{
  "best_global_step": 486,
  "best_metric": 0.8587628865979381,
  "best_model_checkpoint": "./results/checkpoint-486",
  "epoch": 5.0,
  "eval_steps": 500,
  "global_step": 1215,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.411522633744856,
      "grad_norm": 3.3757410049438477,
      "learning_rate": 9.185185185185186e-06,
      "loss": 0.8845,
      "step": 100
    },
    {
      "epoch": 0.823045267489712,
      "grad_norm": 3.417144536972046,
      "learning_rate": 8.362139917695474e-06,
      "loss": 0.5844,
      "step": 200
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.8237113402061855,
      "eval_f1": 0.8244991682852733,
      "eval_loss": 0.44459813833236694,
      "eval_runtime": 2.448,
      "eval_samples_per_second": 396.246,
      "eval_steps_per_second": 24.919,
      "step": 243
    },
    {
      "epoch": 1.2345679012345678,
      "grad_norm": 5.602197647094727,
      "learning_rate": 7.5390946502057615e-06,
      "loss": 0.4285,
      "step": 300
    },
    {
      "epoch": 1.646090534979424,
      "grad_norm": 6.984129905700684,
      "learning_rate": 6.71604938271605e-06,
      "loss": 0.3645,
      "step": 400
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8587628865979381,
      "eval_f1": 0.8582800447235293,
      "eval_loss": 0.3806818127632141,
      "eval_runtime": 2.4825,
      "eval_samples_per_second": 390.733,
      "eval_steps_per_second": 24.572,
      "step": 486
    },
    {
      "epoch": 2.05761316872428,
      "grad_norm": 8.748188018798828,
      "learning_rate": 5.893004115226338e-06,
      "loss": 0.3403,
      "step": 500
    },
    {
      "epoch": 2.4691358024691357,
      "grad_norm": 7.327649116516113,
      "learning_rate": 5.069958847736626e-06,
      "loss": 0.2679,
      "step": 600
    },
    {
      "epoch": 2.8806584362139915,
      "grad_norm": 11.227555274963379,
      "learning_rate": 4.246913580246914e-06,
      "loss": 0.2677,
      "step": 700
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8577319587628865,
      "eval_f1": 0.8591340461713324,
      "eval_loss": 0.3954542577266693,
      "eval_runtime": 2.4954,
      "eval_samples_per_second": 388.715,
      "eval_steps_per_second": 24.445,
      "step": 729
    },
    {
      "epoch": 3.292181069958848,
      "grad_norm": 10.80851936340332,
      "learning_rate": 3.423868312757202e-06,
      "loss": 0.2197,
      "step": 800
    },
    {
      "epoch": 3.7037037037037037,
      "grad_norm": 8.145468711853027,
      "learning_rate": 2.60082304526749e-06,
      "loss": 0.2074,
      "step": 900
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8577319587628865,
      "eval_f1": 0.8574733463132496,
      "eval_loss": 0.4115142524242401,
      "eval_runtime": 2.4956,
      "eval_samples_per_second": 388.69,
      "eval_steps_per_second": 24.443,
      "step": 972
    },
    {
      "epoch": 4.11522633744856,
      "grad_norm": 2.062761068344116,
      "learning_rate": 1.777777777777778e-06,
      "loss": 0.1788,
      "step": 1000
    },
    {
      "epoch": 4.526748971193416,
      "grad_norm": 8.031317710876465,
      "learning_rate": 9.54732510288066e-07,
      "loss": 0.1802,
      "step": 1100
    },
    {
      "epoch": 4.938271604938271,
      "grad_norm": 9.72697925567627,
      "learning_rate": 1.3168724279835392e-07,
      "loss": 0.1668,
      "step": 1200
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.856701030927835,
      "eval_f1": 0.8577030087970102,
      "eval_loss": 0.4302367568016052,
      "eval_runtime": 2.5068,
      "eval_samples_per_second": 386.947,
      "eval_steps_per_second": 24.334,
      "step": 1215
    }
  ],
  "logging_steps": 100,
  "max_steps": 1215,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 5,
  "save_steps": 500,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 749706402995280.0,
  "train_batch_size": 16,
  "trial_name": null,
  "trial_params": null
}