last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.05925925925925926,
  "eval_steps": 8,
  "global_step": 10,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.005925925925925926,
      "eval_loss": 15.123552322387695,
      "eval_runtime": 5.185,
      "eval_samples_per_second": 13.693,
      "eval_steps_per_second": 6.943,
      "step": 1
    },
    {
      "epoch": 0.017777777777777778,
      "grad_norm": 12.838861465454102,
      "learning_rate": 6e-05,
      "loss": 15.0509,
      "step": 3
    },
    {
      "epoch": 0.035555555555555556,
      "grad_norm": 23.51093101501465,
      "learning_rate": 0.00012,
      "loss": 13.5938,
      "step": 6
    },
    {
      "epoch": 0.047407407407407405,
      "eval_loss": 7.749448299407959,
      "eval_runtime": 4.1079,
      "eval_samples_per_second": 17.284,
      "eval_steps_per_second": 8.764,
      "step": 8
    },
    {
      "epoch": 0.05333333333333334,
      "grad_norm": 27.35287094116211,
      "learning_rate": 0.00018,
      "loss": 9.4733,
      "step": 9
    }
  ],
  "logging_steps": 3,
  "max_steps": 30,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 10,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 3709436417802240.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}