{
  "best_metric": 0.037347666919231415,
  "best_model_checkpoint": "doc-topic-model_eval-01_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": 0.981383845685049,
      "eval_f1": 0.0,
      "eval_loss": 0.08976549655199051,
      "eval_precision": 0.0,
      "eval_recall": 0.0,
      "eval_runtime": 12.0327,
      "eval_samples_per_second": 673.996,
      "eval_steps_per_second": 2.659,
      "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": 0.981383845685049,
      "eval_f1": 0.0,
      "eval_loss": 0.07017117738723755,
      "eval_precision": 0.0,
      "eval_recall": 0.0,
      "eval_runtime": 12.2763,
      "eval_samples_per_second": 660.624,
      "eval_steps_per_second": 2.607,
      "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": 0.9819748572076478,
      "eval_f1": 0.06950541410594088,
      "eval_loss": 0.05700037255883217,
      "eval_precision": 0.8911819887429644,
      "eval_recall": 0.03616292348686715,
      "eval_runtime": 14.2939,
      "eval_samples_per_second": 567.375,
      "eval_steps_per_second": 2.239,
      "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": 0.9839533993792253,
      "eval_f1": 0.2863985881759738,
      "eval_loss": 0.049796320497989655,
      "eval_precision": 0.8319296960820213,
      "eval_recall": 0.17297297297297298,
      "eval_runtime": 14.4691,
      "eval_samples_per_second": 560.505,
      "eval_steps_per_second": 2.212,
      "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": 0.9851807758266367,
      "eval_f1": 0.4191111111111111,
      "eval_loss": 0.04676121473312378,
      "eval_precision": 0.7753340184994861,
      "eval_recall": 0.2871716787209745,
      "eval_runtime": 14.3877,
      "eval_samples_per_second": 563.675,
      "eval_steps_per_second": 2.224,
      "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": 0.9861062686905623,
      "eval_f1": 0.4897725498360485,
      "eval_loss": 0.04347195476293564,
      "eval_precision": 0.774103323461665,
      "eval_recall": 0.35820327369623145,
      "eval_runtime": 14.2071,
      "eval_samples_per_second": 570.842,
      "eval_steps_per_second": 2.252,
      "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": 0.9860283175305072,
      "eval_f1": 0.5278735632183909,
      "eval_loss": 0.041799116879701614,
      "eval_precision": 0.7115558424790187,
      "eval_recall": 0.4195660449181576,
      "eval_runtime": 14.3982,
      "eval_samples_per_second": 563.265,
      "eval_steps_per_second": 2.223,
      "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": 0.9865853140014457,
      "eval_f1": 0.5588440922861804,
      "eval_loss": 0.040071167051792145,
      "eval_precision": 0.7205528846153846,
      "eval_recall": 0.45641416063951273,
      "eval_runtime": 14.3408,
      "eval_samples_per_second": 565.521,
      "eval_steps_per_second": 2.231,
      "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": 0.9868985359354848,
      "eval_f1": 0.5774364600475407,
      "eval_loss": 0.03921710327267647,
      "eval_precision": 0.7225717881249285,
      "eval_recall": 0.48085268366958506,
      "eval_runtime": 14.341,
      "eval_samples_per_second": 565.513,
      "eval_steps_per_second": 2.231,
      "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": 0.9873393143132503,
      "eval_f1": 0.5918954726118142,
      "eval_loss": 0.037751466035842896,
      "eval_precision": 0.7400045693397304,
      "eval_recall": 0.49318614389036924,
      "eval_runtime": 14.1834,
      "eval_samples_per_second": 571.794,
      "eval_steps_per_second": 2.256,
      "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": 0.9875731677934153,
      "eval_f1": 0.5988654039710861,
      "eval_loss": 0.037347666919231415,
      "eval_precision": 0.7503152585119798,
      "eval_recall": 0.49828701941377995,
      "eval_runtime": 14.3069,
      "eval_samples_per_second": 566.858,
      "eval_steps_per_second": 2.237,
      "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": 0.9875462392108508,
      "eval_f1": 0.6107641196013289,
      "eval_loss": 0.03775852918624878,
      "eval_precision": 0.7302966101694915,
      "eval_recall": 0.524857251617815,
      "eval_runtime": 14.2377,
      "eval_samples_per_second": 569.612,
      "eval_steps_per_second": 2.248,
      "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": 0.9877163144691526,
      "eval_f1": 0.607952232324603,
      "eval_loss": 0.03745845705270767,
      "eval_precision": 0.7489968791796701,
      "eval_recall": 0.5116102017510468,
      "eval_runtime": 14.2135,
      "eval_samples_per_second": 570.583,
      "eval_steps_per_second": 2.251,
      "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": 0.987587340731607,
      "eval_f1": 0.6144567705581969,
      "eval_loss": 0.03767068684101105,
      "eval_precision": 0.7284208328984448,
      "eval_recall": 0.5313285116102018,
      "eval_runtime": 14.5272,
      "eval_samples_per_second": 558.263,
      "eval_steps_per_second": 2.203,
      "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": 0.9875816715563304,
      "eval_f1": 0.6242710120068611,
      "eval_loss": 0.037651773542165756,
      "eval_precision": 0.7146784486990673,
      "eval_recall": 0.5541682527598021,
      "eval_runtime": 14.1634,
      "eval_samples_per_second": 572.603,
      "eval_steps_per_second": 2.259,
      "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": 0.9878750513769009,
      "eval_f1": 0.6261416772276363,
      "eval_loss": 0.03778611123561859,
      "eval_precision": 0.7349199835863767,
      "eval_recall": 0.5454130186524553,
      "eval_runtime": 14.2528,
      "eval_samples_per_second": 569.01,
      "eval_steps_per_second": 2.245,
      "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
}