{ "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": 0.6211287975311279, "learning_rate": 4.9500000000000004e-05, "loss": 0.2566, "step": 96 }, { "epoch": 1.0, "eval_LOCATION_f1": 0.8977272727272728, "eval_LOCATION_number": 94, "eval_LOCATION_precision": 0.9634146341463414, "eval_LOCATION_recall": 0.8404255319148937, "eval_ORGANIZATION_f1": 0.8781869688385269, "eval_ORGANIZATION_number": 167, "eval_ORGANIZATION_precision": 0.8333333333333334, "eval_ORGANIZATION_recall": 0.9281437125748503, "eval_PERSON_f1": 0.9708029197080292, "eval_PERSON_number": 137, "eval_PERSON_precision": 0.9708029197080292, "eval_PERSON_recall": 0.9708029197080292, "eval_loss": 0.045521095395088196, "eval_overall_accuracy": 0.9842541436464088, "eval_overall_f1": 0.9140722291407224, "eval_overall_precision": 0.9061728395061729, "eval_overall_recall": 0.9221105527638191, "eval_runtime": 0.2965, "eval_samples_per_second": 573.325, "eval_steps_per_second": 10.117, "step": 96 }, { "epoch": 2.0, "grad_norm": 4.20517635345459, "learning_rate": 4.9e-05, "loss": 0.0617, "step": 192 }, { "epoch": 2.0, "eval_LOCATION_f1": 0.8844221105527638, "eval_LOCATION_number": 94, "eval_LOCATION_precision": 0.8380952380952381, "eval_LOCATION_recall": 0.9361702127659575, "eval_ORGANIZATION_f1": 0.8787878787878789, "eval_ORGANIZATION_number": 167, "eval_ORGANIZATION_precision": 0.8895705521472392, "eval_ORGANIZATION_recall": 0.8682634730538922, "eval_PERSON_f1": 0.9852941176470589, "eval_PERSON_number": 137, "eval_PERSON_precision": 0.9925925925925926, "eval_PERSON_recall": 0.9781021897810219, "eval_loss": 0.05185200273990631, "eval_overall_accuracy": 0.9834254143646409, "eval_overall_f1": 0.9163545568039951, "eval_overall_precision": 0.9106699751861043, "eval_overall_recall": 0.9221105527638191, "eval_runtime": 0.28, "eval_samples_per_second": 607.142, "eval_steps_per_second": 10.714, "step": 192 }, { "epoch": 3.0, "grad_norm": 2.140643358230591, "learning_rate": 4.85e-05, "loss": 0.0356, "step": 288 }, { "epoch": 3.0, "eval_LOCATION_f1": 0.9157894736842105, "eval_LOCATION_number": 94, "eval_LOCATION_precision": 0.90625, "eval_LOCATION_recall": 0.925531914893617, "eval_ORGANIZATION_f1": 0.8739495798319327, "eval_ORGANIZATION_number": 167, "eval_ORGANIZATION_precision": 0.8210526315789474, "eval_ORGANIZATION_recall": 0.9341317365269461, "eval_PERSON_f1": 0.9851851851851852, "eval_PERSON_number": 137, "eval_PERSON_precision": 1.0, "eval_PERSON_recall": 0.9708029197080292, "eval_loss": 0.05342954769730568, "eval_overall_accuracy": 0.9839779005524862, "eval_overall_f1": 0.9204406364749081, "eval_overall_precision": 0.8973747016706444, "eval_overall_recall": 0.9447236180904522, "eval_runtime": 0.2759, "eval_samples_per_second": 616.12, "eval_steps_per_second": 10.873, "step": 288 }, { "epoch": 4.0, "grad_norm": 0.18180230259895325, "learning_rate": 4.8e-05, "loss": 0.0235, "step": 384 }, { "epoch": 4.0, "eval_LOCATION_f1": 0.900523560209424, "eval_LOCATION_number": 94, "eval_LOCATION_precision": 0.8865979381443299, "eval_LOCATION_recall": 0.9148936170212766, "eval_ORGANIZATION_f1": 0.9112426035502958, "eval_ORGANIZATION_number": 167, "eval_ORGANIZATION_precision": 0.9005847953216374, "eval_ORGANIZATION_recall": 0.9221556886227545, "eval_PERSON_f1": 0.988929889298893, "eval_PERSON_number": 137, "eval_PERSON_precision": 1.0, "eval_PERSON_recall": 0.9781021897810219, "eval_loss": 0.052525218576192856, "eval_overall_accuracy": 0.9856353591160221, "eval_overall_f1": 0.9349999999999999, "eval_overall_precision": 0.9303482587064676, "eval_overall_recall": 0.9396984924623115, "eval_runtime": 0.2764, "eval_samples_per_second": 615.14, "eval_steps_per_second": 10.855, "step": 384 }, { "epoch": 5.0, "grad_norm": 1.7510687112808228, "learning_rate": 4.75e-05, "loss": 0.0156, "step": 480 }, { "epoch": 5.0, "eval_LOCATION_f1": 0.8983957219251337, "eval_LOCATION_number": 94, "eval_LOCATION_precision": 0.9032258064516129, "eval_LOCATION_recall": 0.8936170212765957, "eval_ORGANIZATION_f1": 0.927710843373494, "eval_ORGANIZATION_number": 167, "eval_ORGANIZATION_precision": 0.9333333333333333, "eval_ORGANIZATION_recall": 0.9221556886227545, "eval_PERSON_f1": 0.9852941176470589, "eval_PERSON_number": 137, "eval_PERSON_precision": 0.9925925925925926, "eval_PERSON_recall": 0.9781021897810219, "eval_loss": 0.062274035066366196, "eval_overall_accuracy": 0.987292817679558, "eval_overall_f1": 0.9405815423514539, "eval_overall_precision": 0.9465648854961832, "eval_overall_recall": 0.9346733668341709, "eval_runtime": 0.2771, "eval_samples_per_second": 613.414, "eval_steps_per_second": 10.825, "step": 480 }, { "epoch": 6.0, "grad_norm": 7.3227858543396, "learning_rate": 4.7e-05, "loss": 0.0101, "step": 576 }, { "epoch": 6.0, "eval_LOCATION_f1": 0.9042553191489362, "eval_LOCATION_number": 94, "eval_LOCATION_precision": 0.9042553191489362, "eval_LOCATION_recall": 0.9042553191489362, "eval_ORGANIZATION_f1": 0.8955223880597015, "eval_ORGANIZATION_number": 167, "eval_ORGANIZATION_precision": 0.8928571428571429, "eval_ORGANIZATION_recall": 0.8982035928143712, "eval_PERSON_f1": 0.9852941176470589, "eval_PERSON_number": 137, "eval_PERSON_precision": 0.9925925925925926, "eval_PERSON_recall": 0.9781021897810219, "eval_loss": 0.05896053463220596, "eval_overall_accuracy": 0.9859116022099448, "eval_overall_f1": 0.9283018867924528, "eval_overall_precision": 0.929471032745592, "eval_overall_recall": 0.9271356783919598, "eval_runtime": 0.2757, "eval_samples_per_second": 616.609, "eval_steps_per_second": 10.881, "step": 576 }, { "epoch": 7.0, "grad_norm": 1.0227528810501099, "learning_rate": 4.6500000000000005e-05, "loss": 0.0091, "step": 672 }, { "epoch": 7.0, "eval_LOCATION_f1": 0.8737864077669905, "eval_LOCATION_number": 94, "eval_LOCATION_precision": 0.8035714285714286, "eval_LOCATION_recall": 0.9574468085106383, "eval_ORGANIZATION_f1": 0.877742946708464, "eval_ORGANIZATION_number": 167, "eval_ORGANIZATION_precision": 0.9210526315789473, "eval_ORGANIZATION_recall": 0.8383233532934131, "eval_PERSON_f1": 0.9747292418772563, "eval_PERSON_number": 137, "eval_PERSON_precision": 0.9642857142857143, "eval_PERSON_recall": 0.9854014598540146, "eval_loss": 0.09551988542079926, "eval_overall_accuracy": 0.9809392265193371, "eval_overall_f1": 0.9102244389027431, "eval_overall_precision": 0.9034653465346535, "eval_overall_recall": 0.9170854271356784, "eval_runtime": 0.2779, "eval_samples_per_second": 611.78, "eval_steps_per_second": 10.796, "step": 672 }, { "epoch": 8.0, "grad_norm": 6.864110469818115, "learning_rate": 4.600000000000001e-05, "loss": 0.0084, "step": 768 }, { "epoch": 8.0, "eval_LOCATION_f1": 0.8787878787878788, "eval_LOCATION_number": 94, "eval_LOCATION_precision": 0.8365384615384616, "eval_LOCATION_recall": 0.925531914893617, "eval_ORGANIZATION_f1": 0.8868501529051986, "eval_ORGANIZATION_number": 167, "eval_ORGANIZATION_precision": 0.90625, "eval_ORGANIZATION_recall": 0.8682634730538922, "eval_PERSON_f1": 0.988929889298893, "eval_PERSON_number": 137, "eval_PERSON_precision": 1.0, "eval_PERSON_recall": 0.9781021897810219, "eval_loss": 0.0871056392788887, "eval_overall_accuracy": 0.9825966850828729, "eval_overall_f1": 0.9195979899497487, "eval_overall_precision": 0.9195979899497487, "eval_overall_recall": 0.9195979899497487, "eval_runtime": 0.2758, "eval_samples_per_second": 616.353, "eval_steps_per_second": 10.877, "step": 768 }, { "epoch": 9.0, "grad_norm": 0.044208019971847534, "learning_rate": 4.55e-05, "loss": 0.007, "step": 864 }, { "epoch": 9.0, "eval_LOCATION_f1": 0.9462365591397849, "eval_LOCATION_number": 94, "eval_LOCATION_precision": 0.9565217391304348, "eval_LOCATION_recall": 0.9361702127659575, "eval_ORGANIZATION_f1": 0.9026548672566371, "eval_ORGANIZATION_number": 167, "eval_ORGANIZATION_precision": 0.8895348837209303, "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": 0.06293077766895294, "eval_overall_accuracy": 0.9881215469613259, "eval_overall_f1": 0.9435382685069008, "eval_overall_precision": 0.9423558897243107, "eval_overall_recall": 0.9447236180904522, "eval_runtime": 0.2804, "eval_samples_per_second": 606.265, "eval_steps_per_second": 10.699, "step": 864 }, { "epoch": 10.0, "grad_norm": 0.006904821842908859, "learning_rate": 4.5e-05, "loss": 0.0047, "step": 960 }, { "epoch": 10.0, "eval_LOCATION_f1": 0.9263157894736843, "eval_LOCATION_number": 94, "eval_LOCATION_precision": 0.9166666666666666, "eval_LOCATION_recall": 0.9361702127659575, "eval_ORGANIZATION_f1": 0.9425981873111783, "eval_ORGANIZATION_number": 167, "eval_ORGANIZATION_precision": 0.9512195121951219, "eval_ORGANIZATION_recall": 0.9341317365269461, "eval_PERSON_f1": 0.988929889298893, "eval_PERSON_number": 137, "eval_PERSON_precision": 1.0, "eval_PERSON_recall": 0.9781021897810219, "eval_loss": 0.056356631219387054, "eval_overall_accuracy": 0.9900552486187846, "eval_overall_f1": 0.9545454545454546, "eval_overall_precision": 0.9593908629441624, "eval_overall_recall": 0.949748743718593, "eval_runtime": 0.2737, "eval_samples_per_second": 621.22, "eval_steps_per_second": 10.963, "step": 960 }, { "epoch": 11.0, "grad_norm": 0.08195316791534424, "learning_rate": 4.4500000000000004e-05, "loss": 0.0043, "step": 1056 }, { "epoch": 11.0, "eval_LOCATION_f1": 0.9206349206349206, "eval_LOCATION_number": 94, "eval_LOCATION_precision": 0.9157894736842105, "eval_LOCATION_recall": 0.925531914893617, "eval_ORGANIZATION_f1": 0.898550724637681, "eval_ORGANIZATION_number": 167, "eval_ORGANIZATION_precision": 0.8707865168539326, "eval_ORGANIZATION_recall": 0.9281437125748503, "eval_PERSON_f1": 0.9852941176470589, "eval_PERSON_number": 137, "eval_PERSON_precision": 0.9925925925925926, "eval_PERSON_recall": 0.9781021897810219, "eval_loss": 0.08287806808948517, "eval_overall_accuracy": 0.9856353591160221, "eval_overall_f1": 0.9330024813895781, "eval_overall_precision": 0.9215686274509803, "eval_overall_recall": 0.9447236180904522, "eval_runtime": 0.289, "eval_samples_per_second": 588.326, "eval_steps_per_second": 10.382, "step": 1056 }, { "epoch": 12.0, "grad_norm": 0.005107543431222439, "learning_rate": 4.4000000000000006e-05, "loss": 0.0034, "step": 1152 }, { "epoch": 12.0, "eval_LOCATION_f1": 0.9197860962566845, "eval_LOCATION_number": 94, "eval_LOCATION_precision": 0.9247311827956989, "eval_LOCATION_recall": 0.9148936170212766, "eval_ORGANIZATION_f1": 0.899135446685879, "eval_ORGANIZATION_number": 167, "eval_ORGANIZATION_precision": 0.8666666666666667, "eval_ORGANIZATION_recall": 0.9341317365269461, "eval_PERSON_f1": 0.9852941176470589, "eval_PERSON_number": 137, "eval_PERSON_precision": 0.9925925925925926, "eval_PERSON_recall": 0.9781021897810219, "eval_loss": 0.07792431116104126, "eval_overall_accuracy": 0.9864640883977901, "eval_overall_f1": 0.9330024813895781, "eval_overall_precision": 0.9215686274509803, "eval_overall_recall": 0.9447236180904522, "eval_runtime": 0.2757, "eval_samples_per_second": 616.704, "eval_steps_per_second": 10.883, "step": 1152 }, { "epoch": 13.0, "grad_norm": 0.008943918161094189, "learning_rate": 4.35e-05, "loss": 0.0047, "step": 1248 }, { "epoch": 13.0, "eval_LOCATION_f1": 0.9285714285714286, "eval_LOCATION_number": 94, "eval_LOCATION_precision": 0.8921568627450981, "eval_LOCATION_recall": 0.9680851063829787, "eval_ORGANIZATION_f1": 0.929663608562691, "eval_ORGANIZATION_number": 167, "eval_ORGANIZATION_precision": 0.95, "eval_ORGANIZATION_recall": 0.9101796407185628, "eval_PERSON_f1": 0.9854014598540146, "eval_PERSON_number": 137, "eval_PERSON_precision": 0.9854014598540146, "eval_PERSON_recall": 0.9854014598540146, "eval_loss": 0.07808144390583038, "eval_overall_accuracy": 0.9861878453038674, "eval_overall_f1": 0.9485570890840652, "eval_overall_precision": 0.9473684210526315, "eval_overall_recall": 0.949748743718593, "eval_runtime": 0.2806, "eval_samples_per_second": 605.748, "eval_steps_per_second": 10.69, "step": 1248 }, { "epoch": 14.0, "grad_norm": 0.007531584240496159, "learning_rate": 4.3e-05, "loss": 0.006, "step": 1344 }, { "epoch": 14.0, "eval_LOCATION_f1": 0.9368421052631578, "eval_LOCATION_number": 94, "eval_LOCATION_precision": 0.9270833333333334, "eval_LOCATION_recall": 0.9468085106382979, "eval_ORGANIZATION_f1": 0.8950617283950617, "eval_ORGANIZATION_number": 167, "eval_ORGANIZATION_precision": 0.9235668789808917, "eval_ORGANIZATION_recall": 0.8682634730538922, "eval_PERSON_f1": 0.988929889298893, "eval_PERSON_number": 137, "eval_PERSON_precision": 1.0, "eval_PERSON_recall": 0.9781021897810219, "eval_loss": 0.0682210773229599, "eval_overall_accuracy": 0.9859116022099448, "eval_overall_f1": 0.9375796178343948, "eval_overall_precision": 0.9509043927648578, "eval_overall_recall": 0.9246231155778895, "eval_runtime": 0.2764, "eval_samples_per_second": 614.962, "eval_steps_per_second": 10.852, "step": 1344 }, { "epoch": 15.0, "grad_norm": 5.901050567626953, "learning_rate": 4.25e-05, "loss": 0.0031, "step": 1440 }, { "epoch": 15.0, "eval_LOCATION_f1": 0.9148936170212766, "eval_LOCATION_number": 94, "eval_LOCATION_precision": 0.9148936170212766, "eval_LOCATION_recall": 0.9148936170212766, "eval_ORGANIZATION_f1": 0.9069767441860466, "eval_ORGANIZATION_number": 167, "eval_ORGANIZATION_precision": 0.8813559322033898, "eval_ORGANIZATION_recall": 0.9341317365269461, "eval_PERSON_f1": 0.9852941176470589, "eval_PERSON_number": 137, "eval_PERSON_precision": 0.9925925925925926, "eval_PERSON_recall": 0.9781021897810219, "eval_loss": 0.07585744559764862, "eval_overall_accuracy": 0.9878453038674033, "eval_overall_f1": 0.9353233830845771, "eval_overall_precision": 0.9261083743842364, "eval_overall_recall": 0.9447236180904522, "eval_runtime": 0.2913, "eval_samples_per_second": 583.672, "eval_steps_per_second": 10.3, "step": 1440 }, { "epoch": 16.0, "grad_norm": 0.19489392638206482, "learning_rate": 4.2e-05, "loss": 0.0049, "step": 1536 }, { "epoch": 16.0, "eval_LOCATION_f1": 0.9270833333333334, "eval_LOCATION_number": 94, "eval_LOCATION_precision": 0.9081632653061225, "eval_LOCATION_recall": 0.9468085106382979, "eval_ORGANIZATION_f1": 0.9134328358208956, "eval_ORGANIZATION_number": 167, "eval_ORGANIZATION_precision": 0.9107142857142857, "eval_ORGANIZATION_recall": 0.9161676646706587, "eval_PERSON_f1": 0.971223021582734, "eval_PERSON_number": 137, "eval_PERSON_precision": 0.9574468085106383, "eval_PERSON_recall": 0.9854014598540146, "eval_loss": 0.08010715246200562, "eval_overall_accuracy": 0.9864640883977901, "eval_overall_f1": 0.9366459627329192, "eval_overall_precision": 0.9262899262899262, "eval_overall_recall": 0.9472361809045227, "eval_runtime": 0.2774, "eval_samples_per_second": 612.779, "eval_steps_per_second": 10.814, "step": 1536 }, { "epoch": 17.0, "grad_norm": 0.04504770413041115, "learning_rate": 4.15e-05, "loss": 0.0036, "step": 1632 }, { "epoch": 17.0, "eval_LOCATION_f1": 0.9424083769633509, "eval_LOCATION_number": 94, "eval_LOCATION_precision": 0.9278350515463918, "eval_LOCATION_recall": 0.9574468085106383, "eval_ORGANIZATION_f1": 0.927710843373494, "eval_ORGANIZATION_number": 167, "eval_ORGANIZATION_precision": 0.9333333333333333, "eval_ORGANIZATION_recall": 0.9221556886227545, "eval_PERSON_f1": 0.9816849816849818, "eval_PERSON_number": 137, "eval_PERSON_precision": 0.9852941176470589, "eval_PERSON_recall": 0.9781021897810219, "eval_loss": 0.0933481752872467, "eval_overall_accuracy": 0.9886740331491712, "eval_overall_f1": 0.949748743718593, "eval_overall_precision": 0.949748743718593, "eval_overall_recall": 0.949748743718593, "eval_runtime": 0.2773, "eval_samples_per_second": 613.019, "eval_steps_per_second": 10.818, "step": 1632 }, { "epoch": 18.0, "grad_norm": 0.03767470270395279, "learning_rate": 4.1e-05, "loss": 0.0033, "step": 1728 }, { "epoch": 18.0, "eval_LOCATION_f1": 0.9263157894736843, "eval_LOCATION_number": 94, "eval_LOCATION_precision": 0.9166666666666666, "eval_LOCATION_recall": 0.9361702127659575, "eval_ORGANIZATION_f1": 0.9194029850746269, "eval_ORGANIZATION_number": 167, "eval_ORGANIZATION_precision": 0.9166666666666666, "eval_ORGANIZATION_recall": 0.9221556886227545, "eval_PERSON_f1": 0.9852941176470589, "eval_PERSON_number": 137, "eval_PERSON_precision": 0.9925925925925926, "eval_PERSON_recall": 0.9781021897810219, "eval_loss": 0.08282637596130371, "eval_overall_accuracy": 0.9870165745856354, "eval_overall_f1": 0.9435382685069008, "eval_overall_precision": 0.9423558897243107, "eval_overall_recall": 0.9447236180904522, "eval_runtime": 0.2764, "eval_samples_per_second": 614.945, "eval_steps_per_second": 10.852, "step": 1728 }, { "epoch": 19.0, "grad_norm": 0.01577984355390072, "learning_rate": 4.05e-05, "loss": 0.0031, "step": 1824 }, { "epoch": 19.0, "eval_LOCATION_f1": 0.9148936170212766, "eval_LOCATION_number": 94, "eval_LOCATION_precision": 0.9148936170212766, "eval_LOCATION_recall": 0.9148936170212766, "eval_ORGANIZATION_f1": 0.9101796407185628, "eval_ORGANIZATION_number": 167, "eval_ORGANIZATION_precision": 0.9101796407185628, "eval_ORGANIZATION_recall": 0.9101796407185628, "eval_PERSON_f1": 0.9708029197080292, "eval_PERSON_number": 137, "eval_PERSON_precision": 0.9708029197080292, "eval_PERSON_recall": 0.9708029197080292, "eval_loss": 0.0819123387336731, "eval_overall_accuracy": 0.987292817679558, "eval_overall_f1": 0.9321608040201005, "eval_overall_precision": 0.9321608040201005, "eval_overall_recall": 0.9321608040201005, "eval_runtime": 0.2771, "eval_samples_per_second": 613.507, "eval_steps_per_second": 10.827, "step": 1824 }, { "epoch": 20.0, "grad_norm": 0.011010286398231983, "learning_rate": 4e-05, "loss": 0.0025, "step": 1920 }, { "epoch": 20.0, "eval_LOCATION_f1": 0.9109947643979057, "eval_LOCATION_number": 94, "eval_LOCATION_precision": 0.8969072164948454, "eval_LOCATION_recall": 0.925531914893617, "eval_ORGANIZATION_f1": 0.9179331306990882, "eval_ORGANIZATION_number": 167, "eval_ORGANIZATION_precision": 0.9320987654320988, "eval_ORGANIZATION_recall": 0.9041916167664671, "eval_PERSON_f1": 0.9708029197080292, "eval_PERSON_number": 137, "eval_PERSON_precision": 0.9708029197080292, "eval_PERSON_recall": 0.9708029197080292, "eval_loss": 0.08713414520025253, "eval_overall_accuracy": 0.9878453038674033, "eval_overall_f1": 0.9345088161209069, "eval_overall_precision": 0.9368686868686869, "eval_overall_recall": 0.9321608040201005, "eval_runtime": 0.2766, "eval_samples_per_second": 614.511, "eval_steps_per_second": 10.844, "step": 1920 }, { "epoch": 21.0, "grad_norm": 0.02176801860332489, "learning_rate": 3.9500000000000005e-05, "loss": 0.0023, "step": 2016 }, { "epoch": 21.0, "eval_LOCATION_f1": 0.9175257731958764, "eval_LOCATION_number": 94, "eval_LOCATION_precision": 0.89, "eval_LOCATION_recall": 0.9468085106382979, "eval_ORGANIZATION_f1": 0.9161676646706587, "eval_ORGANIZATION_number": 167, "eval_ORGANIZATION_precision": 0.9161676646706587, "eval_ORGANIZATION_recall": 0.9161676646706587, "eval_PERSON_f1": 0.967032967032967, "eval_PERSON_number": 137, "eval_PERSON_precision": 0.9705882352941176, "eval_PERSON_recall": 0.9635036496350365, "eval_loss": 0.08127959817647934, "eval_overall_accuracy": 0.987292817679558, "eval_overall_f1": 0.9338327091136079, "eval_overall_precision": 0.9280397022332506, "eval_overall_recall": 0.9396984924623115, "eval_runtime": 0.2768, "eval_samples_per_second": 614.103, "eval_steps_per_second": 10.837, "step": 2016 }, { "epoch": 22.0, "grad_norm": 0.0052330587059259415, "learning_rate": 3.9000000000000006e-05, "loss": 0.0023, "step": 2112 }, { "epoch": 22.0, "eval_LOCATION_f1": 0.9206349206349206, "eval_LOCATION_number": 94, "eval_LOCATION_precision": 0.9157894736842105, "eval_LOCATION_recall": 0.925531914893617, "eval_ORGANIZATION_f1": 0.9069767441860466, "eval_ORGANIZATION_number": 167, "eval_ORGANIZATION_precision": 0.8813559322033898, "eval_ORGANIZATION_recall": 0.9341317365269461, "eval_PERSON_f1": 0.9814126394052044, "eval_PERSON_number": 137, "eval_PERSON_precision": 1.0, "eval_PERSON_recall": 0.9635036496350365, "eval_loss": 0.08847425132989883, "eval_overall_accuracy": 0.9867403314917127, "eval_overall_f1": 0.9351620947630923, "eval_overall_precision": 0.9282178217821783, "eval_overall_recall": 0.9422110552763819, "eval_runtime": 0.2767, "eval_samples_per_second": 614.387, "eval_steps_per_second": 10.842, "step": 2112 }, { "epoch": 23.0, "grad_norm": 5.5226826667785645, "learning_rate": 3.85e-05, "loss": 0.0018, "step": 2208 }, { "epoch": 23.0, "eval_LOCATION_f1": 0.9015544041450778, "eval_LOCATION_number": 94, "eval_LOCATION_precision": 0.8787878787878788, "eval_LOCATION_recall": 0.925531914893617, "eval_ORGANIZATION_f1": 0.9053254437869822, "eval_ORGANIZATION_number": 167, "eval_ORGANIZATION_precision": 0.8947368421052632, "eval_ORGANIZATION_recall": 0.9161676646706587, "eval_PERSON_f1": 0.9743589743589743, "eval_PERSON_number": 137, "eval_PERSON_precision": 0.9779411764705882, "eval_PERSON_recall": 0.9708029197080292, "eval_loss": 0.12085793167352676, "eval_overall_accuracy": 0.9837016574585635, "eval_overall_f1": 0.927860696517413, "eval_overall_precision": 0.9187192118226601, "eval_overall_recall": 0.9371859296482412, "eval_runtime": 0.2767, "eval_samples_per_second": 614.485, "eval_steps_per_second": 10.844, "step": 2208 }, { "epoch": 24.0, "grad_norm": 0.007373427506536245, "learning_rate": 3.8e-05, "loss": 0.0036, "step": 2304 }, { "epoch": 24.0, "eval_LOCATION_f1": 0.9319371727748691, "eval_LOCATION_number": 94, "eval_LOCATION_precision": 0.9175257731958762, "eval_LOCATION_recall": 0.9468085106382979, "eval_ORGANIZATION_f1": 0.9239766081871345, "eval_ORGANIZATION_number": 167, "eval_ORGANIZATION_precision": 0.9028571428571428, "eval_ORGANIZATION_recall": 0.9461077844311377, "eval_PERSON_f1": 0.9816849816849818, "eval_PERSON_number": 137, "eval_PERSON_precision": 0.9852941176470589, "eval_PERSON_recall": 0.9781021897810219, "eval_loss": 0.084134042263031, "eval_overall_accuracy": 0.9878453038674033, "eval_overall_f1": 0.945409429280397, "eval_overall_precision": 0.9338235294117647, "eval_overall_recall": 0.957286432160804, "eval_runtime": 0.2768, "eval_samples_per_second": 614.108, "eval_steps_per_second": 10.837, "step": 2304 }, { "epoch": 25.0, "grad_norm": 0.0022067981772124767, "learning_rate": 3.7500000000000003e-05, "loss": 0.0034, "step": 2400 }, { "epoch": 25.0, "eval_LOCATION_f1": 0.9417989417989417, "eval_LOCATION_number": 94, "eval_LOCATION_precision": 0.9368421052631579, "eval_LOCATION_recall": 0.9468085106382979, "eval_ORGANIZATION_f1": 0.9321533923303835, "eval_ORGANIZATION_number": 167, "eval_ORGANIZATION_precision": 0.9186046511627907, "eval_ORGANIZATION_recall": 0.9461077844311377, "eval_PERSON_f1": 0.9852941176470589, "eval_PERSON_number": 137, "eval_PERSON_precision": 0.9925925925925926, "eval_PERSON_recall": 0.9781021897810219, "eval_loss": 0.08604763448238373, "eval_overall_accuracy": 0.9883977900552486, "eval_overall_f1": 0.9525, "eval_overall_precision": 0.9477611940298507, "eval_overall_recall": 0.957286432160804, "eval_runtime": 0.2883, "eval_samples_per_second": 589.583, "eval_steps_per_second": 10.404, "step": 2400 }, { "epoch": 26.0, "grad_norm": 0.0010795200942084193, "learning_rate": 3.7e-05, "loss": 0.0029, "step": 2496 }, { "epoch": 26.0, "eval_LOCATION_f1": 0.9528795811518325, "eval_LOCATION_number": 94, "eval_LOCATION_precision": 0.9381443298969072, "eval_LOCATION_recall": 0.9680851063829787, "eval_ORGANIZATION_f1": 0.9258160237388723, "eval_ORGANIZATION_number": 167, "eval_ORGANIZATION_precision": 0.9176470588235294, "eval_ORGANIZATION_recall": 0.9341317365269461, "eval_PERSON_f1": 0.9852941176470589, "eval_PERSON_number": 137, "eval_PERSON_precision": 0.9925925925925926, "eval_PERSON_recall": 0.9781021897810219, "eval_loss": 0.0684206485748291, "eval_overall_accuracy": 0.9897790055248619, "eval_overall_f1": 0.9525, "eval_overall_precision": 0.9477611940298507, "eval_overall_recall": 0.957286432160804, "eval_runtime": 0.2771, "eval_samples_per_second": 613.45, "eval_steps_per_second": 10.826, "step": 2496 }, { "epoch": 27.0, "grad_norm": 0.03736229985952377, "learning_rate": 3.65e-05, "loss": 0.0031, "step": 2592 }, { "epoch": 27.0, "eval_LOCATION_f1": 0.9424083769633509, "eval_LOCATION_number": 94, "eval_LOCATION_precision": 0.9278350515463918, "eval_LOCATION_recall": 0.9574468085106383, "eval_ORGANIZATION_f1": 0.9217391304347825, "eval_ORGANIZATION_number": 167, "eval_ORGANIZATION_precision": 0.8932584269662921, "eval_ORGANIZATION_recall": 0.9520958083832335, "eval_PERSON_f1": 0.9852941176470589, "eval_PERSON_number": 137, "eval_PERSON_precision": 0.9925925925925926, "eval_PERSON_recall": 0.9781021897810219, "eval_loss": 0.11582713574171066, "eval_overall_accuracy": 0.9864640883977901, "eval_overall_f1": 0.948019801980198, "eval_overall_precision": 0.9341463414634147, "eval_overall_recall": 0.9623115577889447, "eval_runtime": 0.2768, "eval_samples_per_second": 614.053, "eval_steps_per_second": 10.836, "step": 2592 }, { "epoch": 28.0, "grad_norm": 1.129722237586975, "learning_rate": 3.6e-05, "loss": 0.0045, "step": 2688 }, { "epoch": 28.0, "eval_LOCATION_f1": 0.9312169312169313, "eval_LOCATION_number": 94, "eval_LOCATION_precision": 0.9263157894736842, "eval_LOCATION_recall": 0.9361702127659575, "eval_ORGANIZATION_f1": 0.8882175226586103, "eval_ORGANIZATION_number": 167, "eval_ORGANIZATION_precision": 0.8963414634146342, "eval_ORGANIZATION_recall": 0.8802395209580839, "eval_PERSON_f1": 0.9852941176470589, "eval_PERSON_number": 137, "eval_PERSON_precision": 0.9925925925925926, "eval_PERSON_recall": 0.9781021897810219, "eval_loss": 0.08601514250040054, "eval_overall_accuracy": 0.9853591160220995, "eval_overall_f1": 0.9318181818181818, "eval_overall_precision": 0.9365482233502538, "eval_overall_recall": 0.9271356783919598, "eval_runtime": 0.2772, "eval_samples_per_second": 613.353, "eval_steps_per_second": 10.824, "step": 2688 }, { "epoch": 29.0, "grad_norm": 0.004048886243253946, "learning_rate": 3.55e-05, "loss": 0.0018, "step": 2784 }, { "epoch": 29.0, "eval_LOCATION_f1": 0.9368421052631578, "eval_LOCATION_number": 94, "eval_LOCATION_precision": 0.9270833333333334, "eval_LOCATION_recall": 0.9468085106382979, "eval_ORGANIZATION_f1": 0.9345238095238095, "eval_ORGANIZATION_number": 167, "eval_ORGANIZATION_precision": 0.9289940828402367, "eval_ORGANIZATION_recall": 0.9401197604790419, "eval_PERSON_f1": 0.9852941176470589, "eval_PERSON_number": 137, "eval_PERSON_precision": 0.9925925925925926, "eval_PERSON_recall": 0.9781021897810219, "eval_loss": 0.08694750815629959, "eval_overall_accuracy": 0.9883977900552486, "eval_overall_f1": 0.9523809523809523, "eval_overall_precision": 0.95, "eval_overall_recall": 0.9547738693467337, "eval_runtime": 0.2746, "eval_samples_per_second": 619.169, "eval_steps_per_second": 10.927, "step": 2784 }, { "epoch": 30.0, "grad_norm": 0.0011311785783618689, "learning_rate": 3.5e-05, "loss": 0.0023, "step": 2880 }, { "epoch": 30.0, "eval_LOCATION_f1": 0.9375000000000001, "eval_LOCATION_number": 94, "eval_LOCATION_precision": 0.9183673469387755, "eval_LOCATION_recall": 0.9574468085106383, "eval_ORGANIZATION_f1": 0.9337349397590362, "eval_ORGANIZATION_number": 167, "eval_ORGANIZATION_precision": 0.9393939393939394, "eval_ORGANIZATION_recall": 0.9281437125748503, "eval_PERSON_f1": 0.988929889298893, "eval_PERSON_number": 137, "eval_PERSON_precision": 1.0, "eval_PERSON_recall": 0.9781021897810219, "eval_loss": 0.1041567325592041, "eval_overall_accuracy": 0.9881215469613259, "eval_overall_f1": 0.9534591194968555, "eval_overall_precision": 0.9546599496221663, "eval_overall_recall": 0.9522613065326633, "eval_runtime": 0.2734, "eval_samples_per_second": 621.774, "eval_steps_per_second": 10.972, "step": 2880 }, { "epoch": 31.0, "grad_norm": 0.00352740241214633, "learning_rate": 3.45e-05, "loss": 0.0028, "step": 2976 }, { "epoch": 31.0, "eval_LOCATION_f1": 0.9387755102040817, "eval_LOCATION_number": 94, "eval_LOCATION_precision": 0.9019607843137255, "eval_LOCATION_recall": 0.9787234042553191, "eval_ORGANIZATION_f1": 0.9198813056379822, "eval_ORGANIZATION_number": 167, "eval_ORGANIZATION_precision": 0.9117647058823529, "eval_ORGANIZATION_recall": 0.9281437125748503, "eval_PERSON_f1": 0.9816849816849818, "eval_PERSON_number": 137, "eval_PERSON_precision": 0.9852941176470589, "eval_PERSON_recall": 0.9781021897810219, "eval_loss": 0.10027153044939041, "eval_overall_accuracy": 0.9861878453038674, "eval_overall_f1": 0.945409429280397, "eval_overall_precision": 0.9338235294117647, "eval_overall_recall": 0.957286432160804, "eval_runtime": 0.2893, "eval_samples_per_second": 587.582, "eval_steps_per_second": 10.369, "step": 2976 }, { "epoch": 32.0, "grad_norm": 0.004691402893513441, "learning_rate": 3.4000000000000007e-05, "loss": 0.0015, "step": 3072 }, { "epoch": 32.0, "eval_LOCATION_f1": 0.9381443298969071, "eval_LOCATION_number": 94, "eval_LOCATION_precision": 0.91, "eval_LOCATION_recall": 0.9680851063829787, "eval_ORGANIZATION_f1": 0.9436201780415431, "eval_ORGANIZATION_number": 167, "eval_ORGANIZATION_precision": 0.9352941176470588, "eval_ORGANIZATION_recall": 0.9520958083832335, "eval_PERSON_f1": 0.9816849816849818, "eval_PERSON_number": 137, "eval_PERSON_precision": 0.9852941176470589, "eval_PERSON_recall": 0.9781021897810219, "eval_loss": 0.0801883116364479, "eval_overall_accuracy": 0.988950276243094, "eval_overall_f1": 0.9552238805970149, "eval_overall_precision": 0.9458128078817734, "eval_overall_recall": 0.964824120603015, "eval_runtime": 0.2728, "eval_samples_per_second": 623.209, "eval_steps_per_second": 10.998, "step": 3072 }, { "epoch": 33.0, "grad_norm": 0.05244644358754158, "learning_rate": 3.35e-05, "loss": 0.0025, "step": 3168 }, { "epoch": 33.0, "eval_LOCATION_f1": 0.9145728643216081, "eval_LOCATION_number": 94, "eval_LOCATION_precision": 0.8666666666666667, "eval_LOCATION_recall": 0.9680851063829787, "eval_ORGANIZATION_f1": 0.9174311926605505, "eval_ORGANIZATION_number": 167, "eval_ORGANIZATION_precision": 0.9375, "eval_ORGANIZATION_recall": 0.8982035928143712, "eval_PERSON_f1": 0.988929889298893, "eval_PERSON_number": 137, "eval_PERSON_precision": 1.0, "eval_PERSON_recall": 0.9781021897810219, "eval_loss": 0.09586746245622635, "eval_overall_accuracy": 0.9861878453038674, "eval_overall_f1": 0.9410288582183186, "eval_overall_precision": 0.9398496240601504, "eval_overall_recall": 0.9422110552763819, "eval_runtime": 0.276, "eval_samples_per_second": 615.889, "eval_steps_per_second": 10.869, "step": 3168 }, { "epoch": 34.0, "grad_norm": 0.0031330811325460672, "learning_rate": 3.3e-05, "loss": 0.0014, "step": 3264 }, { "epoch": 34.0, "eval_LOCATION_f1": 0.9375000000000001, "eval_LOCATION_number": 94, "eval_LOCATION_precision": 0.9183673469387755, "eval_LOCATION_recall": 0.9574468085106383, "eval_ORGANIZATION_f1": 0.9313432835820896, "eval_ORGANIZATION_number": 167, "eval_ORGANIZATION_precision": 0.9285714285714286, "eval_ORGANIZATION_recall": 0.9341317365269461, "eval_PERSON_f1": 0.988929889298893, "eval_PERSON_number": 137, "eval_PERSON_precision": 1.0, "eval_PERSON_recall": 0.9781021897810219, "eval_loss": 0.0969545915722847, "eval_overall_accuracy": 0.9881215469613259, "eval_overall_f1": 0.9523809523809523, "eval_overall_precision": 0.95, "eval_overall_recall": 0.9547738693467337, "eval_runtime": 0.2737, "eval_samples_per_second": 621.117, "eval_steps_per_second": 10.961, "step": 3264 }, { "epoch": 35.0, "grad_norm": 0.00067303137620911, "learning_rate": 3.2500000000000004e-05, "loss": 0.0017, "step": 3360 }, { "epoch": 35.0, "eval_LOCATION_f1": 0.9518716577540107, "eval_LOCATION_number": 94, "eval_LOCATION_precision": 0.956989247311828, "eval_LOCATION_recall": 0.9468085106382979, "eval_ORGANIZATION_f1": 0.923076923076923, "eval_ORGANIZATION_number": 167, "eval_ORGANIZATION_precision": 0.9122807017543859, "eval_ORGANIZATION_recall": 0.9341317365269461, "eval_PERSON_f1": 0.9852941176470589, "eval_PERSON_number": 137, "eval_PERSON_precision": 0.9925925925925926, "eval_PERSON_recall": 0.9781021897810219, "eval_loss": 0.07898663729429245, "eval_overall_accuracy": 0.988950276243094, "eval_overall_f1": 0.9510664993726474, "eval_overall_precision": 0.949874686716792, "eval_overall_recall": 0.9522613065326633, "eval_runtime": 0.2772, "eval_samples_per_second": 613.196, "eval_steps_per_second": 10.821, "step": 3360 }, { "epoch": 36.0, "grad_norm": 0.0009424611926078796, "learning_rate": 3.2000000000000005e-05, "loss": 0.002, "step": 3456 }, { "epoch": 36.0, "eval_LOCATION_f1": 0.9333333333333335, "eval_LOCATION_number": 94, "eval_LOCATION_precision": 0.900990099009901, "eval_LOCATION_recall": 0.9680851063829787, "eval_ORGANIZATION_f1": 0.9146341463414633, "eval_ORGANIZATION_number": 167, "eval_ORGANIZATION_precision": 0.9316770186335404, "eval_ORGANIZATION_recall": 0.8982035928143712, "eval_PERSON_f1": 0.9816849816849818, "eval_PERSON_number": 137, "eval_PERSON_precision": 0.9852941176470589, "eval_PERSON_recall": 0.9781021897810219, "eval_loss": 0.09117559343576431, "eval_overall_accuracy": 0.9870165745856354, "eval_overall_f1": 0.9422110552763819, "eval_overall_precision": 0.9422110552763819, "eval_overall_recall": 0.9422110552763819, "eval_runtime": 0.2735, "eval_samples_per_second": 621.585, "eval_steps_per_second": 10.969, "step": 3456 }, { "epoch": 37.0, "grad_norm": 0.0027360552921891212, "learning_rate": 3.15e-05, "loss": 0.0025, "step": 3552 }, { "epoch": 37.0, "eval_LOCATION_f1": 0.9368421052631578, "eval_LOCATION_number": 94, "eval_LOCATION_precision": 0.9270833333333334, "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": 0.988929889298893, "eval_PERSON_number": 137, "eval_PERSON_precision": 1.0, "eval_PERSON_recall": 0.9781021897810219, "eval_loss": 0.10611312836408615, "eval_overall_accuracy": 0.9864640883977901, "eval_overall_f1": 0.9382093316519546, "eval_overall_precision": 0.9417721518987342, "eval_overall_recall": 0.9346733668341709, "eval_runtime": 0.2767, "eval_samples_per_second": 614.331, "eval_steps_per_second": 10.841, "step": 3552 }, { "epoch": 38.0, "grad_norm": 0.0012210855493322015, "learning_rate": 3.1e-05, "loss": 0.0028, "step": 3648 }, { "epoch": 38.0, "eval_LOCATION_f1": 0.9375000000000001, "eval_LOCATION_number": 94, "eval_LOCATION_precision": 0.9183673469387755, "eval_LOCATION_recall": 0.9574468085106383, "eval_ORGANIZATION_f1": 0.9003021148036255, "eval_ORGANIZATION_number": 167, "eval_ORGANIZATION_precision": 0.9085365853658537, "eval_ORGANIZATION_recall": 0.8922155688622755, "eval_PERSON_f1": 0.988929889298893, "eval_PERSON_number": 137, "eval_PERSON_precision": 1.0, "eval_PERSON_recall": 0.9781021897810219, "eval_loss": 0.09818021208047867, "eval_overall_accuracy": 0.9870165745856354, "eval_overall_f1": 0.9395465994962218, "eval_overall_precision": 0.9419191919191919, "eval_overall_recall": 0.9371859296482412, "eval_runtime": 0.273, "eval_samples_per_second": 622.607, "eval_steps_per_second": 10.987, "step": 3648 }, { "epoch": 39.0, "grad_norm": 0.0014512796187773347, "learning_rate": 3.05e-05, "loss": 0.0022, "step": 3744 }, { "epoch": 39.0, "eval_LOCATION_f1": 0.9109947643979057, "eval_LOCATION_number": 94, "eval_LOCATION_precision": 0.8969072164948454, "eval_LOCATION_recall": 0.925531914893617, "eval_ORGANIZATION_f1": 0.9085545722713864, "eval_ORGANIZATION_number": 167, "eval_ORGANIZATION_precision": 0.8953488372093024, "eval_ORGANIZATION_recall": 0.9221556886227545, "eval_PERSON_f1": 0.988929889298893, "eval_PERSON_number": 137, "eval_PERSON_precision": 1.0, "eval_PERSON_recall": 0.9781021897810219, "eval_loss": 0.10608251392841339, "eval_overall_accuracy": 0.9848066298342542, "eval_overall_f1": 0.9363295880149813, "eval_overall_precision": 0.9305210918114144, "eval_overall_recall": 0.9422110552763819, "eval_runtime": 0.2896, "eval_samples_per_second": 587.051, "eval_steps_per_second": 10.36, "step": 3744 }, { "epoch": 40.0, "grad_norm": 0.001278189243748784, "learning_rate": 3e-05, "loss": 0.0018, "step": 3840 }, { "epoch": 40.0, "eval_LOCATION_f1": 0.9166666666666666, "eval_LOCATION_number": 94, "eval_LOCATION_precision": 0.8979591836734694, "eval_LOCATION_recall": 0.9361702127659575, "eval_ORGANIZATION_f1": 0.9090909090909091, "eval_ORGANIZATION_number": 167, "eval_ORGANIZATION_precision": 0.9202453987730062, "eval_ORGANIZATION_recall": 0.8982035928143712, "eval_PERSON_f1": 0.988929889298893, "eval_PERSON_number": 137, "eval_PERSON_precision": 1.0, "eval_PERSON_recall": 0.9781021897810219, "eval_loss": 0.10771890729665756, "eval_overall_accuracy": 0.9861878453038674, "eval_overall_f1": 0.9382093316519546, "eval_overall_precision": 0.9417721518987342, "eval_overall_recall": 0.9346733668341709, "eval_runtime": 0.2777, "eval_samples_per_second": 612.211, "eval_steps_per_second": 10.804, "step": 3840 }, { "epoch": 41.0, "grad_norm": 0.008502389304339886, "learning_rate": 2.95e-05, "loss": 0.002, "step": 3936 }, { "epoch": 41.0, "eval_LOCATION_f1": 0.9166666666666666, "eval_LOCATION_number": 94, "eval_LOCATION_precision": 0.8979591836734694, "eval_LOCATION_recall": 0.9361702127659575, "eval_ORGANIZATION_f1": 0.9212121212121211, "eval_ORGANIZATION_number": 167, "eval_ORGANIZATION_precision": 0.9325153374233128, "eval_ORGANIZATION_recall": 0.9101796407185628, "eval_PERSON_f1": 0.988929889298893, "eval_PERSON_number": 137, "eval_PERSON_precision": 1.0, "eval_PERSON_recall": 0.9781021897810219, "eval_loss": 0.09228293597698212, "eval_overall_accuracy": 0.9870165745856354, "eval_overall_f1": 0.9432534678436318, "eval_overall_precision": 0.9468354430379747, "eval_overall_recall": 0.9396984924623115, "eval_runtime": 0.2844, "eval_samples_per_second": 597.692, "eval_steps_per_second": 10.547, "step": 3936 }, { "epoch": 42.0, "grad_norm": 0.023928137496113777, "learning_rate": 2.9e-05, "loss": 0.003, "step": 4032 }, { "epoch": 42.0, "eval_LOCATION_f1": 0.91005291005291, "eval_LOCATION_number": 94, "eval_LOCATION_precision": 0.9052631578947369, "eval_LOCATION_recall": 0.9148936170212766, "eval_ORGANIZATION_f1": 0.9166666666666667, "eval_ORGANIZATION_number": 167, "eval_ORGANIZATION_precision": 0.9112426035502958, "eval_ORGANIZATION_recall": 0.9221556886227545, "eval_PERSON_f1": 0.9816849816849818, "eval_PERSON_number": 137, "eval_PERSON_precision": 0.9852941176470589, "eval_PERSON_recall": 0.9781021897810219, "eval_loss": 0.08988003432750702, "eval_overall_accuracy": 0.9861878453038674, "eval_overall_f1": 0.93734335839599, "eval_overall_precision": 0.935, "eval_overall_recall": 0.9396984924623115, "eval_runtime": 0.2777, "eval_samples_per_second": 612.18, "eval_steps_per_second": 10.803, "step": 4032 }, { "epoch": 43.0, "grad_norm": 0.051936984062194824, "learning_rate": 2.8499999999999998e-05, "loss": 0.0027, "step": 4128 }, { "epoch": 43.0, "eval_LOCATION_f1": 0.9304812834224598, "eval_LOCATION_number": 94, "eval_LOCATION_precision": 0.9354838709677419, "eval_LOCATION_recall": 0.925531914893617, "eval_ORGANIZATION_f1": 0.924924924924925, "eval_ORGANIZATION_number": 167, "eval_ORGANIZATION_precision": 0.927710843373494, "eval_ORGANIZATION_recall": 0.9221556886227545, "eval_PERSON_f1": 0.988929889298893, "eval_PERSON_number": 137, "eval_PERSON_precision": 1.0, "eval_PERSON_recall": 0.9781021897810219, "eval_loss": 0.08266886323690414, "eval_overall_accuracy": 0.9878453038674033, "eval_overall_f1": 0.9481668773704173, "eval_overall_precision": 0.9541984732824428, "eval_overall_recall": 0.9422110552763819, "eval_runtime": 0.2785, "eval_samples_per_second": 610.397, "eval_steps_per_second": 10.772, "step": 4128 }, { "epoch": 44.0, "grad_norm": 0.0016257220413535833, "learning_rate": 2.8000000000000003e-05, "loss": 0.0015, "step": 4224 }, { "epoch": 44.0, "eval_LOCATION_f1": 0.9148936170212766, "eval_LOCATION_number": 94, "eval_LOCATION_precision": 0.9148936170212766, "eval_LOCATION_recall": 0.9148936170212766, "eval_ORGANIZATION_f1": 0.9101796407185628, "eval_ORGANIZATION_number": 167, "eval_ORGANIZATION_precision": 0.9101796407185628, "eval_ORGANIZATION_recall": 0.9101796407185628, "eval_PERSON_f1": 0.988929889298893, "eval_PERSON_number": 137, "eval_PERSON_precision": 1.0, "eval_PERSON_recall": 0.9781021897810219, "eval_loss": 0.07984667271375656, "eval_overall_accuracy": 0.9878453038674033, "eval_overall_f1": 0.9382093316519546, "eval_overall_precision": 0.9417721518987342, "eval_overall_recall": 0.9346733668341709, "eval_runtime": 0.2741, "eval_samples_per_second": 620.147, "eval_steps_per_second": 10.944, "step": 4224 }, { "epoch": 45.0, "grad_norm": 0.0029611666686832905, "learning_rate": 2.7500000000000004e-05, "loss": 0.0011, "step": 4320 }, { "epoch": 45.0, "eval_LOCATION_f1": 0.9052631578947369, "eval_LOCATION_number": 94, "eval_LOCATION_precision": 0.8958333333333334, "eval_LOCATION_recall": 0.9148936170212766, "eval_ORGANIZATION_f1": 0.9113149847094801, "eval_ORGANIZATION_number": 167, "eval_ORGANIZATION_precision": 0.93125, "eval_ORGANIZATION_recall": 0.8922155688622755, "eval_PERSON_f1": 0.9816849816849818, "eval_PERSON_number": 137, "eval_PERSON_precision": 0.9852941176470589, "eval_PERSON_recall": 0.9781021897810219, "eval_loss": 0.08682180196046829, "eval_overall_accuracy": 0.9881215469613259, "eval_overall_f1": 0.9341772151898734, "eval_overall_precision": 0.9413265306122449, "eval_overall_recall": 0.9271356783919598, "eval_runtime": 0.2856, "eval_samples_per_second": 595.304, "eval_steps_per_second": 10.505, "step": 4320 }, { "epoch": 46.0, "grad_norm": 0.0017064587445929646, "learning_rate": 2.7000000000000002e-05, "loss": 0.0012, "step": 4416 }, { "epoch": 46.0, "eval_LOCATION_f1": 0.9285714285714286, "eval_LOCATION_number": 94, "eval_LOCATION_precision": 0.8921568627450981, "eval_LOCATION_recall": 0.9680851063829787, "eval_ORGANIZATION_f1": 0.934984520123839, "eval_ORGANIZATION_number": 167, "eval_ORGANIZATION_precision": 0.967948717948718, "eval_ORGANIZATION_recall": 0.9041916167664671, "eval_PERSON_f1": 0.9779411764705882, "eval_PERSON_number": 137, "eval_PERSON_precision": 0.9851851851851852, "eval_PERSON_recall": 0.9708029197080292, "eval_loss": 0.07430663704872131, "eval_overall_accuracy": 0.9903314917127072, "eval_overall_f1": 0.9481668773704173, "eval_overall_precision": 0.9541984732824428, "eval_overall_recall": 0.9422110552763819, "eval_runtime": 0.2721, "eval_samples_per_second": 624.819, "eval_steps_per_second": 11.026, "step": 4416 }, { "epoch": 47.0, "grad_norm": 0.253894567489624, "learning_rate": 2.6500000000000004e-05, "loss": 0.0012, "step": 4512 }, { "epoch": 47.0, "eval_LOCATION_f1": 0.9214659685863875, "eval_LOCATION_number": 94, "eval_LOCATION_precision": 0.9072164948453608, "eval_LOCATION_recall": 0.9361702127659575, "eval_ORGANIZATION_f1": 0.9174311926605505, "eval_ORGANIZATION_number": 167, "eval_ORGANIZATION_precision": 0.9375, "eval_ORGANIZATION_recall": 0.8982035928143712, "eval_PERSON_f1": 0.9816849816849818, "eval_PERSON_number": 137, "eval_PERSON_precision": 0.9852941176470589, "eval_PERSON_recall": 0.9781021897810219, "eval_loss": 0.08702202886343002, "eval_overall_accuracy": 0.9883977900552486, "eval_overall_f1": 0.9405815423514539, "eval_overall_precision": 0.9465648854961832, "eval_overall_recall": 0.9346733668341709, "eval_runtime": 0.2936, "eval_samples_per_second": 579.062, "eval_steps_per_second": 10.219, "step": 4512 }, { "epoch": 48.0, "grad_norm": 0.006299526896327734, "learning_rate": 2.6000000000000002e-05, "loss": 0.0019, "step": 4608 }, { "epoch": 48.0, "eval_LOCATION_f1": 0.9175257731958764, "eval_LOCATION_number": 94, "eval_LOCATION_precision": 0.89, "eval_LOCATION_recall": 0.9468085106382979, "eval_ORGANIZATION_f1": 0.9079754601226994, "eval_ORGANIZATION_number": 167, "eval_ORGANIZATION_precision": 0.9308176100628931, "eval_ORGANIZATION_recall": 0.8862275449101796, "eval_PERSON_f1": 0.9743589743589743, "eval_PERSON_number": 137, "eval_PERSON_precision": 0.9779411764705882, "eval_PERSON_recall": 0.9708029197080292, "eval_loss": 0.07586066424846649, "eval_overall_accuracy": 0.9881215469613259, "eval_overall_f1": 0.9331651954602774, "eval_overall_precision": 0.9367088607594937, "eval_overall_recall": 0.9296482412060302, "eval_runtime": 0.2729, "eval_samples_per_second": 622.883, "eval_steps_per_second": 10.992, "step": 4608 }, { "epoch": 49.0, "grad_norm": 0.0033371802419424057, "learning_rate": 2.5500000000000003e-05, "loss": 0.0015, "step": 4704 }, { "epoch": 49.0, "eval_LOCATION_f1": 0.9368421052631578, "eval_LOCATION_number": 94, "eval_LOCATION_precision": 0.9270833333333334, "eval_LOCATION_recall": 0.9468085106382979, "eval_ORGANIZATION_f1": 0.9258160237388723, "eval_ORGANIZATION_number": 167, "eval_ORGANIZATION_precision": 0.9176470588235294, "eval_ORGANIZATION_recall": 0.9341317365269461, "eval_PERSON_f1": 0.988929889298893, "eval_PERSON_number": 137, "eval_PERSON_precision": 1.0, "eval_PERSON_recall": 0.9781021897810219, "eval_loss": 0.08100084215402603, "eval_overall_accuracy": 0.9895027624309393, "eval_overall_f1": 0.9498746867167919, "eval_overall_precision": 0.9475, "eval_overall_recall": 0.9522613065326633, "eval_runtime": 0.2754, "eval_samples_per_second": 617.325, "eval_steps_per_second": 10.894, "step": 4704 }, { "epoch": 50.0, "grad_norm": 0.0009251247392967343, "learning_rate": 2.5e-05, "loss": 0.0011, "step": 4800 }, { "epoch": 50.0, "eval_LOCATION_f1": 0.9270833333333334, "eval_LOCATION_number": 94, "eval_LOCATION_precision": 0.9081632653061225, "eval_LOCATION_recall": 0.9468085106382979, "eval_ORGANIZATION_f1": 0.9361702127659574, "eval_ORGANIZATION_number": 167, "eval_ORGANIZATION_precision": 0.9506172839506173, "eval_ORGANIZATION_recall": 0.9221556886227545, "eval_PERSON_f1": 0.9816849816849818, "eval_PERSON_number": 137, "eval_PERSON_precision": 0.9852941176470589, "eval_PERSON_recall": 0.9781021897810219, "eval_loss": 0.08902022242546082, "eval_overall_accuracy": 0.988950276243094, "eval_overall_f1": 0.9496221662468514, "eval_overall_precision": 0.952020202020202, "eval_overall_recall": 0.9472361809045227, "eval_runtime": 0.2757, "eval_samples_per_second": 616.53, "eval_steps_per_second": 10.88, "step": 4800 }, { "epoch": 51.0, "grad_norm": 0.012115373276174068, "learning_rate": 2.45e-05, "loss": 0.0007, "step": 4896 }, { "epoch": 51.0, "eval_LOCATION_f1": 0.9263157894736843, "eval_LOCATION_number": 94, "eval_LOCATION_precision": 0.9166666666666666, "eval_LOCATION_recall": 0.9361702127659575, "eval_ORGANIZATION_f1": 0.9341317365269461, "eval_ORGANIZATION_number": 167, "eval_ORGANIZATION_precision": 0.9341317365269461, "eval_ORGANIZATION_recall": 0.9341317365269461, "eval_PERSON_f1": 0.9816849816849818, "eval_PERSON_number": 137, "eval_PERSON_precision": 0.9852941176470589, "eval_PERSON_recall": 0.9781021897810219, "eval_loss": 0.08269887417554855, "eval_overall_accuracy": 0.9895027624309393, "eval_overall_f1": 0.9485570890840652, "eval_overall_precision": 0.9473684210526315, "eval_overall_recall": 0.949748743718593, "eval_runtime": 0.2748, "eval_samples_per_second": 618.534, "eval_steps_per_second": 10.915, "step": 4896 }, { "epoch": 52.0, "grad_norm": 0.0006735413917340338, "learning_rate": 2.4e-05, "loss": 0.001, "step": 4992 }, { "epoch": 52.0, "eval_LOCATION_f1": 0.9166666666666666, "eval_LOCATION_number": 94, "eval_LOCATION_precision": 0.8979591836734694, "eval_LOCATION_recall": 0.9361702127659575, "eval_ORGANIZATION_f1": 0.9281437125748502, "eval_ORGANIZATION_number": 167, "eval_ORGANIZATION_precision": 0.9281437125748503, "eval_ORGANIZATION_recall": 0.9281437125748503, "eval_PERSON_f1": 0.9852941176470589, "eval_PERSON_number": 137, "eval_PERSON_precision": 0.9925925925925926, "eval_PERSON_recall": 0.9781021897810219, "eval_loss": 0.08734384179115295, "eval_overall_accuracy": 0.9886740331491712, "eval_overall_f1": 0.9448621553884713, "eval_overall_precision": 0.9425, "eval_overall_recall": 0.9472361809045227, "eval_runtime": 0.2733, "eval_samples_per_second": 622.115, "eval_steps_per_second": 10.979, "step": 4992 }, { "epoch": 53.0, "grad_norm": 0.003732080804184079, "learning_rate": 2.35e-05, "loss": 0.001, "step": 5088 }, { "epoch": 53.0, "eval_LOCATION_f1": 0.9166666666666666, "eval_LOCATION_number": 94, "eval_LOCATION_precision": 0.8979591836734694, "eval_LOCATION_recall": 0.9361702127659575, "eval_ORGANIZATION_f1": 0.9337349397590362, "eval_ORGANIZATION_number": 167, "eval_ORGANIZATION_precision": 0.9393939393939394, "eval_ORGANIZATION_recall": 0.9281437125748503, "eval_PERSON_f1": 0.9779411764705882, "eval_PERSON_number": 137, "eval_PERSON_precision": 0.9851851851851852, "eval_PERSON_recall": 0.9708029197080292, "eval_loss": 0.08197535574436188, "eval_overall_accuracy": 0.988950276243094, "eval_overall_f1": 0.9447236180904522, "eval_overall_precision": 0.9447236180904522, "eval_overall_recall": 0.9447236180904522, "eval_runtime": 0.2828, "eval_samples_per_second": 601.1, "eval_steps_per_second": 10.608, "step": 5088 }, { "epoch": 54.0, "grad_norm": 0.00532448198646307, "learning_rate": 2.3000000000000003e-05, "loss": 0.0004, "step": 5184 }, { "epoch": 54.0, "eval_LOCATION_f1": 0.9230769230769231, "eval_LOCATION_number": 94, "eval_LOCATION_precision": 0.8910891089108911, "eval_LOCATION_recall": 0.9574468085106383, "eval_ORGANIZATION_f1": 0.920245398773006, "eval_ORGANIZATION_number": 167, "eval_ORGANIZATION_precision": 0.9433962264150944, "eval_ORGANIZATION_recall": 0.8982035928143712, "eval_PERSON_f1": 0.9816849816849818, "eval_PERSON_number": 137, "eval_PERSON_precision": 0.9852941176470589, "eval_PERSON_recall": 0.9781021897810219, "eval_loss": 0.0916755199432373, "eval_overall_accuracy": 0.9867403314917127, "eval_overall_f1": 0.9420654911838791, "eval_overall_precision": 0.9444444444444444, "eval_overall_recall": 0.9396984924623115, "eval_runtime": 0.2734, "eval_samples_per_second": 621.692, "eval_steps_per_second": 10.971, "step": 5184 }, { "epoch": 55.0, "grad_norm": 0.6883693933486938, "learning_rate": 2.25e-05, "loss": 0.0006, "step": 5280 }, { "epoch": 55.0, "eval_LOCATION_f1": 0.9166666666666666, "eval_LOCATION_number": 94, "eval_LOCATION_precision": 0.8979591836734694, "eval_LOCATION_recall": 0.9361702127659575, "eval_ORGANIZATION_f1": 0.927710843373494, "eval_ORGANIZATION_number": 167, "eval_ORGANIZATION_precision": 0.9333333333333333, "eval_ORGANIZATION_recall": 0.9221556886227545, "eval_PERSON_f1": 0.9852941176470589, "eval_PERSON_number": 137, "eval_PERSON_precision": 0.9925925925925926, "eval_PERSON_recall": 0.9781021897810219, "eval_loss": 0.10530148446559906, "eval_overall_accuracy": 0.9883977900552486, "eval_overall_f1": 0.9447236180904522, "eval_overall_precision": 0.9447236180904522, "eval_overall_recall": 0.9447236180904522, "eval_runtime": 0.2767, "eval_samples_per_second": 614.365, "eval_steps_per_second": 10.842, "step": 5280 }, { "epoch": 56.0, "grad_norm": 0.0025715897791087627, "learning_rate": 2.2000000000000003e-05, "loss": 0.001, "step": 5376 }, { "epoch": 56.0, "eval_LOCATION_f1": 0.9222797927461138, "eval_LOCATION_number": 94, "eval_LOCATION_precision": 0.898989898989899, "eval_LOCATION_recall": 0.9468085106382979, "eval_ORGANIZATION_f1": 0.927710843373494, "eval_ORGANIZATION_number": 167, "eval_ORGANIZATION_precision": 0.9333333333333333, "eval_ORGANIZATION_recall": 0.9221556886227545, "eval_PERSON_f1": 0.9816849816849818, "eval_PERSON_number": 137, "eval_PERSON_precision": 0.9852941176470589, "eval_PERSON_recall": 0.9781021897810219, "eval_loss": 0.10397301614284515, "eval_overall_accuracy": 0.9881215469613259, "eval_overall_f1": 0.9448621553884713, "eval_overall_precision": 0.9425, "eval_overall_recall": 0.9472361809045227, "eval_runtime": 0.2773, "eval_samples_per_second": 613.09, "eval_steps_per_second": 10.819, "step": 5376 }, { "epoch": 57.0, "grad_norm": 0.000477120338473469, "learning_rate": 2.15e-05, "loss": 0.0005, "step": 5472 }, { "epoch": 57.0, "eval_LOCATION_f1": 0.9222797927461138, "eval_LOCATION_number": 94, "eval_LOCATION_precision": 0.898989898989899, "eval_LOCATION_recall": 0.9468085106382979, "eval_ORGANIZATION_f1": 0.9309309309309309, "eval_ORGANIZATION_number": 167, "eval_ORGANIZATION_precision": 0.9337349397590361, "eval_ORGANIZATION_recall": 0.9281437125748503, "eval_PERSON_f1": 0.9852941176470589, "eval_PERSON_number": 137, "eval_PERSON_precision": 0.9925925925925926, "eval_PERSON_recall": 0.9781021897810219, "eval_loss": 0.10418053716421127, "eval_overall_accuracy": 0.9883977900552486, "eval_overall_f1": 0.9473684210526316, "eval_overall_precision": 0.945, "eval_overall_recall": 0.949748743718593, "eval_runtime": 0.2948, "eval_samples_per_second": 576.643, "eval_steps_per_second": 10.176, "step": 5472 }, { "epoch": 58.0, "grad_norm": 0.002485807752236724, "learning_rate": 2.1e-05, "loss": 0.0009, "step": 5568 }, { "epoch": 58.0, "eval_LOCATION_f1": 0.9270833333333334, "eval_LOCATION_number": 94, "eval_LOCATION_precision": 0.9081632653061225, "eval_LOCATION_recall": 0.9468085106382979, "eval_ORGANIZATION_f1": 0.9090909090909091, "eval_ORGANIZATION_number": 167, "eval_ORGANIZATION_precision": 0.9202453987730062, "eval_ORGANIZATION_recall": 0.8982035928143712, "eval_PERSON_f1": 0.9816849816849818, "eval_PERSON_number": 137, "eval_PERSON_precision": 0.9852941176470589, "eval_PERSON_recall": 0.9781021897810219, "eval_loss": 0.10568208247423172, "eval_overall_accuracy": 0.9875690607734806, "eval_overall_f1": 0.9383647798742139, "eval_overall_precision": 0.9395465994962217, "eval_overall_recall": 0.9371859296482412, "eval_runtime": 0.2734, "eval_samples_per_second": 621.777, "eval_steps_per_second": 10.973, "step": 5568 }, { "epoch": 59.0, "grad_norm": 0.0032608220353722572, "learning_rate": 2.05e-05, "loss": 0.001, "step": 5664 }, { "epoch": 59.0, "eval_LOCATION_f1": 0.9230769230769231, "eval_LOCATION_number": 94, "eval_LOCATION_precision": 0.8910891089108911, "eval_LOCATION_recall": 0.9574468085106383, "eval_ORGANIZATION_f1": 0.924924924924925, "eval_ORGANIZATION_number": 167, "eval_ORGANIZATION_precision": 0.927710843373494, "eval_ORGANIZATION_recall": 0.9221556886227545, "eval_PERSON_f1": 0.988929889298893, "eval_PERSON_number": 137, "eval_PERSON_precision": 1.0, "eval_PERSON_recall": 0.9781021897810219, "eval_loss": 0.10343037545681, "eval_overall_accuracy": 0.987292817679558, "eval_overall_f1": 0.9461827284105131, "eval_overall_precision": 0.942643391521197, "eval_overall_recall": 0.949748743718593, "eval_runtime": 0.2778, "eval_samples_per_second": 611.903, "eval_steps_per_second": 10.798, "step": 5664 }, { "epoch": 60.0, "grad_norm": 5.73241662979126, "learning_rate": 2e-05, "loss": 0.0012, "step": 5760 }, { "epoch": 60.0, "eval_LOCATION_f1": 0.9214659685863875, "eval_LOCATION_number": 94, "eval_LOCATION_precision": 0.9072164948453608, "eval_LOCATION_recall": 0.9361702127659575, "eval_ORGANIZATION_f1": 0.9309309309309309, "eval_ORGANIZATION_number": 167, "eval_ORGANIZATION_precision": 0.9337349397590361, "eval_ORGANIZATION_recall": 0.9281437125748503, "eval_PERSON_f1": 0.9743589743589743, "eval_PERSON_number": 137, "eval_PERSON_precision": 0.9779411764705882, "eval_PERSON_recall": 0.9708029197080292, "eval_loss": 0.09098115563392639, "eval_overall_accuracy": 0.9886740331491712, "eval_overall_f1": 0.9435382685069008, "eval_overall_precision": 0.9423558897243107, "eval_overall_recall": 0.9447236180904522, "eval_runtime": 0.2731, "eval_samples_per_second": 622.561, "eval_steps_per_second": 10.986, "step": 5760 }, { "epoch": 61.0, "grad_norm": 0.0008896036306396127, "learning_rate": 1.9500000000000003e-05, "loss": 0.0008, "step": 5856 }, { "epoch": 61.0, "eval_LOCATION_f1": 0.9197860962566845, "eval_LOCATION_number": 94, "eval_LOCATION_precision": 0.9247311827956989, "eval_LOCATION_recall": 0.9148936170212766, "eval_ORGANIZATION_f1": 0.9101796407185628, "eval_ORGANIZATION_number": 167, "eval_ORGANIZATION_precision": 0.9101796407185628, "eval_ORGANIZATION_recall": 0.9101796407185628, "eval_PERSON_f1": 0.9743589743589743, "eval_PERSON_number": 137, "eval_PERSON_precision": 0.9779411764705882, "eval_PERSON_recall": 0.9708029197080292, "eval_loss": 0.09867171943187714, "eval_overall_accuracy": 0.9861878453038674, "eval_overall_f1": 0.9345088161209069, "eval_overall_precision": 0.9368686868686869, "eval_overall_recall": 0.9321608040201005, "eval_runtime": 0.2763, "eval_samples_per_second": 615.351, "eval_steps_per_second": 10.859, "step": 5856 }, { "epoch": 62.0, "grad_norm": 0.00039030349580571055, "learning_rate": 1.9e-05, "loss": 0.0005, "step": 5952 }, { "epoch": 62.0, "eval_LOCATION_f1": 0.9119170984455959, "eval_LOCATION_number": 94, "eval_LOCATION_precision": 0.8888888888888888, "eval_LOCATION_recall": 0.9361702127659575, "eval_ORGANIZATION_f1": 0.9272727272727274, "eval_ORGANIZATION_number": 167, "eval_ORGANIZATION_precision": 0.9386503067484663, "eval_ORGANIZATION_recall": 0.9161676646706587, "eval_PERSON_f1": 0.988929889298893, "eval_PERSON_number": 137, "eval_PERSON_precision": 1.0, "eval_PERSON_recall": 0.9781021897810219, "eval_loss": 0.10562293231487274, "eval_overall_accuracy": 0.9875690607734806, "eval_overall_f1": 0.9445843828715367, "eval_overall_precision": 0.946969696969697, "eval_overall_recall": 0.9422110552763819, "eval_runtime": 0.2721, "eval_samples_per_second": 624.673, "eval_steps_per_second": 11.024, "step": 5952 }, { "epoch": 63.0, "grad_norm": 0.0003558107418939471, "learning_rate": 1.85e-05, "loss": 0.0006, "step": 6048 }, { "epoch": 63.0, "eval_LOCATION_f1": 0.9166666666666666, "eval_LOCATION_number": 94, "eval_LOCATION_precision": 0.8979591836734694, "eval_LOCATION_recall": 0.9361702127659575, "eval_ORGANIZATION_f1": 0.918429003021148, "eval_ORGANIZATION_number": 167, "eval_ORGANIZATION_precision": 0.926829268292683, "eval_ORGANIZATION_recall": 0.9101796407185628, "eval_PERSON_f1": 0.9852941176470589, "eval_PERSON_number": 137, "eval_PERSON_precision": 0.9925925925925926, "eval_PERSON_recall": 0.9781021897810219, "eval_loss": 0.10495664179325104, "eval_overall_accuracy": 0.987292817679558, "eval_overall_f1": 0.9408805031446541, "eval_overall_precision": 0.9420654911838791, "eval_overall_recall": 0.9396984924623115, "eval_runtime": 0.2764, "eval_samples_per_second": 614.992, "eval_steps_per_second": 10.853, "step": 6048 }, { "epoch": 64.0, "grad_norm": 0.00027060168213211, "learning_rate": 1.8e-05, "loss": 0.0013, "step": 6144 }, { "epoch": 64.0, "eval_LOCATION_f1": 0.9214659685863875, "eval_LOCATION_number": 94, "eval_LOCATION_precision": 0.9072164948453608, "eval_LOCATION_recall": 0.9361702127659575, "eval_ORGANIZATION_f1": 0.9244712990936556, "eval_ORGANIZATION_number": 167, "eval_ORGANIZATION_precision": 0.9329268292682927, "eval_ORGANIZATION_recall": 0.9161676646706587, "eval_PERSON_f1": 0.988929889298893, "eval_PERSON_number": 137, "eval_PERSON_precision": 1.0, "eval_PERSON_recall": 0.9781021897810219, "eval_loss": 0.0956372618675232, "eval_overall_accuracy": 0.9883977900552486, "eval_overall_f1": 0.9457755359394704, "eval_overall_precision": 0.9493670886075949, "eval_overall_recall": 0.9422110552763819, "eval_runtime": 0.2714, "eval_samples_per_second": 626.464, "eval_steps_per_second": 11.055, "step": 6144 }, { "epoch": 65.0, "grad_norm": 0.0006868442869745195, "learning_rate": 1.75e-05, "loss": 0.0006, "step": 6240 }, { "epoch": 65.0, "eval_LOCATION_f1": 0.9270833333333334, "eval_LOCATION_number": 94, "eval_LOCATION_precision": 0.9081632653061225, "eval_LOCATION_recall": 0.9468085106382979, "eval_ORGANIZATION_f1": 0.9113149847094801, "eval_ORGANIZATION_number": 167, "eval_ORGANIZATION_precision": 0.93125, "eval_ORGANIZATION_recall": 0.8922155688622755, "eval_PERSON_f1": 0.988929889298893, "eval_PERSON_number": 137, "eval_PERSON_precision": 1.0, "eval_PERSON_recall": 0.9781021897810219, "eval_loss": 0.10608125478029251, "eval_overall_accuracy": 0.9853591160220995, "eval_overall_f1": 0.9417721518987343, "eval_overall_precision": 0.9489795918367347, "eval_overall_recall": 0.9346733668341709, "eval_runtime": 0.2751, "eval_samples_per_second": 617.909, "eval_steps_per_second": 10.904, "step": 6240 }, { "epoch": 66.0, "grad_norm": 0.00017280179599765688, "learning_rate": 1.7000000000000003e-05, "loss": 0.0008, "step": 6336 }, { "epoch": 66.0, "eval_LOCATION_f1": 0.9166666666666666, "eval_LOCATION_number": 94, "eval_LOCATION_precision": 0.8979591836734694, "eval_LOCATION_recall": 0.9361702127659575, "eval_ORGANIZATION_f1": 0.9212121212121211, "eval_ORGANIZATION_number": 167, "eval_ORGANIZATION_precision": 0.9325153374233128, "eval_ORGANIZATION_recall": 0.9101796407185628, "eval_PERSON_f1": 0.9852941176470589, "eval_PERSON_number": 137, "eval_PERSON_precision": 0.9925925925925926, "eval_PERSON_recall": 0.9781021897810219, "eval_loss": 0.10315535962581635, "eval_overall_accuracy": 0.9881215469613259, "eval_overall_f1": 0.9420654911838791, "eval_overall_precision": 0.9444444444444444, "eval_overall_recall": 0.9396984924623115, "eval_runtime": 0.2837, "eval_samples_per_second": 599.243, "eval_steps_per_second": 10.575, "step": 6336 }, { "epoch": 67.0, "grad_norm": 0.0002649941889103502, "learning_rate": 1.65e-05, "loss": 0.0004, "step": 6432 }, { "epoch": 67.0, "eval_LOCATION_f1": 0.9166666666666666, "eval_LOCATION_number": 94, "eval_LOCATION_precision": 0.8979591836734694, "eval_LOCATION_recall": 0.9361702127659575, "eval_ORGANIZATION_f1": 0.9216867469879517, "eval_ORGANIZATION_number": 167, "eval_ORGANIZATION_precision": 0.9272727272727272, "eval_ORGANIZATION_recall": 0.9161676646706587, "eval_PERSON_f1": 0.988929889298893, "eval_PERSON_number": 137, "eval_PERSON_precision": 1.0, "eval_PERSON_recall": 0.9781021897810219, "eval_loss": 0.09613554179668427, "eval_overall_accuracy": 0.988950276243094, "eval_overall_f1": 0.9433962264150944, "eval_overall_precision": 0.9445843828715366, "eval_overall_recall": 0.9422110552763819, "eval_runtime": 0.2727, "eval_samples_per_second": 623.354, "eval_steps_per_second": 11.0, "step": 6432 }, { "epoch": 68.0, "grad_norm": 0.00015600069309584796, "learning_rate": 1.6000000000000003e-05, "loss": 0.0008, "step": 6528 }, { "epoch": 68.0, "eval_LOCATION_f1": 0.9072164948453608, "eval_LOCATION_number": 94, "eval_LOCATION_precision": 0.88, "eval_LOCATION_recall": 0.9361702127659575, "eval_ORGANIZATION_f1": 0.9051987767584098, "eval_ORGANIZATION_number": 167, "eval_ORGANIZATION_precision": 0.925, "eval_ORGANIZATION_recall": 0.8862275449101796, "eval_PERSON_f1": 0.9852941176470589, "eval_PERSON_number": 137, "eval_PERSON_precision": 0.9925925925925926, "eval_PERSON_recall": 0.9781021897810219, "eval_loss": 0.09787864983081818, "eval_overall_accuracy": 0.9870165745856354, "eval_overall_f1": 0.9331651954602774, "eval_overall_precision": 0.9367088607594937, "eval_overall_recall": 0.9296482412060302, "eval_runtime": 0.2811, "eval_samples_per_second": 604.852, "eval_steps_per_second": 10.674, "step": 6528 }, { "epoch": 69.0, "grad_norm": 0.0005482492269948125, "learning_rate": 1.55e-05, "loss": 0.0013, "step": 6624 }, { "epoch": 69.0, "eval_LOCATION_f1": 0.9175257731958764, "eval_LOCATION_number": 94, "eval_LOCATION_precision": 0.89, "eval_LOCATION_recall": 0.9468085106382979, "eval_ORGANIZATION_f1": 0.9161676646706587, "eval_ORGANIZATION_number": 167, "eval_ORGANIZATION_precision": 0.9161676646706587, "eval_ORGANIZATION_recall": 0.9161676646706587, "eval_PERSON_f1": 0.988929889298893, "eval_PERSON_number": 137, "eval_PERSON_precision": 1.0, "eval_PERSON_recall": 0.9781021897810219, "eval_loss": 0.10209070146083832, "eval_overall_accuracy": 0.9870165745856354, "eval_overall_f1": 0.9411764705882354, "eval_overall_precision": 0.9376558603491272, "eval_overall_recall": 0.9447236180904522, "eval_runtime": 0.2743, "eval_samples_per_second": 619.718, "eval_steps_per_second": 10.936, "step": 6624 }, { "epoch": 70.0, "grad_norm": 0.0002447032311465591, "learning_rate": 1.5e-05, "loss": 0.0004, "step": 6720 }, { "epoch": 70.0, "eval_LOCATION_f1": 0.9072164948453608, "eval_LOCATION_number": 94, "eval_LOCATION_precision": 0.88, "eval_LOCATION_recall": 0.9361702127659575, "eval_ORGANIZATION_f1": 0.9151515151515152, "eval_ORGANIZATION_number": 167, "eval_ORGANIZATION_precision": 0.9263803680981595, "eval_ORGANIZATION_recall": 0.9041916167664671, "eval_PERSON_f1": 0.988929889298893, "eval_PERSON_number": 137, "eval_PERSON_precision": 1.0, "eval_PERSON_recall": 0.9781021897810219, "eval_loss": 0.09332281351089478, "eval_overall_accuracy": 0.9881215469613259, "eval_overall_f1": 0.9383647798742139, "eval_overall_precision": 0.9395465994962217, "eval_overall_recall": 0.9371859296482412, "eval_runtime": 0.2721, "eval_samples_per_second": 624.759, "eval_steps_per_second": 11.025, "step": 6720 }, { "epoch": 71.0, "grad_norm": 0.0005908278399147093, "learning_rate": 1.45e-05, "loss": 0.001, "step": 6816 }, { "epoch": 71.0, "eval_LOCATION_f1": 0.9015544041450778, "eval_LOCATION_number": 94, "eval_LOCATION_precision": 0.8787878787878788, "eval_LOCATION_recall": 0.925531914893617, "eval_ORGANIZATION_f1": 0.9151515151515152, "eval_ORGANIZATION_number": 167, "eval_ORGANIZATION_precision": 0.9263803680981595, "eval_ORGANIZATION_recall": 0.9041916167664671, "eval_PERSON_f1": 0.9779411764705882, "eval_PERSON_number": 137, "eval_PERSON_precision": 0.9851851851851852, "eval_PERSON_recall": 0.9708029197080292, "eval_loss": 0.0892067402601242, "eval_overall_accuracy": 0.9881215469613259, "eval_overall_f1": 0.9333333333333332, "eval_overall_precision": 0.9345088161209067, "eval_overall_recall": 0.9321608040201005, "eval_runtime": 0.2723, "eval_samples_per_second": 624.326, "eval_steps_per_second": 11.018, "step": 6816 }, { "epoch": 72.0, "grad_norm": 0.0008771267021074891, "learning_rate": 1.4000000000000001e-05, "loss": 0.0006, "step": 6912 }, { "epoch": 72.0, "eval_LOCATION_f1": 0.9326424870466321, "eval_LOCATION_number": 94, "eval_LOCATION_precision": 0.9090909090909091, "eval_LOCATION_recall": 0.9574468085106383, "eval_ORGANIZATION_f1": 0.9393939393939393, "eval_ORGANIZATION_number": 167, "eval_ORGANIZATION_precision": 0.950920245398773, "eval_ORGANIZATION_recall": 0.9281437125748503, "eval_PERSON_f1": 0.9852941176470589, "eval_PERSON_number": 137, "eval_PERSON_precision": 0.9925925925925926, "eval_PERSON_recall": 0.9781021897810219, "eval_loss": 0.09660927206277847, "eval_overall_accuracy": 0.9892265193370166, "eval_overall_f1": 0.9534591194968555, "eval_overall_precision": 0.9546599496221663, "eval_overall_recall": 0.9522613065326633, "eval_runtime": 0.2728, "eval_samples_per_second": 623.094, "eval_steps_per_second": 10.996, "step": 6912 }, { "epoch": 73.0, "grad_norm": 0.0006964039639569819, "learning_rate": 1.3500000000000001e-05, "loss": 0.0006, "step": 7008 }, { "epoch": 73.0, "eval_LOCATION_f1": 0.9230769230769231, "eval_LOCATION_number": 94, "eval_LOCATION_precision": 0.8910891089108911, "eval_LOCATION_recall": 0.9574468085106383, "eval_ORGANIZATION_f1": 0.9268292682926829, "eval_ORGANIZATION_number": 167, "eval_ORGANIZATION_precision": 0.9440993788819876, "eval_ORGANIZATION_recall": 0.9101796407185628, "eval_PERSON_f1": 0.988929889298893, "eval_PERSON_number": 137, "eval_PERSON_precision": 1.0, "eval_PERSON_recall": 0.9781021897810219, "eval_loss": 0.09969379007816315, "eval_overall_accuracy": 0.9883977900552486, "eval_overall_f1": 0.947103274559194, "eval_overall_precision": 0.9494949494949495, "eval_overall_recall": 0.9447236180904522, "eval_runtime": 0.2748, "eval_samples_per_second": 618.723, "eval_steps_per_second": 10.919, "step": 7008 }, { "epoch": 74.0, "grad_norm": 0.0003081047907471657, "learning_rate": 1.3000000000000001e-05, "loss": 0.0004, "step": 7104 }, { "epoch": 74.0, "eval_LOCATION_f1": 0.9183673469387754, "eval_LOCATION_number": 94, "eval_LOCATION_precision": 0.8823529411764706, "eval_LOCATION_recall": 0.9574468085106383, "eval_ORGANIZATION_f1": 0.9263803680981595, "eval_ORGANIZATION_number": 167, "eval_ORGANIZATION_precision": 0.949685534591195, "eval_ORGANIZATION_recall": 0.9041916167664671, "eval_PERSON_f1": 0.9852941176470589, "eval_PERSON_number": 137, "eval_PERSON_precision": 0.9925925925925926, "eval_PERSON_recall": 0.9781021897810219, "eval_loss": 0.10354497283697128, "eval_overall_accuracy": 0.9881215469613259, "eval_overall_f1": 0.9445843828715367, "eval_overall_precision": 0.946969696969697, "eval_overall_recall": 0.9422110552763819, "eval_runtime": 0.2742, "eval_samples_per_second": 619.921, "eval_steps_per_second": 10.94, "step": 7104 }, { "epoch": 75.0, "grad_norm": 0.000309989380184561, "learning_rate": 1.25e-05, "loss": 0.0005, "step": 7200 }, { "epoch": 75.0, "eval_LOCATION_f1": 0.9015544041450778, "eval_LOCATION_number": 94, "eval_LOCATION_precision": 0.8787878787878788, "eval_LOCATION_recall": 0.925531914893617, "eval_ORGANIZATION_f1": 0.9141104294478527, "eval_ORGANIZATION_number": 167, "eval_ORGANIZATION_precision": 0.9371069182389937, "eval_ORGANIZATION_recall": 0.8922155688622755, "eval_PERSON_f1": 0.9779411764705882, "eval_PERSON_number": 137, "eval_PERSON_precision": 0.9851851851851852, "eval_PERSON_recall": 0.9708029197080292, "eval_loss": 0.10363327711820602, "eval_overall_accuracy": 0.9870165745856354, "eval_overall_f1": 0.9329962073324906, "eval_overall_precision": 0.9389312977099237, "eval_overall_recall": 0.9271356783919598, "eval_runtime": 0.2916, "eval_samples_per_second": 582.907, "eval_steps_per_second": 10.287, "step": 7200 }, { "epoch": 76.0, "grad_norm": 0.004243805538862944, "learning_rate": 1.2e-05, "loss": 0.0004, "step": 7296 }, { "epoch": 76.0, "eval_LOCATION_f1": 0.9015544041450778, "eval_LOCATION_number": 94, "eval_LOCATION_precision": 0.8787878787878788, "eval_LOCATION_recall": 0.925531914893617, "eval_ORGANIZATION_f1": 0.9146341463414633, "eval_ORGANIZATION_number": 167, "eval_ORGANIZATION_precision": 0.9316770186335404, "eval_ORGANIZATION_recall": 0.8982035928143712, "eval_PERSON_f1": 0.9672727272727273, "eval_PERSON_number": 137, "eval_PERSON_precision": 0.9637681159420289, "eval_PERSON_recall": 0.9708029197080292, "eval_loss": 0.09777092933654785, "eval_overall_accuracy": 0.9867403314917127, "eval_overall_f1": 0.9296482412060302, "eval_overall_precision": 0.9296482412060302, "eval_overall_recall": 0.9296482412060302, "eval_runtime": 0.2749, "eval_samples_per_second": 618.412, "eval_steps_per_second": 10.913, "step": 7296 }, { "epoch": 77.0, "grad_norm": 0.0024527793284505606, "learning_rate": 1.1500000000000002e-05, "loss": 0.0004, "step": 7392 }, { "epoch": 77.0, "eval_LOCATION_f1": 0.9072164948453608, "eval_LOCATION_number": 94, "eval_LOCATION_precision": 0.88, "eval_LOCATION_recall": 0.9361702127659575, "eval_ORGANIZATION_f1": 0.9216867469879517, "eval_ORGANIZATION_number": 167, "eval_ORGANIZATION_precision": 0.9272727272727272, "eval_ORGANIZATION_recall": 0.9161676646706587, "eval_PERSON_f1": 0.9852941176470589, "eval_PERSON_number": 137, "eval_PERSON_precision": 0.9925925925925926, "eval_PERSON_recall": 0.9781021897810219, "eval_loss": 0.08957943320274353, "eval_overall_accuracy": 0.9886740331491712, "eval_overall_f1": 0.9398496240601504, "eval_overall_precision": 0.9375, "eval_overall_recall": 0.9422110552763819, "eval_runtime": 0.2759, "eval_samples_per_second": 616.179, "eval_steps_per_second": 10.874, "step": 7392 }, { "epoch": 78.0, "grad_norm": 0.0007596246432512999, "learning_rate": 1.1000000000000001e-05, "loss": 0.0007, "step": 7488 }, { "epoch": 78.0, "eval_LOCATION_f1": 0.9119170984455959, "eval_LOCATION_number": 94, "eval_LOCATION_precision": 0.8888888888888888, "eval_LOCATION_recall": 0.9361702127659575, "eval_ORGANIZATION_f1": 0.9079754601226994, "eval_ORGANIZATION_number": 167, "eval_ORGANIZATION_precision": 0.9308176100628931, "eval_ORGANIZATION_recall": 0.8862275449101796, "eval_PERSON_f1": 0.988929889298893, "eval_PERSON_number": 137, "eval_PERSON_precision": 1.0, "eval_PERSON_recall": 0.9781021897810219, "eval_loss": 0.10337463766336441, "eval_overall_accuracy": 0.9878453038674033, "eval_overall_f1": 0.9367088607594937, "eval_overall_precision": 0.9438775510204082, "eval_overall_recall": 0.9296482412060302, "eval_runtime": 0.276, "eval_samples_per_second": 615.988, "eval_steps_per_second": 10.87, "step": 7488 }, { "epoch": 79.0, "grad_norm": 0.0006302375695668161, "learning_rate": 1.05e-05, "loss": 0.0004, "step": 7584 }, { "epoch": 79.0, "eval_LOCATION_f1": 0.9128205128205128, "eval_LOCATION_number": 94, "eval_LOCATION_precision": 0.8811881188118812, "eval_LOCATION_recall": 0.9468085106382979, "eval_ORGANIZATION_f1": 0.9118541033434651, "eval_ORGANIZATION_number": 167, "eval_ORGANIZATION_precision": 0.9259259259259259, "eval_ORGANIZATION_recall": 0.8982035928143712, "eval_PERSON_f1": 0.988929889298893, "eval_PERSON_number": 137, "eval_PERSON_precision": 1.0, "eval_PERSON_recall": 0.9781021897810219, "eval_loss": 0.11167038977146149, "eval_overall_accuracy": 0.987292817679558, "eval_overall_f1": 0.9383647798742139, "eval_overall_precision": 0.9395465994962217, "eval_overall_recall": 0.9371859296482412, "eval_runtime": 0.2741, "eval_samples_per_second": 620.126, "eval_steps_per_second": 10.943, "step": 7584 }, { "epoch": 80.0, "grad_norm": 0.0015559011371806264, "learning_rate": 1e-05, "loss": 0.0006, "step": 7680 }, { "epoch": 80.0, "eval_LOCATION_f1": 0.9166666666666666, "eval_LOCATION_number": 94, "eval_LOCATION_precision": 0.8979591836734694, "eval_LOCATION_recall": 0.9361702127659575, "eval_ORGANIZATION_f1": 0.9176470588235294, "eval_ORGANIZATION_number": 167, "eval_ORGANIZATION_precision": 0.9017341040462428, "eval_ORGANIZATION_recall": 0.9341317365269461, "eval_PERSON_f1": 0.988929889298893, "eval_PERSON_number": 137, "eval_PERSON_precision": 1.0, "eval_PERSON_recall": 0.9781021897810219, "eval_loss": 0.10534064471721649, "eval_overall_accuracy": 0.987292817679558, "eval_overall_f1": 0.9414694894146949, "eval_overall_precision": 0.9333333333333333, "eval_overall_recall": 0.949748743718593, "eval_runtime": 0.276, "eval_samples_per_second": 615.985, "eval_steps_per_second": 10.87, "step": 7680 }, { "epoch": 81.0, "grad_norm": 0.0003092308179475367, "learning_rate": 9.5e-06, "loss": 0.0003, "step": 7776 }, { "epoch": 81.0, "eval_LOCATION_f1": 0.9166666666666666, "eval_LOCATION_number": 94, "eval_LOCATION_precision": 0.8979591836734694, "eval_LOCATION_recall": 0.9361702127659575, "eval_ORGANIZATION_f1": 0.9221556886227545, "eval_ORGANIZATION_number": 167, "eval_ORGANIZATION_precision": 0.9221556886227545, "eval_ORGANIZATION_recall": 0.9221556886227545, "eval_PERSON_f1": 0.988929889298893, "eval_PERSON_number": 137, "eval_PERSON_precision": 1.0, "eval_PERSON_recall": 0.9781021897810219, "eval_loss": 0.10229864716529846, "eval_overall_accuracy": 0.9883977900552486, "eval_overall_f1": 0.9435382685069008, "eval_overall_precision": 0.9423558897243107, "eval_overall_recall": 0.9447236180904522, "eval_runtime": 0.2832, "eval_samples_per_second": 600.229, "eval_steps_per_second": 10.592, "step": 7776 }, { "epoch": 82.0, "grad_norm": 0.0020309356041252613, "learning_rate": 9e-06, "loss": 0.0005, "step": 7872 }, { "epoch": 82.0, "eval_LOCATION_f1": 0.9222797927461138, "eval_LOCATION_number": 94, "eval_LOCATION_precision": 0.898989898989899, "eval_LOCATION_recall": 0.9468085106382979, "eval_ORGANIZATION_f1": 0.9281437125748502, "eval_ORGANIZATION_number": 167, "eval_ORGANIZATION_precision": 0.9281437125748503, "eval_ORGANIZATION_recall": 0.9281437125748503, "eval_PERSON_f1": 0.988929889298893, "eval_PERSON_number": 137, "eval_PERSON_precision": 1.0, "eval_PERSON_recall": 0.9781021897810219, "eval_loss": 0.09981951862573624, "eval_overall_accuracy": 0.9886740331491712, "eval_overall_f1": 0.9473684210526316, "eval_overall_precision": 0.945, "eval_overall_recall": 0.949748743718593, "eval_runtime": 0.2783, "eval_samples_per_second": 610.758, "eval_steps_per_second": 10.778, "step": 7872 }, { "epoch": 83.0, "grad_norm": 0.0008582601440139115, "learning_rate": 8.500000000000002e-06, "loss": 0.0004, "step": 7968 }, { "epoch": 83.0, "eval_LOCATION_f1": 0.9166666666666666, "eval_LOCATION_number": 94, "eval_LOCATION_precision": 0.8979591836734694, "eval_LOCATION_recall": 0.9361702127659575, "eval_ORGANIZATION_f1": 0.9221556886227545, "eval_ORGANIZATION_number": 167, "eval_ORGANIZATION_precision": 0.9221556886227545, "eval_ORGANIZATION_recall": 0.9221556886227545, "eval_PERSON_f1": 0.988929889298893, "eval_PERSON_number": 137, "eval_PERSON_precision": 1.0, "eval_PERSON_recall": 0.9781021897810219, "eval_loss": 0.10308824479579926, "eval_overall_accuracy": 0.9883977900552486, "eval_overall_f1": 0.9435382685069008, "eval_overall_precision": 0.9423558897243107, "eval_overall_recall": 0.9447236180904522, "eval_runtime": 0.2773, "eval_samples_per_second": 613.12, "eval_steps_per_second": 10.82, "step": 7968 }, { "epoch": 84.0, "grad_norm": 0.0821322500705719, "learning_rate": 8.000000000000001e-06, "loss": 0.0002, "step": 8064 }, { "epoch": 84.0, "eval_LOCATION_f1": 0.9214659685863875, "eval_LOCATION_number": 94, "eval_LOCATION_precision": 0.9072164948453608, "eval_LOCATION_recall": 0.9361702127659575, "eval_ORGANIZATION_f1": 0.9216867469879517, "eval_ORGANIZATION_number": 167, "eval_ORGANIZATION_precision": 0.9272727272727272, "eval_ORGANIZATION_recall": 0.9161676646706587, "eval_PERSON_f1": 0.988929889298893, "eval_PERSON_number": 137, "eval_PERSON_precision": 1.0, "eval_PERSON_recall": 0.9781021897810219, "eval_loss": 0.10764861106872559, "eval_overall_accuracy": 0.988950276243094, "eval_overall_f1": 0.9445843828715367, "eval_overall_precision": 0.946969696969697, "eval_overall_recall": 0.9422110552763819, "eval_runtime": 0.2736, "eval_samples_per_second": 621.255, "eval_steps_per_second": 10.963, "step": 8064 }, { "epoch": 85.0, "grad_norm": 0.0002442681579850614, "learning_rate": 7.5e-06, "loss": 0.0008, "step": 8160 }, { "epoch": 85.0, "eval_LOCATION_f1": 0.9157894736842105, "eval_LOCATION_number": 94, "eval_LOCATION_precision": 0.90625, "eval_LOCATION_recall": 0.925531914893617, "eval_ORGANIZATION_f1": 0.9216867469879517, "eval_ORGANIZATION_number": 167, "eval_ORGANIZATION_precision": 0.9272727272727272, "eval_ORGANIZATION_recall": 0.9161676646706587, "eval_PERSON_f1": 0.981549815498155, "eval_PERSON_number": 137, "eval_PERSON_precision": 0.9925373134328358, "eval_PERSON_recall": 0.9708029197080292, "eval_loss": 0.10311637073755264, "eval_overall_accuracy": 0.9886740331491712, "eval_overall_f1": 0.9407313997477933, "eval_overall_precision": 0.9443037974683545, "eval_overall_recall": 0.9371859296482412, "eval_runtime": 0.2849, "eval_samples_per_second": 596.714, "eval_steps_per_second": 10.53, "step": 8160 }, { "epoch": 86.0, "grad_norm": 0.004653602372854948, "learning_rate": 7.000000000000001e-06, "loss": 0.0003, "step": 8256 }, { "epoch": 86.0, "eval_LOCATION_f1": 0.9157894736842105, "eval_LOCATION_number": 94, "eval_LOCATION_precision": 0.90625, "eval_LOCATION_recall": 0.925531914893617, "eval_ORGANIZATION_f1": 0.9240121580547112, "eval_ORGANIZATION_number": 167, "eval_ORGANIZATION_precision": 0.9382716049382716, "eval_ORGANIZATION_recall": 0.9101796407185628, "eval_PERSON_f1": 0.981549815498155, "eval_PERSON_number": 137, "eval_PERSON_precision": 0.9925373134328358, "eval_PERSON_recall": 0.9708029197080292, "eval_loss": 0.09666076302528381, "eval_overall_accuracy": 0.9892265193370166, "eval_overall_f1": 0.9417721518987343, "eval_overall_precision": 0.9489795918367347, "eval_overall_recall": 0.9346733668341709, "eval_runtime": 0.2723, "eval_samples_per_second": 624.208, "eval_steps_per_second": 11.015, "step": 8256 }, { "epoch": 87.0, "grad_norm": 0.0002296812308486551, "learning_rate": 6.5000000000000004e-06, "loss": 0.0005, "step": 8352 }, { "epoch": 87.0, "eval_LOCATION_f1": 0.9119170984455959, "eval_LOCATION_number": 94, "eval_LOCATION_precision": 0.8888888888888888, "eval_LOCATION_recall": 0.9361702127659575, "eval_ORGANIZATION_f1": 0.9146341463414633, "eval_ORGANIZATION_number": 167, "eval_ORGANIZATION_precision": 0.9316770186335404, "eval_ORGANIZATION_recall": 0.8982035928143712, "eval_PERSON_f1": 0.988929889298893, "eval_PERSON_number": 137, "eval_PERSON_precision": 1.0, "eval_PERSON_recall": 0.9781021897810219, "eval_loss": 0.09778623282909393, "eval_overall_accuracy": 0.9883977900552486, "eval_overall_f1": 0.9393939393939393, "eval_overall_precision": 0.9441624365482234, "eval_overall_recall": 0.9346733668341709, "eval_runtime": 0.2811, "eval_samples_per_second": 604.867, "eval_steps_per_second": 10.674, "step": 8352 }, { "epoch": 88.0, "grad_norm": 0.0004456902388483286, "learning_rate": 6e-06, "loss": 0.0003, "step": 8448 }, { "epoch": 88.0, "eval_LOCATION_f1": 0.9119170984455959, "eval_LOCATION_number": 94, "eval_LOCATION_precision": 0.8888888888888888, "eval_LOCATION_recall": 0.9361702127659575, "eval_ORGANIZATION_f1": 0.9174311926605505, "eval_ORGANIZATION_number": 167, "eval_ORGANIZATION_precision": 0.9375, "eval_ORGANIZATION_recall": 0.8982035928143712, "eval_PERSON_f1": 0.988929889298893, "eval_PERSON_number": 137, "eval_PERSON_precision": 1.0, "eval_PERSON_recall": 0.9781021897810219, "eval_loss": 0.11038926243782043, "eval_overall_accuracy": 0.9881215469613259, "eval_overall_f1": 0.9405815423514539, "eval_overall_precision": 0.9465648854961832, "eval_overall_recall": 0.9346733668341709, "eval_runtime": 0.2718, "eval_samples_per_second": 625.458, "eval_steps_per_second": 11.037, "step": 8448 }, { "epoch": 89.0, "grad_norm": 0.2186516970396042, "learning_rate": 5.500000000000001e-06, "loss": 0.0005, "step": 8544 }, { "epoch": 89.0, "eval_LOCATION_f1": 0.9175257731958764, "eval_LOCATION_number": 94, "eval_LOCATION_precision": 0.89, "eval_LOCATION_recall": 0.9468085106382979, "eval_ORGANIZATION_f1": 0.9268292682926829, "eval_ORGANIZATION_number": 167, "eval_ORGANIZATION_precision": 0.9440993788819876, "eval_ORGANIZATION_recall": 0.9101796407185628, "eval_PERSON_f1": 0.988929889298893, "eval_PERSON_number": 137, "eval_PERSON_precision": 1.0, "eval_PERSON_recall": 0.9781021897810219, "eval_loss": 0.10691658407449722, "eval_overall_accuracy": 0.9886740331491712, "eval_overall_f1": 0.9457755359394704, "eval_overall_precision": 0.9493670886075949, "eval_overall_recall": 0.9422110552763819, "eval_runtime": 0.2772, "eval_samples_per_second": 613.193, "eval_steps_per_second": 10.821, "step": 8544 }, { "epoch": 90.0, "grad_norm": 0.0001421573106199503, "learning_rate": 5e-06, "loss": 0.0003, "step": 8640 }, { "epoch": 90.0, "eval_LOCATION_f1": 0.9175257731958764, "eval_LOCATION_number": 94, "eval_LOCATION_precision": 0.89, "eval_LOCATION_recall": 0.9468085106382979, "eval_ORGANIZATION_f1": 0.9268292682926829, "eval_ORGANIZATION_number": 167, "eval_ORGANIZATION_precision": 0.9440993788819876, "eval_ORGANIZATION_recall": 0.9101796407185628, "eval_PERSON_f1": 0.988929889298893, "eval_PERSON_number": 137, "eval_PERSON_precision": 1.0, "eval_PERSON_recall": 0.9781021897810219, "eval_loss": 0.10707202553749084, "eval_overall_accuracy": 0.9886740331491712, "eval_overall_f1": 0.9457755359394704, "eval_overall_precision": 0.9493670886075949, "eval_overall_recall": 0.9422110552763819, "eval_runtime": 0.2764, "eval_samples_per_second": 615.111, "eval_steps_per_second": 10.855, "step": 8640 }, { "epoch": 91.0, "grad_norm": 0.0003911272215191275, "learning_rate": 4.5e-06, "loss": 0.0005, "step": 8736 }, { "epoch": 91.0, "eval_LOCATION_f1": 0.9175257731958764, "eval_LOCATION_number": 94, "eval_LOCATION_precision": 0.89, "eval_LOCATION_recall": 0.9468085106382979, "eval_ORGANIZATION_f1": 0.9268292682926829, "eval_ORGANIZATION_number": 167, "eval_ORGANIZATION_precision": 0.9440993788819876, "eval_ORGANIZATION_recall": 0.9101796407185628, "eval_PERSON_f1": 0.988929889298893, "eval_PERSON_number": 137, "eval_PERSON_precision": 1.0, "eval_PERSON_recall": 0.9781021897810219, "eval_loss": 0.10682348161935806, "eval_overall_accuracy": 0.9886740331491712, "eval_overall_f1": 0.9457755359394704, "eval_overall_precision": 0.9493670886075949, "eval_overall_recall": 0.9422110552763819, "eval_runtime": 0.2772, "eval_samples_per_second": 613.199, "eval_steps_per_second": 10.821, "step": 8736 }, { "epoch": 92.0, "grad_norm": 0.0001311097585130483, "learning_rate": 4.000000000000001e-06, "loss": 0.0004, "step": 8832 }, { "epoch": 92.0, "eval_LOCATION_f1": 0.9175257731958764, "eval_LOCATION_number": 94, "eval_LOCATION_precision": 0.89, "eval_LOCATION_recall": 0.9468085106382979, "eval_ORGANIZATION_f1": 0.9300911854103343, "eval_ORGANIZATION_number": 167, "eval_ORGANIZATION_precision": 0.9444444444444444, "eval_ORGANIZATION_recall": 0.9161676646706587, "eval_PERSON_f1": 0.988929889298893, "eval_PERSON_number": 137, "eval_PERSON_precision": 1.0, "eval_PERSON_recall": 0.9781021897810219, "eval_loss": 0.10780012607574463, "eval_overall_accuracy": 0.988950276243094, "eval_overall_f1": 0.947103274559194, "eval_overall_precision": 0.9494949494949495, "eval_overall_recall": 0.9447236180904522, "eval_runtime": 0.272, "eval_samples_per_second": 625.09, "eval_steps_per_second": 11.031, "step": 8832 }, { "epoch": 93.0, "grad_norm": 0.0003295471833553165, "learning_rate": 3.5000000000000004e-06, "loss": 0.0003, "step": 8928 }, { "epoch": 93.0, "eval_LOCATION_f1": 0.9175257731958764, "eval_LOCATION_number": 94, "eval_LOCATION_precision": 0.89, "eval_LOCATION_recall": 0.9468085106382979, "eval_ORGANIZATION_f1": 0.9300911854103343, "eval_ORGANIZATION_number": 167, "eval_ORGANIZATION_precision": 0.9444444444444444, "eval_ORGANIZATION_recall": 0.9161676646706587, "eval_PERSON_f1": 0.988929889298893, "eval_PERSON_number": 137, "eval_PERSON_precision": 1.0, "eval_PERSON_recall": 0.9781021897810219, "eval_loss": 0.107907734811306, "eval_overall_accuracy": 0.988950276243094, "eval_overall_f1": 0.947103274559194, "eval_overall_precision": 0.9494949494949495, "eval_overall_recall": 0.9447236180904522, "eval_runtime": 0.2755, "eval_samples_per_second": 617.144, "eval_steps_per_second": 10.891, "step": 8928 }, { "epoch": 94.0, "grad_norm": 0.00015319112571887672, "learning_rate": 3e-06, "loss": 0.0004, "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": 0.9272727272727274, "eval_ORGANIZATION_number": 167, "eval_ORGANIZATION_precision": 0.9386503067484663, "eval_ORGANIZATION_recall": 0.9161676646706587, "eval_PERSON_f1": 0.988929889298893, "eval_PERSON_number": 137, "eval_PERSON_precision": 1.0, "eval_PERSON_recall": 0.9781021897810219, "eval_loss": 0.1081731989979744, "eval_overall_accuracy": 0.9886740331491712, "eval_overall_f1": 0.9459119496855346, "eval_overall_precision": 0.947103274559194, "eval_overall_recall": 0.9447236180904522, "eval_runtime": 0.2791, "eval_samples_per_second": 609.077, "eval_steps_per_second": 10.748, "step": 9024 }, { "epoch": 95.0, "grad_norm": 0.0003005734470207244, "learning_rate": 2.5e-06, "loss": 0.0003, "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": 0.9272727272727274, "eval_ORGANIZATION_number": 167, "eval_ORGANIZATION_precision": 0.9386503067484663, "eval_ORGANIZATION_recall": 0.9161676646706587, "eval_PERSON_f1": 0.988929889298893, "eval_PERSON_number": 137, "eval_PERSON_precision": 1.0, "eval_PERSON_recall": 0.9781021897810219, "eval_loss": 0.10803940892219543, "eval_overall_accuracy": 0.9886740331491712, "eval_overall_f1": 0.9459119496855346, "eval_overall_precision": 0.947103274559194, "eval_overall_recall": 0.9447236180904522, "eval_runtime": 0.2786, "eval_samples_per_second": 610.125, "eval_steps_per_second": 10.767, "step": 9120 }, { "epoch": 96.0, "grad_norm": 0.00013263874279800802, "learning_rate": 2.0000000000000003e-06, "loss": 0.0003, "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": 0.9272727272727274, "eval_ORGANIZATION_number": 167, "eval_ORGANIZATION_precision": 0.9386503067484663, "eval_ORGANIZATION_recall": 0.9161676646706587, "eval_PERSON_f1": 0.988929889298893, "eval_PERSON_number": 137, "eval_PERSON_precision": 1.0, "eval_PERSON_recall": 0.9781021897810219, "eval_loss": 0.1081809550523758, "eval_overall_accuracy": 0.9886740331491712, "eval_overall_f1": 0.9459119496855346, "eval_overall_precision": 0.947103274559194, "eval_overall_recall": 0.9447236180904522, "eval_runtime": 0.2873, "eval_samples_per_second": 591.671, "eval_steps_per_second": 10.441, "step": 9216 }, { "epoch": 97.0, "grad_norm": 0.005492445081472397, "learning_rate": 1.5e-06, "loss": 0.0002, "step": 9312 }, { "epoch": 97.0, "eval_LOCATION_f1": 0.9175257731958764, "eval_LOCATION_number": 94, "eval_LOCATION_precision": 0.89, "eval_LOCATION_recall": 0.9468085106382979, "eval_ORGANIZATION_f1": 0.9272727272727274, "eval_ORGANIZATION_number": 167, "eval_ORGANIZATION_precision": 0.9386503067484663, "eval_ORGANIZATION_recall": 0.9161676646706587, "eval_PERSON_f1": 0.988929889298893, "eval_PERSON_number": 137, "eval_PERSON_precision": 1.0, "eval_PERSON_recall": 0.9781021897810219, "eval_loss": 0.1080157533288002, "eval_overall_accuracy": 0.9886740331491712, "eval_overall_f1": 0.9459119496855346, "eval_overall_precision": 0.947103274559194, "eval_overall_recall": 0.9447236180904522, "eval_runtime": 0.2745, "eval_samples_per_second": 619.199, "eval_steps_per_second": 10.927, "step": 9312 }, { "epoch": 98.0, "grad_norm": 0.00014524870493914932, "learning_rate": 1.0000000000000002e-06, "loss": 0.0003, "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": 0.9300911854103343, "eval_ORGANIZATION_number": 167, "eval_ORGANIZATION_precision": 0.9444444444444444, "eval_ORGANIZATION_recall": 0.9161676646706587, "eval_PERSON_f1": 0.988929889298893, "eval_PERSON_number": 137, "eval_PERSON_precision": 1.0, "eval_PERSON_recall": 0.9781021897810219, "eval_loss": 0.10801573097705841, "eval_overall_accuracy": 0.988950276243094, "eval_overall_f1": 0.947103274559194, "eval_overall_precision": 0.9494949494949495, "eval_overall_recall": 0.9447236180904522, "eval_runtime": 0.2733, "eval_samples_per_second": 622.014, "eval_steps_per_second": 10.977, "step": 9408 }, { "epoch": 99.0, "grad_norm": 0.00017239197040908039, "learning_rate": 5.000000000000001e-07, "loss": 0.0003, "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": 0.9272727272727274, "eval_ORGANIZATION_number": 167, "eval_ORGANIZATION_precision": 0.9386503067484663, "eval_ORGANIZATION_recall": 0.9161676646706587, "eval_PERSON_f1": 0.988929889298893, "eval_PERSON_number": 137, "eval_PERSON_precision": 1.0, "eval_PERSON_recall": 0.9781021897810219, "eval_loss": 0.10847752541303635, "eval_overall_accuracy": 0.9886740331491712, "eval_overall_f1": 0.9459119496855346, "eval_overall_precision": 0.947103274559194, "eval_overall_recall": 0.9447236180904522, "eval_runtime": 0.2767, "eval_samples_per_second": 614.479, "eval_steps_per_second": 10.844, "step": 9504 }, { "epoch": 100.0, "grad_norm": 0.00010377541184425354, "learning_rate": 0.0, "loss": 0.0002, "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": 0.9272727272727274, "eval_ORGANIZATION_number": 167, "eval_ORGANIZATION_precision": 0.9386503067484663, "eval_ORGANIZATION_recall": 0.9161676646706587, "eval_PERSON_f1": 0.988929889298893, "eval_PERSON_number": 137, "eval_PERSON_precision": 1.0, "eval_PERSON_recall": 0.9781021897810219, "eval_loss": 0.10839837789535522, "eval_overall_accuracy": 0.9886740331491712, "eval_overall_f1": 0.9459119496855346, "eval_overall_precision": 0.947103274559194, "eval_overall_recall": 0.9447236180904522, "eval_runtime": 0.2745, "eval_samples_per_second": 619.357, "eval_steps_per_second": 10.93, "step": 9600 }, { "epoch": 100.0, "step": 9600, "total_flos": 3851325939318660.0, "train_loss": 0.0056626121466979385, "train_runtime": 904.5312, "train_samples_per_second": 169.148, "train_steps_per_second": 10.613 } ], "logging_steps": 500, "max_steps": 9600, "num_input_tokens_seen": 0, "num_train_epochs": 100, "save_steps": 500, "total_flos": 3851325939318660.0, "train_batch_size": 16, "trial_name": null, "trial_params": null }