{
  "best_global_step": 7670,
  "best_metric": 0.8086211681365967,
  "best_model_checkpoint": "distilbert_km_50_v2_mnli/checkpoint-7670",
  "epoch": 10.0,
  "eval_steps": 500,
  "global_step": 15340,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "grad_norm": 0.9368550181388855,
      "learning_rate": 4.9e-05,
      "loss": 0.9937,
      "step": 1534
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.5672949566989303,
      "eval_loss": 0.918012797832489,
      "eval_runtime": 4.0769,
      "eval_samples_per_second": 2407.444,
      "eval_steps_per_second": 9.566,
      "step": 1534
    },
    {
      "epoch": 2.0,
      "grad_norm": 1.4801536798477173,
      "learning_rate": 4.8e-05,
      "loss": 0.8879,
      "step": 3068
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.6020376974019358,
      "eval_loss": 0.8699061274528503,
      "eval_runtime": 3.9759,
      "eval_samples_per_second": 2468.647,
      "eval_steps_per_second": 9.809,
      "step": 3068
    },
    {
      "epoch": 3.0,
      "grad_norm": 1.5542188882827759,
      "learning_rate": 4.7e-05,
      "loss": 0.816,
      "step": 4602
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.6311767702496179,
      "eval_loss": 0.8278898000717163,
      "eval_runtime": 3.961,
      "eval_samples_per_second": 2477.884,
      "eval_steps_per_second": 9.846,
      "step": 4602
    },
    {
      "epoch": 4.0,
      "grad_norm": 1.9876985549926758,
      "learning_rate": 4.600000000000001e-05,
      "loss": 0.7463,
      "step": 6136
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.6448293428425879,
      "eval_loss": 0.8165294528007507,
      "eval_runtime": 3.9648,
      "eval_samples_per_second": 2475.547,
      "eval_steps_per_second": 9.837,
      "step": 6136
    },
    {
      "epoch": 5.0,
      "grad_norm": 1.9823399782180786,
      "learning_rate": 4.5e-05,
      "loss": 0.6806,
      "step": 7670
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.6547121752419766,
      "eval_loss": 0.8086211681365967,
      "eval_runtime": 3.9669,
      "eval_samples_per_second": 2474.241,
      "eval_steps_per_second": 9.831,
      "step": 7670
    },
    {
      "epoch": 6.0,
      "grad_norm": 2.4311466217041016,
      "learning_rate": 4.4000000000000006e-05,
      "loss": 0.6141,
      "step": 9204
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.6548140601120733,
      "eval_loss": 0.8516751527786255,
      "eval_runtime": 3.9487,
      "eval_samples_per_second": 2485.638,
      "eval_steps_per_second": 9.877,
      "step": 9204
    },
    {
      "epoch": 7.0,
      "grad_norm": 2.3432135581970215,
      "learning_rate": 4.3e-05,
      "loss": 0.5484,
      "step": 10738
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.658889454915945,
      "eval_loss": 0.8841859698295593,
      "eval_runtime": 3.9628,
      "eval_samples_per_second": 2476.778,
      "eval_steps_per_second": 9.842,
      "step": 10738
    },
    {
      "epoch": 8.0,
      "grad_norm": 2.830547571182251,
      "learning_rate": 4.2e-05,
      "loss": 0.4816,
      "step": 12272
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.6513499745287825,
      "eval_loss": 0.9944413900375366,
      "eval_runtime": 3.9717,
      "eval_samples_per_second": 2471.229,
      "eval_steps_per_second": 9.819,
      "step": 12272
    },
    {
      "epoch": 9.0,
      "grad_norm": 3.5015695095062256,
      "learning_rate": 4.1e-05,
      "loss": 0.4196,
      "step": 13806
    },
    {
      "epoch": 9.0,
      "eval_accuracy": 0.6543046357615894,
      "eval_loss": 1.0646898746490479,
      "eval_runtime": 3.9792,
      "eval_samples_per_second": 2466.573,
      "eval_steps_per_second": 9.801,
      "step": 13806
    },
    {
      "epoch": 10.0,
      "grad_norm": 4.094024181365967,
      "learning_rate": 4e-05,
      "loss": 0.3638,
      "step": 15340
    },
    {
      "epoch": 10.0,
      "eval_accuracy": 0.6497198166072339,
      "eval_loss": 1.1240673065185547,
      "eval_runtime": 3.9527,
      "eval_samples_per_second": 2483.142,
      "eval_steps_per_second": 9.867,
      "step": 15340
    },
    {
      "epoch": 10.0,
      "step": 15340,
      "total_flos": 2.6010570046823424e+17,
      "train_loss": 0.65518792462007,
      "train_runtime": 3604.7116,
      "train_samples_per_second": 5447.065,
      "train_steps_per_second": 21.278
    }
  ],
  "logging_steps": 1,
  "max_steps": 76700,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 50,
  "save_steps": 500,
  "stateful_callbacks": {
    "EarlyStoppingCallback": {
      "args": {
        "early_stopping_patience": 5,
        "early_stopping_threshold": 0.0
      },
      "attributes": {
        "early_stopping_patience_counter": 5
      }
    },
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 2.6010570046823424e+17,
  "train_batch_size": 256,
  "trial_name": null,
  "trial_params": null
}