{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 2.997384481255449,
  "eval_steps": 50,
  "global_step": 858,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.17436791630340018,
      "grad_norm": 0.31028223037719727,
      "learning_rate": 0.0002,
      "loss": 1.7522,
      "mean_token_accuracy": 0.6208396552503109,
      "step": 50
    },
    {
      "epoch": 0.17436791630340018,
      "eval_loss": 1.5662453174591064,
      "eval_mean_token_accuracy": 0.6466238599197537,
      "eval_runtime": 216.8159,
      "eval_samples_per_second": 9.404,
      "eval_steps_per_second": 1.176,
      "step": 50
    },
    {
      "epoch": 0.34873583260680036,
      "grad_norm": 0.20454441010951996,
      "learning_rate": 0.0002,
      "loss": 1.4597,
      "mean_token_accuracy": 0.6637434582412243,
      "step": 100
    },
    {
      "epoch": 0.34873583260680036,
      "eval_loss": 1.4656527042388916,
      "eval_mean_token_accuracy": 0.6612562953257093,
      "eval_runtime": 219.0107,
      "eval_samples_per_second": 9.31,
      "eval_steps_per_second": 1.164,
      "step": 100
    },
    {
      "epoch": 0.5231037489102005,
      "grad_norm": 0.13001485168933868,
      "learning_rate": 0.0002,
      "loss": 1.4281,
      "mean_token_accuracy": 0.6697324779629708,
      "step": 150
    },
    {
      "epoch": 0.5231037489102005,
      "eval_loss": 1.4386416673660278,
      "eval_mean_token_accuracy": 0.6682294864280551,
      "eval_runtime": 218.6278,
      "eval_samples_per_second": 9.326,
      "eval_steps_per_second": 1.166,
      "step": 150
    },
    {
      "epoch": 0.6974716652136007,
      "grad_norm": 0.2524537146091461,
      "learning_rate": 0.0002,
      "loss": 1.4169,
      "mean_token_accuracy": 0.6710615785419941,
      "step": 200
    },
    {
      "epoch": 0.6974716652136007,
      "eval_loss": 1.4218348264694214,
      "eval_mean_token_accuracy": 0.6715860612252179,
      "eval_runtime": 215.794,
      "eval_samples_per_second": 9.449,
      "eval_steps_per_second": 1.182,
      "step": 200
    },
    {
      "epoch": 0.8718395815170009,
      "grad_norm": 0.16508391499519348,
      "learning_rate": 0.0002,
      "loss": 1.398,
      "mean_token_accuracy": 0.674775720089674,
      "step": 250
    },
    {
      "epoch": 0.8718395815170009,
      "eval_loss": 1.4061472415924072,
      "eval_mean_token_accuracy": 0.6714201174530329,
      "eval_runtime": 217.972,
      "eval_samples_per_second": 9.354,
      "eval_steps_per_second": 1.17,
      "step": 250
    },
    {
      "epoch": 1.045335658238884,
      "grad_norm": 0.06276454031467438,
      "learning_rate": 0.0002,
      "loss": 1.4208,
      "mean_token_accuracy": 0.6656776975147688,
      "step": 300
    },
    {
      "epoch": 1.045335658238884,
      "eval_loss": 1.3979426622390747,
      "eval_mean_token_accuracy": 0.6722868853924321,
      "eval_runtime": 221.6256,
      "eval_samples_per_second": 9.2,
      "eval_steps_per_second": 1.151,
      "step": 300
    },
    {
      "epoch": 1.2197035745422842,
      "grad_norm": 0.0702725201845169,
      "learning_rate": 0.0002,
      "loss": 1.3814,
      "mean_token_accuracy": 0.675048353523016,
      "step": 350
    },
    {
      "epoch": 1.2197035745422842,
      "eval_loss": 1.3879132270812988,
      "eval_mean_token_accuracy": 0.6748333828122008,
      "eval_runtime": 211.3323,
      "eval_samples_per_second": 9.648,
      "eval_steps_per_second": 1.207,
      "step": 350
    },
    {
      "epoch": 1.3940714908456844,
      "grad_norm": 0.07022608816623688,
      "learning_rate": 0.0002,
      "loss": 1.376,
      "mean_token_accuracy": 0.6765765248239041,
      "step": 400
    },
    {
      "epoch": 1.3940714908456844,
      "eval_loss": 1.3835749626159668,
      "eval_mean_token_accuracy": 0.6750268143766066,
      "eval_runtime": 211.0785,
      "eval_samples_per_second": 9.66,
      "eval_steps_per_second": 1.208,
      "step": 400
    },
    {
      "epoch": 1.5684394071490846,
      "grad_norm": 0.07153413444757462,
      "learning_rate": 0.0002,
      "loss": 1.3735,
      "mean_token_accuracy": 0.6764040923118592,
      "step": 450
    },
    {
      "epoch": 1.5684394071490846,
      "eval_loss": 1.3785712718963623,
      "eval_mean_token_accuracy": 0.6757427355822395,
      "eval_runtime": 209.9753,
      "eval_samples_per_second": 9.711,
      "eval_steps_per_second": 1.214,
      "step": 450
    },
    {
      "epoch": 1.7462946817785527,
      "grad_norm": 0.0622311495244503,
      "learning_rate": 0.0002,
      "loss": 1.3769,
      "mean_token_accuracy": 0.6763439090549945,
      "step": 500
    },
    {
      "epoch": 1.7462946817785527,
      "eval_loss": 1.3926106691360474,
      "eval_mean_token_accuracy": 0.6742508411407471,
      "eval_runtime": 211.3856,
      "eval_samples_per_second": 9.646,
      "eval_steps_per_second": 1.206,
      "step": 500
    },
    {
      "epoch": 1.920662598081953,
      "grad_norm": 0.07205487042665482,
      "learning_rate": 0.0002,
      "loss": 1.3622,
      "mean_token_accuracy": 0.6805365033447742,
      "step": 550
    },
    {
      "epoch": 1.920662598081953,
      "eval_loss": 1.3798006772994995,
      "eval_mean_token_accuracy": 0.6765407457071192,
      "eval_runtime": 211.4314,
      "eval_samples_per_second": 9.644,
      "eval_steps_per_second": 1.206,
      "step": 550
    },
    {
      "epoch": 2.0976460331299043,
      "grad_norm": 0.05617905408143997,
      "learning_rate": 0.0002,
      "loss": 1.3978,
      "mean_token_accuracy": 0.6721828516774577,
      "step": 600
    },
    {
      "epoch": 2.0976460331299043,
      "eval_loss": 1.376914620399475,
      "eval_mean_token_accuracy": 0.6766514093268151,
      "eval_runtime": 211.4284,
      "eval_samples_per_second": 9.644,
      "eval_steps_per_second": 1.206,
      "step": 600
    },
    {
      "epoch": 2.2720139494333043,
      "grad_norm": 0.06081470474600792,
      "learning_rate": 0.0002,
      "loss": 1.3542,
      "mean_token_accuracy": 0.6793570870161056,
      "step": 650
    },
    {
      "epoch": 2.2720139494333043,
      "eval_loss": 1.3769170045852661,
      "eval_mean_token_accuracy": 0.6753617045926112,
      "eval_runtime": 211.9023,
      "eval_samples_per_second": 9.622,
      "eval_steps_per_second": 1.203,
      "step": 650
    },
    {
      "epoch": 2.4463818657367042,
      "grad_norm": 0.05523712560534477,
      "learning_rate": 0.0002,
      "loss": 1.35,
      "mean_token_accuracy": 0.6805331887304783,
      "step": 700
    },
    {
      "epoch": 2.4463818657367042,
      "eval_loss": 1.3597146272659302,
      "eval_mean_token_accuracy": 0.6796985892688527,
      "eval_runtime": 211.7167,
      "eval_samples_per_second": 9.631,
      "eval_steps_per_second": 1.204,
      "step": 700
    },
    {
      "epoch": 2.6207497820401047,
      "grad_norm": 0.05956251546740532,
      "learning_rate": 0.0002,
      "loss": 1.3365,
      "mean_token_accuracy": 0.6847609375417233,
      "step": 750
    },
    {
      "epoch": 2.6207497820401047,
      "eval_loss": 1.353187084197998,
      "eval_mean_token_accuracy": 0.6810896275090236,
      "eval_runtime": 223.086,
      "eval_samples_per_second": 9.14,
      "eval_steps_per_second": 1.143,
      "step": 750
    },
    {
      "epoch": 2.7951176983435047,
      "grad_norm": 0.06622714549303055,
      "learning_rate": 0.0002,
      "loss": 1.3297,
      "mean_token_accuracy": 0.6856365065276623,
      "step": 800
    },
    {
      "epoch": 2.7951176983435047,
      "eval_loss": 1.348921537399292,
      "eval_mean_token_accuracy": 0.6817485306777206,
      "eval_runtime": 216.8187,
      "eval_samples_per_second": 9.404,
      "eval_steps_per_second": 1.176,
      "step": 800
    },
    {
      "epoch": 2.969485614646905,
      "grad_norm": 0.06617969274520874,
      "learning_rate": 0.0002,
      "loss": 1.3277,
      "mean_token_accuracy": 0.6844626396894455,
      "step": 850
    },
    {
      "epoch": 2.969485614646905,
      "eval_loss": 1.3416334390640259,
      "eval_mean_token_accuracy": 0.6825305232814715,
      "eval_runtime": 218.0494,
      "eval_samples_per_second": 9.351,
      "eval_steps_per_second": 1.169,
      "step": 850
    }
  ],
  "logging_steps": 50,
  "max_steps": 858,
  "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": 1.2240557351095665e+18,
  "train_batch_size": 8,
  "trial_name": null,
  "trial_params": null
}