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": 0.1174,
      "eval_gen_len": 19.0,
      "eval_loss": 1.0150784254074097,
      "eval_runtime": 39.1976,
      "eval_samples_per_second": 20.282,
      "eval_steps_per_second": 1.276,
      "step": 398
    },
    {
      "epoch": 1.26,
      "learning_rate": 4.3718592964824124e-05,
      "loss": 1.4581,
      "step": 500
    },
    {
      "epoch": 2.0,
      "eval_bleu": 0.1875,
      "eval_gen_len": 19.0,
      "eval_loss": 0.8913741707801819,
      "eval_runtime": 38.8632,
      "eval_samples_per_second": 20.456,
      "eval_steps_per_second": 1.287,
      "step": 796
    },
    {
      "epoch": 2.51,
      "learning_rate": 3.7437185929648245e-05,
      "loss": 1.0266,
      "step": 1000
    },
    {
      "epoch": 3.0,
      "eval_bleu": 0.192,
      "eval_gen_len": 19.0,
      "eval_loss": 0.8158484101295471,
      "eval_runtime": 39.4544,
      "eval_samples_per_second": 20.15,
      "eval_steps_per_second": 1.267,
      "step": 1194
    },
    {
      "epoch": 3.77,
      "learning_rate": 3.1155778894472366e-05,
      "loss": 0.9168,
      "step": 1500
    },
    {
      "epoch": 4.0,
      "eval_bleu": 0.2543,
      "eval_gen_len": 19.0,
      "eval_loss": 0.7705791592597961,
      "eval_runtime": 39.6994,
      "eval_samples_per_second": 20.025,
      "eval_steps_per_second": 1.259,
      "step": 1592
    },
    {
      "epoch": 5.0,
      "eval_bleu": 0.3107,
      "eval_gen_len": 19.0,
      "eval_loss": 0.743384838104248,
      "eval_runtime": 39.1506,
      "eval_samples_per_second": 20.306,
      "eval_steps_per_second": 1.277,
      "step": 1990
    },
    {
      "epoch": 5.03,
      "learning_rate": 2.4874371859296484e-05,
      "loss": 0.857,
      "step": 2000
    },
    {
      "epoch": 6.0,
      "eval_bleu": 0.2904,
      "eval_gen_len": 19.0,
      "eval_loss": 0.7259423136711121,
      "eval_runtime": 39.0753,
      "eval_samples_per_second": 20.345,
      "eval_steps_per_second": 1.28,
      "step": 2388
    },
    {
      "epoch": 6.28,
      "learning_rate": 1.8592964824120602e-05,
      "loss": 0.8163,
      "step": 2500
    },
    {
      "epoch": 7.0,
      "eval_bleu": 0.308,
      "eval_gen_len": 19.0,
      "eval_loss": 0.7096051573753357,
      "eval_runtime": 38.9862,
      "eval_samples_per_second": 20.392,
      "eval_steps_per_second": 1.283,
      "step": 2786
    },
    {
      "epoch": 7.54,
      "learning_rate": 1.2311557788944725e-05,
      "loss": 0.7885,
      "step": 3000
    },
    {
      "epoch": 8.0,
      "eval_bleu": 0.2693,
      "eval_gen_len": 19.0,
      "eval_loss": 0.6969049572944641,
      "eval_runtime": 38.964,
      "eval_samples_per_second": 20.403,
      "eval_steps_per_second": 1.283,
      "step": 3184
    },
    {
      "epoch": 8.79,
      "learning_rate": 6.030150753768844e-06,
      "loss": 0.7718,
      "step": 3500
    },
    {
      "epoch": 9.0,
      "eval_bleu": 0.3197,
      "eval_gen_len": 19.0,
      "eval_loss": 0.693242073059082,
      "eval_runtime": 39.3887,
      "eval_samples_per_second": 20.183,
      "eval_steps_per_second": 1.269,
      "step": 3582
    },
    {
      "epoch": 10.0,
      "eval_bleu": 0.2997,
      "eval_gen_len": 19.0,
      "eval_loss": 0.6918787360191345,
      "eval_runtime": 38.8658,
      "eval_samples_per_second": 20.455,
      "eval_steps_per_second": 1.286,
      "step": 3980
    },
    {
      "epoch": 10.0,
      "step": 3980,
      "total_flos": 2.260259355820032e+16,
      "train_loss": 0.9253955016783134,
      "train_runtime": 2104.4642,
      "train_samples_per_second": 30.202,
      "train_steps_per_second": 1.891
    }
  ],
  "max_steps": 3980,
  "num_train_epochs": 10,
  "total_flos": 2.260259355820032e+16,
  "trial_name": null,
  "trial_params": null
}