trainer_state.json

{
  "best_metric": 0.6991622239146992,
  "best_model_checkpoint": "logs/ecthr_a/roberta-base/seed_1/checkpoint-1128",
  "epoch": 7.0,
  "eval_steps": 500,
  "global_step": 1974,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "eval_loss": 0.17883636057376862,
      "eval_macro-f1": 0.5361152463526995,
      "eval_micro-f1": 0.6690590111642744,
      "eval_runtime": 8.5677,
      "eval_samples_per_second": 116.717,
      "eval_steps_per_second": 3.735,
      "step": 282
    },
    {
      "epoch": 1.773049645390071,
      "grad_norm": 1.5047844648361206,
      "learning_rate": 2.7340425531914897e-05,
      "loss": 0.1598,
      "step": 500
    },
    {
      "epoch": 2.0,
      "eval_loss": 0.165725439786911,
      "eval_macro-f1": 0.5864998053589437,
      "eval_micro-f1": 0.6876456876456877,
      "eval_runtime": 8.5405,
      "eval_samples_per_second": 117.089,
      "eval_steps_per_second": 3.747,
      "step": 564
    },
    {
      "epoch": 3.0,
      "eval_loss": 0.1847357153892517,
      "eval_macro-f1": 0.619725405380703,
      "eval_micro-f1": 0.6802973977695167,
      "eval_runtime": 9.7669,
      "eval_samples_per_second": 102.386,
      "eval_steps_per_second": 3.276,
      "step": 846
    },
    {
      "epoch": 3.546099290780142,
      "grad_norm": 1.266696572303772,
      "learning_rate": 2.4680851063829786e-05,
      "loss": 0.1038,
      "step": 1000
    },
    {
      "epoch": 4.0,
      "eval_loss": 0.17048540711402893,
      "eval_macro-f1": 0.6382923768808028,
      "eval_micro-f1": 0.6991622239146992,
      "eval_runtime": 8.5116,
      "eval_samples_per_second": 117.487,
      "eval_steps_per_second": 3.76,
      "step": 1128
    },
    {
      "epoch": 5.0,
      "eval_loss": 0.18128305673599243,
      "eval_macro-f1": 0.6483583317279754,
      "eval_micro-f1": 0.6948249619482496,
      "eval_runtime": 9.0422,
      "eval_samples_per_second": 110.592,
      "eval_steps_per_second": 3.539,
      "step": 1410
    },
    {
      "epoch": 5.319148936170213,
      "grad_norm": 1.974063754081726,
      "learning_rate": 2.2026595744680854e-05,
      "loss": 0.0835,
      "step": 1500
    },
    {
      "epoch": 6.0,
      "eval_loss": 0.1945626437664032,
      "eval_macro-f1": 0.6427139982243243,
      "eval_micro-f1": 0.6928838951310862,
      "eval_runtime": 8.6426,
      "eval_samples_per_second": 115.707,
      "eval_steps_per_second": 3.703,
      "step": 1692
    },
    {
      "epoch": 7.0,
      "eval_loss": 0.20862846076488495,
      "eval_macro-f1": 0.6248697036429669,
      "eval_micro-f1": 0.6922798115259152,
      "eval_runtime": 8.4606,
      "eval_samples_per_second": 118.195,
      "eval_steps_per_second": 3.782,
      "step": 1974
    },
    {
      "epoch": 7.0,
      "step": 1974,
      "total_flos": 3.099603884499272e+17,
      "train_loss": 0.1036064192396046,
      "train_runtime": 1167.625,
      "train_samples_per_second": 154.159,
      "train_steps_per_second": 4.83
    }
  ],
  "logging_steps": 500,
  "max_steps": 5640,
  "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": 3.099603884499272e+17,
  "train_batch_size": 16,
  "trial_name": null,
  "trial_params": null
}