{
  "best_global_step": 6136,
  "best_metric": 0.815567672252655,
  "best_model_checkpoint": "distilbert_rand_20_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": 1.167750597000122,
      "learning_rate": 4.9e-05,
      "loss": 0.9775,
      "step": 1534
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.5703515028018339,
      "eval_loss": 0.9091921448707581,
      "eval_runtime": 4.0573,
      "eval_samples_per_second": 2419.124,
      "eval_steps_per_second": 9.612,
      "step": 1534
    },
    {
      "epoch": 2.0,
      "grad_norm": 1.5929653644561768,
      "learning_rate": 4.8e-05,
      "loss": 0.8776,
      "step": 3068
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.6022414671421293,
      "eval_loss": 0.8706146478652954,
      "eval_runtime": 3.9473,
      "eval_samples_per_second": 2486.511,
      "eval_steps_per_second": 9.88,
      "step": 3068
    },
    {
      "epoch": 3.0,
      "grad_norm": 1.806445598602295,
      "learning_rate": 4.7e-05,
      "loss": 0.7966,
      "step": 4602
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.6406520631686194,
      "eval_loss": 0.816186249256134,
      "eval_runtime": 4.0056,
      "eval_samples_per_second": 2450.346,
      "eval_steps_per_second": 9.736,
      "step": 4602
    },
    {
      "epoch": 4.0,
      "grad_norm": 2.592050552368164,
      "learning_rate": 4.600000000000001e-05,
      "loss": 0.7182,
      "step": 6136
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.6444218033622007,
      "eval_loss": 0.815567672252655,
      "eval_runtime": 3.9636,
      "eval_samples_per_second": 2476.276,
      "eval_steps_per_second": 9.84,
      "step": 6136
    },
    {
      "epoch": 5.0,
      "grad_norm": 2.0119593143463135,
      "learning_rate": 4.5e-05,
      "loss": 0.6474,
      "step": 7670
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.6535914416709119,
      "eval_loss": 0.8175016045570374,
      "eval_runtime": 3.994,
      "eval_samples_per_second": 2457.428,
      "eval_steps_per_second": 9.765,
      "step": 7670
    },
    {
      "epoch": 6.0,
      "grad_norm": 2.5608832836151123,
      "learning_rate": 4.4000000000000006e-05,
      "loss": 0.5759,
      "step": 9204
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.655934793683138,
      "eval_loss": 0.854282557964325,
      "eval_runtime": 4.0032,
      "eval_samples_per_second": 2451.787,
      "eval_steps_per_second": 9.742,
      "step": 9204
    },
    {
      "epoch": 7.0,
      "grad_norm": 2.502145290374756,
      "learning_rate": 4.3e-05,
      "loss": 0.5059,
      "step": 10738
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.6519612837493632,
      "eval_loss": 0.9202938675880432,
      "eval_runtime": 3.9823,
      "eval_samples_per_second": 2464.677,
      "eval_steps_per_second": 9.793,
      "step": 10738
    },
    {
      "epoch": 8.0,
      "grad_norm": 3.229653835296631,
      "learning_rate": 4.2e-05,
      "loss": 0.4401,
      "step": 12272
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.645236882322975,
      "eval_loss": 1.0363502502441406,
      "eval_runtime": 4.0043,
      "eval_samples_per_second": 2451.105,
      "eval_steps_per_second": 9.739,
      "step": 12272
    },
    {
      "epoch": 9.0,
      "grad_norm": 4.310319900512695,
      "learning_rate": 4.1e-05,
      "loss": 0.3795,
      "step": 13806
    },
    {
      "epoch": 9.0,
      "eval_accuracy": 0.6416709118695874,
      "eval_loss": 1.1485828161239624,
      "eval_runtime": 4.0834,
      "eval_samples_per_second": 2403.607,
      "eval_steps_per_second": 9.551,
      "step": 13806
    },
    {
      "epoch": 9.0,
      "step": 13806,
      "total_flos": 2.340951304214108e+17,
      "train_loss": 0.6576309789872836,
      "train_runtime": 3243.9043,
      "train_samples_per_second": 6052.922,
      "train_steps_per_second": 23.644
    }
  ],
  "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
}