trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 10.0,
  "global_step": 9110,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.55,
      "learning_rate": 4.725576289791438e-05,
      "loss": 2.4428,
      "step": 500
    },
    {
      "epoch": 1.1,
      "learning_rate": 4.451152579582876e-05,
      "loss": 2.2742,
      "step": 1000
    },
    {
      "epoch": 1.65,
      "learning_rate": 4.176728869374314e-05,
      "loss": 2.1079,
      "step": 1500
    },
    {
      "epoch": 2.2,
      "learning_rate": 3.902305159165752e-05,
      "loss": 2.0447,
      "step": 2000
    },
    {
      "epoch": 2.74,
      "learning_rate": 3.62788144895719e-05,
      "loss": 1.9388,
      "step": 2500
    },
    {
      "epoch": 3.29,
      "learning_rate": 3.353457738748628e-05,
      "loss": 1.8627,
      "step": 3000
    },
    {
      "epoch": 3.84,
      "learning_rate": 3.079034028540066e-05,
      "loss": 1.805,
      "step": 3500
    },
    {
      "epoch": 4.39,
      "learning_rate": 2.8046103183315044e-05,
      "loss": 1.7183,
      "step": 4000
    },
    {
      "epoch": 4.94,
      "learning_rate": 2.530186608122942e-05,
      "loss": 1.6946,
      "step": 4500
    },
    {
      "epoch": 5.49,
      "learning_rate": 2.25576289791438e-05,
      "loss": 1.5987,
      "step": 5000
    },
    {
      "epoch": 6.04,
      "learning_rate": 1.981339187705818e-05,
      "loss": 1.5931,
      "step": 5500
    },
    {
      "epoch": 6.59,
      "learning_rate": 1.7069154774972558e-05,
      "loss": 1.5127,
      "step": 6000
    },
    {
      "epoch": 7.14,
      "learning_rate": 1.432491767288694e-05,
      "loss": 1.5066,
      "step": 6500
    },
    {
      "epoch": 7.68,
      "learning_rate": 1.1580680570801318e-05,
      "loss": 1.4534,
      "step": 7000
    },
    {
      "epoch": 8.23,
      "learning_rate": 8.836443468715698e-06,
      "loss": 1.4371,
      "step": 7500
    },
    {
      "epoch": 8.78,
      "learning_rate": 6.092206366630078e-06,
      "loss": 1.4109,
      "step": 8000
    },
    {
      "epoch": 9.33,
      "learning_rate": 3.3479692645444565e-06,
      "loss": 1.395,
      "step": 8500
    },
    {
      "epoch": 9.88,
      "learning_rate": 6.037321624588365e-07,
      "loss": 1.3817,
      "step": 9000
    },
    {
      "epoch": 10.0,
      "step": 9110,
      "total_flos": 5.074420594900992e+16,
      "train_loss": 1.7279373135446325,
      "train_runtime": 6450.8845,
      "train_samples_per_second": 4.235,
      "train_steps_per_second": 1.412
    }
  ],
  "max_steps": 9110,
  "num_train_epochs": 10,
  "total_flos": 5.074420594900992e+16,
  "trial_name": null,
  "trial_params": null
}