trainer_state_5.json

{
  "best_metric": 0.0059661865234375,
  "best_model_checkpoint": "model_fewrel_1_4-task5/checkpoint-630",
  "epoch": 10.0,
  "eval_steps": 500,
  "global_step": 2100,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "eval_loss": 0.017303466796875,
      "eval_rouge1": 98.4539,
      "eval_rouge2": 97.5762,
      "eval_rougeL": 98.1882,
      "eval_rougeLsum": 98.4482,
      "eval_runtime": 34.4843,
      "eval_samples_per_second": 32.479,
      "eval_steps_per_second": 1.015,
      "step": 210
    },
    {
      "epoch": 2.0,
      "eval_loss": 0.00751495361328125,
      "eval_rouge1": 99.1583,
      "eval_rouge2": 98.6851,
      "eval_rougeL": 99.0047,
      "eval_rougeLsum": 99.1647,
      "eval_runtime": 32.741,
      "eval_samples_per_second": 34.208,
      "eval_steps_per_second": 1.069,
      "step": 420
    },
    {
      "epoch": 2.380952380952381,
      "grad_norm": 0.014579183422029018,
      "learning_rate": 0.0008665259359149131,
      "loss": 0.0573,
      "step": 500
    },
    {
      "epoch": 3.0,
      "eval_loss": 0.0059661865234375,
      "eval_rouge1": 99.3646,
      "eval_rouge2": 98.9205,
      "eval_rougeL": 99.2221,
      "eval_rougeLsum": 99.3603,
      "eval_runtime": 32.9254,
      "eval_samples_per_second": 34.016,
      "eval_steps_per_second": 1.063,
      "step": 630
    },
    {
      "epoch": 4.0,
      "eval_loss": 0.00925445556640625,
      "eval_rouge1": 99.1633,
      "eval_rouge2": 98.6891,
      "eval_rougeL": 99.0235,
      "eval_rougeLsum": 99.1601,
      "eval_runtime": 33.2811,
      "eval_samples_per_second": 33.653,
      "eval_steps_per_second": 1.052,
      "step": 840
    },
    {
      "epoch": 4.761904761904762,
      "grad_norm": 0.1038060188293457,
      "learning_rate": 0.0005373650467932121,
      "loss": 0.0102,
      "step": 1000
    },
    {
      "epoch": 5.0,
      "eval_loss": 0.007312774658203125,
      "eval_rouge1": 99.3523,
      "eval_rouge2": 98.8747,
      "eval_rougeL": 99.1909,
      "eval_rougeLsum": 99.3521,
      "eval_runtime": 33.1107,
      "eval_samples_per_second": 33.826,
      "eval_steps_per_second": 1.057,
      "step": 1050
    },
    {
      "epoch": 6.0,
      "eval_loss": 0.008575439453125,
      "eval_rouge1": 99.4615,
      "eval_rouge2": 99.0736,
      "eval_rougeL": 99.3351,
      "eval_rougeLsum": 99.4494,
      "eval_runtime": 32.8562,
      "eval_samples_per_second": 34.088,
      "eval_steps_per_second": 1.065,
      "step": 1260
    },
    {
      "epoch": 7.0,
      "eval_loss": 0.00952911376953125,
      "eval_rouge1": 99.3799,
      "eval_rouge2": 99.0097,
      "eval_rougeL": 99.2562,
      "eval_rougeLsum": 99.371,
      "eval_runtime": 33.1813,
      "eval_samples_per_second": 33.754,
      "eval_steps_per_second": 1.055,
      "step": 1470
    },
    {
      "epoch": 7.142857142857143,
      "grad_norm": 0.025980567559599876,
      "learning_rate": 0.00018825509907063325,
      "loss": 0.0042,
      "step": 1500
    },
    {
      "epoch": 8.0,
      "eval_loss": 0.00867462158203125,
      "eval_rouge1": 99.4488,
      "eval_rouge2": 99.0827,
      "eval_rougeL": 99.3346,
      "eval_rougeLsum": 99.4596,
      "eval_runtime": 32.7477,
      "eval_samples_per_second": 34.201,
      "eval_steps_per_second": 1.069,
      "step": 1680
    },
    {
      "epoch": 9.0,
      "eval_loss": 0.00786590576171875,
      "eval_rouge1": 99.3538,
      "eval_rouge2": 98.9422,
      "eval_rougeL": 99.2192,
      "eval_rougeLsum": 99.3563,
      "eval_runtime": 33.0744,
      "eval_samples_per_second": 33.863,
      "eval_steps_per_second": 1.058,
      "step": 1890
    },
    {
      "epoch": 9.523809523809524,
      "grad_norm": 0.18442556262016296,
      "learning_rate": 5.5845868874357386e-06,
      "loss": 0.0023,
      "step": 2000
    },
    {
      "epoch": 10.0,
      "eval_loss": 0.007965087890625,
      "eval_rouge1": 99.3538,
      "eval_rouge2": 98.9422,
      "eval_rougeL": 99.2192,
      "eval_rougeLsum": 99.3563,
      "eval_runtime": 33.0331,
      "eval_samples_per_second": 33.905,
      "eval_steps_per_second": 1.06,
      "step": 2100
    }
  ],
  "logging_steps": 500,
  "max_steps": 2100,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 10,
  "save_steps": 500,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 2.3099168784384e+16,
  "train_batch_size": 16,
  "trial_name": null,
  "trial_params": null
}