{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 3.0,
  "eval_steps": 500,
  "global_step": 600,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.125,
      "grad_norm": 0.2043563425540924,
      "learning_rate": 0.00019994755690455152,
      "loss": 0.2003,
      "step": 25
    },
    {
      "epoch": 0.25,
      "grad_norm": 0.16403928399085999,
      "learning_rate": 0.00019860320222975431,
      "loss": 0.1158,
      "step": 50
    },
    {
      "epoch": 0.375,
      "grad_norm": 0.2802892327308655,
      "learning_rate": 0.00019546590805746052,
      "loss": 0.1094,
      "step": 75
    },
    {
      "epoch": 0.5,
      "grad_norm": 0.17004738748073578,
      "learning_rate": 0.0001905927209998447,
      "loss": 0.1016,
      "step": 100
    },
    {
      "epoch": 0.625,
      "grad_norm": 0.14632447063922882,
      "learning_rate": 0.00018407225206909208,
      "loss": 0.1038,
      "step": 125
    },
    {
      "epoch": 0.75,
      "grad_norm": 0.21698755025863647,
      "learning_rate": 0.00017602306542969005,
      "loss": 0.101,
      "step": 150
    },
    {
      "epoch": 0.875,
      "grad_norm": 0.17689374089241028,
      "learning_rate": 0.00016659152250116812,
      "loss": 0.1029,
      "step": 175
    },
    {
      "epoch": 1.0,
      "grad_norm": 0.21386151015758514,
      "learning_rate": 0.00015594912061278626,
      "loss": 0.0918,
      "step": 200
    },
    {
      "epoch": 1.0,
      "eval_loss": 0.09140199422836304,
      "eval_runtime": 58.0967,
      "eval_samples_per_second": 3.443,
      "eval_steps_per_second": 0.861,
      "step": 200
    },
    {
      "epoch": 1.125,
      "grad_norm": 0.22129373252391815,
      "learning_rate": 0.00014428937460242417,
      "loss": 0.0694,
      "step": 225
    },
    {
      "epoch": 1.25,
      "grad_norm": 0.169610857963562,
      "learning_rate": 0.0001318242980627444,
      "loss": 0.0652,
      "step": 250
    },
    {
      "epoch": 1.375,
      "grad_norm": 0.2244400978088379,
      "learning_rate": 0.00011878054821746703,
      "loss": 0.0648,
      "step": 275
    },
    {
      "epoch": 1.5,
      "grad_norm": 0.15404100716114044,
      "learning_rate": 0.00010539530452693625,
      "loss": 0.0647,
      "step": 300
    },
    {
      "epoch": 1.625,
      "grad_norm": 0.1526843011379242,
      "learning_rate": 9.19119559638596e-05,
      "loss": 0.0549,
      "step": 325
    },
    {
      "epoch": 1.75,
      "grad_norm": 0.17617414891719818,
      "learning_rate": 7.857567537912403e-05,
      "loss": 0.0585,
      "step": 350
    },
    {
      "epoch": 1.875,
      "grad_norm": 0.18532665073871613,
      "learning_rate": 6.562896143067734e-05,
      "loss": 0.0606,
      "step": 375
    },
    {
      "epoch": 2.0,
      "grad_norm": 0.13493256270885468,
      "learning_rate": 5.3307229138275936e-05,
      "loss": 0.0531,
      "step": 400
    },
    {
      "epoch": 2.0,
      "eval_loss": 0.08346135914325714,
      "eval_runtime": 58.1824,
      "eval_samples_per_second": 3.437,
      "eval_steps_per_second": 0.859,
      "step": 400
    },
    {
      "epoch": 2.125,
      "grad_norm": 0.23295405507087708,
      "learning_rate": 4.183452924271776e-05,
      "loss": 0.031,
      "step": 425
    },
    {
      "epoch": 2.25,
      "grad_norm": 0.2066780924797058,
      "learning_rate": 3.1419474206078205e-05,
      "loss": 0.0319,
      "step": 450
    },
    {
      "epoch": 2.375,
      "grad_norm": 0.1161380186676979,
      "learning_rate": 2.2251444932035094e-05,
      "loss": 0.0306,
      "step": 475
    },
    {
      "epoch": 2.5,
      "grad_norm": 0.2266598790884018,
      "learning_rate": 1.4497147180928027e-05,
      "loss": 0.0259,
      "step": 500
    },
    {
      "epoch": 2.625,
      "grad_norm": 0.1240050196647644,
      "learning_rate": 8.297580295566575e-06,
      "loss": 0.0274,
      "step": 525
    },
    {
      "epoch": 2.75,
      "grad_norm": 0.13692007958889008,
      "learning_rate": 3.7654733565969826e-06,
      "loss": 0.0255,
      "step": 550
    },
    {
      "epoch": 2.875,
      "grad_norm": 0.1393318921327591,
      "learning_rate": 9.832353867893386e-07,
      "loss": 0.0269,
      "step": 575
    },
    {
      "epoch": 3.0,
      "grad_norm": 0.24243566393852234,
      "learning_rate": 1.4568764593603234e-09,
      "loss": 0.03,
      "step": 600
    },
    {
      "epoch": 3.0,
      "eval_loss": 0.10383956134319305,
      "eval_runtime": 58.2331,
      "eval_samples_per_second": 3.434,
      "eval_steps_per_second": 0.859,
      "step": 600
    }
  ],
  "logging_steps": 25,
  "max_steps": 600,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 3,
  "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": 4.488064275802522e+16,
  "train_batch_size": 4,
  "trial_name": null,
  "trial_params": null
}