last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 6.953982023956468e-05,
  "eval_steps": 3,
  "global_step": 3,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 2.3179940079854893e-05,
      "grad_norm": 1.0077486038208008,
      "learning_rate": 2e-05,
      "loss": 0.451,
      "step": 1
    },
    {
      "epoch": 2.3179940079854893e-05,
      "eval_loss": 0.533516526222229,
      "eval_runtime": 976.7669,
      "eval_samples_per_second": 18.597,
      "eval_steps_per_second": 9.299,
      "step": 1
    },
    {
      "epoch": 4.6359880159709785e-05,
      "grad_norm": 1.2165539264678955,
      "learning_rate": 4e-05,
      "loss": 0.5161,
      "step": 2
    },
    {
      "epoch": 6.953982023956468e-05,
      "grad_norm": 2.0040361881256104,
      "learning_rate": 6e-05,
      "loss": 0.8641,
      "step": 3
    },
    {
      "epoch": 6.953982023956468e-05,
      "eval_loss": 0.5213726758956909,
      "eval_runtime": 974.9167,
      "eval_samples_per_second": 18.632,
      "eval_steps_per_second": 9.317,
      "step": 3
    }
  ],
  "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": false
      },
      "attributes": {}
    }
  },
  "total_flos": 1017260336480256.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}