checkpoint-406/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.9993846153846154,
  "eval_steps": 500,
  "global_step": 406,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.12,
      "learning_rate": 0.00019616438356164385,
      "loss": 1.6793,
      "step": 50
    },
    {
      "epoch": 0.25,
      "learning_rate": 0.0001715068493150685,
      "loss": 1.5075,
      "step": 100
    },
    {
      "epoch": 0.37,
      "learning_rate": 0.00014410958904109591,
      "loss": 1.3192,
      "step": 150
    },
    {
      "epoch": 0.49,
      "learning_rate": 0.0001167123287671233,
      "loss": 1.2865,
      "step": 200
    },
    {
      "epoch": 0.62,
      "learning_rate": 8.93150684931507e-05,
      "loss": 1.2809,
      "step": 250
    },
    {
      "epoch": 0.74,
      "learning_rate": 6.191780821917809e-05,
      "loss": 1.2694,
      "step": 300
    },
    {
      "epoch": 0.86,
      "learning_rate": 3.452054794520549e-05,
      "loss": 1.2547,
      "step": 350
    },
    {
      "epoch": 0.98,
      "learning_rate": 7.123287671232877e-06,
      "loss": 1.2635,
      "step": 400
    }
  ],
  "logging_steps": 50,
  "max_steps": 406,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 500,
  "total_flos": 7.101714901499904e+16,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}