checkpoint-1270/trainer_state.json

{
  "best_metric": 47.5767,
  "best_model_checkpoint": "Models/t5-base-Abstract-Title/checkpoint-762",
  "epoch": 5.0,
  "eval_steps": 500,
  "global_step": 1270,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.3937007874015748,
      "grad_norm": 1.9452705383300781,
      "learning_rate": 3.6850393700787406e-05,
      "loss": 2.3204,
      "step": 100
    },
    {
      "epoch": 0.7874015748031497,
      "grad_norm": 1.7186884880065918,
      "learning_rate": 3.3763779527559055e-05,
      "loss": 1.7361,
      "step": 200
    },
    {
      "epoch": 1.0,
      "eval_gen_len": 15.5789,
      "eval_loss": 1.7090216875076294,
      "eval_rouge1": 45.9911,
      "eval_rouge2": 25.0676,
      "eval_rougeL": 40.5154,
      "eval_rougeLsum": 40.4794,
      "eval_runtime": 26.9645,
      "eval_samples_per_second": 37.605,
      "eval_steps_per_second": 1.187,
      "step": 254
    },
    {
      "epoch": 1.1811023622047245,
      "grad_norm": 1.6301051378250122,
      "learning_rate": 3.061417322834646e-05,
      "loss": 1.6049,
      "step": 300
    },
    {
      "epoch": 1.574803149606299,
      "grad_norm": 1.4477652311325073,
      "learning_rate": 2.755905511811024e-05,
      "loss": 1.5778,
      "step": 400
    },
    {
      "epoch": 1.968503937007874,
      "grad_norm": 1.3940485715866089,
      "learning_rate": 2.447244094488189e-05,
      "loss": 1.6009,
      "step": 500
    },
    {
      "epoch": 2.0,
      "eval_gen_len": 15.8442,
      "eval_loss": 1.6694411039352417,
      "eval_rouge1": 47.0385,
      "eval_rouge2": 25.8258,
      "eval_rougeL": 40.9901,
      "eval_rougeLsum": 40.9956,
      "eval_runtime": 26.8145,
      "eval_samples_per_second": 37.815,
      "eval_steps_per_second": 1.193,
      "step": 508
    },
    {
      "epoch": 2.362204724409449,
      "grad_norm": 1.651326060295105,
      "learning_rate": 2.1417322834645672e-05,
      "loss": 1.5077,
      "step": 600
    },
    {
      "epoch": 2.7559055118110236,
      "grad_norm": 2.0051932334899902,
      "learning_rate": 1.8267716535433072e-05,
      "loss": 1.5208,
      "step": 700
    },
    {
      "epoch": 3.0,
      "eval_gen_len": 16.1598,
      "eval_loss": 1.650003433227539,
      "eval_rouge1": 47.5767,
      "eval_rouge2": 26.1208,
      "eval_rougeL": 41.2726,
      "eval_rougeLsum": 41.3154,
      "eval_runtime": 26.7864,
      "eval_samples_per_second": 37.855,
      "eval_steps_per_second": 1.195,
      "step": 762
    },
    {
      "epoch": 3.1496062992125986,
      "grad_norm": 1.4712779521942139,
      "learning_rate": 1.5181102362204725e-05,
      "loss": 1.5006,
      "step": 800
    },
    {
      "epoch": 3.543307086614173,
      "grad_norm": 2.0347633361816406,
      "learning_rate": 1.215748031496063e-05,
      "loss": 1.4733,
      "step": 900
    },
    {
      "epoch": 3.937007874015748,
      "grad_norm": 1.2465462684631348,
      "learning_rate": 9.10236220472441e-06,
      "loss": 1.4867,
      "step": 1000
    },
    {
      "epoch": 4.0,
      "eval_gen_len": 16.0809,
      "eval_loss": 1.6433463096618652,
      "eval_rouge1": 47.2264,
      "eval_rouge2": 25.4119,
      "eval_rougeL": 41.1461,
      "eval_rougeLsum": 41.1921,
      "eval_runtime": 26.8914,
      "eval_samples_per_second": 37.707,
      "eval_steps_per_second": 1.19,
      "step": 1016
    },
    {
      "epoch": 4.330708661417323,
      "grad_norm": 1.1830518245697021,
      "learning_rate": 5.952755905511811e-06,
      "loss": 1.521,
      "step": 1100
    },
    {
      "epoch": 4.724409448818898,
      "grad_norm": 0.4785144329071045,
      "learning_rate": 2.8661417322834652e-06,
      "loss": 1.4997,
      "step": 1200
    },
    {
      "epoch": 5.0,
      "eval_gen_len": 16.0592,
      "eval_loss": 1.6620988845825195,
      "eval_rouge1": 46.3446,
      "eval_rouge2": 24.6426,
      "eval_rougeL": 39.9898,
      "eval_rougeLsum": 40.0317,
      "eval_runtime": 26.8012,
      "eval_samples_per_second": 37.834,
      "eval_steps_per_second": 1.194,
      "step": 1270
    }
  ],
  "logging_steps": 100,
  "max_steps": 1270,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 5,
  "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": 2.4723690356736e+16,
  "train_batch_size": 32,
  "trial_name": null,
  "trial_params": null
}