trainer_state.json

{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 15.0,
  "eval_steps": 500,
  "global_step": 66674,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.9998875014062323,
      "grad_norm": 181.65682983398438,
      "learning_rate": 1e-05,
      "loss": 45.7848,
      "step": 8889
    },
    {
      "epoch": 3.0,
      "grad_norm": 422.18414306640625,
      "learning_rate": 1e-05,
      "loss": 43.869,
      "step": 13334
    },
    {
      "epoch": 4.0,
      "grad_norm": 199.45924377441406,
      "learning_rate": 1e-05,
      "loss": 42.2688,
      "step": 17779
    },
    {
      "epoch": 5.0,
      "grad_norm": 249.6620635986328,
      "learning_rate": 1e-05,
      "loss": 41.0351,
      "step": 22224
    },
    {
      "epoch": 6.0,
      "grad_norm": 328.8633728027344,
      "learning_rate": 1e-05,
      "loss": 40.109,
      "step": 26669
    },
    {
      "epoch": 7.0,
      "grad_norm": 358.3860778808594,
      "learning_rate": 1e-05,
      "loss": 39.3797,
      "step": 31114
    },
    {
      "epoch": 8.0,
      "grad_norm": 239.27536010742188,
      "learning_rate": 1e-05,
      "loss": 38.5914,
      "step": 35559
    },
    {
      "epoch": 9.0,
      "grad_norm": 316.275634765625,
      "learning_rate": 1e-05,
      "loss": 37.9495,
      "step": 40004
    },
    {
      "epoch": 10.0,
      "grad_norm": 177.41749572753906,
      "learning_rate": 1e-05,
      "loss": 37.4032,
      "step": 44449
    },
    {
      "epoch": 11.0,
      "grad_norm": 193.14991760253906,
      "learning_rate": 1e-05,
      "loss": 36.8619,
      "step": 48894
    },
    {
      "epoch": 12.0,
      "grad_norm": 173.55499267578125,
      "learning_rate": 1e-05,
      "loss": 36.3437,
      "step": 53339
    },
    {
      "epoch": 13.0,
      "grad_norm": 377.0917663574219,
      "learning_rate": 1e-05,
      "loss": 35.8008,
      "step": 57784
    },
    {
      "epoch": 14.0,
      "grad_norm": 326.80535888671875,
      "learning_rate": 1e-05,
      "loss": 35.321,
      "step": 62229
    },
    {
      "epoch": 15.0,
      "grad_norm": 370.6023254394531,
      "learning_rate": 1e-05,
      "loss": 34.8802,
      "step": 66674
    },
    {
      "epoch": 15.0,
      "step": 66674,
      "total_flos": 4.498576409525775e+21,
      "train_loss": 34.2752071881468,
      "train_runtime": 162790.0677,
      "train_samples_per_second": 32.76,
      "train_steps_per_second": 0.41
    }
  ],
  "logging_steps": 500,
  "max_steps": 66675,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 15,
  "save_steps": 2500,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 4.498576409525775e+21,
  "train_batch_size": 5,
  "trial_name": null,
  "trial_params": null
}