last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.013099006658661719,
  "eval_steps": 5,
  "global_step": 15,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0008732671105774479,
      "grad_norm": 0.9595964550971985,
      "learning_rate": 1e-05,
      "loss": 14.5402,
      "step": 1
    },
    {
      "epoch": 0.0008732671105774479,
      "eval_loss": NaN,
      "eval_runtime": 293.4541,
      "eval_samples_per_second": 6.573,
      "eval_steps_per_second": 3.288,
      "step": 1
    },
    {
      "epoch": 0.0017465342211548957,
      "grad_norm": 1.0124590396881104,
      "learning_rate": 2e-05,
      "loss": 14.3901,
      "step": 2
    },
    {
      "epoch": 0.0026198013317323436,
      "grad_norm": 0.8724995255470276,
      "learning_rate": 3e-05,
      "loss": 13.8135,
      "step": 3
    },
    {
      "epoch": 0.0034930684423097914,
      "grad_norm": 1.0120279788970947,
      "learning_rate": 4e-05,
      "loss": 13.1183,
      "step": 4
    },
    {
      "epoch": 0.004366335552887239,
      "grad_norm": 0.9744743704795837,
      "learning_rate": 5e-05,
      "loss": 14.1921,
      "step": 5
    },
    {
      "epoch": 0.004366335552887239,
      "eval_loss": NaN,
      "eval_runtime": 83.9616,
      "eval_samples_per_second": 22.975,
      "eval_steps_per_second": 11.493,
      "step": 5
    },
    {
      "epoch": 0.005239602663464687,
      "grad_norm": 0.8877798318862915,
      "learning_rate": 6e-05,
      "loss": 13.3851,
      "step": 6
    },
    {
      "epoch": 0.006112869774042135,
      "grad_norm": 1.045027732849121,
      "learning_rate": 7e-05,
      "loss": 12.9308,
      "step": 7
    },
    {
      "epoch": 0.006986136884619583,
      "grad_norm": 1.2155197858810425,
      "learning_rate": 8e-05,
      "loss": 13.1983,
      "step": 8
    },
    {
      "epoch": 0.007859403995197032,
      "grad_norm": 1.1796858310699463,
      "learning_rate": 9e-05,
      "loss": 13.7154,
      "step": 9
    },
    {
      "epoch": 0.008732671105774479,
      "grad_norm": 1.4411159753799438,
      "learning_rate": 0.0001,
      "loss": 13.4192,
      "step": 10
    },
    {
      "epoch": 0.008732671105774479,
      "eval_loss": NaN,
      "eval_runtime": 83.9193,
      "eval_samples_per_second": 22.986,
      "eval_steps_per_second": 11.499,
      "step": 10
    },
    {
      "epoch": 0.009605938216351927,
      "grad_norm": 1.8270329236984253,
      "learning_rate": 9.755282581475769e-05,
      "loss": 14.8922,
      "step": 11
    },
    {
      "epoch": 0.010479205326929374,
      "grad_norm": 1.498534917831421,
      "learning_rate": 9.045084971874738e-05,
      "loss": 12.4,
      "step": 12
    },
    {
      "epoch": 0.011352472437506823,
      "grad_norm": 1.495392084121704,
      "learning_rate": 7.938926261462366e-05,
      "loss": 13.5343,
      "step": 13
    },
    {
      "epoch": 0.01222573954808427,
      "grad_norm": 1.452784538269043,
      "learning_rate": 6.545084971874738e-05,
      "loss": 14.3007,
      "step": 14
    },
    {
      "epoch": 0.013099006658661719,
      "grad_norm": 1.3574515581130981,
      "learning_rate": 5e-05,
      "loss": 13.6316,
      "step": 15
    },
    {
      "epoch": 0.013099006658661719,
      "eval_loss": NaN,
      "eval_runtime": 84.1591,
      "eval_samples_per_second": 22.921,
      "eval_steps_per_second": 11.466,
      "step": 15
    }
  ],
  "logging_steps": 1,
  "max_steps": 20,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 5,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 1.9463553442381824e+16,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}