{
  "best_metric": 0.03748522698879242,
  "best_model_checkpoint": "doc-topic-model_eval-00_train-03/checkpoint-11000",
  "epoch": 7.889546351084813,
  "eval_steps": 1000,
  "global_step": 16000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.2465483234714004,
      "grad_norm": 0.3881610631942749,
      "learning_rate": 1.9950690335305722e-05,
      "loss": 0.1669,
      "step": 500
    },
    {
      "epoch": 0.4930966469428008,
      "grad_norm": 0.3564796447753906,
      "learning_rate": 1.9901380670611442e-05,
      "loss": 0.0935,
      "step": 1000
    },
    {
      "epoch": 0.4930966469428008,
      "eval_accuracy": 0.9814660487265615,
      "eval_f1": 0.0,
      "eval_loss": 0.08951831609010696,
      "eval_precision": 0.0,
      "eval_recall": 0.0,
      "eval_runtime": 12.2339,
      "eval_samples_per_second": 662.913,
      "eval_steps_per_second": 2.616,
      "step": 1000
    },
    {
      "epoch": 0.7396449704142012,
      "grad_norm": 0.4097365140914917,
      "learning_rate": 1.9852071005917162e-05,
      "loss": 0.0869,
      "step": 1500
    },
    {
      "epoch": 0.9861932938856016,
      "grad_norm": 0.3858756422996521,
      "learning_rate": 1.980276134122288e-05,
      "loss": 0.0764,
      "step": 2000
    },
    {
      "epoch": 0.9861932938856016,
      "eval_accuracy": 0.9814660487265615,
      "eval_f1": 0.0,
      "eval_loss": 0.07000521570444107,
      "eval_precision": 0.0,
      "eval_recall": 0.0,
      "eval_runtime": 12.0785,
      "eval_samples_per_second": 671.442,
      "eval_steps_per_second": 2.649,
      "step": 2000
    },
    {
      "epoch": 1.232741617357002,
      "grad_norm": 0.3765203356742859,
      "learning_rate": 1.9753451676528602e-05,
      "loss": 0.068,
      "step": 2500
    },
    {
      "epoch": 1.4792899408284024,
      "grad_norm": 0.3380969762802124,
      "learning_rate": 1.9704142011834322e-05,
      "loss": 0.0621,
      "step": 3000
    },
    {
      "epoch": 1.4792899408284024,
      "eval_accuracy": 0.982086823419363,
      "eval_f1": 0.07304730473047305,
      "eval_loss": 0.05673361197113991,
      "eval_precision": 0.8924731182795699,
      "eval_recall": 0.03808212892865336,
      "eval_runtime": 14.4231,
      "eval_samples_per_second": 562.293,
      "eval_steps_per_second": 2.219,
      "step": 3000
    },
    {
      "epoch": 1.725838264299803,
      "grad_norm": 0.3490855395793915,
      "learning_rate": 1.965483234714004e-05,
      "loss": 0.056,
      "step": 3500
    },
    {
      "epoch": 1.972386587771203,
      "grad_norm": 0.4068734347820282,
      "learning_rate": 1.9605522682445763e-05,
      "loss": 0.0542,
      "step": 4000
    },
    {
      "epoch": 1.972386587771203,
      "eval_accuracy": 0.9840823731167708,
      "eval_f1": 0.2891322235584531,
      "eval_loss": 0.04972199350595474,
      "eval_precision": 0.8390889052167524,
      "eval_recall": 0.174657796130611,
      "eval_runtime": 14.6604,
      "eval_samples_per_second": 553.191,
      "eval_steps_per_second": 2.183,
      "step": 4000
    },
    {
      "epoch": 2.2189349112426036,
      "grad_norm": 0.39391544461250305,
      "learning_rate": 1.9556213017751483e-05,
      "loss": 0.0488,
      "step": 4500
    },
    {
      "epoch": 2.465483234714004,
      "grad_norm": 0.2698710858821869,
      "learning_rate": 1.95069033530572e-05,
      "loss": 0.0468,
      "step": 5000
    },
    {
      "epoch": 2.465483234714004,
      "eval_accuracy": 0.9852672307496066,
      "eval_f1": 0.42163244867300953,
      "eval_loss": 0.046487729996442795,
      "eval_precision": 0.7738970588235294,
      "eval_recall": 0.28974535443909155,
      "eval_runtime": 14.5194,
      "eval_samples_per_second": 558.563,
      "eval_steps_per_second": 2.204,
      "step": 5000
    },
    {
      "epoch": 2.712031558185404,
      "grad_norm": 0.39123860001564026,
      "learning_rate": 1.9457593688362923e-05,
      "loss": 0.0456,
      "step": 5500
    },
    {
      "epoch": 2.9585798816568047,
      "grad_norm": 0.4525609016418457,
      "learning_rate": 1.940828402366864e-05,
      "loss": 0.0441,
      "step": 6000
    },
    {
      "epoch": 2.9585798816568047,
      "eval_accuracy": 0.9860793401079978,
      "eval_f1": 0.48790406673618353,
      "eval_loss": 0.04352094233036041,
      "eval_precision": 0.7666721284614124,
      "eval_recall": 0.3578037776248375,
      "eval_runtime": 14.4967,
      "eval_samples_per_second": 559.439,
      "eval_steps_per_second": 2.207,
      "step": 6000
    },
    {
      "epoch": 3.2051282051282053,
      "grad_norm": 0.47004055976867676,
      "learning_rate": 1.935897435897436e-05,
      "loss": 0.0409,
      "step": 6500
    },
    {
      "epoch": 3.4516765285996054,
      "grad_norm": 0.3331545293331146,
      "learning_rate": 1.930966469428008e-05,
      "loss": 0.0395,
      "step": 7000
    },
    {
      "epoch": 3.4516765285996054,
      "eval_accuracy": 0.9862437461910228,
      "eval_f1": 0.5321958743011375,
      "eval_loss": 0.04173528403043747,
      "eval_precision": 0.7197236344674749,
      "eval_recall": 0.42219163416685784,
      "eval_runtime": 14.6186,
      "eval_samples_per_second": 554.772,
      "eval_steps_per_second": 2.189,
      "step": 7000
    },
    {
      "epoch": 3.698224852071006,
      "grad_norm": 0.46907347440719604,
      "learning_rate": 1.92603550295858e-05,
      "loss": 0.0382,
      "step": 7500
    },
    {
      "epoch": 3.9447731755424065,
      "grad_norm": 0.3860171437263489,
      "learning_rate": 1.921104536489152e-05,
      "loss": 0.0384,
      "step": 8000
    },
    {
      "epoch": 3.9447731755424065,
      "eval_accuracy": 0.9866335019912978,
      "eval_f1": 0.5599776046283768,
      "eval_loss": 0.040070790797472,
      "eval_precision": 0.7181665868836764,
      "eval_recall": 0.4588973006041141,
      "eval_runtime": 14.6864,
      "eval_samples_per_second": 552.213,
      "eval_steps_per_second": 2.179,
      "step": 8000
    },
    {
      "epoch": 4.191321499013807,
      "grad_norm": 0.39796024560928345,
      "learning_rate": 1.916173570019724e-05,
      "loss": 0.0348,
      "step": 8500
    },
    {
      "epoch": 4.437869822485207,
      "grad_norm": 0.3643665313720703,
      "learning_rate": 1.911242603550296e-05,
      "loss": 0.0343,
      "step": 9000
    },
    {
      "epoch": 4.437869822485207,
      "eval_accuracy": 0.9869694006264439,
      "eval_f1": 0.5789136209581387,
      "eval_loss": 0.039344046264886856,
      "eval_precision": 0.7217083476076281,
      "eval_recall": 0.4832912747572073,
      "eval_runtime": 14.3905,
      "eval_samples_per_second": 563.565,
      "eval_steps_per_second": 2.224,
      "step": 9000
    },
    {
      "epoch": 4.684418145956608,
      "grad_norm": 0.2764923572540283,
      "learning_rate": 1.906311637080868e-05,
      "loss": 0.0345,
      "step": 9500
    },
    {
      "epoch": 4.930966469428008,
      "grad_norm": 0.432156503200531,
      "learning_rate": 1.90138067061144e-05,
      "loss": 0.0337,
      "step": 10000
    },
    {
      "epoch": 4.930966469428008,
      "eval_accuracy": 0.9873265586688776,
      "eval_f1": 0.5906802160578596,
      "eval_loss": 0.037819646298885345,
      "eval_precision": 0.7357737484319763,
      "eval_recall": 0.4933853330274528,
      "eval_runtime": 14.4562,
      "eval_samples_per_second": 561.004,
      "eval_steps_per_second": 2.214,
      "step": 10000
    },
    {
      "epoch": 5.177514792899408,
      "grad_norm": 0.27451589703559875,
      "learning_rate": 1.896459566074951e-05,
      "loss": 0.0309,
      "step": 10500
    },
    {
      "epoch": 5.424063116370808,
      "grad_norm": 0.49061715602874756,
      "learning_rate": 1.891528599605523e-05,
      "loss": 0.0305,
      "step": 11000
    },
    {
      "epoch": 5.424063116370808,
      "eval_accuracy": 0.9875278143912014,
      "eval_f1": 0.5959595959595959,
      "eval_loss": 0.03748522698879242,
      "eval_precision": 0.7457198667126278,
      "eval_recall": 0.4962911982870689,
      "eval_runtime": 14.4209,
      "eval_samples_per_second": 562.38,
      "eval_steps_per_second": 2.219,
      "step": 11000
    },
    {
      "epoch": 5.670611439842209,
      "grad_norm": 0.42573484778404236,
      "learning_rate": 1.886597633136095e-05,
      "loss": 0.0299,
      "step": 11500
    },
    {
      "epoch": 5.9171597633136095,
      "grad_norm": 0.4632101058959961,
      "learning_rate": 1.881666666666667e-05,
      "loss": 0.0295,
      "step": 12000
    },
    {
      "epoch": 5.9171597633136095,
      "eval_accuracy": 0.98739175418456,
      "eval_f1": 0.6050084361957198,
      "eval_loss": 0.03783571720123291,
      "eval_precision": 0.7213340391741663,
      "eval_recall": 0.5209910529938059,
      "eval_runtime": 14.4511,
      "eval_samples_per_second": 561.202,
      "eval_steps_per_second": 2.214,
      "step": 12000
    },
    {
      "epoch": 6.16370808678501,
      "grad_norm": 0.5974611043930054,
      "learning_rate": 1.876735700197239e-05,
      "loss": 0.0271,
      "step": 12500
    },
    {
      "epoch": 6.410256410256411,
      "grad_norm": 0.42065271735191345,
      "learning_rate": 1.8718145956607497e-05,
      "loss": 0.0271,
      "step": 13000
    },
    {
      "epoch": 6.410256410256411,
      "eval_accuracy": 0.9876794648298539,
      "eval_f1": 0.6048097467836523,
      "eval_loss": 0.037606410682201385,
      "eval_precision": 0.7457399103139013,
      "eval_recall": 0.5086793607096429,
      "eval_runtime": 14.3596,
      "eval_samples_per_second": 564.778,
      "eval_steps_per_second": 2.228,
      "step": 13000
    },
    {
      "epoch": 6.65680473372781,
      "grad_norm": 0.7562135457992554,
      "learning_rate": 1.8668836291913217e-05,
      "loss": 0.0265,
      "step": 13500
    },
    {
      "epoch": 6.903353057199211,
      "grad_norm": 0.5237112641334534,
      "learning_rate": 1.8619526627218937e-05,
      "loss": 0.0257,
      "step": 14000
    },
    {
      "epoch": 6.903353057199211,
      "eval_accuracy": 0.9874909647519027,
      "eval_f1": 0.6067896284415931,
      "eval_loss": 0.0378677099943161,
      "eval_precision": 0.7268651937239834,
      "eval_recall": 0.520761642578573,
      "eval_runtime": 14.2708,
      "eval_samples_per_second": 568.292,
      "eval_steps_per_second": 2.242,
      "step": 14000
    },
    {
      "epoch": 7.149901380670611,
      "grad_norm": 0.38859695196151733,
      "learning_rate": 1.8570216962524657e-05,
      "loss": 0.0239,
      "step": 14500
    },
    {
      "epoch": 7.396449704142012,
      "grad_norm": 0.276382178068161,
      "learning_rate": 1.8520907297830377e-05,
      "loss": 0.0234,
      "step": 15000
    },
    {
      "epoch": 7.396449704142012,
      "eval_accuracy": 0.9875901753192454,
      "eval_f1": 0.6246248821058047,
      "eval_loss": 0.037667978554964066,
      "eval_precision": 0.7108010537613426,
      "eval_recall": 0.5570849583237746,
      "eval_runtime": 14.3321,
      "eval_samples_per_second": 565.864,
      "eval_steps_per_second": 2.233,
      "step": 15000
    },
    {
      "epoch": 7.642998027613412,
      "grad_norm": 0.23438788950443268,
      "learning_rate": 1.8471696252465485e-05,
      "loss": 0.0234,
      "step": 15500
    },
    {
      "epoch": 7.889546351084813,
      "grad_norm": 0.3130985498428345,
      "learning_rate": 1.8422386587771205e-05,
      "loss": 0.0241,
      "step": 16000
    },
    {
      "epoch": 7.889546351084813,
      "eval_accuracy": 0.9877928483353884,
      "eval_f1": 0.6227828143476547,
      "eval_loss": 0.03809000924229622,
      "eval_precision": 0.7287822878228782,
      "eval_recall": 0.5437026841018582,
      "eval_runtime": 14.2635,
      "eval_samples_per_second": 568.582,
      "eval_steps_per_second": 2.243,
      "step": 16000
    }
  ],
  "logging_steps": 500,
  "max_steps": 202800,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 100,
  "save_steps": 1000,
  "stateful_callbacks": {
    "EarlyStoppingCallback": {
      "args": {
        "early_stopping_patience": 5,
        "early_stopping_threshold": 0.0
      },
      "attributes": {
        "early_stopping_patience_counter": 0
      }
    },
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 360703978994412.0,
  "train_batch_size": 4,
  "trial_name": null,
  "trial_params": null
}