, + "eval_overall_recall": 0.7837370744742651, + "eval_runtime": 74.7662, + "eval_samples_per_second": 506.74, + "eval_steps_per_second": 63.344, + "step": 416000 + }, + { + "epoch": 2.36, + "learning_rate": 1.062582423716082e-05, + "loss": 0.1793, + "step": 416200 + }, + { + "epoch": 2.36, + "learning_rate": 1.0606903441503523e-05, + "loss": 0.1816, + "step": 416400 + }, + { + "epoch": 2.36, + "learning_rate": 1.0587982645846224e-05, + "loss": 0.179, + "step": 416600 + }, + { + "epoch": 2.37, + "learning_rate": 1.0569061850188924e-05, + "loss": 0.1802, + "step": 416800 + }, + { + "epoch": 2.37, + "learning_rate": 1.0550141054531627e-05, + "loss": 0.181, + "step": 417000 + }, + { + "epoch": 2.37, + "learning_rate": 1.0531220258874326e-05, + "loss": 0.1738, + "step": 417200 + }, + { + "epoch": 2.37, + "learning_rate": 1.0512299463217028e-05, + "loss": 0.1767, + "step": 417400 + }, + { + "epoch": 2.37, + "learning_rate": 1.0493378667559729e-05, + "loss": 0.1795, + "step": 417600 + }, + { + "epoch": 2.37, + "learning_rate": 1.047445787190243e-05, + "loss": 0.1788, + "step": 417800 + }, + { + "epoch": 2.37, + "learning_rate": 1.0455537076245132e-05, + "loss": 0.1863, + "step": 418000 + }, + { + "epoch": 2.37, + "eval_LOC_f1": 0.8157066449497062, + "eval_LOC_number": 22273, + "eval_LOC_precision": 0.8000863557858376, + "eval_LOC_recall": 0.8319489965428994, + "eval_ORG_f1": 0.6821626814688301, + "eval_ORG_number": 19008, + "eval_ORG_precision": 0.6922118717504333, + "eval_ORG_recall": 0.6724010942760943, + "eval_PER_f1": 0.8142562168955921, + "eval_PER_number": 27575, + "eval_PER_precision": 0.8063655761024182, + "eval_PER_recall": 0.8223028105167725, + "eval_loss": 0.1882677972316742, + "eval_overall_accuracy": 0.942047596664066, + "eval_overall_f1": 0.779018759018759, + "eval_overall_precision": 0.7740594172975453, + "eval_overall_recall": 0.7840420587893575, + "eval_runtime": 73.5108, + "eval_samples_per_second": 515.394, + "eval_steps_per_second": 64.426, + "step": 418000 + }, + { + "epoch": 2.37, + "learning_rate": 1.043661628058783e-05, + "loss": 0.1848, + "step": 418200 + }, + { + "epoch": 2.37, + "learning_rate": 1.0417695484930533e-05, + "loss": 0.186, + "step": 418400 + }, + { + "epoch": 2.38, + "learning_rate": 1.0398774689273234e-05, + "loss": 0.1826, + "step": 418600 + }, + { + "epoch": 2.38, + "learning_rate": 1.0379853893615934e-05, + "loss": 0.1828, + "step": 418800 + }, + { + "epoch": 2.38, + "learning_rate": 1.0360933097958637e-05, + "loss": 0.1795, + "step": 419000 + }, + { + "epoch": 2.38, + "learning_rate": 1.0342012302301336e-05, + "loss": 0.1866, + "step": 419200 + }, + { + "epoch": 2.38, + "learning_rate": 1.0323091506644038e-05, + "loss": 0.1726, + "step": 419400 + }, + { + "epoch": 2.38, + "learning_rate": 1.030417071098674e-05, + "loss": 0.1773, + "step": 419600 + }, + { + "epoch": 2.38, + "learning_rate": 1.028524991532944e-05, + "loss": 0.1723, + "step": 419800 + }, + { + "epoch": 2.38, + "learning_rate": 1.0266329119672142e-05, + "loss": 0.1864, + "step": 420000 + }, + { + "epoch": 2.38, + "eval_LOC_f1": 0.8157709561087167, + "eval_LOC_number": 22273, + "eval_LOC_precision": 0.7994224884713184, + "eval_LOC_recall": 0.8328020473218696, + "eval_ORG_f1": 0.6831212306384337, + "eval_ORG_number": 19008, + "eval_ORG_precision": 0.6900327411303742, + "eval_ORG_recall": 0.6763468013468014, + "eval_PER_f1": 0.8137782947123556, + "eval_PER_number": 27575, + "eval_PER_precision": 0.8032144118685977, + "eval_PER_recall": 0.8246237533998186, + "eval_loss": 0.18839803338050842, + "eval_overall_accuracy": 0.9420815181310241, + "eval_overall_f1": 0.7790503597122301, + "eval_overall_precision": 0.7718978102189781, + "eval_overall_recall": 0.786336702683862, + "eval_runtime": 71.9727, + "eval_samples_per_second": 526.408, + "eval_steps_per_second": 65.803, + "step": 420000 + }, + { + "epoch": 2.39, + "learning_rate": 1.0247408324014842e-05, + "loss": 0.1764, + "step": 420200 + }, + { + "epoch": 2.39, + "learning_rate": 1.0228487528357543e-05, + "loss": 0.1709, + "step": 420400 + }, + { + "epoch": 2.39, + "learning_rate": 1.0209566732700245e-05, + "loss": 0.1803, + "step": 420600 + }, + { + "epoch": 2.39, + "learning_rate": 1.0190645937042944e-05, + "loss": 0.1769, + "step": 420800 + }, + { + "epoch": 2.39, + "learning_rate": 1.0171725141385646e-05, + "loss": 0.1876, + "step": 421000 + }, + { + "epoch": 2.39, + "learning_rate": 1.0152804345728347e-05, + "loss": 0.1774, + "step": 421200 + }, + { + "epoch": 2.39, + "learning_rate": 1.0133883550071048e-05, + "loss": 0.1746, + "step": 421400 + }, + { + "epoch": 2.39, + "learning_rate": 1.011496275441375e-05, + "loss": 0.186, + "step": 421600 + }, + { + "epoch": 2.39, + "learning_rate": 1.0096041958756449e-05, + "loss": 0.1755, + "step": 421800 + }, + { + "epoch": 2.4, + "learning_rate": 1.0077121163099151e-05, + "loss": 0.1773, + "step": 422000 + }, + { + "epoch": 2.4, + "eval_LOC_f1": 0.8162943495400787, + "eval_LOC_number": 22273, + "eval_LOC_precision": 0.7968529524949758, + "eval_LOC_recall": 0.836708121941364, + "eval_ORG_f1": 0.682502616957887, + "eval_ORG_number": 19008, + "eval_ORG_precision": 0.6966957093539372, + "eval_ORG_recall": 0.6688762626262627, + "eval_PER_f1": 0.8144101260345107, + "eval_PER_number": 27575, + "eval_PER_precision": 0.8098687513447608, + "eval_PER_recall": 0.8190027198549411, + "eval_loss": 0.1892368644475937, + "eval_overall_accuracy": 0.9422188193068067, + "eval_overall_f1": 0.7795169752417292, + "eval_overall_precision": 0.7757832053163027, + "eval_overall_recall": 0.7832868595329383, + "eval_runtime": 72.1737, + "eval_samples_per_second": 524.942, + "eval_steps_per_second": 65.62, + "step": 422000 + }, + { + "epoch": 2.4, + "learning_rate": 1.0058200367441852e-05, + "loss": 0.1736, + "step": 422200 + }, + { + "epoch": 2.4, + "learning_rate": 1.0039279571784553e-05, + "loss": 0.1771, + "step": 422400 + }, + { + "epoch": 2.4, + "learning_rate": 1.0020358776127255e-05, + "loss": 0.1766, + "step": 422600 + }, + { + "epoch": 2.4, + "learning_rate": 1.0001437980469956e-05, + "loss": 0.1807, + "step": 422800 + }, + { + "epoch": 2.4, + "learning_rate": 9.982517184812656e-06, + "loss": 0.1891, + "step": 423000 + }, + { + "epoch": 2.4, + "learning_rate": 9.963596389155357e-06, + "loss": 0.1814, + "step": 423200 + }, + { + "epoch": 2.4, + "learning_rate": 9.944675593498058e-06, + "loss": 0.1852, + "step": 423400 + }, + { + "epoch": 2.4, + "learning_rate": 9.92575479784076e-06, + "loss": 0.18, + "step": 423600 + }, + { + "epoch": 2.41, + "learning_rate": 9.90683400218346e-06, + "loss": 0.1739, + "step": 423800 + }, + { + "epoch": 2.41, + "learning_rate": 9.887913206526161e-06, + "loss": 0.1705, + "step": 424000 + }, + { + "epoch": 2.41, + "eval_LOC_f1": 0.8165557954094894, + "eval_LOC_number": 22273, + "eval_LOC_precision": 0.8015135135135135, + "eval_LOC_recall": 0.8321734835899969, + "eval_ORG_f1": 0.6829500900709972, + "eval_ORG_number": 19008, + "eval_ORG_precision": 0.6878335112059765, + "eval_ORG_recall": 0.6781355218855218, + "eval_PER_f1": 0.8133499036400641, + "eval_PER_number": 27575, + "eval_PER_precision": 0.8079510484505833, + "eval_PER_recall": 0.8188213961922031, + "eval_loss": 0.19162102043628693, + "eval_overall_accuracy": 0.9422026662273029, + "eval_overall_f1": 0.7789019737933321, + "eval_overall_precision": 0.7735743650714071, + "eval_overall_recall": 0.7843034739165795, + "eval_runtime": 72.1248, + "eval_samples_per_second": 525.298, + "eval_steps_per_second": 65.664, + "step": 424000 + }, + { + "epoch": 2.41, + "learning_rate": 9.868992410868862e-06, + "loss": 0.1808, + "step": 424200 + }, + { + "epoch": 2.41, + "learning_rate": 9.850071615211563e-06, + "loss": 0.1797, + "step": 424400 + }, + { + "epoch": 2.41, + "learning_rate": 9.831150819554265e-06, + "loss": 0.1755, + "step": 424600 + }, + { + "epoch": 2.41, + "learning_rate": 9.812230023896966e-06, + "loss": 0.1809, + "step": 424800 + }, + { + "epoch": 2.41, + "learning_rate": 9.793309228239666e-06, + "loss": 0.1776, + "step": 425000 + }, + { + "epoch": 2.41, + "learning_rate": 9.774388432582367e-06, + "loss": 0.1834, + "step": 425200 + }, + { + "epoch": 2.41, + "learning_rate": 9.755467636925067e-06, + "loss": 0.1748, + "step": 425400 + }, + { + "epoch": 2.42, + "learning_rate": 9.73654684126777e-06, + "loss": 0.18, + "step": 425600 + }, + { + "epoch": 2.42, + "learning_rate": 9.71762604561047e-06, + "loss": 0.1781, + "step": 425800 + }, + { + "epoch": 2.42, + "learning_rate": 9.698705249953171e-06, + "loss": 0.179, + "step": 426000 + }, + { + "epoch": 2.42, + "eval_LOC_f1": 0.8160116064362963, + "eval_LOC_number": 22273, + "eval_LOC_precision": 0.7993884318876782, + "eval_LOC_recall": 0.8333408162349032, + "eval_ORG_f1": 0.6803090860644818, + "eval_ORG_number": 19008, + "eval_ORG_precision": 0.6892344239283015, + "eval_ORG_recall": 0.6716119528619529, + "eval_PER_f1": 0.8138400331239087, + "eval_PER_number": 27575, + "eval_PER_precision": 0.8080360334596411, + "eval_PER_recall": 0.819728014505893, + "eval_loss": 0.1894346922636032, + "eval_overall_accuracy": 0.9420556732038179, + "eval_overall_f1": 0.7783879743957972, + "eval_overall_precision": 0.7735924836835688, + "eval_overall_recall": 0.783243290345068, + "eval_runtime": 71.7013, + "eval_samples_per_second": 528.4, + "eval_steps_per_second": 66.052, + "step": 426000 + }, + { + "epoch": 2.42, + "learning_rate": 9.679784454295872e-06, + "loss": 0.1903, + "step": 426200 + }, + { + "epoch": 2.42, + "learning_rate": 9.660863658638574e-06, + "loss": 0.1812, + "step": 426400 + }, + { + "epoch": 2.42, + "learning_rate": 9.641942862981275e-06, + "loss": 0.1816, + "step": 426600 + }, + { + "epoch": 2.42, + "learning_rate": 9.623022067323975e-06, + "loss": 0.1768, + "step": 426800 + }, + { + "epoch": 2.42, + "learning_rate": 9.604101271666676e-06, + "loss": 0.1772, + "step": 427000 + }, + { + "epoch": 2.42, + "learning_rate": 9.585180476009378e-06, + "loss": 0.1824, + "step": 427200 + }, + { + "epoch": 2.43, + "learning_rate": 9.566259680352079e-06, + "loss": 0.1833, + "step": 427400 + }, + { + "epoch": 2.43, + "learning_rate": 9.54733888469478e-06, + "loss": 0.1717, + "step": 427600 + }, + { + "epoch": 2.43, + "learning_rate": 9.52841808903748e-06, + "loss": 0.1795, + "step": 427800 + }, + { + "epoch": 2.43, + "learning_rate": 9.509497293380181e-06, + "loss": 0.1822, + "step": 428000 + }, + { + "epoch": 2.43, + "eval_LOC_f1": 0.8158039224298861, + "eval_LOC_number": 22273, + "eval_LOC_precision": 0.7984952708512468, + "eval_LOC_recall": 0.8338795851479369, + "eval_ORG_f1": 0.6815181084137714, + "eval_ORG_number": 19008, + "eval_ORG_precision": 0.6814284962920107, + "eval_ORG_recall": 0.6816077441077442, + "eval_PER_f1": 0.8134744877750172, + "eval_PER_number": 27575, + "eval_PER_precision": 0.8077734472771481, + "eval_PER_recall": 0.8192565729827742, + "eval_loss": 0.1889965832233429, + "eval_overall_accuracy": 0.9418376066305161, + "eval_overall_f1": 0.7781675964801289, + "eval_overall_precision": 0.770501138952164, + "eval_overall_recall": 0.7859881491808993, + "eval_runtime": 71.8564, + "eval_samples_per_second": 527.26, + "eval_steps_per_second": 65.909, + "step": 428000 + }, + { + "epoch": 2.43, + "learning_rate": 9.490576497722883e-06, + "loss": 0.1825, + "step": 428200 + }, + { + "epoch": 2.43, + "learning_rate": 9.471655702065584e-06, + "loss": 0.1716, + "step": 428400 + }, + { + "epoch": 2.43, + "learning_rate": 9.452734906408285e-06, + "loss": 0.1902, + "step": 428600 + }, + { + "epoch": 2.43, + "learning_rate": 9.433814110750985e-06, + "loss": 0.1733, + "step": 428800 + }, + { + "epoch": 2.44, + "learning_rate": 9.414893315093688e-06, + "loss": 0.1747, + "step": 429000 + }, + { + "epoch": 2.44, + "learning_rate": 9.395972519436388e-06, + "loss": 0.1803, + "step": 429200 + }, + { + "epoch": 2.44, + "learning_rate": 9.377051723779089e-06, + "loss": 0.1674, + "step": 429400 + }, + { + "epoch": 2.44, + "learning_rate": 9.35813092812179e-06, + "loss": 0.1802, + "step": 429600 + }, + { + "epoch": 2.44, + "learning_rate": 9.33921013246449e-06, + "loss": 0.1759, + "step": 429800 + }, + { + "epoch": 2.44, + "learning_rate": 9.320289336807193e-06, + "loss": 0.1807, + "step": 430000 + }, + { + "epoch": 2.44, + "eval_LOC_f1": 0.8156781981903032, + "eval_LOC_number": 22273, + "eval_LOC_precision": 0.7965340179717587, + "eval_LOC_recall": 0.835765276343555, + "eval_ORG_f1": 0.682720833890151, + "eval_ORG_number": 19008, + "eval_ORG_precision": 0.6938664638452763, + "eval_ORG_recall": 0.6719276094276094, + "eval_PER_f1": 0.8130497071251112, + "eval_PER_number": 27575, + "eval_PER_precision": 0.8131529309344168, + "eval_PER_recall": 0.8129465095194923, + "eval_loss": 0.1891641616821289, + "eval_overall_accuracy": 0.9422737397771198, + "eval_overall_f1": 0.7786340185671595, + "eval_overall_precision": 0.7758886725791333, + "eval_overall_recall": 0.7813988613918903, + "eval_runtime": 71.5899, + "eval_samples_per_second": 529.223, + "eval_steps_per_second": 66.155, + "step": 430000 + }, + { + "epoch": 2.44, + "learning_rate": 9.301368541149893e-06, + "loss": 0.1819, + "step": 430200 + }, + { + "epoch": 2.44, + "learning_rate": 9.282447745492594e-06, + "loss": 0.1709, + "step": 430400 + }, + { + "epoch": 2.44, + "learning_rate": 9.263526949835294e-06, + "loss": 0.1729, + "step": 430600 + }, + { + "epoch": 2.45, + "learning_rate": 9.244606154177995e-06, + "loss": 0.1814, + "step": 430800 + }, + { + "epoch": 2.45, + "learning_rate": 9.225685358520697e-06, + "loss": 0.1844, + "step": 431000 + }, + { + "epoch": 2.45, + "learning_rate": 9.206764562863398e-06, + "loss": 0.1803, + "step": 431200 + }, + { + "epoch": 2.45, + "learning_rate": 9.187843767206099e-06, + "loss": 0.1835, + "step": 431400 + }, + { + "epoch": 2.45, + "learning_rate": 9.168922971548801e-06, + "loss": 0.1803, + "step": 431600 + }, + { + "epoch": 2.45, + "learning_rate": 9.1500021758915e-06, + "loss": 0.1728, + "step": 431800 + }, + { + "epoch": 2.45, + "learning_rate": 9.131081380234202e-06, + "loss": 0.1708, + "step": 432000 + }, + { + "epoch": 2.45, + "eval_LOC_f1": 0.8167049305783317, + "eval_LOC_number": 22273, + "eval_LOC_precision": 0.7960105702838632, + "eval_LOC_recall": 0.838504018318143, + "eval_ORG_f1": 0.6839508764563871, + "eval_ORG_number": 19008, + "eval_ORG_precision": 0.682303664921466, + "eval_ORG_recall": 0.6856060606060606, + "eval_PER_f1": 0.8147974965829797, + "eval_PER_number": 27575, + "eval_PER_precision": 0.8081986513967676, + "eval_PER_recall": 0.8215049864007253, + "eval_loss": 0.18953382968902588, + "eval_overall_accuracy": 0.9422010509193525, + "eval_overall_f1": 0.7796653925864305, + "eval_overall_precision": 0.7700840050431358, + "eval_overall_recall": 0.7894882072731497, + "eval_runtime": 74.7177, + "eval_samples_per_second": 507.069, + "eval_steps_per_second": 63.385, + "step": 432000 + }, + { + "epoch": 2.45, + "learning_rate": 9.112160584576903e-06, + "loss": 0.1787, + "step": 432200 + }, + { + "epoch": 2.45, + "learning_rate": 9.093239788919604e-06, + "loss": 0.1885, + "step": 432400 + }, + { + "epoch": 2.46, + "learning_rate": 9.074318993262306e-06, + "loss": 0.1772, + "step": 432600 + }, + { + "epoch": 2.46, + "learning_rate": 9.055398197605005e-06, + "loss": 0.1813, + "step": 432800 + }, + { + "epoch": 2.46, + "learning_rate": 9.036477401947707e-06, + "loss": 0.1812, + "step": 433000 + }, + { + "epoch": 2.46, + "learning_rate": 9.017556606290408e-06, + "loss": 0.1751, + "step": 433200 + }, + { + "epoch": 2.46, + "learning_rate": 8.998635810633109e-06, + "loss": 0.1829, + "step": 433400 + }, + { + "epoch": 2.46, + "learning_rate": 8.979715014975811e-06, + "loss": 0.1811, + "step": 433600 + }, + { + "epoch": 2.46, + "learning_rate": 8.96079421931851e-06, + "loss": 0.1715, + "step": 433800 + }, + { + "epoch": 2.46, + "learning_rate": 8.941873423661212e-06, + "loss": 0.1828, + "step": 434000 + }, + { + "epoch": 2.46, + "eval_LOC_f1": 0.8163184950235153, + "eval_LOC_number": 22273, + "eval_LOC_precision": 0.795964508147769, + "eval_LOC_recall": 0.837740762358012, + "eval_ORG_f1": 0.6841656451526942, + "eval_ORG_number": 19008, + "eval_ORG_precision": 0.6932174100874825, + "eval_ORG_recall": 0.6753472222222222, + "eval_PER_f1": 0.8142834141367158, + "eval_PER_number": 27575, + "eval_PER_precision": 0.803544947390721, + "eval_PER_recall": 0.825312783318223, + "eval_loss": 0.18813292682170868, + "eval_overall_accuracy": 0.9422769703930206, + "eval_overall_f1": 0.7798588451752936, + "eval_overall_precision": 0.7719472980279446, + "eval_overall_recall": 0.787934239572441, + "eval_runtime": 72.1769, + "eval_samples_per_second": 524.919, + "eval_steps_per_second": 65.617, + "step": 434000 + }, + { + "epoch": 2.46, + "learning_rate": 8.922952628003915e-06, + "loss": 0.1761, + "step": 434200 + }, + { + "epoch": 2.47, + "learning_rate": 8.904031832346614e-06, + "loss": 0.1825, + "step": 434400 + }, + { + "epoch": 2.47, + "learning_rate": 8.885111036689316e-06, + "loss": 0.1817, + "step": 434600 + }, + { + "epoch": 2.47, + "learning_rate": 8.866190241032017e-06, + "loss": 0.1806, + "step": 434800 + }, + { + "epoch": 2.47, + "learning_rate": 8.847269445374717e-06, + "loss": 0.1807, + "step": 435000 + }, + { + "epoch": 2.47, + "learning_rate": 8.82834864971742e-06, + "loss": 0.1713, + "step": 435200 + }, + { + "epoch": 2.47, + "learning_rate": 8.809427854060119e-06, + "loss": 0.1815, + "step": 435400 + }, + { + "epoch": 2.47, + "learning_rate": 8.79050705840282e-06, + "loss": 0.1735, + "step": 435600 + }, + { + "epoch": 2.47, + "learning_rate": 8.771586262745521e-06, + "loss": 0.1732, + "step": 435800 + }, + { + "epoch": 2.47, + "learning_rate": 8.752665467088222e-06, + "loss": 0.1798, + "step": 436000 + }, + { + "epoch": 2.47, + "eval_LOC_f1": 0.8164523747048018, + "eval_LOC_number": 22273, + "eval_LOC_precision": 0.7958139733151456, + "eval_LOC_recall": 0.8381897364522067, + "eval_ORG_f1": 0.6829021880732726, + "eval_ORG_number": 19008, + "eval_ORG_precision": 0.6902983069067455, + "eval_ORG_recall": 0.6756628787878788, + "eval_PER_f1": 0.8132571264243862, + "eval_PER_number": 27575, + "eval_PER_precision": 0.8061860813170366, + "eval_PER_recall": 0.820453309156845, + "eval_loss": 0.1887345016002655, + "eval_overall_accuracy": 0.9420621344356195, + "eval_overall_f1": 0.7790305289136081, + "eval_overall_precision": 0.7719708528812013, + "eval_overall_recall": 0.7862205181828744, + "eval_runtime": 72.2485, + "eval_samples_per_second": 524.399, + "eval_steps_per_second": 65.552, + "step": 436000 + }, + { + "epoch": 2.48, + "learning_rate": 8.733744671430924e-06, + "loss": 0.1843, + "step": 436200 + }, + { + "epoch": 2.48, + "learning_rate": 8.714823875773623e-06, + "loss": 0.173, + "step": 436400 + }, + { + "epoch": 2.48, + "learning_rate": 8.695903080116326e-06, + "loss": 0.1846, + "step": 436600 + }, + { + "epoch": 2.48, + "learning_rate": 8.676982284459026e-06, + "loss": 0.184, + "step": 436800 + }, + { + "epoch": 2.48, + "learning_rate": 8.658061488801727e-06, + "loss": 0.1858, + "step": 437000 + }, + { + "epoch": 2.48, + "learning_rate": 8.63914069314443e-06, + "loss": 0.1728, + "step": 437200 + }, + { + "epoch": 2.48, + "learning_rate": 8.62021989748713e-06, + "loss": 0.1798, + "step": 437400 + }, + { + "epoch": 2.48, + "learning_rate": 8.60129910182983e-06, + "loss": 0.1766, + "step": 437600 + }, + { + "epoch": 2.49, + "learning_rate": 8.582378306172531e-06, + "loss": 0.1936, + "step": 437800 + }, + { + "epoch": 2.49, + "learning_rate": 8.563457510515232e-06, + "loss": 0.1764, + "step": 438000 + }, + { + "epoch": 2.49, + "eval_LOC_f1": 0.8169990565830756, + "eval_LOC_number": 22273, + "eval_LOC_precision": 0.7988929889298892, + "eval_LOC_recall": 0.8359448659812329, + "eval_ORG_f1": 0.6843907135979459, + "eval_ORG_number": 19008, + "eval_ORG_precision": 0.6960826985854189, + "eval_ORG_recall": 0.6730850168350169, + "eval_PER_f1": 0.8135044180898734, + "eval_PER_number": 27575, + "eval_PER_precision": 0.8074449842812232, + "eval_PER_recall": 0.8196554850407978, + "eval_loss": 0.18866625428199768, + "eval_overall_accuracy": 0.9422478948499137, + "eval_overall_f1": 0.7798085668500152, + "eval_overall_precision": 0.775208817704297, + "eval_overall_recall": 0.7844632276054374, + "eval_runtime": 72.0923, + "eval_samples_per_second": 525.535, + "eval_steps_per_second": 65.694, + "step": 438000 + }, + { + "epoch": 2.49, + "learning_rate": 8.544536714857934e-06, + "loss": 0.1768, + "step": 438200 + }, + { + "epoch": 2.49, + "learning_rate": 8.525615919200635e-06, + "loss": 0.1863, + "step": 438400 + }, + { + "epoch": 2.49, + "learning_rate": 8.506695123543336e-06, + "loss": 0.1858, + "step": 438600 + }, + { + "epoch": 2.49, + "learning_rate": 8.487774327886036e-06, + "loss": 0.1894, + "step": 438800 + }, + { + "epoch": 2.49, + "learning_rate": 8.468853532228737e-06, + "loss": 0.1746, + "step": 439000 + }, + { + "epoch": 2.49, + "learning_rate": 8.44993273657144e-06, + "loss": 0.1753, + "step": 439200 + }, + { + "epoch": 2.49, + "learning_rate": 8.43101194091414e-06, + "loss": 0.1788, + "step": 439400 + }, + { + "epoch": 2.5, + "learning_rate": 8.41209114525684e-06, + "loss": 0.1757, + "step": 439600 + }, + { + "epoch": 2.5, + "learning_rate": 8.393170349599541e-06, + "loss": 0.1821, + "step": 439800 + }, + { + "epoch": 2.5, + "learning_rate": 8.374249553942244e-06, + "loss": 0.1847, + "step": 440000 + }, + { + "epoch": 2.5, + "eval_LOC_f1": 0.8165374677002584, + "eval_LOC_number": 22273, + "eval_LOC_precision": 0.796990552729449, + "eval_LOC_recall": 0.8370673012167198, + "eval_ORG_f1": 0.6838709677419355, + "eval_ORG_number": 19008, + "eval_ORG_precision": 0.6874335530512439, + "eval_ORG_recall": 0.6803451178451179, + "eval_PER_f1": 0.8132155590991347, + "eval_PER_number": 27575, + "eval_PER_precision": 0.8036688150718889, + "eval_PER_recall": 0.8229918404351768, + "eval_loss": 0.1876567006111145, + "eval_overall_accuracy": 0.9422398183101618, + "eval_overall_f1": 0.7791872159886288, + "eval_overall_precision": 0.7704101188194711, + "eval_overall_recall": 0.7881666085744162, + "eval_runtime": 72.316, + "eval_samples_per_second": 523.909, + "eval_steps_per_second": 65.49, + "step": 440000 + }, + { + "epoch": 2.5, + "learning_rate": 8.355328758284944e-06, + "loss": 0.1775, + "step": 440200 + }, + { + "epoch": 2.5, + "learning_rate": 8.336407962627645e-06, + "loss": 0.1743, + "step": 440400 + }, + { + "epoch": 2.5, + "learning_rate": 8.317487166970346e-06, + "loss": 0.1779, + "step": 440600 + }, + { + "epoch": 2.5, + "learning_rate": 8.298566371313046e-06, + "loss": 0.173, + "step": 440800 + }, + { + "epoch": 2.5, + "learning_rate": 8.279645575655748e-06, + "loss": 0.1824, + "step": 441000 + }, + { + "epoch": 2.5, + "learning_rate": 8.260724779998449e-06, + "loss": 0.1762, + "step": 441200 + }, + { + "epoch": 2.51, + "learning_rate": 8.24180398434115e-06, + "loss": 0.1727, + "step": 441400 + }, + { + "epoch": 2.51, + "learning_rate": 8.22288318868385e-06, + "loss": 0.174, + "step": 441600 + }, + { + "epoch": 2.51, + "learning_rate": 8.203962393026551e-06, + "loss": 0.1809, + "step": 441800 + }, + { + "epoch": 2.51, + "learning_rate": 8.185041597369253e-06, + "loss": 0.1822, + "step": 442000 + }, + { + "epoch": 2.51, + "eval_LOC_f1": 0.8163828203668841, + "eval_LOC_number": 22273, + "eval_LOC_precision": 0.7965739672775428, + "eval_LOC_recall": 0.8372019934449783, + "eval_ORG_f1": 0.6833708162455352, + "eval_ORG_number": 19008, + "eval_ORG_precision": 0.6873902166391653, + "eval_ORG_recall": 0.6793981481481481, + "eval_PER_f1": 0.8133078889569005, + "eval_PER_number": 27575, + "eval_PER_precision": 0.8075506774873976, + "eval_PER_recall": 0.8191477787851315, + "eval_loss": 0.18770961463451385, + "eval_overall_accuracy": 0.9422414336181121, + "eval_overall_f1": 0.7789934039691274, + "eval_overall_precision": 0.7717160488548748, + "eval_overall_recall": 0.7864093179969792, + "eval_runtime": 73.5165, + "eval_samples_per_second": 515.354, + "eval_steps_per_second": 64.421, + "step": 442000 + }, + { + "epoch": 2.51, + "learning_rate": 8.166120801711954e-06, + "loss": 0.1811, + "step": 442200 + }, + { + "epoch": 2.51, + "learning_rate": 8.147200006054655e-06, + "loss": 0.1763, + "step": 442400 + }, + { + "epoch": 2.51, + "learning_rate": 8.128279210397357e-06, + "loss": 0.1789, + "step": 442600 + }, + { + "epoch": 2.51, + "learning_rate": 8.109358414740056e-06, + "loss": 0.1831, + "step": 442800 + }, + { + "epoch": 2.51, + "learning_rate": 8.090437619082758e-06, + "loss": 0.1742, + "step": 443000 + }, + { + "epoch": 2.52, + "learning_rate": 8.071516823425459e-06, + "loss": 0.1912, + "step": 443200 + }, + { + "epoch": 2.52, + "learning_rate": 8.05259602776816e-06, + "loss": 0.1791, + "step": 443400 + }, + { + "epoch": 2.52, + "learning_rate": 8.033675232110862e-06, + "loss": 0.1776, + "step": 443600 + }, + { + "epoch": 2.52, + "learning_rate": 8.014754436453561e-06, + "loss": 0.1686, + "step": 443800 + }, + { + "epoch": 2.52, + "learning_rate": 7.995833640796263e-06, + "loss": 0.1744, + "step": 444000 + }, + { + "epoch": 2.52, + "eval_LOC_f1": 0.8162685490744586, + "eval_LOC_number": 22273, + "eval_LOC_precision": 0.7952222789984671, + "eval_LOC_recall": 0.8384591209087235, + "eval_ORG_f1": 0.6854124510551074, + "eval_ORG_number": 19008, + "eval_ORG_precision": 0.6847466526647414, + "eval_ORG_recall": 0.6860795454545454, + "eval_PER_f1": 0.8148908210473258, + "eval_PER_number": 27575, + "eval_PER_precision": 0.8090506148126966, + "eval_PER_recall": 0.820815956482321, + "eval_loss": 0.18882843852043152, + "eval_overall_accuracy": 0.9423318908633337, + "eval_overall_f1": 0.7799886625382998, + "eval_overall_precision": 0.770867314374867, + "eval_overall_recall": 0.7893284535842918, + "eval_runtime": 72.5252, + "eval_samples_per_second": 522.398, + "eval_steps_per_second": 65.301, + "step": 444000 + }, + { + "epoch": 2.52, + "learning_rate": 7.976912845138964e-06, + "loss": 0.1852, + "step": 444200 + }, + { + "epoch": 2.52, + "learning_rate": 7.957992049481665e-06, + "loss": 0.1732, + "step": 444400 + }, + { + "epoch": 2.52, + "learning_rate": 7.939071253824367e-06, + "loss": 0.1841, + "step": 444600 + }, + { + "epoch": 2.52, + "learning_rate": 7.920150458167066e-06, + "loss": 0.1779, + "step": 444800 + }, + { + "epoch": 2.53, + "learning_rate": 7.901229662509768e-06, + "loss": 0.1853, + "step": 445000 + }, + { + "epoch": 2.53, + "learning_rate": 7.88230886685247e-06, + "loss": 0.1763, + "step": 445200 + }, + { + "epoch": 2.53, + "learning_rate": 7.86338807119517e-06, + "loss": 0.1694, + "step": 445400 + }, + { + "epoch": 2.53, + "learning_rate": 7.844467275537872e-06, + "loss": 0.1814, + "step": 445600 + }, + { + "epoch": 2.53, + "learning_rate": 7.82554647988057e-06, + "loss": 0.1721, + "step": 445800 + }, + { + "epoch": 2.53, + "learning_rate": 7.806625684223273e-06, + "loss": 0.1723, + "step": 446000 + }, + { + "epoch": 2.53, + "eval_LOC_f1": 0.814568661971831, + "eval_LOC_number": 22273, + "eval_LOC_precision": 0.7988518150817974, + "eval_LOC_recall": 0.8309163561262515, + "eval_ORG_f1": 0.6839126064645823, + "eval_ORG_number": 19008, + "eval_ORG_precision": 0.6877327375252686, + "eval_ORG_recall": 0.6801346801346801, + "eval_PER_f1": 0.8137302399023882, + "eval_PER_number": 27575, + "eval_PER_precision": 0.8053345645688308, + "eval_PER_recall": 0.8223028105167725, + "eval_loss": 0.18926545977592468, + "eval_overall_accuracy": 0.942278585700971, + "eval_overall_f1": 0.7786899990645936, + "eval_overall_precision": 0.7716661199925843, + "eval_overall_recall": 0.7858429185546648, + "eval_runtime": 74.5504, + "eval_samples_per_second": 508.206, + "eval_steps_per_second": 63.527, + "step": 446000 + }, + { + "epoch": 2.53, + "learning_rate": 7.787704888565975e-06, + "loss": 0.1779, + "step": 446200 + }, + { + "epoch": 2.53, + "learning_rate": 7.768784092908674e-06, + "loss": 0.1808, + "step": 446400 + }, + { + "epoch": 2.54, + "learning_rate": 7.749863297251377e-06, + "loss": 0.1828, + "step": 446600 + }, + { + "epoch": 2.54, + "learning_rate": 7.730942501594077e-06, + "loss": 0.1898, + "step": 446800 + }, + { + "epoch": 2.54, + "learning_rate": 7.712021705936778e-06, + "loss": 0.1745, + "step": 447000 + }, + { + "epoch": 2.54, + "learning_rate": 7.69310091027948e-06, + "loss": 0.1763, + "step": 447200 + }, + { + "epoch": 2.54, + "learning_rate": 7.67418011462218e-06, + "loss": 0.1749, + "step": 447400 + }, + { + "epoch": 2.54, + "learning_rate": 7.655259318964882e-06, + "loss": 0.1769, + "step": 447600 + }, + { + "epoch": 2.54, + "learning_rate": 7.636338523307582e-06, + "loss": 0.1726, + "step": 447800 + }, + { + "epoch": 2.54, + "learning_rate": 7.617417727650283e-06, + "loss": 0.1798, + "step": 448000 + }, + { + "epoch": 2.54, + "eval_LOC_f1": 0.8167888947092719, + "eval_LOC_number": 22273, + "eval_LOC_precision": 0.7947585269506859, + "eval_LOC_recall": 0.8400754276478247, + "eval_ORG_f1": 0.6851038075941374, + "eval_ORG_number": 19008, + "eval_ORG_precision": 0.6853382469070808, + "eval_ORG_recall": 0.6848695286195287, + "eval_PER_f1": 0.8137462724104481, + "eval_PER_number": 27575, + "eval_PER_precision": 0.8062724716101243, + "eval_PER_recall": 0.8213599274705349, + "eval_loss": 0.18768233060836792, + "eval_overall_accuracy": 0.9422898928566237, + "eval_overall_f1": 0.7796967415851166, + "eval_overall_precision": 0.7699103767574226, + "eval_overall_recall": 0.7897350993377483, + "eval_runtime": 73.2129, + "eval_samples_per_second": 517.491, + "eval_steps_per_second": 64.688, + "step": 448000 + }, + { + "epoch": 2.54, + "learning_rate": 7.598496931992985e-06, + "loss": 0.1814, + "step": 448200 + }, + { + "epoch": 2.55, + "learning_rate": 7.579576136335685e-06, + "loss": 0.1751, + "step": 448400 + }, + { + "epoch": 2.55, + "learning_rate": 7.560655340678387e-06, + "loss": 0.1843, + "step": 448600 + }, + { + "epoch": 2.55, + "learning_rate": 7.541734545021087e-06, + "loss": 0.1796, + "step": 448800 + }, + { + "epoch": 2.55, + "learning_rate": 7.522813749363789e-06, + "loss": 0.174, + "step": 449000 + }, + { + "epoch": 2.55, + "learning_rate": 7.50389295370649e-06, + "loss": 0.175, + "step": 449200 + }, + { + "epoch": 2.55, + "learning_rate": 7.48497215804919e-06, + "loss": 0.178, + "step": 449400 + }, + { + "epoch": 2.55, + "learning_rate": 7.466051362391892e-06, + "loss": 0.1811, + "step": 449600 + }, + { + "epoch": 2.55, + "learning_rate": 7.447130566734593e-06, + "loss": 0.1858, + "step": 449800 + }, + { + "epoch": 2.55, + "learning_rate": 7.428209771077294e-06, + "loss": 0.1735, + "step": 450000 + }, + { + "epoch": 2.55, + "eval_LOC_f1": 0.8163336992316136, + "eval_LOC_number": 22273, + "eval_LOC_precision": 0.798728358465438, + "eval_LOC_recall": 0.834732635926907, + "eval_ORG_f1": 0.6821151760442292, + "eval_ORG_number": 19008, + "eval_ORG_precision": 0.6859803139132749, + "eval_ORG_recall": 0.6782933501683501, + "eval_PER_f1": 0.8142662458757711, + "eval_PER_number": 27575, + "eval_PER_precision": 0.805341751498599, + "eval_PER_recall": 0.8233907524932004, + "eval_loss": 0.18951769173145294, + "eval_overall_accuracy": 0.9422947387804748, + "eval_overall_f1": 0.7790340782484312, + "eval_overall_precision": 0.7712232263573614, + "eval_overall_recall": 0.7870047635645405, + "eval_runtime": 71.9012, + "eval_samples_per_second": 526.932, + "eval_steps_per_second": 65.868, + "step": 450000 + }, + { + "epoch": 2.56, + "learning_rate": 7.409288975419995e-06, + "loss": 0.1756, + "step": 450200 + }, + { + "epoch": 2.56, + "learning_rate": 7.390368179762695e-06, + "loss": 0.1764, + "step": 450400 + }, + { + "epoch": 2.56, + "learning_rate": 7.3714473841053965e-06, + "loss": 0.1812, + "step": 450600 + }, + { + "epoch": 2.56, + "learning_rate": 7.352526588448099e-06, + "loss": 0.1744, + "step": 450800 + }, + { + "epoch": 2.56, + "learning_rate": 7.333605792790799e-06, + "loss": 0.1741, + "step": 451000 + }, + { + "epoch": 2.56, + "learning_rate": 7.3146849971335e-06, + "loss": 0.1817, + "step": 451200 + }, + { + "epoch": 2.56, + "learning_rate": 7.295764201476201e-06, + "loss": 0.1862, + "step": 451400 + }, + { + "epoch": 2.56, + "learning_rate": 7.276843405818902e-06, + "loss": 0.186, + "step": 451600 + }, + { + "epoch": 2.56, + "learning_rate": 7.257922610161604e-06, + "loss": 0.1854, + "step": 451800 + }, + { + "epoch": 2.57, + "learning_rate": 7.239001814504304e-06, + "loss": 0.1753, + "step": 452000 + }, + { + "epoch": 2.57, + "eval_LOC_f1": 0.8178324365872406, + "eval_LOC_number": 22273, + "eval_LOC_precision": 0.8004470810764337, + "eval_LOC_recall": 0.8359897633906523, + "eval_ORG_f1": 0.6851739986868023, + "eval_ORG_number": 19008, + "eval_ORG_precision": 0.684113914092411, + "eval_ORG_recall": 0.6862373737373737, + "eval_PER_f1": 0.8143301512388718, + "eval_PER_number": 27575, + "eval_PER_precision": 0.8030180164298558, + "eval_PER_recall": 0.8259655485040798, + "eval_loss": 0.18937131762504578, + "eval_overall_accuracy": 0.9422866622407229, + "eval_overall_f1": 0.7802333247341416, + "eval_overall_precision": 0.7701012844451989, + "eval_overall_recall": 0.790635529220402, + "eval_runtime": 73.3689, + "eval_samples_per_second": 516.39, + "eval_steps_per_second": 64.551, + "step": 452000 + }, + { + "epoch": 2.57, + "learning_rate": 7.220081018847005e-06, + "loss": 0.1745, + "step": 452200 + }, + { + "epoch": 2.57, + "learning_rate": 7.201160223189706e-06, + "loss": 0.1773, + "step": 452400 + }, + { + "epoch": 2.57, + "learning_rate": 7.182239427532407e-06, + "loss": 0.1811, + "step": 452600 + }, + { + "epoch": 2.57, + "learning_rate": 7.163318631875109e-06, + "loss": 0.1773, + "step": 452800 + }, + { + "epoch": 2.57, + "learning_rate": 7.1443978362178085e-06, + "loss": 0.1776, + "step": 453000 + }, + { + "epoch": 2.57, + "learning_rate": 7.12547704056051e-06, + "loss": 0.1866, + "step": 453200 + }, + { + "epoch": 2.57, + "learning_rate": 7.106556244903211e-06, + "loss": 0.1795, + "step": 453400 + }, + { + "epoch": 2.57, + "learning_rate": 7.087635449245912e-06, + "loss": 0.1824, + "step": 453600 + }, + { + "epoch": 2.58, + "learning_rate": 7.068714653588614e-06, + "loss": 0.1745, + "step": 453800 + }, + { + "epoch": 2.58, + "learning_rate": 7.0497938579313135e-06, + "loss": 0.1794, + "step": 454000 + }, + { + "epoch": 2.58, + "eval_LOC_f1": 0.8179565341345205, + "eval_LOC_number": 22273, + "eval_LOC_precision": 0.7994941267255423, + "eval_LOC_recall": 0.8372917882638172, + "eval_ORG_f1": 0.6862074493667504, + "eval_ORG_number": 19008, + "eval_ORG_precision": 0.6971983928765338, + "eval_ORG_recall": 0.6755576599326599, + "eval_PER_f1": 0.8147296981884123, + "eval_PER_number": 27575, + "eval_PER_precision": 0.8068060593129934, + "eval_PER_recall": 0.8228105167724388, + "eval_loss": 0.18821245431900024, + "eval_overall_accuracy": 0.9424691920391163, + "eval_overall_f1": 0.7811161892129584, + "eval_overall_precision": 0.7754701857842154, + "eval_overall_recall": 0.7868450098756826, + "eval_runtime": 72.56, + "eval_samples_per_second": 522.147, + "eval_steps_per_second": 65.27, + "step": 454000 + }, + { + "epoch": 2.58, + "learning_rate": 7.030873062274016e-06, + "loss": 0.1796, + "step": 454200 + }, + { + "epoch": 2.58, + "learning_rate": 7.011952266616716e-06, + "loss": 0.1773, + "step": 454400 + }, + { + "epoch": 2.58, + "learning_rate": 6.993031470959417e-06, + "loss": 0.1873, + "step": 454600 + }, + { + "epoch": 2.58, + "learning_rate": 6.974110675302119e-06, + "loss": 0.1753, + "step": 454800 + }, + { + "epoch": 2.58, + "learning_rate": 6.955189879644819e-06, + "loss": 0.1737, + "step": 455000 + }, + { + "epoch": 2.58, + "learning_rate": 6.936269083987521e-06, + "loss": 0.1789, + "step": 455200 + }, + { + "epoch": 2.58, + "learning_rate": 6.9173482883302205e-06, + "loss": 0.1797, + "step": 455400 + }, + { + "epoch": 2.59, + "learning_rate": 6.898427492672922e-06, + "loss": 0.1744, + "step": 455600 + }, + { + "epoch": 2.59, + "learning_rate": 6.8795066970156235e-06, + "loss": 0.174, + "step": 455800 + }, + { + "epoch": 2.59, + "learning_rate": 6.860585901358324e-06, + "loss": 0.1796, + "step": 456000 + }, + { + "epoch": 2.59, + "eval_LOC_f1": 0.8166721365277323, + "eval_LOC_number": 22273, + "eval_LOC_precision": 0.7964749061113008, + "eval_LOC_recall": 0.8379203519956898, + "eval_ORG_f1": 0.6845669667755556, + "eval_ORG_number": 19008, + "eval_ORG_precision": 0.6912308930008045, + "eval_ORG_recall": 0.678030303030303, + "eval_PER_f1": 0.8155475512069243, + "eval_PER_number": 27575, + "eval_PER_precision": 0.8061144962448633, + "eval_PER_recall": 0.8252039891205802, + "eval_loss": 0.18758217990398407, + "eval_overall_accuracy": 0.942501498198124, + "eval_overall_f1": 0.7804772888955956, + "eval_overall_precision": 0.7724343930019202, + "eval_overall_recall": 0.7886894388288602, + "eval_runtime": 73.5073, + "eval_samples_per_second": 515.418, + "eval_steps_per_second": 64.429, + "step": 456000 + }, + { + "epoch": 2.59, + "learning_rate": 6.841665105701026e-06, + "loss": 0.1751, + "step": 456200 + }, + { + "epoch": 2.59, + "learning_rate": 6.8227443100437255e-06, + "loss": 0.1868, + "step": 456400 + }, + { + "epoch": 2.59, + "learning_rate": 6.803823514386427e-06, + "loss": 0.176, + "step": 456600 + }, + { + "epoch": 2.59, + "learning_rate": 6.784902718729129e-06, + "loss": 0.1761, + "step": 456800 + }, + { + "epoch": 2.59, + "learning_rate": 6.765981923071829e-06, + "loss": 0.1731, + "step": 457000 + }, + { + "epoch": 2.6, + "learning_rate": 6.747061127414531e-06, + "loss": 0.1686, + "step": 457200 + }, + { + "epoch": 2.6, + "learning_rate": 6.728140331757231e-06, + "loss": 0.1827, + "step": 457400 + }, + { + "epoch": 2.6, + "learning_rate": 6.709219536099933e-06, + "loss": 0.1872, + "step": 457600 + }, + { + "epoch": 2.6, + "learning_rate": 6.690298740442634e-06, + "loss": 0.1771, + "step": 457800 + }, + { + "epoch": 2.6, + "learning_rate": 6.671377944785334e-06, + "loss": 0.1795, + "step": 458000 + }, + { + "epoch": 2.6, + "eval_LOC_f1": 0.8168983254897669, + "eval_LOC_number": 22273, + "eval_LOC_precision": 0.7970271655561251, + "eval_LOC_recall": 0.8377856597674315, + "eval_ORG_f1": 0.6854562358879001, + "eval_ORG_number": 19008, + "eval_ORG_precision": 0.6922788002360895, + "eval_ORG_recall": 0.678766835016835, + "eval_PER_f1": 0.8142457097723846, + "eval_PER_number": 27575, + "eval_PER_precision": 0.8061706892261756, + "eval_PER_recall": 0.8224841341795104, + "eval_loss": 0.18816010653972626, + "eval_overall_accuracy": 0.9426097238307998, + "eval_overall_f1": 0.7802471266847911, + "eval_overall_precision": 0.7728762360719272, + "eval_overall_recall": 0.7877599628209597, + "eval_runtime": 74.809, + "eval_samples_per_second": 506.45, + "eval_steps_per_second": 63.308, + "step": 458000 + }, + { + "epoch": 2.6, + "learning_rate": 6.6524571491280355e-06, + "loss": 0.1799, + "step": 458200 + }, + { + "epoch": 2.6, + "learning_rate": 6.633536353470736e-06, + "loss": 0.1832, + "step": 458400 + }, + { + "epoch": 2.6, + "learning_rate": 6.614615557813438e-06, + "loss": 0.1786, + "step": 458600 + }, + { + "epoch": 2.6, + "learning_rate": 6.595694762156139e-06, + "loss": 0.1777, + "step": 458800 + }, + { + "epoch": 2.61, + "learning_rate": 6.576773966498839e-06, + "loss": 0.1748, + "step": 459000 + }, + { + "epoch": 2.61, + "learning_rate": 6.5578531708415405e-06, + "loss": 0.1718, + "step": 459200 + }, + { + "epoch": 2.61, + "learning_rate": 6.538932375184241e-06, + "loss": 0.1846, + "step": 459400 + }, + { + "epoch": 2.61, + "learning_rate": 6.520011579526943e-06, + "loss": 0.1813, + "step": 459600 + }, + { + "epoch": 2.61, + "learning_rate": 6.501090783869644e-06, + "loss": 0.1803, + "step": 459800 + }, + { + "epoch": 2.61, + "learning_rate": 6.482169988212345e-06, + "loss": 0.172, + "step": 460000 + }, + { + "epoch": 2.61, + "eval_LOC_f1": 0.8170582583631758, + "eval_LOC_number": 22273, + "eval_LOC_precision": 0.7953492326659014, + "eval_LOC_recall": 0.8399856328289858, + "eval_ORG_f1": 0.6855695935431119, + "eval_ORG_number": 19008, + "eval_ORG_precision": 0.6874371859296482, + "eval_ORG_recall": 0.6837121212121212, + "eval_PER_f1": 0.814566328818728, + "eval_PER_number": 27575, + "eval_PER_precision": 0.8059637912673057, + "eval_PER_recall": 0.8233544877606528, + "eval_loss": 0.1880265474319458, + "eval_overall_accuracy": 0.942535419665082, + "eval_overall_f1": 0.7803146557287708, + "eval_overall_precision": 0.7706875548882405, + "eval_overall_recall": 0.7901853142790751, + "eval_runtime": 72.4319, + "eval_samples_per_second": 523.071, + "eval_steps_per_second": 65.386, + "step": 460000 + }, + { + "epoch": 2.61, + "learning_rate": 6.463249192555046e-06, + "loss": 0.1845, + "step": 460200 + }, + { + "epoch": 2.61, + "learning_rate": 6.444328396897746e-06, + "loss": 0.1766, + "step": 460400 + }, + { + "epoch": 2.61, + "learning_rate": 6.4254076012404475e-06, + "loss": 0.1816, + "step": 460600 + }, + { + "epoch": 2.62, + "learning_rate": 6.406486805583149e-06, + "loss": 0.1761, + "step": 460800 + }, + { + "epoch": 2.62, + "learning_rate": 6.38756600992585e-06, + "loss": 0.1748, + "step": 461000 + }, + { + "epoch": 2.62, + "learning_rate": 6.368645214268551e-06, + "loss": 0.1873, + "step": 461200 + }, + { + "epoch": 2.62, + "learning_rate": 6.349724418611251e-06, + "loss": 0.1808, + "step": 461400 + }, + { + "epoch": 2.62, + "learning_rate": 6.3308036229539525e-06, + "loss": 0.1868, + "step": 461600 + }, + { + "epoch": 2.62, + "learning_rate": 6.311882827296654e-06, + "loss": 0.1835, + "step": 461800 + }, + { + "epoch": 2.62, + "learning_rate": 6.292962031639355e-06, + "loss": 0.184, + "step": 462000 + }, + { + "epoch": 2.62, + "eval_LOC_f1": 0.8171896914446004, + "eval_LOC_number": 22273, + "eval_LOC_precision": 0.7981934158140331, + "eval_LOC_recall": 0.8371121986261393, + "eval_ORG_f1": 0.6876192495230019, + "eval_ORG_number": 19008, + "eval_ORG_precision": 0.6927595044852627, + "eval_ORG_recall": 0.6825547138047138, + "eval_PER_f1": 0.8149977467327625, + "eval_PER_number": 27575, + "eval_PER_precision": 0.8102508960573477, + "eval_PER_recall": 0.8198005439709882, + "eval_loss": 0.18652905523777008, + "eval_overall_accuracy": 0.9426452606057082, + "eval_overall_f1": 0.7810980748039152, + "eval_overall_precision": 0.7747867461099919, + "eval_overall_recall": 0.7875130707563611, + "eval_runtime": 72.2449, + "eval_samples_per_second": 524.425, + "eval_steps_per_second": 65.555, + "step": 462000 + }, + { + "epoch": 2.62, + "learning_rate": 6.274041235982056e-06, + "loss": 0.1836, + "step": 462200 + }, + { + "epoch": 2.62, + "learning_rate": 6.255120440324756e-06, + "loss": 0.1776, + "step": 462400 + }, + { + "epoch": 2.63, + "learning_rate": 6.236199644667458e-06, + "loss": 0.1777, + "step": 462600 + }, + { + "epoch": 2.63, + "learning_rate": 6.217278849010159e-06, + "loss": 0.1792, + "step": 462800 + }, + { + "epoch": 2.63, + "learning_rate": 6.1983580533528595e-06, + "loss": 0.1841, + "step": 463000 + }, + { + "epoch": 2.63, + "learning_rate": 6.179437257695561e-06, + "loss": 0.1827, + "step": 463200 + }, + { + "epoch": 2.63, + "learning_rate": 6.160516462038262e-06, + "loss": 0.1787, + "step": 463400 + }, + { + "epoch": 2.63, + "learning_rate": 6.141595666380963e-06, + "loss": 0.1805, + "step": 463600 + }, + { + "epoch": 2.63, + "learning_rate": 6.122674870723664e-06, + "loss": 0.1739, + "step": 463800 + }, + { + "epoch": 2.63, + "learning_rate": 6.1037540750663645e-06, + "loss": 0.1822, + "step": 464000 + }, + { + "epoch": 2.63, + "eval_LOC_f1": 0.818189760615062, + "eval_LOC_number": 22273, + "eval_LOC_precision": 0.7966483773552805, + "eval_LOC_recall": 0.8409284784267947, + "eval_ORG_f1": 0.6858896354690228, + "eval_ORG_number": 19008, + "eval_ORG_precision": 0.6897946153027562, + "eval_ORG_recall": 0.6820286195286195, + "eval_PER_f1": 0.8154267997634367, + "eval_PER_number": 27575, + "eval_PER_precision": 0.8060515873015873, + "eval_PER_recall": 0.8250226654578422, + "eval_loss": 0.18639163672924042, + "eval_overall_accuracy": 0.9427082576157731, + "eval_overall_f1": 0.7812031423754349, + "eval_overall_precision": 0.7719377844574572, + "eval_overall_recall": 0.7906936214708958, + "eval_runtime": 71.4968, + "eval_samples_per_second": 529.912, + "eval_steps_per_second": 66.241, + "step": 464000 + }, + { + "epoch": 2.63, + "learning_rate": 6.084833279409066e-06, + "loss": 0.1745, + "step": 464200 + }, + { + "epoch": 2.64, + "learning_rate": 6.0659124837517675e-06, + "loss": 0.1754, + "step": 464400 + }, + { + "epoch": 2.64, + "learning_rate": 6.046991688094468e-06, + "loss": 0.1686, + "step": 464600 + }, + { + "epoch": 2.64, + "learning_rate": 6.028070892437169e-06, + "loss": 0.1698, + "step": 464800 + }, + { + "epoch": 2.64, + "learning_rate": 6.009150096779869e-06, + "loss": 0.1831, + "step": 465000 + }, + { + "epoch": 2.64, + "learning_rate": 5.990229301122571e-06, + "loss": 0.1747, + "step": 465200 + }, + { + "epoch": 2.64, + "learning_rate": 5.971308505465272e-06, + "loss": 0.1738, + "step": 465400 + }, + { + "epoch": 2.64, + "learning_rate": 5.952387709807973e-06, + "loss": 0.1779, + "step": 465600 + }, + { + "epoch": 2.64, + "learning_rate": 5.933466914150674e-06, + "loss": 0.1795, + "step": 465800 + }, + { + "epoch": 2.65, + "learning_rate": 5.914546118493375e-06, + "loss": 0.1809, + "step": 466000 + }, + { + "epoch": 2.65, + "eval_LOC_f1": 0.8176819689300053, + "eval_LOC_number": 22273, + "eval_LOC_precision": 0.7953648287278747, + "eval_LOC_recall": 0.8412876577021506, + "eval_ORG_f1": 0.6873124732104586, + "eval_ORG_number": 19008, + "eval_ORG_precision": 0.7002183406113537, + "eval_ORG_recall": 0.6748737373737373, + "eval_PER_f1": 0.8149567522071597, + "eval_PER_number": 27575, + "eval_PER_precision": 0.8049954008349254, + "eval_PER_recall": 0.8251677243880327, + "eval_loss": 0.18697375059127808, + "eval_overall_accuracy": 0.942913401725472, + "eval_overall_f1": 0.7815771109560363, + "eval_overall_precision": 0.7743958942191176, + "eval_overall_recall": 0.7888927617055885, + "eval_runtime": 72.2726, + "eval_samples_per_second": 524.224, + "eval_steps_per_second": 65.53, + "step": 466000 + }, + { + "epoch": 2.65, + "learning_rate": 5.895625322836077e-06, + "loss": 0.1773, + "step": 466200 + }, + { + "epoch": 2.65, + "learning_rate": 5.876704527178777e-06, + "loss": 0.1724, + "step": 466400 + }, + { + "epoch": 2.65, + "learning_rate": 5.857783731521478e-06, + "loss": 0.1772, + "step": 466600 + }, + { + "epoch": 2.65, + "learning_rate": 5.838862935864179e-06, + "loss": 0.1808, + "step": 466800 + }, + { + "epoch": 2.65, + "learning_rate": 5.819942140206881e-06, + "loss": 0.1815, + "step": 467000 + }, + { + "epoch": 2.65, + "learning_rate": 5.801021344549582e-06, + "loss": 0.1707, + "step": 467200 + }, + { + "epoch": 2.65, + "learning_rate": 5.782100548892282e-06, + "loss": 0.1791, + "step": 467400 + }, + { + "epoch": 2.65, + "learning_rate": 5.763179753234983e-06, + "loss": 0.1868, + "step": 467600 + }, + { + "epoch": 2.66, + "learning_rate": 5.744258957577684e-06, + "loss": 0.1809, + "step": 467800 + }, + { + "epoch": 2.66, + "learning_rate": 5.725338161920386e-06, + "loss": 0.1873, + "step": 468000 + }, + { + "epoch": 2.66, + "eval_LOC_f1": 0.8188518999317916, + "eval_LOC_number": 22273, + "eval_LOC_precision": 0.8028995512599241, + "eval_LOC_recall": 0.8354509944776186, + "eval_ORG_f1": 0.6864406779661016, + "eval_ORG_number": 19008, + "eval_ORG_precision": 0.700109409190372, + "eval_ORG_recall": 0.6732954545454546, + "eval_PER_f1": 0.814182157467882, + "eval_PER_number": 27575, + "eval_PER_precision": 0.8057676599069503, + "eval_PER_recall": 0.8227742520398912, + "eval_loss": 0.18617872893810272, + "eval_overall_accuracy": 0.9427841770894412, + "eval_overall_f1": 0.781315673544259, + "eval_overall_precision": 0.7770675017597288, + "eval_overall_recall": 0.7856105495526897, + "eval_runtime": 72.1673, + "eval_samples_per_second": 524.988, + "eval_steps_per_second": 65.625, + "step": 468000 + }, + { + "epoch": 2.66, + "learning_rate": 5.7064173662630865e-06, + "loss": 0.172, + "step": 468200 + }, + { + "epoch": 2.66, + "learning_rate": 5.687496570605787e-06, + "loss": 0.1772, + "step": 468400 + }, + { + "epoch": 2.66, + "learning_rate": 5.668575774948489e-06, + "loss": 0.1797, + "step": 468600 + }, + { + "epoch": 2.66, + "learning_rate": 5.649654979291189e-06, + "loss": 0.1793, + "step": 468800 + }, + { + "epoch": 2.66, + "learning_rate": 5.630734183633891e-06, + "loss": 0.1729, + "step": 469000 + }, + { + "epoch": 2.66, + "learning_rate": 5.6118133879765915e-06, + "loss": 0.1789, + "step": 469200 + }, + { + "epoch": 2.66, + "learning_rate": 5.592892592319292e-06, + "loss": 0.1768, + "step": 469400 + }, + { + "epoch": 2.67, + "learning_rate": 5.573971796661994e-06, + "loss": 0.168, + "step": 469600 + }, + { + "epoch": 2.67, + "learning_rate": 5.555051001004694e-06, + "loss": 0.1815, + "step": 469800 + }, + { + "epoch": 2.67, + "learning_rate": 5.536130205347396e-06, + "loss": 0.1775, + "step": 470000 + }, + { + "epoch": 2.67, + "eval_LOC_f1": 0.8185355345013298, + "eval_LOC_number": 22273, + "eval_LOC_precision": 0.8018778533896115, + "eval_LOC_recall": 0.8358999685718134, + "eval_ORG_f1": 0.6872658187767164, + "eval_ORG_number": 19008, + "eval_ORG_precision": 0.6894123875066173, + "eval_ORG_recall": 0.6851325757575758, + "eval_PER_f1": 0.8161945646902448, + "eval_PER_number": 27575, + "eval_PER_precision": 0.8085984767599117, + "eval_PER_recall": 0.8239347234814143, + "eval_loss": 0.18668155372142792, + "eval_overall_accuracy": 0.9427502556224832, + "eval_overall_f1": 0.7818239346478549, + "eval_overall_precision": 0.7743070392843916, + "eval_overall_recall": 0.7894882072731497, + "eval_runtime": 71.9522, + "eval_samples_per_second": 526.558, + "eval_steps_per_second": 65.822, + "step": 470000 + }, + { + "epoch": 2.67, + "learning_rate": 5.517209409690096e-06, + "loss": 0.1862, + "step": 470200 + }, + { + "epoch": 2.67, + "learning_rate": 5.498288614032798e-06, + "loss": 0.1774, + "step": 470400 + }, + { + "epoch": 2.67, + "learning_rate": 5.4793678183754985e-06, + "loss": 0.1722, + "step": 470600 + }, + { + "epoch": 2.67, + "learning_rate": 5.460447022718199e-06, + "loss": 0.1758, + "step": 470800 + }, + { + "epoch": 2.67, + "learning_rate": 5.441526227060901e-06, + "loss": 0.1736, + "step": 471000 + }, + { + "epoch": 2.67, + "learning_rate": 5.422605431403602e-06, + "loss": 0.1676, + "step": 471200 + }, + { + "epoch": 2.68, + "learning_rate": 5.403684635746303e-06, + "loss": 0.1755, + "step": 471400 + }, + { + "epoch": 2.68, + "learning_rate": 5.3847638400890035e-06, + "loss": 0.1789, + "step": 471600 + }, + { + "epoch": 2.68, + "learning_rate": 5.365843044431704e-06, + "loss": 0.1834, + "step": 471800 + }, + { + "epoch": 2.68, + "learning_rate": 5.346922248774406e-06, + "loss": 0.1782, + "step": 472000 + }, + { + "epoch": 2.68, + "eval_LOC_f1": 0.8174424942766816, + "eval_LOC_number": 22273, + "eval_LOC_precision": 0.7945913869108172, + "eval_LOC_recall": 0.8416468369775064, + "eval_ORG_f1": 0.6848160322341215, + "eval_ORG_number": 19008, + "eval_ORG_precision": 0.6901581534515923, + "eval_ORG_recall": 0.6795559764309764, + "eval_PER_f1": 0.8148241477374996, + "eval_PER_number": 27575, + "eval_PER_precision": 0.8088839652646249, + "eval_PER_recall": 0.8208522212148686, + "eval_loss": 0.18838666379451752, + "eval_overall_accuracy": 0.9424950369663224, + "eval_overall_f1": 0.7804408287638254, + "eval_overall_precision": 0.7724744277361255, + "eval_overall_recall": 0.7885732543278726, + "eval_runtime": 77.3701, + "eval_samples_per_second": 489.685, + "eval_steps_per_second": 61.212, + "step": 472000 + }, + { + "epoch": 2.68, + "learning_rate": 5.328001453117107e-06, + "loss": 0.1739, + "step": 472200 + }, + { + "epoch": 2.68, + "learning_rate": 5.309080657459808e-06, + "loss": 0.171, + "step": 472400 + }, + { + "epoch": 2.68, + "learning_rate": 5.290159861802508e-06, + "loss": 0.178, + "step": 472600 + }, + { + "epoch": 2.68, + "learning_rate": 5.27123906614521e-06, + "loss": 0.1803, + "step": 472800 + }, + { + "epoch": 2.68, + "learning_rate": 5.252318270487911e-06, + "loss": 0.1835, + "step": 473000 + }, + { + "epoch": 2.69, + "learning_rate": 5.233397474830612e-06, + "loss": 0.1793, + "step": 473200 + }, + { + "epoch": 2.69, + "learning_rate": 5.214476679173313e-06, + "loss": 0.1853, + "step": 473400 + }, + { + "epoch": 2.69, + "learning_rate": 5.195555883516013e-06, + "loss": 0.1771, + "step": 473600 + }, + { + "epoch": 2.69, + "learning_rate": 5.176635087858715e-06, + "loss": 0.1897, + "step": 473800 + }, + { + "epoch": 2.69, + "learning_rate": 5.157714292201416e-06, + "loss": 0.1833, + "step": 474000 + }, + { + "epoch": 2.69, + "eval_LOC_f1": 0.8184570826750922, + "eval_LOC_number": 22273, + "eval_LOC_precision": 0.8003690511951251, + "eval_LOC_recall": 0.8373815830826561, + "eval_ORG_f1": 0.6873690630883873, + "eval_ORG_number": 19008, + "eval_ORG_precision": 0.6930110689267953, + "eval_ORG_recall": 0.6818181818181818, + "eval_PER_f1": 0.8153708348262272, + "eval_PER_number": 27575, + "eval_PER_precision": 0.8057001239157373, + "eval_PER_recall": 0.8252765185856754, + "eval_loss": 0.18666990101337433, + "eval_overall_accuracy": 0.9428649424869604, + "eval_overall_f1": 0.7816829014054133, + "eval_overall_precision": 0.7739327250209967, + "eval_overall_recall": 0.7895898687115139, + "eval_runtime": 74.3144, + "eval_samples_per_second": 509.82, + "eval_steps_per_second": 63.729, + "step": 474000 + }, + { + "epoch": 2.69, + "learning_rate": 5.138793496544117e-06, + "loss": 0.1793, + "step": 474200 + }, + { + "epoch": 2.69, + "learning_rate": 5.119872700886818e-06, + "loss": 0.1784, + "step": 474400 + }, + { + "epoch": 2.69, + "learning_rate": 5.100951905229519e-06, + "loss": 0.1748, + "step": 474600 + }, + { + "epoch": 2.7, + "learning_rate": 5.08203110957222e-06, + "loss": 0.1775, + "step": 474800 + }, + { + "epoch": 2.7, + "learning_rate": 5.063110313914921e-06, + "loss": 0.1747, + "step": 475000 + }, + { + "epoch": 2.7, + "learning_rate": 5.044189518257622e-06, + "loss": 0.1824, + "step": 475200 + }, + { + "epoch": 2.7, + "learning_rate": 5.0252687226003226e-06, + "loss": 0.1767, + "step": 475400 + }, + { + "epoch": 2.7, + "learning_rate": 5.006347926943024e-06, + "loss": 0.1803, + "step": 475600 + }, + { + "epoch": 2.7, + "learning_rate": 4.987427131285725e-06, + "loss": 0.1691, + "step": 475800 + }, + { + "epoch": 2.7, + "learning_rate": 4.968506335628426e-06, + "loss": 0.1793, + "step": 476000 + }, + { + "epoch": 2.7, + "eval_LOC_f1": 0.8190401086504129, + "eval_LOC_number": 22273, + "eval_LOC_precision": 0.7996834630849516, + "eval_LOC_recall": 0.839357069097113, + "eval_ORG_f1": 0.6871784046230572, + "eval_ORG_number": 19008, + "eval_ORG_precision": 0.6893430038646832, + "eval_ORG_recall": 0.6850273569023569, + "eval_PER_f1": 0.8159606194531279, + "eval_PER_number": 27575, + "eval_PER_precision": 0.8085235162174672, + "eval_PER_recall": 0.8235358114233907, + "eval_loss": 0.18768605589866638, + "eval_overall_accuracy": 0.9428180985563993, + "eval_overall_f1": 0.7819122189497881, + "eval_overall_precision": 0.7735878557011684, + "eval_overall_recall": 0.7904176832810503, + "eval_runtime": 72.5389, + "eval_samples_per_second": 522.299, + "eval_steps_per_second": 65.289, + "step": 476000 + }, + { + "epoch": 2.7, + "learning_rate": 4.949585539971127e-06, + "loss": 0.1684, + "step": 476200 + }, + { + "epoch": 2.7, + "learning_rate": 4.930664744313828e-06, + "loss": 0.1774, + "step": 476400 + }, + { + "epoch": 2.71, + "learning_rate": 4.911743948656529e-06, + "loss": 0.1832, + "step": 476600 + }, + { + "epoch": 2.71, + "learning_rate": 4.8928231529992305e-06, + "loss": 0.1773, + "step": 476800 + }, + { + "epoch": 2.71, + "learning_rate": 4.873902357341931e-06, + "loss": 0.1764, + "step": 477000 + }, + { + "epoch": 2.71, + "learning_rate": 4.854981561684633e-06, + "loss": 0.1785, + "step": 477200 + }, + { + "epoch": 2.71, + "learning_rate": 4.836060766027333e-06, + "loss": 0.1822, + "step": 477400 + }, + { + "epoch": 2.71, + "learning_rate": 4.817139970370034e-06, + "loss": 0.1776, + "step": 477600 + }, + { + "epoch": 2.71, + "learning_rate": 4.798219174712735e-06, + "loss": 0.1801, + "step": 477800 + }, + { + "epoch": 2.71, + "learning_rate": 4.779298379055436e-06, + "loss": 0.1811, + "step": 478000 + }, + { + "epoch": 2.71, + "eval_LOC_f1": 0.818596091490667, + "eval_LOC_number": 22273, + "eval_LOC_precision": 0.7993667365538488, + "eval_LOC_recall": 0.83877340277466, + "eval_ORG_f1": 0.6865320944141835, + "eval_ORG_number": 19008, + "eval_ORG_precision": 0.6970288440685317, + "eval_ORG_recall": 0.6763468013468014, + "eval_PER_f1": 0.8150895603401483, + "eval_PER_number": 27575, + "eval_PER_precision": 0.8133237046398267, + "eval_PER_recall": 0.8168631006346329, + "eval_loss": 0.18688270449638367, + "eval_overall_accuracy": 0.9428407128677047, + "eval_overall_f1": 0.781449199947964, + "eval_overall_precision": 0.7777729823047044, + "eval_overall_recall": 0.7851603346113628, + "eval_runtime": 71.5336, + "eval_samples_per_second": 529.639, + "eval_steps_per_second": 66.207, + "step": 478000 + }, + { + "epoch": 2.71, + "learning_rate": 4.7603775833981375e-06, + "loss": 0.1745, + "step": 478200 + }, + { + "epoch": 2.72, + "learning_rate": 4.741456787740838e-06, + "loss": 0.1745, + "step": 478400 + }, + { + "epoch": 2.72, + "learning_rate": 4.722535992083539e-06, + "loss": 0.1777, + "step": 478600 + }, + { + "epoch": 2.72, + "learning_rate": 4.70361519642624e-06, + "loss": 0.1705, + "step": 478800 + }, + { + "epoch": 2.72, + "learning_rate": 4.684694400768942e-06, + "loss": 0.1826, + "step": 479000 + }, + { + "epoch": 2.72, + "learning_rate": 4.6657736051116425e-06, + "loss": 0.1693, + "step": 479200 + }, + { + "epoch": 2.72, + "learning_rate": 4.646852809454343e-06, + "loss": 0.1816, + "step": 479400 + }, + { + "epoch": 2.72, + "learning_rate": 4.627932013797044e-06, + "loss": 0.1771, + "step": 479600 + }, + { + "epoch": 2.72, + "learning_rate": 4.609011218139746e-06, + "loss": 0.1745, + "step": 479800 + }, + { + "epoch": 2.72, + "learning_rate": 4.590090422482447e-06, + "loss": 0.1806, + "step": 480000 + }, + { + "epoch": 2.72, + "eval_LOC_f1": 0.8184434270765206, + "eval_LOC_number": 22273, + "eval_LOC_precision": 0.7954824765860067, + "eval_LOC_recall": 0.8427692722129934, + "eval_ORG_f1": 0.6862776530939404, + "eval_ORG_number": 19008, + "eval_ORG_precision": 0.6972312703583062, + "eval_ORG_recall": 0.6756628787878788, + "eval_PER_f1": 0.8159543341291353, + "eval_PER_number": 27575, + "eval_PER_precision": 0.8078481552569844, + "eval_PER_recall": 0.8242248413417951, + "eval_loss": 0.1860371083021164, + "eval_overall_accuracy": 0.9428875567982659, + "eval_overall_f1": 0.7818599063356522, + "eval_overall_precision": 0.7746432695186098, + "eval_overall_recall": 0.7892122690833043, + "eval_runtime": 72.0458, + "eval_samples_per_second": 525.874, + "eval_steps_per_second": 65.736, + "step": 480000 + }, + { + "epoch": 2.73, + "learning_rate": 4.571169626825147e-06, + "loss": 0.175, + "step": 480200 + }, + { + "epoch": 2.73, + "learning_rate": 4.552248831167848e-06, + "loss": 0.1759, + "step": 480400 + }, + { + "epoch": 2.73, + "learning_rate": 4.5333280355105496e-06, + "loss": 0.1803, + "step": 480600 + }, + { + "epoch": 2.73, + "learning_rate": 4.514407239853251e-06, + "loss": 0.1778, + "step": 480800 + }, + { + "epoch": 2.73, + "learning_rate": 4.495486444195952e-06, + "loss": 0.1784, + "step": 481000 + }, + { + "epoch": 2.73, + "learning_rate": 4.476565648538652e-06, + "loss": 0.1747, + "step": 481200 + }, + { + "epoch": 2.73, + "learning_rate": 4.457644852881354e-06, + "loss": 0.1726, + "step": 481400 + }, + { + "epoch": 2.73, + "learning_rate": 4.4387240572240545e-06, + "loss": 0.1719, + "step": 481600 + }, + { + "epoch": 2.73, + "learning_rate": 4.419803261566756e-06, + "loss": 0.1806, + "step": 481800 + }, + { + "epoch": 2.74, + "learning_rate": 4.400882465909457e-06, + "loss": 0.1805, + "step": 482000 + }, + { + "epoch": 2.74, + "eval_LOC_f1": 0.8196757182556179, + "eval_LOC_number": 22273, + "eval_LOC_precision": 0.7994707187980195, + "eval_LOC_recall": 0.8409284784267947, + "eval_ORG_f1": 0.6855813215579472, + "eval_ORG_number": 19008, + "eval_ORG_precision": 0.6891712295997023, + "eval_ORG_recall": 0.6820286195286195, + "eval_PER_f1": 0.8149655209664931, + "eval_PER_number": 27575, + "eval_PER_precision": 0.8092683973396267, + "eval_PER_recall": 0.8207434270172258, + "eval_loss": 0.18566423654556274, + "eval_overall_accuracy": 0.9428600965631093, + "eval_overall_f1": 0.7813261805970041, + "eval_overall_precision": 0.7738195285236095, + "eval_overall_recall": 0.7889799000813291, + "eval_runtime": 72.1592, + "eval_samples_per_second": 525.047, + "eval_steps_per_second": 65.633, + "step": 482000 + }, + { + "epoch": 2.74, + "learning_rate": 4.381961670252157e-06, + "loss": 0.1759, + "step": 482200 + }, + { + "epoch": 2.74, + "learning_rate": 4.363040874594859e-06, + "loss": 0.1762, + "step": 482400 + }, + { + "epoch": 2.74, + "learning_rate": 4.344120078937559e-06, + "loss": 0.1744, + "step": 482600 + }, + { + "epoch": 2.74, + "learning_rate": 4.325199283280261e-06, + "loss": 0.1816, + "step": 482800 + }, + { + "epoch": 2.74, + "learning_rate": 4.3062784876229616e-06, + "loss": 0.1786, + "step": 483000 + }, + { + "epoch": 2.74, + "learning_rate": 4.287357691965663e-06, + "loss": 0.1805, + "step": 483200 + }, + { + "epoch": 2.74, + "learning_rate": 4.268436896308364e-06, + "loss": 0.1741, + "step": 483400 + }, + { + "epoch": 2.75, + "learning_rate": 4.249516100651064e-06, + "loss": 0.1758, + "step": 483600 + }, + { + "epoch": 2.75, + "learning_rate": 4.230595304993766e-06, + "loss": 0.1804, + "step": 483800 + }, + { + "epoch": 2.75, + "learning_rate": 4.211674509336467e-06, + "loss": 0.1803, + "step": 484000 + }, + { + "epoch": 2.75, + "eval_LOC_f1": 0.8195409374725177, + "eval_LOC_number": 22273, + "eval_LOC_precision": 0.802981345051915, + "eval_LOC_recall": 0.836797916760203, + "eval_ORG_f1": 0.6871041566874235, + "eval_ORG_number": 19008, + "eval_ORG_precision": 0.69520732364028, + "eval_ORG_recall": 0.6791877104377104, + "eval_PER_f1": 0.816050982933679, + "eval_PER_number": 27575, + "eval_PER_precision": 0.8102455939656097, + "eval_PER_recall": 0.8219401631912965, + "eval_loss": 0.1859697550535202, + "eval_overall_accuracy": 0.9430054742786439, + "eval_overall_f1": 0.7822379337710121, + "eval_overall_precision": 0.7772027410614445, + "eval_overall_recall": 0.7873387940048797, + "eval_runtime": 74.0716, + "eval_samples_per_second": 511.492, + "eval_steps_per_second": 63.938, + "step": 484000 + }, + { + "epoch": 2.75, + "learning_rate": 4.192753713679168e-06, + "loss": 0.1825, + "step": 484200 + }, + { + "epoch": 2.75, + "learning_rate": 4.173832918021869e-06, + "loss": 0.179, + "step": 484400 + }, + { + "epoch": 2.75, + "learning_rate": 4.154912122364569e-06, + "loss": 0.177, + "step": 484600 + }, + { + "epoch": 2.75, + "learning_rate": 4.135991326707271e-06, + "loss": 0.1767, + "step": 484800 + }, + { + "epoch": 2.75, + "learning_rate": 4.117070531049972e-06, + "loss": 0.1752, + "step": 485000 + }, + { + "epoch": 2.75, + "learning_rate": 4.098149735392673e-06, + "loss": 0.1842, + "step": 485200 + }, + { + "epoch": 2.76, + "learning_rate": 4.0792289397353736e-06, + "loss": 0.1729, + "step": 485400 + }, + { + "epoch": 2.76, + "learning_rate": 4.060308144078075e-06, + "loss": 0.1711, + "step": 485600 + }, + { + "epoch": 2.76, + "learning_rate": 4.0413873484207766e-06, + "loss": 0.1745, + "step": 485800 + }, + { + "epoch": 2.76, + "learning_rate": 4.022466552763477e-06, + "loss": 0.1848, + "step": 486000 + }, + { + "epoch": 2.76, + "eval_LOC_f1": 0.8187476794723393, + "eval_LOC_number": 22273, + "eval_LOC_precision": 0.7971421280939015, + "eval_LOC_recall": 0.8415570421586674, + "eval_ORG_f1": 0.6870007645058394, + "eval_ORG_number": 19008, + "eval_ORG_precision": 0.6885072655217965, + "eval_ORG_recall": 0.6855008417508418, + "eval_PER_f1": 0.8151733817998597, + "eval_PER_number": 27575, + "eval_PER_precision": 0.8090441491641663, + "eval_PER_recall": 0.8213961922030825, + "eval_loss": 0.1862962245941162, + "eval_overall_accuracy": 0.9428051760927962, + "eval_overall_f1": 0.7814431657465305, + "eval_overall_precision": 0.7726840349258182, + "eval_overall_recall": 0.7904031602184268, + "eval_runtime": 72.8703, + "eval_samples_per_second": 519.924, + "eval_steps_per_second": 64.992, + "step": 486000 + }, + { + "epoch": 2.76, + "learning_rate": 4.003545757106178e-06, + "loss": 0.1754, + "step": 486200 + }, + { + "epoch": 2.76, + "learning_rate": 3.9846249614488785e-06, + "loss": 0.1701, + "step": 486400 + }, + { + "epoch": 2.76, + "learning_rate": 3.96570416579158e-06, + "loss": 0.1744, + "step": 486600 + }, + { + "epoch": 2.76, + "learning_rate": 3.9467833701342815e-06, + "loss": 0.1784, + "step": 486800 + }, + { + "epoch": 2.76, + "learning_rate": 3.927862574476982e-06, + "loss": 0.1755, + "step": 487000 + }, + { + "epoch": 2.77, + "learning_rate": 3.908941778819683e-06, + "loss": 0.1763, + "step": 487200 + }, + { + "epoch": 2.77, + "learning_rate": 3.890020983162384e-06, + "loss": 0.1761, + "step": 487400 + }, + { + "epoch": 2.77, + "learning_rate": 3.871100187505086e-06, + "loss": 0.1772, + "step": 487600 + }, + { + "epoch": 2.77, + "learning_rate": 3.852179391847786e-06, + "loss": 0.1828, + "step": 487800 + }, + { + "epoch": 2.77, + "learning_rate": 3.833258596190487e-06, + "loss": 0.1774, + "step": 488000 + }, + { + "epoch": 2.77, + "eval_LOC_f1": 0.8184969069451147, + "eval_LOC_number": 22273, + "eval_LOC_precision": 0.8002402093252692, + "eval_LOC_recall": 0.8376060701297535, + "eval_ORG_f1": 0.6863635157017335, + "eval_ORG_number": 19008, + "eval_ORG_precision": 0.6927074961153084, + "eval_ORG_recall": 0.6801346801346801, + "eval_PER_f1": 0.8157752760493472, + "eval_PER_number": 27575, + "eval_PER_precision": 0.809139880846206, + "eval_PER_recall": 0.822520398912058, + "eval_loss": 0.18605853617191315, + "eval_overall_accuracy": 0.9429957824309415, + "eval_overall_f1": 0.7815616830977293, + "eval_overall_precision": 0.775136772037082, + "eval_overall_recall": 0.7880939932612989, + "eval_runtime": 74.1924, + "eval_samples_per_second": 510.659, + "eval_steps_per_second": 63.834, + "step": 488000 + }, + { + "epoch": 2.77, + "learning_rate": 3.814337800533188e-06, + "loss": 0.1762, + "step": 488200 + }, + { + "epoch": 2.77, + "learning_rate": 3.795417004875889e-06, + "loss": 0.1819, + "step": 488400 + }, + { + "epoch": 2.77, + "learning_rate": 3.7764962092185907e-06, + "loss": 0.1736, + "step": 488600 + }, + { + "epoch": 2.77, + "learning_rate": 3.7575754135612914e-06, + "loss": 0.1782, + "step": 488800 + }, + { + "epoch": 2.78, + "learning_rate": 3.7386546179039924e-06, + "loss": 0.1819, + "step": 489000 + }, + { + "epoch": 2.78, + "learning_rate": 3.719733822246693e-06, + "loss": 0.1714, + "step": 489200 + }, + { + "epoch": 2.78, + "learning_rate": 3.700813026589394e-06, + "loss": 0.1693, + "step": 489400 + }, + { + "epoch": 2.78, + "learning_rate": 3.6818922309320956e-06, + "loss": 0.1689, + "step": 489600 + }, + { + "epoch": 2.78, + "learning_rate": 3.6629714352747967e-06, + "loss": 0.1791, + "step": 489800 + }, + { + "epoch": 2.78, + "learning_rate": 3.6440506396174974e-06, + "loss": 0.1804, + "step": 490000 + }, + { + "epoch": 2.78, + "eval_LOC_f1": 0.8190877562667369, + "eval_LOC_number": 22273, + "eval_LOC_precision": 0.8012883830792356, + "eval_LOC_recall": 0.8376958649485925, + "eval_ORG_f1": 0.6862974380780269, + "eval_ORG_number": 19008, + "eval_ORG_precision": 0.6934479054779806, + "eval_ORG_recall": 0.6792929292929293, + "eval_PER_f1": 0.8152275505068661, + "eval_PER_number": 27575, + "eval_PER_precision": 0.8081679198888136, + "eval_PER_recall": 0.8224116047144152, + "eval_loss": 0.18630899488925934, + "eval_overall_accuracy": 0.9429747834275866, + "eval_overall_f1": 0.7815475933209433, + "eval_overall_precision": 0.7753480261841466, + "eval_overall_recall": 0.7878471011967003, + "eval_runtime": 73.2852, + "eval_samples_per_second": 516.98, + "eval_steps_per_second": 64.624, + "step": 490000 + }, + { + "epoch": 2.78, + "learning_rate": 3.6251298439601984e-06, + "loss": 0.1776, + "step": 490200 + }, + { + "epoch": 2.78, + "learning_rate": 3.606209048302899e-06, + "loss": 0.1675, + "step": 490400 + }, + { + "epoch": 2.78, + "learning_rate": 3.5872882526456006e-06, + "loss": 0.1801, + "step": 490600 + }, + { + "epoch": 2.79, + "learning_rate": 3.5683674569883016e-06, + "loss": 0.167, + "step": 490800 + }, + { + "epoch": 2.79, + "learning_rate": 3.5494466613310023e-06, + "loss": 0.1704, + "step": 491000 + }, + { + "epoch": 2.79, + "learning_rate": 3.5305258656737034e-06, + "loss": 0.1829, + "step": 491200 + }, + { + "epoch": 2.79, + "learning_rate": 3.5116050700164044e-06, + "loss": 0.1842, + "step": 491400 + }, + { + "epoch": 2.79, + "learning_rate": 3.492684274359106e-06, + "loss": 0.1758, + "step": 491600 + }, + { + "epoch": 2.79, + "learning_rate": 3.4737634787018066e-06, + "loss": 0.1742, + "step": 491800 + }, + { + "epoch": 2.79, + "learning_rate": 3.4548426830445076e-06, + "loss": 0.1758, + "step": 492000 + }, + { + "epoch": 2.79, + "eval_LOC_f1": 0.8202220024178482, + "eval_LOC_number": 22273, + "eval_LOC_precision": 0.8034622340883645, + "eval_LOC_recall": 0.8376958649485925, + "eval_ORG_f1": 0.6879974614591322, + "eval_ORG_number": 19008, + "eval_ORG_precision": 0.6916369822957095, + "eval_ORG_recall": 0.6843960437710438, + "eval_PER_f1": 0.8147536866276776, + "eval_PER_number": 27575, + "eval_PER_precision": 0.8061916427024532, + "eval_PER_recall": 0.8234995466908431, + "eval_loss": 0.1859225183725357, + "eval_overall_accuracy": 0.943086239676163, + "eval_overall_f1": 0.7820702748572498, + "eval_overall_precision": 0.7745947177982279, + "eval_overall_recall": 0.789691530149878, + "eval_runtime": 72.1174, + "eval_samples_per_second": 525.352, + "eval_steps_per_second": 65.671, + "step": 492000 + }, + { + "epoch": 2.79, + "learning_rate": 3.4359218873872083e-06, + "loss": 0.1717, + "step": 492200 + }, + { + "epoch": 2.79, + "learning_rate": 3.4170010917299094e-06, + "loss": 0.1757, + "step": 492400 + }, + { + "epoch": 2.8, + "learning_rate": 3.398080296072611e-06, + "loss": 0.1879, + "step": 492600 + }, + { + "epoch": 2.8, + "learning_rate": 3.379159500415312e-06, + "loss": 0.1811, + "step": 492800 + }, + { + "epoch": 2.8, + "learning_rate": 3.3602387047580126e-06, + "loss": 0.1746, + "step": 493000 + }, + { + "epoch": 2.8, + "learning_rate": 3.3413179091007136e-06, + "loss": 0.1714, + "step": 493200 + }, + { + "epoch": 2.8, + "learning_rate": 3.3223971134434143e-06, + "loss": 0.1821, + "step": 493400 + }, + { + "epoch": 2.8, + "learning_rate": 3.3034763177861158e-06, + "loss": 0.1722, + "step": 493600 + }, + { + "epoch": 2.8, + "learning_rate": 3.284555522128817e-06, + "loss": 0.1815, + "step": 493800 + }, + { + "epoch": 2.8, + "learning_rate": 3.265634726471518e-06, + "loss": 0.1728, + "step": 494000 + }, + { + "epoch": 2.8, + "eval_LOC_f1": 0.8192085449133253, + "eval_LOC_number": 22273, + "eval_LOC_precision": 0.799231262011531, + "eval_LOC_recall": 0.8402101198760832, + "eval_ORG_f1": 0.6876497843105928, + "eval_ORG_number": 19008, + "eval_ORG_precision": 0.6962148172112584, + "eval_ORG_recall": 0.6792929292929293, + "eval_PER_f1": 0.8163287297282736, + "eval_PER_number": 27575, + "eval_PER_precision": 0.8086173770725112, + "eval_PER_recall": 0.8241885766092475, + "eval_loss": 0.18620188534259796, + "eval_overall_accuracy": 0.943018396742247, + "eval_overall_f1": 0.7824910202054375, + "eval_overall_precision": 0.7757289451524969, + "eval_overall_recall": 0.7893720227721622, + "eval_runtime": 72.2302, + "eval_samples_per_second": 524.531, + "eval_steps_per_second": 65.568, + "step": 494000 + }, + { + "epoch": 2.81, + "learning_rate": 3.2467139308142186e-06, + "loss": 0.1753, + "step": 494200 + }, + { + "epoch": 2.81, + "learning_rate": 3.2277931351569196e-06, + "loss": 0.1764, + "step": 494400 + }, + { + "epoch": 2.81, + "learning_rate": 3.208872339499621e-06, + "loss": 0.1774, + "step": 494600 + }, + { + "epoch": 2.81, + "learning_rate": 3.189951543842322e-06, + "loss": 0.1815, + "step": 494800 + }, + { + "epoch": 2.81, + "learning_rate": 3.171030748185023e-06, + "loss": 0.1766, + "step": 495000 + }, + { + "epoch": 2.81, + "learning_rate": 3.1521099525277235e-06, + "loss": 0.1898, + "step": 495200 + }, + { + "epoch": 2.81, + "learning_rate": 3.1331891568704246e-06, + "loss": 0.1729, + "step": 495400 + }, + { + "epoch": 2.81, + "learning_rate": 3.1142683612131256e-06, + "loss": 0.1743, + "step": 495600 + }, + { + "epoch": 2.81, + "learning_rate": 3.095347565555827e-06, + "loss": 0.1745, + "step": 495800 + }, + { + "epoch": 2.82, + "learning_rate": 3.076426769898528e-06, + "loss": 0.1767, + "step": 496000 + }, + { + "epoch": 2.82, + "eval_LOC_f1": 0.8190934968250493, + "eval_LOC_number": 22273, + "eval_LOC_precision": 0.7994187288968672, + "eval_LOC_recall": 0.8397611457818884, + "eval_ORG_f1": 0.6881081797693226, + "eval_ORG_number": 19008, + "eval_ORG_precision": 0.6936440904474261, + "eval_ORG_recall": 0.6826599326599326, + "eval_PER_f1": 0.816561242093157, + "eval_PER_number": 27575, + "eval_PER_precision": 0.8093185623196666, + "eval_PER_recall": 0.8239347234814143, + "eval_loss": 0.18628109991550446, + "eval_overall_accuracy": 0.9430216273581477, + "eval_overall_f1": 0.782548028165975, + "eval_overall_precision": 0.7751827521837639, + "eval_overall_recall": 0.7900546067154641, + "eval_runtime": 73.7595, + "eval_samples_per_second": 513.656, + "eval_steps_per_second": 64.209, + "step": 496000 + }, + { + "epoch": 2.82, + "learning_rate": 3.0575059742412293e-06, + "loss": 0.1763, + "step": 496200 + }, + { + "epoch": 2.82, + "learning_rate": 3.03858517858393e-06, + "loss": 0.1763, + "step": 496400 + }, + { + "epoch": 2.82, + "learning_rate": 3.019664382926631e-06, + "loss": 0.17, + "step": 496600 + }, + { + "epoch": 2.82, + "learning_rate": 3.000743587269332e-06, + "loss": 0.1762, + "step": 496800 + }, + { + "epoch": 2.82, + "learning_rate": 2.981822791612033e-06, + "loss": 0.1796, + "step": 497000 + }, + { + "epoch": 2.82, + "learning_rate": 2.9629019959547342e-06, + "loss": 0.1787, + "step": 497200 + }, + { + "epoch": 2.82, + "learning_rate": 2.943981200297435e-06, + "loss": 0.1698, + "step": 497400 + }, + { + "epoch": 2.82, + "learning_rate": 2.925060404640136e-06, + "loss": 0.1704, + "step": 497600 + }, + { + "epoch": 2.83, + "learning_rate": 2.906139608982837e-06, + "loss": 0.1921, + "step": 497800 + }, + { + "epoch": 2.83, + "learning_rate": 2.887218813325538e-06, + "loss": 0.1764, + "step": 498000 + }, + { + "epoch": 2.83, + "eval_LOC_f1": 0.8203545633617858, + "eval_LOC_number": 22273, + "eval_LOC_precision": 0.8003160097365162, + "eval_LOC_recall": 0.841422349930409, + "eval_ORG_f1": 0.6879614767255217, + "eval_ORG_number": 19008, + "eval_ORG_precision": 0.6998693664271718, + "eval_ORG_recall": 0.6764520202020202, + "eval_PER_f1": 0.816058918627627, + "eval_PER_number": 27575, + "eval_PER_precision": 0.808506851752981, + "eval_PER_recall": 0.8237533998186763, + "eval_loss": 0.18625010550022125, + "eval_overall_accuracy": 0.9431799275372853, + "eval_overall_f1": 0.7829609341213781, + "eval_overall_precision": 0.7772022208230782, + "eval_overall_recall": 0.7888056233298478, + "eval_runtime": 72.1951, + "eval_samples_per_second": 524.786, + "eval_steps_per_second": 65.6, + "step": 498000 + }, + { + "epoch": 2.83, + "learning_rate": 2.868298017668239e-06, + "loss": 0.1745, + "step": 498200 + }, + { + "epoch": 2.83, + "learning_rate": 2.8493772220109402e-06, + "loss": 0.187, + "step": 498400 + }, + { + "epoch": 2.83, + "learning_rate": 2.830456426353641e-06, + "loss": 0.1719, + "step": 498600 + }, + { + "epoch": 2.83, + "learning_rate": 2.8115356306963424e-06, + "loss": 0.1726, + "step": 498800 + }, + { + "epoch": 2.83, + "learning_rate": 2.792614835039043e-06, + "loss": 0.1733, + "step": 499000 + }, + { + "epoch": 2.83, + "learning_rate": 2.7736940393817445e-06, + "loss": 0.1747, + "step": 499200 + }, + { + "epoch": 2.83, + "learning_rate": 2.754773243724445e-06, + "loss": 0.1808, + "step": 499400 + }, + { + "epoch": 2.84, + "learning_rate": 2.7358524480671462e-06, + "loss": 0.1796, + "step": 499600 + }, + { + "epoch": 2.84, + "learning_rate": 2.7169316524098473e-06, + "loss": 0.1714, + "step": 499800 + }, + { + "epoch": 2.84, + "learning_rate": 2.6980108567525484e-06, + "loss": 0.1833, + "step": 500000 + }, + { + "epoch": 2.84, + "eval_LOC_f1": 0.8201758241758241, + "eval_LOC_number": 22273, + "eval_LOC_precision": 0.8033323287553279, + "eval_LOC_recall": 0.837740762358012, + "eval_ORG_f1": 0.6878868729989327, + "eval_ORG_number": 19008, + "eval_ORG_precision": 0.697867042009528, + "eval_ORG_recall": 0.6781881313131313, + "eval_PER_f1": 0.8162261169372217, + "eval_PER_number": 27575, + "eval_PER_precision": 0.8084160352861666, + "eval_PER_recall": 0.8241885766092475, + "eval_loss": 0.1851131021976471, + "eval_overall_accuracy": 0.9431912346929381, + "eval_overall_f1": 0.7828338189055875, + "eval_overall_precision": 0.7774737867415344, + "eval_overall_recall": 0.7882682700127803, + "eval_runtime": 71.9489, + "eval_samples_per_second": 526.582, + "eval_steps_per_second": 65.825, + "step": 500000 + }, + { + "epoch": 2.84, + "learning_rate": 2.6790900610952494e-06, + "loss": 0.1737, + "step": 500200 + }, + { + "epoch": 2.84, + "learning_rate": 2.66016926543795e-06, + "loss": 0.1802, + "step": 500400 + }, + { + "epoch": 2.84, + "learning_rate": 2.641248469780651e-06, + "loss": 0.1728, + "step": 500600 + }, + { + "epoch": 2.84, + "learning_rate": 2.6223276741233522e-06, + "loss": 0.1812, + "step": 500800 + }, + { + "epoch": 2.84, + "learning_rate": 2.6034068784660533e-06, + "loss": 0.1742, + "step": 501000 + }, + { + "epoch": 2.84, + "learning_rate": 2.5844860828087544e-06, + "loss": 0.1867, + "step": 501200 + }, + { + "epoch": 2.85, + "learning_rate": 2.5655652871514554e-06, + "loss": 0.1718, + "step": 501400 + }, + { + "epoch": 2.85, + "learning_rate": 2.5466444914941565e-06, + "loss": 0.1819, + "step": 501600 + }, + { + "epoch": 2.85, + "learning_rate": 2.5277236958368576e-06, + "loss": 0.1769, + "step": 501800 + }, + { + "epoch": 2.85, + "learning_rate": 2.5088029001795582e-06, + "loss": 0.1789, + "step": 502000 + }, + { + "epoch": 2.85, + "eval_LOC_f1": 0.8195670311333596, + "eval_LOC_number": 22273, + "eval_LOC_precision": 0.8019593520388433, + "eval_LOC_recall": 0.8379652494051093, + "eval_ORG_f1": 0.6884627642621906, + "eval_ORG_number": 19008, + "eval_ORG_precision": 0.6951834370306801, + "eval_ORG_recall": 0.6818707912457912, + "eval_PER_f1": 0.8167582516251842, + "eval_PER_number": 27575, + "eval_PER_precision": 0.8089715769627548, + "eval_PER_recall": 0.8246962828649139, + "eval_loss": 0.1853255182504654, + "eval_overall_accuracy": 0.9432170796201442, + "eval_overall_f1": 0.7828979580081219, + "eval_overall_precision": 0.7763466042154566, + "eval_overall_recall": 0.789560822586267, + "eval_runtime": 74.1919, + "eval_samples_per_second": 510.662, + "eval_steps_per_second": 63.834, + "step": 502000 + }, + { + "epoch": 2.85, + "learning_rate": 2.4898821045222597e-06, + "loss": 0.1724, + "step": 502200 + }, + { + "epoch": 2.85, + "learning_rate": 2.4709613088649604e-06, + "loss": 0.1826, + "step": 502400 + }, + { + "epoch": 2.85, + "learning_rate": 2.452040513207662e-06, + "loss": 0.1731, + "step": 502600 + }, + { + "epoch": 2.85, + "learning_rate": 2.4331197175503625e-06, + "loss": 0.1827, + "step": 502800 + }, + { + "epoch": 2.86, + "learning_rate": 2.4141989218930636e-06, + "loss": 0.167, + "step": 503000 + }, + { + "epoch": 2.86, + "learning_rate": 2.3952781262357647e-06, + "loss": 0.1859, + "step": 503200 + }, + { + "epoch": 2.86, + "learning_rate": 2.3763573305784657e-06, + "loss": 0.1788, + "step": 503400 + }, + { + "epoch": 2.86, + "learning_rate": 2.357436534921167e-06, + "loss": 0.1787, + "step": 503600 + }, + { + "epoch": 2.86, + "learning_rate": 2.3385157392638674e-06, + "loss": 0.1819, + "step": 503800 + }, + { + "epoch": 2.86, + "learning_rate": 2.3195949436065685e-06, + "loss": 0.1802, + "step": 504000 + }, + { + "epoch": 2.86, + "eval_LOC_f1": 0.8193308550185874, + "eval_LOC_number": 22273, + "eval_LOC_precision": 0.7986528541586733, + "eval_LOC_recall": 0.8411080680644727, + "eval_ORG_f1": 0.6884396613880952, + "eval_ORG_number": 19008, + "eval_ORG_precision": 0.6990617712457292, + "eval_ORG_recall": 0.6781355218855218, + "eval_PER_f1": 0.8165316045380875, + "eval_PER_number": 27575, + "eval_PER_precision": 0.8109819352530853, + "eval_PER_recall": 0.822157751586582, + "eval_loss": 0.1856091469526291, + "eval_overall_accuracy": 0.9431670050736822, + "eval_overall_f1": 0.78287325080926, + "eval_overall_precision": 0.7772973901590529, + "eval_overall_recall": 0.7885296851400023, + "eval_runtime": 73.5203, + "eval_samples_per_second": 515.327, + "eval_steps_per_second": 64.418, + "step": 504000 + }, + { + "epoch": 2.86, + "learning_rate": 2.3006741479492696e-06, + "loss": 0.1725, + "step": 504200 + }, + { + "epoch": 2.86, + "learning_rate": 2.2817533522919707e-06, + "loss": 0.1775, + "step": 504400 + }, + { + "epoch": 2.86, + "learning_rate": 2.2628325566346717e-06, + "loss": 0.1868, + "step": 504600 + }, + { + "epoch": 2.87, + "learning_rate": 2.243911760977373e-06, + "loss": 0.1805, + "step": 504800 + }, + { + "epoch": 2.87, + "learning_rate": 2.2249909653200734e-06, + "loss": 0.1651, + "step": 505000 + }, + { + "epoch": 2.87, + "learning_rate": 2.206070169662775e-06, + "loss": 0.1692, + "step": 505200 + }, + { + "epoch": 2.87, + "learning_rate": 2.1871493740054756e-06, + "loss": 0.1778, + "step": 505400 + }, + { + "epoch": 2.87, + "learning_rate": 2.168228578348177e-06, + "loss": 0.1798, + "step": 505600 + }, + { + "epoch": 2.87, + "learning_rate": 2.1493077826908777e-06, + "loss": 0.1826, + "step": 505800 + }, + { + "epoch": 2.87, + "learning_rate": 2.130386987033579e-06, + "loss": 0.1848, + "step": 506000 + }, + { + "epoch": 2.87, + "eval_LOC_f1": 0.8192929714072547, + "eval_LOC_number": 22273, + "eval_LOC_precision": 0.8012704953214869, + "eval_LOC_recall": 0.8381448390427872, + "eval_ORG_f1": 0.6877944325481798, + "eval_ORG_number": 19008, + "eval_ORG_precision": 0.7000871839581517, + "eval_ORG_recall": 0.6759259259259259, + "eval_PER_f1": 0.8160437622586508, + "eval_PER_number": 27575, + "eval_PER_precision": 0.8098792770912208, + "eval_PER_recall": 0.8223028105167725, + "eval_loss": 0.18502268195152283, + "eval_overall_accuracy": 0.9432396939314496, + "eval_overall_f1": 0.7825189164211864, + "eval_overall_precision": 0.7780697220307834, + "eval_overall_recall": 0.7870192866271639, + "eval_runtime": 71.4188, + "eval_samples_per_second": 530.491, + "eval_steps_per_second": 66.313, + "step": 506000 + } + ], + "max_steps": 528519, + "num_train_epochs": 3, + "total_flos": 9.372634549859451e+17, + "trial_name": null, + "trial_params": null +}