{
  "best_metric": 1.3489488363265991,
  "best_model_checkpoint": "data/Gemma-2-2B_task-1_60-samples_config-2/checkpoint-23",
  "epoch": 14.956521739130435,
  "eval_steps": 500,
  "global_step": 43,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.34782608695652173,
      "grad_norm": 2.729773759841919,
      "learning_rate": 1e-05,
      "loss": 2.4792,
      "step": 1
    },
    {
      "epoch": 0.6956521739130435,
      "grad_norm": 2.711193084716797,
      "learning_rate": 2e-05,
      "loss": 2.491,
      "step": 2
    },
    {
      "epoch": 0.6956521739130435,
      "eval_loss": 2.296361207962036,
      "eval_runtime": 1.1681,
      "eval_samples_per_second": 10.273,
      "eval_steps_per_second": 10.273,
      "step": 2
    },
    {
      "epoch": 1.391304347826087,
      "grad_norm": 2.750892400741577,
      "learning_rate": 4e-05,
      "loss": 2.421,
      "step": 4
    },
    {
      "epoch": 1.7391304347826086,
      "eval_loss": 2.1329658031463623,
      "eval_runtime": 1.1361,
      "eval_samples_per_second": 10.562,
      "eval_steps_per_second": 10.562,
      "step": 5
    },
    {
      "epoch": 2.0869565217391304,
      "grad_norm": 2.629638671875,
      "learning_rate": 6e-05,
      "loss": 2.2711,
      "step": 6
    },
    {
      "epoch": 2.782608695652174,
      "grad_norm": 2.0117809772491455,
      "learning_rate": 8e-05,
      "loss": 2.0067,
      "step": 8
    },
    {
      "epoch": 2.782608695652174,
      "eval_loss": 1.716484546661377,
      "eval_runtime": 1.149,
      "eval_samples_per_second": 10.444,
      "eval_steps_per_second": 10.444,
      "step": 8
    },
    {
      "epoch": 3.4782608695652173,
      "grad_norm": 1.4361461400985718,
      "learning_rate": 0.0001,
      "loss": 1.6227,
      "step": 10
    },
    {
      "epoch": 3.8260869565217392,
      "eval_loss": 1.510815143585205,
      "eval_runtime": 1.1023,
      "eval_samples_per_second": 10.886,
      "eval_steps_per_second": 10.886,
      "step": 11
    },
    {
      "epoch": 4.173913043478261,
      "grad_norm": 1.3543763160705566,
      "learning_rate": 9.987820251299122e-05,
      "loss": 1.5417,
      "step": 12
    },
    {
      "epoch": 4.869565217391305,
      "grad_norm": 1.1106294393539429,
      "learning_rate": 9.951340343707852e-05,
      "loss": 1.4202,
      "step": 14
    },
    {
      "epoch": 4.869565217391305,
      "eval_loss": 1.4400454759597778,
      "eval_runtime": 1.149,
      "eval_samples_per_second": 10.444,
      "eval_steps_per_second": 10.444,
      "step": 14
    },
    {
      "epoch": 5.565217391304348,
      "grad_norm": 0.9278581738471985,
      "learning_rate": 9.890738003669029e-05,
      "loss": 1.3321,
      "step": 16
    },
    {
      "epoch": 5.913043478260869,
      "eval_loss": 1.3851784467697144,
      "eval_runtime": 1.1603,
      "eval_samples_per_second": 10.342,
      "eval_steps_per_second": 10.342,
      "step": 17
    },
    {
      "epoch": 6.260869565217392,
      "grad_norm": 0.7903153896331787,
      "learning_rate": 9.806308479691595e-05,
      "loss": 1.1657,
      "step": 18
    },
    {
      "epoch": 6.956521739130435,
      "grad_norm": 0.7487183809280396,
      "learning_rate": 9.698463103929542e-05,
      "loss": 1.1652,
      "step": 20
    },
    {
      "epoch": 6.956521739130435,
      "eval_loss": 1.3694738149642944,
      "eval_runtime": 1.1521,
      "eval_samples_per_second": 10.416,
      "eval_steps_per_second": 10.416,
      "step": 20
    },
    {
      "epoch": 7.6521739130434785,
      "grad_norm": 0.7358795404434204,
      "learning_rate": 9.567727288213005e-05,
      "loss": 1.0665,
      "step": 22
    },
    {
      "epoch": 8.0,
      "eval_loss": 1.3489488363265991,
      "eval_runtime": 1.1893,
      "eval_samples_per_second": 10.09,
      "eval_steps_per_second": 10.09,
      "step": 23
    },
    {
      "epoch": 8.347826086956522,
      "grad_norm": 0.7612464427947998,
      "learning_rate": 9.414737964294636e-05,
      "loss": 0.967,
      "step": 24
    },
    {
      "epoch": 8.695652173913043,
      "eval_loss": 1.3493938446044922,
      "eval_runtime": 1.1395,
      "eval_samples_per_second": 10.531,
      "eval_steps_per_second": 10.531,
      "step": 25
    },
    {
      "epoch": 9.043478260869565,
      "grad_norm": 0.8194844722747803,
      "learning_rate": 9.24024048078213e-05,
      "loss": 0.9132,
      "step": 26
    },
    {
      "epoch": 9.73913043478261,
      "grad_norm": 0.778131365776062,
      "learning_rate": 9.045084971874738e-05,
      "loss": 0.8164,
      "step": 28
    },
    {
      "epoch": 9.73913043478261,
      "eval_loss": 1.3844507932662964,
      "eval_runtime": 1.1279,
      "eval_samples_per_second": 10.639,
      "eval_steps_per_second": 10.639,
      "step": 28
    },
    {
      "epoch": 10.434782608695652,
      "grad_norm": 0.8459815382957458,
      "learning_rate": 8.83022221559489e-05,
      "loss": 0.7346,
      "step": 30
    },
    {
      "epoch": 10.782608695652174,
      "eval_loss": 1.439292311668396,
      "eval_runtime": 1.2495,
      "eval_samples_per_second": 9.604,
      "eval_steps_per_second": 9.604,
      "step": 31
    },
    {
      "epoch": 11.130434782608695,
      "grad_norm": 1.00431489944458,
      "learning_rate": 8.596699001693255e-05,
      "loss": 0.7148,
      "step": 32
    },
    {
      "epoch": 11.826086956521738,
      "grad_norm": 0.956935465335846,
      "learning_rate": 8.345653031794292e-05,
      "loss": 0.5691,
      "step": 34
    },
    {
      "epoch": 11.826086956521738,
      "eval_loss": 1.5134762525558472,
      "eval_runtime": 1.0581,
      "eval_samples_per_second": 11.342,
      "eval_steps_per_second": 11.342,
      "step": 34
    },
    {
      "epoch": 12.521739130434783,
      "grad_norm": 1.0083123445510864,
      "learning_rate": 8.07830737662829e-05,
      "loss": 0.4731,
      "step": 36
    },
    {
      "epoch": 12.869565217391305,
      "eval_loss": 1.6298284530639648,
      "eval_runtime": 1.0473,
      "eval_samples_per_second": 11.458,
      "eval_steps_per_second": 11.458,
      "step": 37
    },
    {
      "epoch": 13.217391304347826,
      "grad_norm": 1.2033472061157227,
      "learning_rate": 7.795964517353735e-05,
      "loss": 0.4732,
      "step": 38
    },
    {
      "epoch": 13.91304347826087,
      "grad_norm": 1.0522940158843994,
      "learning_rate": 7.500000000000001e-05,
      "loss": 0.3309,
      "step": 40
    },
    {
      "epoch": 13.91304347826087,
      "eval_loss": 1.8103262186050415,
      "eval_runtime": 1.1704,
      "eval_samples_per_second": 10.253,
      "eval_steps_per_second": 10.253,
      "step": 40
    },
    {
      "epoch": 14.608695652173914,
      "grad_norm": 0.9777385592460632,
      "learning_rate": 7.191855733945387e-05,
      "loss": 0.2569,
      "step": 42
    },
    {
      "epoch": 14.956521739130435,
      "eval_loss": 2.0895440578460693,
      "eval_runtime": 1.1482,
      "eval_samples_per_second": 10.451,
      "eval_steps_per_second": 10.451,
      "step": 43
    },
    {
      "epoch": 14.956521739130435,
      "step": 43,
      "total_flos": 2676459427069952.0,
      "train_loss": 1.1577308184878772,
      "train_runtime": 317.644,
      "train_samples_per_second": 7.241,
      "train_steps_per_second": 0.315
    }
  ],
  "logging_steps": 2,
  "max_steps": 100,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 50,
  "save_steps": 25,
  "stateful_callbacks": {
    "EarlyStoppingCallback": {
      "args": {
        "early_stopping_patience": 7,
        "early_stopping_threshold": 0.0
      },
      "attributes": {
        "early_stopping_patience_counter": 0
      }
    },
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 2676459427069952.0,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}