{
  "best_metric": 0.9149305555555556,
  "best_model_checkpoint": "/gpfs/radev/scratch/cohan/jw3278/modernbert-qwen3-stuck-detector/checkpoint-3188",
  "epoch": 5.0,
  "eval_steps": 500,
  "global_step": 3985,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.12547051442910917,
      "grad_norm": 23.81832504272461,
      "learning_rate": 4.874529485570891e-05,
      "loss": 0.5993,
      "step": 100
    },
    {
      "epoch": 0.25094102885821834,
      "grad_norm": 8.124177932739258,
      "learning_rate": 4.749058971141782e-05,
      "loss": 0.5007,
      "step": 200
    },
    {
      "epoch": 0.37641154328732745,
      "grad_norm": 14.401046752929688,
      "learning_rate": 4.623588456712673e-05,
      "loss": 0.513,
      "step": 300
    },
    {
      "epoch": 0.5018820577164367,
      "grad_norm": 1.8089324235916138,
      "learning_rate": 4.4981179422835633e-05,
      "loss": 0.4509,
      "step": 400
    },
    {
      "epoch": 0.6273525721455459,
      "grad_norm": 3.3621015548706055,
      "learning_rate": 4.3726474278544546e-05,
      "loss": 0.3982,
      "step": 500
    },
    {
      "epoch": 0.7528230865746549,
      "grad_norm": 7.305176258087158,
      "learning_rate": 4.247176913425345e-05,
      "loss": 0.4622,
      "step": 600
    },
    {
      "epoch": 0.8782936010037641,
      "grad_norm": 5.204511642456055,
      "learning_rate": 4.1217063989962365e-05,
      "loss": 0.3382,
      "step": 700
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.8844946641556811,
      "eval_f1": 0.8421955403087479,
      "eval_false_negatives": 82,
      "eval_false_positives": 102,
      "eval_loss": 0.37642502784729004,
      "eval_precision": 0.8279932546374368,
      "eval_recall": 0.8568935427574171,
      "eval_runtime": 11.053,
      "eval_samples_per_second": 144.123,
      "eval_steps_per_second": 9.047,
      "eval_true_negatives": 918,
      "eval_true_positives": 491,
      "step": 797
    },
    {
      "epoch": 1.0037641154328734,
      "grad_norm": 20.917531967163086,
      "learning_rate": 3.996235884567127e-05,
      "loss": 0.3475,
      "step": 800
    },
    {
      "epoch": 1.1292346298619824,
      "grad_norm": 30.933334350585938,
      "learning_rate": 3.870765370138018e-05,
      "loss": 0.3489,
      "step": 900
    },
    {
      "epoch": 1.2547051442910915,
      "grad_norm": 6.089908599853516,
      "learning_rate": 3.745294855708908e-05,
      "loss": 0.3437,
      "step": 1000
    },
    {
      "epoch": 1.3801756587202008,
      "grad_norm": 9.339570999145508,
      "learning_rate": 3.6198243412797996e-05,
      "loss": 0.2801,
      "step": 1100
    },
    {
      "epoch": 1.50564617314931,
      "grad_norm": 21.16028594970703,
      "learning_rate": 3.49435382685069e-05,
      "loss": 0.2289,
      "step": 1200
    },
    {
      "epoch": 1.631116687578419,
      "grad_norm": 7.236953258514404,
      "learning_rate": 3.3688833124215815e-05,
      "loss": 0.2722,
      "step": 1300
    },
    {
      "epoch": 1.7565872020075282,
      "grad_norm": 6.876579761505127,
      "learning_rate": 3.2434127979924714e-05,
      "loss": 0.2701,
      "step": 1400
    },
    {
      "epoch": 1.8820577164366374,
      "grad_norm": 0.009545003063976765,
      "learning_rate": 3.117942283563363e-05,
      "loss": 0.2324,
      "step": 1500
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8738229755178908,
      "eval_f1": 0.7908428720083247,
      "eval_false_negatives": 193,
      "eval_false_positives": 8,
      "eval_loss": 0.39299774169921875,
      "eval_precision": 0.979381443298969,
      "eval_recall": 0.6631762652705061,
      "eval_runtime": 11.004,
      "eval_samples_per_second": 144.766,
      "eval_steps_per_second": 9.088,
      "eval_true_negatives": 1012,
      "eval_true_positives": 380,
      "step": 1594
    },
    {
      "epoch": 2.0075282308657467,
      "grad_norm": 4.334559440612793,
      "learning_rate": 2.9924717691342537e-05,
      "loss": 0.2718,
      "step": 1600
    },
    {
      "epoch": 2.132998745294856,
      "grad_norm": 7.7692952156066895,
      "learning_rate": 2.8670012547051446e-05,
      "loss": 0.1646,
      "step": 1700
    },
    {
      "epoch": 2.258469259723965,
      "grad_norm": 3.7941336631774902,
      "learning_rate": 2.7415307402760356e-05,
      "loss": 0.1964,
      "step": 1800
    },
    {
      "epoch": 2.383939774153074,
      "grad_norm": 0.039175648242235184,
      "learning_rate": 2.6160602258469265e-05,
      "loss": 0.1597,
      "step": 1900
    },
    {
      "epoch": 2.509410288582183,
      "grad_norm": 4.6735148429870605,
      "learning_rate": 2.490589711417817e-05,
      "loss": 0.1378,
      "step": 2000
    },
    {
      "epoch": 2.6348808030112925,
      "grad_norm": 0.053006332367658615,
      "learning_rate": 2.3651191969887077e-05,
      "loss": 0.1572,
      "step": 2100
    },
    {
      "epoch": 2.7603513174404015,
      "grad_norm": 15.980093955993652,
      "learning_rate": 2.2396486825595987e-05,
      "loss": 0.1703,
      "step": 2200
    },
    {
      "epoch": 2.8858218318695106,
      "grad_norm": 9.196004867553711,
      "learning_rate": 2.1141781681304896e-05,
      "loss": 0.163,
      "step": 2300
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.9202762084118016,
      "eval_f1": 0.8964955175224124,
      "eval_false_negatives": 23,
      "eval_false_positives": 104,
      "eval_loss": 0.2875446379184723,
      "eval_precision": 0.8409785932721713,
      "eval_recall": 0.9598603839441536,
      "eval_runtime": 11.0499,
      "eval_samples_per_second": 144.164,
      "eval_steps_per_second": 9.05,
      "eval_true_negatives": 916,
      "eval_true_positives": 550,
      "step": 2391
    },
    {
      "epoch": 3.0112923462986196,
      "grad_norm": 12.78604793548584,
      "learning_rate": 1.9887076537013802e-05,
      "loss": 0.1551,
      "step": 2400
    },
    {
      "epoch": 3.136762860727729,
      "grad_norm": 0.02284233085811138,
      "learning_rate": 1.863237139272271e-05,
      "loss": 0.0755,
      "step": 2500
    },
    {
      "epoch": 3.262233375156838,
      "grad_norm": 0.22233843803405762,
      "learning_rate": 1.737766624843162e-05,
      "loss": 0.0673,
      "step": 2600
    },
    {
      "epoch": 3.3877038895859473,
      "grad_norm": 0.02527988888323307,
      "learning_rate": 1.6122961104140527e-05,
      "loss": 0.062,
      "step": 2700
    },
    {
      "epoch": 3.5131744040150563,
      "grad_norm": 0.020394891500473022,
      "learning_rate": 1.4868255959849436e-05,
      "loss": 0.0364,
      "step": 2800
    },
    {
      "epoch": 3.6386449184441654,
      "grad_norm": 0.5171342492103577,
      "learning_rate": 1.3613550815558346e-05,
      "loss": 0.0665,
      "step": 2900
    },
    {
      "epoch": 3.764115432873275,
      "grad_norm": 0.09603510797023773,
      "learning_rate": 1.2358845671267252e-05,
      "loss": 0.0674,
      "step": 3000
    },
    {
      "epoch": 3.889585947302384,
      "grad_norm": 0.0027797145303338766,
      "learning_rate": 1.1104140526976161e-05,
      "loss": 0.0514,
      "step": 3100
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.938480853735091,
      "eval_f1": 0.9149305555555556,
      "eval_false_negatives": 46,
      "eval_false_positives": 52,
      "eval_loss": 0.36789795756340027,
      "eval_precision": 0.9101899827288429,
      "eval_recall": 0.9197207678883071,
      "eval_runtime": 11.0446,
      "eval_samples_per_second": 144.234,
      "eval_steps_per_second": 9.054,
      "eval_true_negatives": 968,
      "eval_true_positives": 527,
      "step": 3188
    },
    {
      "epoch": 4.015056461731493,
      "grad_norm": 0.014151527546346188,
      "learning_rate": 9.849435382685069e-06,
      "loss": 0.0321,
      "step": 3200
    },
    {
      "epoch": 4.1405269761606025,
      "grad_norm": 1.2528949975967407,
      "learning_rate": 8.594730238393977e-06,
      "loss": 0.0016,
      "step": 3300
    },
    {
      "epoch": 4.265997490589712,
      "grad_norm": 0.006553958635777235,
      "learning_rate": 7.3400250941028855e-06,
      "loss": 0.0018,
      "step": 3400
    },
    {
      "epoch": 4.391468005018821,
      "grad_norm": 0.0017017743084579706,
      "learning_rate": 6.085319949811794e-06,
      "loss": 0.0005,
      "step": 3500
    },
    {
      "epoch": 4.51693851944793,
      "grad_norm": 0.012320107780396938,
      "learning_rate": 4.830614805520703e-06,
      "loss": 0.0278,
      "step": 3600
    },
    {
      "epoch": 4.642409033877039,
      "grad_norm": 0.011312933638691902,
      "learning_rate": 3.5759096612296117e-06,
      "loss": 0.0167,
      "step": 3700
    },
    {
      "epoch": 4.767879548306148,
      "grad_norm": 0.00529805151745677,
      "learning_rate": 2.3212045169385195e-06,
      "loss": 0.0133,
      "step": 3800
    },
    {
      "epoch": 4.893350062735257,
      "grad_norm": 0.8228544592857361,
      "learning_rate": 1.066499372647428e-06,
      "loss": 0.0068,
      "step": 3900
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.9372253609541745,
      "eval_f1": 0.911504424778761,
      "eval_false_negatives": 58,
      "eval_false_positives": 42,
      "eval_loss": 0.4694787561893463,
      "eval_precision": 0.9245960502692998,
      "eval_recall": 0.8987783595113438,
      "eval_runtime": 11.0039,
      "eval_samples_per_second": 144.767,
      "eval_steps_per_second": 9.088,
      "eval_true_negatives": 978,
      "eval_true_positives": 515,
      "step": 3985
    }
  ],
  "logging_steps": 100,
  "max_steps": 3985,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 5,
  "save_steps": 500,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 4.340573442551808e+16,
  "train_batch_size": 8,
  "trial_name": null,
  "trial_params": null
}