trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 10.0,
  "global_step": 3970,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "eval_bleu": 0.0631,
      "eval_gen_len": 19.0,
      "eval_loss": 1.1706956624984741,
      "eval_runtime": 38.634,
      "eval_samples_per_second": 20.552,
      "eval_steps_per_second": 1.294,
      "step": 397
    },
    {
      "epoch": 1.26,
      "learning_rate": 4.370277078085643e-05,
      "loss": 1.6442,
      "step": 500
    },
    {
      "epoch": 2.0,
      "eval_bleu": 0.1214,
      "eval_gen_len": 19.0,
      "eval_loss": 1.0399662256240845,
      "eval_runtime": 38.5465,
      "eval_samples_per_second": 20.599,
      "eval_steps_per_second": 1.297,
      "step": 794
    },
    {
      "epoch": 2.52,
      "learning_rate": 3.7405541561712845e-05,
      "loss": 1.1712,
      "step": 1000
    },
    {
      "epoch": 3.0,
      "eval_bleu": 0.1063,
      "eval_gen_len": 19.0,
      "eval_loss": 0.9846653938293457,
      "eval_runtime": 38.7706,
      "eval_samples_per_second": 20.479,
      "eval_steps_per_second": 1.29,
      "step": 1191
    },
    {
      "epoch": 3.78,
      "learning_rate": 3.1108312342569276e-05,
      "loss": 1.0724,
      "step": 1500
    },
    {
      "epoch": 4.0,
      "eval_bleu": 0.1209,
      "eval_gen_len": 19.0,
      "eval_loss": 0.946679949760437,
      "eval_runtime": 39.0506,
      "eval_samples_per_second": 20.333,
      "eval_steps_per_second": 1.28,
      "step": 1588
    },
    {
      "epoch": 5.0,
      "eval_bleu": 0.1059,
      "eval_gen_len": 19.0,
      "eval_loss": 0.917578399181366,
      "eval_runtime": 39.0497,
      "eval_samples_per_second": 20.333,
      "eval_steps_per_second": 1.28,
      "step": 1985
    },
    {
      "epoch": 5.04,
      "learning_rate": 2.4811083123425694e-05,
      "loss": 1.016,
      "step": 2000
    },
    {
      "epoch": 6.0,
      "eval_bleu": 0.1166,
      "eval_gen_len": 19.0,
      "eval_loss": 0.8976603150367737,
      "eval_runtime": 38.9021,
      "eval_samples_per_second": 20.41,
      "eval_steps_per_second": 1.285,
      "step": 2382
    },
    {
      "epoch": 6.3,
      "learning_rate": 1.8513853904282116e-05,
      "loss": 0.9799,
      "step": 2500
    },
    {
      "epoch": 7.0,
      "eval_bleu": 0.1481,
      "eval_gen_len": 19.0,
      "eval_loss": 0.885191023349762,
      "eval_runtime": 38.7571,
      "eval_samples_per_second": 20.487,
      "eval_steps_per_second": 1.29,
      "step": 2779
    },
    {
      "epoch": 7.56,
      "learning_rate": 1.2216624685138539e-05,
      "loss": 0.9553,
      "step": 3000
    },
    {
      "epoch": 8.0,
      "eval_bleu": 0.1381,
      "eval_gen_len": 19.0,
      "eval_loss": 0.8747804760932922,
      "eval_runtime": 38.765,
      "eval_samples_per_second": 20.482,
      "eval_steps_per_second": 1.29,
      "step": 3176
    },
    {
      "epoch": 8.82,
      "learning_rate": 5.919395465994963e-06,
      "loss": 0.9386,
      "step": 3500
    },
    {
      "epoch": 9.0,
      "eval_bleu": 0.1294,
      "eval_gen_len": 19.0,
      "eval_loss": 0.8715814352035522,
      "eval_runtime": 38.6493,
      "eval_samples_per_second": 20.544,
      "eval_steps_per_second": 1.294,
      "step": 3573
    },
    {
      "epoch": 10.0,
      "eval_bleu": 0.1337,
      "eval_gen_len": 19.0,
      "eval_loss": 0.8693967461585999,
      "eval_runtime": 38.623,
      "eval_samples_per_second": 20.558,
      "eval_steps_per_second": 1.295,
      "step": 3970
    },
    {
      "epoch": 10.0,
      "step": 3970,
      "total_flos": 2.2581256937472e+16,
      "train_loss": 1.0896147682324464,
      "train_runtime": 2100.2334,
      "train_samples_per_second": 30.235,
      "train_steps_per_second": 1.89
    }
  ],
  "max_steps": 3970,
  "num_train_epochs": 10,
  "total_flos": 2.2581256937472e+16,
  "trial_name": null,
  "trial_params": null
}