rte/roberta-base_lr1e-05/trainer_state.json

{
  "best_metric": 0.7550200803212851,
  "best_model_checkpoint": "./nlu_finetuned_models/rte/roberta-base_lr1e-05/checkpoint-1269",
  "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": 0.46987951807228917,
      "eval_loss": 0.6921471357345581,
      "eval_runtime": 0.2588,
      "eval_samples_per_second": 962.195,
      "eval_steps_per_second": 61.828,
      "step": 141
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.6265060240963856,
      "eval_loss": 0.6887008547782898,
      "eval_runtime": 0.2555,
      "eval_samples_per_second": 974.448,
      "eval_steps_per_second": 62.615,
      "step": 282
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.7309236947791165,
      "eval_loss": 0.5638577342033386,
      "eval_runtime": 0.2565,
      "eval_samples_per_second": 970.91,
      "eval_steps_per_second": 62.388,
      "step": 423
    },
    {
      "epoch": 3.546099290780142,
      "grad_norm": 70.71211242675781,
      "learning_rate": 6.867924528301887e-06,
      "loss": 0.5832,
      "step": 500
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.7108433734939759,
      "eval_loss": 0.6645846366882324,
      "eval_runtime": 0.2556,
      "eval_samples_per_second": 974.014,
      "eval_steps_per_second": 62.587,
      "step": 564
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.7228915662650602,
      "eval_loss": 0.8847337365150452,
      "eval_runtime": 0.2566,
      "eval_samples_per_second": 970.404,
      "eval_steps_per_second": 62.355,
      "step": 705
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.7389558232931727,
      "eval_loss": 0.8534902334213257,
      "eval_runtime": 0.252,
      "eval_samples_per_second": 988.035,
      "eval_steps_per_second": 63.488,
      "step": 846
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.7389558232931727,
      "eval_loss": 1.0695092678070068,
      "eval_runtime": 0.2556,
      "eval_samples_per_second": 974.276,
      "eval_steps_per_second": 62.604,
      "step": 987
    },
    {
      "epoch": 7.092198581560283,
      "grad_norm": 8.739114761352539,
      "learning_rate": 3.0943396226415094e-06,
      "loss": 0.2844,
      "step": 1000
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.7108433734939759,
      "eval_loss": 1.4401593208312988,
      "eval_runtime": 0.2559,
      "eval_samples_per_second": 972.889,
      "eval_steps_per_second": 62.515,
      "step": 1128
    },
    {
      "epoch": 9.0,
      "eval_accuracy": 0.7550200803212851,
      "eval_loss": 1.3305011987686157,
      "eval_runtime": 0.2532,
      "eval_samples_per_second": 983.482,
      "eval_steps_per_second": 63.196,
      "step": 1269
    },
    {
      "epoch": 10.0,
      "eval_accuracy": 0.7309236947791165,
      "eval_loss": 1.4962958097457886,
      "eval_runtime": 0.2435,
      "eval_samples_per_second": 1022.502,
      "eval_steps_per_second": 65.703,
      "step": 1410
    },
    {
      "epoch": 10.0,
      "step": 1410,
      "total_flos": 1449613956727380.0,
      "train_loss": 0.348689207793973,
      "train_runtime": 119.2101,
      "train_samples_per_second": 187.987,
      "train_steps_per_second": 11.828
    }
  ],
  "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": 1449613956727380.0,
  "train_batch_size": 16,
  "trial_name": null,
  "trial_params": null
}