{
  "best_metric": 0.709494411945343,
  "best_model_checkpoint": "data/Llama-31-8B_task-2_60-samples_config-1/checkpoint-34",
  "epoch": 12.869565217391305,
  "eval_steps": 500,
  "global_step": 74,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.17391304347826086,
      "grad_norm": 0.7017207741737366,
      "learning_rate": 4.000000000000001e-06,
      "loss": 1.1058,
      "step": 1
    },
    {
      "epoch": 0.34782608695652173,
      "grad_norm": 0.7510694861412048,
      "learning_rate": 8.000000000000001e-06,
      "loss": 1.1306,
      "step": 2
    },
    {
      "epoch": 0.6956521739130435,
      "grad_norm": 0.7055416703224182,
      "learning_rate": 1.6000000000000003e-05,
      "loss": 1.0703,
      "step": 4
    },
    {
      "epoch": 0.8695652173913043,
      "eval_loss": 1.0688236951828003,
      "eval_runtime": 12.9792,
      "eval_samples_per_second": 0.925,
      "eval_steps_per_second": 0.925,
      "step": 5
    },
    {
      "epoch": 1.0434782608695652,
      "grad_norm": 0.6694469451904297,
      "learning_rate": 2.4e-05,
      "loss": 1.0363,
      "step": 6
    },
    {
      "epoch": 1.391304347826087,
      "grad_norm": 0.6965046525001526,
      "learning_rate": 3.2000000000000005e-05,
      "loss": 1.0292,
      "step": 8
    },
    {
      "epoch": 1.7391304347826086,
      "grad_norm": 0.5390875339508057,
      "learning_rate": 4e-05,
      "loss": 1.0003,
      "step": 10
    },
    {
      "epoch": 1.9130434782608696,
      "eval_loss": 0.9208905100822449,
      "eval_runtime": 12.9814,
      "eval_samples_per_second": 0.924,
      "eval_steps_per_second": 0.924,
      "step": 11
    },
    {
      "epoch": 2.0869565217391304,
      "grad_norm": 0.44609975814819336,
      "learning_rate": 4.8e-05,
      "loss": 0.8926,
      "step": 12
    },
    {
      "epoch": 2.4347826086956523,
      "grad_norm": 0.3073306679725647,
      "learning_rate": 5.6000000000000006e-05,
      "loss": 0.8031,
      "step": 14
    },
    {
      "epoch": 2.782608695652174,
      "grad_norm": 0.32205966114997864,
      "learning_rate": 6.400000000000001e-05,
      "loss": 0.9073,
      "step": 16
    },
    {
      "epoch": 2.9565217391304346,
      "eval_loss": 0.8262758851051331,
      "eval_runtime": 12.9798,
      "eval_samples_per_second": 0.925,
      "eval_steps_per_second": 0.925,
      "step": 17
    },
    {
      "epoch": 3.130434782608696,
      "grad_norm": 0.3703043460845947,
      "learning_rate": 7.2e-05,
      "loss": 0.7949,
      "step": 18
    },
    {
      "epoch": 3.4782608695652173,
      "grad_norm": 0.409258097410202,
      "learning_rate": 8e-05,
      "loss": 0.7455,
      "step": 20
    },
    {
      "epoch": 3.8260869565217392,
      "grad_norm": 0.31636568903923035,
      "learning_rate": 8.800000000000001e-05,
      "loss": 0.781,
      "step": 22
    },
    {
      "epoch": 4.0,
      "eval_loss": 0.7580106854438782,
      "eval_runtime": 12.9758,
      "eval_samples_per_second": 0.925,
      "eval_steps_per_second": 0.925,
      "step": 23
    },
    {
      "epoch": 4.173913043478261,
      "grad_norm": 0.2922835350036621,
      "learning_rate": 9.6e-05,
      "loss": 0.7427,
      "step": 24
    },
    {
      "epoch": 4.521739130434782,
      "grad_norm": 0.28064048290252686,
      "learning_rate": 9.999512620046522e-05,
      "loss": 0.707,
      "step": 26
    },
    {
      "epoch": 4.869565217391305,
      "grad_norm": 0.2962977886199951,
      "learning_rate": 9.995614150494293e-05,
      "loss": 0.7117,
      "step": 28
    },
    {
      "epoch": 4.869565217391305,
      "eval_loss": 0.7323630452156067,
      "eval_runtime": 12.9768,
      "eval_samples_per_second": 0.925,
      "eval_steps_per_second": 0.925,
      "step": 28
    },
    {
      "epoch": 5.217391304347826,
      "grad_norm": 0.2961689829826355,
      "learning_rate": 9.987820251299122e-05,
      "loss": 0.6258,
      "step": 30
    },
    {
      "epoch": 5.565217391304348,
      "grad_norm": 0.2860451936721802,
      "learning_rate": 9.976136999909156e-05,
      "loss": 0.6417,
      "step": 32
    },
    {
      "epoch": 5.913043478260869,
      "grad_norm": 0.30177056789398193,
      "learning_rate": 9.96057350657239e-05,
      "loss": 0.6407,
      "step": 34
    },
    {
      "epoch": 5.913043478260869,
      "eval_loss": 0.709494411945343,
      "eval_runtime": 12.9794,
      "eval_samples_per_second": 0.925,
      "eval_steps_per_second": 0.925,
      "step": 34
    },
    {
      "epoch": 6.260869565217392,
      "grad_norm": 0.30244502425193787,
      "learning_rate": 9.941141907232765e-05,
      "loss": 0.5412,
      "step": 36
    },
    {
      "epoch": 6.608695652173913,
      "grad_norm": 0.3832639157772064,
      "learning_rate": 9.917857354066931e-05,
      "loss": 0.5998,
      "step": 38
    },
    {
      "epoch": 6.956521739130435,
      "grad_norm": 0.3918859660625458,
      "learning_rate": 9.890738003669029e-05,
      "loss": 0.5381,
      "step": 40
    },
    {
      "epoch": 6.956521739130435,
      "eval_loss": 0.7098861336708069,
      "eval_runtime": 12.9761,
      "eval_samples_per_second": 0.925,
      "eval_steps_per_second": 0.925,
      "step": 40
    },
    {
      "epoch": 7.304347826086957,
      "grad_norm": 0.4506934881210327,
      "learning_rate": 9.859805002892732e-05,
      "loss": 0.5218,
      "step": 42
    },
    {
      "epoch": 7.6521739130434785,
      "grad_norm": 0.4557631313800812,
      "learning_rate": 9.825082472361557e-05,
      "loss": 0.4882,
      "step": 44
    },
    {
      "epoch": 8.0,
      "grad_norm": 0.45536312460899353,
      "learning_rate": 9.786597487660337e-05,
      "loss": 0.4646,
      "step": 46
    },
    {
      "epoch": 8.0,
      "eval_loss": 0.7408999800682068,
      "eval_runtime": 12.9818,
      "eval_samples_per_second": 0.924,
      "eval_steps_per_second": 0.924,
      "step": 46
    },
    {
      "epoch": 8.347826086956522,
      "grad_norm": 0.48201408982276917,
      "learning_rate": 9.744380058222483e-05,
      "loss": 0.4222,
      "step": 48
    },
    {
      "epoch": 8.695652173913043,
      "grad_norm": 0.5174885392189026,
      "learning_rate": 9.698463103929542e-05,
      "loss": 0.3692,
      "step": 50
    },
    {
      "epoch": 8.869565217391305,
      "eval_loss": 0.7877727150917053,
      "eval_runtime": 12.9772,
      "eval_samples_per_second": 0.925,
      "eval_steps_per_second": 0.925,
      "step": 51
    },
    {
      "epoch": 9.043478260869565,
      "grad_norm": 0.5894734263420105,
      "learning_rate": 9.648882429441257e-05,
      "loss": 0.3609,
      "step": 52
    },
    {
      "epoch": 9.391304347826088,
      "grad_norm": 0.767116129398346,
      "learning_rate": 9.595676696276172e-05,
      "loss": 0.2693,
      "step": 54
    },
    {
      "epoch": 9.73913043478261,
      "grad_norm": 0.7209894061088562,
      "learning_rate": 9.538887392664544e-05,
      "loss": 0.2543,
      "step": 56
    },
    {
      "epoch": 9.91304347826087,
      "eval_loss": 0.8772454261779785,
      "eval_runtime": 12.9771,
      "eval_samples_per_second": 0.925,
      "eval_steps_per_second": 0.925,
      "step": 57
    },
    {
      "epoch": 10.08695652173913,
      "grad_norm": 0.6873154044151306,
      "learning_rate": 9.478558801197065e-05,
      "loss": 0.2524,
      "step": 58
    },
    {
      "epoch": 10.434782608695652,
      "grad_norm": 0.7810208797454834,
      "learning_rate": 9.414737964294636e-05,
      "loss": 0.1535,
      "step": 60
    },
    {
      "epoch": 10.782608695652174,
      "grad_norm": 0.801857590675354,
      "learning_rate": 9.347474647526095e-05,
      "loss": 0.1333,
      "step": 62
    },
    {
      "epoch": 10.956521739130435,
      "eval_loss": 1.002935767173767,
      "eval_runtime": 12.9776,
      "eval_samples_per_second": 0.925,
      "eval_steps_per_second": 0.925,
      "step": 63
    },
    {
      "epoch": 11.130434782608695,
      "grad_norm": 0.7418684959411621,
      "learning_rate": 9.276821300802534e-05,
      "loss": 0.1247,
      "step": 64
    },
    {
      "epoch": 11.478260869565217,
      "grad_norm": 1.154414415359497,
      "learning_rate": 9.202833017478422e-05,
      "loss": 0.0808,
      "step": 66
    },
    {
      "epoch": 11.826086956521738,
      "grad_norm": 1.137251853942871,
      "learning_rate": 9.125567491391476e-05,
      "loss": 0.0576,
      "step": 68
    },
    {
      "epoch": 12.0,
      "eval_loss": 1.1541328430175781,
      "eval_runtime": 12.9777,
      "eval_samples_per_second": 0.925,
      "eval_steps_per_second": 0.925,
      "step": 69
    },
    {
      "epoch": 12.173913043478262,
      "grad_norm": 0.5635364055633545,
      "learning_rate": 9.045084971874738e-05,
      "loss": 0.0419,
      "step": 70
    },
    {
      "epoch": 12.521739130434783,
      "grad_norm": 1.1226357221603394,
      "learning_rate": 8.961448216775954e-05,
      "loss": 0.0388,
      "step": 72
    },
    {
      "epoch": 12.869565217391305,
      "grad_norm": 0.8407390713691711,
      "learning_rate": 8.874722443520899e-05,
      "loss": 0.03,
      "step": 74
    },
    {
      "epoch": 12.869565217391305,
      "eval_loss": 1.4092378616333008,
      "eval_runtime": 12.9815,
      "eval_samples_per_second": 0.924,
      "eval_steps_per_second": 0.924,
      "step": 74
    },
    {
      "epoch": 12.869565217391305,
      "step": 74,
      "total_flos": 5.779480386102886e+16,
      "train_loss": 0.540592097350069,
      "train_runtime": 2077.7847,
      "train_samples_per_second": 1.107,
      "train_steps_per_second": 0.12
    }
  ],
  "logging_steps": 2,
  "max_steps": 250,
  "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": 5.779480386102886e+16,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}