trainer_state.json

{
  "best_metric": 0.6684856753069577,
  "best_model_checkpoint": "logs/ildc/roberta-base/seed_1/checkpoint-1010",
  "epoch": 4.0,
  "eval_steps": 500,
  "global_step": 4040,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.49504950495049505,
      "grad_norm": 0.7472001314163208,
      "learning_rate": 2.925891089108911e-05,
      "loss": 0.6885,
      "step": 500
    },
    {
      "epoch": 0.9900990099009901,
      "grad_norm": 1.0043890476226807,
      "learning_rate": 2.8516336633663364e-05,
      "loss": 0.6863,
      "step": 1000
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.5110663983903421,
      "eval_best_threshold": 0.43776339292526245,
      "eval_f1": 0.6684856753069577,
      "eval_loss": 0.7003927230834961,
      "eval_precision": 0.5056759545923633,
      "eval_recall": 0.9859154929577465,
      "eval_runtime": 8.3061,
      "eval_samples_per_second": 119.671,
      "eval_steps_per_second": 3.853,
      "step": 1010
    },
    {
      "epoch": 1.4851485148514851,
      "grad_norm": 0.5840580463409424,
      "learning_rate": 2.7773762376237626e-05,
      "loss": 0.6843,
      "step": 1500
    },
    {
      "epoch": 1.9801980198019802,
      "grad_norm": 1.2000515460968018,
      "learning_rate": 2.7031188118811882e-05,
      "loss": 0.6812,
      "step": 2000
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.5030181086519114,
      "eval_best_threshold": 0.43332192301750183,
      "eval_f1": 0.6671159029649596,
      "eval_loss": 0.699403703212738,
      "eval_precision": 0.5015197568389058,
      "eval_recall": 0.9959758551307847,
      "eval_runtime": 8.2485,
      "eval_samples_per_second": 120.507,
      "eval_steps_per_second": 3.88,
      "step": 2020
    },
    {
      "epoch": 2.4752475247524752,
      "grad_norm": 1.7046343088150024,
      "learning_rate": 2.628861386138614e-05,
      "loss": 0.6823,
      "step": 2500
    },
    {
      "epoch": 2.9702970297029703,
      "grad_norm": 0.7660498023033142,
      "learning_rate": 2.5546039603960396e-05,
      "loss": 0.6816,
      "step": 3000
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.5030181086519114,
      "eval_best_threshold": 0.332852303981781,
      "eval_f1": 0.6644021739130435,
      "eval_loss": 0.751548707485199,
      "eval_precision": 0.5015384615384615,
      "eval_recall": 0.9839034205231388,
      "eval_runtime": 8.2689,
      "eval_samples_per_second": 120.21,
      "eval_steps_per_second": 3.87,
      "step": 3030
    },
    {
      "epoch": 3.4653465346534653,
      "grad_norm": 0.4976850748062134,
      "learning_rate": 2.480346534653465e-05,
      "loss": 0.6795,
      "step": 3500
    },
    {
      "epoch": 3.9603960396039604,
      "grad_norm": 1.8062459230422974,
      "learning_rate": 2.4060891089108914e-05,
      "loss": 0.6796,
      "step": 4000
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.5030181086519114,
      "eval_best_threshold": 0.4006907343864441,
      "eval_f1": 0.6671159029649596,
      "eval_loss": 0.7038925290107727,
      "eval_precision": 0.5015197568389058,
      "eval_recall": 0.9959758551307847,
      "eval_runtime": 8.2224,
      "eval_samples_per_second": 120.89,
      "eval_steps_per_second": 3.892,
      "step": 4040
    },
    {
      "epoch": 4.0,
      "step": 4040,
      "total_flos": 6.343227643160166e+17,
      "train_loss": 0.6829259282291525,
      "train_runtime": 2722.2879,
      "train_samples_per_second": 237.337,
      "train_steps_per_second": 7.42
    }
  ],
  "logging_steps": 500,
  "max_steps": 20200,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 20,
  "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": 6.343227643160166e+17,
  "train_batch_size": 8,
  "trial_name": null,
  "trial_params": null
}