{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 1.3118706566444476,
  "eval_steps": 3526,
  "global_step": 18500,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.03545596369309318,
      "grad_norm": 1.6790207624435425,
      "learning_rate": 0.0001964544036306907,
      "loss": 4.1281,
      "step": 500
    },
    {
      "epoch": 0.07091192738618636,
      "grad_norm": 1.736142635345459,
      "learning_rate": 0.00019290880726138138,
      "loss": 3.4072,
      "step": 1000
    },
    {
      "epoch": 0.10636789107927953,
      "grad_norm": 1.8541169166564941,
      "learning_rate": 0.00018936321089207204,
      "loss": 3.1885,
      "step": 1500
    },
    {
      "epoch": 0.14182385477237272,
      "grad_norm": 2.1542677879333496,
      "learning_rate": 0.00018581761452276274,
      "loss": 3.1089,
      "step": 2000
    },
    {
      "epoch": 0.1772798184654659,
      "grad_norm": 1.8087749481201172,
      "learning_rate": 0.00018227201815345344,
      "loss": 3.0863,
      "step": 2500
    },
    {
      "epoch": 0.21273578215855907,
      "grad_norm": 2.463923692703247,
      "learning_rate": 0.0001787264217841441,
      "loss": 3.0656,
      "step": 3000
    },
    {
      "epoch": 0.24819174585165224,
      "grad_norm": 2.494748115539551,
      "learning_rate": 0.00017518082541483478,
      "loss": 3.0086,
      "step": 3500
    },
    {
      "epoch": 0.2500354559636931,
      "eval_loss": 2.7105393409729004,
      "eval_runtime": 360.9098,
      "eval_samples_per_second": 9.77,
      "eval_steps_per_second": 9.77,
      "step": 3526
    },
    {
      "epoch": 0.28364770954474544,
      "grad_norm": 1.3893351554870605,
      "learning_rate": 0.00017163522904552547,
      "loss": 2.9823,
      "step": 4000
    },
    {
      "epoch": 0.3191036732378386,
      "grad_norm": 1.878633975982666,
      "learning_rate": 0.00016808963267621617,
      "loss": 2.8684,
      "step": 4500
    },
    {
      "epoch": 0.3545596369309318,
      "grad_norm": 3.777602434158325,
      "learning_rate": 0.0001645440363069068,
      "loss": 2.802,
      "step": 5000
    },
    {
      "epoch": 0.39001560062402496,
      "grad_norm": 1.610077977180481,
      "learning_rate": 0.0001609984399375975,
      "loss": 2.894,
      "step": 5500
    },
    {
      "epoch": 0.42547156431711813,
      "grad_norm": 2.231553316116333,
      "learning_rate": 0.0001574528435682882,
      "loss": 2.8238,
      "step": 6000
    },
    {
      "epoch": 0.4609275280102113,
      "grad_norm": 3.5686066150665283,
      "learning_rate": 0.00015390724719897887,
      "loss": 2.8877,
      "step": 6500
    },
    {
      "epoch": 0.4963834917033045,
      "grad_norm": 1.6852627992630005,
      "learning_rate": 0.00015036165082966954,
      "loss": 2.8127,
      "step": 7000
    },
    {
      "epoch": 0.5000709119273862,
      "eval_loss": 2.5279786586761475,
      "eval_runtime": 360.7362,
      "eval_samples_per_second": 9.774,
      "eval_steps_per_second": 9.774,
      "step": 7052
    },
    {
      "epoch": 0.5318394553963977,
      "grad_norm": 2.0495705604553223,
      "learning_rate": 0.00014681605446036024,
      "loss": 2.7904,
      "step": 7500
    },
    {
      "epoch": 0.5672954190894909,
      "grad_norm": 1.7553592920303345,
      "learning_rate": 0.00014327045809105093,
      "loss": 2.7686,
      "step": 8000
    },
    {
      "epoch": 0.602751382782584,
      "grad_norm": 1.814765453338623,
      "learning_rate": 0.0001397248617217416,
      "loss": 2.7501,
      "step": 8500
    },
    {
      "epoch": 0.6382073464756772,
      "grad_norm": 2.8568713665008545,
      "learning_rate": 0.00013617926535243227,
      "loss": 2.7077,
      "step": 9000
    },
    {
      "epoch": 0.6736633101687703,
      "grad_norm": 2.317953109741211,
      "learning_rate": 0.00013263366898312297,
      "loss": 2.7284,
      "step": 9500
    },
    {
      "epoch": 0.7091192738618636,
      "grad_norm": 2.4004650115966797,
      "learning_rate": 0.00012908807261381367,
      "loss": 2.6601,
      "step": 10000
    },
    {
      "epoch": 0.7445752375549567,
      "grad_norm": 2.3228759765625,
      "learning_rate": 0.00012554247624450433,
      "loss": 2.717,
      "step": 10500
    },
    {
      "epoch": 0.7501063678910793,
      "eval_loss": 2.42594575881958,
      "eval_runtime": 360.6749,
      "eval_samples_per_second": 9.776,
      "eval_steps_per_second": 9.776,
      "step": 10578
    },
    {
      "epoch": 0.7800312012480499,
      "grad_norm": 2.101666212081909,
      "learning_rate": 0.00012199687987519502,
      "loss": 2.692,
      "step": 11000
    },
    {
      "epoch": 0.8154871649411432,
      "grad_norm": 2.441512107849121,
      "learning_rate": 0.0001184512835058857,
      "loss": 2.6627,
      "step": 11500
    },
    {
      "epoch": 0.8509431286342363,
      "grad_norm": 1.934930443763733,
      "learning_rate": 0.00011490568713657637,
      "loss": 2.6634,
      "step": 12000
    },
    {
      "epoch": 0.8863990923273295,
      "grad_norm": 2.1375555992126465,
      "learning_rate": 0.00011136009076726705,
      "loss": 2.6631,
      "step": 12500
    },
    {
      "epoch": 0.9218550560204226,
      "grad_norm": 2.178725242614746,
      "learning_rate": 0.00010781449439795775,
      "loss": 2.6373,
      "step": 13000
    },
    {
      "epoch": 0.9573110197135158,
      "grad_norm": 1.6436994075775146,
      "learning_rate": 0.00010426889802864843,
      "loss": 2.5462,
      "step": 13500
    },
    {
      "epoch": 0.992766983406609,
      "grad_norm": 3.6108076572418213,
      "learning_rate": 0.0001007233016593391,
      "loss": 2.6291,
      "step": 14000
    },
    {
      "epoch": 1.0001418238547723,
      "eval_loss": 2.368666410446167,
      "eval_runtime": 360.6633,
      "eval_samples_per_second": 9.776,
      "eval_steps_per_second": 9.776,
      "step": 14104
    },
    {
      "epoch": 1.0282229470997022,
      "grad_norm": 2.3566112518310547,
      "learning_rate": 9.717770529002978e-05,
      "loss": 2.678,
      "step": 14500
    },
    {
      "epoch": 1.0636789107927953,
      "grad_norm": 2.204242467880249,
      "learning_rate": 9.363210892072048e-05,
      "loss": 2.587,
      "step": 15000
    },
    {
      "epoch": 1.0991348744858884,
      "grad_norm": 1.9366542100906372,
      "learning_rate": 9.008651255141115e-05,
      "loss": 2.6107,
      "step": 15500
    },
    {
      "epoch": 1.1345908381789818,
      "grad_norm": 2.088714838027954,
      "learning_rate": 8.654091618210185e-05,
      "loss": 2.5692,
      "step": 16000
    },
    {
      "epoch": 1.1700468018720749,
      "grad_norm": 2.3588242530822754,
      "learning_rate": 8.299531981279251e-05,
      "loss": 2.5354,
      "step": 16500
    },
    {
      "epoch": 1.205502765565168,
      "grad_norm": 2.398696184158325,
      "learning_rate": 7.94497234434832e-05,
      "loss": 2.5996,
      "step": 17000
    },
    {
      "epoch": 1.2409587292582613,
      "grad_norm": 1.8444238901138306,
      "learning_rate": 7.590412707417388e-05,
      "loss": 2.6186,
      "step": 17500
    },
    {
      "epoch": 1.2501772798184656,
      "eval_loss": 2.3247604370117188,
      "eval_runtime": 360.8279,
      "eval_samples_per_second": 9.772,
      "eval_steps_per_second": 9.772,
      "step": 17630
    },
    {
      "epoch": 1.2764146929513545,
      "grad_norm": 2.0656793117523193,
      "learning_rate": 7.235853070486456e-05,
      "loss": 2.5647,
      "step": 18000
    },
    {
      "epoch": 1.3118706566444476,
      "grad_norm": 2.8596479892730713,
      "learning_rate": 6.881293433555525e-05,
      "loss": 2.555,
      "step": 18500
    }
  ],
  "logging_steps": 500,
  "max_steps": 28204,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 2,
  "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": 8.0374916972544e+16,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}