trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 5.0,
  "eval_steps": 500,
  "global_step": 480,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "grad_norm": 2.027179718017578,
      "learning_rate": 4e-05,
      "loss": 0.241,
      "step": 96
    },
    {
      "epoch": 1.0,
      "eval_LOCATION_f1": 0.8723404255319148,
      "eval_LOCATION_number": 86,
      "eval_LOCATION_precision": 0.803921568627451,
      "eval_LOCATION_recall": 0.9534883720930233,
      "eval_ORGANIZATION_f1": 0.9008498583569405,
      "eval_ORGANIZATION_number": 178,
      "eval_ORGANIZATION_precision": 0.9085714285714286,
      "eval_ORGANIZATION_recall": 0.8932584269662921,
      "eval_PERSON_f1": 0.9612403100775193,
      "eval_PERSON_number": 128,
      "eval_PERSON_precision": 0.9538461538461539,
      "eval_PERSON_recall": 0.96875,
      "eval_loss": 0.05585619434714317,
      "eval_overall_accuracy": 0.9832658569500675,
      "eval_overall_f1": 0.9136420525657072,
      "eval_overall_precision": 0.8968058968058968,
      "eval_overall_recall": 0.9311224489795918,
      "eval_runtime": 1.3404,
      "eval_samples_per_second": 126.825,
      "eval_steps_per_second": 2.238,
      "step": 96
    },
    {
      "epoch": 2.0,
      "grad_norm": 1.5001760721206665,
      "learning_rate": 3e-05,
      "loss": 0.0545,
      "step": 192
    },
    {
      "epoch": 2.0,
      "eval_LOCATION_f1": 0.8961748633879781,
      "eval_LOCATION_number": 86,
      "eval_LOCATION_precision": 0.845360824742268,
      "eval_LOCATION_recall": 0.9534883720930233,
      "eval_ORGANIZATION_f1": 0.9111111111111112,
      "eval_ORGANIZATION_number": 178,
      "eval_ORGANIZATION_precision": 0.9010989010989011,
      "eval_ORGANIZATION_recall": 0.9213483146067416,
      "eval_PERSON_f1": 0.9763779527559054,
      "eval_PERSON_number": 128,
      "eval_PERSON_precision": 0.9841269841269841,
      "eval_PERSON_recall": 0.96875,
      "eval_loss": 0.05511007830500603,
      "eval_overall_accuracy": 0.9827260458839406,
      "eval_overall_f1": 0.9284818067754077,
      "eval_overall_precision": 0.9135802469135802,
      "eval_overall_recall": 0.9438775510204082,
      "eval_runtime": 1.2378,
      "eval_samples_per_second": 137.339,
      "eval_steps_per_second": 2.424,
      "step": 192
    },
    {
      "epoch": 3.0,
      "grad_norm": 3.457658290863037,
      "learning_rate": 2e-05,
      "loss": 0.0286,
      "step": 288
    },
    {
      "epoch": 3.0,
      "eval_LOCATION_f1": 0.9189189189189189,
      "eval_LOCATION_number": 86,
      "eval_LOCATION_precision": 0.8585858585858586,
      "eval_LOCATION_recall": 0.9883720930232558,
      "eval_ORGANIZATION_f1": 0.9344729344729344,
      "eval_ORGANIZATION_number": 178,
      "eval_ORGANIZATION_precision": 0.9479768786127167,
      "eval_ORGANIZATION_recall": 0.9213483146067416,
      "eval_PERSON_f1": 0.9763779527559054,
      "eval_PERSON_number": 128,
      "eval_PERSON_precision": 0.9841269841269841,
      "eval_PERSON_recall": 0.96875,
      "eval_loss": 0.04845225811004639,
      "eval_overall_accuracy": 0.9870445344129555,
      "eval_overall_f1": 0.9443037974683544,
      "eval_overall_precision": 0.9371859296482412,
      "eval_overall_recall": 0.951530612244898,
      "eval_runtime": 1.2338,
      "eval_samples_per_second": 137.782,
      "eval_steps_per_second": 2.431,
      "step": 288
    },
    {
      "epoch": 4.0,
      "grad_norm": 2.5664279460906982,
      "learning_rate": 1e-05,
      "loss": 0.0151,
      "step": 384
    },
    {
      "epoch": 4.0,
      "eval_LOCATION_f1": 0.9378531073446328,
      "eval_LOCATION_number": 86,
      "eval_LOCATION_precision": 0.9120879120879121,
      "eval_LOCATION_recall": 0.9651162790697675,
      "eval_ORGANIZATION_f1": 0.9322033898305084,
      "eval_ORGANIZATION_number": 178,
      "eval_ORGANIZATION_precision": 0.9375,
      "eval_ORGANIZATION_recall": 0.9269662921348315,
      "eval_PERSON_f1": 0.9765625,
      "eval_PERSON_number": 128,
      "eval_PERSON_precision": 0.9765625,
      "eval_PERSON_recall": 0.9765625,
      "eval_loss": 0.057044416666030884,
      "eval_overall_accuracy": 0.9873144399460189,
      "eval_overall_f1": 0.9479034307496824,
      "eval_overall_precision": 0.9443037974683545,
      "eval_overall_recall": 0.951530612244898,
      "eval_runtime": 1.1835,
      "eval_samples_per_second": 143.642,
      "eval_steps_per_second": 2.535,
      "step": 384
    },
    {
      "epoch": 5.0,
      "grad_norm": 0.14279630780220032,
      "learning_rate": 0.0,
      "loss": 0.0088,
      "step": 480
    },
    {
      "epoch": 5.0,
      "eval_LOCATION_f1": 0.9431818181818181,
      "eval_LOCATION_number": 86,
      "eval_LOCATION_precision": 0.9222222222222223,
      "eval_LOCATION_recall": 0.9651162790697675,
      "eval_ORGANIZATION_f1": 0.9385474860335196,
      "eval_ORGANIZATION_number": 178,
      "eval_ORGANIZATION_precision": 0.9333333333333333,
      "eval_ORGANIZATION_recall": 0.9438202247191011,
      "eval_PERSON_f1": 0.9803921568627452,
      "eval_PERSON_number": 128,
      "eval_PERSON_precision": 0.984251968503937,
      "eval_PERSON_recall": 0.9765625,
      "eval_loss": 0.05175752565264702,
      "eval_overall_accuracy": 0.9889338731443995,
      "eval_overall_f1": 0.9531051964512041,
      "eval_overall_precision": 0.947103274559194,
      "eval_overall_recall": 0.9591836734693877,
      "eval_runtime": 1.2661,
      "eval_samples_per_second": 134.27,
      "eval_steps_per_second": 2.369,
      "step": 480
    },
    {
      "epoch": 5.0,
      "step": 480,
      "total_flos": 193448468569026.0,
      "train_loss": 0.06961918647090594,
      "train_runtime": 225.7241,
      "train_samples_per_second": 33.913,
      "train_steps_per_second": 2.126
    }
  ],
  "logging_steps": 500,
  "max_steps": 480,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 5,
  "save_steps": 500,
  "total_flos": 193448468569026.0,
  "train_batch_size": 16,
  "trial_name": null,
  "trial_params": null
}