{
  "best_metric": 0.8972761034965515,
  "best_model_checkpoint": "data/Llama-31-8B_task-1_60-samples_config-1_full/checkpoint-51",
  "epoch": 16.0,
  "eval_steps": 500,
  "global_step": 92,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.17391304347826086,
      "grad_norm": 2.0918707847595215,
      "learning_rate": 4.000000000000001e-06,
      "loss": 2.45,
      "step": 1
    },
    {
      "epoch": 0.34782608695652173,
      "grad_norm": 2.04691481590271,
      "learning_rate": 8.000000000000001e-06,
      "loss": 2.5615,
      "step": 2
    },
    {
      "epoch": 0.6956521739130435,
      "grad_norm": 1.5113601684570312,
      "learning_rate": 1.6000000000000003e-05,
      "loss": 2.5176,
      "step": 4
    },
    {
      "epoch": 0.8695652173913043,
      "eval_loss": 2.328240156173706,
      "eval_runtime": 4.8422,
      "eval_samples_per_second": 2.478,
      "eval_steps_per_second": 2.478,
      "step": 5
    },
    {
      "epoch": 1.0434782608695652,
      "grad_norm": 1.5090464353561401,
      "learning_rate": 2.4e-05,
      "loss": 2.4243,
      "step": 6
    },
    {
      "epoch": 1.391304347826087,
      "grad_norm": 1.785285472869873,
      "learning_rate": 3.2000000000000005e-05,
      "loss": 2.3679,
      "step": 8
    },
    {
      "epoch": 1.7391304347826086,
      "grad_norm": 1.2606208324432373,
      "learning_rate": 4e-05,
      "loss": 2.1942,
      "step": 10
    },
    {
      "epoch": 1.9130434782608696,
      "eval_loss": 1.9932295083999634,
      "eval_runtime": 4.8372,
      "eval_samples_per_second": 2.481,
      "eval_steps_per_second": 2.481,
      "step": 11
    },
    {
      "epoch": 2.0869565217391304,
      "grad_norm": 1.6891443729400635,
      "learning_rate": 4.8e-05,
      "loss": 2.1333,
      "step": 12
    },
    {
      "epoch": 2.4347826086956523,
      "grad_norm": 0.9376276135444641,
      "learning_rate": 5.6000000000000006e-05,
      "loss": 1.9619,
      "step": 14
    },
    {
      "epoch": 2.782608695652174,
      "grad_norm": 0.8051391243934631,
      "learning_rate": 6.400000000000001e-05,
      "loss": 1.8164,
      "step": 16
    },
    {
      "epoch": 2.9565217391304346,
      "eval_loss": 1.623587727546692,
      "eval_runtime": 4.8368,
      "eval_samples_per_second": 2.481,
      "eval_steps_per_second": 2.481,
      "step": 17
    },
    {
      "epoch": 3.130434782608696,
      "grad_norm": 0.9244698286056519,
      "learning_rate": 7.2e-05,
      "loss": 1.7283,
      "step": 18
    },
    {
      "epoch": 3.4782608695652173,
      "grad_norm": 1.1509673595428467,
      "learning_rate": 8e-05,
      "loss": 1.5099,
      "step": 20
    },
    {
      "epoch": 3.8260869565217392,
      "grad_norm": 1.1495587825775146,
      "learning_rate": 8.800000000000001e-05,
      "loss": 1.3441,
      "step": 22
    },
    {
      "epoch": 4.0,
      "eval_loss": 1.1448259353637695,
      "eval_runtime": 4.8374,
      "eval_samples_per_second": 2.481,
      "eval_steps_per_second": 2.481,
      "step": 23
    },
    {
      "epoch": 4.173913043478261,
      "grad_norm": 0.7694088220596313,
      "learning_rate": 9.6e-05,
      "loss": 1.1809,
      "step": 24
    },
    {
      "epoch": 4.521739130434782,
      "grad_norm": 0.5819656252861023,
      "learning_rate": 9.999512620046522e-05,
      "loss": 1.1128,
      "step": 26
    },
    {
      "epoch": 4.869565217391305,
      "grad_norm": 0.5470612049102783,
      "learning_rate": 9.995614150494293e-05,
      "loss": 0.987,
      "step": 28
    },
    {
      "epoch": 4.869565217391305,
      "eval_loss": 1.0040026903152466,
      "eval_runtime": 4.8493,
      "eval_samples_per_second": 2.475,
      "eval_steps_per_second": 2.475,
      "step": 28
    },
    {
      "epoch": 5.217391304347826,
      "grad_norm": 0.5045129060745239,
      "learning_rate": 9.987820251299122e-05,
      "loss": 0.9648,
      "step": 30
    },
    {
      "epoch": 5.565217391304348,
      "grad_norm": 0.48170214891433716,
      "learning_rate": 9.976136999909156e-05,
      "loss": 0.9783,
      "step": 32
    },
    {
      "epoch": 5.913043478260869,
      "grad_norm": 0.46512943506240845,
      "learning_rate": 9.96057350657239e-05,
      "loss": 0.9101,
      "step": 34
    },
    {
      "epoch": 5.913043478260869,
      "eval_loss": 0.9507834911346436,
      "eval_runtime": 4.8456,
      "eval_samples_per_second": 2.476,
      "eval_steps_per_second": 2.476,
      "step": 34
    },
    {
      "epoch": 6.260869565217392,
      "grad_norm": 0.43386533856391907,
      "learning_rate": 9.941141907232765e-05,
      "loss": 0.8386,
      "step": 36
    },
    {
      "epoch": 6.608695652173913,
      "grad_norm": 0.49891090393066406,
      "learning_rate": 9.917857354066931e-05,
      "loss": 0.8758,
      "step": 38
    },
    {
      "epoch": 6.956521739130435,
      "grad_norm": 0.4979923963546753,
      "learning_rate": 9.890738003669029e-05,
      "loss": 0.8517,
      "step": 40
    },
    {
      "epoch": 6.956521739130435,
      "eval_loss": 0.9197065234184265,
      "eval_runtime": 4.8505,
      "eval_samples_per_second": 2.474,
      "eval_steps_per_second": 2.474,
      "step": 40
    },
    {
      "epoch": 7.304347826086957,
      "grad_norm": 0.5483697652816772,
      "learning_rate": 9.859805002892732e-05,
      "loss": 0.8447,
      "step": 42
    },
    {
      "epoch": 7.6521739130434785,
      "grad_norm": 0.4813004732131958,
      "learning_rate": 9.825082472361557e-05,
      "loss": 0.7798,
      "step": 44
    },
    {
      "epoch": 8.0,
      "grad_norm": 0.6186925172805786,
      "learning_rate": 9.786597487660337e-05,
      "loss": 0.7732,
      "step": 46
    },
    {
      "epoch": 8.0,
      "eval_loss": 0.8985572457313538,
      "eval_runtime": 4.8425,
      "eval_samples_per_second": 2.478,
      "eval_steps_per_second": 2.478,
      "step": 46
    },
    {
      "epoch": 8.347826086956522,
      "grad_norm": 0.621152937412262,
      "learning_rate": 9.744380058222483e-05,
      "loss": 0.7231,
      "step": 48
    },
    {
      "epoch": 8.695652173913043,
      "grad_norm": 0.7259064316749573,
      "learning_rate": 9.698463103929542e-05,
      "loss": 0.7365,
      "step": 50
    },
    {
      "epoch": 8.869565217391305,
      "eval_loss": 0.8972761034965515,
      "eval_runtime": 4.8455,
      "eval_samples_per_second": 2.477,
      "eval_steps_per_second": 2.477,
      "step": 51
    },
    {
      "epoch": 9.043478260869565,
      "grad_norm": 0.8167423009872437,
      "learning_rate": 9.648882429441257e-05,
      "loss": 0.6953,
      "step": 52
    },
    {
      "epoch": 9.391304347826088,
      "grad_norm": 0.9113949537277222,
      "learning_rate": 9.595676696276172e-05,
      "loss": 0.6369,
      "step": 54
    },
    {
      "epoch": 9.73913043478261,
      "grad_norm": 0.8619420528411865,
      "learning_rate": 9.538887392664544e-05,
      "loss": 0.6133,
      "step": 56
    },
    {
      "epoch": 9.91304347826087,
      "eval_loss": 0.9108675122261047,
      "eval_runtime": 4.8391,
      "eval_samples_per_second": 2.48,
      "eval_steps_per_second": 2.48,
      "step": 57
    },
    {
      "epoch": 10.08695652173913,
      "grad_norm": 0.7065742015838623,
      "learning_rate": 9.478558801197065e-05,
      "loss": 0.6317,
      "step": 58
    },
    {
      "epoch": 10.434782608695652,
      "grad_norm": 0.6940516233444214,
      "learning_rate": 9.414737964294636e-05,
      "loss": 0.5093,
      "step": 60
    },
    {
      "epoch": 10.782608695652174,
      "grad_norm": 0.8662077188491821,
      "learning_rate": 9.347474647526095e-05,
      "loss": 0.5483,
      "step": 62
    },
    {
      "epoch": 10.956521739130435,
      "eval_loss": 0.9299518465995789,
      "eval_runtime": 4.8371,
      "eval_samples_per_second": 2.481,
      "eval_steps_per_second": 2.481,
      "step": 63
    },
    {
      "epoch": 11.130434782608695,
      "grad_norm": 0.7917555570602417,
      "learning_rate": 9.276821300802534e-05,
      "loss": 0.5342,
      "step": 64
    },
    {
      "epoch": 11.478260869565217,
      "grad_norm": 1.0706347227096558,
      "learning_rate": 9.202833017478422e-05,
      "loss": 0.4226,
      "step": 66
    },
    {
      "epoch": 11.826086956521738,
      "grad_norm": 0.8940456509590149,
      "learning_rate": 9.125567491391476e-05,
      "loss": 0.4109,
      "step": 68
    },
    {
      "epoch": 12.0,
      "eval_loss": 0.9910491108894348,
      "eval_runtime": 4.8689,
      "eval_samples_per_second": 2.465,
      "eval_steps_per_second": 2.465,
      "step": 69
    },
    {
      "epoch": 12.173913043478262,
      "grad_norm": 0.8844466805458069,
      "learning_rate": 9.045084971874738e-05,
      "loss": 0.3587,
      "step": 70
    },
    {
      "epoch": 12.521739130434783,
      "grad_norm": 1.2623813152313232,
      "learning_rate": 8.961448216775954e-05,
      "loss": 0.3159,
      "step": 72
    },
    {
      "epoch": 12.869565217391305,
      "grad_norm": 1.2042642831802368,
      "learning_rate": 8.874722443520899e-05,
      "loss": 0.285,
      "step": 74
    },
    {
      "epoch": 12.869565217391305,
      "eval_loss": 1.0814924240112305,
      "eval_runtime": 4.8399,
      "eval_samples_per_second": 2.479,
      "eval_steps_per_second": 2.479,
      "step": 74
    },
    {
      "epoch": 13.217391304347826,
      "grad_norm": 1.173231601715088,
      "learning_rate": 8.784975278258783e-05,
      "loss": 0.2746,
      "step": 76
    },
    {
      "epoch": 13.565217391304348,
      "grad_norm": 1.5600581169128418,
      "learning_rate": 8.692276703129421e-05,
      "loss": 0.1938,
      "step": 78
    },
    {
      "epoch": 13.91304347826087,
      "grad_norm": 1.2864255905151367,
      "learning_rate": 8.596699001693255e-05,
      "loss": 0.2088,
      "step": 80
    },
    {
      "epoch": 13.91304347826087,
      "eval_loss": 1.233132004737854,
      "eval_runtime": 4.841,
      "eval_samples_per_second": 2.479,
      "eval_steps_per_second": 2.479,
      "step": 80
    },
    {
      "epoch": 14.26086956521739,
      "grad_norm": 1.078246831893921,
      "learning_rate": 8.498316702566828e-05,
      "loss": 0.1525,
      "step": 82
    },
    {
      "epoch": 14.608695652173914,
      "grad_norm": 0.9325866103172302,
      "learning_rate": 8.397206521307584e-05,
      "loss": 0.1325,
      "step": 84
    },
    {
      "epoch": 14.956521739130435,
      "grad_norm": 1.9887562990188599,
      "learning_rate": 8.293447300593402e-05,
      "loss": 0.1666,
      "step": 86
    },
    {
      "epoch": 14.956521739130435,
      "eval_loss": 1.4608489274978638,
      "eval_runtime": 4.8447,
      "eval_samples_per_second": 2.477,
      "eval_steps_per_second": 2.477,
      "step": 86
    },
    {
      "epoch": 15.304347826086957,
      "grad_norm": 0.8603971004486084,
      "learning_rate": 8.18711994874345e-05,
      "loss": 0.1077,
      "step": 88
    },
    {
      "epoch": 15.652173913043478,
      "grad_norm": 0.8751214742660522,
      "learning_rate": 8.07830737662829e-05,
      "loss": 0.0945,
      "step": 90
    },
    {
      "epoch": 16.0,
      "grad_norm": 0.9891760349273682,
      "learning_rate": 7.967094433018508e-05,
      "loss": 0.1074,
      "step": 92
    },
    {
      "epoch": 16.0,
      "eval_loss": 1.5691243410110474,
      "eval_runtime": 4.8414,
      "eval_samples_per_second": 2.479,
      "eval_steps_per_second": 2.479,
      "step": 92
    },
    {
      "epoch": 16.0,
      "step": 92,
      "total_flos": 1.101673640493056e+16,
      "train_loss": 0.9230853515150754,
      "train_runtime": 865.6365,
      "train_samples_per_second": 2.657,
      "train_steps_per_second": 0.289
    }
  ],
  "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": 1.101673640493056e+16,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}