trainer_state.json

{
  "best_metric": 0.6924921870231628,
  "best_model_checkpoint": "hBERTv1_new_pretrain_w_init__rte/checkpoint-40",
  "epoch": 7.0,
  "global_step": 140,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "learning_rate": 0.00049,
      "loss": 5.3184,
      "step": 20
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.4729241877256318,
      "eval_loss": 2.3444929122924805,
      "eval_runtime": 0.5056,
      "eval_samples_per_second": 547.866,
      "eval_steps_per_second": 5.934,
      "step": 20
    },
    {
      "epoch": 2.0,
      "learning_rate": 0.00048,
      "loss": 1.0827,
      "step": 40
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.5270758122743683,
      "eval_loss": 0.6924921870231628,
      "eval_runtime": 0.5111,
      "eval_samples_per_second": 541.947,
      "eval_steps_per_second": 5.869,
      "step": 40
    },
    {
      "epoch": 3.0,
      "learning_rate": 0.00047,
      "loss": 0.7099,
      "step": 60
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.5270758122743683,
      "eval_loss": 0.697836697101593,
      "eval_runtime": 0.5089,
      "eval_samples_per_second": 544.326,
      "eval_steps_per_second": 5.895,
      "step": 60
    },
    {
      "epoch": 4.0,
      "learning_rate": 0.00046,
      "loss": 0.7064,
      "step": 80
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.4729241877256318,
      "eval_loss": 0.6937753558158875,
      "eval_runtime": 0.507,
      "eval_samples_per_second": 546.375,
      "eval_steps_per_second": 5.917,
      "step": 80
    },
    {
      "epoch": 5.0,
      "learning_rate": 0.00045000000000000004,
      "loss": 0.7008,
      "step": 100
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.5270758122743683,
      "eval_loss": 0.6991623640060425,
      "eval_runtime": 0.5084,
      "eval_samples_per_second": 544.873,
      "eval_steps_per_second": 5.901,
      "step": 100
    },
    {
      "epoch": 6.0,
      "learning_rate": 0.00044,
      "loss": 0.7007,
      "step": 120
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.4729241877256318,
      "eval_loss": 0.7100631594657898,
      "eval_runtime": 0.5105,
      "eval_samples_per_second": 542.584,
      "eval_steps_per_second": 5.876,
      "step": 120
    },
    {
      "epoch": 7.0,
      "learning_rate": 0.00043,
      "loss": 0.696,
      "step": 140
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.4729241877256318,
      "eval_loss": 0.699120044708252,
      "eval_runtime": 0.5094,
      "eval_samples_per_second": 543.815,
      "eval_steps_per_second": 5.89,
      "step": 140
    },
    {
      "epoch": 7.0,
      "step": 140,
      "total_flos": 2581326716731392.0,
      "train_loss": 1.4164232867104667,
      "train_runtime": 117.9285,
      "train_samples_per_second": 1055.725,
      "train_steps_per_second": 8.48
    }
  ],
  "max_steps": 1000,
  "num_train_epochs": 50,
  "total_flos": 2581326716731392.0,
  "trial_name": null,
  "trial_params": null
}