last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.0010926155577529269,
  "eval_steps": 10,
  "global_step": 50,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 2.1852311155058537e-05,
      "eval_loss": NaN,
      "eval_runtime": 680.7093,
      "eval_samples_per_second": 28.307,
      "eval_steps_per_second": 14.154,
      "step": 1
    },
    {
      "epoch": 0.00010926155577529269,
      "grad_norm": NaN,
      "learning_rate": 5e-05,
      "loss": 0.0,
      "step": 5
    },
    {
      "epoch": 0.00021852311155058537,
      "grad_norm": NaN,
      "learning_rate": 0.0001,
      "loss": 0.0,
      "step": 10
    },
    {
      "epoch": 0.00021852311155058537,
      "eval_loss": NaN,
      "eval_runtime": 675.8531,
      "eval_samples_per_second": 28.511,
      "eval_steps_per_second": 14.256,
      "step": 10
    },
    {
      "epoch": 0.00032778466732587806,
      "grad_norm": NaN,
      "learning_rate": 9.619397662556435e-05,
      "loss": 0.0,
      "step": 15
    },
    {
      "epoch": 0.00043704622310117075,
      "grad_norm": NaN,
      "learning_rate": 8.535533905932738e-05,
      "loss": 0.0,
      "step": 20
    },
    {
      "epoch": 0.00043704622310117075,
      "eval_loss": NaN,
      "eval_runtime": 675.8123,
      "eval_samples_per_second": 28.512,
      "eval_steps_per_second": 14.257,
      "step": 20
    },
    {
      "epoch": 0.0005463077788764634,
      "grad_norm": NaN,
      "learning_rate": 6.91341716182545e-05,
      "loss": 0.0,
      "step": 25
    },
    {
      "epoch": 0.0006555693346517561,
      "grad_norm": NaN,
      "learning_rate": 5e-05,
      "loss": 0.0,
      "step": 30
    },
    {
      "epoch": 0.0006555693346517561,
      "eval_loss": NaN,
      "eval_runtime": 674.8796,
      "eval_samples_per_second": 28.552,
      "eval_steps_per_second": 14.277,
      "step": 30
    },
    {
      "epoch": 0.0007648308904270488,
      "grad_norm": NaN,
      "learning_rate": 3.086582838174551e-05,
      "loss": 0.0,
      "step": 35
    },
    {
      "epoch": 0.0008740924462023415,
      "grad_norm": NaN,
      "learning_rate": 1.4644660940672627e-05,
      "loss": 0.0,
      "step": 40
    },
    {
      "epoch": 0.0008740924462023415,
      "eval_loss": NaN,
      "eval_runtime": 674.2638,
      "eval_samples_per_second": 28.578,
      "eval_steps_per_second": 14.29,
      "step": 40
    },
    {
      "epoch": 0.000983354001977634,
      "grad_norm": NaN,
      "learning_rate": 3.8060233744356633e-06,
      "loss": 0.0,
      "step": 45
    },
    {
      "epoch": 0.0010926155577529269,
      "grad_norm": NaN,
      "learning_rate": 0.0,
      "loss": 0.0,
      "step": 50
    },
    {
      "epoch": 0.0010926155577529269,
      "eval_loss": NaN,
      "eval_runtime": 672.0205,
      "eval_samples_per_second": 28.673,
      "eval_steps_per_second": 14.337,
      "step": 50
    }
  ],
  "logging_steps": 5,
  "max_steps": 50,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 13,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 549810661490688.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}