rte/roberta-large_lr1e-05/trainer_state.json

{
  "best_metric": 0.8072289156626506,
  "best_model_checkpoint": "./nlu_finetuned_models/rte/roberta-large_lr1e-05/checkpoint-987",
  "epoch": 10.0,
  "eval_steps": 500,
  "global_step": 1410,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "eval_accuracy": 0.46184738955823296,
      "eval_loss": 0.7056757211685181,
      "eval_runtime": 0.6863,
      "eval_samples_per_second": 362.829,
      "eval_steps_per_second": 23.314,
      "step": 141
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.7228915662650602,
      "eval_loss": 0.6255487203598022,
      "eval_runtime": 0.6852,
      "eval_samples_per_second": 363.375,
      "eval_steps_per_second": 23.349,
      "step": 282
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.7791164658634538,
      "eval_loss": 0.5178215503692627,
      "eval_runtime": 0.6747,
      "eval_samples_per_second": 369.072,
      "eval_steps_per_second": 23.715,
      "step": 423
    },
    {
      "epoch": 3.546099290780142,
      "grad_norm": 29.539140701293945,
      "learning_rate": 6.867924528301887e-06,
      "loss": 0.553,
      "step": 500
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.7871485943775101,
      "eval_loss": 0.7929865717887878,
      "eval_runtime": 0.6757,
      "eval_samples_per_second": 368.52,
      "eval_steps_per_second": 23.68,
      "step": 564
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.7751004016064257,
      "eval_loss": 1.1344404220581055,
      "eval_runtime": 0.6864,
      "eval_samples_per_second": 362.743,
      "eval_steps_per_second": 23.309,
      "step": 705
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.7751004016064257,
      "eval_loss": 1.3329523801803589,
      "eval_runtime": 0.6828,
      "eval_samples_per_second": 364.657,
      "eval_steps_per_second": 23.432,
      "step": 846
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.8072289156626506,
      "eval_loss": 1.3685961961746216,
      "eval_runtime": 0.6862,
      "eval_samples_per_second": 362.852,
      "eval_steps_per_second": 23.316,
      "step": 987
    },
    {
      "epoch": 7.092198581560283,
      "grad_norm": 0.07306408137083054,
      "learning_rate": 3.0943396226415094e-06,
      "loss": 0.1419,
      "step": 1000
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.7831325301204819,
      "eval_loss": 1.7352402210235596,
      "eval_runtime": 0.6799,
      "eval_samples_per_second": 366.211,
      "eval_steps_per_second": 23.532,
      "step": 1128
    },
    {
      "epoch": 9.0,
      "eval_accuracy": 0.7911646586345381,
      "eval_loss": 1.6394472122192383,
      "eval_runtime": 0.6809,
      "eval_samples_per_second": 365.676,
      "eval_steps_per_second": 23.497,
      "step": 1269
    },
    {
      "epoch": 10.0,
      "eval_accuracy": 0.7871485943775101,
      "eval_loss": 1.6835201978683472,
      "eval_runtime": 0.6531,
      "eval_samples_per_second": 381.265,
      "eval_steps_per_second": 24.499,
      "step": 1410
    },
    {
      "epoch": 10.0,
      "step": 1410,
      "total_flos": 5134488587504724.0,
      "train_loss": 0.2575254629689751,
      "train_runtime": 283.0673,
      "train_samples_per_second": 79.168,
      "train_steps_per_second": 4.981
    }
  ],
  "logging_steps": 500,
  "max_steps": 1410,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 10,
  "save_steps": 500,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 5134488587504724.0,
  "train_batch_size": 16,
  "trial_name": null,
  "trial_params": null
}