trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.03582945180938732,
  "eval_steps": 500,
  "global_step": 100,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0,
      "learning_rate": 0.0002,
      "loss": 0.7063,
      "step": 10
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.00019396926207859084,
      "loss": 0.7526,
      "step": 20
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0001766044443118978,
      "loss": 0.8085,
      "step": 30
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.00015000000000000001,
      "loss": 0.7688,
      "step": 40
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.00011736481776669306,
      "loss": 0.8688,
      "step": 50
    },
    {
      "epoch": 0.02,
      "learning_rate": 8.263518223330697e-05,
      "loss": 0.738,
      "step": 60
    },
    {
      "epoch": 0.03,
      "learning_rate": 5.000000000000002e-05,
      "loss": 0.7781,
      "step": 70
    },
    {
      "epoch": 0.03,
      "learning_rate": 2.339555568810221e-05,
      "loss": 0.6929,
      "step": 80
    },
    {
      "epoch": 0.03,
      "learning_rate": 6.030737921409169e-06,
      "loss": 0.7981,
      "step": 90
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0,
      "loss": 0.8038,
      "step": 100
    }
  ],
  "logging_steps": 10,
  "max_steps": 100,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 10,
  "total_flos": 7.462704855908352e+16,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}