{
  "best_metric": 0.8140900195694716,
  "best_model_checkpoint": "tiny-bert-sst2-distilled/run-1/checkpoint-288",
  "epoch": 4.0,
  "eval_steps": 500,
  "global_step": 384,
  "is_hyper_param_search": true,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "grad_norm": 2.6617980003356934,
      "learning_rate": 0.0006635023570917325,
      "loss": 0.5037,
      "step": 96
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.7661448140900196,
      "eval_f1": 0.792714657415438,
      "eval_loss": 0.4430159628391266,
      "eval_precision": 0.7118380062305296,
      "eval_recall": 0.8943248532289628,
      "eval_runtime": 28.6854,
      "eval_samples_per_second": 35.628,
      "eval_steps_per_second": 1.116,
      "step": 96
    },
    {
      "epoch": 2.0,
      "grad_norm": 1.12320077419281,
      "learning_rate": 0.0004976267678187994,
      "loss": 0.4391,
      "step": 192
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8082191780821918,
      "eval_f1": 0.8227848101265822,
      "eval_loss": 0.41584911942481995,
      "eval_precision": 0.7647058823529411,
      "eval_recall": 0.8904109589041096,
      "eval_runtime": 28.5215,
      "eval_samples_per_second": 35.833,
      "eval_steps_per_second": 1.122,
      "step": 192
    },
    {
      "epoch": 3.0,
      "grad_norm": 2.946756362915039,
      "learning_rate": 0.00033175117854586627,
      "loss": 0.41,
      "step": 288
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8140900195694716,
      "eval_f1": 0.8260073260073261,
      "eval_loss": 0.43016910552978516,
      "eval_precision": 0.7762478485370051,
      "eval_recall": 0.8825831702544031,
      "eval_runtime": 28.1244,
      "eval_samples_per_second": 36.339,
      "eval_steps_per_second": 1.138,
      "step": 288
    },
    {
      "epoch": 4.0,
      "grad_norm": 3.868119478225708,
      "learning_rate": 0.00016587558927293314,
      "loss": 0.4041,
      "step": 384
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8033268101761253,
      "eval_f1": 0.802747791952895,
      "eval_loss": 0.4142443835735321,
      "eval_precision": 0.8051181102362205,
      "eval_recall": 0.8003913894324853,
      "eval_runtime": 29.1758,
      "eval_samples_per_second": 35.029,
      "eval_steps_per_second": 1.097,
      "step": 384
    }
  ],
  "logging_steps": 500,
  "max_steps": 480,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 5,
  "save_steps": 500,
  "total_flos": 942780789120.0,
  "train_batch_size": 32,
  "trial_name": null,
  "trial_params": {
    "alpha": 0.7818460169984094,
    "learning_rate": 0.0008293779463646656,
    "num_train_epochs": 5,
    "temperature": 3
  }
}