trainer_state_6.json

{
  "best_metric": 0.03607177734375,
  "best_model_checkpoint": "model_fewrel_1_5-task6/checkpoint-1260",
  "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": 0.051544189453125,
      "eval_rouge1": 97.5746,
      "eval_rouge2": 95.9962,
      "eval_rougeL": 96.8014,
      "eval_rougeLsum": 97.5759,
      "eval_runtime": 33.8471,
      "eval_samples_per_second": 33.09,
      "eval_steps_per_second": 1.034,
      "step": 210
    },
    {
      "epoch": 2.0,
      "eval_loss": 0.038238525390625,
      "eval_rouge1": 97.0504,
      "eval_rouge2": 95.0089,
      "eval_rougeL": 96.0484,
      "eval_rougeLsum": 97.0773,
      "eval_runtime": 33.7575,
      "eval_samples_per_second": 33.178,
      "eval_steps_per_second": 1.037,
      "step": 420
    },
    {
      "epoch": 2.380952380952381,
      "grad_norm": 0.15559855103492737,
      "learning_rate": 0.0008665259359149131,
      "loss": 0.0759,
      "step": 500
    },
    {
      "epoch": 3.0,
      "eval_loss": 0.039306640625,
      "eval_rouge1": 97.8822,
      "eval_rouge2": 96.4368,
      "eval_rougeL": 97.1693,
      "eval_rougeLsum": 97.886,
      "eval_runtime": 33.8292,
      "eval_samples_per_second": 33.107,
      "eval_steps_per_second": 1.035,
      "step": 630
    },
    {
      "epoch": 4.0,
      "eval_loss": 0.036468505859375,
      "eval_rouge1": 97.8707,
      "eval_rouge2": 96.3824,
      "eval_rougeL": 97.1565,
      "eval_rougeLsum": 97.88,
      "eval_runtime": 34.5353,
      "eval_samples_per_second": 32.431,
      "eval_steps_per_second": 1.013,
      "step": 840
    },
    {
      "epoch": 4.761904761904762,
      "grad_norm": 0.021474618464708328,
      "learning_rate": 0.0005373650467932121,
      "loss": 0.0211,
      "step": 1000
    },
    {
      "epoch": 5.0,
      "eval_loss": 0.03961181640625,
      "eval_rouge1": 97.5277,
      "eval_rouge2": 95.8576,
      "eval_rougeL": 96.691,
      "eval_rougeLsum": 97.527,
      "eval_runtime": 33.9424,
      "eval_samples_per_second": 32.997,
      "eval_steps_per_second": 1.031,
      "step": 1050
    },
    {
      "epoch": 6.0,
      "eval_loss": 0.03607177734375,
      "eval_rouge1": 98.1371,
      "eval_rouge2": 96.8791,
      "eval_rougeL": 97.5059,
      "eval_rougeLsum": 98.173,
      "eval_runtime": 33.9961,
      "eval_samples_per_second": 32.945,
      "eval_steps_per_second": 1.03,
      "step": 1260
    },
    {
      "epoch": 7.0,
      "eval_loss": 0.040924072265625,
      "eval_rouge1": 98.1004,
      "eval_rouge2": 96.774,
      "eval_rougeL": 97.4333,
      "eval_rougeLsum": 98.1087,
      "eval_runtime": 34.4539,
      "eval_samples_per_second": 32.507,
      "eval_steps_per_second": 1.016,
      "step": 1470
    },
    {
      "epoch": 7.142857142857143,
      "grad_norm": 0.08763577789068222,
      "learning_rate": 0.00018825509907063325,
      "loss": 0.0103,
      "step": 1500
    },
    {
      "epoch": 8.0,
      "eval_loss": 0.03826904296875,
      "eval_rouge1": 97.8337,
      "eval_rouge2": 96.3488,
      "eval_rougeL": 97.096,
      "eval_rougeLsum": 97.8664,
      "eval_runtime": 34.3599,
      "eval_samples_per_second": 32.596,
      "eval_steps_per_second": 1.019,
      "step": 1680
    },
    {
      "epoch": 9.0,
      "eval_loss": 0.038909912109375,
      "eval_rouge1": 97.9644,
      "eval_rouge2": 96.525,
      "eval_rougeL": 97.2236,
      "eval_rougeLsum": 97.9585,
      "eval_runtime": 33.8333,
      "eval_samples_per_second": 33.103,
      "eval_steps_per_second": 1.034,
      "step": 1890
    },
    {
      "epoch": 9.523809523809524,
      "grad_norm": 0.19326545298099518,
      "learning_rate": 5.5845868874357386e-06,
      "loss": 0.0071,
      "step": 2000
    },
    {
      "epoch": 10.0,
      "eval_loss": 0.039276123046875,
      "eval_rouge1": 98.0097,
      "eval_rouge2": 96.6105,
      "eval_rougeL": 97.2833,
      "eval_rougeLsum": 98.0092,
      "eval_runtime": 33.8974,
      "eval_samples_per_second": 33.041,
      "eval_steps_per_second": 1.033,
      "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
}