{
  "best_metric": 0.8000965798413886,
  "best_model_checkpoint": "/content/drive/MyDrive/Projects/ModernBertTuning/2025-02-19_17:10:03/checkpoint-3300",
  "epoch": 5.809859154929578,
  "eval_steps": 100,
  "global_step": 3300,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.176056338028169,
      "grad_norm": 5.5684285163879395,
      "learning_rate": 8.802816901408451e-06,
      "loss": 0.9029,
      "step": 100
    },
    {
      "epoch": 0.176056338028169,
      "eval_action_accuracy": 0.7632042253521126,
      "eval_action_f1": 0.3275,
      "eval_action_precision": 0.6787564766839378,
      "eval_action_recall": 0.2158154859967051,
      "eval_combined_f1": 0.39009832159236646,
      "eval_loss": 1.5983511209487915,
      "eval_runtime": 53.3406,
      "eval_samples_per_second": 42.594,
      "eval_severity_accuracy": 0.4942781690140845,
      "eval_severity_f1": 0.4526966431847329,
      "eval_severity_precision": 0.5190575556900403,
      "eval_severity_recall": 0.4942781690140845,
      "eval_steps_per_second": 2.662,
      "step": 100
    },
    {
      "epoch": 0.352112676056338,
      "grad_norm": 10.437986373901367,
      "learning_rate": 1.7605633802816902e-05,
      "loss": 0.6875,
      "step": 200
    },
    {
      "epoch": 0.352112676056338,
      "eval_action_accuracy": 0.8393485915492958,
      "eval_action_f1": 0.7253574115876599,
      "eval_action_precision": 0.667590027700831,
      "eval_action_recall": 0.7940691927512356,
      "eval_combined_f1": 0.6889564202307465,
      "eval_loss": 1.1708728075027466,
      "eval_runtime": 53.402,
      "eval_samples_per_second": 42.545,
      "eval_severity_accuracy": 0.6615316901408451,
      "eval_severity_f1": 0.6525554288738333,
      "eval_severity_precision": 0.6774842234837531,
      "eval_severity_recall": 0.6615316901408451,
      "eval_steps_per_second": 2.659,
      "step": 200
    },
    {
      "epoch": 0.528169014084507,
      "grad_norm": 8.755282402038574,
      "learning_rate": 2.640845070422535e-05,
      "loss": 0.5914,
      "step": 300
    },
    {
      "epoch": 0.528169014084507,
      "eval_action_accuracy": 0.8653169014084507,
      "eval_action_f1": 0.7688821752265861,
      "eval_action_precision": 0.7099023709902371,
      "eval_action_recall": 0.8385502471169687,
      "eval_combined_f1": 0.7012112761418396,
      "eval_loss": 1.1276360750198364,
      "eval_runtime": 53.4046,
      "eval_samples_per_second": 42.543,
      "eval_severity_accuracy": 0.6307218309859155,
      "eval_severity_f1": 0.6335403770570933,
      "eval_severity_precision": 0.6726339478831171,
      "eval_severity_recall": 0.6307218309859155,
      "eval_steps_per_second": 2.659,
      "step": 300
    },
    {
      "epoch": 0.704225352112676,
      "grad_norm": 4.745133399963379,
      "learning_rate": 3.5211267605633805e-05,
      "loss": 0.5167,
      "step": 400
    },
    {
      "epoch": 0.704225352112676,
      "eval_action_accuracy": 0.8820422535211268,
      "eval_action_f1": 0.750465549348231,
      "eval_action_precision": 0.8629550321199143,
      "eval_action_recall": 0.6639209225700164,
      "eval_combined_f1": 0.7324396418875856,
      "eval_loss": 1.0260754823684692,
      "eval_runtime": 53.3961,
      "eval_samples_per_second": 42.55,
      "eval_severity_accuracy": 0.715669014084507,
      "eval_severity_f1": 0.7144137344269402,
      "eval_severity_precision": 0.717618479930179,
      "eval_severity_recall": 0.715669014084507,
      "eval_steps_per_second": 2.659,
      "step": 400
    },
    {
      "epoch": 0.8802816901408451,
      "grad_norm": 4.491367816925049,
      "learning_rate": 4.4014084507042256e-05,
      "loss": 0.5228,
      "step": 500
    },
    {
      "epoch": 0.8802816901408451,
      "eval_action_accuracy": 0.8811619718309859,
      "eval_action_f1": 0.7652173913043478,
      "eval_action_precision": 0.8103130755064457,
      "eval_action_recall": 0.7248764415156508,
      "eval_combined_f1": 0.7303003562324298,
      "eval_loss": 1.0010697841644287,
      "eval_runtime": 53.3858,
      "eval_samples_per_second": 42.558,
      "eval_severity_accuracy": 0.6941021126760564,
      "eval_severity_f1": 0.6953833211605118,
      "eval_severity_precision": 0.7027972731452775,
      "eval_severity_recall": 0.6941021126760564,
      "eval_steps_per_second": 2.66,
      "step": 500
    },
    {
      "epoch": 1.056338028169014,
      "grad_norm": 7.762014865875244,
      "learning_rate": 4.968701095461659e-05,
      "loss": 0.4707,
      "step": 600
    },
    {
      "epoch": 1.056338028169014,
      "eval_action_accuracy": 0.8816021126760564,
      "eval_action_f1": 0.7497674418604651,
      "eval_action_precision": 0.8611111111111112,
      "eval_action_recall": 0.6639209225700164,
      "eval_combined_f1": 0.7288028162963264,
      "eval_loss": 1.05709969997406,
      "eval_runtime": 53.3852,
      "eval_samples_per_second": 42.559,
      "eval_severity_accuracy": 0.710387323943662,
      "eval_severity_f1": 0.7078381907321877,
      "eval_severity_precision": 0.7185583041734807,
      "eval_severity_recall": 0.710387323943662,
      "eval_steps_per_second": 2.66,
      "step": 600
    },
    {
      "epoch": 1.232394366197183,
      "grad_norm": 9.271745681762695,
      "learning_rate": 4.870892018779343e-05,
      "loss": 0.4103,
      "step": 700
    },
    {
      "epoch": 1.232394366197183,
      "eval_action_accuracy": 0.8503521126760564,
      "eval_action_f1": 0.621380846325167,
      "eval_action_precision": 0.9587628865979382,
      "eval_action_recall": 0.4596375617792422,
      "eval_combined_f1": 0.6340972974514396,
      "eval_loss": 1.1522221565246582,
      "eval_runtime": 53.3981,
      "eval_samples_per_second": 42.548,
      "eval_severity_accuracy": 0.6553697183098591,
      "eval_severity_f1": 0.6468137485777123,
      "eval_severity_precision": 0.7014511458318834,
      "eval_severity_recall": 0.6553697183098591,
      "eval_steps_per_second": 2.659,
      "step": 700
    },
    {
      "epoch": 1.408450704225352,
      "grad_norm": 7.816728115081787,
      "learning_rate": 4.773082942097027e-05,
      "loss": 0.4,
      "step": 800
    },
    {
      "epoch": 1.408450704225352,
      "eval_action_accuracy": 0.8820422535211268,
      "eval_action_f1": 0.7922480620155039,
      "eval_action_precision": 0.7481698389458272,
      "eval_action_recall": 0.841845140032949,
      "eval_combined_f1": 0.7584899763574804,
      "eval_loss": 0.9812976717948914,
      "eval_runtime": 53.3978,
      "eval_samples_per_second": 42.549,
      "eval_severity_accuracy": 0.7235915492957746,
      "eval_severity_f1": 0.724731890699457,
      "eval_severity_precision": 0.7317594548978289,
      "eval_severity_recall": 0.7235915492957746,
      "eval_steps_per_second": 2.659,
      "step": 800
    },
    {
      "epoch": 1.584507042253521,
      "grad_norm": 7.098080158233643,
      "learning_rate": 4.675273865414711e-05,
      "loss": 0.4179,
      "step": 900
    },
    {
      "epoch": 1.584507042253521,
      "eval_action_accuracy": 0.894806338028169,
      "eval_action_f1": 0.7879325643300799,
      "eval_action_precision": 0.8538461538461538,
      "eval_action_recall": 0.7314662273476112,
      "eval_combined_f1": 0.7270919018906684,
      "eval_loss": 1.177077054977417,
      "eval_runtime": 53.4076,
      "eval_samples_per_second": 42.541,
      "eval_severity_accuracy": 0.6822183098591549,
      "eval_severity_f1": 0.666251239451257,
      "eval_severity_precision": 0.7051808476907062,
      "eval_severity_recall": 0.6822183098591549,
      "eval_steps_per_second": 2.659,
      "step": 900
    },
    {
      "epoch": 1.76056338028169,
      "grad_norm": 3.6362385749816895,
      "learning_rate": 4.577464788732395e-05,
      "loss": 0.381,
      "step": 1000
    },
    {
      "epoch": 1.76056338028169,
      "eval_action_accuracy": 0.8934859154929577,
      "eval_action_f1": 0.8,
      "eval_action_precision": 0.802653399668325,
      "eval_action_recall": 0.7973640856672158,
      "eval_combined_f1": 0.7276179625363967,
      "eval_loss": 1.0336343050003052,
      "eval_runtime": 53.3973,
      "eval_samples_per_second": 42.549,
      "eval_severity_accuracy": 0.6597711267605634,
      "eval_severity_f1": 0.6552359250727933,
      "eval_severity_precision": 0.690194073616287,
      "eval_severity_recall": 0.6597711267605634,
      "eval_steps_per_second": 2.659,
      "step": 1000
    },
    {
      "epoch": 1.936619718309859,
      "grad_norm": 4.175089359283447,
      "learning_rate": 4.479655712050079e-05,
      "loss": 0.384,
      "step": 1100
    },
    {
      "epoch": 1.936619718309859,
      "eval_action_accuracy": 0.8899647887323944,
      "eval_action_f1": 0.7987117552334944,
      "eval_action_precision": 0.7811023622047244,
      "eval_action_recall": 0.8171334431630972,
      "eval_combined_f1": 0.7613638710145674,
      "eval_loss": 0.9859868288040161,
      "eval_runtime": 53.3975,
      "eval_samples_per_second": 42.549,
      "eval_severity_accuracy": 0.7240316901408451,
      "eval_severity_f1": 0.7240159867956404,
      "eval_severity_precision": 0.7333723870533466,
      "eval_severity_recall": 0.7240316901408451,
      "eval_steps_per_second": 2.659,
      "step": 1100
    },
    {
      "epoch": 2.112676056338028,
      "grad_norm": 5.42644739151001,
      "learning_rate": 4.3818466353677626e-05,
      "loss": 0.287,
      "step": 1200
    },
    {
      "epoch": 2.112676056338028,
      "eval_action_accuracy": 0.8798415492957746,
      "eval_action_f1": 0.7272727272727273,
      "eval_action_precision": 0.9238578680203046,
      "eval_action_recall": 0.5996705107084019,
      "eval_combined_f1": 0.7194719198906718,
      "eval_loss": 1.2687214612960815,
      "eval_runtime": 53.3984,
      "eval_samples_per_second": 42.548,
      "eval_severity_accuracy": 0.7130281690140845,
      "eval_severity_f1": 0.7116711125086165,
      "eval_severity_precision": 0.7299090747607111,
      "eval_severity_recall": 0.7130281690140845,
      "eval_steps_per_second": 2.659,
      "step": 1200
    },
    {
      "epoch": 2.288732394366197,
      "grad_norm": 4.9831342697143555,
      "learning_rate": 4.284037558685447e-05,
      "loss": 0.234,
      "step": 1300
    },
    {
      "epoch": 2.288732394366197,
      "eval_action_accuracy": 0.8758802816901409,
      "eval_action_f1": 0.7904903417533432,
      "eval_action_precision": 0.7198917456021651,
      "eval_action_recall": 0.8764415156507414,
      "eval_combined_f1": 0.7408713797328776,
      "eval_loss": 1.2081983089447021,
      "eval_runtime": 53.3856,
      "eval_samples_per_second": 42.558,
      "eval_severity_accuracy": 0.6954225352112676,
      "eval_severity_f1": 0.691252417712412,
      "eval_severity_precision": 0.710940220005423,
      "eval_severity_recall": 0.6954225352112676,
      "eval_steps_per_second": 2.66,
      "step": 1300
    },
    {
      "epoch": 2.464788732394366,
      "grad_norm": 8.448158264160156,
      "learning_rate": 4.18622848200313e-05,
      "loss": 0.2284,
      "step": 1400
    },
    {
      "epoch": 2.464788732394366,
      "eval_action_accuracy": 0.9097711267605634,
      "eval_action_f1": 0.820333041191937,
      "eval_action_precision": 0.8764044943820225,
      "eval_action_recall": 0.771004942339374,
      "eval_combined_f1": 0.7709403023211041,
      "eval_loss": 1.1236168146133423,
      "eval_runtime": 53.3827,
      "eval_samples_per_second": 42.561,
      "eval_severity_accuracy": 0.7205105633802817,
      "eval_severity_f1": 0.7215475634502713,
      "eval_severity_precision": 0.7388563066943578,
      "eval_severity_recall": 0.7205105633802817,
      "eval_steps_per_second": 2.66,
      "step": 1400
    },
    {
      "epoch": 2.640845070422535,
      "grad_norm": 3.273317575454712,
      "learning_rate": 4.0884194053208144e-05,
      "loss": 0.2147,
      "step": 1500
    },
    {
      "epoch": 2.640845070422535,
      "eval_action_accuracy": 0.9066901408450704,
      "eval_action_f1": 0.8233333333333334,
      "eval_action_precision": 0.8330522765598651,
      "eval_action_recall": 0.8138385502471169,
      "eval_combined_f1": 0.7802089055740911,
      "eval_loss": 1.08566153049469,
      "eval_runtime": 53.3876,
      "eval_samples_per_second": 42.557,
      "eval_severity_accuracy": 0.7394366197183099,
      "eval_severity_f1": 0.7370844778148489,
      "eval_severity_precision": 0.7443892609781334,
      "eval_severity_recall": 0.7394366197183099,
      "eval_steps_per_second": 2.66,
      "step": 1500
    },
    {
      "epoch": 2.816901408450704,
      "grad_norm": 4.577557563781738,
      "learning_rate": 3.990610328638498e-05,
      "loss": 0.2258,
      "step": 1600
    },
    {
      "epoch": 2.816901408450704,
      "eval_action_accuracy": 0.8939260563380281,
      "eval_action_f1": 0.7872903795233892,
      "eval_action_precision": 0.8479087452471483,
      "eval_action_recall": 0.7347611202635914,
      "eval_combined_f1": 0.7659288423194548,
      "eval_loss": 1.0919924974441528,
      "eval_runtime": 53.4132,
      "eval_samples_per_second": 42.536,
      "eval_severity_accuracy": 0.7442781690140845,
      "eval_severity_f1": 0.7445673051155204,
      "eval_severity_precision": 0.7537282342243937,
      "eval_severity_recall": 0.7442781690140845,
      "eval_steps_per_second": 2.659,
      "step": 1600
    },
    {
      "epoch": 2.992957746478873,
      "grad_norm": 5.347113609313965,
      "learning_rate": 3.892801251956182e-05,
      "loss": 0.2055,
      "step": 1700
    },
    {
      "epoch": 2.992957746478873,
      "eval_action_accuracy": 0.90625,
      "eval_action_f1": 0.813321647677476,
      "eval_action_precision": 0.8689138576779026,
      "eval_action_recall": 0.7644151565074135,
      "eval_combined_f1": 0.7792135732932114,
      "eval_loss": 1.0456966161727905,
      "eval_runtime": 53.4219,
      "eval_samples_per_second": 42.529,
      "eval_severity_accuracy": 0.7447183098591549,
      "eval_severity_f1": 0.7451054989089468,
      "eval_severity_precision": 0.7459307596916318,
      "eval_severity_recall": 0.7447183098591549,
      "eval_steps_per_second": 2.658,
      "step": 1700
    },
    {
      "epoch": 3.169014084507042,
      "grad_norm": 8.061402320861816,
      "learning_rate": 3.7949921752738656e-05,
      "loss": 0.1559,
      "step": 1800
    },
    {
      "epoch": 3.169014084507042,
      "eval_action_accuracy": 0.9049295774647887,
      "eval_action_f1": 0.8036363636363636,
      "eval_action_precision": 0.896551724137931,
      "eval_action_recall": 0.728171334431631,
      "eval_combined_f1": 0.7705730204982484,
      "eval_loss": 1.3266334533691406,
      "eval_runtime": 53.3935,
      "eval_samples_per_second": 42.552,
      "eval_severity_accuracy": 0.7376760563380281,
      "eval_severity_f1": 0.7375096773601331,
      "eval_severity_precision": 0.7543615150494364,
      "eval_severity_recall": 0.7376760563380281,
      "eval_steps_per_second": 2.659,
      "step": 1800
    },
    {
      "epoch": 3.345070422535211,
      "grad_norm": 4.052455902099609,
      "learning_rate": 3.69718309859155e-05,
      "loss": 0.1174,
      "step": 1900
    },
    {
      "epoch": 3.345070422535211,
      "eval_action_accuracy": 0.8992077464788732,
      "eval_action_f1": 0.7893284268629255,
      "eval_action_precision": 0.89375,
      "eval_action_recall": 0.7067545304777595,
      "eval_combined_f1": 0.7628428370446565,
      "eval_loss": 1.4744161367416382,
      "eval_runtime": 53.3995,
      "eval_samples_per_second": 42.547,
      "eval_severity_accuracy": 0.7376760563380281,
      "eval_severity_f1": 0.7363572472263874,
      "eval_severity_precision": 0.7497424115719165,
      "eval_severity_recall": 0.7376760563380281,
      "eval_steps_per_second": 2.659,
      "step": 1900
    },
    {
      "epoch": 3.52112676056338,
      "grad_norm": 6.726060390472412,
      "learning_rate": 3.599374021909233e-05,
      "loss": 0.1325,
      "step": 2000
    },
    {
      "epoch": 3.52112676056338,
      "eval_action_accuracy": 0.9044894366197183,
      "eval_action_f1": 0.8021877848678214,
      "eval_action_precision": 0.8979591836734694,
      "eval_action_recall": 0.7248764415156508,
      "eval_combined_f1": 0.7739176088843763,
      "eval_loss": 1.592252492904663,
      "eval_runtime": 53.4078,
      "eval_samples_per_second": 42.541,
      "eval_severity_accuracy": 0.7460387323943662,
      "eval_severity_f1": 0.7456474329009313,
      "eval_severity_precision": 0.7468342652042663,
      "eval_severity_recall": 0.7460387323943662,
      "eval_steps_per_second": 2.659,
      "step": 2000
    },
    {
      "epoch": 3.697183098591549,
      "grad_norm": 1.6415315866470337,
      "learning_rate": 3.5015649452269174e-05,
      "loss": 0.1318,
      "step": 2100
    },
    {
      "epoch": 3.697183098591549,
      "eval_action_accuracy": 0.909330985915493,
      "eval_action_f1": 0.815742397137746,
      "eval_action_precision": 0.8923679060665362,
      "eval_action_recall": 0.7512355848434926,
      "eval_combined_f1": 0.7744664753736676,
      "eval_loss": 1.3085289001464844,
      "eval_runtime": 53.3948,
      "eval_samples_per_second": 42.551,
      "eval_severity_accuracy": 0.7345950704225352,
      "eval_severity_f1": 0.7331905536095894,
      "eval_severity_precision": 0.7463951981865823,
      "eval_severity_recall": 0.7345950704225352,
      "eval_steps_per_second": 2.659,
      "step": 2100
    },
    {
      "epoch": 3.873239436619718,
      "grad_norm": 1.168869137763977,
      "learning_rate": 3.4037558685446016e-05,
      "loss": 0.1303,
      "step": 2200
    },
    {
      "epoch": 3.873239436619718,
      "eval_action_accuracy": 0.9080105633802817,
      "eval_action_f1": 0.8271298593879239,
      "eval_action_precision": 0.8305647840531561,
      "eval_action_recall": 0.8237232289950577,
      "eval_combined_f1": 0.7891663886796052,
      "eval_loss": 1.2526799440383911,
      "eval_runtime": 53.3936,
      "eval_samples_per_second": 42.552,
      "eval_severity_accuracy": 0.7513204225352113,
      "eval_severity_f1": 0.7512029179712866,
      "eval_severity_precision": 0.755462312536582,
      "eval_severity_recall": 0.7513204225352113,
      "eval_steps_per_second": 2.659,
      "step": 2200
    },
    {
      "epoch": 4.049295774647887,
      "grad_norm": 1.3796656131744385,
      "learning_rate": 3.305946791862285e-05,
      "loss": 0.1226,
      "step": 2300
    },
    {
      "epoch": 4.049295774647887,
      "eval_action_accuracy": 0.9163732394366197,
      "eval_action_f1": 0.8403361344537815,
      "eval_action_precision": 0.8576329331046312,
      "eval_action_recall": 0.8237232289950577,
      "eval_combined_f1": 0.7939814504731293,
      "eval_loss": 1.3127936124801636,
      "eval_runtime": 53.3969,
      "eval_samples_per_second": 42.549,
      "eval_severity_accuracy": 0.7508802816901409,
      "eval_severity_f1": 0.747626766492477,
      "eval_severity_precision": 0.7590158294773308,
      "eval_severity_recall": 0.7508802816901409,
      "eval_steps_per_second": 2.659,
      "step": 2300
    },
    {
      "epoch": 4.225352112676056,
      "grad_norm": 4.387660503387451,
      "learning_rate": 3.208137715179969e-05,
      "loss": 0.0703,
      "step": 2400
    },
    {
      "epoch": 4.225352112676056,
      "eval_action_accuracy": 0.9119718309859155,
      "eval_action_f1": 0.8242530755711776,
      "eval_action_precision": 0.8832391713747646,
      "eval_action_recall": 0.7726523887973641,
      "eval_combined_f1": 0.7795952339344467,
      "eval_loss": 1.73722505569458,
      "eval_runtime": 53.3612,
      "eval_samples_per_second": 42.578,
      "eval_severity_accuracy": 0.7372359154929577,
      "eval_severity_f1": 0.7349373922977159,
      "eval_severity_precision": 0.7478231277482102,
      "eval_severity_recall": 0.7372359154929577,
      "eval_steps_per_second": 2.661,
      "step": 2400
    },
    {
      "epoch": 4.401408450704225,
      "grad_norm": 3.3632314205169678,
      "learning_rate": 3.110328638497653e-05,
      "loss": 0.0749,
      "step": 2500
    },
    {
      "epoch": 4.401408450704225,
      "eval_action_accuracy": 0.9058098591549296,
      "eval_action_f1": 0.8145580589254766,
      "eval_action_precision": 0.8592321755027422,
      "eval_action_recall": 0.7742998352553542,
      "eval_combined_f1": 0.7737646532726048,
      "eval_loss": 1.8485854864120483,
      "eval_runtime": 53.3448,
      "eval_samples_per_second": 42.591,
      "eval_severity_accuracy": 0.7319542253521126,
      "eval_severity_f1": 0.7329712476197329,
      "eval_severity_precision": 0.7371728830594799,
      "eval_severity_recall": 0.7319542253521126,
      "eval_steps_per_second": 2.662,
      "step": 2500
    },
    {
      "epoch": 4.577464788732394,
      "grad_norm": 4.060558319091797,
      "learning_rate": 3.0125195618153366e-05,
      "loss": 0.0776,
      "step": 2600
    },
    {
      "epoch": 4.577464788732394,
      "eval_action_accuracy": 0.8877640845070423,
      "eval_action_f1": 0.746772591857001,
      "eval_action_precision": 0.94,
      "eval_action_recall": 0.6194398682042833,
      "eval_combined_f1": 0.7451658251590665,
      "eval_loss": 2.1151413917541504,
      "eval_runtime": 53.345,
      "eval_samples_per_second": 42.591,
      "eval_severity_accuracy": 0.7438380281690141,
      "eval_severity_f1": 0.7435590584611321,
      "eval_severity_precision": 0.7627996939449937,
      "eval_severity_recall": 0.7438380281690141,
      "eval_steps_per_second": 2.662,
      "step": 2600
    },
    {
      "epoch": 4.753521126760563,
      "grad_norm": 1.934782862663269,
      "learning_rate": 2.9147104851330205e-05,
      "loss": 0.0842,
      "step": 2700
    },
    {
      "epoch": 4.753521126760563,
      "eval_action_accuracy": 0.9154929577464789,
      "eval_action_f1": 0.8449111470113085,
      "eval_action_precision": 0.8288431061806656,
      "eval_action_recall": 0.8616144975288303,
      "eval_combined_f1": 0.7964890047427877,
      "eval_loss": 1.6187307834625244,
      "eval_runtime": 53.3536,
      "eval_samples_per_second": 42.584,
      "eval_severity_accuracy": 0.7477992957746479,
      "eval_severity_f1": 0.748066862474267,
      "eval_severity_precision": 0.751069587929407,
      "eval_severity_recall": 0.7477992957746479,
      "eval_steps_per_second": 2.661,
      "step": 2700
    },
    {
      "epoch": 4.929577464788732,
      "grad_norm": 0.15578486025333405,
      "learning_rate": 2.8169014084507046e-05,
      "loss": 0.0718,
      "step": 2800
    },
    {
      "epoch": 4.929577464788732,
      "eval_action_accuracy": 0.9080105633802817,
      "eval_action_f1": 0.8125560538116592,
      "eval_action_precision": 0.8917322834645669,
      "eval_action_recall": 0.7462932454695222,
      "eval_combined_f1": 0.7831897220338088,
      "eval_loss": 1.7883614301681519,
      "eval_runtime": 53.3648,
      "eval_samples_per_second": 42.575,
      "eval_severity_accuracy": 0.7535211267605634,
      "eval_severity_f1": 0.7538233902559583,
      "eval_severity_precision": 0.7606986906314367,
      "eval_severity_recall": 0.7535211267605634,
      "eval_steps_per_second": 2.661,
      "step": 2800
    },
    {
      "epoch": 5.105633802816901,
      "grad_norm": 0.7947894930839539,
      "learning_rate": 2.719092331768388e-05,
      "loss": 0.0732,
      "step": 2900
    },
    {
      "epoch": 5.105633802816901,
      "eval_action_accuracy": 0.909330985915493,
      "eval_action_f1": 0.83003300330033,
      "eval_action_precision": 0.8314049586776859,
      "eval_action_recall": 0.828665568369028,
      "eval_combined_f1": 0.7889699480003767,
      "eval_loss": 1.63761305809021,
      "eval_runtime": 53.3715,
      "eval_samples_per_second": 42.57,
      "eval_severity_accuracy": 0.7473591549295775,
      "eval_severity_f1": 0.7479068927004233,
      "eval_severity_precision": 0.7576437021292298,
      "eval_severity_recall": 0.7473591549295775,
      "eval_steps_per_second": 2.661,
      "step": 2900
    },
    {
      "epoch": 5.28169014084507,
      "grad_norm": 6.216972827911377,
      "learning_rate": 2.6212832550860723e-05,
      "loss": 0.0433,
      "step": 3000
    },
    {
      "epoch": 5.28169014084507,
      "eval_action_accuracy": 0.9119718309859155,
      "eval_action_f1": 0.823943661971831,
      "eval_action_precision": 0.8846880907372401,
      "eval_action_recall": 0.771004942339374,
      "eval_combined_f1": 0.7916775895426225,
      "eval_loss": 2.1584179401397705,
      "eval_runtime": 53.3702,
      "eval_samples_per_second": 42.571,
      "eval_severity_accuracy": 0.7588028169014085,
      "eval_severity_f1": 0.759411517113414,
      "eval_severity_precision": 0.7667529788427636,
      "eval_severity_recall": 0.7588028169014085,
      "eval_steps_per_second": 2.661,
      "step": 3000
    },
    {
      "epoch": 5.457746478873239,
      "grad_norm": 7.994290351867676,
      "learning_rate": 2.5234741784037558e-05,
      "loss": 0.0437,
      "step": 3100
    },
    {
      "epoch": 5.457746478873239,
      "eval_action_accuracy": 0.9132922535211268,
      "eval_action_f1": 0.8365145228215768,
      "eval_action_precision": 0.842809364548495,
      "eval_action_recall": 0.8303130148270181,
      "eval_combined_f1": 0.7970826496584587,
      "eval_loss": 1.9302828311920166,
      "eval_runtime": 53.3908,
      "eval_samples_per_second": 42.554,
      "eval_severity_accuracy": 0.758362676056338,
      "eval_severity_f1": 0.7576507764953404,
      "eval_severity_precision": 0.7661980956912092,
      "eval_severity_recall": 0.758362676056338,
      "eval_steps_per_second": 2.66,
      "step": 3100
    },
    {
      "epoch": 5.633802816901408,
      "grad_norm": 0.0936056599020958,
      "learning_rate": 2.42566510172144e-05,
      "loss": 0.0329,
      "step": 3200
    },
    {
      "epoch": 5.633802816901408,
      "eval_action_accuracy": 0.9137323943661971,
      "eval_action_f1": 0.833616298811545,
      "eval_action_precision": 0.8598949211908932,
      "eval_action_recall": 0.8088962108731467,
      "eval_combined_f1": 0.7923034119567994,
      "eval_loss": 2.4584481716156006,
      "eval_runtime": 53.4118,
      "eval_samples_per_second": 42.537,
      "eval_severity_accuracy": 0.7522007042253521,
      "eval_severity_f1": 0.7509905251020539,
      "eval_severity_precision": 0.7646859410918823,
      "eval_severity_recall": 0.7522007042253521,
      "eval_steps_per_second": 2.659,
      "step": 3200
    },
    {
      "epoch": 5.809859154929578,
      "grad_norm": 0.029598630964756012,
      "learning_rate": 2.327856025039124e-05,
      "loss": 0.0566,
      "step": 3300
    },
    {
      "epoch": 5.809859154929578,
      "eval_action_accuracy": 0.9132922535211268,
      "eval_action_f1": 0.841000807102502,
      "eval_action_precision": 0.8243670886075949,
      "eval_action_recall": 0.8583196046128501,
      "eval_combined_f1": 0.8000965798413886,
      "eval_loss": 1.6451172828674316,
      "eval_runtime": 53.3855,
      "eval_samples_per_second": 42.558,
      "eval_severity_accuracy": 0.7588028169014085,
      "eval_severity_f1": 0.7591923525802752,
      "eval_severity_precision": 0.7658252855908393,
      "eval_severity_recall": 0.7588028169014085,
      "eval_steps_per_second": 2.66,
      "step": 3300
    }
  ],
  "logging_steps": 100,
  "max_steps": 5680,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 10,
  "save_steps": 100,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 5.581830484199424e+16,
  "train_batch_size": 16,
  "trial_name": null,
  "trial_params": null
}