{
  "best_metric": 1.2923505306243896,
  "best_model_checkpoint": "data/Llama-31-8B_task-1_60-samples_config-3/checkpoint-120",
  "epoch": 28.0,
  "eval_steps": 500,
  "global_step": 161,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.17391304347826086,
      "grad_norm": 1.9816622734069824,
      "learning_rate": 1.3333333333333336e-07,
      "loss": 2.073,
      "step": 1
    },
    {
      "epoch": 0.34782608695652173,
      "grad_norm": 1.8311290740966797,
      "learning_rate": 2.666666666666667e-07,
      "loss": 2.0852,
      "step": 2
    },
    {
      "epoch": 0.6956521739130435,
      "grad_norm": 1.7322129011154175,
      "learning_rate": 5.333333333333335e-07,
      "loss": 2.1691,
      "step": 4
    },
    {
      "epoch": 0.8695652173913043,
      "eval_loss": 2.0740702152252197,
      "eval_runtime": 4.8331,
      "eval_samples_per_second": 2.483,
      "eval_steps_per_second": 2.483,
      "step": 5
    },
    {
      "epoch": 1.0434782608695652,
      "grad_norm": 1.7766757011413574,
      "learning_rate": 8.000000000000001e-07,
      "loss": 2.106,
      "step": 6
    },
    {
      "epoch": 1.391304347826087,
      "grad_norm": 1.8310916423797607,
      "learning_rate": 1.066666666666667e-06,
      "loss": 2.1909,
      "step": 8
    },
    {
      "epoch": 1.7391304347826086,
      "grad_norm": 1.8460414409637451,
      "learning_rate": 1.3333333333333334e-06,
      "loss": 2.0145,
      "step": 10
    },
    {
      "epoch": 1.9130434782608696,
      "eval_loss": 2.0674192905426025,
      "eval_runtime": 4.8304,
      "eval_samples_per_second": 2.484,
      "eval_steps_per_second": 2.484,
      "step": 11
    },
    {
      "epoch": 2.0869565217391304,
      "grad_norm": 1.7287455797195435,
      "learning_rate": 1.6000000000000001e-06,
      "loss": 2.0286,
      "step": 12
    },
    {
      "epoch": 2.4347826086956523,
      "grad_norm": 2.0367045402526855,
      "learning_rate": 1.8666666666666669e-06,
      "loss": 2.1431,
      "step": 14
    },
    {
      "epoch": 2.782608695652174,
      "grad_norm": 1.8663474321365356,
      "learning_rate": 2.133333333333334e-06,
      "loss": 2.0918,
      "step": 16
    },
    {
      "epoch": 2.9565217391304346,
      "eval_loss": 2.0492312908172607,
      "eval_runtime": 4.8248,
      "eval_samples_per_second": 2.487,
      "eval_steps_per_second": 2.487,
      "step": 17
    },
    {
      "epoch": 3.130434782608696,
      "grad_norm": 1.719813346862793,
      "learning_rate": 2.4000000000000003e-06,
      "loss": 2.0165,
      "step": 18
    },
    {
      "epoch": 3.4782608695652173,
      "grad_norm": 1.7898764610290527,
      "learning_rate": 2.666666666666667e-06,
      "loss": 1.998,
      "step": 20
    },
    {
      "epoch": 3.8260869565217392,
      "grad_norm": 1.9577224254608154,
      "learning_rate": 2.9333333333333338e-06,
      "loss": 2.0838,
      "step": 22
    },
    {
      "epoch": 4.0,
      "eval_loss": 2.0192394256591797,
      "eval_runtime": 4.8214,
      "eval_samples_per_second": 2.489,
      "eval_steps_per_second": 2.489,
      "step": 23
    },
    {
      "epoch": 4.173913043478261,
      "grad_norm": 1.7197067737579346,
      "learning_rate": 3.2000000000000003e-06,
      "loss": 2.0749,
      "step": 24
    },
    {
      "epoch": 4.521739130434782,
      "grad_norm": 1.730695128440857,
      "learning_rate": 3.4666666666666672e-06,
      "loss": 2.0657,
      "step": 26
    },
    {
      "epoch": 4.869565217391305,
      "grad_norm": 1.9637165069580078,
      "learning_rate": 3.7333333333333337e-06,
      "loss": 2.0314,
      "step": 28
    },
    {
      "epoch": 4.869565217391305,
      "eval_loss": 1.9791640043258667,
      "eval_runtime": 4.8308,
      "eval_samples_per_second": 2.484,
      "eval_steps_per_second": 2.484,
      "step": 28
    },
    {
      "epoch": 5.217391304347826,
      "grad_norm": 2.0035128593444824,
      "learning_rate": 4.000000000000001e-06,
      "loss": 1.9946,
      "step": 30
    },
    {
      "epoch": 5.565217391304348,
      "grad_norm": 1.9658902883529663,
      "learning_rate": 4.266666666666668e-06,
      "loss": 1.9726,
      "step": 32
    },
    {
      "epoch": 5.913043478260869,
      "grad_norm": 1.9448286294937134,
      "learning_rate": 4.533333333333334e-06,
      "loss": 1.9775,
      "step": 34
    },
    {
      "epoch": 5.913043478260869,
      "eval_loss": 1.919035792350769,
      "eval_runtime": 4.8258,
      "eval_samples_per_second": 2.487,
      "eval_steps_per_second": 2.487,
      "step": 34
    },
    {
      "epoch": 6.260869565217392,
      "grad_norm": 1.8925299644470215,
      "learning_rate": 4.800000000000001e-06,
      "loss": 1.7775,
      "step": 36
    },
    {
      "epoch": 6.608695652173913,
      "grad_norm": 1.940888524055481,
      "learning_rate": 5.0666666666666676e-06,
      "loss": 1.9153,
      "step": 38
    },
    {
      "epoch": 6.956521739130435,
      "grad_norm": 1.9882278442382812,
      "learning_rate": 5.333333333333334e-06,
      "loss": 1.8873,
      "step": 40
    },
    {
      "epoch": 6.956521739130435,
      "eval_loss": 1.8338693380355835,
      "eval_runtime": 4.8237,
      "eval_samples_per_second": 2.488,
      "eval_steps_per_second": 2.488,
      "step": 40
    },
    {
      "epoch": 7.304347826086957,
      "grad_norm": 1.6699305772781372,
      "learning_rate": 5.600000000000001e-06,
      "loss": 1.831,
      "step": 42
    },
    {
      "epoch": 7.6521739130434785,
      "grad_norm": 1.955969214439392,
      "learning_rate": 5.8666666666666675e-06,
      "loss": 1.7779,
      "step": 44
    },
    {
      "epoch": 8.0,
      "grad_norm": 1.5607645511627197,
      "learning_rate": 6.133333333333334e-06,
      "loss": 1.7547,
      "step": 46
    },
    {
      "epoch": 8.0,
      "eval_loss": 1.7314118146896362,
      "eval_runtime": 4.8242,
      "eval_samples_per_second": 2.487,
      "eval_steps_per_second": 2.487,
      "step": 46
    },
    {
      "epoch": 8.347826086956522,
      "grad_norm": 1.6665287017822266,
      "learning_rate": 6.4000000000000006e-06,
      "loss": 1.725,
      "step": 48
    },
    {
      "epoch": 8.695652173913043,
      "grad_norm": 1.44089674949646,
      "learning_rate": 6.666666666666667e-06,
      "loss": 1.6653,
      "step": 50
    },
    {
      "epoch": 8.869565217391305,
      "eval_loss": 1.6434863805770874,
      "eval_runtime": 4.8273,
      "eval_samples_per_second": 2.486,
      "eval_steps_per_second": 2.486,
      "step": 51
    },
    {
      "epoch": 9.043478260869565,
      "grad_norm": 1.455543041229248,
      "learning_rate": 6.9333333333333344e-06,
      "loss": 1.6278,
      "step": 52
    },
    {
      "epoch": 9.391304347826088,
      "grad_norm": 1.1698004007339478,
      "learning_rate": 7.2000000000000005e-06,
      "loss": 1.576,
      "step": 54
    },
    {
      "epoch": 9.73913043478261,
      "grad_norm": 1.042985200881958,
      "learning_rate": 7.4666666666666675e-06,
      "loss": 1.5709,
      "step": 56
    },
    {
      "epoch": 9.91304347826087,
      "eval_loss": 1.5690679550170898,
      "eval_runtime": 4.8278,
      "eval_samples_per_second": 2.486,
      "eval_steps_per_second": 2.486,
      "step": 57
    },
    {
      "epoch": 10.08695652173913,
      "grad_norm": 0.7936043739318848,
      "learning_rate": 7.733333333333334e-06,
      "loss": 1.5431,
      "step": 58
    },
    {
      "epoch": 10.434782608695652,
      "grad_norm": 0.7456988096237183,
      "learning_rate": 8.000000000000001e-06,
      "loss": 1.4372,
      "step": 60
    },
    {
      "epoch": 10.782608695652174,
      "grad_norm": 0.8631324172019958,
      "learning_rate": 8.266666666666667e-06,
      "loss": 1.533,
      "step": 62
    },
    {
      "epoch": 10.956521739130435,
      "eval_loss": 1.5253533124923706,
      "eval_runtime": 4.8305,
      "eval_samples_per_second": 2.484,
      "eval_steps_per_second": 2.484,
      "step": 63
    },
    {
      "epoch": 11.130434782608695,
      "grad_norm": 0.7893415093421936,
      "learning_rate": 8.533333333333335e-06,
      "loss": 1.5894,
      "step": 64
    },
    {
      "epoch": 11.478260869565217,
      "grad_norm": 0.7381595969200134,
      "learning_rate": 8.8e-06,
      "loss": 1.4961,
      "step": 66
    },
    {
      "epoch": 11.826086956521738,
      "grad_norm": 0.6585392355918884,
      "learning_rate": 9.066666666666667e-06,
      "loss": 1.4035,
      "step": 68
    },
    {
      "epoch": 12.0,
      "eval_loss": 1.485979676246643,
      "eval_runtime": 4.824,
      "eval_samples_per_second": 2.488,
      "eval_steps_per_second": 2.488,
      "step": 69
    },
    {
      "epoch": 12.173913043478262,
      "grad_norm": 0.639266848564148,
      "learning_rate": 9.333333333333334e-06,
      "loss": 1.3733,
      "step": 70
    },
    {
      "epoch": 12.521739130434783,
      "grad_norm": 0.7418985962867737,
      "learning_rate": 9.600000000000001e-06,
      "loss": 1.4263,
      "step": 72
    },
    {
      "epoch": 12.869565217391305,
      "grad_norm": 0.7750474214553833,
      "learning_rate": 9.866666666666668e-06,
      "loss": 1.4227,
      "step": 74
    },
    {
      "epoch": 12.869565217391305,
      "eval_loss": 1.457255244255066,
      "eval_runtime": 4.8296,
      "eval_samples_per_second": 2.485,
      "eval_steps_per_second": 2.485,
      "step": 74
    },
    {
      "epoch": 13.217391304347826,
      "grad_norm": 0.7266564965248108,
      "learning_rate": 9.999945845889795e-06,
      "loss": 1.4963,
      "step": 76
    },
    {
      "epoch": 13.565217391304348,
      "grad_norm": 0.7112168073654175,
      "learning_rate": 9.999512620046523e-06,
      "loss": 1.3112,
      "step": 78
    },
    {
      "epoch": 13.91304347826087,
      "grad_norm": 0.6581439971923828,
      "learning_rate": 9.99864620589731e-06,
      "loss": 1.4167,
      "step": 80
    },
    {
      "epoch": 13.91304347826087,
      "eval_loss": 1.421582818031311,
      "eval_runtime": 4.8288,
      "eval_samples_per_second": 2.485,
      "eval_steps_per_second": 2.485,
      "step": 80
    },
    {
      "epoch": 14.26086956521739,
      "grad_norm": 0.7612527012825012,
      "learning_rate": 9.99734667851357e-06,
      "loss": 1.2943,
      "step": 82
    },
    {
      "epoch": 14.608695652173914,
      "grad_norm": 0.7063285708427429,
      "learning_rate": 9.995614150494293e-06,
      "loss": 1.2531,
      "step": 84
    },
    {
      "epoch": 14.956521739130435,
      "grad_norm": 0.6375510096549988,
      "learning_rate": 9.993448771956285e-06,
      "loss": 1.3733,
      "step": 86
    },
    {
      "epoch": 14.956521739130435,
      "eval_loss": 1.3883886337280273,
      "eval_runtime": 4.8268,
      "eval_samples_per_second": 2.486,
      "eval_steps_per_second": 2.486,
      "step": 86
    },
    {
      "epoch": 15.304347826086957,
      "grad_norm": 0.7057134509086609,
      "learning_rate": 9.99085073052117e-06,
      "loss": 1.2699,
      "step": 88
    },
    {
      "epoch": 15.652173913043478,
      "grad_norm": 0.6965241432189941,
      "learning_rate": 9.987820251299121e-06,
      "loss": 1.2864,
      "step": 90
    },
    {
      "epoch": 16.0,
      "grad_norm": 0.7538113594055176,
      "learning_rate": 9.984357596869369e-06,
      "loss": 1.2917,
      "step": 92
    },
    {
      "epoch": 16.0,
      "eval_loss": 1.362096905708313,
      "eval_runtime": 4.8216,
      "eval_samples_per_second": 2.489,
      "eval_steps_per_second": 2.489,
      "step": 92
    },
    {
      "epoch": 16.347826086956523,
      "grad_norm": 0.7437707781791687,
      "learning_rate": 9.980463067257437e-06,
      "loss": 1.2624,
      "step": 94
    },
    {
      "epoch": 16.695652173913043,
      "grad_norm": 0.7640065550804138,
      "learning_rate": 9.976136999909156e-06,
      "loss": 1.2393,
      "step": 96
    },
    {
      "epoch": 16.869565217391305,
      "eval_loss": 1.3432282209396362,
      "eval_runtime": 4.8287,
      "eval_samples_per_second": 2.485,
      "eval_steps_per_second": 2.485,
      "step": 97
    },
    {
      "epoch": 17.043478260869566,
      "grad_norm": 0.8291388750076294,
      "learning_rate": 9.971379769661422e-06,
      "loss": 1.2362,
      "step": 98
    },
    {
      "epoch": 17.391304347826086,
      "grad_norm": 0.7341519594192505,
      "learning_rate": 9.966191788709716e-06,
      "loss": 1.192,
      "step": 100
    },
    {
      "epoch": 17.73913043478261,
      "grad_norm": 0.8155421614646912,
      "learning_rate": 9.960573506572391e-06,
      "loss": 1.1512,
      "step": 102
    },
    {
      "epoch": 17.91304347826087,
      "eval_loss": 1.324644684791565,
      "eval_runtime": 4.8257,
      "eval_samples_per_second": 2.487,
      "eval_steps_per_second": 2.487,
      "step": 103
    },
    {
      "epoch": 18.08695652173913,
      "grad_norm": 0.8011210560798645,
      "learning_rate": 9.95452541005172e-06,
      "loss": 1.2282,
      "step": 104
    },
    {
      "epoch": 18.434782608695652,
      "grad_norm": 0.7885610461235046,
      "learning_rate": 9.948048023191728e-06,
      "loss": 1.1364,
      "step": 106
    },
    {
      "epoch": 18.782608695652176,
      "grad_norm": 0.7404261231422424,
      "learning_rate": 9.941141907232766e-06,
      "loss": 1.1361,
      "step": 108
    },
    {
      "epoch": 18.956521739130434,
      "eval_loss": 1.3080695867538452,
      "eval_runtime": 4.8288,
      "eval_samples_per_second": 2.485,
      "eval_steps_per_second": 2.485,
      "step": 109
    },
    {
      "epoch": 19.130434782608695,
      "grad_norm": 0.8348779678344727,
      "learning_rate": 9.933807660562898e-06,
      "loss": 1.2224,
      "step": 110
    },
    {
      "epoch": 19.47826086956522,
      "grad_norm": 0.8465795516967773,
      "learning_rate": 9.926045918666045e-06,
      "loss": 1.1334,
      "step": 112
    },
    {
      "epoch": 19.82608695652174,
      "grad_norm": 0.903944194316864,
      "learning_rate": 9.91785735406693e-06,
      "loss": 1.089,
      "step": 114
    },
    {
      "epoch": 20.0,
      "eval_loss": 1.2985278367996216,
      "eval_runtime": 4.8249,
      "eval_samples_per_second": 2.487,
      "eval_steps_per_second": 2.487,
      "step": 115
    },
    {
      "epoch": 20.17391304347826,
      "grad_norm": 0.8637501001358032,
      "learning_rate": 9.909242676272797e-06,
      "loss": 1.0314,
      "step": 116
    },
    {
      "epoch": 20.52173913043478,
      "grad_norm": 0.9288212656974792,
      "learning_rate": 9.90020263171194e-06,
      "loss": 1.1512,
      "step": 118
    },
    {
      "epoch": 20.869565217391305,
      "grad_norm": 0.8621946573257446,
      "learning_rate": 9.890738003669029e-06,
      "loss": 1.0272,
      "step": 120
    },
    {
      "epoch": 20.869565217391305,
      "eval_loss": 1.2923505306243896,
      "eval_runtime": 4.8264,
      "eval_samples_per_second": 2.486,
      "eval_steps_per_second": 2.486,
      "step": 120
    },
    {
      "epoch": 21.217391304347824,
      "grad_norm": 0.9357754588127136,
      "learning_rate": 9.880849612217238e-06,
      "loss": 1.0331,
      "step": 122
    },
    {
      "epoch": 21.565217391304348,
      "grad_norm": 0.9565621614456177,
      "learning_rate": 9.870538314147194e-06,
      "loss": 1.0331,
      "step": 124
    },
    {
      "epoch": 21.91304347826087,
      "grad_norm": 0.9126953482627869,
      "learning_rate": 9.859805002892733e-06,
      "loss": 1.0591,
      "step": 126
    },
    {
      "epoch": 21.91304347826087,
      "eval_loss": 1.2933709621429443,
      "eval_runtime": 4.8294,
      "eval_samples_per_second": 2.485,
      "eval_steps_per_second": 2.485,
      "step": 126
    },
    {
      "epoch": 22.26086956521739,
      "grad_norm": 0.9442431926727295,
      "learning_rate": 9.84865060845349e-06,
      "loss": 0.9782,
      "step": 128
    },
    {
      "epoch": 22.608695652173914,
      "grad_norm": 1.0432406663894653,
      "learning_rate": 9.83707609731432e-06,
      "loss": 1.0154,
      "step": 130
    },
    {
      "epoch": 22.956521739130434,
      "grad_norm": 1.1213728189468384,
      "learning_rate": 9.825082472361558e-06,
      "loss": 0.9601,
      "step": 132
    },
    {
      "epoch": 22.956521739130434,
      "eval_loss": 1.3022547960281372,
      "eval_runtime": 4.8277,
      "eval_samples_per_second": 2.486,
      "eval_steps_per_second": 2.486,
      "step": 132
    },
    {
      "epoch": 23.304347826086957,
      "grad_norm": 1.1582452058792114,
      "learning_rate": 9.812670772796113e-06,
      "loss": 0.9306,
      "step": 134
    },
    {
      "epoch": 23.652173913043477,
      "grad_norm": 1.1142038106918335,
      "learning_rate": 9.799842074043438e-06,
      "loss": 0.952,
      "step": 136
    },
    {
      "epoch": 24.0,
      "grad_norm": 1.2423542737960815,
      "learning_rate": 9.786597487660336e-06,
      "loss": 0.9245,
      "step": 138
    },
    {
      "epoch": 24.0,
      "eval_loss": 1.3152189254760742,
      "eval_runtime": 4.8216,
      "eval_samples_per_second": 2.489,
      "eval_steps_per_second": 2.489,
      "step": 138
    },
    {
      "epoch": 24.347826086956523,
      "grad_norm": 1.235395073890686,
      "learning_rate": 9.77293816123866e-06,
      "loss": 0.9178,
      "step": 140
    },
    {
      "epoch": 24.695652173913043,
      "grad_norm": 1.0877219438552856,
      "learning_rate": 9.75886527830587e-06,
      "loss": 0.8188,
      "step": 142
    },
    {
      "epoch": 24.869565217391305,
      "eval_loss": 1.325753092765808,
      "eval_runtime": 4.8235,
      "eval_samples_per_second": 2.488,
      "eval_steps_per_second": 2.488,
      "step": 143
    },
    {
      "epoch": 25.043478260869566,
      "grad_norm": 1.2893208265304565,
      "learning_rate": 9.744380058222483e-06,
      "loss": 0.9219,
      "step": 144
    },
    {
      "epoch": 25.391304347826086,
      "grad_norm": 1.412602186203003,
      "learning_rate": 9.729483756076436e-06,
      "loss": 0.8844,
      "step": 146
    },
    {
      "epoch": 25.73913043478261,
      "grad_norm": 1.337964415550232,
      "learning_rate": 9.714177662574316e-06,
      "loss": 0.8866,
      "step": 148
    },
    {
      "epoch": 25.91304347826087,
      "eval_loss": 1.3491390943527222,
      "eval_runtime": 4.8253,
      "eval_samples_per_second": 2.487,
      "eval_steps_per_second": 2.487,
      "step": 149
    },
    {
      "epoch": 26.08695652173913,
      "grad_norm": 1.539898157119751,
      "learning_rate": 9.698463103929542e-06,
      "loss": 0.7192,
      "step": 150
    },
    {
      "epoch": 26.434782608695652,
      "grad_norm": 1.4601242542266846,
      "learning_rate": 9.682341441747446e-06,
      "loss": 0.733,
      "step": 152
    },
    {
      "epoch": 26.782608695652176,
      "grad_norm": 1.4517402648925781,
      "learning_rate": 9.665814072907293e-06,
      "loss": 0.7508,
      "step": 154
    },
    {
      "epoch": 26.956521739130434,
      "eval_loss": 1.377853512763977,
      "eval_runtime": 4.8238,
      "eval_samples_per_second": 2.488,
      "eval_steps_per_second": 2.488,
      "step": 155
    },
    {
      "epoch": 27.130434782608695,
      "grad_norm": 1.503257393836975,
      "learning_rate": 9.648882429441258e-06,
      "loss": 0.784,
      "step": 156
    },
    {
      "epoch": 27.47826086956522,
      "grad_norm": 1.569117546081543,
      "learning_rate": 9.63154797841033e-06,
      "loss": 0.7133,
      "step": 158
    },
    {
      "epoch": 27.82608695652174,
      "grad_norm": 2.004321575164795,
      "learning_rate": 9.613812221777212e-06,
      "loss": 0.7961,
      "step": 160
    },
    {
      "epoch": 28.0,
      "eval_loss": 1.4175916910171509,
      "eval_runtime": 4.8249,
      "eval_samples_per_second": 2.487,
      "eval_steps_per_second": 2.487,
      "step": 161
    },
    {
      "epoch": 28.0,
      "step": 161,
      "total_flos": 1.927928876105728e+16,
      "train_loss": 1.4128068541147694,
      "train_runtime": 1506.1629,
      "train_samples_per_second": 4.581,
      "train_steps_per_second": 0.498
    }
  ],
  "logging_steps": 2,
  "max_steps": 750,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 150,
  "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": 1.927928876105728e+16,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}