{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.004518344478583047,
  "global_step": 25,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "compression/statistics/magnitude_sparsity/sparsity_level_for_model": 0.7019828954219413,
      "compression/statistics/magnitude_sparsity/sparsity_level_for_sparsified_layers": 0.8999995714352454,
      "compression/statistics/magnitude_sparsity/target_sparsity_level": 0.0,
      "compression_loss": 0.0,
      "epoch": 0.0,
      "label_loss": 0.5308924913406372,
      "learning_rate": 4.2857142857142855e-06,
      "loss": 1.8408,
      "step": 1,
      "teacher_loss": 1.9863686561584473
    },
    {
      "compression/statistics/magnitude_sparsity/sparsity_level_for_model": 0.7019828954219413,
      "compression/statistics/magnitude_sparsity/sparsity_level_for_sparsified_layers": 0.8999995714352454,
      "compression/statistics/magnitude_sparsity/target_sparsity_level": 0.0,
      "compression_loss": 0.0,
      "epoch": 0.0,
      "label_loss": 0.7996352910995483,
      "learning_rate": 8.571428571428571e-06,
      "loss": 1.9171,
      "step": 2,
      "teacher_loss": 2.0412771701812744
    },
    {
      "compression/statistics/magnitude_sparsity/sparsity_level_for_model": 0.7019828954219413,
      "compression/statistics/magnitude_sparsity/sparsity_level_for_sparsified_layers": 0.8999995714352454,
      "compression/statistics/magnitude_sparsity/target_sparsity_level": 0.0,
      "compression_loss": 0.0,
      "epoch": 0.0,
      "label_loss": 0.7973085641860962,
      "learning_rate": 1.2857142857142857e-05,
      "loss": 2.0705,
      "step": 3,
      "teacher_loss": 2.2119908332824707
    },
    {
      "compression/statistics/magnitude_sparsity/sparsity_level_for_model": 0.7019828954219413,
      "compression/statistics/magnitude_sparsity/sparsity_level_for_sparsified_layers": 0.8999995714352454,
      "compression/statistics/magnitude_sparsity/target_sparsity_level": 0.0,
      "compression_loss": 0.0,
      "epoch": 0.0,
      "label_loss": 0.9594857692718506,
      "learning_rate": 1.7142857142857142e-05,
      "loss": 1.8092,
      "step": 4,
      "teacher_loss": 1.9035913944244385
    },
    {
      "compression/statistics/magnitude_sparsity/sparsity_level_for_model": 0.7019828954219413,
      "compression/statistics/magnitude_sparsity/sparsity_level_for_sparsified_layers": 0.8999995714352454,
      "compression/statistics/magnitude_sparsity/target_sparsity_level": 0.0,
      "compression_loss": 0.0,
      "epoch": 0.0,
      "label_loss": 1.0630263090133667,
      "learning_rate": 2.1428571428571428e-05,
      "loss": 1.7453,
      "step": 5,
      "teacher_loss": 1.821063756942749
    },
    {
      "compression/statistics/magnitude_sparsity/sparsity_level_for_model": 0.7019828954219413,
      "compression/statistics/magnitude_sparsity/sparsity_level_for_sparsified_layers": 0.8999995714352454,
      "compression/statistics/magnitude_sparsity/target_sparsity_level": 0.0,
      "compression_loss": 0.0,
      "epoch": 0.0,
      "label_loss": 0.7821776866912842,
      "learning_rate": 2.5714285714285714e-05,
      "loss": 1.8497,
      "step": 6,
      "teacher_loss": 1.9682769775390625
    },
    {
      "compression/statistics/magnitude_sparsity/sparsity_level_for_model": 0.7019828954219413,
      "compression/statistics/magnitude_sparsity/sparsity_level_for_sparsified_layers": 0.8999995714352454,
      "compression/statistics/magnitude_sparsity/target_sparsity_level": 0.0,
      "compression_loss": 0.0,
      "epoch": 0.0,
      "label_loss": 0.8524608612060547,
      "learning_rate": 3e-05,
      "loss": 1.4181,
      "step": 7,
      "teacher_loss": 1.480897307395935
    },
    {
      "compression/statistics/magnitude_sparsity/sparsity_level_for_model": 0.7019828954219413,
      "compression/statistics/magnitude_sparsity/sparsity_level_for_sparsified_layers": 0.8999995714352454,
      "compression/statistics/magnitude_sparsity/target_sparsity_level": 0.0,
      "compression_loss": 0.0,
      "epoch": 0.0,
      "label_loss": 0.6336731314659119,
      "learning_rate": 2.977211629518312e-05,
      "loss": 1.948,
      "step": 8,
      "teacher_loss": 2.094085693359375
    },
    {
      "compression/statistics/magnitude_sparsity/sparsity_level_for_model": 0.7019828954219413,
      "compression/statistics/magnitude_sparsity/sparsity_level_for_sparsified_layers": 0.8999995714352454,
      "compression/statistics/magnitude_sparsity/target_sparsity_level": 0.0,
      "compression_loss": 0.0,
      "epoch": 0.0,
      "label_loss": 0.6002183556556702,
      "learning_rate": 2.9095389311788626e-05,
      "loss": 1.6932,
      "step": 9,
      "teacher_loss": 1.8146612644195557
    },
    {
      "compression/statistics/magnitude_sparsity/sparsity_level_for_model": 0.7019828954219413,
      "compression/statistics/magnitude_sparsity/sparsity_level_for_sparsified_layers": 0.8999995714352454,
      "compression/statistics/magnitude_sparsity/target_sparsity_level": 0.0,
      "compression_loss": 0.0,
      "epoch": 0.0,
      "label_loss": 0.5702627301216125,
      "learning_rate": 2.7990381056766583e-05,
      "loss": 1.2117,
      "step": 10,
      "teacher_loss": 1.2829853296279907
    },
    {
      "compression/statistics/magnitude_sparsity/sparsity_level_for_model": 0.7019828954219413,
      "compression/statistics/magnitude_sparsity/sparsity_level_for_sparsified_layers": 0.8999995714352454,
      "compression/statistics/magnitude_sparsity/target_sparsity_level": 0.0,
      "compression_loss": 0.0,
      "epoch": 0.0,
      "label_loss": 0.8245307207107544,
      "learning_rate": 2.649066664678467e-05,
      "loss": 1.353,
      "step": 11,
      "teacher_loss": 1.4116955995559692
    },
    {
      "compression/statistics/magnitude_sparsity/sparsity_level_for_model": 0.7019828954219413,
      "compression/statistics/magnitude_sparsity/sparsity_level_for_sparsified_layers": 0.8999995714352454,
      "compression/statistics/magnitude_sparsity/target_sparsity_level": 0.0,
      "compression_loss": 0.0,
      "epoch": 0.0,
      "label_loss": 0.4113720655441284,
      "learning_rate": 2.464181414529809e-05,
      "loss": 1.4131,
      "step": 12,
      "teacher_loss": 1.524428129196167
    },
    {
      "compression/statistics/magnitude_sparsity/sparsity_level_for_model": 0.7019828954219413,
      "compression/statistics/magnitude_sparsity/sparsity_level_for_sparsified_layers": 0.8999995714352454,
      "compression/statistics/magnitude_sparsity/target_sparsity_level": 0.0,
      "compression_loss": 0.0,
      "epoch": 0.0,
      "label_loss": 0.5899919867515564,
      "learning_rate": 2.25e-05,
      "loss": 1.748,
      "step": 13,
      "teacher_loss": 1.876697301864624
    },
    {
      "compression/statistics/magnitude_sparsity/sparsity_level_for_model": 0.7019828954219413,
      "compression/statistics/magnitude_sparsity/sparsity_level_for_sparsified_layers": 0.8999995714352454,
      "compression/statistics/magnitude_sparsity/target_sparsity_level": 0.0,
      "compression_loss": 0.0,
      "epoch": 0.0,
      "label_loss": 0.7763683199882507,
      "learning_rate": 2.0130302149885033e-05,
      "loss": 1.2634,
      "step": 14,
      "teacher_loss": 1.317491054534912
    },
    {
      "compression/statistics/magnitude_sparsity/sparsity_level_for_model": 0.7019828954219413,
      "compression/statistics/magnitude_sparsity/sparsity_level_for_sparsified_layers": 0.8999995714352454,
      "compression/statistics/magnitude_sparsity/target_sparsity_level": 0.0,
      "compression_loss": 0.0,
      "epoch": 0.0,
      "label_loss": 1.256364107131958,
      "learning_rate": 1.760472266500396e-05,
      "loss": 1.8216,
      "step": 15,
      "teacher_loss": 1.884453296661377
    },
    {
      "compression/statistics/magnitude_sparsity/sparsity_level_for_model": 0.7019828954219413,
      "compression/statistics/magnitude_sparsity/sparsity_level_for_sparsified_layers": 0.8999995714352454,
      "compression/statistics/magnitude_sparsity/target_sparsity_level": 0.0,
      "compression_loss": 0.0,
      "epoch": 0.0,
      "label_loss": 0.630717396736145,
      "learning_rate": 1.5e-05,
      "loss": 1.052,
      "step": 16,
      "teacher_loss": 1.0988600254058838
    },
    {
      "compression/statistics/magnitude_sparsity/sparsity_level_for_model": 0.7019828954219413,
      "compression/statistics/magnitude_sparsity/sparsity_level_for_sparsified_layers": 0.8999995714352454,
      "compression/statistics/magnitude_sparsity/target_sparsity_level": 0.0,
      "compression_loss": 0.0,
      "epoch": 0.0,
      "label_loss": 2.1178719997406006,
      "learning_rate": 1.2395277334996045e-05,
      "loss": 1.768,
      "step": 17,
      "teacher_loss": 1.729122519493103
    },
    {
      "compression/statistics/magnitude_sparsity/sparsity_level_for_model": 0.7019828954219413,
      "compression/statistics/magnitude_sparsity/sparsity_level_for_sparsified_layers": 0.8999995714352454,
      "compression/statistics/magnitude_sparsity/target_sparsity_level": 0.0,
      "compression_loss": 0.0,
      "epoch": 0.0,
      "label_loss": 0.33149364590644836,
      "learning_rate": 9.86969785011497e-06,
      "loss": 1.0413,
      "step": 18,
      "teacher_loss": 1.1201426982879639
    },
    {
      "compression/statistics/magnitude_sparsity/sparsity_level_for_model": 0.7019828954219413,
      "compression/statistics/magnitude_sparsity/sparsity_level_for_sparsified_layers": 0.8999995714352454,
      "compression/statistics/magnitude_sparsity/target_sparsity_level": 0.0,
      "compression_loss": 0.0,
      "epoch": 0.0,
      "label_loss": 1.1079105138778687,
      "learning_rate": 7.500000000000004e-06,
      "loss": 1.718,
      "step": 19,
      "teacher_loss": 1.7857446670532227
    },
    {
      "compression/statistics/magnitude_sparsity/sparsity_level_for_model": 0.7019828954219413,
      "compression/statistics/magnitude_sparsity/sparsity_level_for_sparsified_layers": 0.8999995714352454,
      "compression/statistics/magnitude_sparsity/target_sparsity_level": 0.0,
      "compression_loss": 0.0,
      "epoch": 0.0,
      "label_loss": 1.143695592880249,
      "learning_rate": 5.3581858547019095e-06,
      "loss": 1.0575,
      "step": 20,
      "teacher_loss": 1.047965407371521
    },
    {
      "compression/statistics/magnitude_sparsity/sparsity_level_for_model": 0.7019828954219413,
      "compression/statistics/magnitude_sparsity/sparsity_level_for_sparsified_layers": 0.8999995714352454,
      "compression/statistics/magnitude_sparsity/target_sparsity_level": 0.0,
      "compression_loss": 0.0,
      "epoch": 0.0,
      "label_loss": 0.5923559665679932,
      "learning_rate": 3.5093333532153316e-06,
      "loss": 0.9105,
      "step": 21,
      "teacher_loss": 0.9459026455879211
    },
    {
      "compression/statistics/magnitude_sparsity/sparsity_level_for_model": 0.7019828954219413,
      "compression/statistics/magnitude_sparsity/sparsity_level_for_sparsified_layers": 0.8999995714352454,
      "compression/statistics/magnitude_sparsity/target_sparsity_level": 0.0,
      "compression_loss": 0.0,
      "epoch": 0.0,
      "label_loss": 0.4449232816696167,
      "learning_rate": 2.0096189432334194e-06,
      "loss": 0.9125,
      "step": 22,
      "teacher_loss": 0.9643977284431458
    },
    {
      "compression/statistics/magnitude_sparsity/sparsity_level_for_model": 0.7019828954219413,
      "compression/statistics/magnitude_sparsity/sparsity_level_for_sparsified_layers": 0.8999995714352454,
      "compression/statistics/magnitude_sparsity/target_sparsity_level": 0.0,
      "compression_loss": 0.0,
      "epoch": 0.0,
      "label_loss": 1.3087965250015259,
      "learning_rate": 9.046106882113753e-07,
      "loss": 1.5437,
      "step": 23,
      "teacher_loss": 1.5698204040527344
    },
    {
      "compression/statistics/magnitude_sparsity/sparsity_level_for_model": 0.7019828954219413,
      "compression/statistics/magnitude_sparsity/sparsity_level_for_sparsified_layers": 0.8999995714352454,
      "compression/statistics/magnitude_sparsity/target_sparsity_level": 0.0,
      "compression_loss": 0.0,
      "epoch": 0.0,
      "label_loss": 1.0976316928863525,
      "learning_rate": 2.278837048168797e-07,
      "loss": 1.6757,
      "step": 24,
      "teacher_loss": 1.7399156093597412
    },
    {
      "compression/statistics/magnitude_sparsity/sparsity_level_for_model": 0.7019828954219413,
      "compression/statistics/magnitude_sparsity/sparsity_level_for_sparsified_layers": 0.8999995714352454,
      "compression/statistics/magnitude_sparsity/target_sparsity_level": 0.0,
      "compression_loss": 0.0,
      "epoch": 0.0,
      "label_loss": 0.7453712224960327,
      "learning_rate": 0.0,
      "loss": 1.3899,
      "step": 25,
      "teacher_loss": 1.4615265130996704
    },
    {
      "epoch": 0.0,
      "step": 25,
      "total_flos": 78389685043200.0,
      "train_loss": 1.526875147819519,
      "train_runtime": 69.4951,
      "train_samples_per_second": 5.756,
      "train_steps_per_second": 0.36
    }
  ],
  "max_steps": 25,
  "num_train_epochs": 1,
  "total_flos": 78389685043200.0,
  "trial_name": null,
  "trial_params": null
}