{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 20.51282051282051,
  "eval_steps": 100,
  "global_step": 4000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.5128205128205128,
      "grad_norm": 73.68424224853516,
      "learning_rate": 9.9907e-06,
      "loss": 3.1457,
      "step": 100
    },
    {
      "epoch": 0.5128205128205128,
      "eval_loss": 2.2357213497161865,
      "eval_runtime": 36.4689,
      "eval_samples_per_second": 10.776,
      "eval_steps_per_second": 1.371,
      "step": 100
    },
    {
      "epoch": 1.0256410256410255,
      "grad_norm": 57.47202682495117,
      "learning_rate": 9.980800000000001e-06,
      "loss": 2.1614,
      "step": 200
    },
    {
      "epoch": 1.0256410256410255,
      "eval_loss": 2.0913825035095215,
      "eval_runtime": 36.349,
      "eval_samples_per_second": 10.812,
      "eval_steps_per_second": 1.376,
      "step": 200
    },
    {
      "epoch": 1.5384615384615383,
      "grad_norm": 176.88357543945312,
      "learning_rate": 9.970800000000001e-06,
      "loss": 2.0388,
      "step": 300
    },
    {
      "epoch": 1.5384615384615383,
      "eval_loss": 2.003911018371582,
      "eval_runtime": 36.2551,
      "eval_samples_per_second": 10.84,
      "eval_steps_per_second": 1.379,
      "step": 300
    },
    {
      "epoch": 2.051282051282051,
      "grad_norm": 55.31132507324219,
      "learning_rate": 9.960800000000001e-06,
      "loss": 1.9285,
      "step": 400
    },
    {
      "epoch": 2.051282051282051,
      "eval_loss": 1.9796075820922852,
      "eval_runtime": 36.4437,
      "eval_samples_per_second": 10.784,
      "eval_steps_per_second": 1.372,
      "step": 400
    },
    {
      "epoch": 2.564102564102564,
      "grad_norm": 41.900753021240234,
      "learning_rate": 9.9508e-06,
      "loss": 1.9523,
      "step": 500
    },
    {
      "epoch": 2.564102564102564,
      "eval_loss": 1.936122179031372,
      "eval_runtime": 36.5845,
      "eval_samples_per_second": 10.742,
      "eval_steps_per_second": 1.367,
      "step": 500
    },
    {
      "epoch": 3.076923076923077,
      "grad_norm": 50.21903991699219,
      "learning_rate": 9.9408e-06,
      "loss": 1.8452,
      "step": 600
    },
    {
      "epoch": 3.076923076923077,
      "eval_loss": 1.8883634805679321,
      "eval_runtime": 36.6015,
      "eval_samples_per_second": 10.737,
      "eval_steps_per_second": 1.366,
      "step": 600
    },
    {
      "epoch": 3.58974358974359,
      "grad_norm": 45.193939208984375,
      "learning_rate": 9.930900000000002e-06,
      "loss": 1.8403,
      "step": 700
    },
    {
      "epoch": 3.58974358974359,
      "eval_loss": 1.8506474494934082,
      "eval_runtime": 36.454,
      "eval_samples_per_second": 10.781,
      "eval_steps_per_second": 1.372,
      "step": 700
    },
    {
      "epoch": 4.102564102564102,
      "grad_norm": 27.302494049072266,
      "learning_rate": 9.920900000000002e-06,
      "loss": 1.7976,
      "step": 800
    },
    {
      "epoch": 4.102564102564102,
      "eval_loss": 1.8370662927627563,
      "eval_runtime": 36.834,
      "eval_samples_per_second": 10.67,
      "eval_steps_per_second": 1.357,
      "step": 800
    },
    {
      "epoch": 4.615384615384615,
      "grad_norm": 52.6607666015625,
      "learning_rate": 9.9109e-06,
      "loss": 1.7508,
      "step": 900
    },
    {
      "epoch": 4.615384615384615,
      "eval_loss": 1.8037244081497192,
      "eval_runtime": 36.9939,
      "eval_samples_per_second": 10.623,
      "eval_steps_per_second": 1.352,
      "step": 900
    },
    {
      "epoch": 5.128205128205128,
      "grad_norm": 59.508033752441406,
      "learning_rate": 9.9009e-06,
      "loss": 1.7383,
      "step": 1000
    },
    {
      "epoch": 5.128205128205128,
      "eval_loss": 1.7986633777618408,
      "eval_runtime": 36.8356,
      "eval_samples_per_second": 10.669,
      "eval_steps_per_second": 1.357,
      "step": 1000
    },
    {
      "epoch": 5.641025641025641,
      "grad_norm": 71.58872985839844,
      "learning_rate": 9.8909e-06,
      "loss": 1.7361,
      "step": 1100
    },
    {
      "epoch": 5.641025641025641,
      "eval_loss": 1.7810852527618408,
      "eval_runtime": 37.0957,
      "eval_samples_per_second": 10.594,
      "eval_steps_per_second": 1.348,
      "step": 1100
    },
    {
      "epoch": 6.153846153846154,
      "grad_norm": 41.782066345214844,
      "learning_rate": 9.8809e-06,
      "loss": 1.682,
      "step": 1200
    },
    {
      "epoch": 6.153846153846154,
      "eval_loss": 1.777554988861084,
      "eval_runtime": 36.9173,
      "eval_samples_per_second": 10.645,
      "eval_steps_per_second": 1.354,
      "step": 1200
    },
    {
      "epoch": 6.666666666666667,
      "grad_norm": 40.28728485107422,
      "learning_rate": 9.8709e-06,
      "loss": 1.7216,
      "step": 1300
    },
    {
      "epoch": 6.666666666666667,
      "eval_loss": 1.7382572889328003,
      "eval_runtime": 36.8918,
      "eval_samples_per_second": 10.653,
      "eval_steps_per_second": 1.355,
      "step": 1300
    },
    {
      "epoch": 7.17948717948718,
      "grad_norm": 104.99211883544922,
      "learning_rate": 9.8609e-06,
      "loss": 1.6534,
      "step": 1400
    },
    {
      "epoch": 7.17948717948718,
      "eval_loss": 1.76559579372406,
      "eval_runtime": 36.3398,
      "eval_samples_per_second": 10.815,
      "eval_steps_per_second": 1.376,
      "step": 1400
    },
    {
      "epoch": 7.6923076923076925,
      "grad_norm": 29.326631546020508,
      "learning_rate": 9.8509e-06,
      "loss": 1.707,
      "step": 1500
    },
    {
      "epoch": 7.6923076923076925,
      "eval_loss": 1.750089406967163,
      "eval_runtime": 36.4098,
      "eval_samples_per_second": 10.794,
      "eval_steps_per_second": 1.373,
      "step": 1500
    },
    {
      "epoch": 8.205128205128204,
      "grad_norm": 37.941261291503906,
      "learning_rate": 9.840900000000001e-06,
      "loss": 1.6554,
      "step": 1600
    },
    {
      "epoch": 8.205128205128204,
      "eval_loss": 1.6900651454925537,
      "eval_runtime": 36.3226,
      "eval_samples_per_second": 10.82,
      "eval_steps_per_second": 1.377,
      "step": 1600
    },
    {
      "epoch": 8.717948717948717,
      "grad_norm": 44.60703659057617,
      "learning_rate": 9.830900000000001e-06,
      "loss": 1.6334,
      "step": 1700
    },
    {
      "epoch": 8.717948717948717,
      "eval_loss": 1.7162973880767822,
      "eval_runtime": 36.2672,
      "eval_samples_per_second": 10.836,
      "eval_steps_per_second": 1.379,
      "step": 1700
    },
    {
      "epoch": 9.23076923076923,
      "grad_norm": 34.127254486083984,
      "learning_rate": 9.820900000000001e-06,
      "loss": 1.6345,
      "step": 1800
    },
    {
      "epoch": 9.23076923076923,
      "eval_loss": 1.6906001567840576,
      "eval_runtime": 36.2264,
      "eval_samples_per_second": 10.848,
      "eval_steps_per_second": 1.38,
      "step": 1800
    },
    {
      "epoch": 9.743589743589745,
      "grad_norm": 60.377540588378906,
      "learning_rate": 9.810900000000001e-06,
      "loss": 1.598,
      "step": 1900
    },
    {
      "epoch": 9.743589743589745,
      "eval_loss": 1.6555503606796265,
      "eval_runtime": 36.3896,
      "eval_samples_per_second": 10.8,
      "eval_steps_per_second": 1.374,
      "step": 1900
    },
    {
      "epoch": 10.256410256410255,
      "grad_norm": 20.264404296875,
      "learning_rate": 9.800900000000001e-06,
      "loss": 1.5466,
      "step": 2000
    },
    {
      "epoch": 10.256410256410255,
      "eval_loss": 1.648037075996399,
      "eval_runtime": 36.4136,
      "eval_samples_per_second": 10.793,
      "eval_steps_per_second": 1.373,
      "step": 2000
    },
    {
      "epoch": 10.76923076923077,
      "grad_norm": 27.18608856201172,
      "learning_rate": 9.790900000000001e-06,
      "loss": 1.5865,
      "step": 2100
    },
    {
      "epoch": 10.76923076923077,
      "eval_loss": 1.6171936988830566,
      "eval_runtime": 36.1051,
      "eval_samples_per_second": 10.885,
      "eval_steps_per_second": 1.385,
      "step": 2100
    },
    {
      "epoch": 11.282051282051283,
      "grad_norm": 32.486331939697266,
      "learning_rate": 9.780900000000002e-06,
      "loss": 1.5284,
      "step": 2200
    },
    {
      "epoch": 11.282051282051283,
      "eval_loss": 1.5915095806121826,
      "eval_runtime": 36.1781,
      "eval_samples_per_second": 10.863,
      "eval_steps_per_second": 1.382,
      "step": 2200
    },
    {
      "epoch": 11.794871794871796,
      "grad_norm": 65.88719940185547,
      "learning_rate": 9.770900000000002e-06,
      "loss": 1.5514,
      "step": 2300
    },
    {
      "epoch": 11.794871794871796,
      "eval_loss": 1.5879931449890137,
      "eval_runtime": 36.3934,
      "eval_samples_per_second": 10.799,
      "eval_steps_per_second": 1.374,
      "step": 2300
    },
    {
      "epoch": 12.307692307692308,
      "grad_norm": 31.737024307250977,
      "learning_rate": 9.760900000000002e-06,
      "loss": 1.4941,
      "step": 2400
    },
    {
      "epoch": 12.307692307692308,
      "eval_loss": 1.583853006362915,
      "eval_runtime": 36.3797,
      "eval_samples_per_second": 10.803,
      "eval_steps_per_second": 1.374,
      "step": 2400
    },
    {
      "epoch": 12.820512820512821,
      "grad_norm": 45.48268508911133,
      "learning_rate": 9.7509e-06,
      "loss": 1.5097,
      "step": 2500
    },
    {
      "epoch": 12.820512820512821,
      "eval_loss": 1.5559026002883911,
      "eval_runtime": 36.1605,
      "eval_samples_per_second": 10.868,
      "eval_steps_per_second": 1.383,
      "step": 2500
    },
    {
      "epoch": 13.333333333333334,
      "grad_norm": 27.500398635864258,
      "learning_rate": 9.7409e-06,
      "loss": 1.5018,
      "step": 2600
    },
    {
      "epoch": 13.333333333333334,
      "eval_loss": 1.5453521013259888,
      "eval_runtime": 36.2887,
      "eval_samples_per_second": 10.83,
      "eval_steps_per_second": 1.378,
      "step": 2600
    },
    {
      "epoch": 13.846153846153847,
      "grad_norm": 32.49728775024414,
      "learning_rate": 9.7309e-06,
      "loss": 1.4804,
      "step": 2700
    },
    {
      "epoch": 13.846153846153847,
      "eval_loss": 1.5424816608428955,
      "eval_runtime": 36.359,
      "eval_samples_per_second": 10.809,
      "eval_steps_per_second": 1.375,
      "step": 2700
    },
    {
      "epoch": 14.35897435897436,
      "grad_norm": 38.46280288696289,
      "learning_rate": 9.7209e-06,
      "loss": 1.4826,
      "step": 2800
    },
    {
      "epoch": 14.35897435897436,
      "eval_loss": 1.5317177772521973,
      "eval_runtime": 36.3362,
      "eval_samples_per_second": 10.816,
      "eval_steps_per_second": 1.376,
      "step": 2800
    },
    {
      "epoch": 14.871794871794872,
      "grad_norm": 16.075960159301758,
      "learning_rate": 9.7109e-06,
      "loss": 1.4568,
      "step": 2900
    },
    {
      "epoch": 14.871794871794872,
      "eval_loss": 1.5241832733154297,
      "eval_runtime": 36.2025,
      "eval_samples_per_second": 10.856,
      "eval_steps_per_second": 1.381,
      "step": 2900
    },
    {
      "epoch": 15.384615384615385,
      "grad_norm": 27.334318161010742,
      "learning_rate": 9.7009e-06,
      "loss": 1.4176,
      "step": 3000
    },
    {
      "epoch": 15.384615384615385,
      "eval_loss": 1.520580768585205,
      "eval_runtime": 36.398,
      "eval_samples_per_second": 10.797,
      "eval_steps_per_second": 1.374,
      "step": 3000
    },
    {
      "epoch": 15.897435897435898,
      "grad_norm": 94.90784454345703,
      "learning_rate": 9.6909e-06,
      "loss": 1.4681,
      "step": 3100
    },
    {
      "epoch": 15.897435897435898,
      "eval_loss": 1.5268648862838745,
      "eval_runtime": 36.0541,
      "eval_samples_per_second": 10.9,
      "eval_steps_per_second": 1.387,
      "step": 3100
    },
    {
      "epoch": 16.41025641025641,
      "grad_norm": 16.697856903076172,
      "learning_rate": 9.6809e-06,
      "loss": 1.454,
      "step": 3200
    },
    {
      "epoch": 16.41025641025641,
      "eval_loss": 1.5157753229141235,
      "eval_runtime": 36.3172,
      "eval_samples_per_second": 10.821,
      "eval_steps_per_second": 1.377,
      "step": 3200
    },
    {
      "epoch": 16.923076923076923,
      "grad_norm": 54.05553436279297,
      "learning_rate": 9.670900000000001e-06,
      "loss": 1.4309,
      "step": 3300
    },
    {
      "epoch": 16.923076923076923,
      "eval_loss": 1.516249179840088,
      "eval_runtime": 36.2632,
      "eval_samples_per_second": 10.837,
      "eval_steps_per_second": 1.379,
      "step": 3300
    },
    {
      "epoch": 17.435897435897434,
      "grad_norm": 47.010475158691406,
      "learning_rate": 9.660900000000001e-06,
      "loss": 1.4571,
      "step": 3400
    },
    {
      "epoch": 17.435897435897434,
      "eval_loss": 1.5018247365951538,
      "eval_runtime": 36.2118,
      "eval_samples_per_second": 10.853,
      "eval_steps_per_second": 1.381,
      "step": 3400
    },
    {
      "epoch": 17.94871794871795,
      "grad_norm": 52.865718841552734,
      "learning_rate": 9.650900000000001e-06,
      "loss": 1.4168,
      "step": 3500
    },
    {
      "epoch": 17.94871794871795,
      "eval_loss": 1.4993616342544556,
      "eval_runtime": 36.2572,
      "eval_samples_per_second": 10.839,
      "eval_steps_per_second": 1.379,
      "step": 3500
    },
    {
      "epoch": 18.46153846153846,
      "grad_norm": 30.117380142211914,
      "learning_rate": 9.640900000000001e-06,
      "loss": 1.4275,
      "step": 3600
    },
    {
      "epoch": 18.46153846153846,
      "eval_loss": 1.4899998903274536,
      "eval_runtime": 36.4696,
      "eval_samples_per_second": 10.776,
      "eval_steps_per_second": 1.371,
      "step": 3600
    },
    {
      "epoch": 18.974358974358974,
      "grad_norm": 31.10028076171875,
      "learning_rate": 9.630900000000001e-06,
      "loss": 1.4148,
      "step": 3700
    },
    {
      "epoch": 18.974358974358974,
      "eval_loss": 1.5231629610061646,
      "eval_runtime": 36.4604,
      "eval_samples_per_second": 10.779,
      "eval_steps_per_second": 1.371,
      "step": 3700
    },
    {
      "epoch": 19.487179487179485,
      "grad_norm": 44.06697082519531,
      "learning_rate": 9.620900000000001e-06,
      "loss": 1.4057,
      "step": 3800
    },
    {
      "epoch": 19.487179487179485,
      "eval_loss": 1.4841217994689941,
      "eval_runtime": 36.3382,
      "eval_samples_per_second": 10.815,
      "eval_steps_per_second": 1.376,
      "step": 3800
    },
    {
      "epoch": 20.0,
      "grad_norm": 53.86429214477539,
      "learning_rate": 9.610900000000001e-06,
      "loss": 1.4302,
      "step": 3900
    },
    {
      "epoch": 20.0,
      "eval_loss": 1.477772831916809,
      "eval_runtime": 36.1794,
      "eval_samples_per_second": 10.863,
      "eval_steps_per_second": 1.382,
      "step": 3900
    },
    {
      "epoch": 20.51282051282051,
      "grad_norm": 80.95457458496094,
      "learning_rate": 9.600900000000002e-06,
      "loss": 1.4076,
      "step": 4000
    },
    {
      "epoch": 20.51282051282051,
      "eval_loss": 1.4769134521484375,
      "eval_runtime": 36.4725,
      "eval_samples_per_second": 10.775,
      "eval_steps_per_second": 1.371,
      "step": 4000
    }
  ],
  "logging_steps": 100,
  "max_steps": 100000,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 513,
  "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": 1.528972664832e+19,
  "train_batch_size": 8,
  "trial_name": null,
  "trial_params": null
}