{
  "best_metric": 0.8140900195694716,
  "best_model_checkpoint": "tiny-bert-sst2-distilled/run-10/checkpoint-288",
  "epoch": 6.0,
  "eval_steps": 500,
  "global_step": 288,
  "is_hyper_param_search": true,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "grad_norm": 1.5607351064682007,
      "learning_rate": 0.0008549093212842401,
      "loss": 0.5996,
      "step": 48
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.7632093933463796,
      "eval_f1": 0.7949152542372881,
      "eval_loss": 0.502604067325592,
      "eval_precision": 0.7010463378176383,
      "eval_recall": 0.9178082191780822,
      "eval_runtime": 27.7883,
      "eval_samples_per_second": 36.778,
      "eval_steps_per_second": 0.576,
      "step": 48
    },
    {
      "epoch": 2.0,
      "grad_norm": 3.883798122406006,
      "learning_rate": 0.0007124244344035335,
      "loss": 0.4919,
      "step": 96
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.7847358121330724,
      "eval_f1": 0.8066783831282953,
      "eval_loss": 0.4441834092140198,
      "eval_precision": 0.7320574162679426,
      "eval_recall": 0.898238747553816,
      "eval_runtime": 27.0257,
      "eval_samples_per_second": 37.816,
      "eval_steps_per_second": 0.592,
      "step": 96
    },
    {
      "epoch": 3.0,
      "grad_norm": 3.711146116256714,
      "learning_rate": 0.0005699395475228268,
      "loss": 0.449,
      "step": 144
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.799412915851272,
      "eval_f1": 0.8222029488291415,
      "eval_loss": 0.46845388412475586,
      "eval_precision": 0.7383177570093458,
      "eval_recall": 0.9275929549902152,
      "eval_runtime": 27.3696,
      "eval_samples_per_second": 37.341,
      "eval_steps_per_second": 0.585,
      "step": 144
    },
    {
      "epoch": 4.0,
      "grad_norm": 4.223331451416016,
      "learning_rate": 0.0004274546606421201,
      "loss": 0.4144,
      "step": 192
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8003913894324853,
      "eval_f1": 0.796812749003984,
      "eval_loss": 0.45222949981689453,
      "eval_precision": 0.8113590263691683,
      "eval_recall": 0.7827788649706457,
      "eval_runtime": 27.0074,
      "eval_samples_per_second": 37.841,
      "eval_steps_per_second": 0.592,
      "step": 192
    },
    {
      "epoch": 5.0,
      "grad_norm": 2.6906895637512207,
      "learning_rate": 0.0002849697737614134,
      "loss": 0.3883,
      "step": 240
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8033268101761253,
      "eval_f1": 0.8280581693755347,
      "eval_loss": 0.4314705431461334,
      "eval_precision": 0.7355623100303952,
      "eval_recall": 0.9471624266144814,
      "eval_runtime": 27.3713,
      "eval_samples_per_second": 37.338,
      "eval_steps_per_second": 0.585,
      "step": 240
    },
    {
      "epoch": 6.0,
      "grad_norm": 2.0857512950897217,
      "learning_rate": 0.0001424848868807067,
      "loss": 0.3715,
      "step": 288
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.8140900195694716,
      "eval_f1": 0.8327464788732395,
      "eval_loss": 0.4203811585903168,
      "eval_precision": 0.7568,
      "eval_recall": 0.9256360078277887,
      "eval_runtime": 27.3476,
      "eval_samples_per_second": 37.371,
      "eval_steps_per_second": 0.585,
      "step": 288
    }
  ],
  "logging_steps": 500,
  "max_steps": 336,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 7,
  "save_steps": 500,
  "total_flos": 1414171183680.0,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": {
    "alpha": 0.9873851004059778,
    "learning_rate": 0.0009973942081649468,
    "num_train_epochs": 7,
    "temperature": 16
  }
}