last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.015432098765432098,
  "eval_steps": 3,
  "global_step": 10,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0015432098765432098,
      "grad_norm": NaN,
      "learning_rate": 2e-05,
      "loss": 0.0,
      "step": 1
    },
    {
      "epoch": 0.0015432098765432098,
      "eval_loss": NaN,
      "eval_runtime": 12.8713,
      "eval_samples_per_second": 21.21,
      "eval_steps_per_second": 10.644,
      "step": 1
    },
    {
      "epoch": 0.0030864197530864196,
      "grad_norm": NaN,
      "learning_rate": 4e-05,
      "loss": 0.0,
      "step": 2
    },
    {
      "epoch": 0.004629629629629629,
      "grad_norm": NaN,
      "learning_rate": 6e-05,
      "loss": 0.0,
      "step": 3
    },
    {
      "epoch": 0.004629629629629629,
      "eval_loss": NaN,
      "eval_runtime": 10.811,
      "eval_samples_per_second": 25.252,
      "eval_steps_per_second": 12.672,
      "step": 3
    },
    {
      "epoch": 0.006172839506172839,
      "grad_norm": NaN,
      "learning_rate": 8e-05,
      "loss": 0.0,
      "step": 4
    },
    {
      "epoch": 0.007716049382716049,
      "grad_norm": NaN,
      "learning_rate": 0.0001,
      "loss": 0.0,
      "step": 5
    },
    {
      "epoch": 0.009259259259259259,
      "grad_norm": NaN,
      "learning_rate": 0.00012,
      "loss": 0.0,
      "step": 6
    },
    {
      "epoch": 0.009259259259259259,
      "eval_loss": NaN,
      "eval_runtime": 10.9706,
      "eval_samples_per_second": 24.885,
      "eval_steps_per_second": 12.488,
      "step": 6
    },
    {
      "epoch": 0.010802469135802469,
      "grad_norm": NaN,
      "learning_rate": 0.00014,
      "loss": 0.0,
      "step": 7
    },
    {
      "epoch": 0.012345679012345678,
      "grad_norm": NaN,
      "learning_rate": 0.00016,
      "loss": 0.0,
      "step": 8
    },
    {
      "epoch": 0.013888888888888888,
      "grad_norm": NaN,
      "learning_rate": 0.00018,
      "loss": 0.0,
      "step": 9
    },
    {
      "epoch": 0.013888888888888888,
      "eval_loss": NaN,
      "eval_runtime": 10.8541,
      "eval_samples_per_second": 25.152,
      "eval_steps_per_second": 12.622,
      "step": 9
    },
    {
      "epoch": 0.015432098765432098,
      "grad_norm": NaN,
      "learning_rate": 0.0002,
      "loss": 0.0,
      "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": 2586328248090624.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}