last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.015105740181268883,
  "eval_steps": 3,
  "global_step": 10,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0015105740181268882,
      "grad_norm": NaN,
      "learning_rate": 2e-05,
      "loss": 0.0,
      "step": 1
    },
    {
      "epoch": 0.0015105740181268882,
      "eval_loss": NaN,
      "eval_runtime": 7.2858,
      "eval_samples_per_second": 38.294,
      "eval_steps_per_second": 19.215,
      "step": 1
    },
    {
      "epoch": 0.0030211480362537764,
      "grad_norm": NaN,
      "learning_rate": 4e-05,
      "loss": 0.0,
      "step": 2
    },
    {
      "epoch": 0.004531722054380665,
      "grad_norm": NaN,
      "learning_rate": 6e-05,
      "loss": 0.0,
      "step": 3
    },
    {
      "epoch": 0.004531722054380665,
      "eval_loss": NaN,
      "eval_runtime": 5.792,
      "eval_samples_per_second": 48.17,
      "eval_steps_per_second": 24.171,
      "step": 3
    },
    {
      "epoch": 0.006042296072507553,
      "grad_norm": NaN,
      "learning_rate": 8e-05,
      "loss": 0.0,
      "step": 4
    },
    {
      "epoch": 0.0075528700906344415,
      "grad_norm": NaN,
      "learning_rate": 0.0001,
      "loss": 0.0,
      "step": 5
    },
    {
      "epoch": 0.00906344410876133,
      "grad_norm": NaN,
      "learning_rate": 0.00012,
      "loss": 0.0,
      "step": 6
    },
    {
      "epoch": 0.00906344410876133,
      "eval_loss": NaN,
      "eval_runtime": 5.9071,
      "eval_samples_per_second": 47.231,
      "eval_steps_per_second": 23.7,
      "step": 6
    },
    {
      "epoch": 0.010574018126888218,
      "grad_norm": NaN,
      "learning_rate": 0.00014,
      "loss": 0.0,
      "step": 7
    },
    {
      "epoch": 0.012084592145015106,
      "grad_norm": NaN,
      "learning_rate": 0.00016,
      "loss": 0.0,
      "step": 8
    },
    {
      "epoch": 0.013595166163141994,
      "grad_norm": NaN,
      "learning_rate": 0.00018,
      "loss": 0.0,
      "step": 9
    },
    {
      "epoch": 0.013595166163141994,
      "eval_loss": NaN,
      "eval_runtime": 5.9005,
      "eval_samples_per_second": 47.284,
      "eval_steps_per_second": 23.727,
      "step": 9
    },
    {
      "epoch": 0.015105740181268883,
      "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": 408023281434624.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}