trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 10.0,
  "global_step": 3950,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "eval_bleu": 0.0232,
      "eval_gen_len": 19.0,
      "eval_loss": 1.0572127103805542,
      "eval_runtime": 39.011,
      "eval_samples_per_second": 20.251,
      "eval_steps_per_second": 1.282,
      "step": 395
    },
    {
      "epoch": 1.27,
      "learning_rate": 4.367088607594937e-05,
      "loss": 1.5347,
      "step": 500
    },
    {
      "epoch": 2.0,
      "eval_bleu": 0.1035,
      "eval_gen_len": 19.0,
      "eval_loss": 0.9248576760292053,
      "eval_runtime": 39.2027,
      "eval_samples_per_second": 20.152,
      "eval_steps_per_second": 1.275,
      "step": 790
    },
    {
      "epoch": 2.53,
      "learning_rate": 3.7341772151898736e-05,
      "loss": 1.0539,
      "step": 1000
    },
    {
      "epoch": 3.0,
      "eval_bleu": 0.1313,
      "eval_gen_len": 19.0,
      "eval_loss": 0.854935884475708,
      "eval_runtime": 39.1519,
      "eval_samples_per_second": 20.178,
      "eval_steps_per_second": 1.277,
      "step": 1185
    },
    {
      "epoch": 3.8,
      "learning_rate": 3.10126582278481e-05,
      "loss": 0.9461,
      "step": 1500
    },
    {
      "epoch": 4.0,
      "eval_bleu": 0.1392,
      "eval_gen_len": 19.0,
      "eval_loss": 0.8109087347984314,
      "eval_runtime": 39.0695,
      "eval_samples_per_second": 20.22,
      "eval_steps_per_second": 1.28,
      "step": 1580
    },
    {
      "epoch": 5.0,
      "eval_bleu": 0.1545,
      "eval_gen_len": 19.0,
      "eval_loss": 0.7860050201416016,
      "eval_runtime": 39.2946,
      "eval_samples_per_second": 20.105,
      "eval_steps_per_second": 1.272,
      "step": 1975
    },
    {
      "epoch": 5.06,
      "learning_rate": 2.468354430379747e-05,
      "loss": 0.8824,
      "step": 2000
    },
    {
      "epoch": 6.0,
      "eval_bleu": 0.1756,
      "eval_gen_len": 19.0,
      "eval_loss": 0.7620316743850708,
      "eval_runtime": 39.148,
      "eval_samples_per_second": 20.18,
      "eval_steps_per_second": 1.277,
      "step": 2370
    },
    {
      "epoch": 6.33,
      "learning_rate": 1.8354430379746836e-05,
      "loss": 0.8447,
      "step": 2500
    },
    {
      "epoch": 7.0,
      "eval_bleu": 0.1705,
      "eval_gen_len": 19.0,
      "eval_loss": 0.7495388984680176,
      "eval_runtime": 39.4044,
      "eval_samples_per_second": 20.049,
      "eval_steps_per_second": 1.269,
      "step": 2765
    },
    {
      "epoch": 7.59,
      "learning_rate": 1.2025316455696203e-05,
      "loss": 0.814,
      "step": 3000
    },
    {
      "epoch": 8.0,
      "eval_bleu": 0.14,
      "eval_gen_len": 19.0,
      "eval_loss": 0.7383524775505066,
      "eval_runtime": 39.2093,
      "eval_samples_per_second": 20.148,
      "eval_steps_per_second": 1.275,
      "step": 3160
    },
    {
      "epoch": 8.86,
      "learning_rate": 5.69620253164557e-06,
      "loss": 0.8031,
      "step": 3500
    },
    {
      "epoch": 9.0,
      "eval_bleu": 0.1807,
      "eval_gen_len": 19.0,
      "eval_loss": 0.7330699563026428,
      "eval_runtime": 39.1087,
      "eval_samples_per_second": 20.2,
      "eval_steps_per_second": 1.278,
      "step": 3555
    },
    {
      "epoch": 10.0,
      "eval_bleu": 0.1841,
      "eval_gen_len": 19.0,
      "eval_loss": 0.7314567565917969,
      "eval_runtime": 39.0954,
      "eval_samples_per_second": 20.207,
      "eval_steps_per_second": 1.279,
      "step": 3950
    },
    {
      "epoch": 10.0,
      "step": 3950,
      "total_flos": 2.246746162692096e+16,
      "train_loss": 0.9606689144086235,
      "train_runtime": 2098.7276,
      "train_samples_per_second": 30.104,
      "train_steps_per_second": 1.882
    }
  ],
  "max_steps": 3950,
  "num_train_epochs": 10,
  "total_flos": 2.246746162692096e+16,
  "trial_name": null,
  "trial_params": null
}