{ "best_metric": null, "best_model_checkpoint": null, "epoch": 100.0, "eval_steps": 500, "global_step": 9600, "is_hyper_param_search": false, "is_local_process_zero": true, "is_world_process_zero": true, "log_history": [ { "epoch": 1.0, "grad_norm": 2.3345937728881836, "learning_rate": 4.9500000000000004e-05, "loss": 0.8465, "step": 96 }, { "epoch": 1.0, "eval_LOCATION_f1": 0.0, "eval_LOCATION_number": 94, "eval_LOCATION_precision": 0.0, "eval_LOCATION_recall": 0.0, "eval_ORGANIZATION_f1": 0.011764705882352941, "eval_ORGANIZATION_number": 167, "eval_ORGANIZATION_precision": 0.3333333333333333, "eval_ORGANIZATION_recall": 0.005988023952095809, "eval_PERSON_f1": 0.0, "eval_PERSON_number": 137, "eval_PERSON_precision": 0.0, "eval_PERSON_recall": 0.0, "eval_loss": 0.5494903326034546, "eval_overall_accuracy": 0.8345303867403315, "eval_overall_f1": 0.0049504950495049506, "eval_overall_precision": 0.16666666666666666, "eval_overall_recall": 0.002512562814070352, "eval_runtime": 0.252, "eval_samples_per_second": 674.584, "eval_steps_per_second": 11.904, "step": 96 }, { "epoch": 2.0, "grad_norm": 4.829708099365234, "learning_rate": 4.9e-05, "loss": 0.4598, "step": 192 }, { "epoch": 2.0, "eval_LOCATION_f1": 0.2047244094488189, "eval_LOCATION_number": 94, "eval_LOCATION_precision": 0.3939393939393939, "eval_LOCATION_recall": 0.13829787234042554, "eval_ORGANIZATION_f1": 0.38873239436619716, "eval_ORGANIZATION_number": 167, "eval_ORGANIZATION_precision": 0.3670212765957447, "eval_ORGANIZATION_recall": 0.41317365269461076, "eval_PERSON_f1": 0.39473684210526316, "eval_PERSON_number": 137, "eval_PERSON_precision": 0.30864197530864196, "eval_PERSON_recall": 0.5474452554744526, "eval_loss": 0.3268411457538605, "eval_overall_accuracy": 0.8922651933701657, "eval_overall_f1": 0.3642691415313225, "eval_overall_precision": 0.33836206896551724, "eval_overall_recall": 0.3944723618090452, "eval_runtime": 0.2465, "eval_samples_per_second": 689.616, "eval_steps_per_second": 12.17, "step": 192 }, { "epoch": 3.0, "grad_norm": 0.9785643815994263, "learning_rate": 4.85e-05, "loss": 0.3128, "step": 288 }, { "epoch": 3.0, "eval_LOCATION_f1": 0.4022346368715084, "eval_LOCATION_number": 94, "eval_LOCATION_precision": 0.4235294117647059, "eval_LOCATION_recall": 0.3829787234042553, "eval_ORGANIZATION_f1": 0.578512396694215, "eval_ORGANIZATION_number": 167, "eval_ORGANIZATION_precision": 0.5357142857142857, "eval_ORGANIZATION_recall": 0.6287425149700598, "eval_PERSON_f1": 0.6793650793650794, "eval_PERSON_number": 137, "eval_PERSON_precision": 0.601123595505618, "eval_PERSON_recall": 0.781021897810219, "eval_loss": 0.23010307550430298, "eval_overall_accuracy": 0.9328729281767956, "eval_overall_f1": 0.5787631271878646, "eval_overall_precision": 0.5403050108932462, "eval_overall_recall": 0.6231155778894473, "eval_runtime": 0.2466, "eval_samples_per_second": 689.359, "eval_steps_per_second": 12.165, "step": 288 }, { "epoch": 4.0, "grad_norm": 0.7525363564491272, "learning_rate": 4.8e-05, "loss": 0.2169, "step": 384 }, { "epoch": 4.0, "eval_LOCATION_f1": 0.5846153846153846, "eval_LOCATION_number": 94, "eval_LOCATION_precision": 0.5643564356435643, "eval_LOCATION_recall": 0.6063829787234043, "eval_ORGANIZATION_f1": 0.7127071823204418, "eval_ORGANIZATION_number": 167, "eval_ORGANIZATION_precision": 0.6615384615384615, "eval_ORGANIZATION_recall": 0.7724550898203593, "eval_PERSON_f1": 0.9014084507042254, "eval_PERSON_number": 137, "eval_PERSON_precision": 0.8707482993197279, "eval_PERSON_recall": 0.9343065693430657, "eval_loss": 0.14087924361228943, "eval_overall_accuracy": 0.9577348066298342, "eval_overall_f1": 0.746730083234245, "eval_overall_precision": 0.708803611738149, "eval_overall_recall": 0.7889447236180904, "eval_runtime": 0.2422, "eval_samples_per_second": 702.008, "eval_steps_per_second": 12.388, "step": 384 }, { "epoch": 5.0, "grad_norm": 0.9961393475532532, "learning_rate": 4.75e-05, "loss": 0.1458, "step": 480 }, { "epoch": 5.0, "eval_LOCATION_f1": 0.8217821782178217, "eval_LOCATION_number": 94, "eval_LOCATION_precision": 0.7685185185185185, "eval_LOCATION_recall": 0.8829787234042553, "eval_ORGANIZATION_f1": 0.7461773700305809, "eval_ORGANIZATION_number": 167, "eval_ORGANIZATION_precision": 0.7625, "eval_ORGANIZATION_recall": 0.7305389221556886, "eval_PERSON_f1": 0.9357142857142857, "eval_PERSON_number": 137, "eval_PERSON_precision": 0.916083916083916, "eval_PERSON_recall": 0.9562043795620438, "eval_loss": 0.10370707511901855, "eval_overall_accuracy": 0.9685082872928177, "eval_overall_f1": 0.830655129789864, "eval_overall_precision": 0.8175182481751825, "eval_overall_recall": 0.8442211055276382, "eval_runtime": 0.2489, "eval_samples_per_second": 682.954, "eval_steps_per_second": 12.052, "step": 480 }, { "epoch": 6.0, "grad_norm": 1.0409573316574097, "learning_rate": 4.7e-05, "loss": 0.1248, "step": 576 }, { "epoch": 6.0, "eval_LOCATION_f1": 0.8383838383838383, "eval_LOCATION_number": 94, "eval_LOCATION_precision": 0.7980769230769231, "eval_LOCATION_recall": 0.8829787234042553, "eval_ORGANIZATION_f1": 0.8069164265129684, "eval_ORGANIZATION_number": 167, "eval_ORGANIZATION_precision": 0.7777777777777778, "eval_ORGANIZATION_recall": 0.8383233532934131, "eval_PERSON_f1": 0.9537366548042704, "eval_PERSON_number": 137, "eval_PERSON_precision": 0.9305555555555556, "eval_PERSON_recall": 0.9781021897810219, "eval_loss": 0.08994247019290924, "eval_overall_accuracy": 0.9718232044198895, "eval_overall_f1": 0.8644067796610169, "eval_overall_precision": 0.8341121495327103, "eval_overall_recall": 0.8969849246231156, "eval_runtime": 0.2475, "eval_samples_per_second": 686.878, "eval_steps_per_second": 12.121, "step": 576 }, { "epoch": 7.0, "grad_norm": 1.0360184907913208, "learning_rate": 4.6500000000000005e-05, "loss": 0.1067, "step": 672 }, { "epoch": 7.0, "eval_LOCATION_f1": 0.848780487804878, "eval_LOCATION_number": 94, "eval_LOCATION_precision": 0.7837837837837838, "eval_LOCATION_recall": 0.925531914893617, "eval_ORGANIZATION_f1": 0.8048048048048048, "eval_ORGANIZATION_number": 167, "eval_ORGANIZATION_precision": 0.8072289156626506, "eval_ORGANIZATION_recall": 0.8023952095808383, "eval_PERSON_f1": 0.9537366548042704, "eval_PERSON_number": 137, "eval_PERSON_precision": 0.9305555555555556, "eval_PERSON_recall": 0.9781021897810219, "eval_loss": 0.08151240646839142, "eval_overall_accuracy": 0.9729281767955801, "eval_overall_f1": 0.8669108669108669, "eval_overall_precision": 0.8432304038004751, "eval_overall_recall": 0.8919597989949749, "eval_runtime": 0.2489, "eval_samples_per_second": 683.079, "eval_steps_per_second": 12.054, "step": 672 }, { "epoch": 8.0, "grad_norm": 0.4758455753326416, "learning_rate": 4.600000000000001e-05, "loss": 0.0964, "step": 768 }, { "epoch": 8.0, "eval_LOCATION_f1": 0.8844221105527638, "eval_LOCATION_number": 94, "eval_LOCATION_precision": 0.8380952380952381, "eval_LOCATION_recall": 0.9361702127659575, "eval_ORGANIZATION_f1": 0.8083832335329342, "eval_ORGANIZATION_number": 167, "eval_ORGANIZATION_precision": 0.8083832335329342, "eval_ORGANIZATION_recall": 0.8083832335329342, "eval_PERSON_f1": 0.9605734767025089, "eval_PERSON_number": 137, "eval_PERSON_precision": 0.9436619718309859, "eval_PERSON_recall": 0.9781021897810219, "eval_loss": 0.0750458613038063, "eval_overall_accuracy": 0.9751381215469613, "eval_overall_f1": 0.8793103448275863, "eval_overall_precision": 0.8623188405797102, "eval_overall_recall": 0.8969849246231156, "eval_runtime": 0.2469, "eval_samples_per_second": 688.478, "eval_steps_per_second": 12.15, "step": 768 }, { "epoch": 9.0, "grad_norm": 0.6534972190856934, "learning_rate": 4.55e-05, "loss": 0.0926, "step": 864 }, { "epoch": 9.0, "eval_LOCATION_f1": 0.864321608040201, "eval_LOCATION_number": 94, "eval_LOCATION_precision": 0.819047619047619, "eval_LOCATION_recall": 0.9148936170212766, "eval_ORGANIZATION_f1": 0.8495575221238938, "eval_ORGANIZATION_number": 167, "eval_ORGANIZATION_precision": 0.8372093023255814, "eval_ORGANIZATION_recall": 0.8622754491017964, "eval_PERSON_f1": 0.971223021582734, "eval_PERSON_number": 137, "eval_PERSON_precision": 0.9574468085106383, "eval_PERSON_recall": 0.9854014598540146, "eval_loss": 0.06837134063243866, "eval_overall_accuracy": 0.976243093922652, "eval_overall_f1": 0.8946078431372548, "eval_overall_precision": 0.8732057416267942, "eval_overall_recall": 0.9170854271356784, "eval_runtime": 0.2475, "eval_samples_per_second": 686.813, "eval_steps_per_second": 12.12, "step": 864 }, { "epoch": 10.0, "grad_norm": 0.6045637726783752, "learning_rate": 4.5e-05, "loss": 0.082, "step": 960 }, { "epoch": 10.0, "eval_LOCATION_f1": 0.878048780487805, "eval_LOCATION_number": 94, "eval_LOCATION_precision": 0.8108108108108109, "eval_LOCATION_recall": 0.9574468085106383, "eval_ORGANIZATION_f1": 0.8298507462686566, "eval_ORGANIZATION_number": 167, "eval_ORGANIZATION_precision": 0.8273809523809523, "eval_ORGANIZATION_recall": 0.8323353293413174, "eval_PERSON_f1": 0.9747292418772563, "eval_PERSON_number": 137, "eval_PERSON_precision": 0.9642857142857143, "eval_PERSON_recall": 0.9854014598540146, "eval_loss": 0.06697890162467957, "eval_overall_accuracy": 0.9765193370165746, "eval_overall_f1": 0.8910648714810282, "eval_overall_precision": 0.8687350835322196, "eval_overall_recall": 0.914572864321608, "eval_runtime": 0.2472, "eval_samples_per_second": 687.821, "eval_steps_per_second": 12.138, "step": 960 }, { "epoch": 11.0, "grad_norm": 1.1809117794036865, "learning_rate": 4.4500000000000004e-05, "loss": 0.0787, "step": 1056 }, { "epoch": 11.0, "eval_LOCATION_f1": 0.8969072164948454, "eval_LOCATION_number": 94, "eval_LOCATION_precision": 0.87, "eval_LOCATION_recall": 0.925531914893617, "eval_ORGANIZATION_f1": 0.8493975903614458, "eval_ORGANIZATION_number": 167, "eval_ORGANIZATION_precision": 0.8545454545454545, "eval_ORGANIZATION_recall": 0.844311377245509, "eval_PERSON_f1": 0.9710144927536232, "eval_PERSON_number": 137, "eval_PERSON_precision": 0.9640287769784173, "eval_PERSON_recall": 0.9781021897810219, "eval_loss": 0.06165817379951477, "eval_overall_accuracy": 0.9792817679558011, "eval_overall_f1": 0.9027431421446384, "eval_overall_precision": 0.8960396039603961, "eval_overall_recall": 0.9095477386934674, "eval_runtime": 0.2479, "eval_samples_per_second": 685.644, "eval_steps_per_second": 12.1, "step": 1056 }, { "epoch": 12.0, "grad_norm": 0.9423420429229736, "learning_rate": 4.4000000000000006e-05, "loss": 0.0759, "step": 1152 }, { "epoch": 12.0, "eval_LOCATION_f1": 0.8855721393034826, "eval_LOCATION_number": 94, "eval_LOCATION_precision": 0.8317757009345794, "eval_LOCATION_recall": 0.9468085106382979, "eval_ORGANIZATION_f1": 0.8520710059171598, "eval_ORGANIZATION_number": 167, "eval_ORGANIZATION_precision": 0.8421052631578947, "eval_ORGANIZATION_recall": 0.8622754491017964, "eval_PERSON_f1": 0.9782608695652174, "eval_PERSON_number": 137, "eval_PERSON_precision": 0.9712230215827338, "eval_PERSON_recall": 0.9854014598540146, "eval_loss": 0.059144098311662674, "eval_overall_accuracy": 0.9781767955801105, "eval_overall_f1": 0.9030674846625767, "eval_overall_precision": 0.882494004796163, "eval_overall_recall": 0.9246231155778895, "eval_runtime": 0.2404, "eval_samples_per_second": 707.165, "eval_steps_per_second": 12.479, "step": 1152 }, { "epoch": 13.0, "grad_norm": 1.4245811700820923, "learning_rate": 4.35e-05, "loss": 0.069, "step": 1248 }, { "epoch": 13.0, "eval_LOCATION_f1": 0.8955223880597015, "eval_LOCATION_number": 94, "eval_LOCATION_precision": 0.8411214953271028, "eval_LOCATION_recall": 0.9574468085106383, "eval_ORGANIZATION_f1": 0.851063829787234, "eval_ORGANIZATION_number": 167, "eval_ORGANIZATION_precision": 0.8641975308641975, "eval_ORGANIZATION_recall": 0.8383233532934131, "eval_PERSON_f1": 0.9745454545454545, "eval_PERSON_number": 137, "eval_PERSON_precision": 0.9710144927536232, "eval_PERSON_recall": 0.9781021897810219, "eval_loss": 0.058085229247808456, "eval_overall_accuracy": 0.9798342541436464, "eval_overall_f1": 0.9043478260869565, "eval_overall_precision": 0.8943488943488943, "eval_overall_recall": 0.914572864321608, "eval_runtime": 0.2488, "eval_samples_per_second": 683.265, "eval_steps_per_second": 12.058, "step": 1248 }, { "epoch": 14.0, "grad_norm": 0.9371987581253052, "learning_rate": 4.3e-05, "loss": 0.0648, "step": 1344 }, { "epoch": 14.0, "eval_LOCATION_f1": 0.8855721393034826, "eval_LOCATION_number": 94, "eval_LOCATION_precision": 0.8317757009345794, "eval_LOCATION_recall": 0.9468085106382979, "eval_ORGANIZATION_f1": 0.8630952380952381, "eval_ORGANIZATION_number": 167, "eval_ORGANIZATION_precision": 0.8579881656804734, "eval_ORGANIZATION_recall": 0.8682634730538922, "eval_PERSON_f1": 0.9818181818181817, "eval_PERSON_number": 137, "eval_PERSON_precision": 0.9782608695652174, "eval_PERSON_recall": 0.9854014598540146, "eval_loss": 0.053421638906002045, "eval_overall_accuracy": 0.980110497237569, "eval_overall_f1": 0.9088669950738916, "eval_overall_precision": 0.8913043478260869, "eval_overall_recall": 0.9271356783919598, "eval_runtime": 0.2473, "eval_samples_per_second": 687.322, "eval_steps_per_second": 12.129, "step": 1344 }, { "epoch": 15.0, "grad_norm": 0.8175836205482483, "learning_rate": 4.25e-05, "loss": 0.0637, "step": 1440 }, { "epoch": 15.0, "eval_LOCATION_f1": 0.8944723618090452, "eval_LOCATION_number": 94, "eval_LOCATION_precision": 0.8476190476190476, "eval_LOCATION_recall": 0.9468085106382979, "eval_ORGANIZATION_f1": 0.8776119402985075, "eval_ORGANIZATION_number": 167, "eval_ORGANIZATION_precision": 0.875, "eval_ORGANIZATION_recall": 0.8802395209580839, "eval_PERSON_f1": 0.9854014598540146, "eval_PERSON_number": 137, "eval_PERSON_precision": 0.9854014598540146, "eval_PERSON_recall": 0.9854014598540146, "eval_loss": 0.05094514787197113, "eval_overall_accuracy": 0.9820441988950276, "eval_overall_f1": 0.9183168316831684, "eval_overall_precision": 0.9048780487804878, "eval_overall_recall": 0.9321608040201005, "eval_runtime": 0.2481, "eval_samples_per_second": 685.255, "eval_steps_per_second": 12.093, "step": 1440 }, { "epoch": 16.0, "grad_norm": 0.5499685406684875, "learning_rate": 4.2e-05, "loss": 0.061, "step": 1536 }, { "epoch": 16.0, "eval_LOCATION_f1": 0.8900523560209425, "eval_LOCATION_number": 94, "eval_LOCATION_precision": 0.8762886597938144, "eval_LOCATION_recall": 0.9042553191489362, "eval_ORGANIZATION_f1": 0.8830409356725146, "eval_ORGANIZATION_number": 167, "eval_ORGANIZATION_precision": 0.8628571428571429, "eval_ORGANIZATION_recall": 0.9041916167664671, "eval_PERSON_f1": 0.9926470588235294, "eval_PERSON_number": 137, "eval_PERSON_precision": 1.0, "eval_PERSON_recall": 0.9854014598540146, "eval_loss": 0.04840411618351936, "eval_overall_accuracy": 0.9825966850828729, "eval_overall_f1": 0.9217391304347826, "eval_overall_precision": 0.9115479115479116, "eval_overall_recall": 0.9321608040201005, "eval_runtime": 0.2468, "eval_samples_per_second": 688.682, "eval_steps_per_second": 12.153, "step": 1536 }, { "epoch": 17.0, "grad_norm": 1.1770079135894775, "learning_rate": 4.15e-05, "loss": 0.056, "step": 1632 }, { "epoch": 17.0, "eval_LOCATION_f1": 0.898989898989899, "eval_LOCATION_number": 94, "eval_LOCATION_precision": 0.8557692307692307, "eval_LOCATION_recall": 0.9468085106382979, "eval_ORGANIZATION_f1": 0.8630952380952381, "eval_ORGANIZATION_number": 167, "eval_ORGANIZATION_precision": 0.8579881656804734, "eval_ORGANIZATION_recall": 0.8682634730538922, "eval_PERSON_f1": 0.9816849816849818, "eval_PERSON_number": 137, "eval_PERSON_precision": 0.9852941176470589, "eval_PERSON_recall": 0.9781021897810219, "eval_loss": 0.050229690968990326, "eval_overall_accuracy": 0.9823204419889503, "eval_overall_f1": 0.9120198265179678, "eval_overall_precision": 0.8997555012224939, "eval_overall_recall": 0.9246231155778895, "eval_runtime": 0.2488, "eval_samples_per_second": 683.35, "eval_steps_per_second": 12.059, "step": 1632 }, { "epoch": 18.0, "grad_norm": 1.0919617414474487, "learning_rate": 4.1e-05, "loss": 0.0566, "step": 1728 }, { "epoch": 18.0, "eval_LOCATION_f1": 0.8934010152284263, "eval_LOCATION_number": 94, "eval_LOCATION_precision": 0.8543689320388349, "eval_LOCATION_recall": 0.9361702127659575, "eval_ORGANIZATION_f1": 0.8716417910447762, "eval_ORGANIZATION_number": 167, "eval_ORGANIZATION_precision": 0.8690476190476191, "eval_ORGANIZATION_recall": 0.874251497005988, "eval_PERSON_f1": 0.9816849816849818, "eval_PERSON_number": 137, "eval_PERSON_precision": 0.9852941176470589, "eval_PERSON_recall": 0.9781021897810219, "eval_loss": 0.048194997012615204, "eval_overall_accuracy": 0.9825966850828729, "eval_overall_f1": 0.9142857142857144, "eval_overall_precision": 0.9041769041769042, "eval_overall_recall": 0.9246231155778895, "eval_runtime": 0.2468, "eval_samples_per_second": 688.802, "eval_steps_per_second": 12.155, "step": 1728 }, { "epoch": 19.0, "grad_norm": 0.33032482862472534, "learning_rate": 4.05e-05, "loss": 0.0508, "step": 1824 }, { "epoch": 19.0, "eval_LOCATION_f1": 0.8923076923076922, "eval_LOCATION_number": 94, "eval_LOCATION_precision": 0.8613861386138614, "eval_LOCATION_recall": 0.925531914893617, "eval_ORGANIZATION_f1": 0.874251497005988, "eval_ORGANIZATION_number": 167, "eval_ORGANIZATION_precision": 0.874251497005988, "eval_ORGANIZATION_recall": 0.874251497005988, "eval_PERSON_f1": 0.9816849816849818, "eval_PERSON_number": 137, "eval_PERSON_precision": 0.9852941176470589, "eval_PERSON_recall": 0.9781021897810219, "eval_loss": 0.0475204698741436, "eval_overall_accuracy": 0.9831491712707182, "eval_overall_f1": 0.915211970074813, "eval_overall_precision": 0.9084158415841584, "eval_overall_recall": 0.9221105527638191, "eval_runtime": 0.242, "eval_samples_per_second": 702.469, "eval_steps_per_second": 12.397, "step": 1824 }, { "epoch": 20.0, "grad_norm": 0.6420110464096069, "learning_rate": 4e-05, "loss": 0.05, "step": 1920 }, { "epoch": 20.0, "eval_LOCATION_f1": 0.8844221105527638, "eval_LOCATION_number": 94, "eval_LOCATION_precision": 0.8380952380952381, "eval_LOCATION_recall": 0.9361702127659575, "eval_ORGANIZATION_f1": 0.8588588588588588, "eval_ORGANIZATION_number": 167, "eval_ORGANIZATION_precision": 0.8614457831325302, "eval_ORGANIZATION_recall": 0.8562874251497006, "eval_PERSON_f1": 0.989010989010989, "eval_PERSON_number": 137, "eval_PERSON_precision": 0.9926470588235294, "eval_PERSON_recall": 0.9854014598540146, "eval_loss": 0.050045743584632874, "eval_overall_accuracy": 0.9820441988950276, "eval_overall_f1": 0.9093167701863355, "eval_overall_precision": 0.8992628992628993, "eval_overall_recall": 0.9195979899497487, "eval_runtime": 0.2477, "eval_samples_per_second": 686.269, "eval_steps_per_second": 12.111, "step": 1920 }, { "epoch": 21.0, "grad_norm": 1.2013027667999268, "learning_rate": 3.9500000000000005e-05, "loss": 0.0481, "step": 2016 }, { "epoch": 21.0, "eval_LOCATION_f1": 0.8712871287128713, "eval_LOCATION_number": 94, "eval_LOCATION_precision": 0.8148148148148148, "eval_LOCATION_recall": 0.9361702127659575, "eval_ORGANIZATION_f1": 0.863768115942029, "eval_ORGANIZATION_number": 167, "eval_ORGANIZATION_precision": 0.8370786516853933, "eval_ORGANIZATION_recall": 0.8922155688622755, "eval_PERSON_f1": 0.9854014598540146, "eval_PERSON_number": 137, "eval_PERSON_precision": 0.9854014598540146, "eval_PERSON_recall": 0.9854014598540146, "eval_loss": 0.052587978541851044, "eval_overall_accuracy": 0.9806629834254144, "eval_overall_f1": 0.9062119366626066, "eval_overall_precision": 0.8794326241134752, "eval_overall_recall": 0.9346733668341709, "eval_runtime": 0.247, "eval_samples_per_second": 688.122, "eval_steps_per_second": 12.143, "step": 2016 }, { "epoch": 22.0, "grad_norm": 0.5056206583976746, "learning_rate": 3.9000000000000006e-05, "loss": 0.0456, "step": 2112 }, { "epoch": 22.0, "eval_LOCATION_f1": 0.9285714285714286, "eval_LOCATION_number": 94, "eval_LOCATION_precision": 0.8921568627450981, "eval_LOCATION_recall": 0.9680851063829787, "eval_ORGANIZATION_f1": 0.875739644970414, "eval_ORGANIZATION_number": 167, "eval_ORGANIZATION_precision": 0.8654970760233918, "eval_ORGANIZATION_recall": 0.8862275449101796, "eval_PERSON_f1": 0.9816849816849818, "eval_PERSON_number": 137, "eval_PERSON_precision": 0.9852941176470589, "eval_PERSON_recall": 0.9781021897810219, "eval_loss": 0.04648258164525032, "eval_overall_accuracy": 0.9842541436464088, "eval_overall_f1": 0.9244114002478315, "eval_overall_precision": 0.9119804400977995, "eval_overall_recall": 0.9371859296482412, "eval_runtime": 0.2482, "eval_samples_per_second": 685.038, "eval_steps_per_second": 12.089, "step": 2112 }, { "epoch": 23.0, "grad_norm": 0.698632001876831, "learning_rate": 3.85e-05, "loss": 0.0438, "step": 2208 }, { "epoch": 23.0, "eval_LOCATION_f1": 0.9045226130653266, "eval_LOCATION_number": 94, "eval_LOCATION_precision": 0.8571428571428571, "eval_LOCATION_recall": 0.9574468085106383, "eval_ORGANIZATION_f1": 0.8343558282208589, "eval_ORGANIZATION_number": 167, "eval_ORGANIZATION_precision": 0.8553459119496856, "eval_ORGANIZATION_recall": 0.8143712574850299, "eval_PERSON_f1": 0.989010989010989, "eval_PERSON_number": 137, "eval_PERSON_precision": 0.9926470588235294, "eval_PERSON_recall": 0.9854014598540146, "eval_loss": 0.055541761219501495, "eval_overall_accuracy": 0.9809392265193371, "eval_overall_f1": 0.9047619047619048, "eval_overall_precision": 0.9025, "eval_overall_recall": 0.907035175879397, "eval_runtime": 0.2481, "eval_samples_per_second": 685.131, "eval_steps_per_second": 12.091, "step": 2208 }, { "epoch": 24.0, "grad_norm": 0.32339999079704285, "learning_rate": 3.8e-05, "loss": 0.0417, "step": 2304 }, { "epoch": 24.0, "eval_LOCATION_f1": 0.9128205128205128, "eval_LOCATION_number": 94, "eval_LOCATION_precision": 0.8811881188118812, "eval_LOCATION_recall": 0.9468085106382979, "eval_ORGANIZATION_f1": 0.9053254437869822, "eval_ORGANIZATION_number": 167, "eval_ORGANIZATION_precision": 0.8947368421052632, "eval_ORGANIZATION_recall": 0.9161676646706587, "eval_PERSON_f1": 0.9926470588235294, "eval_PERSON_number": 137, "eval_PERSON_precision": 1.0, "eval_PERSON_recall": 0.9854014598540146, "eval_loss": 0.04770859703421593, "eval_overall_accuracy": 0.9861878453038674, "eval_overall_f1": 0.9366459627329192, "eval_overall_precision": 0.9262899262899262, "eval_overall_recall": 0.9472361809045227, "eval_runtime": 0.2459, "eval_samples_per_second": 691.217, "eval_steps_per_second": 12.198, "step": 2304 }, { "epoch": 25.0, "grad_norm": 0.3970746695995331, "learning_rate": 3.7500000000000003e-05, "loss": 0.0412, "step": 2400 }, { "epoch": 25.0, "eval_LOCATION_f1": 0.8955223880597015, "eval_LOCATION_number": 94, "eval_LOCATION_precision": 0.8411214953271028, "eval_LOCATION_recall": 0.9574468085106383, "eval_ORGANIZATION_f1": 0.8588588588588588, "eval_ORGANIZATION_number": 167, "eval_ORGANIZATION_precision": 0.8614457831325302, "eval_ORGANIZATION_recall": 0.8562874251497006, "eval_PERSON_f1": 0.989010989010989, "eval_PERSON_number": 137, "eval_PERSON_precision": 0.9926470588235294, "eval_PERSON_recall": 0.9854014598540146, "eval_loss": 0.0511150024831295, "eval_overall_accuracy": 0.9828729281767956, "eval_overall_f1": 0.9120198265179678, "eval_overall_precision": 0.8997555012224939, "eval_overall_recall": 0.9246231155778895, "eval_runtime": 0.2466, "eval_samples_per_second": 689.31, "eval_steps_per_second": 12.164, "step": 2400 }, { "epoch": 26.0, "grad_norm": 0.6808413863182068, "learning_rate": 3.7e-05, "loss": 0.0377, "step": 2496 }, { "epoch": 26.0, "eval_LOCATION_f1": 0.9285714285714286, "eval_LOCATION_number": 94, "eval_LOCATION_precision": 0.8921568627450981, "eval_LOCATION_recall": 0.9680851063829787, "eval_ORGANIZATION_f1": 0.8802395209580839, "eval_ORGANIZATION_number": 167, "eval_ORGANIZATION_precision": 0.8802395209580839, "eval_ORGANIZATION_recall": 0.8802395209580839, "eval_PERSON_f1": 0.9854014598540146, "eval_PERSON_number": 137, "eval_PERSON_precision": 0.9854014598540146, "eval_PERSON_recall": 0.9854014598540146, "eval_loss": 0.04815667122602463, "eval_overall_accuracy": 0.9845303867403314, "eval_overall_f1": 0.927860696517413, "eval_overall_precision": 0.9187192118226601, "eval_overall_recall": 0.9371859296482412, "eval_runtime": 0.2475, "eval_samples_per_second": 686.962, "eval_steps_per_second": 12.123, "step": 2496 }, { "epoch": 27.0, "grad_norm": 0.7148529887199402, "learning_rate": 3.65e-05, "loss": 0.0394, "step": 2592 }, { "epoch": 27.0, "eval_LOCATION_f1": 0.9222797927461138, "eval_LOCATION_number": 94, "eval_LOCATION_precision": 0.898989898989899, "eval_LOCATION_recall": 0.9468085106382979, "eval_ORGANIZATION_f1": 0.8862275449101796, "eval_ORGANIZATION_number": 167, "eval_ORGANIZATION_precision": 0.8862275449101796, "eval_ORGANIZATION_recall": 0.8862275449101796, "eval_PERSON_f1": 0.9816849816849818, "eval_PERSON_number": 137, "eval_PERSON_precision": 0.9852941176470589, "eval_PERSON_recall": 0.9781021897810219, "eval_loss": 0.04498957470059395, "eval_overall_accuracy": 0.9842541436464088, "eval_overall_f1": 0.9275, "eval_overall_precision": 0.9228855721393034, "eval_overall_recall": 0.9321608040201005, "eval_runtime": 0.2463, "eval_samples_per_second": 690.18, "eval_steps_per_second": 12.18, "step": 2592 }, { "epoch": 28.0, "grad_norm": 0.31477001309394836, "learning_rate": 3.6e-05, "loss": 0.0352, "step": 2688 }, { "epoch": 28.0, "eval_LOCATION_f1": 0.9278350515463918, "eval_LOCATION_number": 94, "eval_LOCATION_precision": 0.9, "eval_LOCATION_recall": 0.9574468085106383, "eval_ORGANIZATION_f1": 0.8828828828828829, "eval_ORGANIZATION_number": 167, "eval_ORGANIZATION_precision": 0.8855421686746988, "eval_ORGANIZATION_recall": 0.8802395209580839, "eval_PERSON_f1": 0.9781021897810219, "eval_PERSON_number": 137, "eval_PERSON_precision": 0.9781021897810219, "eval_PERSON_recall": 0.9781021897810219, "eval_loss": 0.04605722054839134, "eval_overall_accuracy": 0.9842541436464088, "eval_overall_f1": 0.9263420724094882, "eval_overall_precision": 0.9205955334987593, "eval_overall_recall": 0.9321608040201005, "eval_runtime": 0.2479, "eval_samples_per_second": 685.773, "eval_steps_per_second": 12.102, "step": 2688 }, { "epoch": 29.0, "grad_norm": 0.630024254322052, "learning_rate": 3.55e-05, "loss": 0.0337, "step": 2784 }, { "epoch": 29.0, "eval_LOCATION_f1": 0.9035532994923857, "eval_LOCATION_number": 94, "eval_LOCATION_precision": 0.8640776699029126, "eval_LOCATION_recall": 0.9468085106382979, "eval_ORGANIZATION_f1": 0.8835820895522388, "eval_ORGANIZATION_number": 167, "eval_ORGANIZATION_precision": 0.8809523809523809, "eval_ORGANIZATION_recall": 0.8862275449101796, "eval_PERSON_f1": 0.989010989010989, "eval_PERSON_number": 137, "eval_PERSON_precision": 0.9926470588235294, "eval_PERSON_recall": 0.9854014598540146, "eval_loss": 0.0465216338634491, "eval_overall_accuracy": 0.9845303867403314, "eval_overall_f1": 0.924223602484472, "eval_overall_precision": 0.914004914004914, "eval_overall_recall": 0.9346733668341709, "eval_runtime": 0.2462, "eval_samples_per_second": 690.434, "eval_steps_per_second": 12.184, "step": 2784 }, { "epoch": 30.0, "grad_norm": 0.8620179295539856, "learning_rate": 3.5e-05, "loss": 0.0329, "step": 2880 }, { "epoch": 30.0, "eval_LOCATION_f1": 0.9072164948453608, "eval_LOCATION_number": 94, "eval_LOCATION_precision": 0.88, "eval_LOCATION_recall": 0.9361702127659575, "eval_ORGANIZATION_f1": 0.8895522388059701, "eval_ORGANIZATION_number": 167, "eval_ORGANIZATION_precision": 0.8869047619047619, "eval_ORGANIZATION_recall": 0.8922155688622755, "eval_PERSON_f1": 0.989010989010989, "eval_PERSON_number": 137, "eval_PERSON_precision": 0.9926470588235294, "eval_PERSON_recall": 0.9854014598540146, "eval_loss": 0.046264734119176865, "eval_overall_accuracy": 0.9839779005524862, "eval_overall_f1": 0.9276807980049875, "eval_overall_precision": 0.9207920792079208, "eval_overall_recall": 0.9346733668341709, "eval_runtime": 0.2469, "eval_samples_per_second": 688.455, "eval_steps_per_second": 12.149, "step": 2880 }, { "epoch": 31.0, "grad_norm": 1.0834476947784424, "learning_rate": 3.45e-05, "loss": 0.0323, "step": 2976 }, { "epoch": 31.0, "eval_LOCATION_f1": 0.9230769230769231, "eval_LOCATION_number": 94, "eval_LOCATION_precision": 0.8910891089108911, "eval_LOCATION_recall": 0.9574468085106383, "eval_ORGANIZATION_f1": 0.8821752265861027, "eval_ORGANIZATION_number": 167, "eval_ORGANIZATION_precision": 0.8902439024390244, "eval_ORGANIZATION_recall": 0.874251497005988, "eval_PERSON_f1": 0.9854014598540146, "eval_PERSON_number": 137, "eval_PERSON_precision": 0.9854014598540146, "eval_PERSON_recall": 0.9854014598540146, "eval_loss": 0.04753027483820915, "eval_overall_accuracy": 0.9850828729281768, "eval_overall_f1": 0.9275, "eval_overall_precision": 0.9228855721393034, "eval_overall_recall": 0.9321608040201005, "eval_runtime": 0.247, "eval_samples_per_second": 688.15, "eval_steps_per_second": 12.144, "step": 2976 }, { "epoch": 32.0, "grad_norm": 0.5082409977912903, "learning_rate": 3.4000000000000007e-05, "loss": 0.0297, "step": 3072 }, { "epoch": 32.0, "eval_LOCATION_f1": 0.9145728643216081, "eval_LOCATION_number": 94, "eval_LOCATION_precision": 0.8666666666666667, "eval_LOCATION_recall": 0.9680851063829787, "eval_ORGANIZATION_f1": 0.8761329305135951, "eval_ORGANIZATION_number": 167, "eval_ORGANIZATION_precision": 0.8841463414634146, "eval_ORGANIZATION_recall": 0.8682634730538922, "eval_PERSON_f1": 0.9781021897810219, "eval_PERSON_number": 137, "eval_PERSON_precision": 0.9781021897810219, "eval_PERSON_recall": 0.9781021897810219, "eval_loss": 0.048817116767168045, "eval_overall_accuracy": 0.9831491712707182, "eval_overall_f1": 0.9203980099502488, "eval_overall_precision": 0.9113300492610837, "eval_overall_recall": 0.9296482412060302, "eval_runtime": 0.2465, "eval_samples_per_second": 689.775, "eval_steps_per_second": 12.172, "step": 3072 }, { "epoch": 33.0, "grad_norm": 0.554415762424469, "learning_rate": 3.35e-05, "loss": 0.0306, "step": 3168 }, { "epoch": 33.0, "eval_LOCATION_f1": 0.9183673469387754, "eval_LOCATION_number": 94, "eval_LOCATION_precision": 0.8823529411764706, "eval_LOCATION_recall": 0.9574468085106383, "eval_ORGANIZATION_f1": 0.8855421686746987, "eval_ORGANIZATION_number": 167, "eval_ORGANIZATION_precision": 0.8909090909090909, "eval_ORGANIZATION_recall": 0.8802395209580839, "eval_PERSON_f1": 0.9854014598540146, "eval_PERSON_number": 137, "eval_PERSON_precision": 0.9854014598540146, "eval_PERSON_recall": 0.9854014598540146, "eval_loss": 0.048499397933483124, "eval_overall_accuracy": 0.9842541436464088, "eval_overall_f1": 0.9276807980049875, "eval_overall_precision": 0.9207920792079208, "eval_overall_recall": 0.9346733668341709, "eval_runtime": 0.2461, "eval_samples_per_second": 690.709, "eval_steps_per_second": 12.189, "step": 3168 }, { "epoch": 34.0, "grad_norm": 0.24838736653327942, "learning_rate": 3.3e-05, "loss": 0.0288, "step": 3264 }, { "epoch": 34.0, "eval_LOCATION_f1": 0.898989898989899, "eval_LOCATION_number": 94, "eval_LOCATION_precision": 0.8557692307692307, "eval_LOCATION_recall": 0.9468085106382979, "eval_ORGANIZATION_f1": 0.8727272727272728, "eval_ORGANIZATION_number": 167, "eval_ORGANIZATION_precision": 0.8834355828220859, "eval_ORGANIZATION_recall": 0.8622754491017964, "eval_PERSON_f1": 0.9854014598540146, "eval_PERSON_number": 137, "eval_PERSON_precision": 0.9854014598540146, "eval_PERSON_recall": 0.9854014598540146, "eval_loss": 0.05253738909959793, "eval_overall_accuracy": 0.9814917127071823, "eval_overall_f1": 0.917705735660848, "eval_overall_precision": 0.9108910891089109, "eval_overall_recall": 0.9246231155778895, "eval_runtime": 0.2468, "eval_samples_per_second": 688.896, "eval_steps_per_second": 12.157, "step": 3264 }, { "epoch": 35.0, "grad_norm": 0.43114426732063293, "learning_rate": 3.2500000000000004e-05, "loss": 0.0283, "step": 3360 }, { "epoch": 35.0, "eval_LOCATION_f1": 0.9230769230769231, "eval_LOCATION_number": 94, "eval_LOCATION_precision": 0.8910891089108911, "eval_LOCATION_recall": 0.9574468085106383, "eval_ORGANIZATION_f1": 0.8814589665653495, "eval_ORGANIZATION_number": 167, "eval_ORGANIZATION_precision": 0.8950617283950617, "eval_ORGANIZATION_recall": 0.8682634730538922, "eval_PERSON_f1": 0.9854014598540146, "eval_PERSON_number": 137, "eval_PERSON_precision": 0.9854014598540146, "eval_PERSON_recall": 0.9854014598540146, "eval_loss": 0.04848175123333931, "eval_overall_accuracy": 0.9845303867403314, "eval_overall_f1": 0.9273182957393483, "eval_overall_precision": 0.925, "eval_overall_recall": 0.9296482412060302, "eval_runtime": 0.2465, "eval_samples_per_second": 689.702, "eval_steps_per_second": 12.171, "step": 3360 }, { "epoch": 36.0, "grad_norm": 0.5155237913131714, "learning_rate": 3.2000000000000005e-05, "loss": 0.0289, "step": 3456 }, { "epoch": 36.0, "eval_LOCATION_f1": 0.8877551020408163, "eval_LOCATION_number": 94, "eval_LOCATION_precision": 0.8529411764705882, "eval_LOCATION_recall": 0.925531914893617, "eval_ORGANIZATION_f1": 0.8823529411764706, "eval_ORGANIZATION_number": 167, "eval_ORGANIZATION_precision": 0.8670520231213873, "eval_ORGANIZATION_recall": 0.8982035928143712, "eval_PERSON_f1": 0.9854014598540146, "eval_PERSON_number": 137, "eval_PERSON_precision": 0.9854014598540146, "eval_PERSON_recall": 0.9854014598540146, "eval_loss": 0.04846788942813873, "eval_overall_accuracy": 0.9820441988950276, "eval_overall_f1": 0.9185185185185186, "eval_overall_precision": 0.9029126213592233, "eval_overall_recall": 0.9346733668341709, "eval_runtime": 0.2465, "eval_samples_per_second": 689.596, "eval_steps_per_second": 12.169, "step": 3456 }, { "epoch": 37.0, "grad_norm": 1.0342140197753906, "learning_rate": 3.15e-05, "loss": 0.0266, "step": 3552 }, { "epoch": 37.0, "eval_LOCATION_f1": 0.8823529411764706, "eval_LOCATION_number": 94, "eval_LOCATION_precision": 0.8181818181818182, "eval_LOCATION_recall": 0.9574468085106383, "eval_ORGANIZATION_f1": 0.8615384615384616, "eval_ORGANIZATION_number": 167, "eval_ORGANIZATION_precision": 0.8860759493670886, "eval_ORGANIZATION_recall": 0.8383233532934131, "eval_PERSON_f1": 0.9854014598540146, "eval_PERSON_number": 137, "eval_PERSON_precision": 0.9854014598540146, "eval_PERSON_recall": 0.9854014598540146, "eval_loss": 0.05281976982951164, "eval_overall_accuracy": 0.9812154696132597, "eval_overall_f1": 0.9090909090909091, "eval_overall_precision": 0.9012345679012346, "eval_overall_recall": 0.9170854271356784, "eval_runtime": 0.2471, "eval_samples_per_second": 688.114, "eval_steps_per_second": 12.143, "step": 3552 }, { "epoch": 38.0, "grad_norm": 0.6417266130447388, "learning_rate": 3.1e-05, "loss": 0.0268, "step": 3648 }, { "epoch": 38.0, "eval_LOCATION_f1": 0.9381443298969071, "eval_LOCATION_number": 94, "eval_LOCATION_precision": 0.91, "eval_LOCATION_recall": 0.9680851063829787, "eval_ORGANIZATION_f1": 0.8841463414634148, "eval_ORGANIZATION_number": 167, "eval_ORGANIZATION_precision": 0.9006211180124224, "eval_ORGANIZATION_recall": 0.8682634730538922, "eval_PERSON_f1": 0.963768115942029, "eval_PERSON_number": 137, "eval_PERSON_precision": 0.9568345323741008, "eval_PERSON_recall": 0.9708029197080292, "eval_loss": 0.04737497866153717, "eval_overall_accuracy": 0.9848066298342542, "eval_overall_f1": 0.924812030075188, "eval_overall_precision": 0.9225, "eval_overall_recall": 0.9271356783919598, "eval_runtime": 0.2481, "eval_samples_per_second": 685.319, "eval_steps_per_second": 12.094, "step": 3648 }, { "epoch": 39.0, "grad_norm": 0.6291232705116272, "learning_rate": 3.05e-05, "loss": 0.0263, "step": 3744 }, { "epoch": 39.0, "eval_LOCATION_f1": 0.8695652173913044, "eval_LOCATION_number": 94, "eval_LOCATION_precision": 0.7964601769911505, "eval_LOCATION_recall": 0.9574468085106383, "eval_ORGANIZATION_f1": 0.8473520249221184, "eval_ORGANIZATION_number": 167, "eval_ORGANIZATION_precision": 0.8831168831168831, "eval_ORGANIZATION_recall": 0.8143712574850299, "eval_PERSON_f1": 0.9745454545454545, "eval_PERSON_number": 137, "eval_PERSON_precision": 0.9710144927536232, "eval_PERSON_recall": 0.9781021897810219, "eval_loss": 0.05302976816892624, "eval_overall_accuracy": 0.9809392265193371, "eval_overall_f1": 0.8966376089663761, "eval_overall_precision": 0.8888888888888888, "eval_overall_recall": 0.9045226130653267, "eval_runtime": 0.241, "eval_samples_per_second": 705.257, "eval_steps_per_second": 12.446, "step": 3744 }, { "epoch": 40.0, "grad_norm": 0.7022667527198792, "learning_rate": 3e-05, "loss": 0.0256, "step": 3840 }, { "epoch": 40.0, "eval_LOCATION_f1": 0.898989898989899, "eval_LOCATION_number": 94, "eval_LOCATION_precision": 0.8557692307692307, "eval_LOCATION_recall": 0.9468085106382979, "eval_ORGANIZATION_f1": 0.875739644970414, "eval_ORGANIZATION_number": 167, "eval_ORGANIZATION_precision": 0.8654970760233918, "eval_ORGANIZATION_recall": 0.8862275449101796, "eval_PERSON_f1": 0.9745454545454545, "eval_PERSON_number": 137, "eval_PERSON_precision": 0.9710144927536232, "eval_PERSON_recall": 0.9781021897810219, "eval_loss": 0.048180822283029556, "eval_overall_accuracy": 0.9831491712707182, "eval_overall_f1": 0.9149198520345253, "eval_overall_precision": 0.8983050847457628, "eval_overall_recall": 0.9321608040201005, "eval_runtime": 0.2465, "eval_samples_per_second": 689.522, "eval_steps_per_second": 12.168, "step": 3840 }, { "epoch": 41.0, "grad_norm": 0.1118118017911911, "learning_rate": 2.95e-05, "loss": 0.0241, "step": 3936 }, { "epoch": 41.0, "eval_LOCATION_f1": 0.89, "eval_LOCATION_number": 94, "eval_LOCATION_precision": 0.839622641509434, "eval_LOCATION_recall": 0.9468085106382979, "eval_ORGANIZATION_f1": 0.855421686746988, "eval_ORGANIZATION_number": 167, "eval_ORGANIZATION_precision": 0.8606060606060606, "eval_ORGANIZATION_recall": 0.8502994011976048, "eval_PERSON_f1": 0.963768115942029, "eval_PERSON_number": 137, "eval_PERSON_precision": 0.9568345323741008, "eval_PERSON_recall": 0.9708029197080292, "eval_loss": 0.04991815239191055, "eval_overall_accuracy": 0.9820441988950276, "eval_overall_f1": 0.9009900990099009, "eval_overall_precision": 0.8878048780487805, "eval_overall_recall": 0.914572864321608, "eval_runtime": 0.2404, "eval_samples_per_second": 707.176, "eval_steps_per_second": 12.48, "step": 3936 }, { "epoch": 42.0, "grad_norm": 1.0873582363128662, "learning_rate": 2.9e-05, "loss": 0.0241, "step": 4032 }, { "epoch": 42.0, "eval_LOCATION_f1": 0.9230769230769231, "eval_LOCATION_number": 94, "eval_LOCATION_precision": 0.8910891089108911, "eval_LOCATION_recall": 0.9574468085106383, "eval_ORGANIZATION_f1": 0.8768768768768769, "eval_ORGANIZATION_number": 167, "eval_ORGANIZATION_precision": 0.8795180722891566, "eval_ORGANIZATION_recall": 0.874251497005988, "eval_PERSON_f1": 0.9854014598540146, "eval_PERSON_number": 137, "eval_PERSON_precision": 0.9854014598540146, "eval_PERSON_recall": 0.9854014598540146, "eval_loss": 0.046265311539173126, "eval_overall_accuracy": 0.9845303867403314, "eval_overall_f1": 0.9251870324189527, "eval_overall_precision": 0.9183168316831684, "eval_overall_recall": 0.9321608040201005, "eval_runtime": 0.2468, "eval_samples_per_second": 688.713, "eval_steps_per_second": 12.154, "step": 4032 }, { "epoch": 43.0, "grad_norm": 0.4210592806339264, "learning_rate": 2.8499999999999998e-05, "loss": 0.0223, "step": 4128 }, { "epoch": 43.0, "eval_LOCATION_f1": 0.9238578680203046, "eval_LOCATION_number": 94, "eval_LOCATION_precision": 0.883495145631068, "eval_LOCATION_recall": 0.9680851063829787, "eval_ORGANIZATION_f1": 0.8814589665653495, "eval_ORGANIZATION_number": 167, "eval_ORGANIZATION_precision": 0.8950617283950617, "eval_ORGANIZATION_recall": 0.8682634730538922, "eval_PERSON_f1": 0.9781021897810219, "eval_PERSON_number": 137, "eval_PERSON_precision": 0.9781021897810219, "eval_PERSON_recall": 0.9781021897810219, "eval_loss": 0.051920872181653976, "eval_overall_accuracy": 0.9848066298342542, "eval_overall_f1": 0.925, "eval_overall_precision": 0.9203980099502488, "eval_overall_recall": 0.9296482412060302, "eval_runtime": 0.247, "eval_samples_per_second": 688.356, "eval_steps_per_second": 12.147, "step": 4128 }, { "epoch": 44.0, "grad_norm": 0.6354262232780457, "learning_rate": 2.8000000000000003e-05, "loss": 0.0211, "step": 4224 }, { "epoch": 44.0, "eval_LOCATION_f1": 0.9381443298969071, "eval_LOCATION_number": 94, "eval_LOCATION_precision": 0.91, "eval_LOCATION_recall": 0.9680851063829787, "eval_ORGANIZATION_f1": 0.8895705521472391, "eval_ORGANIZATION_number": 167, "eval_ORGANIZATION_precision": 0.9119496855345912, "eval_ORGANIZATION_recall": 0.8682634730538922, "eval_PERSON_f1": 0.9781021897810219, "eval_PERSON_number": 137, "eval_PERSON_precision": 0.9781021897810219, "eval_PERSON_recall": 0.9781021897810219, "eval_loss": 0.04981081187725067, "eval_overall_accuracy": 0.9856353591160221, "eval_overall_f1": 0.9319899244332494, "eval_overall_precision": 0.9343434343434344, "eval_overall_recall": 0.9296482412060302, "eval_runtime": 0.2523, "eval_samples_per_second": 673.815, "eval_steps_per_second": 11.891, "step": 4224 }, { "epoch": 45.0, "grad_norm": 0.4316812753677368, "learning_rate": 2.7500000000000004e-05, "loss": 0.0209, "step": 4320 }, { "epoch": 45.0, "eval_LOCATION_f1": 0.9278350515463918, "eval_LOCATION_number": 94, "eval_LOCATION_precision": 0.9, "eval_LOCATION_recall": 0.9574468085106383, "eval_ORGANIZATION_f1": 0.8975903614457832, "eval_ORGANIZATION_number": 167, "eval_ORGANIZATION_precision": 0.9030303030303031, "eval_ORGANIZATION_recall": 0.8922155688622755, "eval_PERSON_f1": 0.9745454545454545, "eval_PERSON_number": 137, "eval_PERSON_precision": 0.9710144927536232, "eval_PERSON_recall": 0.9781021897810219, "eval_loss": 0.04499218985438347, "eval_overall_accuracy": 0.9845303867403314, "eval_overall_f1": 0.9313358302122346, "eval_overall_precision": 0.9255583126550868, "eval_overall_recall": 0.9371859296482412, "eval_runtime": 0.246, "eval_samples_per_second": 691.049, "eval_steps_per_second": 12.195, "step": 4320 }, { "epoch": 46.0, "grad_norm": 0.22505725920200348, "learning_rate": 2.7000000000000002e-05, "loss": 0.0194, "step": 4416 }, { "epoch": 46.0, "eval_LOCATION_f1": 0.9183673469387754, "eval_LOCATION_number": 94, "eval_LOCATION_precision": 0.8823529411764706, "eval_LOCATION_recall": 0.9574468085106383, "eval_ORGANIZATION_f1": 0.8814589665653495, "eval_ORGANIZATION_number": 167, "eval_ORGANIZATION_precision": 0.8950617283950617, "eval_ORGANIZATION_recall": 0.8682634730538922, "eval_PERSON_f1": 0.9745454545454545, "eval_PERSON_number": 137, "eval_PERSON_precision": 0.9710144927536232, "eval_PERSON_recall": 0.9781021897810219, "eval_loss": 0.052171189337968826, "eval_overall_accuracy": 0.9837016574585635, "eval_overall_f1": 0.9225, "eval_overall_precision": 0.917910447761194, "eval_overall_recall": 0.9271356783919598, "eval_runtime": 0.2472, "eval_samples_per_second": 687.807, "eval_steps_per_second": 12.138, "step": 4416 }, { "epoch": 47.0, "grad_norm": 0.38328567147254944, "learning_rate": 2.6500000000000004e-05, "loss": 0.0203, "step": 4512 }, { "epoch": 47.0, "eval_LOCATION_f1": 0.9278350515463918, "eval_LOCATION_number": 94, "eval_LOCATION_precision": 0.9, "eval_LOCATION_recall": 0.9574468085106383, "eval_ORGANIZATION_f1": 0.8948948948948949, "eval_ORGANIZATION_number": 167, "eval_ORGANIZATION_precision": 0.8975903614457831, "eval_ORGANIZATION_recall": 0.8922155688622755, "eval_PERSON_f1": 0.9781021897810219, "eval_PERSON_number": 137, "eval_PERSON_precision": 0.9781021897810219, "eval_PERSON_recall": 0.9781021897810219, "eval_loss": 0.04933099076151848, "eval_overall_accuracy": 0.9853591160220995, "eval_overall_f1": 0.9313358302122346, "eval_overall_precision": 0.9255583126550868, "eval_overall_recall": 0.9371859296482412, "eval_runtime": 0.2484, "eval_samples_per_second": 684.49, "eval_steps_per_second": 12.079, "step": 4512 }, { "epoch": 48.0, "grad_norm": 0.6866722106933594, "learning_rate": 2.6000000000000002e-05, "loss": 0.0198, "step": 4608 }, { "epoch": 48.0, "eval_LOCATION_f1": 0.9230769230769231, "eval_LOCATION_number": 94, "eval_LOCATION_precision": 0.8910891089108911, "eval_LOCATION_recall": 0.9574468085106383, "eval_ORGANIZATION_f1": 0.8814589665653495, "eval_ORGANIZATION_number": 167, "eval_ORGANIZATION_precision": 0.8950617283950617, "eval_ORGANIZATION_recall": 0.8682634730538922, "eval_PERSON_f1": 0.9745454545454545, "eval_PERSON_number": 137, "eval_PERSON_precision": 0.9710144927536232, "eval_PERSON_recall": 0.9781021897810219, "eval_loss": 0.05320937931537628, "eval_overall_accuracy": 0.9845303867403314, "eval_overall_f1": 0.9236545682102627, "eval_overall_precision": 0.9201995012468828, "eval_overall_recall": 0.9271356783919598, "eval_runtime": 0.2474, "eval_samples_per_second": 687.083, "eval_steps_per_second": 12.125, "step": 4608 }, { "epoch": 49.0, "grad_norm": 0.38931208848953247, "learning_rate": 2.5500000000000003e-05, "loss": 0.019, "step": 4704 }, { "epoch": 49.0, "eval_LOCATION_f1": 0.9035532994923857, "eval_LOCATION_number": 94, "eval_LOCATION_precision": 0.8640776699029126, "eval_LOCATION_recall": 0.9468085106382979, "eval_ORGANIZATION_f1": 0.8768768768768769, "eval_ORGANIZATION_number": 167, "eval_ORGANIZATION_precision": 0.8795180722891566, "eval_ORGANIZATION_recall": 0.874251497005988, "eval_PERSON_f1": 0.9854014598540146, "eval_PERSON_number": 137, "eval_PERSON_precision": 0.9854014598540146, "eval_PERSON_recall": 0.9854014598540146, "eval_loss": 0.04946500062942505, "eval_overall_accuracy": 0.9839779005524862, "eval_overall_f1": 0.9203980099502488, "eval_overall_precision": 0.9113300492610837, "eval_overall_recall": 0.9296482412060302, "eval_runtime": 0.2472, "eval_samples_per_second": 687.752, "eval_steps_per_second": 12.137, "step": 4704 }, { "epoch": 50.0, "grad_norm": 1.0583337545394897, "learning_rate": 2.5e-05, "loss": 0.0188, "step": 4800 }, { "epoch": 50.0, "eval_LOCATION_f1": 0.9090909090909091, "eval_LOCATION_number": 94, "eval_LOCATION_precision": 0.8653846153846154, "eval_LOCATION_recall": 0.9574468085106383, "eval_ORGANIZATION_f1": 0.8708708708708709, "eval_ORGANIZATION_number": 167, "eval_ORGANIZATION_precision": 0.8734939759036144, "eval_ORGANIZATION_recall": 0.8682634730538922, "eval_PERSON_f1": 0.9854014598540146, "eval_PERSON_number": 137, "eval_PERSON_precision": 0.9854014598540146, "eval_PERSON_recall": 0.9854014598540146, "eval_loss": 0.05096418038010597, "eval_overall_accuracy": 0.9839779005524862, "eval_overall_f1": 0.9192546583850932, "eval_overall_precision": 0.9090909090909091, "eval_overall_recall": 0.9296482412060302, "eval_runtime": 0.2472, "eval_samples_per_second": 687.719, "eval_steps_per_second": 12.136, "step": 4800 }, { "epoch": 51.0, "grad_norm": 0.035651467740535736, "learning_rate": 2.45e-05, "loss": 0.0174, "step": 4896 }, { "epoch": 51.0, "eval_LOCATION_f1": 0.9183673469387754, "eval_LOCATION_number": 94, "eval_LOCATION_precision": 0.8823529411764706, "eval_LOCATION_recall": 0.9574468085106383, "eval_ORGANIZATION_f1": 0.8753799392097263, "eval_ORGANIZATION_number": 167, "eval_ORGANIZATION_precision": 0.8888888888888888, "eval_ORGANIZATION_recall": 0.8622754491017964, "eval_PERSON_f1": 0.9854014598540146, "eval_PERSON_number": 137, "eval_PERSON_precision": 0.9854014598540146, "eval_PERSON_recall": 0.9854014598540146, "eval_loss": 0.054159507155418396, "eval_overall_accuracy": 0.9842541436464088, "eval_overall_f1": 0.9236545682102627, "eval_overall_precision": 0.9201995012468828, "eval_overall_recall": 0.9271356783919598, "eval_runtime": 0.2461, "eval_samples_per_second": 690.695, "eval_steps_per_second": 12.189, "step": 4896 }, { "epoch": 52.0, "grad_norm": 0.5753004550933838, "learning_rate": 2.4e-05, "loss": 0.0176, "step": 4992 }, { "epoch": 52.0, "eval_LOCATION_f1": 0.9230769230769231, "eval_LOCATION_number": 94, "eval_LOCATION_precision": 0.8910891089108911, "eval_LOCATION_recall": 0.9574468085106383, "eval_ORGANIZATION_f1": 0.8948948948948949, "eval_ORGANIZATION_number": 167, "eval_ORGANIZATION_precision": 0.8975903614457831, "eval_ORGANIZATION_recall": 0.8922155688622755, "eval_PERSON_f1": 0.9854014598540146, "eval_PERSON_number": 137, "eval_PERSON_precision": 0.9854014598540146, "eval_PERSON_recall": 0.9854014598540146, "eval_loss": 0.05155513808131218, "eval_overall_accuracy": 0.9848066298342542, "eval_overall_f1": 0.9326683291770573, "eval_overall_precision": 0.9257425742574258, "eval_overall_recall": 0.9396984924623115, "eval_runtime": 0.2459, "eval_samples_per_second": 691.228, "eval_steps_per_second": 12.198, "step": 4992 }, { "epoch": 53.0, "grad_norm": 0.6415030360221863, "learning_rate": 2.35e-05, "loss": 0.0177, "step": 5088 }, { "epoch": 53.0, "eval_LOCATION_f1": 0.9137055837563451, "eval_LOCATION_number": 94, "eval_LOCATION_precision": 0.8737864077669902, "eval_LOCATION_recall": 0.9574468085106383, "eval_ORGANIZATION_f1": 0.8734939759036144, "eval_ORGANIZATION_number": 167, "eval_ORGANIZATION_precision": 0.8787878787878788, "eval_ORGANIZATION_recall": 0.8682634730538922, "eval_PERSON_f1": 0.9854014598540146, "eval_PERSON_number": 137, "eval_PERSON_precision": 0.9854014598540146, "eval_PERSON_recall": 0.9854014598540146, "eval_loss": 0.05428411439061165, "eval_overall_accuracy": 0.9839779005524862, "eval_overall_f1": 0.9215442092154421, "eval_overall_precision": 0.9135802469135802, "eval_overall_recall": 0.9296482412060302, "eval_runtime": 0.2456, "eval_samples_per_second": 692.316, "eval_steps_per_second": 12.217, "step": 5088 }, { "epoch": 54.0, "grad_norm": 0.752815306186676, "learning_rate": 2.3000000000000003e-05, "loss": 0.0177, "step": 5184 }, { "epoch": 54.0, "eval_LOCATION_f1": 0.9183673469387754, "eval_LOCATION_number": 94, "eval_LOCATION_precision": 0.8823529411764706, "eval_LOCATION_recall": 0.9574468085106383, "eval_ORGANIZATION_f1": 0.8749999999999999, "eval_ORGANIZATION_number": 167, "eval_ORGANIZATION_precision": 0.8698224852071006, "eval_ORGANIZATION_recall": 0.8802395209580839, "eval_PERSON_f1": 0.9854014598540146, "eval_PERSON_number": 137, "eval_PERSON_precision": 0.9854014598540146, "eval_PERSON_recall": 0.9854014598540146, "eval_loss": 0.050988342612981796, "eval_overall_accuracy": 0.9839779005524862, "eval_overall_f1": 0.9230769230769231, "eval_overall_precision": 0.9117647058823529, "eval_overall_recall": 0.9346733668341709, "eval_runtime": 0.2459, "eval_samples_per_second": 691.254, "eval_steps_per_second": 12.199, "step": 5184 }, { "epoch": 55.0, "grad_norm": 0.6094796061515808, "learning_rate": 2.25e-05, "loss": 0.0159, "step": 5280 }, { "epoch": 55.0, "eval_LOCATION_f1": 0.9045226130653266, "eval_LOCATION_number": 94, "eval_LOCATION_precision": 0.8571428571428571, "eval_LOCATION_recall": 0.9574468085106383, "eval_ORGANIZATION_f1": 0.8795180722891567, "eval_ORGANIZATION_number": 167, "eval_ORGANIZATION_precision": 0.8848484848484849, "eval_ORGANIZATION_recall": 0.874251497005988, "eval_PERSON_f1": 0.9854014598540146, "eval_PERSON_number": 137, "eval_PERSON_precision": 0.9854014598540146, "eval_PERSON_recall": 0.9854014598540146, "eval_loss": 0.05016588419675827, "eval_overall_accuracy": 0.9837016574585635, "eval_overall_f1": 0.9217391304347826, "eval_overall_precision": 0.9115479115479116, "eval_overall_recall": 0.9321608040201005, "eval_runtime": 0.2472, "eval_samples_per_second": 687.67, "eval_steps_per_second": 12.135, "step": 5280 }, { "epoch": 56.0, "grad_norm": 0.8017212152481079, "learning_rate": 2.2000000000000003e-05, "loss": 0.0176, "step": 5376 }, { "epoch": 56.0, "eval_LOCATION_f1": 0.9278350515463918, "eval_LOCATION_number": 94, "eval_LOCATION_precision": 0.9, "eval_LOCATION_recall": 0.9574468085106383, "eval_ORGANIZATION_f1": 0.8875379939209727, "eval_ORGANIZATION_number": 167, "eval_ORGANIZATION_precision": 0.9012345679012346, "eval_ORGANIZATION_recall": 0.874251497005988, "eval_PERSON_f1": 0.9854014598540146, "eval_PERSON_number": 137, "eval_PERSON_precision": 0.9854014598540146, "eval_PERSON_recall": 0.9854014598540146, "eval_loss": 0.05175416171550751, "eval_overall_accuracy": 0.9853591160220995, "eval_overall_f1": 0.9309912170639899, "eval_overall_precision": 0.9298245614035088, "eval_overall_recall": 0.9321608040201005, "eval_runtime": 0.2456, "eval_samples_per_second": 692.132, "eval_steps_per_second": 12.214, "step": 5376 }, { "epoch": 57.0, "grad_norm": 0.37985149025917053, "learning_rate": 2.15e-05, "loss": 0.016, "step": 5472 }, { "epoch": 57.0, "eval_LOCATION_f1": 0.8823529411764706, "eval_LOCATION_number": 94, "eval_LOCATION_precision": 0.8181818181818182, "eval_LOCATION_recall": 0.9574468085106383, "eval_ORGANIZATION_f1": 0.8553846153846154, "eval_ORGANIZATION_number": 167, "eval_ORGANIZATION_precision": 0.879746835443038, "eval_ORGANIZATION_recall": 0.8323353293413174, "eval_PERSON_f1": 0.9854014598540146, "eval_PERSON_number": 137, "eval_PERSON_precision": 0.9854014598540146, "eval_PERSON_recall": 0.9854014598540146, "eval_loss": 0.05611170455813408, "eval_overall_accuracy": 0.9823204419889503, "eval_overall_f1": 0.9066002490660026, "eval_overall_precision": 0.8987654320987655, "eval_overall_recall": 0.914572864321608, "eval_runtime": 0.2484, "eval_samples_per_second": 684.254, "eval_steps_per_second": 12.075, "step": 5472 }, { "epoch": 58.0, "grad_norm": 0.21325336396694183, "learning_rate": 2.1e-05, "loss": 0.0149, "step": 5568 }, { "epoch": 58.0, "eval_LOCATION_f1": 0.9137055837563451, "eval_LOCATION_number": 94, "eval_LOCATION_precision": 0.8737864077669902, "eval_LOCATION_recall": 0.9574468085106383, "eval_ORGANIZATION_f1": 0.8795180722891567, "eval_ORGANIZATION_number": 167, "eval_ORGANIZATION_precision": 0.8848484848484849, "eval_ORGANIZATION_recall": 0.874251497005988, "eval_PERSON_f1": 0.9854014598540146, "eval_PERSON_number": 137, "eval_PERSON_precision": 0.9854014598540146, "eval_PERSON_recall": 0.9854014598540146, "eval_loss": 0.05132393538951874, "eval_overall_accuracy": 0.9845303867403314, "eval_overall_f1": 0.9240348692403487, "eval_overall_precision": 0.9160493827160494, "eval_overall_recall": 0.9321608040201005, "eval_runtime": 0.2469, "eval_samples_per_second": 688.436, "eval_steps_per_second": 12.149, "step": 5568 }, { "epoch": 59.0, "grad_norm": 0.6967980861663818, "learning_rate": 2.05e-05, "loss": 0.0154, "step": 5664 }, { "epoch": 59.0, "eval_LOCATION_f1": 0.9183673469387754, "eval_LOCATION_number": 94, "eval_LOCATION_precision": 0.8823529411764706, "eval_LOCATION_recall": 0.9574468085106383, "eval_ORGANIZATION_f1": 0.8848484848484848, "eval_ORGANIZATION_number": 167, "eval_ORGANIZATION_precision": 0.8957055214723927, "eval_ORGANIZATION_recall": 0.874251497005988, "eval_PERSON_f1": 0.9854014598540146, "eval_PERSON_number": 137, "eval_PERSON_precision": 0.9854014598540146, "eval_PERSON_recall": 0.9854014598540146, "eval_loss": 0.05428219959139824, "eval_overall_accuracy": 0.9848066298342542, "eval_overall_f1": 0.9275, "eval_overall_precision": 0.9228855721393034, "eval_overall_recall": 0.9321608040201005, "eval_runtime": 0.2463, "eval_samples_per_second": 690.279, "eval_steps_per_second": 12.181, "step": 5664 }, { "epoch": 60.0, "grad_norm": 1.6469084024429321, "learning_rate": 2e-05, "loss": 0.0156, "step": 5760 }, { "epoch": 60.0, "eval_LOCATION_f1": 0.8955223880597015, "eval_LOCATION_number": 94, "eval_LOCATION_precision": 0.8411214953271028, "eval_LOCATION_recall": 0.9574468085106383, "eval_ORGANIZATION_f1": 0.8615384615384616, "eval_ORGANIZATION_number": 167, "eval_ORGANIZATION_precision": 0.8860759493670886, "eval_ORGANIZATION_recall": 0.8383233532934131, "eval_PERSON_f1": 0.9854014598540146, "eval_PERSON_number": 137, "eval_PERSON_precision": 0.9854014598540146, "eval_PERSON_recall": 0.9854014598540146, "eval_loss": 0.0560385137796402, "eval_overall_accuracy": 0.9828729281767956, "eval_overall_f1": 0.9125, "eval_overall_precision": 0.9079601990049752, "eval_overall_recall": 0.9170854271356784, "eval_runtime": 0.2456, "eval_samples_per_second": 692.273, "eval_steps_per_second": 12.217, "step": 5760 }, { "epoch": 61.0, "grad_norm": 0.47649747133255005, "learning_rate": 1.9500000000000003e-05, "loss": 0.0148, "step": 5856 }, { "epoch": 61.0, "eval_LOCATION_f1": 0.8811881188118812, "eval_LOCATION_number": 94, "eval_LOCATION_precision": 0.8240740740740741, "eval_LOCATION_recall": 0.9468085106382979, "eval_ORGANIZATION_f1": 0.8761329305135951, "eval_ORGANIZATION_number": 167, "eval_ORGANIZATION_precision": 0.8841463414634146, "eval_ORGANIZATION_recall": 0.8682634730538922, "eval_PERSON_f1": 0.9854014598540146, "eval_PERSON_number": 137, "eval_PERSON_precision": 0.9854014598540146, "eval_PERSON_recall": 0.9854014598540146, "eval_loss": 0.05628784000873566, "eval_overall_accuracy": 0.9823204419889503, "eval_overall_f1": 0.9144981412639406, "eval_overall_precision": 0.902200488997555, "eval_overall_recall": 0.9271356783919598, "eval_runtime": 0.2478, "eval_samples_per_second": 685.955, "eval_steps_per_second": 12.105, "step": 5856 }, { "epoch": 62.0, "grad_norm": 0.07543937116861343, "learning_rate": 1.9e-05, "loss": 0.0138, "step": 5952 }, { "epoch": 62.0, "eval_LOCATION_f1": 0.9081632653061226, "eval_LOCATION_number": 94, "eval_LOCATION_precision": 0.8725490196078431, "eval_LOCATION_recall": 0.9468085106382979, "eval_ORGANIZATION_f1": 0.8719512195121951, "eval_ORGANIZATION_number": 167, "eval_ORGANIZATION_precision": 0.8881987577639752, "eval_ORGANIZATION_recall": 0.8562874251497006, "eval_PERSON_f1": 0.9854014598540146, "eval_PERSON_number": 137, "eval_PERSON_precision": 0.9854014598540146, "eval_PERSON_recall": 0.9854014598540146, "eval_loss": 0.05514856055378914, "eval_overall_accuracy": 0.9842541436464088, "eval_overall_f1": 0.9197994987468672, "eval_overall_precision": 0.9175, "eval_overall_recall": 0.9221105527638191, "eval_runtime": 0.2466, "eval_samples_per_second": 689.323, "eval_steps_per_second": 12.165, "step": 5952 }, { "epoch": 63.0, "grad_norm": 0.818498432636261, "learning_rate": 1.85e-05, "loss": 0.014, "step": 6048 }, { "epoch": 63.0, "eval_LOCATION_f1": 0.898989898989899, "eval_LOCATION_number": 94, "eval_LOCATION_precision": 0.8557692307692307, "eval_LOCATION_recall": 0.9468085106382979, "eval_ORGANIZATION_f1": 0.8802395209580839, "eval_ORGANIZATION_number": 167, "eval_ORGANIZATION_precision": 0.8802395209580839, "eval_ORGANIZATION_recall": 0.8802395209580839, "eval_PERSON_f1": 0.9854014598540146, "eval_PERSON_number": 137, "eval_PERSON_precision": 0.9854014598540146, "eval_PERSON_recall": 0.9854014598540146, "eval_loss": 0.055294014513492584, "eval_overall_accuracy": 0.9834254143646409, "eval_overall_f1": 0.9205955334987593, "eval_overall_precision": 0.9093137254901961, "eval_overall_recall": 0.9321608040201005, "eval_runtime": 0.2466, "eval_samples_per_second": 689.473, "eval_steps_per_second": 12.167, "step": 6048 }, { "epoch": 64.0, "grad_norm": 0.07948024570941925, "learning_rate": 1.8e-05, "loss": 0.0133, "step": 6144 }, { "epoch": 64.0, "eval_LOCATION_f1": 0.8855721393034826, "eval_LOCATION_number": 94, "eval_LOCATION_precision": 0.8317757009345794, "eval_LOCATION_recall": 0.9468085106382979, "eval_ORGANIZATION_f1": 0.851063829787234, "eval_ORGANIZATION_number": 167, "eval_ORGANIZATION_precision": 0.8641975308641975, "eval_ORGANIZATION_recall": 0.8383233532934131, "eval_PERSON_f1": 0.9854014598540146, "eval_PERSON_number": 137, "eval_PERSON_precision": 0.9854014598540146, "eval_PERSON_recall": 0.9854014598540146, "eval_loss": 0.058841772377491, "eval_overall_accuracy": 0.981767955801105, "eval_overall_f1": 0.9054726368159204, "eval_overall_precision": 0.896551724137931, "eval_overall_recall": 0.914572864321608, "eval_runtime": 0.2469, "eval_samples_per_second": 688.627, "eval_steps_per_second": 12.152, "step": 6144 }, { "epoch": 65.0, "grad_norm": 0.42106878757476807, "learning_rate": 1.75e-05, "loss": 0.0132, "step": 6240 }, { "epoch": 65.0, "eval_LOCATION_f1": 0.8955223880597015, "eval_LOCATION_number": 94, "eval_LOCATION_precision": 0.8411214953271028, "eval_LOCATION_recall": 0.9574468085106383, "eval_ORGANIZATION_f1": 0.8685015290519879, "eval_ORGANIZATION_number": 167, "eval_ORGANIZATION_precision": 0.8875, "eval_ORGANIZATION_recall": 0.8502994011976048, "eval_PERSON_f1": 0.9854014598540146, "eval_PERSON_number": 137, "eval_PERSON_precision": 0.9854014598540146, "eval_PERSON_recall": 0.9854014598540146, "eval_loss": 0.05472525954246521, "eval_overall_accuracy": 0.9828729281767956, "eval_overall_f1": 0.915211970074813, "eval_overall_precision": 0.9084158415841584, "eval_overall_recall": 0.9221105527638191, "eval_runtime": 0.2462, "eval_samples_per_second": 690.363, "eval_steps_per_second": 12.183, "step": 6240 }, { "epoch": 66.0, "grad_norm": 0.7309462428092957, "learning_rate": 1.7000000000000003e-05, "loss": 0.0139, "step": 6336 }, { "epoch": 66.0, "eval_LOCATION_f1": 0.898989898989899, "eval_LOCATION_number": 94, "eval_LOCATION_precision": 0.8557692307692307, "eval_LOCATION_recall": 0.9468085106382979, "eval_ORGANIZATION_f1": 0.8597560975609756, "eval_ORGANIZATION_number": 167, "eval_ORGANIZATION_precision": 0.8757763975155279, "eval_ORGANIZATION_recall": 0.844311377245509, "eval_PERSON_f1": 0.989010989010989, "eval_PERSON_number": 137, "eval_PERSON_precision": 0.9926470588235294, "eval_PERSON_recall": 0.9854014598540146, "eval_loss": 0.05803309753537178, "eval_overall_accuracy": 0.9831491712707182, "eval_overall_f1": 0.9136420525657072, "eval_overall_precision": 0.9102244389027432, "eval_overall_recall": 0.9170854271356784, "eval_runtime": 0.2458, "eval_samples_per_second": 691.602, "eval_steps_per_second": 12.205, "step": 6336 }, { "epoch": 67.0, "grad_norm": 0.0585617758333683, "learning_rate": 1.65e-05, "loss": 0.0127, "step": 6432 }, { "epoch": 67.0, "eval_LOCATION_f1": 0.9081632653061226, "eval_LOCATION_number": 94, "eval_LOCATION_precision": 0.8725490196078431, "eval_LOCATION_recall": 0.9468085106382979, "eval_ORGANIZATION_f1": 0.874251497005988, "eval_ORGANIZATION_number": 167, "eval_ORGANIZATION_precision": 0.874251497005988, "eval_ORGANIZATION_recall": 0.874251497005988, "eval_PERSON_f1": 0.9854014598540146, "eval_PERSON_number": 137, "eval_PERSON_precision": 0.9854014598540146, "eval_PERSON_recall": 0.9854014598540146, "eval_loss": 0.05596785247325897, "eval_overall_accuracy": 0.9837016574585635, "eval_overall_f1": 0.9203980099502488, "eval_overall_precision": 0.9113300492610837, "eval_overall_recall": 0.9296482412060302, "eval_runtime": 0.2473, "eval_samples_per_second": 687.434, "eval_steps_per_second": 12.131, "step": 6432 }, { "epoch": 68.0, "grad_norm": 0.20063190162181854, "learning_rate": 1.6000000000000003e-05, "loss": 0.0124, "step": 6528 }, { "epoch": 68.0, "eval_LOCATION_f1": 0.898989898989899, "eval_LOCATION_number": 94, "eval_LOCATION_precision": 0.8557692307692307, "eval_LOCATION_recall": 0.9468085106382979, "eval_ORGANIZATION_f1": 0.8761329305135951, "eval_ORGANIZATION_number": 167, "eval_ORGANIZATION_precision": 0.8841463414634146, "eval_ORGANIZATION_recall": 0.8682634730538922, "eval_PERSON_f1": 0.989010989010989, "eval_PERSON_number": 137, "eval_PERSON_precision": 0.9926470588235294, "eval_PERSON_recall": 0.9854014598540146, "eval_loss": 0.05695489048957825, "eval_overall_accuracy": 0.9839779005524862, "eval_overall_f1": 0.9201995012468827, "eval_overall_precision": 0.9133663366336634, "eval_overall_recall": 0.9271356783919598, "eval_runtime": 0.2473, "eval_samples_per_second": 687.428, "eval_steps_per_second": 12.131, "step": 6528 }, { "epoch": 69.0, "grad_norm": 0.7651628255844116, "learning_rate": 1.55e-05, "loss": 0.0132, "step": 6624 }, { "epoch": 69.0, "eval_LOCATION_f1": 0.8944723618090452, "eval_LOCATION_number": 94, "eval_LOCATION_precision": 0.8476190476190476, "eval_LOCATION_recall": 0.9468085106382979, "eval_ORGANIZATION_f1": 0.896969696969697, "eval_ORGANIZATION_number": 167, "eval_ORGANIZATION_precision": 0.9079754601226994, "eval_ORGANIZATION_recall": 0.8862275449101796, "eval_PERSON_f1": 0.9854014598540146, "eval_PERSON_number": 137, "eval_PERSON_precision": 0.9854014598540146, "eval_PERSON_recall": 0.9854014598540146, "eval_loss": 0.05245853215456009, "eval_overall_accuracy": 0.9845303867403314, "eval_overall_f1": 0.9265255292652553, "eval_overall_precision": 0.9185185185185185, "eval_overall_recall": 0.9346733668341709, "eval_runtime": 0.2469, "eval_samples_per_second": 688.492, "eval_steps_per_second": 12.15, "step": 6624 }, { "epoch": 70.0, "grad_norm": 0.047926392406225204, "learning_rate": 1.5e-05, "loss": 0.0131, "step": 6720 }, { "epoch": 70.0, "eval_LOCATION_f1": 0.9128205128205128, "eval_LOCATION_number": 94, "eval_LOCATION_precision": 0.8811881188118812, "eval_LOCATION_recall": 0.9468085106382979, "eval_ORGANIZATION_f1": 0.8821752265861027, "eval_ORGANIZATION_number": 167, "eval_ORGANIZATION_precision": 0.8902439024390244, "eval_ORGANIZATION_recall": 0.874251497005988, "eval_PERSON_f1": 0.9854014598540146, "eval_PERSON_number": 137, "eval_PERSON_precision": 0.9854014598540146, "eval_PERSON_recall": 0.9854014598540146, "eval_loss": 0.05109546333551407, "eval_overall_accuracy": 0.9853591160220995, "eval_overall_f1": 0.925, "eval_overall_precision": 0.9203980099502488, "eval_overall_recall": 0.9296482412060302, "eval_runtime": 0.2468, "eval_samples_per_second": 688.917, "eval_steps_per_second": 12.157, "step": 6720 }, { "epoch": 71.0, "grad_norm": 0.45357948541641235, "learning_rate": 1.45e-05, "loss": 0.0125, "step": 6816 }, { "epoch": 71.0, "eval_LOCATION_f1": 0.9175257731958764, "eval_LOCATION_number": 94, "eval_LOCATION_precision": 0.89, "eval_LOCATION_recall": 0.9468085106382979, "eval_ORGANIZATION_f1": 0.8975903614457832, "eval_ORGANIZATION_number": 167, "eval_ORGANIZATION_precision": 0.9030303030303031, "eval_ORGANIZATION_recall": 0.8922155688622755, "eval_PERSON_f1": 0.9854014598540146, "eval_PERSON_number": 137, "eval_PERSON_precision": 0.9854014598540146, "eval_PERSON_recall": 0.9854014598540146, "eval_loss": 0.05031238868832588, "eval_overall_accuracy": 0.9859116022099448, "eval_overall_f1": 0.9325, "eval_overall_precision": 0.927860696517413, "eval_overall_recall": 0.9371859296482412, "eval_runtime": 0.2464, "eval_samples_per_second": 690.074, "eval_steps_per_second": 12.178, "step": 6816 }, { "epoch": 72.0, "grad_norm": 0.899650514125824, "learning_rate": 1.4000000000000001e-05, "loss": 0.014, "step": 6912 }, { "epoch": 72.0, "eval_LOCATION_f1": 0.898989898989899, "eval_LOCATION_number": 94, "eval_LOCATION_precision": 0.8557692307692307, "eval_LOCATION_recall": 0.9468085106382979, "eval_ORGANIZATION_f1": 0.8848484848484848, "eval_ORGANIZATION_number": 167, "eval_ORGANIZATION_precision": 0.8957055214723927, "eval_ORGANIZATION_recall": 0.874251497005988, "eval_PERSON_f1": 0.989010989010989, "eval_PERSON_number": 137, "eval_PERSON_precision": 0.9926470588235294, "eval_PERSON_recall": 0.9854014598540146, "eval_loss": 0.052724651992321014, "eval_overall_accuracy": 0.9848066298342542, "eval_overall_f1": 0.9238451935081149, "eval_overall_precision": 0.9181141439205955, "eval_overall_recall": 0.9296482412060302, "eval_runtime": 0.2468, "eval_samples_per_second": 688.791, "eval_steps_per_second": 12.155, "step": 6912 }, { "epoch": 73.0, "grad_norm": 0.8117560148239136, "learning_rate": 1.3500000000000001e-05, "loss": 0.0123, "step": 7008 }, { "epoch": 73.0, "eval_LOCATION_f1": 0.9081632653061226, "eval_LOCATION_number": 94, "eval_LOCATION_precision": 0.8725490196078431, "eval_LOCATION_recall": 0.9468085106382979, "eval_ORGANIZATION_f1": 0.8700906344410877, "eval_ORGANIZATION_number": 167, "eval_ORGANIZATION_precision": 0.8780487804878049, "eval_ORGANIZATION_recall": 0.8622754491017964, "eval_PERSON_f1": 0.9854014598540146, "eval_PERSON_number": 137, "eval_PERSON_precision": 0.9854014598540146, "eval_PERSON_recall": 0.9854014598540146, "eval_loss": 0.05363898724317551, "eval_overall_accuracy": 0.9839779005524862, "eval_overall_f1": 0.9188514357053683, "eval_overall_precision": 0.913151364764268, "eval_overall_recall": 0.9246231155778895, "eval_runtime": 0.2416, "eval_samples_per_second": 703.667, "eval_steps_per_second": 12.418, "step": 7008 }, { "epoch": 74.0, "grad_norm": 0.7420791387557983, "learning_rate": 1.3000000000000001e-05, "loss": 0.0121, "step": 7104 }, { "epoch": 74.0, "eval_LOCATION_f1": 0.9222797927461138, "eval_LOCATION_number": 94, "eval_LOCATION_precision": 0.898989898989899, "eval_LOCATION_recall": 0.9468085106382979, "eval_ORGANIZATION_f1": 0.8821752265861027, "eval_ORGANIZATION_number": 167, "eval_ORGANIZATION_precision": 0.8902439024390244, "eval_ORGANIZATION_recall": 0.874251497005988, "eval_PERSON_f1": 0.9854014598540146, "eval_PERSON_number": 137, "eval_PERSON_precision": 0.9854014598540146, "eval_PERSON_recall": 0.9854014598540146, "eval_loss": 0.05422784760594368, "eval_overall_accuracy": 0.9842541436464088, "eval_overall_f1": 0.9273182957393483, "eval_overall_precision": 0.925, "eval_overall_recall": 0.9296482412060302, "eval_runtime": 0.2463, "eval_samples_per_second": 690.136, "eval_steps_per_second": 12.179, "step": 7104 }, { "epoch": 75.0, "grad_norm": 0.7961884140968323, "learning_rate": 1.25e-05, "loss": 0.0115, "step": 7200 }, { "epoch": 75.0, "eval_LOCATION_f1": 0.9230769230769231, "eval_LOCATION_number": 94, "eval_LOCATION_precision": 0.8910891089108911, "eval_LOCATION_recall": 0.9574468085106383, "eval_ORGANIZATION_f1": 0.8936170212765959, "eval_ORGANIZATION_number": 167, "eval_ORGANIZATION_precision": 0.9074074074074074, "eval_ORGANIZATION_recall": 0.8802395209580839, "eval_PERSON_f1": 0.9854014598540146, "eval_PERSON_number": 137, "eval_PERSON_precision": 0.9854014598540146, "eval_PERSON_recall": 0.9854014598540146, "eval_loss": 0.05263187363743782, "eval_overall_accuracy": 0.9850828729281768, "eval_overall_f1": 0.9323308270676693, "eval_overall_precision": 0.93, "eval_overall_recall": 0.9346733668341709, "eval_runtime": 0.2479, "eval_samples_per_second": 685.787, "eval_steps_per_second": 12.102, "step": 7200 }, { "epoch": 76.0, "grad_norm": 0.4262893795967102, "learning_rate": 1.2e-05, "loss": 0.0115, "step": 7296 }, { "epoch": 76.0, "eval_LOCATION_f1": 0.9175257731958764, "eval_LOCATION_number": 94, "eval_LOCATION_precision": 0.89, "eval_LOCATION_recall": 0.9468085106382979, "eval_ORGANIZATION_f1": 0.8882175226586103, "eval_ORGANIZATION_number": 167, "eval_ORGANIZATION_precision": 0.8963414634146342, "eval_ORGANIZATION_recall": 0.8802395209580839, "eval_PERSON_f1": 0.9854014598540146, "eval_PERSON_number": 137, "eval_PERSON_precision": 0.9854014598540146, "eval_PERSON_recall": 0.9854014598540146, "eval_loss": 0.05334838479757309, "eval_overall_accuracy": 0.9842541436464088, "eval_overall_f1": 0.9286608260325407, "eval_overall_precision": 0.9251870324189526, "eval_overall_recall": 0.9321608040201005, "eval_runtime": 0.2398, "eval_samples_per_second": 709.06, "eval_steps_per_second": 12.513, "step": 7296 }, { "epoch": 77.0, "grad_norm": 0.271272212266922, "learning_rate": 1.1500000000000002e-05, "loss": 0.0129, "step": 7392 }, { "epoch": 77.0, "eval_LOCATION_f1": 0.8944723618090452, "eval_LOCATION_number": 94, "eval_LOCATION_precision": 0.8476190476190476, "eval_LOCATION_recall": 0.9468085106382979, "eval_ORGANIZATION_f1": 0.8746177370030581, "eval_ORGANIZATION_number": 167, "eval_ORGANIZATION_precision": 0.89375, "eval_ORGANIZATION_recall": 0.8562874251497006, "eval_PERSON_f1": 0.9854014598540146, "eval_PERSON_number": 137, "eval_PERSON_precision": 0.9854014598540146, "eval_PERSON_recall": 0.9854014598540146, "eval_loss": 0.05538813769817352, "eval_overall_accuracy": 0.9831491712707182, "eval_overall_f1": 0.9175, "eval_overall_precision": 0.9129353233830846, "eval_overall_recall": 0.9221105527638191, "eval_runtime": 0.2465, "eval_samples_per_second": 689.528, "eval_steps_per_second": 12.168, "step": 7392 }, { "epoch": 78.0, "grad_norm": 0.8887650370597839, "learning_rate": 1.1000000000000001e-05, "loss": 0.0117, "step": 7488 }, { "epoch": 78.0, "eval_LOCATION_f1": 0.9035532994923857, "eval_LOCATION_number": 94, "eval_LOCATION_precision": 0.8640776699029126, "eval_LOCATION_recall": 0.9468085106382979, "eval_ORGANIZATION_f1": 0.888888888888889, "eval_ORGANIZATION_number": 167, "eval_ORGANIZATION_precision": 0.891566265060241, "eval_ORGANIZATION_recall": 0.8862275449101796, "eval_PERSON_f1": 0.9854014598540146, "eval_PERSON_number": 137, "eval_PERSON_precision": 0.9854014598540146, "eval_PERSON_recall": 0.9854014598540146, "eval_loss": 0.054783519357442856, "eval_overall_accuracy": 0.9842541436464088, "eval_overall_f1": 0.9253731343283583, "eval_overall_precision": 0.916256157635468, "eval_overall_recall": 0.9346733668341709, "eval_runtime": 0.2466, "eval_samples_per_second": 689.343, "eval_steps_per_second": 12.165, "step": 7488 }, { "epoch": 79.0, "grad_norm": 0.4379172623157501, "learning_rate": 1.05e-05, "loss": 0.0107, "step": 7584 }, { "epoch": 79.0, "eval_LOCATION_f1": 0.9222797927461138, "eval_LOCATION_number": 94, "eval_LOCATION_precision": 0.898989898989899, "eval_LOCATION_recall": 0.9468085106382979, "eval_ORGANIZATION_f1": 0.891566265060241, "eval_ORGANIZATION_number": 167, "eval_ORGANIZATION_precision": 0.896969696969697, "eval_ORGANIZATION_recall": 0.8862275449101796, "eval_PERSON_f1": 0.9854014598540146, "eval_PERSON_number": 137, "eval_PERSON_precision": 0.9854014598540146, "eval_PERSON_recall": 0.9854014598540146, "eval_loss": 0.055064547806978226, "eval_overall_accuracy": 0.9848066298342542, "eval_overall_f1": 0.9311639549436797, "eval_overall_precision": 0.9276807980049875, "eval_overall_recall": 0.9346733668341709, "eval_runtime": 0.2474, "eval_samples_per_second": 687.222, "eval_steps_per_second": 12.127, "step": 7584 }, { "epoch": 80.0, "grad_norm": 0.9737386703491211, "learning_rate": 1e-05, "loss": 0.012, "step": 7680 }, { "epoch": 80.0, "eval_LOCATION_f1": 0.9035532994923857, "eval_LOCATION_number": 94, "eval_LOCATION_precision": 0.8640776699029126, "eval_LOCATION_recall": 0.9468085106382979, "eval_ORGANIZATION_f1": 0.880952380952381, "eval_ORGANIZATION_number": 167, "eval_ORGANIZATION_precision": 0.8757396449704142, "eval_ORGANIZATION_recall": 0.8862275449101796, "eval_PERSON_f1": 0.9854014598540146, "eval_PERSON_number": 137, "eval_PERSON_precision": 0.9854014598540146, "eval_PERSON_recall": 0.9854014598540146, "eval_loss": 0.05387932434678078, "eval_overall_accuracy": 0.9834254143646409, "eval_overall_f1": 0.9219330855018587, "eval_overall_precision": 0.9095354523227384, "eval_overall_recall": 0.9346733668341709, "eval_runtime": 0.246, "eval_samples_per_second": 690.926, "eval_steps_per_second": 12.193, "step": 7680 }, { "epoch": 81.0, "grad_norm": 0.8605514168739319, "learning_rate": 9.5e-06, "loss": 0.0107, "step": 7776 }, { "epoch": 81.0, "eval_LOCATION_f1": 0.898989898989899, "eval_LOCATION_number": 94, "eval_LOCATION_precision": 0.8557692307692307, "eval_LOCATION_recall": 0.9468085106382979, "eval_ORGANIZATION_f1": 0.8711656441717791, "eval_ORGANIZATION_number": 167, "eval_ORGANIZATION_precision": 0.8930817610062893, "eval_ORGANIZATION_recall": 0.8502994011976048, "eval_PERSON_f1": 0.9854014598540146, "eval_PERSON_number": 137, "eval_PERSON_precision": 0.9854014598540146, "eval_PERSON_recall": 0.9854014598540146, "eval_loss": 0.055788103491067886, "eval_overall_accuracy": 0.9837016574585635, "eval_overall_f1": 0.9172932330827068, "eval_overall_precision": 0.915, "eval_overall_recall": 0.9195979899497487, "eval_runtime": 0.2462, "eval_samples_per_second": 690.475, "eval_steps_per_second": 12.185, "step": 7776 }, { "epoch": 82.0, "grad_norm": 0.13176043331623077, "learning_rate": 9e-06, "loss": 0.0107, "step": 7872 }, { "epoch": 82.0, "eval_LOCATION_f1": 0.9035532994923857, "eval_LOCATION_number": 94, "eval_LOCATION_precision": 0.8640776699029126, "eval_LOCATION_recall": 0.9468085106382979, "eval_ORGANIZATION_f1": 0.8693009118541033, "eval_ORGANIZATION_number": 167, "eval_ORGANIZATION_precision": 0.8827160493827161, "eval_ORGANIZATION_recall": 0.8562874251497006, "eval_PERSON_f1": 0.9854014598540146, "eval_PERSON_number": 137, "eval_PERSON_precision": 0.9854014598540146, "eval_PERSON_recall": 0.9854014598540146, "eval_loss": 0.05625952407717705, "eval_overall_accuracy": 0.9834254143646409, "eval_overall_f1": 0.9175, "eval_overall_precision": 0.9129353233830846, "eval_overall_recall": 0.9221105527638191, "eval_runtime": 0.2474, "eval_samples_per_second": 687.097, "eval_steps_per_second": 12.125, "step": 7872 }, { "epoch": 83.0, "grad_norm": 0.5827556848526001, "learning_rate": 8.500000000000002e-06, "loss": 0.0102, "step": 7968 }, { "epoch": 83.0, "eval_LOCATION_f1": 0.9081632653061226, "eval_LOCATION_number": 94, "eval_LOCATION_precision": 0.8725490196078431, "eval_LOCATION_recall": 0.9468085106382979, "eval_ORGANIZATION_f1": 0.8780487804878049, "eval_ORGANIZATION_number": 167, "eval_ORGANIZATION_precision": 0.8944099378881988, "eval_ORGANIZATION_recall": 0.8622754491017964, "eval_PERSON_f1": 0.9854014598540146, "eval_PERSON_number": 137, "eval_PERSON_precision": 0.9854014598540146, "eval_PERSON_recall": 0.9854014598540146, "eval_loss": 0.056949179619550705, "eval_overall_accuracy": 0.9839779005524862, "eval_overall_f1": 0.9223057644110277, "eval_overall_precision": 0.92, "eval_overall_recall": 0.9246231155778895, "eval_runtime": 0.2468, "eval_samples_per_second": 688.934, "eval_steps_per_second": 12.158, "step": 7968 }, { "epoch": 84.0, "grad_norm": 0.29158249497413635, "learning_rate": 8.000000000000001e-06, "loss": 0.0103, "step": 8064 }, { "epoch": 84.0, "eval_LOCATION_f1": 0.9175257731958764, "eval_LOCATION_number": 94, "eval_LOCATION_precision": 0.89, "eval_LOCATION_recall": 0.9468085106382979, "eval_ORGANIZATION_f1": 0.8753799392097263, "eval_ORGANIZATION_number": 167, "eval_ORGANIZATION_precision": 0.8888888888888888, "eval_ORGANIZATION_recall": 0.8622754491017964, "eval_PERSON_f1": 0.9854014598540146, "eval_PERSON_number": 137, "eval_PERSON_precision": 0.9854014598540146, "eval_PERSON_recall": 0.9854014598540146, "eval_loss": 0.05674479901790619, "eval_overall_accuracy": 0.9839779005524862, "eval_overall_f1": 0.9234629861982434, "eval_overall_precision": 0.9223057644110275, "eval_overall_recall": 0.9246231155778895, "eval_runtime": 0.2464, "eval_samples_per_second": 689.801, "eval_steps_per_second": 12.173, "step": 8064 }, { "epoch": 85.0, "grad_norm": 0.1342383176088333, "learning_rate": 7.5e-06, "loss": 0.0105, "step": 8160 }, { "epoch": 85.0, "eval_LOCATION_f1": 0.9128205128205128, "eval_LOCATION_number": 94, "eval_LOCATION_precision": 0.8811881188118812, "eval_LOCATION_recall": 0.9468085106382979, "eval_ORGANIZATION_f1": 0.8719512195121951, "eval_ORGANIZATION_number": 167, "eval_ORGANIZATION_precision": 0.8881987577639752, "eval_ORGANIZATION_recall": 0.8562874251497006, "eval_PERSON_f1": 0.989010989010989, "eval_PERSON_number": 137, "eval_PERSON_precision": 0.9926470588235294, "eval_PERSON_recall": 0.9854014598540146, "eval_loss": 0.05965152755379677, "eval_overall_accuracy": 0.9839779005524862, "eval_overall_f1": 0.9221105527638191, "eval_overall_precision": 0.9221105527638191, "eval_overall_recall": 0.9221105527638191, "eval_runtime": 0.2474, "eval_samples_per_second": 687.04, "eval_steps_per_second": 12.124, "step": 8160 }, { "epoch": 86.0, "grad_norm": 1.1350373029708862, "learning_rate": 7.000000000000001e-06, "loss": 0.0113, "step": 8256 }, { "epoch": 86.0, "eval_LOCATION_f1": 0.9175257731958764, "eval_LOCATION_number": 94, "eval_LOCATION_precision": 0.89, "eval_LOCATION_recall": 0.9468085106382979, "eval_ORGANIZATION_f1": 0.8821752265861027, "eval_ORGANIZATION_number": 167, "eval_ORGANIZATION_precision": 0.8902439024390244, "eval_ORGANIZATION_recall": 0.874251497005988, "eval_PERSON_f1": 0.9854014598540146, "eval_PERSON_number": 137, "eval_PERSON_precision": 0.9854014598540146, "eval_PERSON_recall": 0.9854014598540146, "eval_loss": 0.05516235530376434, "eval_overall_accuracy": 0.9842541436464088, "eval_overall_f1": 0.9261576971214018, "eval_overall_precision": 0.9226932668329177, "eval_overall_recall": 0.9296482412060302, "eval_runtime": 0.246, "eval_samples_per_second": 691.043, "eval_steps_per_second": 12.195, "step": 8256 }, { "epoch": 87.0, "grad_norm": 0.5880078673362732, "learning_rate": 6.5000000000000004e-06, "loss": 0.0103, "step": 8352 }, { "epoch": 87.0, "eval_LOCATION_f1": 0.9081632653061226, "eval_LOCATION_number": 94, "eval_LOCATION_precision": 0.8725490196078431, "eval_LOCATION_recall": 0.9468085106382979, "eval_ORGANIZATION_f1": 0.8700906344410877, "eval_ORGANIZATION_number": 167, "eval_ORGANIZATION_precision": 0.8780487804878049, "eval_ORGANIZATION_recall": 0.8622754491017964, "eval_PERSON_f1": 0.9854014598540146, "eval_PERSON_number": 137, "eval_PERSON_precision": 0.9854014598540146, "eval_PERSON_recall": 0.9854014598540146, "eval_loss": 0.05637901648879051, "eval_overall_accuracy": 0.9834254143646409, "eval_overall_f1": 0.9188514357053683, "eval_overall_precision": 0.913151364764268, "eval_overall_recall": 0.9246231155778895, "eval_runtime": 0.2459, "eval_samples_per_second": 691.365, "eval_steps_per_second": 12.201, "step": 8352 }, { "epoch": 88.0, "grad_norm": 0.25769877433776855, "learning_rate": 6e-06, "loss": 0.0109, "step": 8448 }, { "epoch": 88.0, "eval_LOCATION_f1": 0.9128205128205128, "eval_LOCATION_number": 94, "eval_LOCATION_precision": 0.8811881188118812, "eval_LOCATION_recall": 0.9468085106382979, "eval_ORGANIZATION_f1": 0.8787878787878789, "eval_ORGANIZATION_number": 167, "eval_ORGANIZATION_precision": 0.8895705521472392, "eval_ORGANIZATION_recall": 0.8682634730538922, "eval_PERSON_f1": 0.9854014598540146, "eval_PERSON_number": 137, "eval_PERSON_precision": 0.9854014598540146, "eval_PERSON_recall": 0.9854014598540146, "eval_loss": 0.05501656234264374, "eval_overall_accuracy": 0.9837016574585635, "eval_overall_f1": 0.9236545682102627, "eval_overall_precision": 0.9201995012468828, "eval_overall_recall": 0.9271356783919598, "eval_runtime": 0.2484, "eval_samples_per_second": 684.429, "eval_steps_per_second": 12.078, "step": 8448 }, { "epoch": 89.0, "grad_norm": 0.6412275433540344, "learning_rate": 5.500000000000001e-06, "loss": 0.0099, "step": 8544 }, { "epoch": 89.0, "eval_LOCATION_f1": 0.9128205128205128, "eval_LOCATION_number": 94, "eval_LOCATION_precision": 0.8811881188118812, "eval_LOCATION_recall": 0.9468085106382979, "eval_ORGANIZATION_f1": 0.8882175226586103, "eval_ORGANIZATION_number": 167, "eval_ORGANIZATION_precision": 0.8963414634146342, "eval_ORGANIZATION_recall": 0.8802395209580839, "eval_PERSON_f1": 0.9854014598540146, "eval_PERSON_number": 137, "eval_PERSON_precision": 0.9854014598540146, "eval_PERSON_recall": 0.9854014598540146, "eval_loss": 0.05412000045180321, "eval_overall_accuracy": 0.9842541436464088, "eval_overall_f1": 0.9275, "eval_overall_precision": 0.9228855721393034, "eval_overall_recall": 0.9321608040201005, "eval_runtime": 0.2465, "eval_samples_per_second": 689.683, "eval_steps_per_second": 12.171, "step": 8544 }, { "epoch": 90.0, "grad_norm": 0.06483729183673859, "learning_rate": 5e-06, "loss": 0.0096, "step": 8640 }, { "epoch": 90.0, "eval_LOCATION_f1": 0.9128205128205128, "eval_LOCATION_number": 94, "eval_LOCATION_precision": 0.8811881188118812, "eval_LOCATION_recall": 0.9468085106382979, "eval_ORGANIZATION_f1": 0.8648648648648649, "eval_ORGANIZATION_number": 167, "eval_ORGANIZATION_precision": 0.8674698795180723, "eval_ORGANIZATION_recall": 0.8622754491017964, "eval_PERSON_f1": 0.9854014598540146, "eval_PERSON_number": 137, "eval_PERSON_precision": 0.9854014598540146, "eval_PERSON_recall": 0.9854014598540146, "eval_loss": 0.05568243935704231, "eval_overall_accuracy": 0.9831491712707182, "eval_overall_f1": 0.917705735660848, "eval_overall_precision": 0.9108910891089109, "eval_overall_recall": 0.9246231155778895, "eval_runtime": 0.2474, "eval_samples_per_second": 687.099, "eval_steps_per_second": 12.125, "step": 8640 }, { "epoch": 91.0, "grad_norm": 0.06773576885461807, "learning_rate": 4.5e-06, "loss": 0.0101, "step": 8736 }, { "epoch": 91.0, "eval_LOCATION_f1": 0.9128205128205128, "eval_LOCATION_number": 94, "eval_LOCATION_precision": 0.8811881188118812, "eval_LOCATION_recall": 0.9468085106382979, "eval_ORGANIZATION_f1": 0.8666666666666667, "eval_ORGANIZATION_number": 167, "eval_ORGANIZATION_precision": 0.8773006134969326, "eval_ORGANIZATION_recall": 0.8562874251497006, "eval_PERSON_f1": 0.9854014598540146, "eval_PERSON_number": 137, "eval_PERSON_precision": 0.9854014598540146, "eval_PERSON_recall": 0.9854014598540146, "eval_loss": 0.05824797600507736, "eval_overall_accuracy": 0.9834254143646409, "eval_overall_f1": 0.918648310387985, "eval_overall_precision": 0.9152119700748129, "eval_overall_recall": 0.9221105527638191, "eval_runtime": 0.246, "eval_samples_per_second": 691.164, "eval_steps_per_second": 12.197, "step": 8736 }, { "epoch": 92.0, "grad_norm": 0.17988841235637665, "learning_rate": 4.000000000000001e-06, "loss": 0.0106, "step": 8832 }, { "epoch": 92.0, "eval_LOCATION_f1": 0.9081632653061226, "eval_LOCATION_number": 94, "eval_LOCATION_precision": 0.8725490196078431, "eval_LOCATION_recall": 0.9468085106382979, "eval_ORGANIZATION_f1": 0.8674698795180722, "eval_ORGANIZATION_number": 167, "eval_ORGANIZATION_precision": 0.8727272727272727, "eval_ORGANIZATION_recall": 0.8622754491017964, "eval_PERSON_f1": 0.9854014598540146, "eval_PERSON_number": 137, "eval_PERSON_precision": 0.9854014598540146, "eval_PERSON_recall": 0.9854014598540146, "eval_loss": 0.05633843317627907, "eval_overall_accuracy": 0.9831491712707182, "eval_overall_f1": 0.917705735660848, "eval_overall_precision": 0.9108910891089109, "eval_overall_recall": 0.9246231155778895, "eval_runtime": 0.2471, "eval_samples_per_second": 687.846, "eval_steps_per_second": 12.138, "step": 8832 }, { "epoch": 93.0, "grad_norm": 0.3765294849872589, "learning_rate": 3.5000000000000004e-06, "loss": 0.0101, "step": 8928 }, { "epoch": 93.0, "eval_LOCATION_f1": 0.9081632653061226, "eval_LOCATION_number": 94, "eval_LOCATION_precision": 0.8725490196078431, "eval_LOCATION_recall": 0.9468085106382979, "eval_ORGANIZATION_f1": 0.8761329305135951, "eval_ORGANIZATION_number": 167, "eval_ORGANIZATION_precision": 0.8841463414634146, "eval_ORGANIZATION_recall": 0.8682634730538922, "eval_PERSON_f1": 0.9854014598540146, "eval_PERSON_number": 137, "eval_PERSON_precision": 0.9854014598540146, "eval_PERSON_recall": 0.9854014598540146, "eval_loss": 0.05586003512144089, "eval_overall_accuracy": 0.9834254143646409, "eval_overall_f1": 0.9213483146067416, "eval_overall_precision": 0.9156327543424317, "eval_overall_recall": 0.9271356783919598, "eval_runtime": 0.2458, "eval_samples_per_second": 691.574, "eval_steps_per_second": 12.204, "step": 8928 }, { "epoch": 94.0, "grad_norm": 0.9919960498809814, "learning_rate": 3e-06, "loss": 0.0088, "step": 9024 }, { "epoch": 94.0, "eval_LOCATION_f1": 0.9175257731958764, "eval_LOCATION_number": 94, "eval_LOCATION_precision": 0.89, "eval_LOCATION_recall": 0.9468085106382979, "eval_ORGANIZATION_f1": 0.8787878787878789, "eval_ORGANIZATION_number": 167, "eval_ORGANIZATION_precision": 0.8895705521472392, "eval_ORGANIZATION_recall": 0.8682634730538922, "eval_PERSON_f1": 0.9854014598540146, "eval_PERSON_number": 137, "eval_PERSON_precision": 0.9854014598540146, "eval_PERSON_recall": 0.9854014598540146, "eval_loss": 0.05623457208275795, "eval_overall_accuracy": 0.9839779005524862, "eval_overall_f1": 0.924812030075188, "eval_overall_precision": 0.9225, "eval_overall_recall": 0.9271356783919598, "eval_runtime": 0.2463, "eval_samples_per_second": 690.164, "eval_steps_per_second": 12.179, "step": 9024 }, { "epoch": 95.0, "grad_norm": 0.08972230553627014, "learning_rate": 2.5e-06, "loss": 0.0086, "step": 9120 }, { "epoch": 95.0, "eval_LOCATION_f1": 0.9175257731958764, "eval_LOCATION_number": 94, "eval_LOCATION_precision": 0.89, "eval_LOCATION_recall": 0.9468085106382979, "eval_ORGANIZATION_f1": 0.8700906344410877, "eval_ORGANIZATION_number": 167, "eval_ORGANIZATION_precision": 0.8780487804878049, "eval_ORGANIZATION_recall": 0.8622754491017964, "eval_PERSON_f1": 0.9854014598540146, "eval_PERSON_number": 137, "eval_PERSON_precision": 0.9854014598540146, "eval_PERSON_recall": 0.9854014598540146, "eval_loss": 0.05714541673660278, "eval_overall_accuracy": 0.9837016574585635, "eval_overall_f1": 0.9211514392991239, "eval_overall_precision": 0.9177057356608479, "eval_overall_recall": 0.9246231155778895, "eval_runtime": 0.2466, "eval_samples_per_second": 689.426, "eval_steps_per_second": 12.166, "step": 9120 }, { "epoch": 96.0, "grad_norm": 1.009540319442749, "learning_rate": 2.0000000000000003e-06, "loss": 0.0101, "step": 9216 }, { "epoch": 96.0, "eval_LOCATION_f1": 0.9175257731958764, "eval_LOCATION_number": 94, "eval_LOCATION_precision": 0.89, "eval_LOCATION_recall": 0.9468085106382979, "eval_ORGANIZATION_f1": 0.8795180722891567, "eval_ORGANIZATION_number": 167, "eval_ORGANIZATION_precision": 0.8848484848484849, "eval_ORGANIZATION_recall": 0.874251497005988, "eval_PERSON_f1": 0.9854014598540146, "eval_PERSON_number": 137, "eval_PERSON_precision": 0.9854014598540146, "eval_PERSON_recall": 0.9854014598540146, "eval_loss": 0.05605379119515419, "eval_overall_accuracy": 0.9839779005524862, "eval_overall_f1": 0.925, "eval_overall_precision": 0.9203980099502488, "eval_overall_recall": 0.9296482412060302, "eval_runtime": 0.246, "eval_samples_per_second": 691.132, "eval_steps_per_second": 12.196, "step": 9216 }, { "epoch": 97.0, "grad_norm": 0.7972439527511597, "learning_rate": 1.5e-06, "loss": 0.0101, "step": 9312 }, { "epoch": 97.0, "eval_LOCATION_f1": 0.9128205128205128, "eval_LOCATION_number": 94, "eval_LOCATION_precision": 0.8811881188118812, "eval_LOCATION_recall": 0.9468085106382979, "eval_ORGANIZATION_f1": 0.8640483383685801, "eval_ORGANIZATION_number": 167, "eval_ORGANIZATION_precision": 0.8719512195121951, "eval_ORGANIZATION_recall": 0.8562874251497006, "eval_PERSON_f1": 0.9854014598540146, "eval_PERSON_number": 137, "eval_PERSON_precision": 0.9854014598540146, "eval_PERSON_recall": 0.9854014598540146, "eval_loss": 0.05796745419502258, "eval_overall_accuracy": 0.9831491712707182, "eval_overall_f1": 0.9175, "eval_overall_precision": 0.9129353233830846, "eval_overall_recall": 0.9221105527638191, "eval_runtime": 0.2476, "eval_samples_per_second": 686.709, "eval_steps_per_second": 12.118, "step": 9312 }, { "epoch": 98.0, "grad_norm": 0.3534379303455353, "learning_rate": 1.0000000000000002e-06, "loss": 0.0099, "step": 9408 }, { "epoch": 98.0, "eval_LOCATION_f1": 0.9175257731958764, "eval_LOCATION_number": 94, "eval_LOCATION_precision": 0.89, "eval_LOCATION_recall": 0.9468085106382979, "eval_ORGANIZATION_f1": 0.8795180722891567, "eval_ORGANIZATION_number": 167, "eval_ORGANIZATION_precision": 0.8848484848484849, "eval_ORGANIZATION_recall": 0.874251497005988, "eval_PERSON_f1": 0.9854014598540146, "eval_PERSON_number": 137, "eval_PERSON_precision": 0.9854014598540146, "eval_PERSON_recall": 0.9854014598540146, "eval_loss": 0.05727317929267883, "eval_overall_accuracy": 0.9837016574585635, "eval_overall_f1": 0.925, "eval_overall_precision": 0.9203980099502488, "eval_overall_recall": 0.9296482412060302, "eval_runtime": 0.2467, "eval_samples_per_second": 688.969, "eval_steps_per_second": 12.158, "step": 9408 }, { "epoch": 99.0, "grad_norm": 0.6432434320449829, "learning_rate": 5.000000000000001e-07, "loss": 0.0113, "step": 9504 }, { "epoch": 99.0, "eval_LOCATION_f1": 0.9175257731958764, "eval_LOCATION_number": 94, "eval_LOCATION_precision": 0.89, "eval_LOCATION_recall": 0.9468085106382979, "eval_ORGANIZATION_f1": 0.8795180722891567, "eval_ORGANIZATION_number": 167, "eval_ORGANIZATION_precision": 0.8848484848484849, "eval_ORGANIZATION_recall": 0.874251497005988, "eval_PERSON_f1": 0.9854014598540146, "eval_PERSON_number": 137, "eval_PERSON_precision": 0.9854014598540146, "eval_PERSON_recall": 0.9854014598540146, "eval_loss": 0.0567387193441391, "eval_overall_accuracy": 0.9839779005524862, "eval_overall_f1": 0.925, "eval_overall_precision": 0.9203980099502488, "eval_overall_recall": 0.9296482412060302, "eval_runtime": 0.2474, "eval_samples_per_second": 687.09, "eval_steps_per_second": 12.125, "step": 9504 }, { "epoch": 100.0, "grad_norm": 0.33665773272514343, "learning_rate": 0.0, "loss": 0.0091, "step": 9600 }, { "epoch": 100.0, "eval_LOCATION_f1": 0.9175257731958764, "eval_LOCATION_number": 94, "eval_LOCATION_precision": 0.89, "eval_LOCATION_recall": 0.9468085106382979, "eval_ORGANIZATION_f1": 0.8795180722891567, "eval_ORGANIZATION_number": 167, "eval_ORGANIZATION_precision": 0.8848484848484849, "eval_ORGANIZATION_recall": 0.874251497005988, "eval_PERSON_f1": 0.9854014598540146, "eval_PERSON_number": 137, "eval_PERSON_precision": 0.9854014598540146, "eval_PERSON_recall": 0.9854014598540146, "eval_loss": 0.056667208671569824, "eval_overall_accuracy": 0.9839779005524862, "eval_overall_f1": 0.925, "eval_overall_precision": 0.9203980099502488, "eval_overall_recall": 0.9296482412060302, "eval_runtime": 0.2462, "eval_samples_per_second": 690.381, "eval_steps_per_second": 12.183, "step": 9600 }, { "epoch": 100.0, "step": 9600, "total_flos": 3891827501465220.0, "train_loss": 0.04628125069042047, "train_runtime": 564.024, "train_samples_per_second": 271.265, "train_steps_per_second": 17.021 } ], "logging_steps": 500, "max_steps": 9600, "num_input_tokens_seen": 0, "num_train_epochs": 100, "save_steps": 500, "total_flos": 3891827501465220.0, "train_batch_size": 16, "trial_name": null, "trial_params": null }