{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 4.0,
  "eval_steps": 500,
  "global_step": 2116,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.1890359168241966,
      "grad_norm": 0.5107442140579224,
      "learning_rate": 5e-05,
      "loss": 2.0482,
      "step": 100
    },
    {
      "epoch": 0.3780718336483932,
      "grad_norm": 0.5563424229621887,
      "learning_rate": 5e-05,
      "loss": 1.8835,
      "step": 200
    },
    {
      "epoch": 0.5671077504725898,
      "grad_norm": 0.4735371172428131,
      "learning_rate": 5e-05,
      "loss": 1.8512,
      "step": 300
    },
    {
      "epoch": 0.7561436672967864,
      "grad_norm": 0.4543495178222656,
      "learning_rate": 5e-05,
      "loss": 1.8489,
      "step": 400
    },
    {
      "epoch": 0.945179584120983,
      "grad_norm": 0.5007112622261047,
      "learning_rate": 5e-05,
      "loss": 1.8439,
      "step": 500
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.5733333333333334,
      "eval_loss": 1.6194567680358887,
      "eval_runtime": 5.4154,
      "eval_samples_per_second": 92.33,
      "eval_steps_per_second": 11.634,
      "step": 529
    },
    {
      "epoch": 1.0,
      "eval_exact_match": 17.0,
      "eval_f1": 26.880952380952387,
      "step": 529
    },
    {
      "epoch": 1.1342155009451795,
      "grad_norm": 0.5181360244750977,
      "learning_rate": 5e-05,
      "loss": 1.8217,
      "step": 600
    },
    {
      "epoch": 1.3232514177693762,
      "grad_norm": 1.1348918676376343,
      "learning_rate": 5e-05,
      "loss": 1.7866,
      "step": 700
    },
    {
      "epoch": 1.5122873345935728,
      "grad_norm": 0.661758303642273,
      "learning_rate": 5e-05,
      "loss": 1.7934,
      "step": 800
    },
    {
      "epoch": 1.7013232514177694,
      "grad_norm": 0.5634772777557373,
      "learning_rate": 5e-05,
      "loss": 1.7731,
      "step": 900
    },
    {
      "epoch": 1.8903591682419658,
      "grad_norm": 0.6698930263519287,
      "learning_rate": 5e-05,
      "loss": 1.7969,
      "step": 1000
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.5751794871794872,
      "eval_loss": 1.603736162185669,
      "eval_runtime": 5.0986,
      "eval_samples_per_second": 98.066,
      "eval_steps_per_second": 12.356,
      "step": 1058
    },
    {
      "epoch": 2.0,
      "eval_exact_match": 17.6,
      "eval_f1": 27.218095238095238,
      "step": 1058
    },
    {
      "epoch": 2.0793950850661624,
      "grad_norm": 0.6265193223953247,
      "learning_rate": 5e-05,
      "loss": 1.7416,
      "step": 1100
    },
    {
      "epoch": 2.268431001890359,
      "grad_norm": 0.812364935874939,
      "learning_rate": 5e-05,
      "loss": 1.6957,
      "step": 1200
    },
    {
      "epoch": 2.4574669187145557,
      "grad_norm": 1.0972926616668701,
      "learning_rate": 5e-05,
      "loss": 1.7033,
      "step": 1300
    },
    {
      "epoch": 2.6465028355387523,
      "grad_norm": 1.0831658840179443,
      "learning_rate": 5e-05,
      "loss": 1.6935,
      "step": 1400
    },
    {
      "epoch": 2.835538752362949,
      "grad_norm": 1.0022494792938232,
      "learning_rate": 5e-05,
      "loss": 1.6799,
      "step": 1500
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.573025641025641,
      "eval_loss": 1.6428505182266235,
      "eval_runtime": 5.1488,
      "eval_samples_per_second": 97.111,
      "eval_steps_per_second": 12.236,
      "step": 1587
    },
    {
      "epoch": 3.0,
      "eval_exact_match": 20.2,
      "eval_f1": 27.93809523809523,
      "step": 1587
    },
    {
      "epoch": 3.0245746691871456,
      "grad_norm": 0.8240587711334229,
      "learning_rate": 5e-05,
      "loss": 1.6618,
      "step": 1600
    },
    {
      "epoch": 3.213610586011342,
      "grad_norm": 1.2994645833969116,
      "learning_rate": 5e-05,
      "loss": 1.5736,
      "step": 1700
    },
    {
      "epoch": 3.402646502835539,
      "grad_norm": 1.0891159772872925,
      "learning_rate": 5e-05,
      "loss": 1.5798,
      "step": 1800
    },
    {
      "epoch": 3.5916824196597354,
      "grad_norm": 1.3826475143432617,
      "learning_rate": 5e-05,
      "loss": 1.5957,
      "step": 1900
    },
    {
      "epoch": 3.780718336483932,
      "grad_norm": 0.9901586174964905,
      "learning_rate": 5e-05,
      "loss": 1.6026,
      "step": 2000
    },
    {
      "epoch": 3.9697542533081287,
      "grad_norm": 1.0607813596725464,
      "learning_rate": 5e-05,
      "loss": 1.6106,
      "step": 2100
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.5708717948717948,
      "eval_loss": 1.7143596410751343,
      "eval_runtime": 5.3942,
      "eval_samples_per_second": 92.693,
      "eval_steps_per_second": 11.679,
      "step": 2116
    },
    {
      "epoch": 4.0,
      "eval_exact_match": 19.0,
      "eval_f1": 27.71428571428571,
      "step": 2116
    }
  ],
  "logging_steps": 100,
  "max_steps": 26450,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 50,
  "save_steps": 500,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 1.3034850571255808e+17,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}