{
  "best_metric": 0.721001221001221,
  "best_model_checkpoint": "logs/indian_ner/MHGanainy/roberta-base-legal-multi/seed_1/checkpoint-3440",
  "epoch": 20.0,
  "eval_steps": 500,
  "global_step": 3440,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "eval_accuracy": 0.8307266822776638,
      "eval_f1": 0.19708233973909384,
      "eval_loss": 0.23097991943359375,
      "eval_precision": 0.12332133766347757,
      "eval_recall": 0.49040139616055844,
      "eval_runtime": 2.7434,
      "eval_samples_per_second": 391.481,
      "eval_steps_per_second": 6.197,
      "step": 172
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8764707582897285,
      "eval_f1": 0.2899502674298583,
      "eval_loss": 0.1928548961877823,
      "eval_precision": 0.19828028747433266,
      "eval_recall": 0.5392670157068062,
      "eval_runtime": 2.6548,
      "eval_samples_per_second": 404.551,
      "eval_steps_per_second": 6.404,
      "step": 344
    },
    {
      "epoch": 2.9069767441860463,
      "grad_norm": 1.780714511871338,
      "learning_rate": 2.5656976744186047e-05,
      "loss": 0.4324,
      "step": 500
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.87381110693533,
      "eval_f1": 0.26039346696362287,
      "eval_loss": 0.16671840846538544,
      "eval_precision": 0.17734799646062444,
      "eval_recall": 0.4897033158813264,
      "eval_runtime": 2.6947,
      "eval_samples_per_second": 398.561,
      "eval_steps_per_second": 6.309,
      "step": 516
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.9081071184219403,
      "eval_f1": 0.3974813966800229,
      "eval_loss": 0.1836225837469101,
      "eval_precision": 0.2957410562180579,
      "eval_recall": 0.605933682373473,
      "eval_runtime": 2.6664,
      "eval_samples_per_second": 402.785,
      "eval_steps_per_second": 6.376,
      "step": 688
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.9137155136692588,
      "eval_f1": 0.3787023977433004,
      "eval_loss": 0.20050308108329773,
      "eval_precision": 0.28548644338118023,
      "eval_recall": 0.562303664921466,
      "eval_runtime": 2.651,
      "eval_samples_per_second": 405.126,
      "eval_steps_per_second": 6.413,
      "step": 860
    },
    {
      "epoch": 5.813953488372093,
      "grad_norm": 2.750487804412842,
      "learning_rate": 2.1296511627906976e-05,
      "loss": 0.1106,
      "step": 1000
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.9323427095679994,
      "eval_f1": 0.49676777722526105,
      "eval_loss": 0.20030471682548523,
      "eval_precision": 0.3857887623093261,
      "eval_recall": 0.6973821989528796,
      "eval_runtime": 2.6529,
      "eval_samples_per_second": 404.843,
      "eval_steps_per_second": 6.408,
      "step": 1032
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.9428367687163328,
      "eval_f1": 0.515532940546331,
      "eval_loss": 0.22236432135105133,
      "eval_precision": 0.41820551814034324,
      "eval_recall": 0.6719022687609075,
      "eval_runtime": 2.7707,
      "eval_samples_per_second": 387.631,
      "eval_steps_per_second": 6.136,
      "step": 1204
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.931224885085716,
      "eval_f1": 0.4334235786364755,
      "eval_loss": 0.22214500606060028,
      "eval_precision": 0.3347272381676487,
      "eval_recall": 0.6146596858638743,
      "eval_runtime": 2.6397,
      "eval_samples_per_second": 406.868,
      "eval_steps_per_second": 6.44,
      "step": 1376
    },
    {
      "epoch": 8.720930232558139,
      "grad_norm": 1.4176898002624512,
      "learning_rate": 1.6936046511627905e-05,
      "loss": 0.0589,
      "step": 1500
    },
    {
      "epoch": 9.0,
      "eval_accuracy": 0.9403602093029979,
      "eval_f1": 0.5151631477927063,
      "eval_loss": 0.19603897631168365,
      "eval_precision": 0.4066666666666667,
      "eval_recall": 0.7026178010471205,
      "eval_runtime": 2.6818,
      "eval_samples_per_second": 400.482,
      "eval_steps_per_second": 6.339,
      "step": 1548
    },
    {
      "epoch": 10.0,
      "eval_accuracy": 0.9524153681593478,
      "eval_f1": 0.6005657708628006,
      "eval_loss": 0.19039033353328705,
      "eval_precision": 0.5048751486325802,
      "eval_recall": 0.7410122164048866,
      "eval_runtime": 2.6725,
      "eval_samples_per_second": 401.865,
      "eval_steps_per_second": 6.361,
      "step": 1720
    },
    {
      "epoch": 11.0,
      "eval_accuracy": 0.9565493914602065,
      "eval_f1": 0.6306769494430162,
      "eval_loss": 0.2274240404367447,
      "eval_precision": 0.5337200870195794,
      "eval_recall": 0.7706806282722513,
      "eval_runtime": 2.6498,
      "eval_samples_per_second": 405.32,
      "eval_steps_per_second": 6.416,
      "step": 1892
    },
    {
      "epoch": 11.627906976744185,
      "grad_norm": 0.5939443111419678,
      "learning_rate": 1.2575581395348837e-05,
      "loss": 0.0359,
      "step": 2000
    },
    {
      "epoch": 12.0,
      "eval_accuracy": 0.9574744875834754,
      "eval_f1": 0.6432322053675613,
      "eval_loss": 0.2470693439245224,
      "eval_precision": 0.5524931094963669,
      "eval_recall": 0.7696335078534031,
      "eval_runtime": 2.7137,
      "eval_samples_per_second": 395.77,
      "eval_steps_per_second": 6.265,
      "step": 2064
    },
    {
      "epoch": 13.0,
      "eval_accuracy": 0.9590644965453442,
      "eval_f1": 0.6507842556969518,
      "eval_loss": 0.2351667284965515,
      "eval_precision": 0.5648600051374262,
      "eval_recall": 0.7675392670157068,
      "eval_runtime": 3.026,
      "eval_samples_per_second": 354.925,
      "eval_steps_per_second": 5.618,
      "step": 2236
    },
    {
      "epoch": 14.0,
      "eval_accuracy": 0.9586308577375618,
      "eval_f1": 0.6423763535264853,
      "eval_loss": 0.22973008453845978,
      "eval_precision": 0.5530360292265054,
      "eval_recall": 0.7661431064572426,
      "eval_runtime": 2.6379,
      "eval_samples_per_second": 407.14,
      "eval_steps_per_second": 6.444,
      "step": 2408
    },
    {
      "epoch": 14.534883720930232,
      "grad_norm": 0.7274452447891235,
      "learning_rate": 8.215116279069767e-06,
      "loss": 0.0224,
      "step": 2500
    },
    {
      "epoch": 15.0,
      "eval_accuracy": 0.9597486822198452,
      "eval_f1": 0.6631609699094361,
      "eval_loss": 0.23491716384887695,
      "eval_precision": 0.5702084903290631,
      "eval_recall": 0.7923211169284468,
      "eval_runtime": 2.701,
      "eval_samples_per_second": 397.624,
      "eval_steps_per_second": 6.294,
      "step": 2580
    },
    {
      "epoch": 16.0,
      "eval_accuracy": 0.9624083335742438,
      "eval_f1": 0.6897892061593661,
      "eval_loss": 0.2465096265077591,
      "eval_precision": 0.6032949790794979,
      "eval_recall": 0.8052356020942408,
      "eval_runtime": 2.6696,
      "eval_samples_per_second": 402.31,
      "eval_steps_per_second": 6.368,
      "step": 2752
    },
    {
      "epoch": 17.0,
      "eval_accuracy": 0.964701801046515,
      "eval_f1": 0.6958608278344331,
      "eval_loss": 0.24279330670833588,
      "eval_precision": 0.6100447015514068,
      "eval_recall": 0.8097731239092496,
      "eval_runtime": 2.6397,
      "eval_samples_per_second": 406.87,
      "eval_steps_per_second": 6.44,
      "step": 2924
    },
    {
      "epoch": 17.441860465116278,
      "grad_norm": 0.6849046349525452,
      "learning_rate": 3.854651162790698e-06,
      "loss": 0.0143,
      "step": 3000
    },
    {
      "epoch": 18.0,
      "eval_accuracy": 0.9646150732849585,
      "eval_f1": 0.7068078668683813,
      "eval_loss": 0.25429967045783997,
      "eval_precision": 0.6237650200267023,
      "eval_recall": 0.8153577661431065,
      "eval_runtime": 2.6691,
      "eval_samples_per_second": 402.382,
      "eval_steps_per_second": 6.369,
      "step": 3096
    },
    {
      "epoch": 19.0,
      "eval_accuracy": 0.9650679849286423,
      "eval_f1": 0.711395101171459,
      "eval_loss": 0.25257599353790283,
      "eval_precision": 0.6305285868392665,
      "eval_recall": 0.8160558464223385,
      "eval_runtime": 2.6493,
      "eval_samples_per_second": 405.386,
      "eval_steps_per_second": 6.417,
      "step": 3268
    },
    {
      "epoch": 20.0,
      "eval_accuracy": 0.9663303556801865,
      "eval_f1": 0.721001221001221,
      "eval_loss": 0.2526082694530487,
      "eval_precision": 0.640629237862761,
      "eval_recall": 0.8244328097731239,
      "eval_runtime": 2.6583,
      "eval_samples_per_second": 404.01,
      "eval_steps_per_second": 6.395,
      "step": 3440
    },
    {
      "epoch": 20.0,
      "step": 3440,
      "total_flos": 5.74731828937728e+16,
      "train_loss": 0.09935806185700173,
      "train_runtime": 802.219,
      "train_samples_per_second": 274.115,
      "train_steps_per_second": 4.288
    }
  ],
  "logging_steps": 500,
  "max_steps": 3440,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 20,
  "save_steps": 500,
  "stateful_callbacks": {
    "EarlyStoppingCallback": {
      "args": {
        "early_stopping_patience": 3,
        "early_stopping_threshold": 0.0
      },
      "attributes": {
        "early_stopping_patience_counter": 0
      }
    },
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 5.74731828937728e+16,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}