zhongjing_7_26/trainer_state.json

{
  "best_metric": 1.051451325416565,
  "best_model_checkpoint": "./checkpoints/zhongjing_7_26/checkpoint-100",
  "epoch": 100.0,
  "global_step": 100,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 10.0,
      "learning_rate": 9.755282581475769e-05,
      "loss": 0.803,
      "step": 10
    },
    {
      "epoch": 10.0,
      "eval_loss": 0.6868318915367126,
      "eval_runtime": 0.263,
      "eval_samples_per_second": 26.615,
      "eval_steps_per_second": 3.802,
      "step": 10
    },
    {
      "epoch": 20.0,
      "learning_rate": 9.045084971874738e-05,
      "loss": 0.5757,
      "step": 20
    },
    {
      "epoch": 20.0,
      "eval_loss": 0.5540383458137512,
      "eval_runtime": 0.2628,
      "eval_samples_per_second": 26.639,
      "eval_steps_per_second": 3.806,
      "step": 20
    },
    {
      "epoch": 30.0,
      "learning_rate": 7.938926261462366e-05,
      "loss": 0.3858,
      "step": 30
    },
    {
      "epoch": 30.0,
      "eval_loss": 0.5853456854820251,
      "eval_runtime": 0.2612,
      "eval_samples_per_second": 26.799,
      "eval_steps_per_second": 3.828,
      "step": 30
    },
    {
      "epoch": 40.0,
      "learning_rate": 6.545084971874738e-05,
      "loss": 0.2198,
      "step": 40
    },
    {
      "epoch": 40.0,
      "eval_loss": 0.6874333620071411,
      "eval_runtime": 0.2605,
      "eval_samples_per_second": 26.873,
      "eval_steps_per_second": 3.839,
      "step": 40
    },
    {
      "epoch": 50.0,
      "learning_rate": 5e-05,
      "loss": 0.0973,
      "step": 50
    },
    {
      "epoch": 50.0,
      "eval_loss": 0.8139681816101074,
      "eval_runtime": 0.2601,
      "eval_samples_per_second": 26.91,
      "eval_steps_per_second": 3.844,
      "step": 50
    },
    {
      "epoch": 60.0,
      "learning_rate": 3.4549150281252636e-05,
      "loss": 0.0398,
      "step": 60
    },
    {
      "epoch": 60.0,
      "eval_loss": 0.9051175713539124,
      "eval_runtime": 0.2596,
      "eval_samples_per_second": 26.963,
      "eval_steps_per_second": 3.852,
      "step": 60
    },
    {
      "epoch": 70.0,
      "learning_rate": 2.061073738537635e-05,
      "loss": 0.02,
      "step": 70
    },
    {
      "epoch": 70.0,
      "eval_loss": 1.000623345375061,
      "eval_runtime": 0.2584,
      "eval_samples_per_second": 27.089,
      "eval_steps_per_second": 3.87,
      "step": 70
    },
    {
      "epoch": 80.0,
      "learning_rate": 9.549150281252633e-06,
      "loss": 0.0135,
      "step": 80
    },
    {
      "epoch": 80.0,
      "eval_loss": 1.0337673425674438,
      "eval_runtime": 0.2607,
      "eval_samples_per_second": 26.848,
      "eval_steps_per_second": 3.835,
      "step": 80
    },
    {
      "epoch": 90.0,
      "learning_rate": 2.4471741852423237e-06,
      "loss": 0.0115,
      "step": 90
    },
    {
      "epoch": 90.0,
      "eval_loss": 1.0473424196243286,
      "eval_runtime": 0.2628,
      "eval_samples_per_second": 26.64,
      "eval_steps_per_second": 3.806,
      "step": 90
    },
    {
      "epoch": 100.0,
      "learning_rate": 0.0,
      "loss": 0.0107,
      "step": 100
    },
    {
      "epoch": 100.0,
      "eval_loss": 1.051451325416565,
      "eval_runtime": 0.2626,
      "eval_samples_per_second": 26.659,
      "eval_steps_per_second": 3.808,
      "step": 100
    },
    {
      "epoch": 100.0,
      "step": 100,
      "total_flos": 8.61449315165184e+16,
      "train_loss": 0.2177230092138052,
      "train_runtime": 600.5927,
      "train_samples_per_second": 8.991,
      "train_steps_per_second": 0.167
    }
  ],
  "max_steps": 100,
  "num_train_epochs": 100,
  "total_flos": 8.61449315165184e+16,
  "trial_name": null,
  "trial_params": null
}