trainer_state.json

{
  "best_metric": 0.339957594871521,
  "best_model_checkpoint": "bert-multilingual-base-cased-finetuned-toxicity-classification/checkpoint-4015",
  "epoch": 2.0,
  "eval_steps": 500,
  "global_step": 8030,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.12453300124533001,
      "grad_norm": 7.769826412200928,
      "learning_rate": 1.950186799501868e-05,
      "loss": 0.5596,
      "step": 500
    },
    {
      "epoch": 0.24906600249066002,
      "grad_norm": 13.72755241394043,
      "learning_rate": 1.9003735990037362e-05,
      "loss": 0.4566,
      "step": 1000
    },
    {
      "epoch": 0.37359900373599003,
      "grad_norm": 6.368185043334961,
      "learning_rate": 1.850560398505604e-05,
      "loss": 0.4185,
      "step": 1500
    },
    {
      "epoch": 0.49813200498132004,
      "grad_norm": 5.925895690917969,
      "learning_rate": 1.8007471980074722e-05,
      "loss": 0.3982,
      "step": 2000
    },
    {
      "epoch": 0.6226650062266501,
      "grad_norm": 2.7038302421569824,
      "learning_rate": 1.75093399750934e-05,
      "loss": 0.3906,
      "step": 2500
    },
    {
      "epoch": 0.7471980074719801,
      "grad_norm": 3.163766622543335,
      "learning_rate": 1.701120797011208e-05,
      "loss": 0.3678,
      "step": 3000
    },
    {
      "epoch": 0.8717310087173101,
      "grad_norm": 3.3420841693878174,
      "learning_rate": 1.651307596513076e-05,
      "loss": 0.3629,
      "step": 3500
    },
    {
      "epoch": 0.9962640099626401,
      "grad_norm": 6.345468997955322,
      "learning_rate": 1.601494396014944e-05,
      "loss": 0.3694,
      "step": 4000
    },
    {
      "epoch": 1.0,
      "eval_loss": 0.339957594871521,
      "eval_runtime": 174.5055,
      "eval_samples_per_second": 40.91,
      "eval_steps_per_second": 5.117,
      "step": 4015
    },
    {
      "epoch": 1.1207970112079702,
      "grad_norm": 6.267732620239258,
      "learning_rate": 1.551681195516812e-05,
      "loss": 0.2756,
      "step": 4500
    },
    {
      "epoch": 1.2453300124533002,
      "grad_norm": 34.6484375,
      "learning_rate": 1.5018679950186801e-05,
      "loss": 0.2826,
      "step": 5000
    },
    {
      "epoch": 1.36986301369863,
      "grad_norm": 12.416611671447754,
      "learning_rate": 1.4520547945205482e-05,
      "loss": 0.2705,
      "step": 5500
    },
    {
      "epoch": 1.4943960149439601,
      "grad_norm": 4.286174297332764,
      "learning_rate": 1.4022415940224162e-05,
      "loss": 0.2735,
      "step": 6000
    },
    {
      "epoch": 1.6189290161892902,
      "grad_norm": 3.970327377319336,
      "learning_rate": 1.3524283935242842e-05,
      "loss": 0.273,
      "step": 6500
    },
    {
      "epoch": 1.74346201743462,
      "grad_norm": 16.260072708129883,
      "learning_rate": 1.3026151930261522e-05,
      "loss": 0.2733,
      "step": 7000
    },
    {
      "epoch": 1.86799501867995,
      "grad_norm": 0.7613949775695801,
      "learning_rate": 1.2528019925280199e-05,
      "loss": 0.2659,
      "step": 7500
    },
    {
      "epoch": 1.9925280199252802,
      "grad_norm": 12.411490440368652,
      "learning_rate": 1.2029887920298879e-05,
      "loss": 0.2632,
      "step": 8000
    },
    {
      "epoch": 2.0,
      "eval_loss": 0.3527267277240753,
      "eval_runtime": 174.8977,
      "eval_samples_per_second": 40.818,
      "eval_steps_per_second": 5.106,
      "step": 8030
    }
  ],
  "logging_steps": 500,
  "max_steps": 20075,
  "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": false
      },
      "attributes": {}
    }
  },
  "total_flos": 3.38018772820992e+16,
  "train_batch_size": 8,
  "trial_name": null,
  "trial_params": null
}