last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.08097165991902834,
  "eval_steps": 3,
  "global_step": 10,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.008097165991902834,
      "grad_norm": NaN,
      "learning_rate": 0.0,
      "loss": 5.6119,
      "step": 1
    },
    {
      "epoch": 0.008097165991902834,
      "eval_loss": 5.563105583190918,
      "eval_runtime": 6.1798,
      "eval_samples_per_second": 33.658,
      "eval_steps_per_second": 4.207,
      "step": 1
    },
    {
      "epoch": 0.016194331983805668,
      "grad_norm": NaN,
      "learning_rate": 0.0,
      "loss": 5.7242,
      "step": 2
    },
    {
      "epoch": 0.024291497975708502,
      "grad_norm": NaN,
      "learning_rate": 0.0,
      "loss": 5.5118,
      "step": 3
    },
    {
      "epoch": 0.024291497975708502,
      "eval_loss": 5.563105583190918,
      "eval_runtime": 6.1784,
      "eval_samples_per_second": 33.665,
      "eval_steps_per_second": 4.208,
      "step": 3
    },
    {
      "epoch": 0.032388663967611336,
      "grad_norm": 7.104294776916504,
      "learning_rate": 2e-05,
      "loss": 5.596,
      "step": 4
    },
    {
      "epoch": 0.04048582995951417,
      "grad_norm": 7.289365291595459,
      "learning_rate": 4e-05,
      "loss": 5.7354,
      "step": 5
    },
    {
      "epoch": 0.048582995951417005,
      "grad_norm": 7.546472549438477,
      "learning_rate": 6e-05,
      "loss": 5.2959,
      "step": 6
    },
    {
      "epoch": 0.048582995951417005,
      "eval_loss": 4.9584832191467285,
      "eval_runtime": 6.0022,
      "eval_samples_per_second": 34.654,
      "eval_steps_per_second": 4.332,
      "step": 6
    },
    {
      "epoch": 0.05668016194331984,
      "grad_norm": 6.355536460876465,
      "learning_rate": 8e-05,
      "loss": 4.8219,
      "step": 7
    },
    {
      "epoch": 0.06477732793522267,
      "grad_norm": 6.963198661804199,
      "learning_rate": 0.0001,
      "loss": 4.199,
      "step": 8
    },
    {
      "epoch": 0.0728744939271255,
      "grad_norm": 7.163834095001221,
      "learning_rate": 0.00012,
      "loss": 3.4092,
      "step": 9
    },
    {
      "epoch": 0.0728744939271255,
      "eval_loss": 2.5691821575164795,
      "eval_runtime": 5.9932,
      "eval_samples_per_second": 34.706,
      "eval_steps_per_second": 4.338,
      "step": 9
    },
    {
      "epoch": 0.08097165991902834,
      "grad_norm": 7.162381649017334,
      "learning_rate": 0.00014,
      "loss": 2.6516,
      "step": 10
    }
  ],
  "logging_steps": 1,
  "max_steps": 10,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 3,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 5.935098268483584e+16,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}