{
  "best_global_step": 6136,
  "best_metric": 0.7739408016204834,
  "best_model_checkpoint": "distilbert_km_5_v2_mnli/checkpoint-6136",
  "epoch": 9.0,
  "eval_steps": 500,
  "global_step": 13806,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "grad_norm": 1.2396316528320312,
      "learning_rate": 4.9e-05,
      "loss": 0.9748,
      "step": 1534
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.5751400916963831,
      "eval_loss": 0.8968575596809387,
      "eval_runtime": 4.0906,
      "eval_samples_per_second": 2399.419,
      "eval_steps_per_second": 9.534,
      "step": 1534
    },
    {
      "epoch": 2.0,
      "grad_norm": 1.8438777923583984,
      "learning_rate": 4.8e-05,
      "loss": 0.8527,
      "step": 3068
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.6402445236882323,
      "eval_loss": 0.8092689514160156,
      "eval_runtime": 3.9266,
      "eval_samples_per_second": 2499.636,
      "eval_steps_per_second": 9.932,
      "step": 3068
    },
    {
      "epoch": 3.0,
      "grad_norm": 2.0112054347991943,
      "learning_rate": 4.7e-05,
      "loss": 0.7526,
      "step": 4602
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.6682628629648497,
      "eval_loss": 0.7755393385887146,
      "eval_runtime": 3.95,
      "eval_samples_per_second": 2484.839,
      "eval_steps_per_second": 9.874,
      "step": 4602
    },
    {
      "epoch": 4.0,
      "grad_norm": 2.4817464351654053,
      "learning_rate": 4.600000000000001e-05,
      "loss": 0.6686,
      "step": 6136
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.6713194090677534,
      "eval_loss": 0.7739408016204834,
      "eval_runtime": 3.9266,
      "eval_samples_per_second": 2499.601,
      "eval_steps_per_second": 9.932,
      "step": 6136
    },
    {
      "epoch": 5.0,
      "grad_norm": 2.417473316192627,
      "learning_rate": 4.5e-05,
      "loss": 0.5878,
      "step": 7670
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.6785532348446256,
      "eval_loss": 0.7808324694633484,
      "eval_runtime": 3.9403,
      "eval_samples_per_second": 2490.9,
      "eval_steps_per_second": 9.898,
      "step": 7670
    },
    {
      "epoch": 6.0,
      "grad_norm": 2.7710580825805664,
      "learning_rate": 4.4000000000000006e-05,
      "loss": 0.5109,
      "step": 9204
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.6777381558838512,
      "eval_loss": 0.8452056646347046,
      "eval_runtime": 3.9333,
      "eval_samples_per_second": 2495.368,
      "eval_steps_per_second": 9.915,
      "step": 9204
    },
    {
      "epoch": 7.0,
      "grad_norm": 3.401859760284424,
      "learning_rate": 4.3e-05,
      "loss": 0.4361,
      "step": 10738
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.6761079979623026,
      "eval_loss": 0.893120288848877,
      "eval_runtime": 3.9435,
      "eval_samples_per_second": 2488.912,
      "eval_steps_per_second": 9.89,
      "step": 10738
    },
    {
      "epoch": 8.0,
      "grad_norm": 3.13527250289917,
      "learning_rate": 4.2e-05,
      "loss": 0.368,
      "step": 12272
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.6752929190015283,
      "eval_loss": 0.9989559054374695,
      "eval_runtime": 3.9379,
      "eval_samples_per_second": 2492.476,
      "eval_steps_per_second": 9.904,
      "step": 12272
    },
    {
      "epoch": 9.0,
      "grad_norm": 4.258540153503418,
      "learning_rate": 4.1e-05,
      "loss": 0.3074,
      "step": 13806
    },
    {
      "epoch": 9.0,
      "eval_accuracy": 0.6691798267957209,
      "eval_loss": 1.1551142930984497,
      "eval_runtime": 3.9503,
      "eval_samples_per_second": 2484.636,
      "eval_steps_per_second": 9.873,
      "step": 13806
    },
    {
      "epoch": 9.0,
      "step": 13806,
      "total_flos": 2.340951304214108e+17,
      "train_loss": 0.6065415536150213,
      "train_runtime": 3244.9194,
      "train_samples_per_second": 6051.028,
      "train_steps_per_second": 23.637
    }
  ],
  "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": 2.340951304214108e+17,
  "train_batch_size": 256,
  "trial_name": null,
  "trial_params": null
}