trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 10.0,
  "global_step": 3930,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "eval_bleu": 0.0452,
      "eval_gen_len": 19.0,
      "eval_loss": 1.0331772565841675,
      "eval_runtime": 41.1909,
      "eval_samples_per_second": 19.058,
      "eval_steps_per_second": 1.214,
      "step": 393
    },
    {
      "epoch": 1.27,
      "learning_rate": 4.3638676844783716e-05,
      "loss": 1.5229,
      "step": 500
    },
    {
      "epoch": 2.0,
      "eval_bleu": 0.1159,
      "eval_gen_len": 19.0,
      "eval_loss": 0.912860095500946,
      "eval_runtime": 38.6643,
      "eval_samples_per_second": 20.303,
      "eval_steps_per_second": 1.293,
      "step": 786
    },
    {
      "epoch": 2.54,
      "learning_rate": 3.727735368956743e-05,
      "loss": 1.0464,
      "step": 1000
    },
    {
      "epoch": 3.0,
      "eval_bleu": 0.1479,
      "eval_gen_len": 19.0,
      "eval_loss": 0.8561409115791321,
      "eval_runtime": 38.5663,
      "eval_samples_per_second": 20.355,
      "eval_steps_per_second": 1.296,
      "step": 1179
    },
    {
      "epoch": 3.82,
      "learning_rate": 3.091603053435115e-05,
      "loss": 0.9477,
      "step": 1500
    },
    {
      "epoch": 4.0,
      "eval_bleu": 0.1678,
      "eval_gen_len": 19.0,
      "eval_loss": 0.8148155808448792,
      "eval_runtime": 38.0933,
      "eval_samples_per_second": 20.607,
      "eval_steps_per_second": 1.313,
      "step": 1572
    },
    {
      "epoch": 5.0,
      "eval_bleu": 0.1791,
      "eval_gen_len": 19.0,
      "eval_loss": 0.7879176735877991,
      "eval_runtime": 38.3603,
      "eval_samples_per_second": 20.464,
      "eval_steps_per_second": 1.303,
      "step": 1965
    },
    {
      "epoch": 5.09,
      "learning_rate": 2.455470737913486e-05,
      "loss": 0.8888,
      "step": 2000
    },
    {
      "epoch": 6.0,
      "eval_bleu": 0.1628,
      "eval_gen_len": 19.0,
      "eval_loss": 0.7707892656326294,
      "eval_runtime": 38.214,
      "eval_samples_per_second": 20.542,
      "eval_steps_per_second": 1.308,
      "step": 2358
    },
    {
      "epoch": 6.36,
      "learning_rate": 1.8193384223918574e-05,
      "loss": 0.8517,
      "step": 2500
    },
    {
      "epoch": 7.0,
      "eval_bleu": 0.1582,
      "eval_gen_len": 19.0,
      "eval_loss": 0.7597165107727051,
      "eval_runtime": 38.4006,
      "eval_samples_per_second": 20.442,
      "eval_steps_per_second": 1.302,
      "step": 2751
    },
    {
      "epoch": 7.63,
      "learning_rate": 1.1832061068702292e-05,
      "loss": 0.8279,
      "step": 3000
    },
    {
      "epoch": 8.0,
      "eval_bleu": 0.1683,
      "eval_gen_len": 19.0,
      "eval_loss": 0.7492462992668152,
      "eval_runtime": 38.3938,
      "eval_samples_per_second": 20.446,
      "eval_steps_per_second": 1.302,
      "step": 3144
    },
    {
      "epoch": 8.91,
      "learning_rate": 5.470737913486006e-06,
      "loss": 0.8107,
      "step": 3500
    },
    {
      "epoch": 9.0,
      "eval_bleu": 0.1704,
      "eval_gen_len": 19.0,
      "eval_loss": 0.7442331314086914,
      "eval_runtime": 38.261,
      "eval_samples_per_second": 20.517,
      "eval_steps_per_second": 1.307,
      "step": 3537
    },
    {
      "epoch": 10.0,
      "eval_bleu": 0.1683,
      "eval_gen_len": 19.0,
      "eval_loss": 0.7417959570884705,
      "eval_runtime": 38.3788,
      "eval_samples_per_second": 20.454,
      "eval_steps_per_second": 1.303,
      "step": 3930
    },
    {
      "epoch": 10.0,
      "step": 3930,
      "total_flos": 2.232877359218688e+16,
      "train_loss": 0.9653779240964934,
      "train_runtime": 2075.6207,
      "train_samples_per_second": 30.251,
      "train_steps_per_second": 1.893
    }
  ],
  "max_steps": 3930,
  "num_train_epochs": 10,
  "total_flos": 2.232877359218688e+16,
  "trial_name": null,
  "trial_params": null
}