last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.006264588509901363,
  "eval_steps": 13,
  "global_step": 26,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0002409457119192832,
      "grad_norm": NaN,
      "learning_rate": 1e-05,
      "loss": 0.0,
      "step": 1
    },
    {
      "epoch": 0.0002409457119192832,
      "eval_loss": NaN,
      "eval_runtime": 142.96,
      "eval_samples_per_second": 48.895,
      "eval_steps_per_second": 24.447,
      "step": 1
    },
    {
      "epoch": 0.0004818914238385664,
      "grad_norm": NaN,
      "learning_rate": 2e-05,
      "loss": 0.0,
      "step": 2
    },
    {
      "epoch": 0.0007228371357578496,
      "grad_norm": NaN,
      "learning_rate": 3e-05,
      "loss": 0.0,
      "step": 3
    },
    {
      "epoch": 0.0009637828476771327,
      "grad_norm": NaN,
      "learning_rate": 4e-05,
      "loss": 0.0,
      "step": 4
    },
    {
      "epoch": 0.0012047285595964158,
      "grad_norm": NaN,
      "learning_rate": 5e-05,
      "loss": 0.0,
      "step": 5
    },
    {
      "epoch": 0.0014456742715156991,
      "grad_norm": NaN,
      "learning_rate": 6e-05,
      "loss": 279.5375,
      "step": 6
    },
    {
      "epoch": 0.0016866199834349822,
      "grad_norm": NaN,
      "learning_rate": 7e-05,
      "loss": 188.3483,
      "step": 7
    },
    {
      "epoch": 0.0019275656953542655,
      "grad_norm": NaN,
      "learning_rate": 8e-05,
      "loss": 329.8409,
      "step": 8
    },
    {
      "epoch": 0.002168511407273549,
      "grad_norm": NaN,
      "learning_rate": 9e-05,
      "loss": 616.5905,
      "step": 9
    },
    {
      "epoch": 0.0024094571191928317,
      "grad_norm": NaN,
      "learning_rate": 0.0001,
      "loss": 21.2405,
      "step": 10
    },
    {
      "epoch": 0.002650402831112115,
      "grad_norm": NaN,
      "learning_rate": 9.98458666866564e-05,
      "loss": 0.0,
      "step": 11
    },
    {
      "epoch": 0.0028913485430313982,
      "grad_norm": NaN,
      "learning_rate": 9.938441702975689e-05,
      "loss": 0.0,
      "step": 12
    },
    {
      "epoch": 0.0031322942549506815,
      "grad_norm": NaN,
      "learning_rate": 9.861849601988383e-05,
      "loss": 0.0789,
      "step": 13
    },
    {
      "epoch": 0.0031322942549506815,
      "eval_loss": NaN,
      "eval_runtime": 82.341,
      "eval_samples_per_second": 84.891,
      "eval_steps_per_second": 42.445,
      "step": 13
    },
    {
      "epoch": 0.0033732399668699644,
      "grad_norm": NaN,
      "learning_rate": 9.755282581475769e-05,
      "loss": 2686.054,
      "step": 14
    },
    {
      "epoch": 0.0036141856787892477,
      "grad_norm": NaN,
      "learning_rate": 9.619397662556435e-05,
      "loss": 0.0,
      "step": 15
    },
    {
      "epoch": 0.003855131390708531,
      "grad_norm": NaN,
      "learning_rate": 9.45503262094184e-05,
      "loss": 3.0458,
      "step": 16
    },
    {
      "epoch": 0.004096077102627814,
      "grad_norm": NaN,
      "learning_rate": 9.263200821770461e-05,
      "loss": 32.6457,
      "step": 17
    },
    {
      "epoch": 0.004337022814547098,
      "grad_norm": NaN,
      "learning_rate": 9.045084971874738e-05,
      "loss": 39.4524,
      "step": 18
    },
    {
      "epoch": 0.004577968526466381,
      "grad_norm": NaN,
      "learning_rate": 8.802029828000156e-05,
      "loss": 0.0,
      "step": 19
    },
    {
      "epoch": 0.004818914238385663,
      "grad_norm": NaN,
      "learning_rate": 8.535533905932738e-05,
      "loss": 0.1735,
      "step": 20
    },
    {
      "epoch": 0.005059859950304947,
      "grad_norm": NaN,
      "learning_rate": 8.247240241650918e-05,
      "loss": 0.0,
      "step": 21
    },
    {
      "epoch": 0.00530080566222423,
      "grad_norm": NaN,
      "learning_rate": 7.938926261462366e-05,
      "loss": 0.0,
      "step": 22
    },
    {
      "epoch": 0.005541751374143513,
      "grad_norm": NaN,
      "learning_rate": 7.612492823579745e-05,
      "loss": 0.0,
      "step": 23
    },
    {
      "epoch": 0.0057826970860627965,
      "grad_norm": NaN,
      "learning_rate": 7.269952498697734e-05,
      "loss": 495.0417,
      "step": 24
    },
    {
      "epoch": 0.00602364279798208,
      "grad_norm": NaN,
      "learning_rate": 6.91341716182545e-05,
      "loss": 0.0657,
      "step": 25
    },
    {
      "epoch": 0.006264588509901363,
      "grad_norm": NaN,
      "learning_rate": 6.545084971874738e-05,
      "loss": 0.2915,
      "step": 26
    },
    {
      "epoch": 0.006264588509901363,
      "eval_loss": NaN,
      "eval_runtime": 82.3576,
      "eval_samples_per_second": 84.874,
      "eval_steps_per_second": 42.437,
      "step": 26
    }
  ],
  "logging_steps": 1,
  "max_steps": 50,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 13,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 4856544378224640.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}