{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 3.0,
  "eval_steps": 1000,
  "global_step": 15153,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.1,
      "learning_rate": 4.835016168415495e-05,
      "loss": 0.8808,
      "step": 500
    },
    {
      "epoch": 0.2,
      "learning_rate": 4.670032336830991e-05,
      "loss": 0.6038,
      "step": 1000
    },
    {
      "epoch": 0.2,
      "eval_COMMENT": {
        "f1": 0.5911795137092601,
        "number": 6900,
        "precision": 0.5337459131247081,
        "recall": 0.662463768115942
      },
      "eval_NAME": {
        "f1": 0.7920792079207921,
        "number": 8840,
        "precision": 0.7791639308382579,
        "recall": 0.8054298642533937
      },
      "eval_QTY": {
        "f1": 0.9777005904828066,
        "number": 7169,
        "precision": 0.9738444505950733,
        "recall": 0.9815873901520435
      },
      "eval_RANGE_END": {
        "f1": 0.5829145728643217,
        "number": 94,
        "precision": 0.5523809523809524,
        "recall": 0.6170212765957447
      },
      "eval_UNIT": {
        "f1": 0.9443229604709841,
        "number": 5778,
        "precision": 0.918520942408377,
        "recall": 0.9716164762893735
      },
      "eval_loss": 0.5653780698776245,
      "eval_overall_accuracy": 0.7991183879093199,
      "eval_overall_f1": 0.8143376831425424,
      "eval_overall_precision": 0.7834323326376625,
      "eval_overall_recall": 0.8477815225322262,
      "eval_runtime": 9.1463,
      "eval_samples_per_second": 929.995,
      "eval_steps_per_second": 29.083,
      "step": 1000
    },
    {
      "epoch": 0.3,
      "learning_rate": 4.505048505246486e-05,
      "loss": 0.5757,
      "step": 1500
    },
    {
      "epoch": 0.4,
      "learning_rate": 4.3400646736619816e-05,
      "loss": 0.5583,
      "step": 2000
    },
    {
      "epoch": 0.4,
      "eval_COMMENT": {
        "f1": 0.6243035287874767,
        "number": 6900,
        "precision": 0.575587084148728,
        "recall": 0.6820289855072463
      },
      "eval_NAME": {
        "f1": 0.7976755880873889,
        "number": 8840,
        "precision": 0.7881196864303853,
        "recall": 0.8074660633484163
      },
      "eval_QTY": {
        "f1": 0.9807208717518859,
        "number": 7169,
        "precision": 0.982230306422275,
        "recall": 0.9792160691867764
      },
      "eval_RANGE_END": {
        "f1": 0.6640625,
        "number": 94,
        "precision": 0.5246913580246914,
        "recall": 0.9042553191489362
      },
      "eval_UNIT": {
        "f1": 0.9456703443664326,
        "number": 5778,
        "precision": 0.9144843194309732,
        "recall": 0.9790584977500866
      },
      "eval_loss": 0.5386040806770325,
      "eval_overall_accuracy": 0.8086405617891764,
      "eval_overall_f1": 0.8269673494765498,
      "eval_overall_precision": 0.8007680291590732,
      "eval_overall_recall": 0.8549390222716375,
      "eval_runtime": 12.5089,
      "eval_samples_per_second": 679.995,
      "eval_steps_per_second": 21.265,
      "step": 2000
    },
    {
      "epoch": 0.49,
      "learning_rate": 4.1750808420774766e-05,
      "loss": 0.535,
      "step": 2500
    },
    {
      "epoch": 0.59,
      "learning_rate": 4.010097010492972e-05,
      "loss": 0.5351,
      "step": 3000
    },
    {
      "epoch": 0.59,
      "eval_COMMENT": {
        "f1": 0.6408767695218214,
        "number": 6900,
        "precision": 0.5827500296595088,
        "recall": 0.7118840579710145
      },
      "eval_NAME": {
        "f1": 0.7941633377718598,
        "number": 8840,
        "precision": 0.7792900696864111,
        "recall": 0.8096153846153846
      },
      "eval_QTY": {
        "f1": 0.9805879090404881,
        "number": 7169,
        "precision": 0.974776016540317,
        "recall": 0.9864695215511229
      },
      "eval_RANGE_END": {
        "f1": 0.7333333333333333,
        "number": 94,
        "precision": 0.6027397260273972,
        "recall": 0.9361702127659575
      },
      "eval_UNIT": {
        "f1": 0.946524064171123,
        "number": 5778,
        "precision": 0.9150242326332795,
        "recall": 0.980269989615784
      },
      "eval_loss": 0.5204864740371704,
      "eval_overall_accuracy": 0.8152240287000992,
      "eval_overall_f1": 0.8299741602067184,
      "eval_overall_precision": 0.797750288424561,
      "eval_overall_recall": 0.8649108787047011,
      "eval_runtime": 9.7563,
      "eval_samples_per_second": 871.849,
      "eval_steps_per_second": 27.264,
      "step": 3000
    },
    {
      "epoch": 0.69,
      "learning_rate": 3.845113178908467e-05,
      "loss": 0.5267,
      "step": 3500
    },
    {
      "epoch": 0.79,
      "learning_rate": 3.680129347323962e-05,
      "loss": 0.5238,
      "step": 4000
    },
    {
      "epoch": 0.79,
      "eval_COMMENT": {
        "f1": 0.64383198621969,
        "number": 6900,
        "precision": 0.592994874298267,
        "recall": 0.7042028985507246
      },
      "eval_NAME": {
        "f1": 0.8008923591745678,
        "number": 8840,
        "precision": 0.7898789878987899,
        "recall": 0.8122171945701357
      },
      "eval_QTY": {
        "f1": 0.9833298458533863,
        "number": 7169,
        "precision": 0.9833984375,
        "recall": 0.983261263774585
      },
      "eval_RANGE_END": {
        "f1": 0.7445887445887446,
        "number": 94,
        "precision": 0.6277372262773723,
        "recall": 0.9148936170212766
      },
      "eval_UNIT": {
        "f1": 0.948472641826116,
        "number": 5778,
        "precision": 0.9206581948517433,
        "recall": 0.9780200761509172
      },
      "eval_loss": 0.507017195224762,
      "eval_overall_accuracy": 0.820261812075414,
      "eval_overall_f1": 0.83434160112926,
      "eval_overall_precision": 0.8079213720831842,
      "eval_overall_recall": 0.8625482088878079,
      "eval_runtime": 7.298,
      "eval_samples_per_second": 1165.532,
      "eval_steps_per_second": 36.449,
      "step": 4000
    },
    {
      "epoch": 0.89,
      "learning_rate": 3.515145515739457e-05,
      "loss": 0.5177,
      "step": 4500
    },
    {
      "epoch": 0.99,
      "learning_rate": 3.3501616841549535e-05,
      "loss": 0.5134,
      "step": 5000
    },
    {
      "epoch": 0.99,
      "eval_COMMENT": {
        "f1": 0.6618449602209219,
        "number": 6900,
        "precision": 0.6057287278854254,
        "recall": 0.7294202898550725
      },
      "eval_NAME": {
        "f1": 0.8053840591801545,
        "number": 8840,
        "precision": 0.7922092132618448,
        "recall": 0.8190045248868778
      },
      "eval_QTY": {
        "f1": 0.9847285853116757,
        "number": 7169,
        "precision": 0.980102252314495,
        "recall": 0.9893988003905705
      },
      "eval_RANGE_END": {
        "f1": 0.7627906976744186,
        "number": 94,
        "precision": 0.6776859504132231,
        "recall": 0.8723404255319149
      },
      "eval_UNIT": {
        "f1": 0.9481246345334559,
        "number": 5778,
        "precision": 0.9163571774584208,
        "recall": 0.9821737625475944
      },
      "eval_loss": 0.49935489892959595,
      "eval_overall_accuracy": 0.8229906114037097,
      "eval_overall_f1": 0.8405152224824355,
      "eval_overall_precision": 0.8104454982418787,
      "eval_overall_recall": 0.8729022619088982,
      "eval_runtime": 9.399,
      "eval_samples_per_second": 904.995,
      "eval_steps_per_second": 28.301,
      "step": 5000
    },
    {
      "epoch": 1.09,
      "learning_rate": 3.1851778525704485e-05,
      "loss": 0.5019,
      "step": 5500
    },
    {
      "epoch": 1.19,
      "learning_rate": 3.0201940209859435e-05,
      "loss": 0.4954,
      "step": 6000
    },
    {
      "epoch": 1.19,
      "eval_COMMENT": {
        "f1": 0.6517109756914216,
        "number": 6900,
        "precision": 0.6057512759865554,
        "recall": 0.7052173913043478
      },
      "eval_NAME": {
        "f1": 0.7966298985643812,
        "number": 8840,
        "precision": 0.7810020649929356,
        "recall": 0.81289592760181
      },
      "eval_QTY": {
        "f1": 0.9840166782487838,
        "number": 7169,
        "precision": 0.9804736186123806,
        "recall": 0.9875854372994839
      },
      "eval_RANGE_END": {
        "f1": 0.7906976744186046,
        "number": 94,
        "precision": 0.7024793388429752,
        "recall": 0.9042553191489362
      },
      "eval_UNIT": {
        "f1": 0.9480010015858442,
        "number": 5778,
        "precision": 0.9155247460906013,
        "recall": 0.9828660436137072
      },
      "eval_loss": 0.502167820930481,
      "eval_overall_accuracy": 0.8203381421265552,
      "eval_overall_f1": 0.8359973136333109,
      "eval_overall_precision": 0.8088631859384645,
      "eval_overall_recall": 0.8650151141377992,
      "eval_runtime": 6.9755,
      "eval_samples_per_second": 1219.408,
      "eval_steps_per_second": 38.133,
      "step": 6000
    },
    {
      "epoch": 1.29,
      "learning_rate": 2.855210189401439e-05,
      "loss": 0.5008,
      "step": 6500
    },
    {
      "epoch": 1.39,
      "learning_rate": 2.690226357816934e-05,
      "loss": 0.4998,
      "step": 7000
    },
    {
      "epoch": 1.39,
      "eval_COMMENT": {
        "f1": 0.672593378889405,
        "number": 6900,
        "precision": 0.6202128961213753,
        "recall": 0.7346376811594203
      },
      "eval_NAME": {
        "f1": 0.8033342595165324,
        "number": 8840,
        "precision": 0.7895139268159476,
        "recall": 0.8176470588235294
      },
      "eval_QTY": {
        "f1": 0.9836797321802204,
        "number": 7169,
        "precision": 0.9836797321802204,
        "recall": 0.9836797321802204
      },
      "eval_RANGE_END": {
        "f1": 0.7565217391304346,
        "number": 94,
        "precision": 0.6397058823529411,
        "recall": 0.925531914893617
      },
      "eval_UNIT": {
        "f1": 0.9496971736204576,
        "number": 5778,
        "precision": 0.9238952536824877,
        "recall": 0.976981654551748
      },
      "eval_loss": 0.4921141266822815,
      "eval_overall_accuracy": 0.825166017861232,
      "eval_overall_f1": 0.8427189032995095,
      "eval_overall_precision": 0.8158279933643431,
      "eval_overall_recall": 0.871442965845523,
      "eval_runtime": 8.6591,
      "eval_samples_per_second": 982.314,
      "eval_steps_per_second": 30.719,
      "step": 7000
    },
    {
      "epoch": 1.48,
      "learning_rate": 2.5252425262324292e-05,
      "loss": 0.4981,
      "step": 7500
    },
    {
      "epoch": 1.58,
      "learning_rate": 2.3602586946479245e-05,
      "loss": 0.4912,
      "step": 8000
    },
    {
      "epoch": 1.58,
      "eval_COMMENT": {
        "f1": 0.6753212169629186,
        "number": 6900,
        "precision": 0.6245536264006896,
        "recall": 0.735072463768116
      },
      "eval_NAME": {
        "f1": 0.8065398136264716,
        "number": 8840,
        "precision": 0.7958374628344896,
        "recall": 0.8175339366515837
      },
      "eval_QTY": {
        "f1": 0.9842043003270474,
        "number": 7169,
        "precision": 0.9819494584837545,
        "recall": 0.9864695215511229
      },
      "eval_RANGE_END": {
        "f1": 0.7665198237885463,
        "number": 94,
        "precision": 0.6541353383458647,
        "recall": 0.925531914893617
      },
      "eval_UNIT": {
        "f1": 0.9495798319327731,
        "number": 5778,
        "precision": 0.9229010127409344,
        "recall": 0.9778470058843891
      },
      "eval_loss": 0.4943971037864685,
      "eval_overall_accuracy": 0.8256049156552935,
      "eval_overall_f1": 0.8448183041722745,
      "eval_overall_precision": 0.818943866401383,
      "eval_overall_recall": 0.8723810847434071,
      "eval_runtime": 8.4356,
      "eval_samples_per_second": 1008.348,
      "eval_steps_per_second": 31.533,
      "step": 8000
    },
    {
      "epoch": 1.68,
      "learning_rate": 2.19527486306342e-05,
      "loss": 0.4874,
      "step": 8500
    },
    {
      "epoch": 1.78,
      "learning_rate": 2.0302910314789152e-05,
      "loss": 0.4974,
      "step": 9000
    },
    {
      "epoch": 1.78,
      "eval_COMMENT": {
        "f1": 0.6773682124455014,
        "number": 6900,
        "precision": 0.6223597960670065,
        "recall": 0.7430434782608696
      },
      "eval_NAME": {
        "f1": 0.8045016435456014,
        "number": 8840,
        "precision": 0.7926226808650785,
        "recall": 0.8167420814479638
      },
      "eval_QTY": {
        "f1": 0.984450177811868,
        "number": 7169,
        "precision": 0.9842442833240379,
        "recall": 0.9846561584600363
      },
      "eval_RANGE_END": {
        "f1": 0.7531380753138076,
        "number": 94,
        "precision": 0.6206896551724138,
        "recall": 0.9574468085106383
      },
      "eval_UNIT": {
        "f1": 0.949092741935484,
        "number": 5778,
        "precision": 0.9221351616062684,
        "recall": 0.9776739356178609
      },
      "eval_loss": 0.4863777756690979,
      "eval_overall_accuracy": 0.8257766582703611,
      "eval_overall_f1": 0.8442027160866864,
      "eval_overall_precision": 0.8166612536537837,
      "eval_overall_recall": 0.8736666550849519,
      "eval_runtime": 8.6924,
      "eval_samples_per_second": 978.559,
      "eval_steps_per_second": 30.602,
      "step": 9000
    },
    {
      "epoch": 1.88,
      "learning_rate": 1.8653071998944105e-05,
      "loss": 0.4939,
      "step": 9500
    },
    {
      "epoch": 1.98,
      "learning_rate": 1.700323368309906e-05,
      "loss": 0.4872,
      "step": 10000
    },
    {
      "epoch": 1.98,
      "eval_COMMENT": {
        "f1": 0.6780160857908847,
        "number": 6900,
        "precision": 0.6306733167082295,
        "recall": 0.7330434782608696
      },
      "eval_NAME": {
        "f1": 0.8058138887341983,
        "number": 8840,
        "precision": 0.7935724470768893,
        "recall": 0.8184389140271493
      },
      "eval_QTY": {
        "f1": 0.9854324945981738,
        "number": 7169,
        "precision": 0.9848147116188354,
        "recall": 0.9860510531454875
      },
      "eval_RANGE_END": {
        "f1": 0.7733333333333333,
        "number": 94,
        "precision": 0.6641221374045801,
        "recall": 0.925531914893617
      },
      "eval_UNIT": {
        "f1": 0.949941206114564,
        "number": 5778,
        "precision": 0.922813315926893,
        "recall": 0.9787123572170301
      },
      "eval_loss": 0.4847513437271118,
      "eval_overall_accuracy": 0.8277421570872453,
      "eval_overall_f1": 0.8458933535506697,
      "eval_overall_precision": 0.8210898148753843,
      "eval_overall_recall": 0.8722421041659428,
      "eval_runtime": 7.9849,
      "eval_samples_per_second": 1065.267,
      "eval_steps_per_second": 33.313,
      "step": 10000
    },
    {
      "epoch": 2.08,
      "learning_rate": 1.535339536725401e-05,
      "loss": 0.4725,
      "step": 10500
    },
    {
      "epoch": 2.18,
      "learning_rate": 1.3703557051408963e-05,
      "loss": 0.4777,
      "step": 11000
    },
    {
      "epoch": 2.18,
      "eval_COMMENT": {
        "f1": 0.6794315831609847,
        "number": 6900,
        "precision": 0.6294968475707752,
        "recall": 0.7379710144927536
      },
      "eval_NAME": {
        "f1": 0.8072463768115943,
        "number": 8840,
        "precision": 0.7957142857142857,
        "recall": 0.8191176470588235
      },
      "eval_QTY": {
        "f1": 0.9849162011173185,
        "number": 7169,
        "precision": 0.9861557824080548,
        "recall": 0.9836797321802204
      },
      "eval_RANGE_END": {
        "f1": 0.7837837837837838,
        "number": 94,
        "precision": 0.6796875,
        "recall": 0.925531914893617
      },
      "eval_UNIT": {
        "f1": 0.949970570924073,
        "number": 5778,
        "precision": 0.9237939493049877,
        "recall": 0.9776739356178609
      },
      "eval_loss": 0.4820418059825897,
      "eval_overall_accuracy": 0.8306617815433937,
      "eval_overall_f1": 0.8463378478539182,
      "eval_overall_precision": 0.8214040480005231,
      "eval_overall_recall": 0.872832771620166,
      "eval_runtime": 7.6368,
      "eval_samples_per_second": 1113.817,
      "eval_steps_per_second": 34.831,
      "step": 11000
    },
    {
      "epoch": 2.28,
      "learning_rate": 1.2053718735563915e-05,
      "loss": 0.4827,
      "step": 11500
    },
    {
      "epoch": 2.38,
      "learning_rate": 1.0403880419718868e-05,
      "loss": 0.4781,
      "step": 12000
    },
    {
      "epoch": 2.38,
      "eval_COMMENT": {
        "f1": 0.6840429765220851,
        "number": 6900,
        "precision": 0.6305942773294204,
        "recall": 0.7473913043478261
      },
      "eval_NAME": {
        "f1": 0.8081213743864346,
        "number": 8840,
        "precision": 0.7970950704225352,
        "recall": 0.8194570135746606
      },
      "eval_QTY": {
        "f1": 0.9850746268656716,
        "number": 7169,
        "precision": 0.9850746268656716,
        "recall": 0.9850746268656716
      },
      "eval_RANGE_END": {
        "f1": 0.7705627705627707,
        "number": 94,
        "precision": 0.6496350364963503,
        "recall": 0.9468085106382979
      },
      "eval_UNIT": {
        "f1": 0.9505249895002099,
        "number": 5778,
        "precision": 0.923453566182471,
        "recall": 0.9792315680166147
      },
      "eval_loss": 0.48209530115127563,
      "eval_overall_accuracy": 0.8303755438516144,
      "eval_overall_f1": 0.847679892400807,
      "eval_overall_precision": 0.821199387602202,
      "eval_overall_recall": 0.8759250894687467,
      "eval_runtime": 9.3635,
      "eval_samples_per_second": 908.425,
      "eval_steps_per_second": 28.408,
      "step": 12000
    },
    {
      "epoch": 2.47,
      "learning_rate": 8.75404210387382e-06,
      "loss": 0.472,
      "step": 12500
    },
    {
      "epoch": 2.57,
      "learning_rate": 7.104203788028774e-06,
      "loss": 0.4804,
      "step": 13000
    },
    {
      "epoch": 2.57,
      "eval_COMMENT": {
        "f1": 0.6835392320534225,
        "number": 6900,
        "precision": 0.6338080495356038,
        "recall": 0.7417391304347826
      },
      "eval_NAME": {
        "f1": 0.8073610023492561,
        "number": 8840,
        "precision": 0.7985173710998008,
        "recall": 0.8164027149321267
      },
      "eval_QTY": {
        "f1": 0.9856365918281969,
        "number": 7169,
        "precision": 0.9853617733166039,
        "recall": 0.9859115636769424
      },
      "eval_RANGE_END": {
        "f1": 0.7659574468085107,
        "number": 94,
        "precision": 0.6382978723404256,
        "recall": 0.9574468085106383
      },
      "eval_UNIT": {
        "f1": 0.9502095557418274,
        "number": 5778,
        "precision": 0.9213263979193758,
        "recall": 0.9809622706818969
      },
      "eval_loss": 0.4809066951274872,
      "eval_overall_accuracy": 0.8298221509808411,
      "eval_overall_f1": 0.8477425876010781,
      "eval_overall_precision": 0.8228195820661238,
      "eval_overall_recall": 0.8742225773948091,
      "eval_runtime": 7.6878,
      "eval_samples_per_second": 1106.424,
      "eval_steps_per_second": 34.6,
      "step": 13000
    },
    {
      "epoch": 2.67,
      "learning_rate": 5.4543654721837265e-06,
      "loss": 0.4797,
      "step": 13500
    },
    {
      "epoch": 2.77,
      "learning_rate": 3.804527156338679e-06,
      "loss": 0.4721,
      "step": 14000
    },
    {
      "epoch": 2.77,
      "eval_COMMENT": {
        "f1": 0.6857370941990419,
        "number": 6900,
        "precision": 0.6337924249877029,
        "recall": 0.7469565217391304
      },
      "eval_NAME": {
        "f1": 0.8065774804905238,
        "number": 8840,
        "precision": 0.795054945054945,
        "recall": 0.8184389140271493
      },
      "eval_QTY": {
        "f1": 0.9850059278889741,
        "number": 7169,
        "precision": 0.9849372384937238,
        "recall": 0.9850746268656716
      },
      "eval_RANGE_END": {
        "f1": 0.7753303964757708,
        "number": 94,
        "precision": 0.6616541353383458,
        "recall": 0.9361702127659575
      },
      "eval_UNIT": {
        "f1": 0.9506711409395974,
        "number": 5778,
        "precision": 0.9225008140670792,
        "recall": 0.9806161301488404
      },
      "eval_loss": 0.479864239692688,
      "eval_overall_accuracy": 0.8301847187237615,
      "eval_overall_f1": 0.8478253557132766,
      "eval_overall_precision": 0.8216253219024025,
      "eval_overall_recall": 0.8757513637469164,
      "eval_runtime": 9.172,
      "eval_samples_per_second": 927.389,
      "eval_steps_per_second": 29.001,
      "step": 14000
    },
    {
      "epoch": 2.87,
      "learning_rate": 2.154688840493632e-06,
      "loss": 0.4635,
      "step": 14500
    },
    {
      "epoch": 2.97,
      "learning_rate": 5.048505246485845e-07,
      "loss": 0.4792,
      "step": 15000
    },
    {
      "epoch": 2.97,
      "eval_COMMENT": {
        "f1": 0.6867325146823279,
        "number": 6900,
        "precision": 0.6364423552696685,
        "recall": 0.7456521739130435
      },
      "eval_NAME": {
        "f1": 0.8062162312705399,
        "number": 8840,
        "precision": 0.7941402392186986,
        "recall": 0.8186651583710407
      },
      "eval_QTY": {
        "f1": 0.9851495503032839,
        "number": 7169,
        "precision": 0.9848062447727907,
        "recall": 0.985493095271307
      },
      "eval_RANGE_END": {
        "f1": 0.7753303964757708,
        "number": 94,
        "precision": 0.6616541353383458,
        "recall": 0.9361702127659575
      },
      "eval_UNIT": {
        "f1": 0.9499538938720764,
        "number": 5778,
        "precision": 0.921151032352463,
        "recall": 0.9806161301488404
      },
      "eval_loss": 0.4792616069316864,
      "eval_overall_accuracy": 0.8303755438516144,
      "eval_overall_f1": 0.8480045763510331,
      "eval_overall_precision": 0.8220844886641657,
      "eval_overall_recall": 0.8756123831694521,
      "eval_runtime": 9.9787,
      "eval_samples_per_second": 852.419,
      "eval_steps_per_second": 26.657,
      "step": 15000
    },
    {
      "epoch": 3.0,
      "step": 15153,
      "total_flos": 468631722156180.0,
      "train_loss": 0.5153773278912515,
      "train_runtime": 629.31,
      "train_samples_per_second": 770.382,
      "train_steps_per_second": 24.079
    }
  ],
  "logging_steps": 500,
  "max_steps": 15153,
  "num_train_epochs": 3,
  "save_steps": 500,
  "total_flos": 468631722156180.0,
  "trial_name": null,
  "trial_params": null
}