trainer_state.json

{
  "best_global_step": 264,
  "best_metric": 0.4740709066390991,
  "best_model_checkpoint": "distilbert_km_50_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": 6.5697431564331055,
      "learning_rate": 4.9e-05,
      "loss": 0.4425,
      "step": 264
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.7935779816513762,
      "eval_loss": 0.4740709066390991,
      "eval_runtime": 0.3931,
      "eval_samples_per_second": 2218.104,
      "eval_steps_per_second": 10.175,
      "step": 264
    },
    {
      "epoch": 2.0,
      "grad_norm": 6.0281805992126465,
      "learning_rate": 4.8e-05,
      "loss": 0.2392,
      "step": 528
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8027522935779816,
      "eval_loss": 0.501961886882782,
      "eval_runtime": 0.3628,
      "eval_samples_per_second": 2403.216,
      "eval_steps_per_second": 11.024,
      "step": 528
    },
    {
      "epoch": 3.0,
      "grad_norm": 3.6847054958343506,
      "learning_rate": 4.7e-05,
      "loss": 0.1775,
      "step": 792
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.7981651376146789,
      "eval_loss": 0.5157281160354614,
      "eval_runtime": 0.3694,
      "eval_samples_per_second": 2360.787,
      "eval_steps_per_second": 10.829,
      "step": 792
    },
    {
      "epoch": 4.0,
      "grad_norm": 6.764742374420166,
      "learning_rate": 4.600000000000001e-05,
      "loss": 0.1405,
      "step": 1056
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.7912844036697247,
      "eval_loss": 0.6457065939903259,
      "eval_runtime": 0.3977,
      "eval_samples_per_second": 2192.648,
      "eval_steps_per_second": 10.058,
      "step": 1056
    },
    {
      "epoch": 5.0,
      "grad_norm": 8.774916648864746,
      "learning_rate": 4.5e-05,
      "loss": 0.1102,
      "step": 1320
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.7901376146788991,
      "eval_loss": 0.6386711597442627,
      "eval_runtime": 0.3679,
      "eval_samples_per_second": 2370.413,
      "eval_steps_per_second": 10.873,
      "step": 1320
    },
    {
      "epoch": 6.0,
      "grad_norm": 1.2579424381256104,
      "learning_rate": 4.4000000000000006e-05,
      "loss": 0.0865,
      "step": 1584
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.7958715596330275,
      "eval_loss": 0.7057285308837891,
      "eval_runtime": 0.3702,
      "eval_samples_per_second": 2355.584,
      "eval_steps_per_second": 10.805,
      "step": 1584
    },
    {
      "epoch": 6.0,
      "step": 1584,
      "total_flos": 2.676464049624883e+16,
      "train_loss": 0.19940086928280917,
      "train_runtime": 379.6413,
      "train_samples_per_second": 8870.084,
      "train_steps_per_second": 34.77
    }
  ],
  "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
}