trainer_state.json

{
  "best_global_step": 264,
  "best_metric": 0.4479246437549591,
  "best_model_checkpoint": "distilbert_rand_10_v2_sst2/checkpoint-264",
  "epoch": 6.0,
  "eval_steps": 500,
  "global_step": 1584,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "grad_norm": 7.783900737762451,
      "learning_rate": 4.9e-05,
      "loss": 0.3879,
      "step": 264
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.8119266055045872,
      "eval_loss": 0.4479246437549591,
      "eval_runtime": 0.3757,
      "eval_samples_per_second": 2320.758,
      "eval_steps_per_second": 10.646,
      "step": 264
    },
    {
      "epoch": 2.0,
      "grad_norm": 7.653947830200195,
      "learning_rate": 4.8e-05,
      "loss": 0.2214,
      "step": 528
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8061926605504587,
      "eval_loss": 0.4917372465133667,
      "eval_runtime": 0.3669,
      "eval_samples_per_second": 2376.727,
      "eval_steps_per_second": 10.902,
      "step": 528
    },
    {
      "epoch": 3.0,
      "grad_norm": 3.9446558952331543,
      "learning_rate": 4.7e-05,
      "loss": 0.1641,
      "step": 792
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.7970183486238532,
      "eval_loss": 0.5171812772750854,
      "eval_runtime": 0.3827,
      "eval_samples_per_second": 2278.251,
      "eval_steps_per_second": 10.451,
      "step": 792
    },
    {
      "epoch": 4.0,
      "grad_norm": 5.524185657501221,
      "learning_rate": 4.600000000000001e-05,
      "loss": 0.1248,
      "step": 1056
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8222477064220184,
      "eval_loss": 0.595367968082428,
      "eval_runtime": 0.3803,
      "eval_samples_per_second": 2293.031,
      "eval_steps_per_second": 10.518,
      "step": 1056
    },
    {
      "epoch": 5.0,
      "grad_norm": 6.436934471130371,
      "learning_rate": 4.5e-05,
      "loss": 0.094,
      "step": 1320
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8061926605504587,
      "eval_loss": 0.6349753737449646,
      "eval_runtime": 0.3707,
      "eval_samples_per_second": 2352.097,
      "eval_steps_per_second": 10.789,
      "step": 1320
    },
    {
      "epoch": 6.0,
      "grad_norm": 3.756779670715332,
      "learning_rate": 4.4000000000000006e-05,
      "loss": 0.0752,
      "step": 1584
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.7958715596330275,
      "eval_loss": 0.7473843097686768,
      "eval_runtime": 0.3627,
      "eval_samples_per_second": 2404.442,
      "eval_steps_per_second": 11.03,
      "step": 1584
    },
    {
      "epoch": 6.0,
      "step": 1584,
      "total_flos": 2.676464049624883e+16,
      "train_loss": 0.17787643394084893,
      "train_runtime": 377.4144,
      "train_samples_per_second": 8922.42,
      "train_steps_per_second": 34.975
    }
  ],
  "logging_steps": 1,
  "max_steps": 13200,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 50,
  "save_steps": 500,
  "stateful_callbacks": {
    "EarlyStoppingCallback": {
      "args": {
        "early_stopping_patience": 5,
        "early_stopping_threshold": 0.0
      },
      "attributes": {
        "early_stopping_patience_counter": 5
      }
    },
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 2.676464049624883e+16,
  "train_batch_size": 256,
  "trial_name": null,
  "trial_params": null
}