{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 1.1998400213304894,
  "eval_steps": 500,
  "global_step": 4500,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.03,
      "grad_norm": 0.5563874321368991,
      "learning_rate": 4.000000000000001e-06,
      "loss": 1.58,
      "step": 100
    },
    {
      "epoch": 0.05,
      "grad_norm": 0.4957053999852371,
      "learning_rate": 8.000000000000001e-06,
      "loss": 1.4463,
      "step": 200
    },
    {
      "epoch": 0.08,
      "grad_norm": 0.589756030703148,
      "learning_rate": 1.2e-05,
      "loss": 1.3542,
      "step": 300
    },
    {
      "epoch": 0.11,
      "grad_norm": 0.6420302678887735,
      "learning_rate": 1.6000000000000003e-05,
      "loss": 1.3331,
      "step": 400
    },
    {
      "epoch": 0.13,
      "grad_norm": 0.7649283840590947,
      "learning_rate": 2e-05,
      "loss": 1.3253,
      "step": 500
    },
    {
      "epoch": 0.13,
      "eval_loss": 1.3249897956848145,
      "eval_runtime": 2000.6322,
      "eval_samples_per_second": 6.665,
      "eval_steps_per_second": 0.833,
      "step": 500
    },
    {
      "epoch": 0.16,
      "grad_norm": 0.7507878377667085,
      "learning_rate": 1.9714285714285718e-05,
      "loss": 1.3221,
      "step": 600
    },
    {
      "epoch": 0.19,
      "grad_norm": 0.6176603895840452,
      "learning_rate": 1.942857142857143e-05,
      "loss": 1.3126,
      "step": 700
    },
    {
      "epoch": 0.21,
      "grad_norm": 0.7327398805078954,
      "learning_rate": 1.9142857142857146e-05,
      "loss": 1.3055,
      "step": 800
    },
    {
      "epoch": 0.24,
      "grad_norm": 0.6839292302496646,
      "learning_rate": 1.885714285714286e-05,
      "loss": 1.3093,
      "step": 900
    },
    {
      "epoch": 0.27,
      "grad_norm": 0.607074650571682,
      "learning_rate": 1.8571428571428575e-05,
      "loss": 1.2985,
      "step": 1000
    },
    {
      "epoch": 0.27,
      "eval_loss": 1.2995541095733643,
      "eval_runtime": 1951.5754,
      "eval_samples_per_second": 6.833,
      "eval_steps_per_second": 0.854,
      "step": 1000
    },
    {
      "epoch": 0.29,
      "grad_norm": 0.74727636749315,
      "learning_rate": 1.8285714285714288e-05,
      "loss": 1.3017,
      "step": 1100
    },
    {
      "epoch": 0.32,
      "grad_norm": 0.6311499573020434,
      "learning_rate": 1.8e-05,
      "loss": 1.2896,
      "step": 1200
    },
    {
      "epoch": 0.35,
      "grad_norm": 0.5915205156004358,
      "learning_rate": 1.7714285714285717e-05,
      "loss": 1.2834,
      "step": 1300
    },
    {
      "epoch": 0.37,
      "grad_norm": 0.5946814358633083,
      "learning_rate": 1.742857142857143e-05,
      "loss": 1.3073,
      "step": 1400
    },
    {
      "epoch": 0.4,
      "grad_norm": 0.6097257778288048,
      "learning_rate": 1.7142857142857142e-05,
      "loss": 1.2843,
      "step": 1500
    },
    {
      "epoch": 0.4,
      "eval_loss": 1.2924432754516602,
      "eval_runtime": 1987.1893,
      "eval_samples_per_second": 6.71,
      "eval_steps_per_second": 0.839,
      "step": 1500
    },
    {
      "epoch": 0.43,
      "grad_norm": 0.596713525589904,
      "learning_rate": 1.6857142857142858e-05,
      "loss": 1.2748,
      "step": 1600
    },
    {
      "epoch": 0.45,
      "grad_norm": 0.6872784300870175,
      "learning_rate": 1.6571428571428574e-05,
      "loss": 1.2971,
      "step": 1700
    },
    {
      "epoch": 0.48,
      "grad_norm": 0.5519473130229932,
      "learning_rate": 1.6285714285714287e-05,
      "loss": 1.2746,
      "step": 1800
    },
    {
      "epoch": 0.51,
      "grad_norm": 0.6633345934082727,
      "learning_rate": 1.6000000000000003e-05,
      "loss": 1.2906,
      "step": 1900
    },
    {
      "epoch": 0.53,
      "grad_norm": 0.6476833370052498,
      "learning_rate": 1.5714285714285715e-05,
      "loss": 1.2813,
      "step": 2000
    },
    {
      "epoch": 0.53,
      "eval_loss": 1.2873681783676147,
      "eval_runtime": 1954.0241,
      "eval_samples_per_second": 6.824,
      "eval_steps_per_second": 0.853,
      "step": 2000
    },
    {
      "epoch": 0.56,
      "grad_norm": 0.6831512981144761,
      "learning_rate": 1.542857142857143e-05,
      "loss": 1.2973,
      "step": 2100
    },
    {
      "epoch": 0.59,
      "grad_norm": 0.6111369458671693,
      "learning_rate": 1.5142857142857144e-05,
      "loss": 1.2847,
      "step": 2200
    },
    {
      "epoch": 0.61,
      "grad_norm": 0.6468631061437224,
      "learning_rate": 1.4857142857142858e-05,
      "loss": 1.2754,
      "step": 2300
    },
    {
      "epoch": 0.64,
      "grad_norm": 0.6799634395265591,
      "learning_rate": 1.4571428571428573e-05,
      "loss": 1.2857,
      "step": 2400
    },
    {
      "epoch": 0.67,
      "grad_norm": 0.7108542425010391,
      "learning_rate": 1.4285714285714287e-05,
      "loss": 1.289,
      "step": 2500
    },
    {
      "epoch": 0.67,
      "eval_loss": 1.283585786819458,
      "eval_runtime": 1870.2159,
      "eval_samples_per_second": 7.13,
      "eval_steps_per_second": 0.891,
      "step": 2500
    },
    {
      "epoch": 0.69,
      "grad_norm": 0.7253629154301096,
      "learning_rate": 1.4e-05,
      "loss": 1.2793,
      "step": 2600
    },
    {
      "epoch": 0.72,
      "grad_norm": 0.614991663938332,
      "learning_rate": 1.3714285714285716e-05,
      "loss": 1.2767,
      "step": 2700
    },
    {
      "epoch": 0.75,
      "grad_norm": 0.5699407855446496,
      "learning_rate": 1.3428571428571429e-05,
      "loss": 1.2719,
      "step": 2800
    },
    {
      "epoch": 0.77,
      "grad_norm": 0.763568295530126,
      "learning_rate": 1.3142857142857145e-05,
      "loss": 1.2733,
      "step": 2900
    },
    {
      "epoch": 0.8,
      "grad_norm": 0.6538876064837678,
      "learning_rate": 1.2857142857142859e-05,
      "loss": 1.2822,
      "step": 3000
    },
    {
      "epoch": 0.8,
      "eval_loss": 1.2801251411437988,
      "eval_runtime": 1889.0286,
      "eval_samples_per_second": 7.059,
      "eval_steps_per_second": 0.882,
      "step": 3000
    },
    {
      "epoch": 0.83,
      "grad_norm": 0.6413250796291871,
      "learning_rate": 1.2571428571428572e-05,
      "loss": 1.2709,
      "step": 3100
    },
    {
      "epoch": 0.85,
      "grad_norm": 0.7125599551490435,
      "learning_rate": 1.2285714285714288e-05,
      "loss": 1.2829,
      "step": 3200
    },
    {
      "epoch": 0.88,
      "grad_norm": 0.6792031678140497,
      "learning_rate": 1.2e-05,
      "loss": 1.2728,
      "step": 3300
    },
    {
      "epoch": 0.91,
      "grad_norm": 0.6989586202409576,
      "learning_rate": 1.1714285714285716e-05,
      "loss": 1.2835,
      "step": 3400
    },
    {
      "epoch": 0.93,
      "grad_norm": 0.7569471708203911,
      "learning_rate": 1.1428571428571429e-05,
      "loss": 1.2793,
      "step": 3500
    },
    {
      "epoch": 0.93,
      "eval_loss": 1.2773550748825073,
      "eval_runtime": 1896.8518,
      "eval_samples_per_second": 7.03,
      "eval_steps_per_second": 0.879,
      "step": 3500
    },
    {
      "epoch": 0.96,
      "grad_norm": 0.6242374298012554,
      "learning_rate": 1.1142857142857143e-05,
      "loss": 1.2805,
      "step": 3600
    },
    {
      "epoch": 0.99,
      "grad_norm": 0.6585613373985719,
      "learning_rate": 1.0857142857142858e-05,
      "loss": 1.265,
      "step": 3700
    },
    {
      "epoch": 1.01,
      "grad_norm": 0.7452400076297672,
      "learning_rate": 1.0571428571428572e-05,
      "loss": 1.2589,
      "step": 3800
    },
    {
      "epoch": 1.04,
      "grad_norm": 0.7861020289489161,
      "learning_rate": 1.0285714285714285e-05,
      "loss": 1.2616,
      "step": 3900
    },
    {
      "epoch": 1.07,
      "grad_norm": 0.7757094950303476,
      "learning_rate": 1e-05,
      "loss": 1.2721,
      "step": 4000
    },
    {
      "epoch": 1.07,
      "eval_loss": 1.2751243114471436,
      "eval_runtime": 1895.5553,
      "eval_samples_per_second": 7.035,
      "eval_steps_per_second": 0.879,
      "step": 4000
    },
    {
      "epoch": 1.09,
      "grad_norm": 0.6999523427612176,
      "learning_rate": 9.714285714285715e-06,
      "loss": 1.2649,
      "step": 4100
    },
    {
      "epoch": 1.12,
      "grad_norm": 0.6528269639884913,
      "learning_rate": 9.42857142857143e-06,
      "loss": 1.2675,
      "step": 4200
    },
    {
      "epoch": 1.15,
      "grad_norm": 0.8019204201326198,
      "learning_rate": 9.142857142857144e-06,
      "loss": 1.2775,
      "step": 4300
    },
    {
      "epoch": 1.17,
      "grad_norm": 0.8453041527524593,
      "learning_rate": 8.857142857142858e-06,
      "loss": 1.2682,
      "step": 4400
    },
    {
      "epoch": 1.2,
      "grad_norm": 0.6858153075780669,
      "learning_rate": 8.571428571428571e-06,
      "loss": 1.2753,
      "step": 4500
    },
    {
      "epoch": 1.2,
      "eval_loss": 1.2731624841690063,
      "eval_runtime": 2031.5498,
      "eval_samples_per_second": 6.564,
      "eval_steps_per_second": 0.821,
      "step": 4500
    }
  ],
  "logging_steps": 100,
  "max_steps": 7500,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 2,
  "save_steps": 500,
  "total_flos": 6.185622662887244e+18,
  "train_batch_size": 8,
  "trial_name": null,
  "trial_params": null
}