last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.014154281670205236,
  "eval_steps": 3,
  "global_step": 10,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0014154281670205238,
      "grad_norm": 0.35757556557655334,
      "learning_rate": 2e-05,
      "loss": 6.0737,
      "step": 1
    },
    {
      "epoch": 0.0014154281670205238,
      "eval_loss": 1.550566554069519,
      "eval_runtime": 33.1784,
      "eval_samples_per_second": 8.982,
      "eval_steps_per_second": 4.491,
      "step": 1
    },
    {
      "epoch": 0.0028308563340410475,
      "grad_norm": 0.3364515006542206,
      "learning_rate": 4e-05,
      "loss": 5.9829,
      "step": 2
    },
    {
      "epoch": 0.004246284501061571,
      "grad_norm": 0.4029381573200226,
      "learning_rate": 6e-05,
      "loss": 5.6525,
      "step": 3
    },
    {
      "epoch": 0.004246284501061571,
      "eval_loss": 1.5499653816223145,
      "eval_runtime": 33.159,
      "eval_samples_per_second": 8.987,
      "eval_steps_per_second": 4.494,
      "step": 3
    },
    {
      "epoch": 0.005661712668082095,
      "grad_norm": 0.3250284194946289,
      "learning_rate": 8e-05,
      "loss": 6.2793,
      "step": 4
    },
    {
      "epoch": 0.007077140835102618,
      "grad_norm": 0.39584881067276,
      "learning_rate": 0.0001,
      "loss": 6.1844,
      "step": 5
    },
    {
      "epoch": 0.008492569002123142,
      "grad_norm": 0.46705353260040283,
      "learning_rate": 0.00012,
      "loss": 6.5416,
      "step": 6
    },
    {
      "epoch": 0.008492569002123142,
      "eval_loss": 1.5393693447113037,
      "eval_runtime": 33.188,
      "eval_samples_per_second": 8.979,
      "eval_steps_per_second": 4.49,
      "step": 6
    },
    {
      "epoch": 0.009907997169143666,
      "grad_norm": 0.4628216326236725,
      "learning_rate": 0.00014,
      "loss": 5.9544,
      "step": 7
    },
    {
      "epoch": 0.01132342533616419,
      "grad_norm": 0.6455901861190796,
      "learning_rate": 0.00016,
      "loss": 6.0813,
      "step": 8
    },
    {
      "epoch": 0.012738853503184714,
      "grad_norm": 0.606442391872406,
      "learning_rate": 0.00018,
      "loss": 6.0511,
      "step": 9
    },
    {
      "epoch": 0.012738853503184714,
      "eval_loss": 1.4868851900100708,
      "eval_runtime": 33.1917,
      "eval_samples_per_second": 8.978,
      "eval_steps_per_second": 4.489,
      "step": 9
    },
    {
      "epoch": 0.014154281670205236,
      "grad_norm": 0.6529233455657959,
      "learning_rate": 0.0002,
      "loss": 5.8752,
      "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": 6332448610713600.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}