last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.0001396028995522237,
  "eval_steps": 3,
  "global_step": 10,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.396028995522237e-05,
      "grad_norm": NaN,
      "learning_rate": 0.0,
      "loss": 1.6888,
      "step": 1
    },
    {
      "epoch": 1.396028995522237e-05,
      "eval_loss": 1.870382308959961,
      "eval_runtime": 1878.2928,
      "eval_samples_per_second": 8.029,
      "eval_steps_per_second": 8.029,
      "step": 1
    },
    {
      "epoch": 2.792057991044474e-05,
      "grad_norm": NaN,
      "learning_rate": 0.0,
      "loss": 1.6735,
      "step": 2
    },
    {
      "epoch": 4.188086986566711e-05,
      "grad_norm": NaN,
      "learning_rate": 0.0,
      "loss": 1.8269,
      "step": 3
    },
    {
      "epoch": 4.188086986566711e-05,
      "eval_loss": 1.870382308959961,
      "eval_runtime": 1814.1651,
      "eval_samples_per_second": 8.313,
      "eval_steps_per_second": 8.313,
      "step": 3
    },
    {
      "epoch": 5.584115982088948e-05,
      "grad_norm": 0.47626373171806335,
      "learning_rate": 2e-05,
      "loss": 1.6525,
      "step": 4
    },
    {
      "epoch": 6.980144977611185e-05,
      "grad_norm": NaN,
      "learning_rate": 2e-05,
      "loss": 1.6793,
      "step": 5
    },
    {
      "epoch": 8.376173973133422e-05,
      "grad_norm": 0.632369339466095,
      "learning_rate": 4e-05,
      "loss": 1.5735,
      "step": 6
    },
    {
      "epoch": 8.376173973133422e-05,
      "eval_loss": 1.8685113191604614,
      "eval_runtime": 1813.0543,
      "eval_samples_per_second": 8.318,
      "eval_steps_per_second": 8.318,
      "step": 6
    },
    {
      "epoch": 9.77220296865566e-05,
      "grad_norm": NaN,
      "learning_rate": 4e-05,
      "loss": 1.7572,
      "step": 7
    },
    {
      "epoch": 0.00011168231964177897,
      "grad_norm": 0.616361141204834,
      "learning_rate": 6e-05,
      "loss": 1.8794,
      "step": 8
    },
    {
      "epoch": 0.00012564260959700132,
      "grad_norm": NaN,
      "learning_rate": 6e-05,
      "loss": 1.6309,
      "step": 9
    },
    {
      "epoch": 0.00012564260959700132,
      "eval_loss": 1.8624062538146973,
      "eval_runtime": 1806.2004,
      "eval_samples_per_second": 8.35,
      "eval_steps_per_second": 8.35,
      "step": 9
    },
    {
      "epoch": 0.0001396028995522237,
      "grad_norm": 0.5328972935676575,
      "learning_rate": 8e-05,
      "loss": 1.7707,
      "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": 1628733135912960.0,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}