trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 22.598870056497177,
  "global_step": 100000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.68,
      "learning_rate": 9.830508474576272e-06,
      "loss": 3.108,
      "step": 3000
    },
    {
      "epoch": 1.36,
      "learning_rate": 9.661016949152544e-06,
      "loss": 2.0197,
      "step": 6000
    },
    {
      "epoch": 2.03,
      "learning_rate": 9.491525423728815e-06,
      "loss": 1.6483,
      "step": 9000
    },
    {
      "epoch": 2.26,
      "learning_rate": 9.435028248587572e-06,
      "loss": 1.4748,
      "step": 10000
    },
    {
      "epoch": 4.52,
      "learning_rate": 8.870056497175143e-06,
      "loss": 1.2986,
      "step": 20000
    },
    {
      "epoch": 6.78,
      "learning_rate": 8.305084745762712e-06,
      "loss": 1.074,
      "step": 30000
    },
    {
      "epoch": 9.04,
      "learning_rate": 7.740112994350283e-06,
      "loss": 0.9305,
      "step": 40000
    },
    {
      "epoch": 11.3,
      "learning_rate": 7.175141242937854e-06,
      "loss": 0.8309,
      "step": 50000
    },
    {
      "epoch": 13.56,
      "learning_rate": 6.610169491525424e-06,
      "loss": 0.7535,
      "step": 60000
    },
    {
      "epoch": 15.82,
      "learning_rate": 6.045197740112995e-06,
      "loss": 0.69,
      "step": 70000
    },
    {
      "epoch": 18.08,
      "learning_rate": 5.480225988700565e-06,
      "loss": 0.6365,
      "step": 80000
    },
    {
      "epoch": 20.34,
      "learning_rate": 4.915254237288136e-06,
      "loss": 0.5881,
      "step": 90000
    },
    {
      "epoch": 22.6,
      "learning_rate": 4.350282485875706e-06,
      "loss": 0.5565,
      "step": 100000
    }
  ],
  "max_steps": 177000,
  "num_train_epochs": 40,
  "total_flos": 1.2400055882758914e+20,
  "trial_name": null,
  "trial_params": null
}