last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.03558718861209965,
  "eval_steps": 3,
  "global_step": 10,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0035587188612099642,
      "grad_norm": NaN,
      "learning_rate": 0.0,
      "loss": 9.2341,
      "step": 1
    },
    {
      "epoch": 0.0035587188612099642,
      "eval_loss": 10.556809425354004,
      "eval_runtime": 14.4703,
      "eval_samples_per_second": 32.757,
      "eval_steps_per_second": 4.146,
      "step": 1
    },
    {
      "epoch": 0.0071174377224199285,
      "grad_norm": NaN,
      "learning_rate": 0.0,
      "loss": 8.3346,
      "step": 2
    },
    {
      "epoch": 0.010676156583629894,
      "grad_norm": NaN,
      "learning_rate": 0.0,
      "loss": 9.1849,
      "step": 3
    },
    {
      "epoch": 0.010676156583629894,
      "eval_loss": 10.556809425354004,
      "eval_runtime": 15.1727,
      "eval_samples_per_second": 31.24,
      "eval_steps_per_second": 3.954,
      "step": 3
    },
    {
      "epoch": 0.014234875444839857,
      "grad_norm": NaN,
      "learning_rate": 0.0,
      "loss": 8.7306,
      "step": 4
    },
    {
      "epoch": 0.017793594306049824,
      "grad_norm": 14.29753589630127,
      "learning_rate": 2e-05,
      "loss": 9.3229,
      "step": 5
    },
    {
      "epoch": 0.021352313167259787,
      "grad_norm": NaN,
      "learning_rate": 2e-05,
      "loss": 9.4419,
      "step": 6
    },
    {
      "epoch": 0.021352313167259787,
      "eval_loss": 10.556809425354004,
      "eval_runtime": 14.0309,
      "eval_samples_per_second": 33.783,
      "eval_steps_per_second": 4.276,
      "step": 6
    },
    {
      "epoch": 0.02491103202846975,
      "grad_norm": NaN,
      "learning_rate": 2e-05,
      "loss": 9.3528,
      "step": 7
    },
    {
      "epoch": 0.028469750889679714,
      "grad_norm": 16.465744018554688,
      "learning_rate": 4e-05,
      "loss": 10.3572,
      "step": 8
    },
    {
      "epoch": 0.03202846975088968,
      "grad_norm": 12.530702590942383,
      "learning_rate": 6e-05,
      "loss": 9.0991,
      "step": 9
    },
    {
      "epoch": 0.03202846975088968,
      "eval_loss": 9.917542457580566,
      "eval_runtime": 14.0571,
      "eval_samples_per_second": 33.72,
      "eval_steps_per_second": 4.268,
      "step": 9
    },
    {
      "epoch": 0.03558718861209965,
      "grad_norm": 11.652721405029297,
      "learning_rate": 8e-05,
      "loss": 8.351,
      "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": 5.935098268483584e+16,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}