{
  "best_global_step": 2045,
  "best_metric": 0.1048,
  "best_model_checkpoint": "results_T5-small/checkpoint-2045",
  "epoch": 10.0,
  "eval_steps": 409,
  "global_step": 4090,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "grad_norm": 0.4738257825374603,
      "learning_rate": 3.6009779951100247e-05,
      "loss": 1.2153,
      "step": 409
    },
    {
      "epoch": 1.0,
      "eval_gen_len": 20.0,
      "eval_loss": 0.9282543659210205,
      "eval_rouge1": 0.1023,
      "eval_rouge2": 0.0711,
      "eval_rougeL": 0.0935,
      "eval_rougeLsum": 0.1013,
      "eval_runtime": 88.6924,
      "eval_samples_per_second": 18.435,
      "eval_steps_per_second": 1.161,
      "step": 409
    },
    {
      "epoch": 2.0,
      "grad_norm": 0.5024434924125671,
      "learning_rate": 3.200977995110025e-05,
      "loss": 1.0542,
      "step": 818
    },
    {
      "epoch": 2.0,
      "eval_gen_len": 20.0,
      "eval_loss": 0.8425189256668091,
      "eval_rouge1": 0.1045,
      "eval_rouge2": 0.0736,
      "eval_rougeL": 0.0946,
      "eval_rougeLsum": 0.1038,
      "eval_runtime": 88.6494,
      "eval_samples_per_second": 18.443,
      "eval_steps_per_second": 1.162,
      "step": 818
    },
    {
      "epoch": 3.0,
      "grad_norm": 0.4162379503250122,
      "learning_rate": 2.800977995110025e-05,
      "loss": 0.9748,
      "step": 1227
    },
    {
      "epoch": 3.0,
      "eval_gen_len": 20.0,
      "eval_loss": 0.797379732131958,
      "eval_rouge1": 0.103,
      "eval_rouge2": 0.0728,
      "eval_rougeL": 0.0936,
      "eval_rougeLsum": 0.1022,
      "eval_runtime": 88.7962,
      "eval_samples_per_second": 18.413,
      "eval_steps_per_second": 1.16,
      "step": 1227
    },
    {
      "epoch": 4.0,
      "grad_norm": 0.5149953961372375,
      "learning_rate": 2.401955990220049e-05,
      "loss": 0.928,
      "step": 1636
    },
    {
      "epoch": 4.0,
      "eval_gen_len": 20.0,
      "eval_loss": 0.7670938372612,
      "eval_rouge1": 0.1029,
      "eval_rouge2": 0.073,
      "eval_rougeL": 0.0942,
      "eval_rougeLsum": 0.102,
      "eval_runtime": 88.649,
      "eval_samples_per_second": 18.444,
      "eval_steps_per_second": 1.162,
      "step": 1636
    },
    {
      "epoch": 5.0,
      "grad_norm": 0.4161084294319153,
      "learning_rate": 2.001955990220049e-05,
      "loss": 0.895,
      "step": 2045
    },
    {
      "epoch": 5.0,
      "eval_gen_len": 20.0,
      "eval_loss": 0.746514618396759,
      "eval_rouge1": 0.1048,
      "eval_rouge2": 0.0747,
      "eval_rougeL": 0.0956,
      "eval_rougeLsum": 0.1039,
      "eval_runtime": 88.6188,
      "eval_samples_per_second": 18.45,
      "eval_steps_per_second": 1.162,
      "step": 2045
    },
    {
      "epoch": 6.0,
      "grad_norm": 0.38424187898635864,
      "learning_rate": 1.601955990220049e-05,
      "loss": 0.873,
      "step": 2454
    },
    {
      "epoch": 6.0,
      "eval_gen_len": 20.0,
      "eval_loss": 0.7311462759971619,
      "eval_rouge1": 0.1036,
      "eval_rouge2": 0.0738,
      "eval_rougeL": 0.0947,
      "eval_rougeLsum": 0.1027,
      "eval_runtime": 88.5139,
      "eval_samples_per_second": 18.472,
      "eval_steps_per_second": 1.164,
      "step": 2454
    },
    {
      "epoch": 7.0,
      "grad_norm": 0.4080784320831299,
      "learning_rate": 1.201955990220049e-05,
      "loss": 0.8575,
      "step": 2863
    },
    {
      "epoch": 7.0,
      "eval_gen_len": 20.0,
      "eval_loss": 0.7203021049499512,
      "eval_rouge1": 0.1041,
      "eval_rouge2": 0.0743,
      "eval_rougeL": 0.0951,
      "eval_rougeLsum": 0.1032,
      "eval_runtime": 88.4525,
      "eval_samples_per_second": 18.485,
      "eval_steps_per_second": 1.164,
      "step": 2863
    },
    {
      "epoch": 8.0,
      "grad_norm": 0.5001679062843323,
      "learning_rate": 8.01955990220049e-06,
      "loss": 0.8454,
      "step": 3272
    },
    {
      "epoch": 8.0,
      "eval_gen_len": 20.0,
      "eval_loss": 0.714043915271759,
      "eval_rouge1": 0.1046,
      "eval_rouge2": 0.0746,
      "eval_rougeL": 0.0953,
      "eval_rougeLsum": 0.1037,
      "eval_runtime": 87.8037,
      "eval_samples_per_second": 18.621,
      "eval_steps_per_second": 1.173,
      "step": 3272
    },
    {
      "epoch": 9.0,
      "grad_norm": 0.4022293984889984,
      "learning_rate": 4.019559902200489e-06,
      "loss": 0.8393,
      "step": 3681
    },
    {
      "epoch": 9.0,
      "eval_gen_len": 20.0,
      "eval_loss": 0.7095059156417847,
      "eval_rouge1": 0.1044,
      "eval_rouge2": 0.0745,
      "eval_rougeL": 0.0951,
      "eval_rougeLsum": 0.1035,
      "eval_runtime": 88.118,
      "eval_samples_per_second": 18.555,
      "eval_steps_per_second": 1.169,
      "step": 3681
    },
    {
      "epoch": 10.0,
      "grad_norm": 0.46870842576026917,
      "learning_rate": 2.9339853300733502e-08,
      "loss": 0.8351,
      "step": 4090
    },
    {
      "epoch": 10.0,
      "eval_gen_len": 20.0,
      "eval_loss": 0.7084447145462036,
      "eval_rouge1": 0.1046,
      "eval_rouge2": 0.0747,
      "eval_rougeL": 0.0953,
      "eval_rougeLsum": 0.1037,
      "eval_runtime": 87.9897,
      "eval_samples_per_second": 18.582,
      "eval_steps_per_second": 1.171,
      "step": 4090
    }
  ],
  "logging_steps": 409,
  "max_steps": 4090,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 10,
  "save_steps": 409,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 1.25855187075072e+16,
  "train_batch_size": 16,
  "trial_name": null,
  "trial_params": null
}