{
  "best_global_step": 4950,
  "best_metric": 3.057368755340576,
  "best_model_checkpoint": null,
  "epoch": 15.151515151515152,
  "eval_steps": 500,
  "global_step": 7500,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.00202020202020202,
      "grad_norm": 29.595918655395508,
      "learning_rate": 0.0,
      "loss": 10.3213,
      "step": 1
    },
    {
      "epoch": 1.0,
      "grad_norm": 1.2207119464874268,
      "learning_rate": 9.88e-05,
      "loss": 4.8361,
      "step": 495
    },
    {
      "epoch": 1.0,
      "eval_loss": 3.7897000312805176,
      "eval_runtime": 19.6861,
      "eval_samples_per_second": 1614.843,
      "eval_steps_per_second": 6.35,
      "step": 495
    },
    {
      "epoch": 2.0,
      "grad_norm": 1.0401272773742676,
      "learning_rate": 9.478678038379531e-05,
      "loss": 3.538,
      "step": 990
    },
    {
      "epoch": 2.0,
      "eval_loss": 3.4326586723327637,
      "eval_runtime": 19.7059,
      "eval_samples_per_second": 1613.223,
      "eval_steps_per_second": 6.343,
      "step": 990
    },
    {
      "epoch": 3.0,
      "grad_norm": 1.0095438957214355,
      "learning_rate": 8.950959488272922e-05,
      "loss": 3.2785,
      "step": 1485
    },
    {
      "epoch": 3.0,
      "eval_loss": 3.287219285964966,
      "eval_runtime": 19.5988,
      "eval_samples_per_second": 1622.034,
      "eval_steps_per_second": 6.378,
      "step": 1485
    },
    {
      "epoch": 4.0,
      "grad_norm": 1.0992753505706787,
      "learning_rate": 8.42324093816631e-05,
      "loss": 3.1342,
      "step": 1980
    },
    {
      "epoch": 4.0,
      "eval_loss": 3.207063674926758,
      "eval_runtime": 19.9631,
      "eval_samples_per_second": 1592.44,
      "eval_steps_per_second": 6.262,
      "step": 1980
    },
    {
      "epoch": 5.0,
      "grad_norm": 1.065517783164978,
      "learning_rate": 7.895522388059702e-05,
      "loss": 3.032,
      "step": 2475
    },
    {
      "epoch": 5.0,
      "eval_loss": 3.151052713394165,
      "eval_runtime": 19.6853,
      "eval_samples_per_second": 1614.911,
      "eval_steps_per_second": 6.35,
      "step": 2475
    },
    {
      "epoch": 6.0,
      "grad_norm": 1.0753278732299805,
      "learning_rate": 7.367803837953093e-05,
      "loss": 2.9514,
      "step": 2970
    },
    {
      "epoch": 6.0,
      "eval_loss": 3.113241195678711,
      "eval_runtime": 19.795,
      "eval_samples_per_second": 1605.965,
      "eval_steps_per_second": 6.315,
      "step": 2970
    },
    {
      "epoch": 7.0,
      "grad_norm": 1.1069707870483398,
      "learning_rate": 6.840085287846483e-05,
      "loss": 2.8836,
      "step": 3465
    },
    {
      "epoch": 7.0,
      "eval_loss": 3.090135097503662,
      "eval_runtime": 19.7634,
      "eval_samples_per_second": 1608.533,
      "eval_steps_per_second": 6.325,
      "step": 3465
    },
    {
      "epoch": 8.0,
      "grad_norm": 1.1921679973602295,
      "learning_rate": 6.312366737739872e-05,
      "loss": 2.8231,
      "step": 3960
    },
    {
      "epoch": 8.0,
      "eval_loss": 3.0735952854156494,
      "eval_runtime": 19.9524,
      "eval_samples_per_second": 1593.29,
      "eval_steps_per_second": 6.265,
      "step": 3960
    },
    {
      "epoch": 9.0,
      "grad_norm": 1.208296537399292,
      "learning_rate": 5.784648187633263e-05,
      "loss": 2.7678,
      "step": 4455
    },
    {
      "epoch": 9.0,
      "eval_loss": 3.0633397102355957,
      "eval_runtime": 19.7455,
      "eval_samples_per_second": 1609.99,
      "eval_steps_per_second": 6.331,
      "step": 4455
    },
    {
      "epoch": 10.0,
      "grad_norm": 1.3137807846069336,
      "learning_rate": 5.256929637526653e-05,
      "loss": 2.7164,
      "step": 4950
    },
    {
      "epoch": 10.0,
      "eval_loss": 3.057368755340576,
      "eval_runtime": 19.795,
      "eval_samples_per_second": 1605.965,
      "eval_steps_per_second": 6.315,
      "step": 4950
    },
    {
      "epoch": 11.0,
      "grad_norm": 1.3745468854904175,
      "learning_rate": 4.7292110874200426e-05,
      "loss": 2.6678,
      "step": 5445
    },
    {
      "epoch": 11.0,
      "eval_loss": 3.0588877201080322,
      "eval_runtime": 19.7143,
      "eval_samples_per_second": 1612.535,
      "eval_steps_per_second": 6.341,
      "step": 5445
    },
    {
      "epoch": 12.0,
      "grad_norm": 1.4381449222564697,
      "learning_rate": 4.201492537313433e-05,
      "loss": 2.6213,
      "step": 5940
    },
    {
      "epoch": 12.0,
      "eval_loss": 3.0614066123962402,
      "eval_runtime": 19.6703,
      "eval_samples_per_second": 1616.141,
      "eval_steps_per_second": 6.355,
      "step": 5940
    },
    {
      "epoch": 13.0,
      "grad_norm": 1.5913258790969849,
      "learning_rate": 3.673773987206823e-05,
      "loss": 2.5765,
      "step": 6435
    },
    {
      "epoch": 13.0,
      "eval_loss": 3.0655486583709717,
      "eval_runtime": 19.7216,
      "eval_samples_per_second": 1611.937,
      "eval_steps_per_second": 6.338,
      "step": 6435
    },
    {
      "epoch": 14.0,
      "grad_norm": 1.6668187379837036,
      "learning_rate": 3.1460554371002134e-05,
      "loss": 2.5352,
      "step": 6930
    },
    {
      "epoch": 14.0,
      "eval_loss": 3.0753347873687744,
      "eval_runtime": 19.8126,
      "eval_samples_per_second": 1604.537,
      "eval_steps_per_second": 6.309,
      "step": 6930
    },
    {
      "epoch": 15.0,
      "grad_norm": 1.680243730545044,
      "learning_rate": 2.6183368869936037e-05,
      "loss": 2.4983,
      "step": 7425
    },
    {
      "epoch": 15.0,
      "eval_loss": 3.081855058670044,
      "eval_runtime": 19.7547,
      "eval_samples_per_second": 1609.24,
      "eval_steps_per_second": 6.328,
      "step": 7425
    }
  ],
  "logging_steps": 500,
  "max_steps": 9900,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 20,
  "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": 1.2527613812736e+17,
  "train_batch_size": 256,
  "trial_name": null,
  "trial_params": null
}