last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.14746543778801843,
  "eval_steps": 3,
  "global_step": 10,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.014746543778801843,
      "grad_norm": 2.212881326675415,
      "learning_rate": 1e-05,
      "loss": 3.5787,
      "step": 1
    },
    {
      "epoch": 0.014746543778801843,
      "eval_loss": NaN,
      "eval_runtime": 78.8407,
      "eval_samples_per_second": 1.459,
      "eval_steps_per_second": 0.736,
      "step": 1
    },
    {
      "epoch": 0.029493087557603687,
      "grad_norm": 2.1416406631469727,
      "learning_rate": 2e-05,
      "loss": 3.488,
      "step": 2
    },
    {
      "epoch": 0.04423963133640553,
      "grad_norm": 1.9657536745071411,
      "learning_rate": 3e-05,
      "loss": 3.2804,
      "step": 3
    },
    {
      "epoch": 0.04423963133640553,
      "eval_loss": NaN,
      "eval_runtime": 4.6465,
      "eval_samples_per_second": 24.75,
      "eval_steps_per_second": 12.483,
      "step": 3
    },
    {
      "epoch": 0.05898617511520737,
      "grad_norm": NaN,
      "learning_rate": 4e-05,
      "loss": 4.1871,
      "step": 4
    },
    {
      "epoch": 0.07373271889400922,
      "grad_norm": 2.0105156898498535,
      "learning_rate": 5e-05,
      "loss": 3.3107,
      "step": 5
    },
    {
      "epoch": 0.08847926267281106,
      "grad_norm": NaN,
      "learning_rate": 6e-05,
      "loss": 10.3852,
      "step": 6
    },
    {
      "epoch": 0.08847926267281106,
      "eval_loss": NaN,
      "eval_runtime": 4.6268,
      "eval_samples_per_second": 24.855,
      "eval_steps_per_second": 12.536,
      "step": 6
    },
    {
      "epoch": 0.1032258064516129,
      "grad_norm": 2.1391870975494385,
      "learning_rate": 7e-05,
      "loss": 3.2799,
      "step": 7
    },
    {
      "epoch": 0.11797235023041475,
      "grad_norm": NaN,
      "learning_rate": 8e-05,
      "loss": 3.136,
      "step": 8
    },
    {
      "epoch": 0.1327188940092166,
      "grad_norm": 2.1525681018829346,
      "learning_rate": 9e-05,
      "loss": 3.2295,
      "step": 9
    },
    {
      "epoch": 0.1327188940092166,
      "eval_loss": NaN,
      "eval_runtime": 4.6219,
      "eval_samples_per_second": 24.882,
      "eval_steps_per_second": 12.549,
      "step": 9
    },
    {
      "epoch": 0.14746543778801843,
      "grad_norm": NaN,
      "learning_rate": 0.0001,
      "loss": 4.6221,
      "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": 1.394108836872192e+16,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}