{
  "best_global_step": 7000,
  "best_metric": 0.4164,
  "best_model_checkpoint": "models/facebook/bart-large-cnn_30_5e-05/checkpoint-7000",
  "epoch": 28.225806451612904,
  "eval_steps": 500,
  "global_step": 7000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 2.0161290322580645,
      "grad_norm": 2.271665573120117,
      "learning_rate": 4.911573288058857e-05,
      "loss": 0.6372,
      "step": 500
    },
    {
      "epoch": 2.0161290322580645,
      "eval_gen_len": 86.6301,
      "eval_loss": 0.6856603622436523,
      "eval_rouge1": 0.3918,
      "eval_rouge2": 0.3151,
      "eval_rougeL": 0.3653,
      "eval_rougeLsum": 0.3829,
      "eval_runtime": 470.8787,
      "eval_samples_per_second": 2.4,
      "eval_steps_per_second": 0.076,
      "step": 500
    },
    {
      "epoch": 4.032258064516129,
      "grad_norm": 1.920154333114624,
      "learning_rate": 4.557866440294284e-05,
      "loss": 0.3093,
      "step": 1000
    },
    {
      "epoch": 4.032258064516129,
      "eval_gen_len": 86.4681,
      "eval_loss": 0.8830690383911133,
      "eval_rouge1": 0.4027,
      "eval_rouge2": 0.328,
      "eval_rougeL": 0.3763,
      "eval_rougeLsum": 0.3948,
      "eval_runtime": 463.6435,
      "eval_samples_per_second": 2.437,
      "eval_steps_per_second": 0.078,
      "step": 1000
    },
    {
      "epoch": 6.048387096774194,
      "grad_norm": 2.545297622680664,
      "learning_rate": 4.204867006225241e-05,
      "loss": 0.1443,
      "step": 1500
    },
    {
      "epoch": 6.048387096774194,
      "eval_gen_len": 85.0513,
      "eval_loss": 1.0781320333480835,
      "eval_rouge1": 0.4129,
      "eval_rouge2": 0.3395,
      "eval_rougeL": 0.3857,
      "eval_rougeLsum": 0.4057,
      "eval_runtime": 467.3383,
      "eval_samples_per_second": 2.418,
      "eval_steps_per_second": 0.077,
      "step": 1500
    },
    {
      "epoch": 8.064516129032258,
      "grad_norm": 1.484631061553955,
      "learning_rate": 3.8511601584606685e-05,
      "loss": 0.076,
      "step": 2000
    },
    {
      "epoch": 8.064516129032258,
      "eval_gen_len": 77.6938,
      "eval_loss": 1.1164271831512451,
      "eval_rouge1": 0.4342,
      "eval_rouge2": 0.3594,
      "eval_rougeL": 0.4087,
      "eval_rougeLsum": 0.4259,
      "eval_runtime": 433.3086,
      "eval_samples_per_second": 2.608,
      "eval_steps_per_second": 0.083,
      "step": 2000
    },
    {
      "epoch": 10.080645161290322,
      "grad_norm": 0.9293203949928284,
      "learning_rate": 3.4974533106960955e-05,
      "loss": 0.046,
      "step": 2500
    },
    {
      "epoch": 10.080645161290322,
      "eval_gen_len": 79.4717,
      "eval_loss": 1.4387677907943726,
      "eval_rouge1": 0.4304,
      "eval_rouge2": 0.3557,
      "eval_rougeL": 0.406,
      "eval_rougeLsum": 0.423,
      "eval_runtime": 431.9944,
      "eval_samples_per_second": 2.616,
      "eval_steps_per_second": 0.083,
      "step": 2500
    },
    {
      "epoch": 12.096774193548388,
      "grad_norm": 1.3663330078125,
      "learning_rate": 3.143746462931523e-05,
      "loss": 0.0322,
      "step": 3000
    },
    {
      "epoch": 12.096774193548388,
      "eval_gen_len": 78.7823,
      "eval_loss": 1.2941193580627441,
      "eval_rouge1": 0.4297,
      "eval_rouge2": 0.3536,
      "eval_rougeL": 0.4034,
      "eval_rougeLsum": 0.4216,
      "eval_runtime": 429.3366,
      "eval_samples_per_second": 2.632,
      "eval_steps_per_second": 0.084,
      "step": 3000
    },
    {
      "epoch": 14.112903225806452,
      "grad_norm": 3.2132232189178467,
      "learning_rate": 2.79003961516695e-05,
      "loss": 0.0231,
      "step": 3500
    },
    {
      "epoch": 14.112903225806452,
      "eval_gen_len": 80.5938,
      "eval_loss": 1.4391534328460693,
      "eval_rouge1": 0.4162,
      "eval_rouge2": 0.342,
      "eval_rougeL": 0.3912,
      "eval_rougeLsum": 0.4083,
      "eval_runtime": 429.1979,
      "eval_samples_per_second": 2.633,
      "eval_steps_per_second": 0.084,
      "step": 3500
    },
    {
      "epoch": 16.129032258064516,
      "grad_norm": 0.9062560796737671,
      "learning_rate": 2.436332767402377e-05,
      "loss": 0.0173,
      "step": 4000
    },
    {
      "epoch": 16.129032258064516,
      "eval_gen_len": 81.5655,
      "eval_loss": 1.5033551454544067,
      "eval_rouge1": 0.4273,
      "eval_rouge2": 0.3537,
      "eval_rougeL": 0.4024,
      "eval_rougeLsum": 0.4199,
      "eval_runtime": 427.5276,
      "eval_samples_per_second": 2.643,
      "eval_steps_per_second": 0.084,
      "step": 4000
    },
    {
      "epoch": 18.14516129032258,
      "grad_norm": Infinity,
      "learning_rate": 2.082625919637804e-05,
      "loss": 0.0126,
      "step": 4500
    },
    {
      "epoch": 18.14516129032258,
      "eval_gen_len": 81.192,
      "eval_loss": 1.4981940984725952,
      "eval_rouge1": 0.4252,
      "eval_rouge2": 0.3528,
      "eval_rougeL": 0.4013,
      "eval_rougeLsum": 0.4175,
      "eval_runtime": 437.7384,
      "eval_samples_per_second": 2.581,
      "eval_steps_per_second": 0.082,
      "step": 4500
    },
    {
      "epoch": 20.161290322580644,
      "grad_norm": 0.3979549705982208,
      "learning_rate": 1.729626485568761e-05,
      "loss": 0.009,
      "step": 5000
    },
    {
      "epoch": 20.161290322580644,
      "eval_gen_len": 81.9372,
      "eval_loss": 1.5800154209136963,
      "eval_rouge1": 0.4257,
      "eval_rouge2": 0.3534,
      "eval_rougeL": 0.401,
      "eval_rougeLsum": 0.418,
      "eval_runtime": 434.4426,
      "eval_samples_per_second": 2.601,
      "eval_steps_per_second": 0.083,
      "step": 5000
    },
    {
      "epoch": 22.177419354838708,
      "grad_norm": 1.6955211162567139,
      "learning_rate": 1.375919637804188e-05,
      "loss": 0.0063,
      "step": 5500
    },
    {
      "epoch": 22.177419354838708,
      "eval_gen_len": 79.0938,
      "eval_loss": 1.576202154159546,
      "eval_rouge1": 0.4323,
      "eval_rouge2": 0.3582,
      "eval_rougeL": 0.4074,
      "eval_rougeLsum": 0.4245,
      "eval_runtime": 428.2115,
      "eval_samples_per_second": 2.639,
      "eval_steps_per_second": 0.084,
      "step": 5500
    },
    {
      "epoch": 24.193548387096776,
      "grad_norm": 0.2606065273284912,
      "learning_rate": 1.0222127900396152e-05,
      "loss": 0.0046,
      "step": 6000
    },
    {
      "epoch": 24.193548387096776,
      "eval_gen_len": 80.6867,
      "eval_loss": 1.5904020071029663,
      "eval_rouge1": 0.4385,
      "eval_rouge2": 0.3675,
      "eval_rougeL": 0.4145,
      "eval_rougeLsum": 0.4313,
      "eval_runtime": 427.4723,
      "eval_samples_per_second": 2.643,
      "eval_steps_per_second": 0.084,
      "step": 6000
    },
    {
      "epoch": 26.20967741935484,
      "grad_norm": 0.12859472632408142,
      "learning_rate": 6.685059422750426e-06,
      "loss": 0.0033,
      "step": 6500
    },
    {
      "epoch": 26.20967741935484,
      "eval_gen_len": 79.9637,
      "eval_loss": 1.5917092561721802,
      "eval_rouge1": 0.4393,
      "eval_rouge2": 0.3686,
      "eval_rougeL": 0.415,
      "eval_rougeLsum": 0.4312,
      "eval_runtime": 433.2177,
      "eval_samples_per_second": 2.608,
      "eval_steps_per_second": 0.083,
      "step": 6500
    },
    {
      "epoch": 28.225806451612904,
      "grad_norm": 0.3688865005970001,
      "learning_rate": 3.1479909451046973e-06,
      "loss": 0.002,
      "step": 7000
    },
    {
      "epoch": 28.225806451612904,
      "eval_gen_len": 80.7027,
      "eval_loss": 1.6072598695755005,
      "eval_rouge1": 0.4404,
      "eval_rouge2": 0.3687,
      "eval_rougeL": 0.4164,
      "eval_rougeLsum": 0.4327,
      "eval_runtime": 441.5366,
      "eval_samples_per_second": 2.559,
      "eval_steps_per_second": 0.082,
      "step": 7000
    }
  ],
  "logging_steps": 500,
  "max_steps": 7440,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 30,
  "save_steps": 500,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 4.704962073115853e+17,
  "train_batch_size": 32,
  "trial_name": null,
  "trial_params": null
}