trainer_state.json

{
  "best_global_step": 2750,
  "best_metric": 3.8293724060058594,
  "best_model_checkpoint": "/work/10368/siyuansong/vista/models/opt-clean_plus_vocab_error_5pct_42_1e-3/checkpoint-2750",
  "epoch": 10.0,
  "eval_steps": 500,
  "global_step": 2750,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "eval_loss": 7.02741813659668,
      "eval_runtime": 34.4123,
      "eval_samples_per_second": 1507.106,
      "eval_steps_per_second": 5.899,
      "step": 275
    },
    {
      "epoch": 2.0,
      "eval_loss": 5.511542320251465,
      "eval_runtime": 34.6866,
      "eval_samples_per_second": 1495.186,
      "eval_steps_per_second": 5.852,
      "step": 550
    },
    {
      "epoch": 3.0,
      "eval_loss": 4.973597049713135,
      "eval_runtime": 34.6026,
      "eval_samples_per_second": 1498.818,
      "eval_steps_per_second": 5.867,
      "step": 825
    },
    {
      "epoch": 3.6375227686703098,
      "grad_norm": 0.28458723425865173,
      "learning_rate": 3.121875e-05,
      "loss": 6.3514,
      "step": 1000
    },
    {
      "epoch": 4.0,
      "eval_loss": 4.6719231605529785,
      "eval_runtime": 34.7122,
      "eval_samples_per_second": 1494.084,
      "eval_steps_per_second": 5.848,
      "step": 1100
    },
    {
      "epoch": 5.0,
      "eval_loss": 4.446601867675781,
      "eval_runtime": 34.7852,
      "eval_samples_per_second": 1490.95,
      "eval_steps_per_second": 5.836,
      "step": 1375
    },
    {
      "epoch": 6.0,
      "eval_loss": 4.2719502449035645,
      "eval_runtime": 34.6807,
      "eval_samples_per_second": 1495.443,
      "eval_steps_per_second": 5.853,
      "step": 1650
    },
    {
      "epoch": 7.0,
      "eval_loss": 4.134940147399902,
      "eval_runtime": 34.5514,
      "eval_samples_per_second": 1501.039,
      "eval_steps_per_second": 5.875,
      "step": 1925
    },
    {
      "epoch": 7.273224043715847,
      "grad_norm": 0.487853467464447,
      "learning_rate": 6.246875000000001e-05,
      "loss": 4.412,
      "step": 2000
    },
    {
      "epoch": 8.0,
      "eval_loss": 4.018152713775635,
      "eval_runtime": 34.6739,
      "eval_samples_per_second": 1495.736,
      "eval_steps_per_second": 5.855,
      "step": 2200
    },
    {
      "epoch": 9.0,
      "eval_loss": 3.918316602706909,
      "eval_runtime": 34.8253,
      "eval_samples_per_second": 1489.235,
      "eval_steps_per_second": 5.829,
      "step": 2475
    },
    {
      "epoch": 10.0,
      "eval_loss": 3.8293724060058594,
      "eval_runtime": 34.673,
      "eval_samples_per_second": 1495.776,
      "eval_steps_per_second": 5.855,
      "step": 2750
    },
    {
      "epoch": 10.0,
      "step": 2750,
      "total_flos": 7.3446054690816e+17,
      "train_loss": 5.0002771661931815,
      "train_runtime": 7255.6111,
      "train_samples_per_second": 774.851,
      "train_steps_per_second": 0.379
    }
  ],
  "logging_steps": 1000,
  "max_steps": 2750,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 10,
  "save_steps": 500,
  "stateful_callbacks": {
    "EarlyStoppingCallback": {
      "args": {
        "early_stopping_patience": 3,
        "early_stopping_threshold": 0.0
      },
      "attributes": {
        "early_stopping_patience_counter": 0
      }
    },
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 7.3446054690816e+17,
  "train_batch_size": 256,
  "trial_name": null,
  "trial_params": null
}