trainer_state.json

{
  "best_metric": 0.716598119152416,
  "best_model_checkpoint": "logs/eurlex/roberta-base/seed_1/checkpoint-6876",
  "epoch": 2.0,
  "eval_steps": 500,
  "global_step": 6876,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.14543339150668994,
      "grad_norm": 10488.72265625,
      "learning_rate": 2.7818499127399653e-05,
      "loss": 0.1782,
      "step": 500
    },
    {
      "epoch": 0.29086678301337987,
      "grad_norm": 8601.787109375,
      "learning_rate": 2.56369982547993e-05,
      "loss": 0.109,
      "step": 1000
    },
    {
      "epoch": 0.4363001745200698,
      "grad_norm": 13134.970703125,
      "learning_rate": 2.3455497382198953e-05,
      "loss": 0.0926,
      "step": 1500
    },
    {
      "epoch": 0.5817335660267597,
      "grad_norm": 16072.458984375,
      "learning_rate": 2.1273996509598605e-05,
      "loss": 0.0851,
      "step": 2000
    },
    {
      "epoch": 0.7271669575334497,
      "grad_norm": 12878.12109375,
      "learning_rate": 1.9092495636998257e-05,
      "loss": 0.078,
      "step": 2500
    },
    {
      "epoch": 0.8726003490401396,
      "grad_norm": 12653.638671875,
      "learning_rate": 1.6910994764397905e-05,
      "loss": 0.0758,
      "step": 3000
    },
    {
      "epoch": 1.0,
      "eval_loss": 0.08141529560089111,
      "eval_macro-f1": 0.296432741440559,
      "eval_micro-f1": 0.6862200624783061,
      "eval_runtime": 20.7429,
      "eval_samples_per_second": 241.047,
      "eval_steps_per_second": 15.09,
      "step": 3438
    },
    {
      "epoch": 1.0180337405468296,
      "grad_norm": 14271.0712890625,
      "learning_rate": 1.4729493891797557e-05,
      "loss": 0.0715,
      "step": 3500
    },
    {
      "epoch": 1.1634671320535195,
      "grad_norm": 15767.826171875,
      "learning_rate": 1.2547993019197209e-05,
      "loss": 0.0679,
      "step": 4000
    },
    {
      "epoch": 1.3089005235602094,
      "grad_norm": 13860.2705078125,
      "learning_rate": 1.0366492146596857e-05,
      "loss": 0.0652,
      "step": 4500
    },
    {
      "epoch": 1.4543339150668992,
      "grad_norm": 11340.1640625,
      "learning_rate": 8.18499127399651e-06,
      "loss": 0.0651,
      "step": 5000
    },
    {
      "epoch": 1.5997673065735893,
      "grad_norm": 14212.9599609375,
      "learning_rate": 6.003490401396161e-06,
      "loss": 0.0633,
      "step": 5500
    },
    {
      "epoch": 1.7452006980802792,
      "grad_norm": 16454.734375,
      "learning_rate": 3.821989528795812e-06,
      "loss": 0.0624,
      "step": 6000
    },
    {
      "epoch": 1.8906340895869693,
      "grad_norm": 11580.310546875,
      "learning_rate": 1.6404886561954625e-06,
      "loss": 0.0618,
      "step": 6500
    },
    {
      "epoch": 2.0,
      "eval_loss": 0.07441535592079163,
      "eval_macro-f1": 0.35506427791546535,
      "eval_micro-f1": 0.716598119152416,
      "eval_runtime": 19.4037,
      "eval_samples_per_second": 257.682,
      "eval_steps_per_second": 16.131,
      "step": 6876
    },
    {
      "epoch": 2.0,
      "step": 6876,
      "total_flos": 2.896768241664e+16,
      "train_loss": 0.08159883026469232,
      "train_runtime": 766.6169,
      "train_samples_per_second": 143.488,
      "train_steps_per_second": 8.969
    }
  ],
  "logging_steps": 500,
  "max_steps": 6876,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 2,
  "save_steps": 500,
  "stateful_callbacks": {
    "EarlyStoppingCallback": {
      "args": {
        "early_stopping_patience": 3,
        "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": 2.896768241664e+16,
  "train_batch_size": 16,
  "trial_name": null,
  "trial_params": null
}