tmp-checkpoint-300/trainer_state.json

{
  "best_metric": NaN,
  "best_model_checkpoint": "mgh6/TCS_Pairing_VAE/checkpoint-100",
  "epoch": 0.0027706715184203476,
  "eval_steps": 100,
  "global_step": 300,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0,
      "learning_rate": 0.0009999692147609064,
      "loss": 12653.28,
      "step": 100
    },
    {
      "epoch": 0.0,
      "eval_loss": NaN,
      "eval_runtime": 60.2212,
      "eval_samples_per_second": 53.636,
      "eval_steps_per_second": 53.636,
      "step": 100
    },
    {
      "epoch": 0.0,
      "learning_rate": 0.000999938429521813,
      "loss": 0.0,
      "step": 200
    },
    {
      "epoch": 0.0,
      "eval_loss": NaN,
      "eval_runtime": 62.8807,
      "eval_samples_per_second": 51.367,
      "eval_steps_per_second": 51.367,
      "step": 200
    },
    {
      "epoch": 0.0,
      "learning_rate": 0.0009999076442827194,
      "loss": 0.0,
      "step": 300
    },
    {
      "epoch": 0.0,
      "eval_loss": NaN,
      "eval_runtime": 62.677,
      "eval_samples_per_second": 51.534,
      "eval_steps_per_second": 51.534,
      "step": 300
    }
  ],
  "logging_steps": 100,
  "max_steps": 3248310,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 30,
  "save_steps": 100,
  "total_flos": 0.0,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}