last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.03666361136571952,
  "eval_steps": 3,
  "global_step": 10,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0036663611365719525,
      "grad_norm": 11.825860977172852,
      "learning_rate": 2e-05,
      "loss": 10.8658,
      "step": 1
    },
    {
      "epoch": 0.0036663611365719525,
      "eval_loss": 11.236878395080566,
      "eval_runtime": 12.396,
      "eval_samples_per_second": 9.277,
      "eval_steps_per_second": 4.679,
      "step": 1
    },
    {
      "epoch": 0.007332722273143905,
      "grad_norm": 11.643694877624512,
      "learning_rate": 4e-05,
      "loss": 11.5997,
      "step": 2
    },
    {
      "epoch": 0.010999083409715857,
      "grad_norm": 13.432502746582031,
      "learning_rate": 6e-05,
      "loss": 12.1139,
      "step": 3
    },
    {
      "epoch": 0.010999083409715857,
      "eval_loss": 11.0986967086792,
      "eval_runtime": 11.9026,
      "eval_samples_per_second": 9.662,
      "eval_steps_per_second": 4.873,
      "step": 3
    },
    {
      "epoch": 0.01466544454628781,
      "grad_norm": 16.266033172607422,
      "learning_rate": 8e-05,
      "loss": 12.442,
      "step": 4
    },
    {
      "epoch": 0.01833180568285976,
      "grad_norm": 14.242290496826172,
      "learning_rate": 0.0001,
      "loss": 9.6975,
      "step": 5
    },
    {
      "epoch": 0.021998166819431713,
      "grad_norm": 19.96150779724121,
      "learning_rate": 0.00012,
      "loss": 9.8631,
      "step": 6
    },
    {
      "epoch": 0.021998166819431713,
      "eval_loss": 8.199483871459961,
      "eval_runtime": 12.1413,
      "eval_samples_per_second": 9.472,
      "eval_steps_per_second": 4.777,
      "step": 6
    },
    {
      "epoch": 0.025664527956003668,
      "grad_norm": 21.698095321655273,
      "learning_rate": 0.00014,
      "loss": 8.2158,
      "step": 7
    },
    {
      "epoch": 0.02933088909257562,
      "grad_norm": 32.11186218261719,
      "learning_rate": 0.00016,
      "loss": 6.0142,
      "step": 8
    },
    {
      "epoch": 0.03299725022914757,
      "grad_norm": 14.958884239196777,
      "learning_rate": 0.00018,
      "loss": 2.9791,
      "step": 9
    },
    {
      "epoch": 0.03299725022914757,
      "eval_loss": 2.25606632232666,
      "eval_runtime": 12.2047,
      "eval_samples_per_second": 9.423,
      "eval_steps_per_second": 4.752,
      "step": 9
    },
    {
      "epoch": 0.03666361136571952,
      "grad_norm": 13.798932075500488,
      "learning_rate": 0.0002,
      "loss": 2.7228,
      "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": 1849564248145920.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}