last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 1.0526315789473684,
  "eval_steps": 4,
  "global_step": 15,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.07017543859649122,
      "grad_norm": NaN,
      "learning_rate": 5e-05,
      "loss": 0.0,
      "step": 1
    },
    {
      "epoch": 0.07017543859649122,
      "eval_loss": NaN,
      "eval_runtime": 10.4393,
      "eval_samples_per_second": 9.292,
      "eval_steps_per_second": 2.395,
      "step": 1
    },
    {
      "epoch": 0.14035087719298245,
      "grad_norm": NaN,
      "learning_rate": 0.0001,
      "loss": 0.0,
      "step": 2
    },
    {
      "epoch": 0.21052631578947367,
      "grad_norm": NaN,
      "learning_rate": 9.85470908713026e-05,
      "loss": 0.0,
      "step": 3
    },
    {
      "epoch": 0.2807017543859649,
      "grad_norm": NaN,
      "learning_rate": 9.42728012826605e-05,
      "loss": 0.0,
      "step": 4
    },
    {
      "epoch": 0.2807017543859649,
      "eval_loss": NaN,
      "eval_runtime": 9.7444,
      "eval_samples_per_second": 9.954,
      "eval_steps_per_second": 2.566,
      "step": 4
    },
    {
      "epoch": 0.3508771929824561,
      "grad_norm": NaN,
      "learning_rate": 8.742553740855506e-05,
      "loss": 0.0,
      "step": 5
    },
    {
      "epoch": 0.42105263157894735,
      "grad_norm": NaN,
      "learning_rate": 7.840323733655778e-05,
      "loss": 0.0,
      "step": 6
    },
    {
      "epoch": 0.49122807017543857,
      "grad_norm": NaN,
      "learning_rate": 6.773024435212678e-05,
      "loss": 0.0,
      "step": 7
    },
    {
      "epoch": 0.5614035087719298,
      "grad_norm": NaN,
      "learning_rate": 5.602683401276615e-05,
      "loss": 0.0,
      "step": 8
    },
    {
      "epoch": 0.5614035087719298,
      "eval_loss": NaN,
      "eval_runtime": 9.7423,
      "eval_samples_per_second": 9.957,
      "eval_steps_per_second": 2.566,
      "step": 8
    },
    {
      "epoch": 0.631578947368421,
      "grad_norm": NaN,
      "learning_rate": 4.397316598723385e-05,
      "loss": 0.0,
      "step": 9
    },
    {
      "epoch": 0.7017543859649122,
      "grad_norm": NaN,
      "learning_rate": 3.226975564787322e-05,
      "loss": 0.0,
      "step": 10
    },
    {
      "epoch": 0.7719298245614035,
      "grad_norm": NaN,
      "learning_rate": 2.1596762663442218e-05,
      "loss": 0.0,
      "step": 11
    },
    {
      "epoch": 0.8421052631578947,
      "grad_norm": NaN,
      "learning_rate": 1.257446259144494e-05,
      "loss": 0.0,
      "step": 12
    },
    {
      "epoch": 0.8421052631578947,
      "eval_loss": NaN,
      "eval_runtime": 9.752,
      "eval_samples_per_second": 9.947,
      "eval_steps_per_second": 2.564,
      "step": 12
    },
    {
      "epoch": 0.9122807017543859,
      "grad_norm": NaN,
      "learning_rate": 5.727198717339511e-06,
      "loss": 0.0,
      "step": 13
    },
    {
      "epoch": 0.9824561403508771,
      "grad_norm": NaN,
      "learning_rate": 1.4529091286973995e-06,
      "loss": 0.0,
      "step": 14
    },
    {
      "epoch": 1.0526315789473684,
      "grad_norm": NaN,
      "learning_rate": 0.0,
      "loss": 0.0,
      "step": 15
    }
  ],
  "logging_steps": 1,
  "max_steps": 15,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 2,
  "save_steps": 4,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 3.3553856474382336e+17,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}