last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.23952095808383234,
  "eval_steps": 3,
  "global_step": 10,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.023952095808383235,
      "grad_norm": 12127.7060546875,
      "learning_rate": 2e-05,
      "loss": 19.1507,
      "step": 1
    },
    {
      "epoch": 0.023952095808383235,
      "eval_loss": 4.564117908477783,
      "eval_runtime": 0.7252,
      "eval_samples_per_second": 24.82,
      "eval_steps_per_second": 12.41,
      "step": 1
    },
    {
      "epoch": 0.04790419161676647,
      "grad_norm": 17571.01171875,
      "learning_rate": 4e-05,
      "loss": 18.2525,
      "step": 2
    },
    {
      "epoch": 0.0718562874251497,
      "grad_norm": 20290.41796875,
      "learning_rate": 6e-05,
      "loss": 18.0966,
      "step": 3
    },
    {
      "epoch": 0.0718562874251497,
      "eval_loss": 4.592726707458496,
      "eval_runtime": 0.2393,
      "eval_samples_per_second": 75.207,
      "eval_steps_per_second": 37.603,
      "step": 3
    },
    {
      "epoch": 0.09580838323353294,
      "grad_norm": 33943.01171875,
      "learning_rate": 8e-05,
      "loss": 18.4686,
      "step": 4
    },
    {
      "epoch": 0.11976047904191617,
      "grad_norm": 17252.619140625,
      "learning_rate": 0.0001,
      "loss": 17.8588,
      "step": 5
    },
    {
      "epoch": 0.1437125748502994,
      "grad_norm": 18595.34375,
      "learning_rate": 0.00012,
      "loss": 17.6699,
      "step": 6
    },
    {
      "epoch": 0.1437125748502994,
      "eval_loss": 4.403322219848633,
      "eval_runtime": 0.2201,
      "eval_samples_per_second": 81.784,
      "eval_steps_per_second": 40.892,
      "step": 6
    },
    {
      "epoch": 0.16766467065868262,
      "grad_norm": 14934.6083984375,
      "learning_rate": 0.00014,
      "loss": 18.3235,
      "step": 7
    },
    {
      "epoch": 0.19161676646706588,
      "grad_norm": 26373.935546875,
      "learning_rate": 0.00016,
      "loss": 17.102,
      "step": 8
    },
    {
      "epoch": 0.2155688622754491,
      "grad_norm": 14985.6845703125,
      "learning_rate": 0.00018,
      "loss": 17.8228,
      "step": 9
    },
    {
      "epoch": 0.2155688622754491,
      "eval_loss": 4.397857666015625,
      "eval_runtime": 0.2368,
      "eval_samples_per_second": 76.008,
      "eval_steps_per_second": 38.004,
      "step": 9
    },
    {
      "epoch": 0.23952095808383234,
      "grad_norm": 59956.89453125,
      "learning_rate": 0.0002,
      "loss": 17.0715,
      "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": 1899013079040.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}