trainer_state.json

{
  "best_metric": 1.0991482734680176,
  "best_model_checkpoint": "hBERTv2_new_pretrain_mnli/checkpoint-9204",
  "epoch": 8.0,
  "global_step": 24544,
  "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": 70.7498,
      "step": 3068
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.3273560876209883,
      "eval_loss": 30.0480899810791,
      "eval_runtime": 16.8669,
      "eval_samples_per_second": 581.909,
      "eval_steps_per_second": 4.565,
      "step": 3068
    },
    {
      "epoch": 2.0,
      "learning_rate": 0.00048,
      "loss": 5.5287,
      "step": 6136
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.3273560876209883,
      "eval_loss": 1.1252200603485107,
      "eval_runtime": 17.0029,
      "eval_samples_per_second": 577.254,
      "eval_steps_per_second": 4.529,
      "step": 6136
    },
    {
      "epoch": 3.0,
      "learning_rate": 0.00047,
      "loss": 1.354,
      "step": 9204
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.31818644931227713,
      "eval_loss": 1.0991482734680176,
      "eval_runtime": 16.9792,
      "eval_samples_per_second": 578.06,
      "eval_steps_per_second": 4.535,
      "step": 9204
    },
    {
      "epoch": 4.0,
      "learning_rate": 0.00046,
      "loss": 3.0766,
      "step": 12272
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.31818644931227713,
      "eval_loss": 1.2373435497283936,
      "eval_runtime": 16.9102,
      "eval_samples_per_second": 580.418,
      "eval_steps_per_second": 4.553,
      "step": 12272
    },
    {
      "epoch": 5.0,
      "learning_rate": 0.00045000000000000004,
      "loss": 4.2409,
      "step": 15340
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.3273560876209883,
      "eval_loss": 1.191184163093567,
      "eval_runtime": 16.9866,
      "eval_samples_per_second": 577.809,
      "eval_steps_per_second": 4.533,
      "step": 15340
    },
    {
      "epoch": 6.0,
      "learning_rate": 0.00044,
      "loss": 5.4957,
      "step": 18408
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.3544574630667346,
      "eval_loss": 1.1716059446334839,
      "eval_runtime": 16.9674,
      "eval_samples_per_second": 578.463,
      "eval_steps_per_second": 4.538,
      "step": 18408
    },
    {
      "epoch": 7.0,
      "learning_rate": 0.00043,
      "loss": 3.0327,
      "step": 21476
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.3273560876209883,
      "eval_loss": 1.315549373626709,
      "eval_runtime": 16.8941,
      "eval_samples_per_second": 580.971,
      "eval_steps_per_second": 4.558,
      "step": 21476
    },
    {
      "epoch": 8.0,
      "learning_rate": 0.00042,
      "loss": 1.571,
      "step": 24544
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.3273560876209883,
      "eval_loss": 1.3787474632263184,
      "eval_runtime": 16.9205,
      "eval_samples_per_second": 580.066,
      "eval_steps_per_second": 4.551,
      "step": 24544
    },
    {
      "epoch": 8.0,
      "step": 24544,
      "total_flos": 4.597451878333153e+17,
      "train_loss": 11.881152813251202,
      "train_runtime": 15762.7066,
      "train_samples_per_second": 1245.668,
      "train_steps_per_second": 9.732
    }
  ],
  "max_steps": 153400,
  "num_train_epochs": 50,
  "total_flos": 4.597451878333153e+17,
  "trial_name": null,
  "trial_params": null
}