checkpoint-300/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.11632415664986429,
  "global_step": 300,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.02,
      "learning_rate": 0.0001999867761371633,
      "loss": 1.0435,
      "step": 50
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.00019993306018843102,
      "loss": 0.8918,
      "step": 100
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.00019983804784290833,
      "loss": 0.8874,
      "step": 150
    },
    {
      "epoch": 0.08,
      "learning_rate": 0.00019970177836355307,
      "loss": 0.8839,
      "step": 200
    },
    {
      "epoch": 0.09,
      "learning_rate": 0.00019961818913082012,
      "loss": 0.8801,
      "step": 225
    },
    {
      "epoch": 0.1,
      "learning_rate": 0.00019952430806244534,
      "loss": 0.8753,
      "step": 250
    },
    {
      "epoch": 0.11,
      "learning_rate": 0.00019942014485754635,
      "loss": 0.8754,
      "step": 275
    },
    {
      "epoch": 0.12,
      "learning_rate": 0.00019930571027751713,
      "loss": 0.8751,
      "step": 300
    }
  ],
  "max_steps": 7737,
  "num_train_epochs": 3,
  "total_flos": 1.290261776910254e+18,
  "trial_name": null,
  "trial_params": null
}