last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.02313475997686524,
  "eval_steps": 3,
  "global_step": 10,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.002313475997686524,
      "grad_norm": 0.009941955097019672,
      "learning_rate": 2e-05,
      "loss": 11.9345,
      "step": 1
    },
    {
      "epoch": 0.002313475997686524,
      "eval_loss": NaN,
      "eval_runtime": 1.6946,
      "eval_samples_per_second": 54.291,
      "eval_steps_per_second": 54.291,
      "step": 1
    },
    {
      "epoch": 0.004626951995373048,
      "grad_norm": 0.011387641541659832,
      "learning_rate": 4e-05,
      "loss": 11.9332,
      "step": 2
    },
    {
      "epoch": 0.006940427993059572,
      "grad_norm": 0.011258460581302643,
      "learning_rate": 6e-05,
      "loss": 11.93,
      "step": 3
    },
    {
      "epoch": 0.006940427993059572,
      "eval_loss": NaN,
      "eval_runtime": 1.7039,
      "eval_samples_per_second": 53.994,
      "eval_steps_per_second": 53.994,
      "step": 3
    },
    {
      "epoch": 0.009253903990746095,
      "grad_norm": 0.009373195469379425,
      "learning_rate": 8e-05,
      "loss": 11.9335,
      "step": 4
    },
    {
      "epoch": 0.01156737998843262,
      "grad_norm": 0.008356277830898762,
      "learning_rate": 0.0001,
      "loss": 11.9314,
      "step": 5
    },
    {
      "epoch": 0.013880855986119144,
      "grad_norm": 0.01255944836884737,
      "learning_rate": 0.00012,
      "loss": 11.9327,
      "step": 6
    },
    {
      "epoch": 0.013880855986119144,
      "eval_loss": NaN,
      "eval_runtime": 1.7199,
      "eval_samples_per_second": 53.49,
      "eval_steps_per_second": 53.49,
      "step": 6
    },
    {
      "epoch": 0.016194331983805668,
      "grad_norm": 0.013384313322603703,
      "learning_rate": 0.00014,
      "loss": 11.9338,
      "step": 7
    },
    {
      "epoch": 0.01850780798149219,
      "grad_norm": 0.011861835606396198,
      "learning_rate": 0.00016,
      "loss": 11.9333,
      "step": 8
    },
    {
      "epoch": 0.020821283979178717,
      "grad_norm": 0.006169832777231932,
      "learning_rate": 0.00018,
      "loss": 11.9316,
      "step": 9
    },
    {
      "epoch": 0.020821283979178717,
      "eval_loss": NaN,
      "eval_runtime": 1.723,
      "eval_samples_per_second": 53.397,
      "eval_steps_per_second": 53.397,
      "step": 9
    },
    {
      "epoch": 0.02313475997686524,
      "grad_norm": 0.010204005986452103,
      "learning_rate": 0.0002,
      "loss": 11.9312,
      "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": 7494696960.0,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}