{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 1.303030303030303,
  "eval_steps": 15,
  "global_step": 75,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.017316017316017316,
      "eval_loss": 0.8123326897621155,
      "eval_runtime": 21.4215,
      "eval_samples_per_second": 4.528,
      "eval_steps_per_second": 0.607,
      "step": 1
    },
    {
      "epoch": 0.05194805194805195,
      "grad_norm": 1.44769287109375,
      "learning_rate": 3e-05,
      "loss": 3.0463,
      "step": 3
    },
    {
      "epoch": 0.1038961038961039,
      "grad_norm": 0.9673851728439331,
      "learning_rate": 6e-05,
      "loss": 2.883,
      "step": 6
    },
    {
      "epoch": 0.15584415584415584,
      "grad_norm": 1.2237026691436768,
      "learning_rate": 9e-05,
      "loss": 3.0048,
      "step": 9
    },
    {
      "epoch": 0.2077922077922078,
      "grad_norm": 1.0807669162750244,
      "learning_rate": 9.9962857531815e-05,
      "loss": 2.8812,
      "step": 12
    },
    {
      "epoch": 0.2597402597402597,
      "grad_norm": 0.8111446499824524,
      "learning_rate": 9.976801044672608e-05,
      "loss": 2.8254,
      "step": 15
    },
    {
      "epoch": 0.2597402597402597,
      "eval_loss": 0.7221285700798035,
      "eval_runtime": 21.8495,
      "eval_samples_per_second": 4.439,
      "eval_steps_per_second": 0.595,
      "step": 15
    },
    {
      "epoch": 0.3116883116883117,
      "grad_norm": 0.7993184924125671,
      "learning_rate": 9.940682350363912e-05,
      "loss": 2.5783,
      "step": 18
    },
    {
      "epoch": 0.36363636363636365,
      "grad_norm": 0.88717120885849,
      "learning_rate": 9.888050389939172e-05,
      "loss": 3.0288,
      "step": 21
    },
    {
      "epoch": 0.4155844155844156,
      "grad_norm": 0.9446913003921509,
      "learning_rate": 9.819081075450014e-05,
      "loss": 2.6752,
      "step": 24
    },
    {
      "epoch": 0.4675324675324675,
      "grad_norm": 0.7598215937614441,
      "learning_rate": 9.734004923364257e-05,
      "loss": 2.6636,
      "step": 27
    },
    {
      "epoch": 0.5194805194805194,
      "grad_norm": 0.9138664603233337,
      "learning_rate": 9.63310628410961e-05,
      "loss": 2.6451,
      "step": 30
    },
    {
      "epoch": 0.5194805194805194,
      "eval_loss": 0.6947051286697388,
      "eval_runtime": 21.8569,
      "eval_samples_per_second": 4.438,
      "eval_steps_per_second": 0.595,
      "step": 30
    },
    {
      "epoch": 0.5714285714285714,
      "grad_norm": 0.7083372473716736,
      "learning_rate": 9.516722391687902e-05,
      "loss": 2.5942,
      "step": 33
    },
    {
      "epoch": 0.6233766233766234,
      "grad_norm": 0.6996327042579651,
      "learning_rate": 9.38524223653626e-05,
      "loss": 2.648,
      "step": 36
    },
    {
      "epoch": 0.6753246753246753,
      "grad_norm": 0.7574095726013184,
      "learning_rate": 9.239105265402525e-05,
      "loss": 2.5124,
      "step": 39
    },
    {
      "epoch": 0.7272727272727273,
      "grad_norm": 0.7147637605667114,
      "learning_rate": 9.078799912580304e-05,
      "loss": 2.652,
      "step": 42
    },
    {
      "epoch": 0.7792207792207793,
      "grad_norm": 0.8316633701324463,
      "learning_rate": 8.904861967412703e-05,
      "loss": 2.5992,
      "step": 45
    },
    {
      "epoch": 0.7792207792207793,
      "eval_loss": 0.6810405254364014,
      "eval_runtime": 21.865,
      "eval_samples_per_second": 4.436,
      "eval_steps_per_second": 0.595,
      "step": 45
    },
    {
      "epoch": 0.8311688311688312,
      "grad_norm": 0.7336857318878174,
      "learning_rate": 8.717872783521047e-05,
      "loss": 2.4763,
      "step": 48
    },
    {
      "epoch": 0.8831168831168831,
      "grad_norm": 0.6515845060348511,
      "learning_rate": 8.518457335743926e-05,
      "loss": 2.5329,
      "step": 51
    },
    {
      "epoch": 0.935064935064935,
      "grad_norm": 0.7598447203636169,
      "learning_rate": 8.307282131280804e-05,
      "loss": 2.6017,
      "step": 54
    },
    {
      "epoch": 0.987012987012987,
      "grad_norm": 0.6881215572357178,
      "learning_rate": 8.085052982021847e-05,
      "loss": 2.5885,
      "step": 57
    },
    {
      "epoch": 1.0432900432900434,
      "grad_norm": 0.6885347962379456,
      "learning_rate": 7.85251264550948e-05,
      "loss": 2.3153,
      "step": 60
    },
    {
      "epoch": 1.0432900432900434,
      "eval_loss": 0.6708703637123108,
      "eval_runtime": 21.8845,
      "eval_samples_per_second": 4.432,
      "eval_steps_per_second": 0.594,
      "step": 60
    },
    {
      "epoch": 1.0952380952380953,
      "grad_norm": 0.8033732175827026,
      "learning_rate": 7.610438342416319e-05,
      "loss": 2.1058,
      "step": 63
    },
    {
      "epoch": 1.1471861471861473,
      "grad_norm": 0.7723002433776855,
      "learning_rate": 7.359639158836828e-05,
      "loss": 2.0976,
      "step": 66
    },
    {
      "epoch": 1.199134199134199,
      "grad_norm": 0.8295950889587402,
      "learning_rate": 7.10095334207501e-05,
      "loss": 2.1156,
      "step": 69
    },
    {
      "epoch": 1.251082251082251,
      "grad_norm": 0.8891462683677673,
      "learning_rate": 6.835245498966461e-05,
      "loss": 2.3097,
      "step": 72
    },
    {
      "epoch": 1.303030303030303,
      "grad_norm": 0.8947754502296448,
      "learning_rate": 6.563403706098833e-05,
      "loss": 2.2014,
      "step": 75
    },
    {
      "epoch": 1.303030303030303,
      "eval_loss": 0.6706818342208862,
      "eval_runtime": 21.8884,
      "eval_samples_per_second": 4.432,
      "eval_steps_per_second": 0.594,
      "step": 75
    }
  ],
  "logging_steps": 3,
  "max_steps": 173,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 4,
  "save_steps": 15,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 3.86404668407808e+17,
  "train_batch_size": 8,
  "trial_name": null,
  "trial_params": null
}