trainer_state.json

{
  "best_metric": 0.6954379081726074,
  "best_model_checkpoint": "bert_tiny_lda_100_v1_sst2/checkpoint-1056",
  "epoch": 9.0,
  "eval_steps": 500,
  "global_step": 2376,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "grad_norm": 1.3844283819198608,
      "learning_rate": 0.00098,
      "loss": 0.6923,
      "step": 264
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.5091743119266054,
      "eval_loss": 0.6971402168273926,
      "eval_runtime": 0.2617,
      "eval_samples_per_second": 3332.644,
      "eval_steps_per_second": 15.287,
      "step": 264
    },
    {
      "epoch": 2.0,
      "grad_norm": 0.17217020690441132,
      "learning_rate": 0.00096,
      "loss": 0.6871,
      "step": 528
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.5091743119266054,
      "eval_loss": 0.700759768486023,
      "eval_runtime": 0.271,
      "eval_samples_per_second": 3217.292,
      "eval_steps_per_second": 14.758,
      "step": 528
    },
    {
      "epoch": 3.0,
      "grad_norm": 0.30297428369522095,
      "learning_rate": 0.00094,
      "loss": 0.6868,
      "step": 792
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.5091743119266054,
      "eval_loss": 0.6972835659980774,
      "eval_runtime": 0.2641,
      "eval_samples_per_second": 3301.314,
      "eval_steps_per_second": 15.144,
      "step": 792
    },
    {
      "epoch": 4.0,
      "grad_norm": 0.40916410088539124,
      "learning_rate": 0.00092,
      "loss": 0.6868,
      "step": 1056
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.5091743119266054,
      "eval_loss": 0.6954379081726074,
      "eval_runtime": 0.2711,
      "eval_samples_per_second": 3216.861,
      "eval_steps_per_second": 14.756,
      "step": 1056
    },
    {
      "epoch": 5.0,
      "grad_norm": 0.10967253148555756,
      "learning_rate": 0.0009000000000000001,
      "loss": 0.6867,
      "step": 1320
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.5091743119266054,
      "eval_loss": 0.6955096125602722,
      "eval_runtime": 0.3123,
      "eval_samples_per_second": 2791.787,
      "eval_steps_per_second": 12.806,
      "step": 1320
    },
    {
      "epoch": 6.0,
      "grad_norm": 0.08872182667255402,
      "learning_rate": 0.00088,
      "loss": 0.6866,
      "step": 1584
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.5091743119266054,
      "eval_loss": 0.6971402168273926,
      "eval_runtime": 0.2648,
      "eval_samples_per_second": 3292.65,
      "eval_steps_per_second": 15.104,
      "step": 1584
    },
    {
      "epoch": 7.0,
      "grad_norm": 0.1451181173324585,
      "learning_rate": 0.00086,
      "loss": 0.6865,
      "step": 1848
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.5091743119266054,
      "eval_loss": 0.698842465877533,
      "eval_runtime": 0.2611,
      "eval_samples_per_second": 3339.088,
      "eval_steps_per_second": 15.317,
      "step": 1848
    },
    {
      "epoch": 8.0,
      "grad_norm": 0.049795445054769516,
      "learning_rate": 0.00084,
      "loss": 0.6865,
      "step": 2112
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.5091743119266054,
      "eval_loss": 0.6989858150482178,
      "eval_runtime": 0.2626,
      "eval_samples_per_second": 3320.526,
      "eval_steps_per_second": 15.232,
      "step": 2112
    },
    {
      "epoch": 9.0,
      "grad_norm": 0.18644928932189941,
      "learning_rate": 0.00082,
      "loss": 0.6866,
      "step": 2376
    },
    {
      "epoch": 9.0,
      "eval_accuracy": 0.5091743119266054,
      "eval_loss": 0.6969968676567078,
      "eval_runtime": 0.2637,
      "eval_samples_per_second": 3306.531,
      "eval_steps_per_second": 15.168,
      "step": 2376
    },
    {
      "epoch": 9.0,
      "step": 2376,
      "total_flos": 1.5895185860385792e+16,
      "train_loss": 0.6873250633779199,
      "train_runtime": 324.1575,
      "train_samples_per_second": 10388.315,
      "train_steps_per_second": 40.721
    }
  ],
  "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": 1.5895185860385792e+16,
  "train_batch_size": 256,
  "trial_name": null,
  "trial_params": null
}