trainer_state.json

{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 14.955483170466884,
  "eval_steps": 500,
  "global_step": 1725,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.7296416938110748,
      "grad_norm": 70699.7265625,
      "learning_rate": 8.846376811594204e-05,
      "loss": 1.4382,
      "step": 200
    },
    {
      "epoch": 3.451682953311618,
      "grad_norm": 72228.890625,
      "learning_rate": 7.686956521739131e-05,
      "loss": 0.878,
      "step": 400
    },
    {
      "epoch": 5.217155266015201,
      "grad_norm": 81379.8359375,
      "learning_rate": 6.527536231884058e-05,
      "loss": 0.404,
      "step": 600
    },
    {
      "epoch": 6.946796959826276,
      "grad_norm": 81457.5703125,
      "learning_rate": 5.368115942028986e-05,
      "loss": 0.2967,
      "step": 800
    },
    {
      "epoch": 8.668838219326819,
      "grad_norm": 63213.9140625,
      "learning_rate": 4.208695652173914e-05,
      "loss": 0.1453,
      "step": 1000
    },
    {
      "epoch": 10.425624321389794,
      "grad_norm": 54476.24609375,
      "learning_rate": 3.049275362318841e-05,
      "loss": 0.079,
      "step": 1200
    },
    {
      "epoch": 12.147665580890337,
      "grad_norm": 46802.86328125,
      "learning_rate": 1.889855072463768e-05,
      "loss": 0.0489,
      "step": 1400
    },
    {
      "epoch": 13.877307274701412,
      "grad_norm": 35309.5546875,
      "learning_rate": 7.304347826086957e-06,
      "loss": 0.0294,
      "step": 1600
    }
  ],
  "logging_steps": 200,
  "max_steps": 1725,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 15,
  "save_steps": 500,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 2.3775363971049062e+17,
  "train_batch_size": 16,
  "trial_name": null,
  "trial_params": null
}