{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 3.0,
  "eval_steps": 500,
  "global_step": 183,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "entropy": 0.34310978651046753,
      "epoch": 0.016597510373443983,
      "grad_norm": 30.463375091552734,
      "learning_rate": 0.0,
      "loss": 2.5902838706970215,
      "mean_token_accuracy": 0.7455968856811523,
      "num_tokens": 8192.0,
      "step": 1
    },
    {
      "entropy": 0.30017997283074593,
      "epoch": 0.16597510373443983,
      "grad_norm": 7.794386863708496,
      "learning_rate": 4.5e-05,
      "loss": 2.039700402153863,
      "mean_token_accuracy": 0.7990324033631219,
      "num_tokens": 81920.0,
      "step": 10
    },
    {
      "entropy": 0.15870278999209403,
      "epoch": 0.33195020746887965,
      "grad_norm": 1.6451208591461182,
      "learning_rate": 9.5e-05,
      "loss": 0.4637583255767822,
      "mean_token_accuracy": 0.940655580163002,
      "num_tokens": 163840.0,
      "step": 20
    },
    {
      "entropy": 0.08129047341644764,
      "epoch": 0.4979253112033195,
      "grad_norm": 0.6042754650115967,
      "learning_rate": 9.924965614670629e-05,
      "loss": 0.051254814863204955,
      "mean_token_accuracy": 0.9927592992782592,
      "num_tokens": 245760.0,
      "step": 30
    },
    {
      "entropy": 0.023062964738346636,
      "epoch": 0.6639004149377593,
      "grad_norm": 0.07445944845676422,
      "learning_rate": 9.66847747759402e-05,
      "loss": 0.006664524227380753,
      "mean_token_accuracy": 0.9981164455413818,
      "num_tokens": 327680.0,
      "step": 40
    },
    {
      "entropy": 0.007560247508808971,
      "epoch": 0.8298755186721992,
      "grad_norm": 0.02006595768034458,
      "learning_rate": 9.239105265402525e-05,
      "loss": 0.0012541665695607663,
      "mean_token_accuracy": 1.0,
      "num_tokens": 409600.0,
      "step": 50
    },
    {
      "entropy": 0.0038804779178462923,
      "epoch": 0.995850622406639,
      "grad_norm": 0.006827245932072401,
      "learning_rate": 8.65274958340934e-05,
      "loss": 0.0004053693264722824,
      "mean_token_accuracy": 1.0,
      "num_tokens": 491520.0,
      "step": 60
    },
    {
      "entropy": 0.0023303336237330695,
      "epoch": 1.1493775933609958,
      "grad_norm": 0.004071525298058987,
      "learning_rate": 7.931124482984802e-05,
      "loss": 0.00023144541773945093,
      "mean_token_accuracy": 1.0,
      "num_tokens": 565760.0,
      "step": 70
    },
    {
      "entropy": 0.0017596775403944775,
      "epoch": 1.3153526970954357,
      "grad_norm": 0.0030074911192059517,
      "learning_rate": 7.10095334207501e-05,
      "loss": 0.00016952784499153496,
      "mean_token_accuracy": 1.0,
      "num_tokens": 647680.0,
      "step": 80
    },
    {
      "entropy": 0.0014692362194182352,
      "epoch": 1.4813278008298756,
      "grad_norm": 0.002476779278367758,
      "learning_rate": 6.192979239247243e-05,
      "loss": 0.00013985750265419484,
      "mean_token_accuracy": 1.0,
      "num_tokens": 729600.0,
      "step": 90
    },
    {
      "entropy": 0.0013017851480981335,
      "epoch": 1.6473029045643153,
      "grad_norm": 0.002169998362660408,
      "learning_rate": 5.240826469306187e-05,
      "loss": 0.00012233242159709334,
      "mean_token_accuracy": 1.0,
      "num_tokens": 811520.0,
      "step": 100
    },
    {
      "entropy": 0.0011935859685763717,
      "epoch": 1.8132780082987552,
      "grad_norm": 0.001988000003620982,
      "learning_rate": 4.27975536108268e-05,
      "loss": 0.00011085805017501116,
      "mean_token_accuracy": 1.0,
      "num_tokens": 893440.0,
      "step": 110
    },
    {
      "entropy": 0.0011260211758781224,
      "epoch": 1.979253112033195,
      "grad_norm": 0.0018395803635939956,
      "learning_rate": 3.345356509253959e-05,
      "loss": 0.00010404185159131884,
      "mean_token_accuracy": 1.0,
      "num_tokens": 975360.0,
      "step": 120
    },
    {
      "entropy": 0.001064722627602719,
      "epoch": 2.132780082987552,
      "grad_norm": 0.0017558144172653556,
      "learning_rate": 2.472232775687119e-05,
      "loss": 9.810116607695818e-05,
      "mean_token_accuracy": 1.0,
      "num_tokens": 1049600.0,
      "step": 130
    },
    {
      "entropy": 0.0010282685951096937,
      "epoch": 2.2987551867219915,
      "grad_norm": 0.0016462178900837898,
      "learning_rate": 1.692717868719195e-05,
      "loss": 9.387803729623556e-05,
      "mean_token_accuracy": 1.0,
      "num_tokens": 1131520.0,
      "step": 140
    },
    {
      "entropy": 0.001004051680502016,
      "epoch": 2.4647302904564317,
      "grad_norm": 0.00162224390078336,
      "learning_rate": 1.0356789542251938e-05,
      "loss": 9.128837846219539e-05,
      "mean_token_accuracy": 1.0,
      "num_tokens": 1213440.0,
      "step": 150
    },
    {
      "entropy": 0.0009880667523248122,
      "epoch": 2.6307053941908713,
      "grad_norm": 0.001601163879968226,
      "learning_rate": 5.25447640441834e-06,
      "loss": 8.967729518190026e-05,
      "mean_token_accuracy": 1.0,
      "num_tokens": 1295360.0,
      "step": 160
    },
    {
      "entropy": 0.0009832712908973916,
      "epoch": 2.796680497925311,
      "grad_norm": 0.0015979394083842635,
      "learning_rate": 1.8091892454998594e-06,
      "loss": 8.923084242269397e-05,
      "mean_token_accuracy": 1.0,
      "num_tokens": 1377280.0,
      "step": 170
    },
    {
      "entropy": 0.0009787017304915934,
      "epoch": 2.962655601659751,
      "grad_norm": 0.0015788950258865952,
      "learning_rate": 1.4851469022234e-07,
      "loss": 8.886267896741628e-05,
      "mean_token_accuracy": 1.0,
      "num_tokens": 1459200.0,
      "step": 180
    },
    {
      "entropy": 0.0009836054329449933,
      "epoch": 3.0,
      "mean_token_accuracy": 1.0,
      "num_tokens": 1476096.0,
      "step": 183,
      "total_flos": 922173333700608.0,
      "train_loss": 0.14314490778558084,
      "train_runtime": 252.6415,
      "train_samples_per_second": 11.411,
      "train_steps_per_second": 0.724
    }
  ],
  "logging_steps": 10,
  "max_steps": 183,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 3,
  "save_steps": 50,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 922173333700608.0,
  "train_batch_size": 4,
  "trial_name": null,
  "trial_params": null
}