{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 10.0,
  "eval_steps": 500,
  "global_step": 140,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.07142857142857142,
      "grad_norm": 179.0,
      "learning_rate": 1.4285714285714285e-05,
      "loss": 46.5124,
      "step": 1
    },
    {
      "epoch": 0.35714285714285715,
      "grad_norm": 134.0,
      "learning_rate": 7.142857142857143e-05,
      "loss": 45.4999,
      "step": 5
    },
    {
      "epoch": 0.7142857142857143,
      "grad_norm": 19.5,
      "learning_rate": 0.00014285714285714287,
      "loss": 30.8653,
      "step": 10
    },
    {
      "epoch": 1.0,
      "eval_loss": 10.663816452026367,
      "eval_runtime": 0.2491,
      "eval_samples_per_second": 40.143,
      "eval_steps_per_second": 4.014,
      "step": 14
    },
    {
      "epoch": 1.0714285714285714,
      "grad_norm": 12.125,
      "learning_rate": 0.00019996891820008164,
      "loss": 22.9738,
      "step": 15
    },
    {
      "epoch": 1.4285714285714286,
      "grad_norm": 4.53125,
      "learning_rate": 0.00019888308262251285,
      "loss": 19.8669,
      "step": 20
    },
    {
      "epoch": 1.7857142857142856,
      "grad_norm": 5.5625,
      "learning_rate": 0.0001962624246950012,
      "loss": 18.5328,
      "step": 25
    },
    {
      "epoch": 2.0,
      "eval_loss": 7.303110599517822,
      "eval_runtime": 0.2339,
      "eval_samples_per_second": 42.753,
      "eval_steps_per_second": 4.275,
      "step": 28
    },
    {
      "epoch": 2.142857142857143,
      "grad_norm": 10.25,
      "learning_rate": 0.00019214762118704076,
      "loss": 17.3052,
      "step": 30
    },
    {
      "epoch": 2.5,
      "grad_norm": 17.5,
      "learning_rate": 0.00018660254037844388,
      "loss": 15.1835,
      "step": 35
    },
    {
      "epoch": 2.857142857142857,
      "grad_norm": 24.875,
      "learning_rate": 0.00017971325072229226,
      "loss": 11.486,
      "step": 40
    },
    {
      "epoch": 3.0,
      "eval_loss": 6.628046989440918,
      "eval_runtime": 0.2415,
      "eval_samples_per_second": 41.402,
      "eval_steps_per_second": 4.14,
      "step": 42
    },
    {
      "epoch": 3.2142857142857144,
      "grad_norm": 26.5,
      "learning_rate": 0.00017158668492597186,
      "loss": 6.607,
      "step": 45
    },
    {
      "epoch": 3.571428571428571,
      "grad_norm": 7.125,
      "learning_rate": 0.00016234898018587337,
      "loss": 3.2178,
      "step": 50
    },
    {
      "epoch": 3.928571428571429,
      "grad_norm": 5.4375,
      "learning_rate": 0.0001521435203379498,
      "loss": 2.4959,
      "step": 55
    },
    {
      "epoch": 4.0,
      "eval_loss": 3.508697032928467,
      "eval_runtime": 0.2457,
      "eval_samples_per_second": 40.698,
      "eval_steps_per_second": 4.07,
      "step": 56
    },
    {
      "epoch": 4.285714285714286,
      "grad_norm": 2.421875,
      "learning_rate": 0.00014112871031306119,
      "loss": 2.101,
      "step": 60
    },
    {
      "epoch": 4.642857142857143,
      "grad_norm": 1.03125,
      "learning_rate": 0.00012947551744109043,
      "loss": 1.8643,
      "step": 65
    },
    {
      "epoch": 5.0,
      "grad_norm": 0.65234375,
      "learning_rate": 0.00011736481776669306,
      "loss": 1.742,
      "step": 70
    },
    {
      "epoch": 5.0,
      "eval_loss": 3.0216174125671387,
      "eval_runtime": 0.2345,
      "eval_samples_per_second": 42.653,
      "eval_steps_per_second": 4.265,
      "step": 70
    },
    {
      "epoch": 5.357142857142857,
      "grad_norm": 0.75390625,
      "learning_rate": 0.00010498458856606972,
      "loss": 1.6462,
      "step": 75
    },
    {
      "epoch": 5.714285714285714,
      "grad_norm": 0.73046875,
      "learning_rate": 9.252699064135758e-05,
      "loss": 1.5971,
      "step": 80
    },
    {
      "epoch": 6.0,
      "eval_loss": 2.880188465118408,
      "eval_runtime": 0.2347,
      "eval_samples_per_second": 42.612,
      "eval_steps_per_second": 4.261,
      "step": 84
    },
    {
      "epoch": 6.071428571428571,
      "grad_norm": 0.474609375,
      "learning_rate": 8.018538568006027e-05,
      "loss": 1.5391,
      "step": 85
    },
    {
      "epoch": 6.428571428571429,
      "grad_norm": 0.62109375,
      "learning_rate": 6.815133497483157e-05,
      "loss": 1.5124,
      "step": 90
    },
    {
      "epoch": 6.785714285714286,
      "grad_norm": 0.408203125,
      "learning_rate": 5.6611626088244194e-05,
      "loss": 1.4792,
      "step": 95
    },
    {
      "epoch": 7.0,
      "eval_loss": 2.8307275772094727,
      "eval_runtime": 0.235,
      "eval_samples_per_second": 42.562,
      "eval_steps_per_second": 4.256,
      "step": 98
    },
    {
      "epoch": 7.142857142857143,
      "grad_norm": 0.443359375,
      "learning_rate": 4.574537361342407e-05,
      "loss": 1.4514,
      "step": 100
    },
    {
      "epoch": 7.5,
      "grad_norm": 0.60546875,
      "learning_rate": 3.5721239031346066e-05,
      "loss": 1.4383,
      "step": 105
    },
    {
      "epoch": 7.857142857142857,
      "grad_norm": 0.8828125,
      "learning_rate": 2.669481281701739e-05,
      "loss": 1.4333,
      "step": 110
    },
    {
      "epoch": 8.0,
      "eval_loss": 2.808138608932495,
      "eval_runtime": 0.2344,
      "eval_samples_per_second": 42.659,
      "eval_steps_per_second": 4.266,
      "step": 112
    },
    {
      "epoch": 8.214285714285714,
      "grad_norm": 0.482421875,
      "learning_rate": 1.880619942841435e-05,
      "loss": 1.4232,
      "step": 115
    },
    {
      "epoch": 8.571428571428571,
      "grad_norm": 0.78125,
      "learning_rate": 1.2177842662977135e-05,
      "loss": 1.4159,
      "step": 120
    },
    {
      "epoch": 8.928571428571429,
      "grad_norm": 0.53515625,
      "learning_rate": 6.9126251355795864e-06,
      "loss": 1.4129,
      "step": 125
    },
    {
      "epoch": 9.0,
      "eval_loss": 2.8151421546936035,
      "eval_runtime": 0.2356,
      "eval_samples_per_second": 42.443,
      "eval_steps_per_second": 4.244,
      "step": 126
    },
    {
      "epoch": 9.285714285714286,
      "grad_norm": 0.427734375,
      "learning_rate": 3.092271377092215e-06,
      "loss": 1.4193,
      "step": 130
    },
    {
      "epoch": 9.642857142857142,
      "grad_norm": 0.6875,
      "learning_rate": 7.760793399827937e-07,
      "loss": 1.4139,
      "step": 135
    },
    {
      "epoch": 10.0,
      "grad_norm": 0.376953125,
      "learning_rate": 0.0,
      "loss": 1.4048,
      "step": 140
    },
    {
      "epoch": 10.0,
      "eval_loss": 2.812880277633667,
      "eval_runtime": 0.2405,
      "eval_samples_per_second": 41.578,
      "eval_steps_per_second": 4.158,
      "step": 140
    },
    {
      "epoch": 10.0,
      "step": 140,
      "total_flos": 4.268849030789857e+17,
      "train_loss": 7.876109651156834,
      "train_runtime": 340.0833,
      "train_samples_per_second": 25.758,
      "train_steps_per_second": 0.412
    }
  ],
  "logging_steps": 5,
  "max_steps": 140,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 10,
  "save_steps": 100,
  "total_flos": 4.268849030789857e+17,
  "train_batch_size": 4,
  "trial_name": null,
  "trial_params": null
}