{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 10.0,
  "eval_steps": 500,
  "global_step": 11320,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.8833922261484098,
      "grad_norm": 13501.556640625,
      "learning_rate": 0.0005993999999999999,
      "loss": 1.1808,
      "step": 1000
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.22445885182474118,
      "eval_loss": 4.3484416007995605,
      "eval_runtime": 7.0864,
      "eval_samples_per_second": 36.972,
      "eval_steps_per_second": 1.27,
      "step": 1132
    },
    {
      "epoch": 1.76678445229682,
      "grad_norm": 13690.3916015625,
      "learning_rate": 0.0005419186046511628,
      "loss": 0.9761,
      "step": 2000
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.27711716287477034,
      "eval_loss": 3.9408960342407227,
      "eval_runtime": 6.3793,
      "eval_samples_per_second": 41.07,
      "eval_steps_per_second": 1.411,
      "step": 2264
    },
    {
      "epoch": 2.65017667844523,
      "grad_norm": 11980.6181640625,
      "learning_rate": 0.00048377906976744176,
      "loss": 0.9067,
      "step": 3000
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.3071958889171061,
      "eval_loss": 3.7296414375305176,
      "eval_runtime": 6.3284,
      "eval_samples_per_second": 41.401,
      "eval_steps_per_second": 1.422,
      "step": 3396
    },
    {
      "epoch": 3.53356890459364,
      "grad_norm": 10367.0361328125,
      "learning_rate": 0.0004256395348837209,
      "loss": 0.8612,
      "step": 4000
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.3292451561823098,
      "eval_loss": 3.5685932636260986,
      "eval_runtime": 6.4008,
      "eval_samples_per_second": 40.933,
      "eval_steps_per_second": 1.406,
      "step": 4528
    },
    {
      "epoch": 4.41696113074205,
      "grad_norm": 10069.853515625,
      "learning_rate": 0.0003675,
      "loss": 0.8249,
      "step": 5000
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.3458941455908935,
      "eval_loss": 3.458752393722534,
      "eval_runtime": 6.5228,
      "eval_samples_per_second": 40.167,
      "eval_steps_per_second": 1.38,
      "step": 5660
    },
    {
      "epoch": 5.30035335689046,
      "grad_norm": 10210.6884765625,
      "learning_rate": 0.00030936046511627905,
      "loss": 0.7961,
      "step": 6000
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.3591072735692625,
      "eval_loss": 3.375469446182251,
      "eval_runtime": 6.438,
      "eval_samples_per_second": 40.696,
      "eval_steps_per_second": 1.398,
      "step": 6792
    },
    {
      "epoch": 6.18374558303887,
      "grad_norm": 10369.4931640625,
      "learning_rate": 0.0002512209302325581,
      "loss": 0.7736,
      "step": 7000
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.3692057184685021,
      "eval_loss": 3.306090831756592,
      "eval_runtime": 6.274,
      "eval_samples_per_second": 41.76,
      "eval_steps_per_second": 1.434,
      "step": 7924
    },
    {
      "epoch": 7.067137809187279,
      "grad_norm": 10323.146484375,
      "learning_rate": 0.00019308139534883722,
      "loss": 0.7547,
      "step": 8000
    },
    {
      "epoch": 7.950530035335689,
      "grad_norm": 10506.6484375,
      "learning_rate": 0.00013494186046511625,
      "loss": 0.7365,
      "step": 9000
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.3775862326526344,
      "eval_loss": 3.2525575160980225,
      "eval_runtime": 6.6899,
      "eval_samples_per_second": 39.164,
      "eval_steps_per_second": 1.345,
      "step": 9056
    },
    {
      "epoch": 8.8339222614841,
      "grad_norm": 10966.0029296875,
      "learning_rate": 7.680232558139534e-05,
      "loss": 0.7209,
      "step": 10000
    },
    {
      "epoch": 9.0,
      "eval_accuracy": 0.3828520637576373,
      "eval_loss": 3.2175469398498535,
      "eval_runtime": 6.6835,
      "eval_samples_per_second": 39.201,
      "eval_steps_per_second": 1.347,
      "step": 10188
    },
    {
      "epoch": 9.717314487632509,
      "grad_norm": 11004.7744140625,
      "learning_rate": 1.8662790697674418e-05,
      "loss": 0.7088,
      "step": 11000
    },
    {
      "epoch": 10.0,
      "eval_accuracy": 0.3858024230292347,
      "eval_loss": 3.1980981826782227,
      "eval_runtime": 6.3685,
      "eval_samples_per_second": 41.14,
      "eval_steps_per_second": 1.413,
      "step": 11320
    },
    {
      "epoch": 10.0,
      "step": 11320,
      "total_flos": 9.464519983104e+16,
      "train_loss": 0.8361858516194374,
      "train_runtime": 10174.633,
      "train_samples_per_second": 35.6,
      "train_steps_per_second": 1.113
    }
  ],
  "logging_steps": 1000,
  "max_steps": 11320,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 10,
  "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": 9.464519983104e+16,
  "train_batch_size": 32,
  "trial_name": null,
  "trial_params": null
}