trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 10.0,
  "global_step": 3980,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "eval_bleu": 0.0,
      "eval_gen_len": 19.0,
      "eval_loss": 1.1777664422988892,
      "eval_runtime": 38.2539,
      "eval_samples_per_second": 20.782,
      "eval_steps_per_second": 1.307,
      "step": 398
    },
    {
      "epoch": 1.26,
      "learning_rate": 4.3718592964824124e-05,
      "loss": 1.6138,
      "step": 500
    },
    {
      "epoch": 2.0,
      "eval_bleu": 0.0729,
      "eval_gen_len": 19.0,
      "eval_loss": 1.050442099571228,
      "eval_runtime": 38.3177,
      "eval_samples_per_second": 20.748,
      "eval_steps_per_second": 1.305,
      "step": 796
    },
    {
      "epoch": 2.51,
      "learning_rate": 3.7437185929648245e-05,
      "loss": 1.1972,
      "step": 1000
    },
    {
      "epoch": 3.0,
      "eval_bleu": 0.0547,
      "eval_gen_len": 19.0,
      "eval_loss": 0.9887745380401611,
      "eval_runtime": 38.6368,
      "eval_samples_per_second": 20.576,
      "eval_steps_per_second": 1.294,
      "step": 1194
    },
    {
      "epoch": 3.77,
      "learning_rate": 3.1155778894472366e-05,
      "loss": 1.0888,
      "step": 1500
    },
    {
      "epoch": 4.0,
      "eval_bleu": 0.0841,
      "eval_gen_len": 19.0,
      "eval_loss": 0.9423935413360596,
      "eval_runtime": 38.2901,
      "eval_samples_per_second": 20.763,
      "eval_steps_per_second": 1.306,
      "step": 1592
    },
    {
      "epoch": 5.0,
      "eval_bleu": 0.0757,
      "eval_gen_len": 19.0,
      "eval_loss": 0.907833456993103,
      "eval_runtime": 38.2162,
      "eval_samples_per_second": 20.803,
      "eval_steps_per_second": 1.308,
      "step": 1990
    },
    {
      "epoch": 5.03,
      "learning_rate": 2.4874371859296484e-05,
      "loss": 1.0299,
      "step": 2000
    },
    {
      "epoch": 6.0,
      "eval_bleu": 0.0784,
      "eval_gen_len": 19.0,
      "eval_loss": 0.8801307082176208,
      "eval_runtime": 38.4667,
      "eval_samples_per_second": 20.667,
      "eval_steps_per_second": 1.3,
      "step": 2388
    },
    {
      "epoch": 6.28,
      "learning_rate": 1.8592964824120602e-05,
      "loss": 0.9817,
      "step": 2500
    },
    {
      "epoch": 7.0,
      "eval_bleu": 0.0764,
      "eval_gen_len": 19.0,
      "eval_loss": 0.8627172112464905,
      "eval_runtime": 38.3001,
      "eval_samples_per_second": 20.757,
      "eval_steps_per_second": 1.305,
      "step": 2786
    },
    {
      "epoch": 7.54,
      "learning_rate": 1.2311557788944725e-05,
      "loss": 0.9466,
      "step": 3000
    },
    {
      "epoch": 8.0,
      "eval_bleu": 0.0784,
      "eval_gen_len": 19.0,
      "eval_loss": 0.8503021597862244,
      "eval_runtime": 38.3277,
      "eval_samples_per_second": 20.742,
      "eval_steps_per_second": 1.305,
      "step": 3184
    },
    {
      "epoch": 8.79,
      "learning_rate": 6.030150753768844e-06,
      "loss": 0.9286,
      "step": 3500
    },
    {
      "epoch": 9.0,
      "eval_bleu": 0.0781,
      "eval_gen_len": 19.0,
      "eval_loss": 0.8432070016860962,
      "eval_runtime": 38.4246,
      "eval_samples_per_second": 20.69,
      "eval_steps_per_second": 1.301,
      "step": 3582
    },
    {
      "epoch": 10.0,
      "eval_bleu": 0.0773,
      "eval_gen_len": 19.0,
      "eval_loss": 0.8411539793014526,
      "eval_runtime": 38.1433,
      "eval_samples_per_second": 20.842,
      "eval_steps_per_second": 1.311,
      "step": 3980
    },
    {
      "epoch": 10.0,
      "step": 3980,
      "total_flos": 2.262748628238336e+16,
      "train_loss": 1.0886857229261542,
      "train_runtime": 2095.3725,
      "train_samples_per_second": 30.367,
      "train_steps_per_second": 1.899
    }
  ],
  "max_steps": 3980,
  "num_train_epochs": 10,
  "total_flos": 2.262748628238336e+16,
  "trial_name": null,
  "trial_params": null
}