last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.028303555634176542,
  "eval_steps": 3,
  "global_step": 10,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0028303555634176542,
      "grad_norm": NaN,
      "learning_rate": 1e-05,
      "loss": 0.0,
      "step": 1
    },
    {
      "epoch": 0.0028303555634176542,
      "eval_loss": NaN,
      "eval_runtime": 94.5226,
      "eval_samples_per_second": 6.305,
      "eval_steps_per_second": 3.153,
      "step": 1
    },
    {
      "epoch": 0.0056607111268353084,
      "grad_norm": NaN,
      "learning_rate": 2e-05,
      "loss": 0.0,
      "step": 2
    },
    {
      "epoch": 0.008491066690252963,
      "grad_norm": NaN,
      "learning_rate": 3e-05,
      "loss": 0.0,
      "step": 3
    },
    {
      "epoch": 0.008491066690252963,
      "eval_loss": NaN,
      "eval_runtime": 13.5254,
      "eval_samples_per_second": 44.065,
      "eval_steps_per_second": 22.033,
      "step": 3
    },
    {
      "epoch": 0.011321422253670617,
      "grad_norm": NaN,
      "learning_rate": 4e-05,
      "loss": 0.0,
      "step": 4
    },
    {
      "epoch": 0.014151777817088271,
      "grad_norm": NaN,
      "learning_rate": 5e-05,
      "loss": 0.0,
      "step": 5
    },
    {
      "epoch": 0.016982133380505925,
      "grad_norm": NaN,
      "learning_rate": 6e-05,
      "loss": 0.0,
      "step": 6
    },
    {
      "epoch": 0.016982133380505925,
      "eval_loss": NaN,
      "eval_runtime": 13.5837,
      "eval_samples_per_second": 43.876,
      "eval_steps_per_second": 21.938,
      "step": 6
    },
    {
      "epoch": 0.01981248894392358,
      "grad_norm": NaN,
      "learning_rate": 7e-05,
      "loss": 0.0,
      "step": 7
    },
    {
      "epoch": 0.022642844507341234,
      "grad_norm": NaN,
      "learning_rate": 8e-05,
      "loss": 0.0,
      "step": 8
    },
    {
      "epoch": 0.02547320007075889,
      "grad_norm": NaN,
      "learning_rate": 9e-05,
      "loss": 0.0,
      "step": 9
    },
    {
      "epoch": 0.02547320007075889,
      "eval_loss": NaN,
      "eval_runtime": 13.5203,
      "eval_samples_per_second": 44.082,
      "eval_steps_per_second": 22.041,
      "step": 9
    },
    {
      "epoch": 0.028303555634176542,
      "grad_norm": NaN,
      "learning_rate": 0.0001,
      "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": 7362637294731264.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}