{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 3.0,
  "eval_steps": 500,
  "global_step": 16464,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.09110787172011661,
      "grad_norm": 0.5542775988578796,
      "learning_rate": 4.848153547133139e-05,
      "loss": 7.6369,
      "step": 500
    },
    {
      "epoch": 0.18221574344023322,
      "grad_norm": 0.6718146800994873,
      "learning_rate": 4.696307094266278e-05,
      "loss": 6.0229,
      "step": 1000
    },
    {
      "epoch": 0.27332361516034986,
      "grad_norm": 0.6730484366416931,
      "learning_rate": 4.5444606413994175e-05,
      "loss": 5.5062,
      "step": 1500
    },
    {
      "epoch": 0.36443148688046645,
      "grad_norm": 0.7009643912315369,
      "learning_rate": 4.3926141885325564e-05,
      "loss": 5.2064,
      "step": 2000
    },
    {
      "epoch": 0.4555393586005831,
      "grad_norm": 0.7892385721206665,
      "learning_rate": 4.240767735665695e-05,
      "loss": 4.9923,
      "step": 2500
    },
    {
      "epoch": 0.5466472303206997,
      "grad_norm": 0.840638279914856,
      "learning_rate": 4.088921282798834e-05,
      "loss": 4.8437,
      "step": 3000
    },
    {
      "epoch": 0.6377551020408163,
      "grad_norm": 0.8709360957145691,
      "learning_rate": 3.937074829931973e-05,
      "loss": 4.7268,
      "step": 3500
    },
    {
      "epoch": 0.7288629737609329,
      "grad_norm": 0.9013066291809082,
      "learning_rate": 3.785228377065112e-05,
      "loss": 4.6324,
      "step": 4000
    },
    {
      "epoch": 0.8199708454810496,
      "grad_norm": 0.9396289587020874,
      "learning_rate": 3.6333819241982507e-05,
      "loss": 4.556,
      "step": 4500
    },
    {
      "epoch": 0.9110787172011662,
      "grad_norm": 1.0571647882461548,
      "learning_rate": 3.4815354713313895e-05,
      "loss": 4.4846,
      "step": 5000
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.23895378723078822,
      "eval_loss": 4.422997951507568,
      "eval_runtime": 213.5442,
      "eval_samples_per_second": 89.185,
      "eval_steps_per_second": 2.791,
      "step": 5488
    },
    {
      "epoch": 1.0021865889212829,
      "grad_norm": 0.9928910732269287,
      "learning_rate": 3.329689018464529e-05,
      "loss": 4.4267,
      "step": 5500
    },
    {
      "epoch": 1.0932944606413995,
      "grad_norm": 1.0094648599624634,
      "learning_rate": 3.177842565597668e-05,
      "loss": 4.3598,
      "step": 6000
    },
    {
      "epoch": 1.184402332361516,
      "grad_norm": 1.1073040962219238,
      "learning_rate": 3.0259961127308068e-05,
      "loss": 4.3227,
      "step": 6500
    },
    {
      "epoch": 1.2755102040816326,
      "grad_norm": 1.0472743511199951,
      "learning_rate": 2.8741496598639456e-05,
      "loss": 4.2846,
      "step": 7000
    },
    {
      "epoch": 1.3666180758017492,
      "grad_norm": 1.0852700471878052,
      "learning_rate": 2.7223032069970845e-05,
      "loss": 4.2512,
      "step": 7500
    },
    {
      "epoch": 1.4577259475218658,
      "grad_norm": 1.0805084705352783,
      "learning_rate": 2.5704567541302237e-05,
      "loss": 4.2266,
      "step": 8000
    },
    {
      "epoch": 1.5488338192419824,
      "grad_norm": 1.1083316802978516,
      "learning_rate": 2.4186103012633625e-05,
      "loss": 4.1916,
      "step": 8500
    },
    {
      "epoch": 1.639941690962099,
      "grad_norm": 1.1683896780014038,
      "learning_rate": 2.2667638483965014e-05,
      "loss": 4.1699,
      "step": 9000
    },
    {
      "epoch": 1.7310495626822158,
      "grad_norm": 1.2223966121673584,
      "learning_rate": 2.1149173955296406e-05,
      "loss": 4.1517,
      "step": 9500
    },
    {
      "epoch": 1.8221574344023324,
      "grad_norm": 1.1384276151657104,
      "learning_rate": 1.9630709426627795e-05,
      "loss": 4.1228,
      "step": 10000
    },
    {
      "epoch": 1.913265306122449,
      "grad_norm": 1.2079554796218872,
      "learning_rate": 1.8112244897959187e-05,
      "loss": 4.1079,
      "step": 10500
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.27172573472254197,
      "eval_loss": 4.11556339263916,
      "eval_runtime": 213.5538,
      "eval_samples_per_second": 89.181,
      "eval_steps_per_second": 2.791,
      "step": 10976
    },
    {
      "epoch": 2.0043731778425657,
      "grad_norm": 1.2157140970230103,
      "learning_rate": 1.6593780369290575e-05,
      "loss": 4.0902,
      "step": 11000
    },
    {
      "epoch": 2.0954810495626823,
      "grad_norm": 1.2627109289169312,
      "learning_rate": 1.5075315840621965e-05,
      "loss": 4.0556,
      "step": 11500
    },
    {
      "epoch": 2.186588921282799,
      "grad_norm": 1.2118178606033325,
      "learning_rate": 1.3556851311953352e-05,
      "loss": 4.0461,
      "step": 12000
    },
    {
      "epoch": 2.2776967930029155,
      "grad_norm": 1.3033051490783691,
      "learning_rate": 1.2038386783284743e-05,
      "loss": 4.0326,
      "step": 12500
    },
    {
      "epoch": 2.368804664723032,
      "grad_norm": 1.2471331357955933,
      "learning_rate": 1.0519922254616133e-05,
      "loss": 4.0282,
      "step": 13000
    },
    {
      "epoch": 2.4599125364431487,
      "grad_norm": 1.3048267364501953,
      "learning_rate": 9.001457725947522e-06,
      "loss": 4.0197,
      "step": 13500
    },
    {
      "epoch": 2.5510204081632653,
      "grad_norm": 1.1969304084777832,
      "learning_rate": 7.482993197278912e-06,
      "loss": 4.0076,
      "step": 14000
    },
    {
      "epoch": 2.642128279883382,
      "grad_norm": 1.264858603477478,
      "learning_rate": 5.964528668610301e-06,
      "loss": 3.9948,
      "step": 14500
    },
    {
      "epoch": 2.7332361516034984,
      "grad_norm": 1.244125485420227,
      "learning_rate": 4.4460641399416915e-06,
      "loss": 3.9953,
      "step": 15000
    },
    {
      "epoch": 2.824344023323615,
      "grad_norm": 1.2413588762283325,
      "learning_rate": 2.9275996112730806e-06,
      "loss": 3.9837,
      "step": 15500
    },
    {
      "epoch": 2.9154518950437316,
      "grad_norm": 1.233183741569519,
      "learning_rate": 1.4091350826044704e-06,
      "loss": 3.9841,
      "step": 16000
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.2825701437173418,
      "eval_loss": 4.0273261070251465,
      "eval_runtime": 215.4599,
      "eval_samples_per_second": 88.392,
      "eval_steps_per_second": 2.766,
      "step": 16464
    },
    {
      "epoch": 3.0,
      "step": 16464,
      "total_flos": 1.44589665695957e+17,
      "train_loss": 4.469071887788319,
      "train_runtime": 13360.3551,
      "train_samples_per_second": 39.431,
      "train_steps_per_second": 1.232
    }
  ],
  "logging_steps": 500,
  "max_steps": 16464,
  "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": true
      },
      "attributes": {}
    }
  },
  "total_flos": 1.44589665695957e+17,
  "train_batch_size": 32,
  "trial_name": null,
  "trial_params": null
}