{
  "best_metric": 0.9704542756080627,
  "best_model_checkpoint": "data/Gemma-2-2B_task-1_60-samples_config-2_auto/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": 1.7917019128799438,
      "learning_rate": 1e-05,
      "loss": 1.6556,
      "step": 1
    },
    {
      "epoch": 0.6956521739130435,
      "grad_norm": 1.750916600227356,
      "learning_rate": 2e-05,
      "loss": 1.5832,
      "step": 2
    },
    {
      "epoch": 0.6956521739130435,
      "eval_loss": 1.6094385385513306,
      "eval_runtime": 1.0683,
      "eval_samples_per_second": 11.232,
      "eval_steps_per_second": 11.232,
      "step": 2
    },
    {
      "epoch": 1.391304347826087,
      "grad_norm": 1.8976556062698364,
      "learning_rate": 4e-05,
      "loss": 1.5979,
      "step": 4
    },
    {
      "epoch": 1.7391304347826086,
      "eval_loss": 1.4841265678405762,
      "eval_runtime": 1.0646,
      "eval_samples_per_second": 11.271,
      "eval_steps_per_second": 11.271,
      "step": 5
    },
    {
      "epoch": 2.0869565217391304,
      "grad_norm": 1.7825384140014648,
      "learning_rate": 6e-05,
      "loss": 1.5783,
      "step": 6
    },
    {
      "epoch": 2.782608695652174,
      "grad_norm": 1.1850117444992065,
      "learning_rate": 8e-05,
      "loss": 1.304,
      "step": 8
    },
    {
      "epoch": 2.782608695652174,
      "eval_loss": 1.1954673528671265,
      "eval_runtime": 1.0905,
      "eval_samples_per_second": 11.004,
      "eval_steps_per_second": 11.004,
      "step": 8
    },
    {
      "epoch": 3.4782608695652173,
      "grad_norm": 1.138503909111023,
      "learning_rate": 0.0001,
      "loss": 1.1472,
      "step": 10
    },
    {
      "epoch": 3.8260869565217392,
      "eval_loss": 1.0738513469696045,
      "eval_runtime": 1.0625,
      "eval_samples_per_second": 11.294,
      "eval_steps_per_second": 11.294,
      "step": 11
    },
    {
      "epoch": 4.173913043478261,
      "grad_norm": 1.0497030019760132,
      "learning_rate": 9.987820251299122e-05,
      "loss": 1.0905,
      "step": 12
    },
    {
      "epoch": 4.869565217391305,
      "grad_norm": 0.8257514238357544,
      "learning_rate": 9.951340343707852e-05,
      "loss": 0.925,
      "step": 14
    },
    {
      "epoch": 4.869565217391305,
      "eval_loss": 1.0284658670425415,
      "eval_runtime": 1.0598,
      "eval_samples_per_second": 11.323,
      "eval_steps_per_second": 11.323,
      "step": 14
    },
    {
      "epoch": 5.565217391304348,
      "grad_norm": 0.7231950759887695,
      "learning_rate": 9.890738003669029e-05,
      "loss": 0.8884,
      "step": 16
    },
    {
      "epoch": 5.913043478260869,
      "eval_loss": 1.0012325048446655,
      "eval_runtime": 1.0964,
      "eval_samples_per_second": 10.945,
      "eval_steps_per_second": 10.945,
      "step": 17
    },
    {
      "epoch": 6.260869565217392,
      "grad_norm": 0.7096526026725769,
      "learning_rate": 9.806308479691595e-05,
      "loss": 0.8296,
      "step": 18
    },
    {
      "epoch": 6.956521739130435,
      "grad_norm": 0.668281078338623,
      "learning_rate": 9.698463103929542e-05,
      "loss": 0.7766,
      "step": 20
    },
    {
      "epoch": 6.956521739130435,
      "eval_loss": 0.9815215468406677,
      "eval_runtime": 1.0799,
      "eval_samples_per_second": 11.112,
      "eval_steps_per_second": 11.112,
      "step": 20
    },
    {
      "epoch": 7.6521739130434785,
      "grad_norm": 0.5610716342926025,
      "learning_rate": 9.567727288213005e-05,
      "loss": 0.7111,
      "step": 22
    },
    {
      "epoch": 8.0,
      "eval_loss": 0.9704542756080627,
      "eval_runtime": 1.0748,
      "eval_samples_per_second": 11.165,
      "eval_steps_per_second": 11.165,
      "step": 23
    },
    {
      "epoch": 8.347826086956522,
      "grad_norm": 0.6033831834793091,
      "learning_rate": 9.414737964294636e-05,
      "loss": 0.6764,
      "step": 24
    },
    {
      "epoch": 8.695652173913043,
      "eval_loss": 0.9787288308143616,
      "eval_runtime": 1.0794,
      "eval_samples_per_second": 11.117,
      "eval_steps_per_second": 11.117,
      "step": 25
    },
    {
      "epoch": 9.043478260869565,
      "grad_norm": 0.6936924457550049,
      "learning_rate": 9.24024048078213e-05,
      "loss": 0.6381,
      "step": 26
    },
    {
      "epoch": 9.73913043478261,
      "grad_norm": 0.7001102566719055,
      "learning_rate": 9.045084971874738e-05,
      "loss": 0.5511,
      "step": 28
    },
    {
      "epoch": 9.73913043478261,
      "eval_loss": 0.9882747530937195,
      "eval_runtime": 1.0855,
      "eval_samples_per_second": 11.055,
      "eval_steps_per_second": 11.055,
      "step": 28
    },
    {
      "epoch": 10.434782608695652,
      "grad_norm": 0.6346513628959656,
      "learning_rate": 8.83022221559489e-05,
      "loss": 0.5029,
      "step": 30
    },
    {
      "epoch": 10.782608695652174,
      "eval_loss": 1.0112475156784058,
      "eval_runtime": 1.086,
      "eval_samples_per_second": 11.05,
      "eval_steps_per_second": 11.05,
      "step": 31
    },
    {
      "epoch": 11.130434782608695,
      "grad_norm": 0.7861942052841187,
      "learning_rate": 8.596699001693255e-05,
      "loss": 0.4735,
      "step": 32
    },
    {
      "epoch": 11.826086956521738,
      "grad_norm": 0.6800621151924133,
      "learning_rate": 8.345653031794292e-05,
      "loss": 0.3922,
      "step": 34
    },
    {
      "epoch": 11.826086956521738,
      "eval_loss": 1.0753231048583984,
      "eval_runtime": 1.0844,
      "eval_samples_per_second": 11.066,
      "eval_steps_per_second": 11.066,
      "step": 34
    },
    {
      "epoch": 12.521739130434783,
      "grad_norm": 0.6756002902984619,
      "learning_rate": 8.07830737662829e-05,
      "loss": 0.3437,
      "step": 36
    },
    {
      "epoch": 12.869565217391305,
      "eval_loss": 1.1804890632629395,
      "eval_runtime": 1.1035,
      "eval_samples_per_second": 10.875,
      "eval_steps_per_second": 10.875,
      "step": 37
    },
    {
      "epoch": 13.217391304347826,
      "grad_norm": 0.8871679902076721,
      "learning_rate": 7.795964517353735e-05,
      "loss": 0.2938,
      "step": 38
    },
    {
      "epoch": 13.91304347826087,
      "grad_norm": 0.763042688369751,
      "learning_rate": 7.500000000000001e-05,
      "loss": 0.2457,
      "step": 40
    },
    {
      "epoch": 13.91304347826087,
      "eval_loss": 1.3212419748306274,
      "eval_runtime": 1.0831,
      "eval_samples_per_second": 11.079,
      "eval_steps_per_second": 11.079,
      "step": 40
    },
    {
      "epoch": 14.608695652173914,
      "grad_norm": 0.7653207778930664,
      "learning_rate": 7.191855733945387e-05,
      "loss": 0.1916,
      "step": 42
    },
    {
      "epoch": 14.956521739130435,
      "eval_loss": 1.502553939819336,
      "eval_runtime": 1.0839,
      "eval_samples_per_second": 11.071,
      "eval_steps_per_second": 11.071,
      "step": 43
    },
    {
      "epoch": 14.956521739130435,
      "step": 43,
      "total_flos": 2945762076917760.0,
      "train_loss": 0.7853282565294311,
      "train_runtime": 311.077,
      "train_samples_per_second": 7.394,
      "train_steps_per_second": 0.321
    }
  ],
  "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": 2945762076917760.0,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}