{
  "best_metric": 2.3113090991973877,
  "best_model_checkpoint": "epochmetrics/task-embedder/checkpoint-1245",
  "epoch": 15.0,
  "eval_steps": 1,
  "global_step": 1245,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "grad_norm": 6.888462066650391,
      "learning_rate": 4.666666666666667e-05,
      "loss": 5.6652,
      "step": 83
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.36489910020092603,
      "eval_loss": 4.285048484802246,
      "eval_runtime": 9.4538,
      "eval_samples_per_second": 17.348,
      "eval_steps_per_second": 2.221,
      "step": 83
    },
    {
      "epoch": 2.0,
      "grad_norm": 7.599656581878662,
      "learning_rate": 4.3333333333333334e-05,
      "loss": 3.9443,
      "step": 166
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.4281380027739251,
      "eval_loss": 3.5406606197357178,
      "eval_runtime": 9.4735,
      "eval_samples_per_second": 17.311,
      "eval_steps_per_second": 2.217,
      "step": 166
    },
    {
      "epoch": 3.0,
      "grad_norm": 7.9922637939453125,
      "learning_rate": 4e-05,
      "loss": 3.3575,
      "step": 249
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.4709654748308882,
      "eval_loss": 3.109172821044922,
      "eval_runtime": 10.2035,
      "eval_samples_per_second": 16.073,
      "eval_steps_per_second": 2.058,
      "step": 249
    },
    {
      "epoch": 4.0,
      "grad_norm": 7.56485652923584,
      "learning_rate": 3.6666666666666666e-05,
      "loss": 3.084,
      "step": 332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.4962124510230736,
      "eval_loss": 2.8743200302124023,
      "eval_runtime": 9.3602,
      "eval_samples_per_second": 17.521,
      "eval_steps_per_second": 2.244,
      "step": 332
    },
    {
      "epoch": 5.0,
      "grad_norm": 7.547119617462158,
      "learning_rate": 3.3333333333333335e-05,
      "loss": 2.8764,
      "step": 415
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.5210972307154713,
      "eval_loss": 2.7019972801208496,
      "eval_runtime": 10.277,
      "eval_samples_per_second": 15.958,
      "eval_steps_per_second": 2.043,
      "step": 415
    },
    {
      "epoch": 6.0,
      "grad_norm": 6.975924968719482,
      "learning_rate": 3e-05,
      "loss": 2.7367,
      "step": 498
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.5187872505830526,
      "eval_loss": 2.669877767562866,
      "eval_runtime": 3.8263,
      "eval_samples_per_second": 42.861,
      "eval_steps_per_second": 5.488,
      "step": 498
    },
    {
      "epoch": 7.0,
      "grad_norm": 7.427117347717285,
      "learning_rate": 2.6666666666666667e-05,
      "loss": 2.6275,
      "step": 581
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.5403870967741935,
      "eval_loss": 2.5638315677642822,
      "eval_runtime": 3.776,
      "eval_samples_per_second": 43.433,
      "eval_steps_per_second": 5.562,
      "step": 581
    },
    {
      "epoch": 8.0,
      "grad_norm": 7.4915266036987305,
      "learning_rate": 2.3333333333333336e-05,
      "loss": 2.5257,
      "step": 664
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.5430055462628752,
      "eval_loss": 2.5348384380340576,
      "eval_runtime": 3.924,
      "eval_samples_per_second": 41.794,
      "eval_steps_per_second": 5.352,
      "step": 664
    },
    {
      "epoch": 9.0,
      "grad_norm": 7.47868013381958,
      "learning_rate": 2e-05,
      "loss": 2.4742,
      "step": 747
    },
    {
      "epoch": 9.0,
      "eval_accuracy": 0.5590811583839829,
      "eval_loss": 2.4301819801330566,
      "eval_runtime": 3.7824,
      "eval_samples_per_second": 43.359,
      "eval_steps_per_second": 5.552,
      "step": 747
    },
    {
      "epoch": 10.0,
      "grad_norm": 7.228312015533447,
      "learning_rate": 1.6666666666666667e-05,
      "loss": 2.4238,
      "step": 830
    },
    {
      "epoch": 10.0,
      "eval_accuracy": 0.5576721426074799,
      "eval_loss": 2.4159433841705322,
      "eval_runtime": 3.7919,
      "eval_samples_per_second": 43.251,
      "eval_steps_per_second": 5.538,
      "step": 830
    },
    {
      "epoch": 11.0,
      "grad_norm": 7.564913272857666,
      "learning_rate": 1.3333333333333333e-05,
      "loss": 2.3516,
      "step": 913
    },
    {
      "epoch": 11.0,
      "eval_accuracy": 0.5740578439964943,
      "eval_loss": 2.3461461067199707,
      "eval_runtime": 3.8232,
      "eval_samples_per_second": 42.896,
      "eval_steps_per_second": 5.493,
      "step": 913
    },
    {
      "epoch": 12.0,
      "grad_norm": 7.104005336761475,
      "learning_rate": 1e-05,
      "loss": 2.3115,
      "step": 996
    },
    {
      "epoch": 12.0,
      "eval_accuracy": 0.572778166550035,
      "eval_loss": 2.329103469848633,
      "eval_runtime": 3.8201,
      "eval_samples_per_second": 42.93,
      "eval_steps_per_second": 5.497,
      "step": 996
    },
    {
      "epoch": 13.0,
      "grad_norm": 7.211333751678467,
      "learning_rate": 6.666666666666667e-06,
      "loss": 2.29,
      "step": 1079
    },
    {
      "epoch": 13.0,
      "eval_accuracy": 0.5698073370282396,
      "eval_loss": 2.3577311038970947,
      "eval_runtime": 3.8954,
      "eval_samples_per_second": 42.101,
      "eval_steps_per_second": 5.391,
      "step": 1079
    },
    {
      "epoch": 14.0,
      "grad_norm": 7.1609063148498535,
      "learning_rate": 3.3333333333333333e-06,
      "loss": 2.2412,
      "step": 1162
    },
    {
      "epoch": 14.0,
      "eval_accuracy": 0.5673802421477452,
      "eval_loss": 2.347292423248291,
      "eval_runtime": 3.8169,
      "eval_samples_per_second": 42.967,
      "eval_steps_per_second": 5.502,
      "step": 1162
    },
    {
      "epoch": 15.0,
      "grad_norm": 6.575444221496582,
      "learning_rate": 0.0,
      "loss": 2.245,
      "step": 1245
    },
    {
      "epoch": 15.0,
      "eval_accuracy": 0.5719677022994558,
      "eval_loss": 2.3113090991973877,
      "eval_runtime": 3.8203,
      "eval_samples_per_second": 42.928,
      "eval_steps_per_second": 5.497,
      "step": 1245
    },
    {
      "epoch": 15.0,
      "step": 1245,
      "total_flos": 2605727798784000.0,
      "train_loss": 2.876972158654148,
      "train_runtime": 3151.4946,
      "train_samples_per_second": 3.141,
      "train_steps_per_second": 0.395
    }
  ],
  "logging_steps": 1,
  "max_steps": 1245,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 15,
  "save_steps": 1,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 2605727798784000.0,
  "train_batch_size": 8,
  "trial_name": null,
  "trial_params": null
}