checkpoint-300/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 9.025,
  "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": 4.9999999999999996e-06,
      "loss": 11.1049,
      "step": 1
    },
    {
      "epoch": 1.02,
      "learning_rate": 0.00025,
      "loss": 8.1845,
      "step": 50
    },
    {
      "epoch": 3.01,
      "learning_rate": 0.0005,
      "loss": 5.9529,
      "step": 100
    },
    {
      "epoch": 4.03,
      "learning_rate": 0.0005833333333333333,
      "loss": 5.1965,
      "step": 150
    },
    {
      "epoch": 6.02,
      "learning_rate": 0.0005555555555555556,
      "loss": 4.8925,
      "step": 200
    },
    {
      "epoch": 8.01,
      "learning_rate": 0.0005277777777777777,
      "loss": 4.7182,
      "step": 250
    },
    {
      "epoch": 9.03,
      "learning_rate": 0.0005,
      "loss": 4.561,
      "step": 300
    }
  ],
  "max_steps": 1200,
  "num_train_epochs": 9223372036854775807,
  "total_flos": 8.1124904927232e+16,
  "trial_name": null,
  "trial_params": null
}