{ "best_metric": null, "best_model_checkpoint": null, "epoch": 100.0, "eval_steps": 500, "global_step": 9600, "is_hyper_param_search": false, "is_local_process_zero": true, "is_world_process_zero": true, "log_history": [ { "epoch": 1.0, "grad_norm": 1.1241459846496582, "learning_rate": 4.9500000000000004e-05, "loss": 0.8279, "step": 96 }, { "epoch": 1.0, "eval_LOCATION_f1": 0.06153846153846154, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.2857142857142857, "eval_LOCATION_recall": 0.034482758620689655, "eval_ORGANIZATION_f1": 0.2803738317757009, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.2222222222222222, "eval_ORGANIZATION_recall": 0.379746835443038, "eval_PERSON_f1": 0.27536231884057977, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.25, "eval_PERSON_recall": 0.3064516129032258, "eval_loss": 0.38678956031799316, "eval_overall_accuracy": 0.8676914630798792, "eval_overall_f1": 0.2446043165467626, "eval_overall_precision": 0.23394495412844038, "eval_overall_recall": 0.2562814070351759, "eval_runtime": 0.6078, "eval_samples_per_second": 279.698, "eval_steps_per_second": 4.936, "step": 96 }, { "epoch": 2.0, "grad_norm": 1.16001296043396, "learning_rate": 4.9e-05, "loss": 0.3613, "step": 192 }, { "epoch": 2.0, "eval_LOCATION_f1": 0.3918367346938776, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.37209302325581395, "eval_LOCATION_recall": 0.41379310344827586, "eval_ORGANIZATION_f1": 0.6139817629179332, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.5906432748538012, "eval_ORGANIZATION_recall": 0.6392405063291139, "eval_PERSON_f1": 0.6046511627906977, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.5141242937853108, "eval_PERSON_recall": 0.7338709677419355, "eval_loss": 0.20645824074745178, "eval_overall_accuracy": 0.9346692286576997, "eval_overall_f1": 0.5485714285714286, "eval_overall_precision": 0.5031446540880503, "eval_overall_recall": 0.6030150753768844, "eval_runtime": 0.6374, "eval_samples_per_second": 266.713, "eval_steps_per_second": 4.707, "step": 192 }, { "epoch": 3.0, "grad_norm": 1.1462738513946533, "learning_rate": 4.85e-05, "loss": 0.1915, "step": 288 }, { "epoch": 3.0, "eval_LOCATION_f1": 0.7945205479452053, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.8446601941747572, "eval_LOCATION_recall": 0.75, "eval_ORGANIZATION_f1": 0.7861271676300579, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.723404255319149, "eval_ORGANIZATION_recall": 0.8607594936708861, "eval_PERSON_f1": 0.9531249999999999, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.9242424242424242, "eval_PERSON_recall": 0.9838709677419355, "eval_loss": 0.0954388678073883, "eval_overall_accuracy": 0.9703541037606368, "eval_overall_f1": 0.8404384896467721, "eval_overall_precision": 0.8156028368794326, "eval_overall_recall": 0.8668341708542714, "eval_runtime": 0.6679, "eval_samples_per_second": 254.517, "eval_steps_per_second": 4.491, "step": 288 }, { "epoch": 4.0, "grad_norm": 1.801932692527771, "learning_rate": 4.8e-05, "loss": 0.1356, "step": 384 }, { "epoch": 4.0, "eval_LOCATION_f1": 0.8267716535433071, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.7608695652173914, "eval_LOCATION_recall": 0.9051724137931034, "eval_ORGANIZATION_f1": 0.8338368580060425, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.7976878612716763, "eval_ORGANIZATION_recall": 0.8734177215189873, "eval_PERSON_f1": 0.9685039370078741, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.9461538461538461, "eval_PERSON_recall": 0.9919354838709677, "eval_loss": 0.08161529898643494, "eval_overall_accuracy": 0.9717265989569036, "eval_overall_f1": 0.872467222884386, "eval_overall_precision": 0.8299319727891157, "eval_overall_recall": 0.9195979899497487, "eval_runtime": 0.6601, "eval_samples_per_second": 257.52, "eval_steps_per_second": 4.544, "step": 384 }, { "epoch": 5.0, "grad_norm": 1.1523765325546265, "learning_rate": 4.75e-05, "loss": 0.1094, "step": 480 }, { "epoch": 5.0, "eval_LOCATION_f1": 0.881057268722467, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.9009009009009009, "eval_LOCATION_recall": 0.8620689655172413, "eval_ORGANIZATION_f1": 0.8706624605678234, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.8679245283018868, "eval_ORGANIZATION_recall": 0.8734177215189873, "eval_PERSON_f1": 0.9800796812749005, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.968503937007874, "eval_PERSON_recall": 0.9919354838709677, "eval_loss": 0.05813028663396835, "eval_overall_accuracy": 0.981059566291518, "eval_overall_f1": 0.9081761006289307, "eval_overall_precision": 0.9093198992443325, "eval_overall_recall": 0.907035175879397, "eval_runtime": 0.6364, "eval_samples_per_second": 267.142, "eval_steps_per_second": 4.714, "step": 480 }, { "epoch": 6.0, "grad_norm": 0.9123748540878296, "learning_rate": 4.7e-05, "loss": 0.1026, "step": 576 }, { "epoch": 6.0, "eval_LOCATION_f1": 0.8806584362139918, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.84251968503937, "eval_LOCATION_recall": 0.9224137931034483, "eval_ORGANIZATION_f1": 0.9078947368421053, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.9452054794520548, "eval_ORGANIZATION_recall": 0.8734177215189873, "eval_PERSON_f1": 0.9761904761904763, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.9609375, "eval_PERSON_recall": 0.9919354838709677, "eval_loss": 0.05239282548427582, "eval_overall_accuracy": 0.9838045566840516, "eval_overall_f1": 0.9211514392991239, "eval_overall_precision": 0.9177057356608479, "eval_overall_recall": 0.9246231155778895, "eval_runtime": 0.6848, "eval_samples_per_second": 248.24, "eval_steps_per_second": 4.381, "step": 576 }, { "epoch": 7.0, "grad_norm": 0.8782280087471008, "learning_rate": 4.6500000000000005e-05, "loss": 0.0877, "step": 672 }, { "epoch": 7.0, "eval_LOCATION_f1": 0.8849557522123893, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.9090909090909091, "eval_LOCATION_recall": 0.8620689655172413, "eval_ORGANIZATION_f1": 0.9024390243902438, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.8705882352941177, "eval_ORGANIZATION_recall": 0.9367088607594937, "eval_PERSON_f1": 0.9840000000000001, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.9761904761904762, "eval_PERSON_recall": 0.9919354838709677, "eval_loss": 0.04487282410264015, "eval_overall_accuracy": 0.9843535547625584, "eval_overall_f1": 0.9228855721393034, "eval_overall_precision": 0.9137931034482759, "eval_overall_recall": 0.9321608040201005, "eval_runtime": 0.614, "eval_samples_per_second": 276.895, "eval_steps_per_second": 4.886, "step": 672 }, { "epoch": 8.0, "grad_norm": 0.32933393120765686, "learning_rate": 4.600000000000001e-05, "loss": 0.0774, "step": 768 }, { "epoch": 8.0, "eval_LOCATION_f1": 0.9288702928870294, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.9024390243902439, "eval_LOCATION_recall": 0.9568965517241379, "eval_ORGANIZATION_f1": 0.9206349206349206, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.9235668789808917, "eval_ORGANIZATION_recall": 0.9177215189873418, "eval_PERSON_f1": 0.9840000000000001, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.9761904761904762, "eval_PERSON_recall": 0.9919354838709677, "eval_loss": 0.041816890239715576, "eval_overall_accuracy": 0.9862750480373319, "eval_overall_f1": 0.9427860696517413, "eval_overall_precision": 0.9334975369458128, "eval_overall_recall": 0.9522613065326633, "eval_runtime": 0.6235, "eval_samples_per_second": 272.633, "eval_steps_per_second": 4.811, "step": 768 }, { "epoch": 9.0, "grad_norm": 0.42530685663223267, "learning_rate": 4.55e-05, "loss": 0.0735, "step": 864 }, { "epoch": 9.0, "eval_LOCATION_f1": 0.9152542372881356, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.9, "eval_LOCATION_recall": 0.9310344827586207, "eval_ORGANIZATION_f1": 0.937888198757764, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.9207317073170732, "eval_ORGANIZATION_recall": 0.9556962025316456, "eval_PERSON_f1": 0.9840000000000001, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.9761904761904762, "eval_PERSON_recall": 0.9919354838709677, "eval_loss": 0.036840181797742844, "eval_overall_accuracy": 0.9887455393906122, "eval_overall_f1": 0.9455445544554456, "eval_overall_precision": 0.9317073170731708, "eval_overall_recall": 0.9597989949748744, "eval_runtime": 0.63, "eval_samples_per_second": 269.862, "eval_steps_per_second": 4.762, "step": 864 }, { "epoch": 10.0, "grad_norm": 0.7166396379470825, "learning_rate": 4.5e-05, "loss": 0.0657, "step": 960 }, { "epoch": 10.0, "eval_LOCATION_f1": 0.9113924050632911, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.8925619834710744, "eval_LOCATION_recall": 0.9310344827586207, "eval_ORGANIZATION_f1": 0.9363057324840764, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.9423076923076923, "eval_ORGANIZATION_recall": 0.930379746835443, "eval_PERSON_f1": 0.9685039370078741, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.9461538461538461, "eval_PERSON_recall": 0.9919354838709677, "eval_loss": 0.041209977120161057, "eval_overall_accuracy": 0.9865495470765853, "eval_overall_f1": 0.9391304347826087, "eval_overall_precision": 0.9287469287469288, "eval_overall_recall": 0.949748743718593, "eval_runtime": 0.6304, "eval_samples_per_second": 269.682, "eval_steps_per_second": 4.759, "step": 960 }, { "epoch": 11.0, "grad_norm": 0.4892464280128479, "learning_rate": 4.4500000000000004e-05, "loss": 0.0605, "step": 1056 }, { "epoch": 11.0, "eval_LOCATION_f1": 0.9367088607594937, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.9173553719008265, "eval_LOCATION_recall": 0.9568965517241379, "eval_ORGANIZATION_f1": 0.9483870967741935, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.9671052631578947, "eval_ORGANIZATION_recall": 0.930379746835443, "eval_PERSON_f1": 0.9840000000000001, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.9761904761904762, "eval_PERSON_recall": 0.9919354838709677, "eval_loss": 0.03784897178411484, "eval_overall_accuracy": 0.9898435355476256, "eval_overall_f1": 0.9560853199498118, "eval_overall_precision": 0.9548872180451128, "eval_overall_recall": 0.957286432160804, "eval_runtime": 0.6318, "eval_samples_per_second": 269.083, "eval_steps_per_second": 4.749, "step": 1056 }, { "epoch": 12.0, "grad_norm": 0.5705476403236389, "learning_rate": 4.4000000000000006e-05, "loss": 0.0567, "step": 1152 }, { "epoch": 12.0, "eval_LOCATION_f1": 0.8936170212765958, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.8823529411764706, "eval_LOCATION_recall": 0.9051724137931034, "eval_ORGANIZATION_f1": 0.9433962264150944, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.9375, "eval_ORGANIZATION_recall": 0.9493670886075949, "eval_PERSON_f1": 0.9761904761904763, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.9609375, "eval_PERSON_recall": 0.9919354838709677, "eval_loss": 0.03704560920596123, "eval_overall_accuracy": 0.9881965413121054, "eval_overall_f1": 0.9391304347826087, "eval_overall_precision": 0.9287469287469288, "eval_overall_recall": 0.949748743718593, "eval_runtime": 0.6389, "eval_samples_per_second": 266.097, "eval_steps_per_second": 4.696, "step": 1152 }, { "epoch": 13.0, "grad_norm": 0.22472934424877167, "learning_rate": 4.35e-05, "loss": 0.0546, "step": 1248 }, { "epoch": 13.0, "eval_LOCATION_f1": 0.9276595744680851, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.9159663865546218, "eval_LOCATION_recall": 0.9396551724137931, "eval_ORGANIZATION_f1": 0.9426751592356688, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.9487179487179487, "eval_ORGANIZATION_recall": 0.9367088607594937, "eval_PERSON_f1": 0.9606299212598426, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.9384615384615385, "eval_PERSON_recall": 0.9838709677419355, "eval_loss": 0.036021873354911804, "eval_overall_accuracy": 0.9898435355476256, "eval_overall_f1": 0.9439601494396015, "eval_overall_precision": 0.9358024691358025, "eval_overall_recall": 0.9522613065326633, "eval_runtime": 0.6226, "eval_samples_per_second": 273.06, "eval_steps_per_second": 4.819, "step": 1248 }, { "epoch": 14.0, "grad_norm": 0.7680674195289612, "learning_rate": 4.3e-05, "loss": 0.0504, "step": 1344 }, { "epoch": 14.0, "eval_LOCATION_f1": 0.9406779661016949, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.925, "eval_LOCATION_recall": 0.9568965517241379, "eval_ORGANIZATION_f1": 0.9483870967741935, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.9671052631578947, "eval_ORGANIZATION_recall": 0.930379746835443, "eval_PERSON_f1": 0.9761904761904763, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.9609375, "eval_PERSON_recall": 0.9919354838709677, "eval_loss": 0.036000605672597885, "eval_overall_accuracy": 0.9906670326653857, "eval_overall_f1": 0.9548872180451128, "eval_overall_precision": 0.9525, "eval_overall_recall": 0.957286432160804, "eval_runtime": 0.6275, "eval_samples_per_second": 270.932, "eval_steps_per_second": 4.781, "step": 1344 }, { "epoch": 15.0, "grad_norm": 0.4943349361419678, "learning_rate": 4.25e-05, "loss": 0.0484, "step": 1440 }, { "epoch": 15.0, "eval_LOCATION_f1": 0.9322033898305084, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.9166666666666666, "eval_LOCATION_recall": 0.9482758620689655, "eval_ORGANIZATION_f1": 0.9487179487179488, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.961038961038961, "eval_ORGANIZATION_recall": 0.9367088607594937, "eval_PERSON_f1": 0.9800796812749005, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.968503937007874, "eval_PERSON_recall": 0.9919354838709677, "eval_loss": 0.03694455325603485, "eval_overall_accuracy": 0.990118034586879, "eval_overall_f1": 0.9536921151439299, "eval_overall_precision": 0.9501246882793017, "eval_overall_recall": 0.957286432160804, "eval_runtime": 0.6341, "eval_samples_per_second": 268.103, "eval_steps_per_second": 4.731, "step": 1440 }, { "epoch": 16.0, "grad_norm": 0.4439069330692291, "learning_rate": 4.2e-05, "loss": 0.0449, "step": 1536 }, { "epoch": 16.0, "eval_LOCATION_f1": 0.9075630252100839, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.8852459016393442, "eval_LOCATION_recall": 0.9310344827586207, "eval_ORGANIZATION_f1": 0.9483870967741935, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.9671052631578947, "eval_ORGANIZATION_recall": 0.930379746835443, "eval_PERSON_f1": 0.976, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.9682539682539683, "eval_PERSON_recall": 0.9838709677419355, "eval_loss": 0.0340009480714798, "eval_overall_accuracy": 0.9884710403513588, "eval_overall_f1": 0.9448621553884713, "eval_overall_precision": 0.9425, "eval_overall_recall": 0.9472361809045227, "eval_runtime": 0.6457, "eval_samples_per_second": 263.281, "eval_steps_per_second": 4.646, "step": 1536 }, { "epoch": 17.0, "grad_norm": 0.40939855575561523, "learning_rate": 4.15e-05, "loss": 0.0432, "step": 1632 }, { "epoch": 17.0, "eval_LOCATION_f1": 0.9322033898305084, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.9166666666666666, "eval_LOCATION_recall": 0.9482758620689655, "eval_ORGANIZATION_f1": 0.9453376205787781, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.9607843137254902, "eval_ORGANIZATION_recall": 0.930379746835443, "eval_PERSON_f1": 0.9879518072289156, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.984, "eval_PERSON_recall": 0.9919354838709677, "eval_loss": 0.03531834855675697, "eval_overall_accuracy": 0.990118034586879, "eval_overall_f1": 0.9547738693467337, "eval_overall_precision": 0.9547738693467337, "eval_overall_recall": 0.9547738693467337, "eval_runtime": 0.7197, "eval_samples_per_second": 236.21, "eval_steps_per_second": 4.168, "step": 1632 }, { "epoch": 18.0, "grad_norm": 0.4544738531112671, "learning_rate": 4.1e-05, "loss": 0.0427, "step": 1728 }, { "epoch": 18.0, "eval_LOCATION_f1": 0.9377593360995852, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.904, "eval_LOCATION_recall": 0.9741379310344828, "eval_ORGANIZATION_f1": 0.9315960912052117, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.959731543624161, "eval_ORGANIZATION_recall": 0.9050632911392406, "eval_PERSON_f1": 0.9879518072289156, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.984, "eval_PERSON_recall": 0.9919354838709677, "eval_loss": 0.03867640718817711, "eval_overall_accuracy": 0.9884710403513588, "eval_overall_f1": 0.9510664993726474, "eval_overall_precision": 0.949874686716792, "eval_overall_recall": 0.9522613065326633, "eval_runtime": 0.6523, "eval_samples_per_second": 260.635, "eval_steps_per_second": 4.599, "step": 1728 }, { "epoch": 19.0, "grad_norm": 0.4176604449748993, "learning_rate": 4.05e-05, "loss": 0.0421, "step": 1824 }, { "epoch": 19.0, "eval_LOCATION_f1": 0.9478260869565217, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.956140350877193, "eval_LOCATION_recall": 0.9396551724137931, "eval_ORGANIZATION_f1": 0.9490445859872612, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.9551282051282052, "eval_ORGANIZATION_recall": 0.9430379746835443, "eval_PERSON_f1": 0.9879518072289156, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.984, "eval_PERSON_recall": 0.9919354838709677, "eval_loss": 0.036288563162088394, "eval_overall_accuracy": 0.9898435355476256, "eval_overall_f1": 0.9609079445145018, "eval_overall_precision": 0.9645569620253165, "eval_overall_recall": 0.957286432160804, "eval_runtime": 0.6382, "eval_samples_per_second": 266.365, "eval_steps_per_second": 4.701, "step": 1824 }, { "epoch": 20.0, "grad_norm": 0.5175350904464722, "learning_rate": 4e-05, "loss": 0.0363, "step": 1920 }, { "epoch": 20.0, "eval_LOCATION_f1": 0.944206008583691, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.9401709401709402, "eval_LOCATION_recall": 0.9482758620689655, "eval_ORGANIZATION_f1": 0.9433962264150944, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.9375, "eval_ORGANIZATION_recall": 0.9493670886075949, "eval_PERSON_f1": 0.9879518072289156, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.984, "eval_PERSON_recall": 0.9919354838709677, "eval_loss": 0.037881579250097275, "eval_overall_accuracy": 0.9892945374691189, "eval_overall_f1": 0.9575, "eval_overall_precision": 0.9527363184079602, "eval_overall_recall": 0.9623115577889447, "eval_runtime": 0.6226, "eval_samples_per_second": 273.058, "eval_steps_per_second": 4.819, "step": 1920 }, { "epoch": 21.0, "grad_norm": 0.48619720339775085, "learning_rate": 3.9500000000000005e-05, "loss": 0.0349, "step": 2016 }, { "epoch": 21.0, "eval_LOCATION_f1": 0.9322033898305084, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.9166666666666666, "eval_LOCATION_recall": 0.9482758620689655, "eval_ORGANIZATION_f1": 0.9456869009584665, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.9548387096774194, "eval_ORGANIZATION_recall": 0.9367088607594937, "eval_PERSON_f1": 0.9879518072289156, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.984, "eval_PERSON_recall": 0.9919354838709677, "eval_loss": 0.036370955407619476, "eval_overall_accuracy": 0.9898435355476256, "eval_overall_f1": 0.9548872180451128, "eval_overall_precision": 0.9525, "eval_overall_recall": 0.957286432160804, "eval_runtime": 0.6299, "eval_samples_per_second": 269.897, "eval_steps_per_second": 4.763, "step": 2016 }, { "epoch": 22.0, "grad_norm": 0.2978288233280182, "learning_rate": 3.9000000000000006e-05, "loss": 0.0373, "step": 2112 }, { "epoch": 22.0, "eval_LOCATION_f1": 0.9482758620689655, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.9482758620689655, "eval_LOCATION_recall": 0.9482758620689655, "eval_ORGANIZATION_f1": 0.9493670886075949, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.9493670886075949, "eval_ORGANIZATION_recall": 0.9493670886075949, "eval_PERSON_f1": 0.9879518072289156, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.984, "eval_PERSON_recall": 0.9919354838709677, "eval_loss": 0.0324406698346138, "eval_overall_accuracy": 0.9917650288223991, "eval_overall_f1": 0.9611041405269761, "eval_overall_precision": 0.9598997493734336, "eval_overall_recall": 0.9623115577889447, "eval_runtime": 0.6317, "eval_samples_per_second": 269.119, "eval_steps_per_second": 4.749, "step": 2112 }, { "epoch": 23.0, "grad_norm": 0.6087857484817505, "learning_rate": 3.85e-05, "loss": 0.0331, "step": 2208 }, { "epoch": 23.0, "eval_LOCATION_f1": 0.9310344827586207, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.9310344827586207, "eval_LOCATION_recall": 0.9310344827586207, "eval_ORGANIZATION_f1": 0.946031746031746, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.9490445859872612, "eval_ORGANIZATION_recall": 0.9430379746835443, "eval_PERSON_f1": 0.9879518072289156, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.984, "eval_PERSON_recall": 0.9919354838709677, "eval_loss": 0.03451100364327431, "eval_overall_accuracy": 0.990118034586879, "eval_overall_f1": 0.9547738693467337, "eval_overall_precision": 0.9547738693467337, "eval_overall_recall": 0.9547738693467337, "eval_runtime": 0.618, "eval_samples_per_second": 275.094, "eval_steps_per_second": 4.855, "step": 2208 }, { "epoch": 24.0, "grad_norm": 1.0620331764221191, "learning_rate": 3.8e-05, "loss": 0.0317, "step": 2304 }, { "epoch": 24.0, "eval_LOCATION_f1": 0.9406779661016949, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.925, "eval_LOCATION_recall": 0.9568965517241379, "eval_ORGANIZATION_f1": 0.9490445859872612, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.9551282051282052, "eval_ORGANIZATION_recall": 0.9430379746835443, "eval_PERSON_f1": 0.9840000000000001, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.9761904761904762, "eval_PERSON_recall": 0.9919354838709677, "eval_loss": 0.04056801646947861, "eval_overall_accuracy": 0.9890200384298655, "eval_overall_f1": 0.9575, "eval_overall_precision": 0.9527363184079602, "eval_overall_recall": 0.9623115577889447, "eval_runtime": 0.6283, "eval_samples_per_second": 270.582, "eval_steps_per_second": 4.775, "step": 2304 }, { "epoch": 25.0, "grad_norm": 0.5072365403175354, "learning_rate": 3.7500000000000003e-05, "loss": 0.0328, "step": 2400 }, { "epoch": 25.0, "eval_LOCATION_f1": 0.94017094017094, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.9322033898305084, "eval_LOCATION_recall": 0.9482758620689655, "eval_ORGANIZATION_f1": 0.9290322580645162, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.9473684210526315, "eval_ORGANIZATION_recall": 0.9113924050632911, "eval_PERSON_f1": 0.9879518072289156, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.984, "eval_PERSON_recall": 0.9919354838709677, "eval_loss": 0.03679350018501282, "eval_overall_accuracy": 0.9898435355476256, "eval_overall_f1": 0.9508196721311476, "eval_overall_precision": 0.9544303797468354, "eval_overall_recall": 0.9472361809045227, "eval_runtime": 0.6265, "eval_samples_per_second": 271.346, "eval_steps_per_second": 4.788, "step": 2400 }, { "epoch": 26.0, "grad_norm": 0.4231331944465637, "learning_rate": 3.7e-05, "loss": 0.0299, "step": 2496 }, { "epoch": 26.0, "eval_LOCATION_f1": 0.9322033898305084, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.9166666666666666, "eval_LOCATION_recall": 0.9482758620689655, "eval_ORGANIZATION_f1": 0.9423076923076923, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.9545454545454546, "eval_ORGANIZATION_recall": 0.930379746835443, "eval_PERSON_f1": 0.9879518072289156, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.984, "eval_PERSON_recall": 0.9919354838709677, "eval_loss": 0.0363524928689003, "eval_overall_accuracy": 0.9906670326653857, "eval_overall_f1": 0.9535759096612295, "eval_overall_precision": 0.9523809523809523, "eval_overall_recall": 0.9547738693467337, "eval_runtime": 0.6227, "eval_samples_per_second": 272.987, "eval_steps_per_second": 4.817, "step": 2496 }, { "epoch": 27.0, "grad_norm": 1.0756925344467163, "learning_rate": 3.65e-05, "loss": 0.0314, "step": 2592 }, { "epoch": 27.0, "eval_LOCATION_f1": 0.9482758620689655, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.9482758620689655, "eval_LOCATION_recall": 0.9482758620689655, "eval_ORGANIZATION_f1": 0.939297124600639, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.9483870967741935, "eval_ORGANIZATION_recall": 0.930379746835443, "eval_PERSON_f1": 0.9761904761904763, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.9609375, "eval_PERSON_recall": 0.9919354838709677, "eval_loss": 0.03862130269408226, "eval_overall_accuracy": 0.9906670326653857, "eval_overall_f1": 0.9535759096612295, "eval_overall_precision": 0.9523809523809523, "eval_overall_recall": 0.9547738693467337, "eval_runtime": 0.6347, "eval_samples_per_second": 267.857, "eval_steps_per_second": 4.727, "step": 2592 }, { "epoch": 28.0, "grad_norm": 0.5345285534858704, "learning_rate": 3.6e-05, "loss": 0.0297, "step": 2688 }, { "epoch": 28.0, "eval_LOCATION_f1": 0.9276595744680851, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.9159663865546218, "eval_LOCATION_recall": 0.9396551724137931, "eval_ORGANIZATION_f1": 0.9260450160771704, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.9411764705882353, "eval_ORGANIZATION_recall": 0.9113924050632911, "eval_PERSON_f1": 0.9840000000000001, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.9761904761904762, "eval_PERSON_recall": 0.9919354838709677, "eval_loss": 0.039158355444669724, "eval_overall_accuracy": 0.9887455393906122, "eval_overall_f1": 0.9447236180904522, "eval_overall_precision": 0.9447236180904522, "eval_overall_recall": 0.9447236180904522, "eval_runtime": 0.6883, "eval_samples_per_second": 246.999, "eval_steps_per_second": 4.359, "step": 2688 }, { "epoch": 29.0, "grad_norm": 0.18525943160057068, "learning_rate": 3.55e-05, "loss": 0.0263, "step": 2784 }, { "epoch": 29.0, "eval_LOCATION_f1": 0.9173553719008265, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.8809523809523809, "eval_LOCATION_recall": 0.9568965517241379, "eval_ORGANIZATION_f1": 0.9235668789808917, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.9294871794871795, "eval_ORGANIZATION_recall": 0.9177215189873418, "eval_PERSON_f1": 0.9879518072289156, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.984, "eval_PERSON_recall": 0.9919354838709677, "eval_loss": 0.040079668164253235, "eval_overall_accuracy": 0.9890200384298655, "eval_overall_f1": 0.9416149068322981, "eval_overall_precision": 0.9312039312039312, "eval_overall_recall": 0.9522613065326633, "eval_runtime": 0.6355, "eval_samples_per_second": 267.502, "eval_steps_per_second": 4.721, "step": 2784 }, { "epoch": 30.0, "grad_norm": 0.8972023725509644, "learning_rate": 3.5e-05, "loss": 0.026, "step": 2880 }, { "epoch": 30.0, "eval_LOCATION_f1": 0.9356223175965666, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.9316239316239316, "eval_LOCATION_recall": 0.9396551724137931, "eval_ORGANIZATION_f1": 0.9404388714733543, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.9316770186335404, "eval_ORGANIZATION_recall": 0.9493670886075949, "eval_PERSON_f1": 0.9840000000000001, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.9761904761904762, "eval_PERSON_recall": 0.9919354838709677, "eval_loss": 0.04829633980989456, "eval_overall_accuracy": 0.9887455393906122, "eval_overall_f1": 0.9526184538653367, "eval_overall_precision": 0.9455445544554455, "eval_overall_recall": 0.9597989949748744, "eval_runtime": 0.6301, "eval_samples_per_second": 269.783, "eval_steps_per_second": 4.761, "step": 2880 }, { "epoch": 31.0, "grad_norm": 0.6168848276138306, "learning_rate": 3.45e-05, "loss": 0.0271, "step": 2976 }, { "epoch": 31.0, "eval_LOCATION_f1": 0.9159663865546218, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.8934426229508197, "eval_LOCATION_recall": 0.9396551724137931, "eval_ORGANIZATION_f1": 0.9245283018867925, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.91875, "eval_ORGANIZATION_recall": 0.930379746835443, "eval_PERSON_f1": 0.9840000000000001, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.9761904761904762, "eval_PERSON_recall": 0.9919354838709677, "eval_loss": 0.04206063225865364, "eval_overall_accuracy": 0.9873730441943454, "eval_overall_f1": 0.9404466501240695, "eval_overall_precision": 0.928921568627451, "eval_overall_recall": 0.9522613065326633, "eval_runtime": 0.623, "eval_samples_per_second": 272.874, "eval_steps_per_second": 4.815, "step": 2976 }, { "epoch": 32.0, "grad_norm": 0.5183842778205872, "learning_rate": 3.4000000000000007e-05, "loss": 0.0263, "step": 3072 }, { "epoch": 32.0, "eval_LOCATION_f1": 0.923076923076923, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.9152542372881356, "eval_LOCATION_recall": 0.9310344827586207, "eval_ORGANIZATION_f1": 0.9517684887459807, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.9673202614379085, "eval_ORGANIZATION_recall": 0.9367088607594937, "eval_PERSON_f1": 0.9879518072289156, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.984, "eval_PERSON_recall": 0.9919354838709677, "eval_loss": 0.04044005274772644, "eval_overall_accuracy": 0.9895690365083722, "eval_overall_f1": 0.9546599496221662, "eval_overall_precision": 0.9570707070707071, "eval_overall_recall": 0.9522613065326633, "eval_runtime": 0.6405, "eval_samples_per_second": 265.414, "eval_steps_per_second": 4.684, "step": 3072 }, { "epoch": 33.0, "grad_norm": 0.46202486753463745, "learning_rate": 3.35e-05, "loss": 0.0255, "step": 3168 }, { "epoch": 33.0, "eval_LOCATION_f1": 0.923076923076923, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.9152542372881356, "eval_LOCATION_recall": 0.9310344827586207, "eval_ORGANIZATION_f1": 0.9389067524115756, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.954248366013072, "eval_ORGANIZATION_recall": 0.9240506329113924, "eval_PERSON_f1": 0.9879518072289156, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.984, "eval_PERSON_recall": 0.9919354838709677, "eval_loss": 0.042875442653894424, "eval_overall_accuracy": 0.9895690365083722, "eval_overall_f1": 0.9496221662468514, "eval_overall_precision": 0.952020202020202, "eval_overall_recall": 0.9472361809045227, "eval_runtime": 0.6305, "eval_samples_per_second": 269.62, "eval_steps_per_second": 4.758, "step": 3168 }, { "epoch": 34.0, "grad_norm": 0.2009233981370926, "learning_rate": 3.3e-05, "loss": 0.0256, "step": 3264 }, { "epoch": 34.0, "eval_LOCATION_f1": 0.9396551724137931, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.9396551724137931, "eval_LOCATION_recall": 0.9396551724137931, "eval_ORGANIZATION_f1": 0.9367088607594937, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.9367088607594937, "eval_ORGANIZATION_recall": 0.9367088607594937, "eval_PERSON_f1": 0.9840000000000001, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.9761904761904762, "eval_PERSON_recall": 0.9919354838709677, "eval_loss": 0.0402030423283577, "eval_overall_accuracy": 0.9895690365083722, "eval_overall_f1": 0.9523809523809523, "eval_overall_precision": 0.95, "eval_overall_recall": 0.9547738693467337, "eval_runtime": 0.6297, "eval_samples_per_second": 269.963, "eval_steps_per_second": 4.764, "step": 3264 }, { "epoch": 35.0, "grad_norm": 0.7572520971298218, "learning_rate": 3.2500000000000004e-05, "loss": 0.0202, "step": 3360 }, { "epoch": 35.0, "eval_LOCATION_f1": 0.9264069264069263, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.9304347826086956, "eval_LOCATION_recall": 0.9224137931034483, "eval_ORGANIZATION_f1": 0.9433962264150944, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.9375, "eval_ORGANIZATION_recall": 0.9493670886075949, "eval_PERSON_f1": 0.9800796812749005, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.968503937007874, "eval_PERSON_recall": 0.9919354838709677, "eval_loss": 0.0406881608068943, "eval_overall_accuracy": 0.9903925336261323, "eval_overall_f1": 0.95, "eval_overall_precision": 0.945273631840796, "eval_overall_recall": 0.9547738693467337, "eval_runtime": 0.6335, "eval_samples_per_second": 268.343, "eval_steps_per_second": 4.735, "step": 3360 }, { "epoch": 36.0, "grad_norm": 0.3844459354877472, "learning_rate": 3.2000000000000005e-05, "loss": 0.0232, "step": 3456 }, { "epoch": 36.0, "eval_LOCATION_f1": 0.9322033898305084, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.9166666666666666, "eval_LOCATION_recall": 0.9482758620689655, "eval_ORGANIZATION_f1": 0.9453376205787781, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.9607843137254902, "eval_ORGANIZATION_recall": 0.930379746835443, "eval_PERSON_f1": 0.9840000000000001, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.9761904761904762, "eval_PERSON_recall": 0.9919354838709677, "eval_loss": 0.03911978378891945, "eval_overall_accuracy": 0.9898435355476256, "eval_overall_f1": 0.9535759096612295, "eval_overall_precision": 0.9523809523809523, "eval_overall_recall": 0.9547738693467337, "eval_runtime": 0.6376, "eval_samples_per_second": 266.617, "eval_steps_per_second": 4.705, "step": 3456 }, { "epoch": 37.0, "grad_norm": 0.2559301257133484, "learning_rate": 3.15e-05, "loss": 0.0216, "step": 3552 }, { "epoch": 37.0, "eval_LOCATION_f1": 0.923076923076923, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.9152542372881356, "eval_LOCATION_recall": 0.9310344827586207, "eval_ORGANIZATION_f1": 0.946031746031746, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.9490445859872612, "eval_ORGANIZATION_recall": 0.9430379746835443, "eval_PERSON_f1": 0.9840000000000001, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.9761904761904762, "eval_PERSON_recall": 0.9919354838709677, "eval_loss": 0.03514598309993744, "eval_overall_accuracy": 0.9898435355476256, "eval_overall_f1": 0.951188986232791, "eval_overall_precision": 0.9476309226932669, "eval_overall_recall": 0.9547738693467337, "eval_runtime": 0.6284, "eval_samples_per_second": 270.545, "eval_steps_per_second": 4.774, "step": 3552 }, { "epoch": 38.0, "grad_norm": 0.49328336119651794, "learning_rate": 3.1e-05, "loss": 0.0218, "step": 3648 }, { "epoch": 38.0, "eval_LOCATION_f1": 0.9264069264069263, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.9304347826086956, "eval_LOCATION_recall": 0.9224137931034483, "eval_ORGANIZATION_f1": 0.9493670886075949, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.9493670886075949, "eval_ORGANIZATION_recall": 0.9493670886075949, "eval_PERSON_f1": 0.9840000000000001, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.9761904761904762, "eval_PERSON_recall": 0.9919354838709677, "eval_loss": 0.03533317148685455, "eval_overall_accuracy": 0.9909415317046391, "eval_overall_f1": 0.9535759096612295, "eval_overall_precision": 0.9523809523809523, "eval_overall_recall": 0.9547738693467337, "eval_runtime": 0.6336, "eval_samples_per_second": 268.297, "eval_steps_per_second": 4.735, "step": 3648 }, { "epoch": 39.0, "grad_norm": 0.43509092926979065, "learning_rate": 3.05e-05, "loss": 0.0193, "step": 3744 }, { "epoch": 39.0, "eval_LOCATION_f1": 0.9316239316239315, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.923728813559322, "eval_LOCATION_recall": 0.9396551724137931, "eval_ORGANIZATION_f1": 0.9358974358974359, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.948051948051948, "eval_ORGANIZATION_recall": 0.9240506329113924, "eval_PERSON_f1": 0.9840000000000001, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.9761904761904762, "eval_PERSON_recall": 0.9919354838709677, "eval_loss": 0.04069233313202858, "eval_overall_accuracy": 0.9903925336261323, "eval_overall_f1": 0.949748743718593, "eval_overall_precision": 0.949748743718593, "eval_overall_recall": 0.949748743718593, "eval_runtime": 0.6329, "eval_samples_per_second": 268.625, "eval_steps_per_second": 4.74, "step": 3744 }, { "epoch": 40.0, "grad_norm": 0.10171817988157272, "learning_rate": 3e-05, "loss": 0.0214, "step": 3840 }, { "epoch": 40.0, "eval_LOCATION_f1": 0.9482758620689655, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.9482758620689655, "eval_LOCATION_recall": 0.9482758620689655, "eval_ORGANIZATION_f1": 0.9648562300319489, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.9741935483870968, "eval_ORGANIZATION_recall": 0.9556962025316456, "eval_PERSON_f1": 0.9800796812749005, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.968503937007874, "eval_PERSON_recall": 0.9919354838709677, "eval_loss": 0.03859032317996025, "eval_overall_accuracy": 0.9917650288223991, "eval_overall_f1": 0.964824120603015, "eval_overall_precision": 0.964824120603015, "eval_overall_recall": 0.964824120603015, "eval_runtime": 0.6305, "eval_samples_per_second": 269.629, "eval_steps_per_second": 4.758, "step": 3840 }, { "epoch": 41.0, "grad_norm": 0.06850344687700272, "learning_rate": 2.95e-05, "loss": 0.0201, "step": 3936 }, { "epoch": 41.0, "eval_LOCATION_f1": 0.9316239316239315, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.923728813559322, "eval_LOCATION_recall": 0.9396551724137931, "eval_ORGANIZATION_f1": 0.946031746031746, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.9490445859872612, "eval_ORGANIZATION_recall": 0.9430379746835443, "eval_PERSON_f1": 0.9800796812749005, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.968503937007874, "eval_PERSON_recall": 0.9919354838709677, "eval_loss": 0.03688988834619522, "eval_overall_accuracy": 0.990118034586879, "eval_overall_f1": 0.9525, "eval_overall_precision": 0.9477611940298507, "eval_overall_recall": 0.957286432160804, "eval_runtime": 0.6358, "eval_samples_per_second": 267.37, "eval_steps_per_second": 4.718, "step": 3936 }, { "epoch": 42.0, "grad_norm": 0.5080527067184448, "learning_rate": 2.9e-05, "loss": 0.0193, "step": 4032 }, { "epoch": 42.0, "eval_LOCATION_f1": 0.94017094017094, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.9322033898305084, "eval_LOCATION_recall": 0.9482758620689655, "eval_ORGANIZATION_f1": 0.9456869009584665, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.9548387096774194, "eval_ORGANIZATION_recall": 0.9367088607594937, "eval_PERSON_f1": 0.9879518072289156, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.984, "eval_PERSON_recall": 0.9919354838709677, "eval_loss": 0.03985198587179184, "eval_overall_accuracy": 0.990118034586879, "eval_overall_f1": 0.957286432160804, "eval_overall_precision": 0.957286432160804, "eval_overall_recall": 0.957286432160804, "eval_runtime": 0.6343, "eval_samples_per_second": 268.017, "eval_steps_per_second": 4.73, "step": 4032 }, { "epoch": 43.0, "grad_norm": 0.12518884241580963, "learning_rate": 2.8499999999999998e-05, "loss": 0.0198, "step": 4128 }, { "epoch": 43.0, "eval_LOCATION_f1": 0.9361702127659575, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.9243697478991597, "eval_LOCATION_recall": 0.9482758620689655, "eval_ORGANIZATION_f1": 0.9551282051282052, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.9675324675324676, "eval_ORGANIZATION_recall": 0.9430379746835443, "eval_PERSON_f1": 0.9879518072289156, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.984, "eval_PERSON_recall": 0.9919354838709677, "eval_loss": 0.03934216871857643, "eval_overall_accuracy": 0.9914905297831458, "eval_overall_f1": 0.9597989949748744, "eval_overall_precision": 0.9597989949748744, "eval_overall_recall": 0.9597989949748744, "eval_runtime": 0.6288, "eval_samples_per_second": 270.355, "eval_steps_per_second": 4.771, "step": 4128 }, { "epoch": 44.0, "grad_norm": 0.36400577425956726, "learning_rate": 2.8000000000000003e-05, "loss": 0.0195, "step": 4224 }, { "epoch": 44.0, "eval_LOCATION_f1": 0.9191489361702128, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.907563025210084, "eval_LOCATION_recall": 0.9310344827586207, "eval_ORGANIZATION_f1": 0.932475884244373, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.9477124183006536, "eval_ORGANIZATION_recall": 0.9177215189873418, "eval_PERSON_f1": 0.9840000000000001, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.9761904761904762, "eval_PERSON_recall": 0.9919354838709677, "eval_loss": 0.041267186403274536, "eval_overall_accuracy": 0.9895690365083722, "eval_overall_f1": 0.9447236180904522, "eval_overall_precision": 0.9447236180904522, "eval_overall_recall": 0.9447236180904522, "eval_runtime": 0.6258, "eval_samples_per_second": 271.639, "eval_steps_per_second": 4.794, "step": 4224 }, { "epoch": 45.0, "grad_norm": 0.7251652479171753, "learning_rate": 2.7500000000000004e-05, "loss": 0.0173, "step": 4320 }, { "epoch": 45.0, "eval_LOCATION_f1": 0.9361702127659575, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.9243697478991597, "eval_LOCATION_recall": 0.9482758620689655, "eval_ORGANIZATION_f1": 0.932475884244373, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.9477124183006536, "eval_ORGANIZATION_recall": 0.9177215189873418, "eval_PERSON_f1": 0.9840000000000001, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.9761904761904762, "eval_PERSON_recall": 0.9919354838709677, "eval_loss": 0.03837477043271065, "eval_overall_accuracy": 0.9898435355476256, "eval_overall_f1": 0.949748743718593, "eval_overall_precision": 0.949748743718593, "eval_overall_recall": 0.949748743718593, "eval_runtime": 0.6183, "eval_samples_per_second": 274.94, "eval_steps_per_second": 4.852, "step": 4320 }, { "epoch": 46.0, "grad_norm": 0.0785386711359024, "learning_rate": 2.7000000000000002e-05, "loss": 0.0199, "step": 4416 }, { "epoch": 46.0, "eval_LOCATION_f1": 0.9243697478991596, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.9016393442622951, "eval_LOCATION_recall": 0.9482758620689655, "eval_ORGANIZATION_f1": 0.9483870967741935, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.9671052631578947, "eval_ORGANIZATION_recall": 0.930379746835443, "eval_PERSON_f1": 0.9879518072289156, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.984, "eval_PERSON_recall": 0.9919354838709677, "eval_loss": 0.040848102420568466, "eval_overall_accuracy": 0.9903925336261323, "eval_overall_f1": 0.9535759096612295, "eval_overall_precision": 0.9523809523809523, "eval_overall_recall": 0.9547738693467337, "eval_runtime": 0.6564, "eval_samples_per_second": 258.978, "eval_steps_per_second": 4.57, "step": 4416 }, { "epoch": 47.0, "grad_norm": 0.5049558877944946, "learning_rate": 2.6500000000000004e-05, "loss": 0.0183, "step": 4512 }, { "epoch": 47.0, "eval_LOCATION_f1": 0.944206008583691, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.9401709401709402, "eval_LOCATION_recall": 0.9482758620689655, "eval_ORGANIZATION_f1": 0.9523809523809524, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.9554140127388535, "eval_ORGANIZATION_recall": 0.9493670886075949, "eval_PERSON_f1": 0.9800796812749005, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.968503937007874, "eval_PERSON_recall": 0.9919354838709677, "eval_loss": 0.036147408187389374, "eval_overall_accuracy": 0.9912160307438924, "eval_overall_f1": 0.9586983729662076, "eval_overall_precision": 0.9551122194513716, "eval_overall_recall": 0.9623115577889447, "eval_runtime": 0.6239, "eval_samples_per_second": 272.49, "eval_steps_per_second": 4.809, "step": 4512 }, { "epoch": 48.0, "grad_norm": 0.6900705099105835, "learning_rate": 2.6000000000000002e-05, "loss": 0.0194, "step": 4608 }, { "epoch": 48.0, "eval_LOCATION_f1": 0.9361702127659575, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.9243697478991597, "eval_LOCATION_recall": 0.9482758620689655, "eval_ORGANIZATION_f1": 0.9456869009584665, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.9548387096774194, "eval_ORGANIZATION_recall": 0.9367088607594937, "eval_PERSON_f1": 0.9879518072289156, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.984, "eval_PERSON_recall": 0.9919354838709677, "eval_loss": 0.04444650933146477, "eval_overall_accuracy": 0.9892945374691189, "eval_overall_f1": 0.9560853199498118, "eval_overall_precision": 0.9548872180451128, "eval_overall_recall": 0.957286432160804, "eval_runtime": 0.6369, "eval_samples_per_second": 266.925, "eval_steps_per_second": 4.71, "step": 4608 }, { "epoch": 49.0, "grad_norm": 0.20286892354488373, "learning_rate": 2.5500000000000003e-05, "loss": 0.0158, "step": 4704 }, { "epoch": 49.0, "eval_LOCATION_f1": 0.9316239316239315, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.923728813559322, "eval_LOCATION_recall": 0.9396551724137931, "eval_ORGANIZATION_f1": 0.9389067524115756, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.954248366013072, "eval_ORGANIZATION_recall": 0.9240506329113924, "eval_PERSON_f1": 0.9800796812749005, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.968503937007874, "eval_PERSON_recall": 0.9919354838709677, "eval_loss": 0.04167434573173523, "eval_overall_accuracy": 0.9898435355476256, "eval_overall_f1": 0.949748743718593, "eval_overall_precision": 0.949748743718593, "eval_overall_recall": 0.949748743718593, "eval_runtime": 0.6386, "eval_samples_per_second": 266.209, "eval_steps_per_second": 4.698, "step": 4704 }, { "epoch": 50.0, "grad_norm": 0.03262986242771149, "learning_rate": 2.5e-05, "loss": 0.0159, "step": 4800 }, { "epoch": 50.0, "eval_LOCATION_f1": 0.94017094017094, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.9322033898305084, "eval_LOCATION_recall": 0.9482758620689655, "eval_ORGANIZATION_f1": 0.939297124600639, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.9483870967741935, "eval_ORGANIZATION_recall": 0.930379746835443, "eval_PERSON_f1": 0.9799196787148594, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.976, "eval_PERSON_recall": 0.9838709677419355, "eval_loss": 0.04414771497249603, "eval_overall_accuracy": 0.9890200384298655, "eval_overall_f1": 0.9522613065326633, "eval_overall_precision": 0.9522613065326633, "eval_overall_recall": 0.9522613065326633, "eval_runtime": 0.6276, "eval_samples_per_second": 270.875, "eval_steps_per_second": 4.78, "step": 4800 }, { "epoch": 51.0, "grad_norm": 0.46143171191215515, "learning_rate": 2.45e-05, "loss": 0.0147, "step": 4896 }, { "epoch": 51.0, "eval_LOCATION_f1": 0.927038626609442, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.9230769230769231, "eval_LOCATION_recall": 0.9310344827586207, "eval_ORGANIZATION_f1": 0.930379746835443, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.930379746835443, "eval_ORGANIZATION_recall": 0.930379746835443, "eval_PERSON_f1": 0.9800796812749005, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.968503937007874, "eval_PERSON_recall": 0.9919354838709677, "eval_loss": 0.04490748047828674, "eval_overall_accuracy": 0.9884710403513588, "eval_overall_f1": 0.9450000000000001, "eval_overall_precision": 0.9402985074626866, "eval_overall_recall": 0.949748743718593, "eval_runtime": 0.6391, "eval_samples_per_second": 266.005, "eval_steps_per_second": 4.694, "step": 4896 }, { "epoch": 52.0, "grad_norm": 0.7441356778144836, "learning_rate": 2.4e-05, "loss": 0.0161, "step": 4992 }, { "epoch": 52.0, "eval_LOCATION_f1": 0.9098712446351932, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.905982905982906, "eval_LOCATION_recall": 0.9137931034482759, "eval_ORGANIZATION_f1": 0.9449838187702266, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.9668874172185431, "eval_ORGANIZATION_recall": 0.9240506329113924, "eval_PERSON_f1": 0.9840000000000001, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.9761904761904762, "eval_PERSON_recall": 0.9919354838709677, "eval_loss": 0.05439521744847298, "eval_overall_accuracy": 0.9873730441943454, "eval_overall_f1": 0.946969696969697, "eval_overall_precision": 0.9517766497461929, "eval_overall_recall": 0.9422110552763819, "eval_runtime": 0.6358, "eval_samples_per_second": 267.376, "eval_steps_per_second": 4.718, "step": 4992 }, { "epoch": 53.0, "grad_norm": 0.630390465259552, "learning_rate": 2.35e-05, "loss": 0.0157, "step": 5088 }, { "epoch": 53.0, "eval_LOCATION_f1": 0.9191489361702128, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.907563025210084, "eval_LOCATION_recall": 0.9310344827586207, "eval_ORGANIZATION_f1": 0.9329073482428115, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.9419354838709677, "eval_ORGANIZATION_recall": 0.9240506329113924, "eval_PERSON_f1": 0.9800796812749005, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.968503937007874, "eval_PERSON_recall": 0.9919354838709677, "eval_loss": 0.04217611625790596, "eval_overall_accuracy": 0.9892945374691189, "eval_overall_f1": 0.9436795994993743, "eval_overall_precision": 0.940149625935162, "eval_overall_recall": 0.9472361809045227, "eval_runtime": 0.6309, "eval_samples_per_second": 269.448, "eval_steps_per_second": 4.755, "step": 5088 }, { "epoch": 54.0, "grad_norm": 0.1506197303533554, "learning_rate": 2.3000000000000003e-05, "loss": 0.0163, "step": 5184 }, { "epoch": 54.0, "eval_LOCATION_f1": 0.9145299145299146, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.9067796610169492, "eval_LOCATION_recall": 0.9224137931034483, "eval_ORGANIZATION_f1": 0.9367088607594937, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.9367088607594937, "eval_ORGANIZATION_recall": 0.9367088607594937, "eval_PERSON_f1": 0.9840000000000001, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.9761904761904762, "eval_PERSON_recall": 0.9919354838709677, "eval_loss": 0.04696401581168175, "eval_overall_accuracy": 0.9887455393906122, "eval_overall_f1": 0.9450000000000001, "eval_overall_precision": 0.9402985074626866, "eval_overall_recall": 0.949748743718593, "eval_runtime": 0.6366, "eval_samples_per_second": 267.032, "eval_steps_per_second": 4.712, "step": 5184 }, { "epoch": 55.0, "grad_norm": 0.9606214165687561, "learning_rate": 2.25e-05, "loss": 0.0158, "step": 5280 }, { "epoch": 55.0, "eval_LOCATION_f1": 0.927038626609442, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.9230769230769231, "eval_LOCATION_recall": 0.9310344827586207, "eval_ORGANIZATION_f1": 0.9182389937106918, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.9125, "eval_ORGANIZATION_recall": 0.9240506329113924, "eval_PERSON_f1": 0.9800796812749005, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.968503937007874, "eval_PERSON_recall": 0.9919354838709677, "eval_loss": 0.04361223429441452, "eval_overall_accuracy": 0.9890200384298655, "eval_overall_f1": 0.9401496259351622, "eval_overall_precision": 0.9331683168316832, "eval_overall_recall": 0.9472361809045227, "eval_runtime": 0.6249, "eval_samples_per_second": 272.022, "eval_steps_per_second": 4.8, "step": 5280 }, { "epoch": 56.0, "grad_norm": 0.6902320981025696, "learning_rate": 2.2000000000000003e-05, "loss": 0.016, "step": 5376 }, { "epoch": 56.0, "eval_LOCATION_f1": 0.9316239316239315, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.923728813559322, "eval_LOCATION_recall": 0.9396551724137931, "eval_ORGANIZATION_f1": 0.9206349206349206, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.9235668789808917, "eval_ORGANIZATION_recall": 0.9177215189873418, "eval_PERSON_f1": 0.9840000000000001, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.9761904761904762, "eval_PERSON_recall": 0.9919354838709677, "eval_loss": 0.04493039846420288, "eval_overall_accuracy": 0.9887455393906122, "eval_overall_f1": 0.9436795994993743, "eval_overall_precision": 0.940149625935162, "eval_overall_recall": 0.9472361809045227, "eval_runtime": 0.6303, "eval_samples_per_second": 269.693, "eval_steps_per_second": 4.759, "step": 5376 }, { "epoch": 57.0, "grad_norm": 0.07920908182859421, "learning_rate": 2.15e-05, "loss": 0.0148, "step": 5472 }, { "epoch": 57.0, "eval_LOCATION_f1": 0.9356223175965666, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.9316239316239316, "eval_LOCATION_recall": 0.9396551724137931, "eval_ORGANIZATION_f1": 0.9113924050632911, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.9113924050632911, "eval_ORGANIZATION_recall": 0.9113924050632911, "eval_PERSON_f1": 0.9800796812749005, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.968503937007874, "eval_PERSON_recall": 0.9919354838709677, "eval_loss": 0.0447387620806694, "eval_overall_accuracy": 0.9887455393906122, "eval_overall_f1": 0.94, "eval_overall_precision": 0.9353233830845771, "eval_overall_recall": 0.9447236180904522, "eval_runtime": 0.6366, "eval_samples_per_second": 267.057, "eval_steps_per_second": 4.713, "step": 5472 }, { "epoch": 58.0, "grad_norm": 0.04797344282269478, "learning_rate": 2.1e-05, "loss": 0.015, "step": 5568 }, { "epoch": 58.0, "eval_LOCATION_f1": 0.9316239316239315, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.923728813559322, "eval_LOCATION_recall": 0.9396551724137931, "eval_ORGANIZATION_f1": 0.9320388349514563, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.9536423841059603, "eval_ORGANIZATION_recall": 0.9113924050632911, "eval_PERSON_f1": 0.9840000000000001, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.9761904761904762, "eval_PERSON_recall": 0.9919354838709677, "eval_loss": 0.044603314250707626, "eval_overall_accuracy": 0.9898435355476256, "eval_overall_f1": 0.948297604035309, "eval_overall_precision": 0.9518987341772152, "eval_overall_recall": 0.9447236180904522, "eval_runtime": 0.624, "eval_samples_per_second": 272.414, "eval_steps_per_second": 4.807, "step": 5568 }, { "epoch": 59.0, "grad_norm": 0.2126830667257309, "learning_rate": 2.05e-05, "loss": 0.0145, "step": 5664 }, { "epoch": 59.0, "eval_LOCATION_f1": 0.944206008583691, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.9401709401709402, "eval_LOCATION_recall": 0.9482758620689655, "eval_ORGANIZATION_f1": 0.9206349206349206, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.9235668789808917, "eval_ORGANIZATION_recall": 0.9177215189873418, "eval_PERSON_f1": 0.9840000000000001, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.9761904761904762, "eval_PERSON_recall": 0.9919354838709677, "eval_loss": 0.04263820871710777, "eval_overall_accuracy": 0.9898435355476256, "eval_overall_f1": 0.9473684210526316, "eval_overall_precision": 0.945, "eval_overall_recall": 0.949748743718593, "eval_runtime": 0.6415, "eval_samples_per_second": 265.006, "eval_steps_per_second": 4.677, "step": 5664 }, { "epoch": 60.0, "grad_norm": 0.29798468947410583, "learning_rate": 2e-05, "loss": 0.0142, "step": 5760 }, { "epoch": 60.0, "eval_LOCATION_f1": 0.9613733905579399, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.9572649572649573, "eval_LOCATION_recall": 0.9655172413793104, "eval_ORGANIZATION_f1": 0.9363057324840764, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.9423076923076923, "eval_ORGANIZATION_recall": 0.930379746835443, "eval_PERSON_f1": 0.9879518072289156, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.984, "eval_PERSON_recall": 0.9919354838709677, "eval_loss": 0.04414110630750656, "eval_overall_accuracy": 0.9912160307438924, "eval_overall_f1": 0.9597989949748744, "eval_overall_precision": 0.9597989949748744, "eval_overall_recall": 0.9597989949748744, "eval_runtime": 0.6433, "eval_samples_per_second": 264.274, "eval_steps_per_second": 4.664, "step": 5760 }, { "epoch": 61.0, "grad_norm": 1.3759386539459229, "learning_rate": 1.9500000000000003e-05, "loss": 0.0127, "step": 5856 }, { "epoch": 61.0, "eval_LOCATION_f1": 0.9527896995708154, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.9487179487179487, "eval_LOCATION_recall": 0.9568965517241379, "eval_ORGANIZATION_f1": 0.9367088607594937, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.9367088607594937, "eval_ORGANIZATION_recall": 0.9367088607594937, "eval_PERSON_f1": 0.9800796812749005, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.968503937007874, "eval_PERSON_recall": 0.9919354838709677, "eval_loss": 0.04610952362418175, "eval_overall_accuracy": 0.9909415317046391, "eval_overall_f1": 0.9550000000000001, "eval_overall_precision": 0.9502487562189055, "eval_overall_recall": 0.9597989949748744, "eval_runtime": 0.6253, "eval_samples_per_second": 271.882, "eval_steps_per_second": 4.798, "step": 5856 }, { "epoch": 62.0, "grad_norm": 0.600804328918457, "learning_rate": 1.9e-05, "loss": 0.0131, "step": 5952 }, { "epoch": 62.0, "eval_LOCATION_f1": 0.9356223175965666, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.9316239316239316, "eval_LOCATION_recall": 0.9396551724137931, "eval_ORGANIZATION_f1": 0.9329073482428115, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.9419354838709677, "eval_ORGANIZATION_recall": 0.9240506329113924, "eval_PERSON_f1": 0.9800796812749005, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.968503937007874, "eval_PERSON_recall": 0.9919354838709677, "eval_loss": 0.04702340438961983, "eval_overall_accuracy": 0.9895690365083722, "eval_overall_f1": 0.9485570890840652, "eval_overall_precision": 0.9473684210526315, "eval_overall_recall": 0.949748743718593, "eval_runtime": 0.6314, "eval_samples_per_second": 269.228, "eval_steps_per_second": 4.751, "step": 5952 }, { "epoch": 63.0, "grad_norm": 0.023430807515978813, "learning_rate": 1.85e-05, "loss": 0.0137, "step": 6048 }, { "epoch": 63.0, "eval_LOCATION_f1": 0.94017094017094, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.9322033898305084, "eval_LOCATION_recall": 0.9482758620689655, "eval_ORGANIZATION_f1": 0.9453376205787781, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.9607843137254902, "eval_ORGANIZATION_recall": 0.930379746835443, "eval_PERSON_f1": 0.9840000000000001, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.9761904761904762, "eval_PERSON_recall": 0.9919354838709677, "eval_loss": 0.0426749587059021, "eval_overall_accuracy": 0.9912160307438924, "eval_overall_f1": 0.9559748427672956, "eval_overall_precision": 0.9571788413098237, "eval_overall_recall": 0.9547738693467337, "eval_runtime": 0.6271, "eval_samples_per_second": 271.077, "eval_steps_per_second": 4.784, "step": 6048 }, { "epoch": 64.0, "grad_norm": 0.7301260828971863, "learning_rate": 1.8e-05, "loss": 0.0117, "step": 6144 }, { "epoch": 64.0, "eval_LOCATION_f1": 0.9478260869565217, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.956140350877193, "eval_LOCATION_recall": 0.9396551724137931, "eval_ORGANIZATION_f1": 0.9274447949526813, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.9245283018867925, "eval_ORGANIZATION_recall": 0.930379746835443, "eval_PERSON_f1": 0.976, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.9682539682539683, "eval_PERSON_recall": 0.9838709677419355, "eval_loss": 0.04693343862891197, "eval_overall_accuracy": 0.9903925336261323, "eval_overall_f1": 0.9485570890840652, "eval_overall_precision": 0.9473684210526315, "eval_overall_recall": 0.949748743718593, "eval_runtime": 0.6306, "eval_samples_per_second": 269.575, "eval_steps_per_second": 4.757, "step": 6144 }, { "epoch": 65.0, "grad_norm": 0.34183019399642944, "learning_rate": 1.75e-05, "loss": 0.0131, "step": 6240 }, { "epoch": 65.0, "eval_LOCATION_f1": 0.9451476793248945, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.9256198347107438, "eval_LOCATION_recall": 0.9655172413793104, "eval_ORGANIZATION_f1": 0.9137380191693291, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.9225806451612903, "eval_ORGANIZATION_recall": 0.9050632911392406, "eval_PERSON_f1": 0.9800796812749005, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.968503937007874, "eval_PERSON_recall": 0.9919354838709677, "eval_loss": 0.043672408908605576, "eval_overall_accuracy": 0.9898435355476256, "eval_overall_f1": 0.9438202247191011, "eval_overall_precision": 0.9379652605459057, "eval_overall_recall": 0.949748743718593, "eval_runtime": 0.6449, "eval_samples_per_second": 263.61, "eval_steps_per_second": 4.652, "step": 6240 }, { "epoch": 66.0, "grad_norm": 0.3577764928340912, "learning_rate": 1.7000000000000003e-05, "loss": 0.0132, "step": 6336 }, { "epoch": 66.0, "eval_LOCATION_f1": 0.9523809523809523, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.9565217391304348, "eval_LOCATION_recall": 0.9482758620689655, "eval_ORGANIZATION_f1": 0.9260450160771704, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.9411764705882353, "eval_ORGANIZATION_recall": 0.9113924050632911, "eval_PERSON_f1": 0.9761904761904763, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.9609375, "eval_PERSON_recall": 0.9919354838709677, "eval_loss": 0.043462902307510376, "eval_overall_accuracy": 0.990118034586879, "eval_overall_f1": 0.9496221662468514, "eval_overall_precision": 0.952020202020202, "eval_overall_recall": 0.9472361809045227, "eval_runtime": 0.6284, "eval_samples_per_second": 270.545, "eval_steps_per_second": 4.774, "step": 6336 }, { "epoch": 67.0, "grad_norm": 0.035872530192136765, "learning_rate": 1.65e-05, "loss": 0.0117, "step": 6432 }, { "epoch": 67.0, "eval_LOCATION_f1": 0.9446808510638298, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.9327731092436975, "eval_LOCATION_recall": 0.9568965517241379, "eval_ORGANIZATION_f1": 0.9142857142857144, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.9171974522292994, "eval_ORGANIZATION_recall": 0.9113924050632911, "eval_PERSON_f1": 0.976, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.9682539682539683, "eval_PERSON_recall": 0.9838709677419355, "eval_loss": 0.041273586452007294, "eval_overall_accuracy": 0.9898435355476256, "eval_overall_f1": 0.9425, "eval_overall_precision": 0.9378109452736318, "eval_overall_recall": 0.9472361809045227, "eval_runtime": 0.6969, "eval_samples_per_second": 243.94, "eval_steps_per_second": 4.305, "step": 6432 }, { "epoch": 68.0, "grad_norm": 0.42826521396636963, "learning_rate": 1.6000000000000003e-05, "loss": 0.0124, "step": 6528 }, { "epoch": 68.0, "eval_LOCATION_f1": 0.9446808510638298, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.9327731092436975, "eval_LOCATION_recall": 0.9568965517241379, "eval_ORGANIZATION_f1": 0.9423076923076923, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.9545454545454546, "eval_ORGANIZATION_recall": 0.930379746835443, "eval_PERSON_f1": 0.9800796812749005, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.968503937007874, "eval_PERSON_recall": 0.9919354838709677, "eval_loss": 0.04395867511630058, "eval_overall_accuracy": 0.9906670326653857, "eval_overall_f1": 0.9548872180451128, "eval_overall_precision": 0.9525, "eval_overall_recall": 0.957286432160804, "eval_runtime": 0.6382, "eval_samples_per_second": 266.365, "eval_steps_per_second": 4.701, "step": 6528 }, { "epoch": 69.0, "grad_norm": 0.7718914747238159, "learning_rate": 1.55e-05, "loss": 0.0122, "step": 6624 }, { "epoch": 69.0, "eval_LOCATION_f1": 0.943231441048035, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.9557522123893806, "eval_LOCATION_recall": 0.9310344827586207, "eval_ORGANIZATION_f1": 0.9367088607594937, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.9367088607594937, "eval_ORGANIZATION_recall": 0.9367088607594937, "eval_PERSON_f1": 0.9800796812749005, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.968503937007874, "eval_PERSON_recall": 0.9919354838709677, "eval_loss": 0.042728327214717865, "eval_overall_accuracy": 0.9906670326653857, "eval_overall_f1": 0.9522613065326633, "eval_overall_precision": 0.9522613065326633, "eval_overall_recall": 0.9522613065326633, "eval_runtime": 0.6377, "eval_samples_per_second": 266.589, "eval_steps_per_second": 4.705, "step": 6624 }, { "epoch": 70.0, "grad_norm": 0.8137780427932739, "learning_rate": 1.5e-05, "loss": 0.0105, "step": 6720 }, { "epoch": 70.0, "eval_LOCATION_f1": 0.923076923076923, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.9152542372881356, "eval_LOCATION_recall": 0.9310344827586207, "eval_ORGANIZATION_f1": 0.9294871794871796, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.9415584415584416, "eval_ORGANIZATION_recall": 0.9177215189873418, "eval_PERSON_f1": 0.976, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.9682539682539683, "eval_PERSON_recall": 0.9838709677419355, "eval_loss": 0.04645096883177757, "eval_overall_accuracy": 0.9887455393906122, "eval_overall_f1": 0.9422110552763819, "eval_overall_precision": 0.9422110552763819, "eval_overall_recall": 0.9422110552763819, "eval_runtime": 0.6326, "eval_samples_per_second": 268.714, "eval_steps_per_second": 4.742, "step": 6720 }, { "epoch": 71.0, "grad_norm": 0.49729543924331665, "learning_rate": 1.45e-05, "loss": 0.0115, "step": 6816 }, { "epoch": 71.0, "eval_LOCATION_f1": 0.9568965517241379, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.9568965517241379, "eval_LOCATION_recall": 0.9568965517241379, "eval_ORGANIZATION_f1": 0.9354838709677419, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.9539473684210527, "eval_ORGANIZATION_recall": 0.9177215189873418, "eval_PERSON_f1": 0.976, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.9682539682539683, "eval_PERSON_recall": 0.9838709677419355, "eval_loss": 0.04674909636378288, "eval_overall_accuracy": 0.9903925336261323, "eval_overall_f1": 0.9545454545454546, "eval_overall_precision": 0.9593908629441624, "eval_overall_recall": 0.949748743718593, "eval_runtime": 0.637, "eval_samples_per_second": 266.869, "eval_steps_per_second": 4.709, "step": 6816 }, { "epoch": 72.0, "grad_norm": 0.05852510780096054, "learning_rate": 1.4000000000000001e-05, "loss": 0.01, "step": 6912 }, { "epoch": 72.0, "eval_LOCATION_f1": 0.9356223175965666, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.9316239316239316, "eval_LOCATION_recall": 0.9396551724137931, "eval_ORGANIZATION_f1": 0.9389067524115756, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.954248366013072, "eval_ORGANIZATION_recall": 0.9240506329113924, "eval_PERSON_f1": 0.976, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.9682539682539683, "eval_PERSON_recall": 0.9838709677419355, "eval_loss": 0.04796583577990532, "eval_overall_accuracy": 0.9906670326653857, "eval_overall_f1": 0.9496221662468514, "eval_overall_precision": 0.952020202020202, "eval_overall_recall": 0.9472361809045227, "eval_runtime": 0.6313, "eval_samples_per_second": 269.288, "eval_steps_per_second": 4.752, "step": 6912 }, { "epoch": 73.0, "grad_norm": 0.11066675186157227, "learning_rate": 1.3500000000000001e-05, "loss": 0.0112, "step": 7008 }, { "epoch": 73.0, "eval_LOCATION_f1": 0.94017094017094, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.9322033898305084, "eval_LOCATION_recall": 0.9482758620689655, "eval_ORGANIZATION_f1": 0.9354838709677419, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.9539473684210527, "eval_ORGANIZATION_recall": 0.9177215189873418, "eval_PERSON_f1": 0.976, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.9682539682539683, "eval_PERSON_recall": 0.9838709677419355, "eval_loss": 0.046430543065071106, "eval_overall_accuracy": 0.9903925336261323, "eval_overall_f1": 0.9496221662468514, "eval_overall_precision": 0.952020202020202, "eval_overall_recall": 0.9472361809045227, "eval_runtime": 0.6274, "eval_samples_per_second": 270.942, "eval_steps_per_second": 4.781, "step": 7008 }, { "epoch": 74.0, "grad_norm": 0.022412816062569618, "learning_rate": 1.3000000000000001e-05, "loss": 0.0104, "step": 7104 }, { "epoch": 74.0, "eval_LOCATION_f1": 0.9396551724137931, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.9396551724137931, "eval_LOCATION_recall": 0.9396551724137931, "eval_ORGANIZATION_f1": 0.9517684887459807, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.9673202614379085, "eval_ORGANIZATION_recall": 0.9367088607594937, "eval_PERSON_f1": 0.976, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.9682539682539683, "eval_PERSON_recall": 0.9838709677419355, "eval_loss": 0.04746369644999504, "eval_overall_accuracy": 0.9903925336261323, "eval_overall_f1": 0.9558638083228247, "eval_overall_precision": 0.959493670886076, "eval_overall_recall": 0.9522613065326633, "eval_runtime": 0.6393, "eval_samples_per_second": 265.909, "eval_steps_per_second": 4.693, "step": 7104 }, { "epoch": 75.0, "grad_norm": 0.42441970109939575, "learning_rate": 1.25e-05, "loss": 0.0103, "step": 7200 }, { "epoch": 75.0, "eval_LOCATION_f1": 0.9322033898305084, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.9166666666666666, "eval_LOCATION_recall": 0.9482758620689655, "eval_ORGANIZATION_f1": 0.9320388349514563, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.9536423841059603, "eval_ORGANIZATION_recall": 0.9113924050632911, "eval_PERSON_f1": 0.976, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.9682539682539683, "eval_PERSON_recall": 0.9838709677419355, "eval_loss": 0.051750294864177704, "eval_overall_accuracy": 0.9890200384298655, "eval_overall_f1": 0.9459119496855346, "eval_overall_precision": 0.947103274559194, "eval_overall_recall": 0.9447236180904522, "eval_runtime": 0.6372, "eval_samples_per_second": 266.81, "eval_steps_per_second": 4.708, "step": 7200 }, { "epoch": 76.0, "grad_norm": 1.392417073249817, "learning_rate": 1.2e-05, "loss": 0.0106, "step": 7296 }, { "epoch": 76.0, "eval_LOCATION_f1": 0.94017094017094, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.9322033898305084, "eval_LOCATION_recall": 0.9482758620689655, "eval_ORGANIZATION_f1": 0.9354838709677419, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.9539473684210527, "eval_ORGANIZATION_recall": 0.9177215189873418, "eval_PERSON_f1": 0.976, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.9682539682539683, "eval_PERSON_recall": 0.9838709677419355, "eval_loss": 0.04839925840497017, "eval_overall_accuracy": 0.990118034586879, "eval_overall_f1": 0.9496221662468514, "eval_overall_precision": 0.952020202020202, "eval_overall_recall": 0.9472361809045227, "eval_runtime": 0.632, "eval_samples_per_second": 268.998, "eval_steps_per_second": 4.747, "step": 7296 }, { "epoch": 77.0, "grad_norm": 0.22566446661949158, "learning_rate": 1.1500000000000002e-05, "loss": 0.0099, "step": 7392 }, { "epoch": 77.0, "eval_LOCATION_f1": 0.9406779661016949, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.925, "eval_LOCATION_recall": 0.9568965517241379, "eval_ORGANIZATION_f1": 0.9255663430420711, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.9470198675496688, "eval_ORGANIZATION_recall": 0.9050632911392406, "eval_PERSON_f1": 0.976, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.9682539682539683, "eval_PERSON_recall": 0.9838709677419355, "eval_loss": 0.049101416021585464, "eval_overall_accuracy": 0.9892945374691189, "eval_overall_f1": 0.9459119496855346, "eval_overall_precision": 0.947103274559194, "eval_overall_recall": 0.9447236180904522, "eval_runtime": 0.6289, "eval_samples_per_second": 270.32, "eval_steps_per_second": 4.77, "step": 7392 }, { "epoch": 78.0, "grad_norm": 0.487615168094635, "learning_rate": 1.1000000000000001e-05, "loss": 0.0097, "step": 7488 }, { "epoch": 78.0, "eval_LOCATION_f1": 0.9361702127659575, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.9243697478991597, "eval_LOCATION_recall": 0.9482758620689655, "eval_ORGANIZATION_f1": 0.9320388349514563, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.9536423841059603, "eval_ORGANIZATION_recall": 0.9113924050632911, "eval_PERSON_f1": 0.976, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.9682539682539683, "eval_PERSON_recall": 0.9838709677419355, "eval_loss": 0.04617413133382797, "eval_overall_accuracy": 0.9898435355476256, "eval_overall_f1": 0.947103274559194, "eval_overall_precision": 0.9494949494949495, "eval_overall_recall": 0.9447236180904522, "eval_runtime": 0.6976, "eval_samples_per_second": 243.689, "eval_steps_per_second": 4.3, "step": 7488 }, { "epoch": 79.0, "grad_norm": 0.5360363125801086, "learning_rate": 1.05e-05, "loss": 0.0086, "step": 7584 }, { "epoch": 79.0, "eval_LOCATION_f1": 0.9310344827586207, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.9310344827586207, "eval_LOCATION_recall": 0.9310344827586207, "eval_ORGANIZATION_f1": 0.9329073482428115, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.9419354838709677, "eval_ORGANIZATION_recall": 0.9240506329113924, "eval_PERSON_f1": 0.976, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.9682539682539683, "eval_PERSON_recall": 0.9838709677419355, "eval_loss": 0.04512044042348862, "eval_overall_accuracy": 0.990118034586879, "eval_overall_f1": 0.9459119496855346, "eval_overall_precision": 0.947103274559194, "eval_overall_recall": 0.9447236180904522, "eval_runtime": 0.6381, "eval_samples_per_second": 266.398, "eval_steps_per_second": 4.701, "step": 7584 }, { "epoch": 80.0, "grad_norm": 0.9246647953987122, "learning_rate": 1e-05, "loss": 0.0094, "step": 7680 }, { "epoch": 80.0, "eval_LOCATION_f1": 0.927038626609442, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.9230769230769231, "eval_LOCATION_recall": 0.9310344827586207, "eval_ORGANIZATION_f1": 0.9290322580645162, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.9473684210526315, "eval_ORGANIZATION_recall": 0.9113924050632911, "eval_PERSON_f1": 0.976, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.9682539682539683, "eval_PERSON_recall": 0.9838709677419355, "eval_loss": 0.04963825270533562, "eval_overall_accuracy": 0.9887455393906122, "eval_overall_f1": 0.9432534678436318, "eval_overall_precision": 0.9468354430379747, "eval_overall_recall": 0.9396984924623115, "eval_runtime": 0.632, "eval_samples_per_second": 268.974, "eval_steps_per_second": 4.747, "step": 7680 }, { "epoch": 81.0, "grad_norm": 1.0030280351638794, "learning_rate": 9.5e-06, "loss": 0.0079, "step": 7776 }, { "epoch": 81.0, "eval_LOCATION_f1": 0.9437229437229439, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.9478260869565217, "eval_LOCATION_recall": 0.9396551724137931, "eval_ORGANIZATION_f1": 0.9389067524115756, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.954248366013072, "eval_ORGANIZATION_recall": 0.9240506329113924, "eval_PERSON_f1": 0.976, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.9682539682539683, "eval_PERSON_recall": 0.9838709677419355, "eval_loss": 0.045180778950452805, "eval_overall_accuracy": 0.990118034586879, "eval_overall_f1": 0.952020202020202, "eval_overall_precision": 0.9568527918781726, "eval_overall_recall": 0.9472361809045227, "eval_runtime": 0.6235, "eval_samples_per_second": 272.674, "eval_steps_per_second": 4.812, "step": 7776 }, { "epoch": 82.0, "grad_norm": 0.25211992859840393, "learning_rate": 9e-06, "loss": 0.0075, "step": 7872 }, { "epoch": 82.0, "eval_LOCATION_f1": 0.9437229437229439, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.9478260869565217, "eval_LOCATION_recall": 0.9396551724137931, "eval_ORGANIZATION_f1": 0.9354838709677419, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.9539473684210527, "eval_ORGANIZATION_recall": 0.9177215189873418, "eval_PERSON_f1": 0.976, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.9682539682539683, "eval_PERSON_recall": 0.9838709677419355, "eval_loss": 0.0471523217856884, "eval_overall_accuracy": 0.990118034586879, "eval_overall_f1": 0.9506953223767383, "eval_overall_precision": 0.9567430025445293, "eval_overall_recall": 0.9447236180904522, "eval_runtime": 0.6256, "eval_samples_per_second": 271.741, "eval_steps_per_second": 4.795, "step": 7872 }, { "epoch": 83.0, "grad_norm": 1.0778411626815796, "learning_rate": 8.500000000000002e-06, "loss": 0.0083, "step": 7968 }, { "epoch": 83.0, "eval_LOCATION_f1": 0.9482758620689655, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.9482758620689655, "eval_LOCATION_recall": 0.9482758620689655, "eval_ORGANIZATION_f1": 0.9260450160771704, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.9411764705882353, "eval_ORGANIZATION_recall": 0.9113924050632911, "eval_PERSON_f1": 0.976, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.9682539682539683, "eval_PERSON_recall": 0.9838709677419355, "eval_loss": 0.05099722743034363, "eval_overall_accuracy": 0.9898435355476256, "eval_overall_f1": 0.948297604035309, "eval_overall_precision": 0.9518987341772152, "eval_overall_recall": 0.9447236180904522, "eval_runtime": 0.6334, "eval_samples_per_second": 268.414, "eval_steps_per_second": 4.737, "step": 7968 }, { "epoch": 84.0, "grad_norm": 0.1189350113272667, "learning_rate": 8.000000000000001e-06, "loss": 0.0085, "step": 8064 }, { "epoch": 84.0, "eval_LOCATION_f1": 0.9361702127659575, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.9243697478991597, "eval_LOCATION_recall": 0.9482758620689655, "eval_ORGANIZATION_f1": 0.926517571884984, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.9354838709677419, "eval_ORGANIZATION_recall": 0.9177215189873418, "eval_PERSON_f1": 0.976, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.9682539682539683, "eval_PERSON_recall": 0.9838709677419355, "eval_loss": 0.04737309366464615, "eval_overall_accuracy": 0.990118034586879, "eval_overall_f1": 0.9448621553884713, "eval_overall_precision": 0.9425, "eval_overall_recall": 0.9472361809045227, "eval_runtime": 0.6297, "eval_samples_per_second": 269.959, "eval_steps_per_second": 4.764, "step": 8064 }, { "epoch": 85.0, "grad_norm": 0.39557117223739624, "learning_rate": 7.5e-06, "loss": 0.0085, "step": 8160 }, { "epoch": 85.0, "eval_LOCATION_f1": 0.9310344827586207, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.9310344827586207, "eval_LOCATION_recall": 0.9310344827586207, "eval_ORGANIZATION_f1": 0.9329073482428115, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.9419354838709677, "eval_ORGANIZATION_recall": 0.9240506329113924, "eval_PERSON_f1": 0.976, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.9682539682539683, "eval_PERSON_recall": 0.9838709677419355, "eval_loss": 0.04798389971256256, "eval_overall_accuracy": 0.9898435355476256, "eval_overall_f1": 0.9459119496855346, "eval_overall_precision": 0.947103274559194, "eval_overall_recall": 0.9447236180904522, "eval_runtime": 0.6362, "eval_samples_per_second": 267.228, "eval_steps_per_second": 4.716, "step": 8160 }, { "epoch": 86.0, "grad_norm": 0.25670096278190613, "learning_rate": 7.000000000000001e-06, "loss": 0.0092, "step": 8256 }, { "epoch": 86.0, "eval_LOCATION_f1": 0.9356223175965666, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.9316239316239316, "eval_LOCATION_recall": 0.9396551724137931, "eval_ORGANIZATION_f1": 0.932475884244373, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.9477124183006536, "eval_ORGANIZATION_recall": 0.9177215189873418, "eval_PERSON_f1": 0.976, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.9682539682539683, "eval_PERSON_recall": 0.9838709677419355, "eval_loss": 0.04724019393324852, "eval_overall_accuracy": 0.9892945374691189, "eval_overall_f1": 0.947103274559194, "eval_overall_precision": 0.9494949494949495, "eval_overall_recall": 0.9447236180904522, "eval_runtime": 0.6265, "eval_samples_per_second": 271.352, "eval_steps_per_second": 4.789, "step": 8256 }, { "epoch": 87.0, "grad_norm": 0.2632824182510376, "learning_rate": 6.5000000000000004e-06, "loss": 0.0083, "step": 8352 }, { "epoch": 87.0, "eval_LOCATION_f1": 0.944206008583691, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.9401709401709402, "eval_LOCATION_recall": 0.9482758620689655, "eval_ORGANIZATION_f1": 0.932475884244373, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.9477124183006536, "eval_ORGANIZATION_recall": 0.9177215189873418, "eval_PERSON_f1": 0.976, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.9682539682539683, "eval_PERSON_recall": 0.9838709677419355, "eval_loss": 0.04643959552049637, "eval_overall_accuracy": 0.9903925336261323, "eval_overall_f1": 0.9496221662468514, "eval_overall_precision": 0.952020202020202, "eval_overall_recall": 0.9472361809045227, "eval_runtime": 0.6261, "eval_samples_per_second": 271.504, "eval_steps_per_second": 4.791, "step": 8352 }, { "epoch": 88.0, "grad_norm": 0.0976150631904602, "learning_rate": 6e-06, "loss": 0.0081, "step": 8448 }, { "epoch": 88.0, "eval_LOCATION_f1": 0.9446808510638298, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.9327731092436975, "eval_LOCATION_recall": 0.9568965517241379, "eval_ORGANIZATION_f1": 0.926517571884984, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.9354838709677419, "eval_ORGANIZATION_recall": 0.9177215189873418, "eval_PERSON_f1": 0.976, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.9682539682539683, "eval_PERSON_recall": 0.9838709677419355, "eval_loss": 0.044152893126010895, "eval_overall_accuracy": 0.9909415317046391, "eval_overall_f1": 0.9473684210526316, "eval_overall_precision": 0.945, "eval_overall_recall": 0.949748743718593, "eval_runtime": 0.631, "eval_samples_per_second": 269.414, "eval_steps_per_second": 4.754, "step": 8448 }, { "epoch": 89.0, "grad_norm": 0.5976393818855286, "learning_rate": 5.500000000000001e-06, "loss": 0.0085, "step": 8544 }, { "epoch": 89.0, "eval_LOCATION_f1": 0.9316239316239315, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.923728813559322, "eval_LOCATION_recall": 0.9396551724137931, "eval_ORGANIZATION_f1": 0.9235668789808917, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.9294871794871795, "eval_ORGANIZATION_recall": 0.9177215189873418, "eval_PERSON_f1": 0.976, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.9682539682539683, "eval_PERSON_recall": 0.9838709677419355, "eval_loss": 0.0436055026948452, "eval_overall_accuracy": 0.9903925336261323, "eval_overall_f1": 0.9423558897243106, "eval_overall_precision": 0.94, "eval_overall_recall": 0.9447236180904522, "eval_runtime": 0.7134, "eval_samples_per_second": 238.309, "eval_steps_per_second": 4.205, "step": 8544 }, { "epoch": 90.0, "grad_norm": 0.27976006269454956, "learning_rate": 5e-06, "loss": 0.0092, "step": 8640 }, { "epoch": 90.0, "eval_LOCATION_f1": 0.9531914893617022, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.9411764705882353, "eval_LOCATION_recall": 0.9655172413793104, "eval_ORGANIZATION_f1": 0.9260450160771704, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.9411764705882353, "eval_ORGANIZATION_recall": 0.9113924050632911, "eval_PERSON_f1": 0.976, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.9682539682539683, "eval_PERSON_recall": 0.9838709677419355, "eval_loss": 0.04579922929406166, "eval_overall_accuracy": 0.9906670326653857, "eval_overall_f1": 0.949748743718593, "eval_overall_precision": 0.949748743718593, "eval_overall_recall": 0.949748743718593, "eval_runtime": 0.6328, "eval_samples_per_second": 268.661, "eval_steps_per_second": 4.741, "step": 8640 }, { "epoch": 91.0, "grad_norm": 0.4517710208892822, "learning_rate": 4.5e-06, "loss": 0.0087, "step": 8736 }, { "epoch": 91.0, "eval_LOCATION_f1": 0.9391304347826087, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.9473684210526315, "eval_LOCATION_recall": 0.9310344827586207, "eval_ORGANIZATION_f1": 0.9294871794871796, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.9415584415584416, "eval_ORGANIZATION_recall": 0.9177215189873418, "eval_PERSON_f1": 0.976, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.9682539682539683, "eval_PERSON_recall": 0.9838709677419355, "eval_loss": 0.047343574464321136, "eval_overall_accuracy": 0.9895690365083722, "eval_overall_f1": 0.946969696969697, "eval_overall_precision": 0.9517766497461929, "eval_overall_recall": 0.9422110552763819, "eval_runtime": 0.622, "eval_samples_per_second": 273.308, "eval_steps_per_second": 4.823, "step": 8736 }, { "epoch": 92.0, "grad_norm": 0.05907585844397545, "learning_rate": 4.000000000000001e-06, "loss": 0.0084, "step": 8832 }, { "epoch": 92.0, "eval_LOCATION_f1": 0.9396551724137931, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.9396551724137931, "eval_LOCATION_recall": 0.9396551724137931, "eval_ORGANIZATION_f1": 0.9358974358974359, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.948051948051948, "eval_ORGANIZATION_recall": 0.9240506329113924, "eval_PERSON_f1": 0.976, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.9682539682539683, "eval_PERSON_recall": 0.9838709677419355, "eval_loss": 0.0475139319896698, "eval_overall_accuracy": 0.9903925336261323, "eval_overall_f1": 0.9496221662468514, "eval_overall_precision": 0.952020202020202, "eval_overall_recall": 0.9472361809045227, "eval_runtime": 0.6304, "eval_samples_per_second": 269.657, "eval_steps_per_second": 4.759, "step": 8832 }, { "epoch": 93.0, "grad_norm": 0.4860309660434723, "learning_rate": 3.5000000000000004e-06, "loss": 0.0087, "step": 8928 }, { "epoch": 93.0, "eval_LOCATION_f1": 0.944206008583691, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.9401709401709402, "eval_LOCATION_recall": 0.9482758620689655, "eval_ORGANIZATION_f1": 0.923076923076923, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.935064935064935, "eval_ORGANIZATION_recall": 0.9113924050632911, "eval_PERSON_f1": 0.976, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.9682539682539683, "eval_PERSON_recall": 0.9838709677419355, "eval_loss": 0.047235433012247086, "eval_overall_accuracy": 0.990118034586879, "eval_overall_f1": 0.9459119496855346, "eval_overall_precision": 0.947103274559194, "eval_overall_recall": 0.9447236180904522, "eval_runtime": 0.6379, "eval_samples_per_second": 266.497, "eval_steps_per_second": 4.703, "step": 8928 }, { "epoch": 94.0, "grad_norm": 1.2510859966278076, "learning_rate": 3e-06, "loss": 0.008, "step": 9024 }, { "epoch": 94.0, "eval_LOCATION_f1": 0.9316239316239315, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.923728813559322, "eval_LOCATION_recall": 0.9396551724137931, "eval_ORGANIZATION_f1": 0.926517571884984, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.9354838709677419, "eval_ORGANIZATION_recall": 0.9177215189873418, "eval_PERSON_f1": 0.976, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.9682539682539683, "eval_PERSON_recall": 0.9838709677419355, "eval_loss": 0.04603435844182968, "eval_overall_accuracy": 0.990118034586879, "eval_overall_f1": 0.9435382685069008, "eval_overall_precision": 0.9423558897243107, "eval_overall_recall": 0.9447236180904522, "eval_runtime": 0.6286, "eval_samples_per_second": 270.434, "eval_steps_per_second": 4.772, "step": 9024 }, { "epoch": 95.0, "grad_norm": 0.07371649146080017, "learning_rate": 2.5e-06, "loss": 0.0075, "step": 9120 }, { "epoch": 95.0, "eval_LOCATION_f1": 0.94017094017094, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.9322033898305084, "eval_LOCATION_recall": 0.9482758620689655, "eval_ORGANIZATION_f1": 0.926517571884984, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.9354838709677419, "eval_ORGANIZATION_recall": 0.9177215189873418, "eval_PERSON_f1": 0.976, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.9682539682539683, "eval_PERSON_recall": 0.9838709677419355, "eval_loss": 0.04587940499186516, "eval_overall_accuracy": 0.990118034586879, "eval_overall_f1": 0.9460476787954831, "eval_overall_precision": 0.9448621553884712, "eval_overall_recall": 0.9472361809045227, "eval_runtime": 0.6255, "eval_samples_per_second": 271.792, "eval_steps_per_second": 4.796, "step": 9120 }, { "epoch": 96.0, "grad_norm": 0.04202267527580261, "learning_rate": 2.0000000000000003e-06, "loss": 0.0088, "step": 9216 }, { "epoch": 96.0, "eval_LOCATION_f1": 0.94017094017094, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.9322033898305084, "eval_LOCATION_recall": 0.9482758620689655, "eval_ORGANIZATION_f1": 0.932475884244373, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.9477124183006536, "eval_ORGANIZATION_recall": 0.9177215189873418, "eval_PERSON_f1": 0.976, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.9682539682539683, "eval_PERSON_recall": 0.9838709677419355, "eval_loss": 0.04721646010875702, "eval_overall_accuracy": 0.9898435355476256, "eval_overall_f1": 0.9484276729559749, "eval_overall_precision": 0.9496221662468514, "eval_overall_recall": 0.9472361809045227, "eval_runtime": 0.6301, "eval_samples_per_second": 269.807, "eval_steps_per_second": 4.761, "step": 9216 }, { "epoch": 97.0, "grad_norm": 0.5052945613861084, "learning_rate": 1.5e-06, "loss": 0.0079, "step": 9312 }, { "epoch": 97.0, "eval_LOCATION_f1": 0.9316239316239315, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.923728813559322, "eval_LOCATION_recall": 0.9396551724137931, "eval_ORGANIZATION_f1": 0.932475884244373, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.9477124183006536, "eval_ORGANIZATION_recall": 0.9177215189873418, "eval_PERSON_f1": 0.976, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.9682539682539683, "eval_PERSON_recall": 0.9838709677419355, "eval_loss": 0.0471828393638134, "eval_overall_accuracy": 0.9898435355476256, "eval_overall_f1": 0.9459119496855346, "eval_overall_precision": 0.947103274559194, "eval_overall_recall": 0.9447236180904522, "eval_runtime": 0.6361, "eval_samples_per_second": 267.246, "eval_steps_per_second": 4.716, "step": 9312 }, { "epoch": 98.0, "grad_norm": 0.050978515297174454, "learning_rate": 1.0000000000000002e-06, "loss": 0.0075, "step": 9408 }, { "epoch": 98.0, "eval_LOCATION_f1": 0.94017094017094, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.9322033898305084, "eval_LOCATION_recall": 0.9482758620689655, "eval_ORGANIZATION_f1": 0.932475884244373, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.9477124183006536, "eval_ORGANIZATION_recall": 0.9177215189873418, "eval_PERSON_f1": 0.976, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.9682539682539683, "eval_PERSON_recall": 0.9838709677419355, "eval_loss": 0.04793929308652878, "eval_overall_accuracy": 0.990118034586879, "eval_overall_f1": 0.9484276729559749, "eval_overall_precision": 0.9496221662468514, "eval_overall_recall": 0.9472361809045227, "eval_runtime": 0.6282, "eval_samples_per_second": 270.61, "eval_steps_per_second": 4.775, "step": 9408 }, { "epoch": 99.0, "grad_norm": 0.024095578119158745, "learning_rate": 5.000000000000001e-07, "loss": 0.0075, "step": 9504 }, { "epoch": 99.0, "eval_LOCATION_f1": 0.9356223175965666, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.9316239316239316, "eval_LOCATION_recall": 0.9396551724137931, "eval_ORGANIZATION_f1": 0.932475884244373, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.9477124183006536, "eval_ORGANIZATION_recall": 0.9177215189873418, "eval_PERSON_f1": 0.976, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.9682539682539683, "eval_PERSON_recall": 0.9838709677419355, "eval_loss": 0.04786060005426407, "eval_overall_accuracy": 0.9898435355476256, "eval_overall_f1": 0.947103274559194, "eval_overall_precision": 0.9494949494949495, "eval_overall_recall": 0.9447236180904522, "eval_runtime": 0.6932, "eval_samples_per_second": 245.243, "eval_steps_per_second": 4.328, "step": 9504 }, { "epoch": 100.0, "grad_norm": 0.4516121745109558, "learning_rate": 0.0, "loss": 0.0079, "step": 9600 }, { "epoch": 100.0, "eval_LOCATION_f1": 0.9356223175965666, "eval_LOCATION_number": 116, "eval_LOCATION_precision": 0.9316239316239316, "eval_LOCATION_recall": 0.9396551724137931, "eval_ORGANIZATION_f1": 0.932475884244373, "eval_ORGANIZATION_number": 158, "eval_ORGANIZATION_precision": 0.9477124183006536, "eval_ORGANIZATION_recall": 0.9177215189873418, "eval_PERSON_f1": 0.976, "eval_PERSON_number": 124, "eval_PERSON_precision": 0.9682539682539683, "eval_PERSON_recall": 0.9838709677419355, "eval_loss": 0.04794635623693466, "eval_overall_accuracy": 0.9898435355476256, "eval_overall_f1": 0.947103274559194, "eval_overall_precision": 0.9494949494949495, "eval_overall_recall": 0.9447236180904522, "eval_runtime": 0.6898, "eval_samples_per_second": 246.445, "eval_steps_per_second": 4.349, "step": 9600 }, { "epoch": 100.0, "step": 9600, "total_flos": 4293597283743744.0, "train_loss": 0.037866643778979775, "train_runtime": 1083.6064, "train_samples_per_second": 141.287, "train_steps_per_second": 8.859 } ], "logging_steps": 500, "max_steps": 9600, "num_input_tokens_seen": 0, "num_train_epochs": 100, "save_steps": 500, "total_flos": 4293597283743744.0, "train_batch_size": 16, "trial_name": null, "trial_params": null }