{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 2.0,
  "eval_steps": 500,
  "global_step": 1408,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.07102272727272728,
      "grad_norm": 21.179128646850586,
      "learning_rate": 1.953598484848485e-05,
      "loss": 1.104847640991211,
      "step": 50
    },
    {
      "epoch": 0.14204545454545456,
      "grad_norm": 8.350769996643066,
      "learning_rate": 1.9062500000000003e-05,
      "loss": 0.7109419250488281,
      "step": 100
    },
    {
      "epoch": 0.21306818181818182,
      "grad_norm": 10.24179744720459,
      "learning_rate": 1.8589015151515152e-05,
      "loss": 0.646075439453125,
      "step": 150
    },
    {
      "epoch": 0.2840909090909091,
      "grad_norm": 7.701420783996582,
      "learning_rate": 1.8115530303030304e-05,
      "loss": 0.5862005233764649,
      "step": 200
    },
    {
      "epoch": 0.35511363636363635,
      "grad_norm": 6.683117389678955,
      "learning_rate": 1.7642045454545457e-05,
      "loss": 0.5390690994262696,
      "step": 250
    },
    {
      "epoch": 0.42613636363636365,
      "grad_norm": 12.960431098937988,
      "learning_rate": 1.7168560606060606e-05,
      "loss": 0.5759765625,
      "step": 300
    },
    {
      "epoch": 0.4971590909090909,
      "grad_norm": 7.7473015785217285,
      "learning_rate": 1.669507575757576e-05,
      "loss": 0.5988656616210938,
      "step": 350
    },
    {
      "epoch": 0.5681818181818182,
      "grad_norm": 9.293025970458984,
      "learning_rate": 1.622159090909091e-05,
      "loss": 0.5481339263916015,
      "step": 400
    },
    {
      "epoch": 0.6392045454545454,
      "grad_norm": 6.079025745391846,
      "learning_rate": 1.574810606060606e-05,
      "loss": 0.4792301940917969,
      "step": 450
    },
    {
      "epoch": 0.7102272727272727,
      "grad_norm": 7.443867206573486,
      "learning_rate": 1.5274621212121213e-05,
      "loss": 0.4998136901855469,
      "step": 500
    },
    {
      "epoch": 0.78125,
      "grad_norm": 17.248519897460938,
      "learning_rate": 1.4801136363636365e-05,
      "loss": 0.5037076187133789,
      "step": 550
    },
    {
      "epoch": 0.8522727272727273,
      "grad_norm": 9.058268547058105,
      "learning_rate": 1.4327651515151518e-05,
      "loss": 0.5227883911132812,
      "step": 600
    },
    {
      "epoch": 0.9232954545454546,
      "grad_norm": 5.971234321594238,
      "learning_rate": 1.3854166666666669e-05,
      "loss": 0.5120198059082032,
      "step": 650
    },
    {
      "epoch": 0.9943181818181818,
      "grad_norm": 11.57950210571289,
      "learning_rate": 1.338068181818182e-05,
      "loss": 0.5006477737426758,
      "step": 700
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.9088,
      "eval_loss": 0.4469251036643982,
      "eval_runtime": 10.1253,
      "eval_samples_per_second": 246.906,
      "eval_steps_per_second": 1.975,
      "step": 704
    },
    {
      "epoch": 1.0653409090909092,
      "grad_norm": 15.682221412658691,
      "learning_rate": 1.2907196969696972e-05,
      "loss": 0.36200748443603514,
      "step": 750
    },
    {
      "epoch": 1.1363636363636362,
      "grad_norm": 9.524371147155762,
      "learning_rate": 1.2433712121212123e-05,
      "loss": 0.37806861877441406,
      "step": 800
    },
    {
      "epoch": 1.2073863636363638,
      "grad_norm": 6.594772815704346,
      "learning_rate": 1.1960227272727274e-05,
      "loss": 0.3531819534301758,
      "step": 850
    },
    {
      "epoch": 1.2784090909090908,
      "grad_norm": 22.462160110473633,
      "learning_rate": 1.1486742424242424e-05,
      "loss": 0.310184326171875,
      "step": 900
    },
    {
      "epoch": 1.3494318181818181,
      "grad_norm": 2.5592997074127197,
      "learning_rate": 1.1013257575757577e-05,
      "loss": 0.3397782516479492,
      "step": 950
    },
    {
      "epoch": 1.4204545454545454,
      "grad_norm": 2.6996567249298096,
      "learning_rate": 1.0539772727272728e-05,
      "loss": 0.36845054626464846,
      "step": 1000
    },
    {
      "epoch": 1.4914772727272727,
      "grad_norm": 8.247840881347656,
      "learning_rate": 1.0066287878787879e-05,
      "loss": 0.33971687316894533,
      "step": 1050
    },
    {
      "epoch": 1.5625,
      "grad_norm": 6.193925380706787,
      "learning_rate": 9.592803030303031e-06,
      "loss": 0.36116703033447267,
      "step": 1100
    },
    {
      "epoch": 1.6335227272727273,
      "grad_norm": 11.514214515686035,
      "learning_rate": 9.119318181818184e-06,
      "loss": 0.38261314392089846,
      "step": 1150
    },
    {
      "epoch": 1.7045454545454546,
      "grad_norm": 12.636307716369629,
      "learning_rate": 8.645833333333335e-06,
      "loss": 0.3271693420410156,
      "step": 1200
    },
    {
      "epoch": 1.7755681818181817,
      "grad_norm": 13.377233505249023,
      "learning_rate": 8.172348484848485e-06,
      "loss": 0.3878217315673828,
      "step": 1250
    },
    {
      "epoch": 1.8465909090909092,
      "grad_norm": 5.174794673919678,
      "learning_rate": 7.698863636363638e-06,
      "loss": 0.3461064147949219,
      "step": 1300
    },
    {
      "epoch": 1.9176136363636362,
      "grad_norm": 6.655035972595215,
      "learning_rate": 7.225378787878789e-06,
      "loss": 0.35596923828125,
      "step": 1350
    },
    {
      "epoch": 1.9886363636363638,
      "grad_norm": 15.530329704284668,
      "learning_rate": 6.75189393939394e-06,
      "loss": 0.33932086944580075,
      "step": 1400
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.91,
      "eval_loss": 0.46745410561561584,
      "eval_runtime": 10.0775,
      "eval_samples_per_second": 248.078,
      "eval_steps_per_second": 1.985,
      "step": 1408
    }
  ],
  "logging_steps": 50,
  "max_steps": 2112,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 3,
  "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": 2980516469760000.0,
  "train_batch_size": 32,
  "trial_name": null,
  "trial_params": null
}