last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.10526315789473684,
  "eval_steps": 5,
  "global_step": 15,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.007017543859649123,
      "grad_norm": NaN,
      "learning_rate": 1e-05,
      "loss": 0.0,
      "step": 1
    },
    {
      "epoch": 0.007017543859649123,
      "eval_loss": NaN,
      "eval_runtime": 137.8307,
      "eval_samples_per_second": 1.741,
      "eval_steps_per_second": 0.871,
      "step": 1
    },
    {
      "epoch": 0.014035087719298246,
      "grad_norm": NaN,
      "learning_rate": 2e-05,
      "loss": 0.0,
      "step": 2
    },
    {
      "epoch": 0.021052631578947368,
      "grad_norm": NaN,
      "learning_rate": 3e-05,
      "loss": 0.0,
      "step": 3
    },
    {
      "epoch": 0.028070175438596492,
      "grad_norm": NaN,
      "learning_rate": 4e-05,
      "loss": 0.0,
      "step": 4
    },
    {
      "epoch": 0.03508771929824561,
      "grad_norm": NaN,
      "learning_rate": 5e-05,
      "loss": 0.0,
      "step": 5
    },
    {
      "epoch": 0.03508771929824561,
      "eval_loss": NaN,
      "eval_runtime": 12.8355,
      "eval_samples_per_second": 18.698,
      "eval_steps_per_second": 9.349,
      "step": 5
    },
    {
      "epoch": 0.042105263157894736,
      "grad_norm": NaN,
      "learning_rate": 6e-05,
      "loss": 0.0,
      "step": 6
    },
    {
      "epoch": 0.04912280701754386,
      "grad_norm": NaN,
      "learning_rate": 7e-05,
      "loss": 0.0,
      "step": 7
    },
    {
      "epoch": 0.056140350877192984,
      "grad_norm": NaN,
      "learning_rate": 8e-05,
      "loss": 0.0,
      "step": 8
    },
    {
      "epoch": 0.06315789473684211,
      "grad_norm": NaN,
      "learning_rate": 9e-05,
      "loss": 0.0,
      "step": 9
    },
    {
      "epoch": 0.07017543859649122,
      "grad_norm": NaN,
      "learning_rate": 0.0001,
      "loss": 0.0,
      "step": 10
    },
    {
      "epoch": 0.07017543859649122,
      "eval_loss": NaN,
      "eval_runtime": 12.8432,
      "eval_samples_per_second": 18.687,
      "eval_steps_per_second": 9.343,
      "step": 10
    },
    {
      "epoch": 0.07719298245614035,
      "grad_norm": NaN,
      "learning_rate": 9.755282581475769e-05,
      "loss": 0.0,
      "step": 11
    },
    {
      "epoch": 0.08421052631578947,
      "grad_norm": NaN,
      "learning_rate": 9.045084971874738e-05,
      "loss": 0.0,
      "step": 12
    },
    {
      "epoch": 0.0912280701754386,
      "grad_norm": NaN,
      "learning_rate": 7.938926261462366e-05,
      "loss": 0.0,
      "step": 13
    },
    {
      "epoch": 0.09824561403508772,
      "grad_norm": NaN,
      "learning_rate": 6.545084971874738e-05,
      "loss": 0.0,
      "step": 14
    },
    {
      "epoch": 0.10526315789473684,
      "grad_norm": NaN,
      "learning_rate": 5e-05,
      "loss": 0.0,
      "step": 15
    },
    {
      "epoch": 0.10526315789473684,
      "eval_loss": NaN,
      "eval_runtime": 14.785,
      "eval_samples_per_second": 16.233,
      "eval_steps_per_second": 8.116,
      "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": 2.7827364213817344e+16,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}