{
  "best_metric": 0.4578779637813568,
  "best_model_checkpoint": "data/Gemma-2-2B_task-3_180-samples_config-1/checkpoint-85",
  "epoch": 12.0,
  "eval_steps": 500,
  "global_step": 204,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.058823529411764705,
      "grad_norm": 7.895924091339111,
      "learning_rate": 1.1764705882352942e-06,
      "loss": 3.679,
      "step": 1
    },
    {
      "epoch": 0.11764705882352941,
      "grad_norm": 3.5310511589050293,
      "learning_rate": 2.3529411764705885e-06,
      "loss": 2.495,
      "step": 2
    },
    {
      "epoch": 0.23529411764705882,
      "grad_norm": 7.643738746643066,
      "learning_rate": 4.705882352941177e-06,
      "loss": 3.8582,
      "step": 4
    },
    {
      "epoch": 0.35294117647058826,
      "grad_norm": 3.6761486530303955,
      "learning_rate": 7.058823529411765e-06,
      "loss": 3.5124,
      "step": 6
    },
    {
      "epoch": 0.47058823529411764,
      "grad_norm": 7.808310508728027,
      "learning_rate": 9.411764705882354e-06,
      "loss": 3.2096,
      "step": 8
    },
    {
      "epoch": 0.5882352941176471,
      "grad_norm": 10.513341903686523,
      "learning_rate": 1.1764705882352942e-05,
      "loss": 3.4439,
      "step": 10
    },
    {
      "epoch": 0.7058823529411765,
      "grad_norm": 6.278796672821045,
      "learning_rate": 1.411764705882353e-05,
      "loss": 2.6674,
      "step": 12
    },
    {
      "epoch": 0.8235294117647058,
      "grad_norm": 8.531153678894043,
      "learning_rate": 1.647058823529412e-05,
      "loss": 3.1711,
      "step": 14
    },
    {
      "epoch": 0.9411764705882353,
      "grad_norm": 3.3822810649871826,
      "learning_rate": 1.8823529411764708e-05,
      "loss": 2.5602,
      "step": 16
    },
    {
      "epoch": 1.0,
      "eval_loss": 2.2180635929107666,
      "eval_runtime": 11.3037,
      "eval_samples_per_second": 3.185,
      "eval_steps_per_second": 3.185,
      "step": 17
    },
    {
      "epoch": 1.0588235294117647,
      "grad_norm": 2.2841455936431885,
      "learning_rate": 2.1176470588235296e-05,
      "loss": 2.3249,
      "step": 18
    },
    {
      "epoch": 1.1764705882352942,
      "grad_norm": 3.8521552085876465,
      "learning_rate": 2.3529411764705884e-05,
      "loss": 1.975,
      "step": 20
    },
    {
      "epoch": 1.2941176470588236,
      "grad_norm": 2.7394566535949707,
      "learning_rate": 2.5882352941176475e-05,
      "loss": 1.6755,
      "step": 22
    },
    {
      "epoch": 1.4117647058823528,
      "grad_norm": 3.5609967708587646,
      "learning_rate": 2.823529411764706e-05,
      "loss": 1.4273,
      "step": 24
    },
    {
      "epoch": 1.5294117647058822,
      "grad_norm": 3.0554816722869873,
      "learning_rate": 3.058823529411765e-05,
      "loss": 1.1777,
      "step": 26
    },
    {
      "epoch": 1.6470588235294117,
      "grad_norm": 2.1725094318389893,
      "learning_rate": 3.294117647058824e-05,
      "loss": 1.1747,
      "step": 28
    },
    {
      "epoch": 1.7647058823529411,
      "grad_norm": 1.2551171779632568,
      "learning_rate": 3.529411764705883e-05,
      "loss": 1.0888,
      "step": 30
    },
    {
      "epoch": 1.8823529411764706,
      "grad_norm": 1.0792454481124878,
      "learning_rate": 3.7647058823529415e-05,
      "loss": 0.593,
      "step": 32
    },
    {
      "epoch": 2.0,
      "grad_norm": 0.9411457777023315,
      "learning_rate": 4e-05,
      "loss": 0.7021,
      "step": 34
    },
    {
      "epoch": 2.0,
      "eval_loss": 0.6751887798309326,
      "eval_runtime": 11.2873,
      "eval_samples_per_second": 3.189,
      "eval_steps_per_second": 3.189,
      "step": 34
    },
    {
      "epoch": 2.1176470588235294,
      "grad_norm": 0.7325714230537415,
      "learning_rate": 4.235294117647059e-05,
      "loss": 0.6256,
      "step": 36
    },
    {
      "epoch": 2.235294117647059,
      "grad_norm": 0.6819780468940735,
      "learning_rate": 4.470588235294118e-05,
      "loss": 0.69,
      "step": 38
    },
    {
      "epoch": 2.3529411764705883,
      "grad_norm": 0.9401460886001587,
      "learning_rate": 4.705882352941177e-05,
      "loss": 0.5963,
      "step": 40
    },
    {
      "epoch": 2.4705882352941178,
      "grad_norm": 0.5545613169670105,
      "learning_rate": 4.9411764705882355e-05,
      "loss": 0.6241,
      "step": 42
    },
    {
      "epoch": 2.588235294117647,
      "grad_norm": 0.4665645658969879,
      "learning_rate": 5.176470588235295e-05,
      "loss": 0.4712,
      "step": 44
    },
    {
      "epoch": 2.7058823529411766,
      "grad_norm": 0.512626588344574,
      "learning_rate": 5.411764705882353e-05,
      "loss": 0.5095,
      "step": 46
    },
    {
      "epoch": 2.8235294117647056,
      "grad_norm": 0.7327120900154114,
      "learning_rate": 5.647058823529412e-05,
      "loss": 0.4916,
      "step": 48
    },
    {
      "epoch": 2.9411764705882355,
      "grad_norm": 1.0528010129928589,
      "learning_rate": 5.882352941176471e-05,
      "loss": 0.3994,
      "step": 50
    },
    {
      "epoch": 3.0,
      "eval_loss": 0.5229716300964355,
      "eval_runtime": 11.2839,
      "eval_samples_per_second": 3.19,
      "eval_steps_per_second": 3.19,
      "step": 51
    },
    {
      "epoch": 3.0588235294117645,
      "grad_norm": 0.5702608227729797,
      "learning_rate": 6.11764705882353e-05,
      "loss": 0.3743,
      "step": 52
    },
    {
      "epoch": 3.176470588235294,
      "grad_norm": 0.8179157376289368,
      "learning_rate": 6.352941176470588e-05,
      "loss": 0.6003,
      "step": 54
    },
    {
      "epoch": 3.2941176470588234,
      "grad_norm": 0.7172646522521973,
      "learning_rate": 6.588235294117648e-05,
      "loss": 0.4653,
      "step": 56
    },
    {
      "epoch": 3.411764705882353,
      "grad_norm": 1.637542486190796,
      "learning_rate": 6.823529411764707e-05,
      "loss": 0.6361,
      "step": 58
    },
    {
      "epoch": 3.5294117647058822,
      "grad_norm": 0.382338285446167,
      "learning_rate": 7.058823529411765e-05,
      "loss": 0.1991,
      "step": 60
    },
    {
      "epoch": 3.6470588235294117,
      "grad_norm": 0.5550841689109802,
      "learning_rate": 7.294117647058823e-05,
      "loss": 0.3704,
      "step": 62
    },
    {
      "epoch": 3.764705882352941,
      "grad_norm": 0.6467725038528442,
      "learning_rate": 7.529411764705883e-05,
      "loss": 0.5009,
      "step": 64
    },
    {
      "epoch": 3.8823529411764706,
      "grad_norm": 0.6327499747276306,
      "learning_rate": 7.764705882352942e-05,
      "loss": 0.2861,
      "step": 66
    },
    {
      "epoch": 4.0,
      "grad_norm": 0.3987816572189331,
      "learning_rate": 8e-05,
      "loss": 0.291,
      "step": 68
    },
    {
      "epoch": 4.0,
      "eval_loss": 0.46538543701171875,
      "eval_runtime": 11.2871,
      "eval_samples_per_second": 3.189,
      "eval_steps_per_second": 3.189,
      "step": 68
    },
    {
      "epoch": 4.117647058823529,
      "grad_norm": 0.8739417195320129,
      "learning_rate": 8.23529411764706e-05,
      "loss": 0.434,
      "step": 70
    },
    {
      "epoch": 4.235294117647059,
      "grad_norm": 0.5957450270652771,
      "learning_rate": 8.470588235294118e-05,
      "loss": 0.3252,
      "step": 72
    },
    {
      "epoch": 4.352941176470588,
      "grad_norm": 0.7681469321250916,
      "learning_rate": 8.705882352941177e-05,
      "loss": 0.4662,
      "step": 74
    },
    {
      "epoch": 4.470588235294118,
      "grad_norm": 0.5825863480567932,
      "learning_rate": 8.941176470588236e-05,
      "loss": 0.3018,
      "step": 76
    },
    {
      "epoch": 4.588235294117647,
      "grad_norm": 0.500510036945343,
      "learning_rate": 9.176470588235295e-05,
      "loss": 0.3442,
      "step": 78
    },
    {
      "epoch": 4.705882352941177,
      "grad_norm": 0.6307154297828674,
      "learning_rate": 9.411764705882353e-05,
      "loss": 0.2709,
      "step": 80
    },
    {
      "epoch": 4.823529411764706,
      "grad_norm": 0.7237655520439148,
      "learning_rate": 9.647058823529412e-05,
      "loss": 0.3004,
      "step": 82
    },
    {
      "epoch": 4.9411764705882355,
      "grad_norm": 2.7985341548919678,
      "learning_rate": 9.882352941176471e-05,
      "loss": 0.2886,
      "step": 84
    },
    {
      "epoch": 5.0,
      "eval_loss": 0.4578779637813568,
      "eval_runtime": 11.299,
      "eval_samples_per_second": 3.186,
      "eval_steps_per_second": 3.186,
      "step": 85
    },
    {
      "epoch": 5.0588235294117645,
      "grad_norm": 0.8119005560874939,
      "learning_rate": 9.99995783847866e-05,
      "loss": 0.2726,
      "step": 86
    },
    {
      "epoch": 5.176470588235294,
      "grad_norm": 0.683082103729248,
      "learning_rate": 9.999620550574153e-05,
      "loss": 0.3215,
      "step": 88
    },
    {
      "epoch": 5.294117647058823,
      "grad_norm": 0.7073795199394226,
      "learning_rate": 9.998945997517956e-05,
      "loss": 0.2464,
      "step": 90
    },
    {
      "epoch": 5.411764705882353,
      "grad_norm": 0.049651362001895905,
      "learning_rate": 9.997934224814173e-05,
      "loss": 0.0941,
      "step": 92
    },
    {
      "epoch": 5.529411764705882,
      "grad_norm": 1.3247442245483398,
      "learning_rate": 9.996585300715116e-05,
      "loss": 0.2315,
      "step": 94
    },
    {
      "epoch": 5.647058823529412,
      "grad_norm": 0.6855947971343994,
      "learning_rate": 9.994899316216708e-05,
      "loss": 0.1483,
      "step": 96
    },
    {
      "epoch": 5.764705882352941,
      "grad_norm": 0.904195249080658,
      "learning_rate": 9.992876385052345e-05,
      "loss": 0.2587,
      "step": 98
    },
    {
      "epoch": 5.882352941176471,
      "grad_norm": 1.3182839155197144,
      "learning_rate": 9.990516643685222e-05,
      "loss": 0.3598,
      "step": 100
    },
    {
      "epoch": 6.0,
      "grad_norm": 0.974306583404541,
      "learning_rate": 9.987820251299122e-05,
      "loss": 0.3088,
      "step": 102
    },
    {
      "epoch": 6.0,
      "eval_loss": 0.4717375338077545,
      "eval_runtime": 11.3011,
      "eval_samples_per_second": 3.186,
      "eval_steps_per_second": 3.186,
      "step": 102
    },
    {
      "epoch": 6.117647058823529,
      "grad_norm": 0.7311336994171143,
      "learning_rate": 9.984787389787688e-05,
      "loss": 0.1157,
      "step": 104
    },
    {
      "epoch": 6.235294117647059,
      "grad_norm": 1.191741943359375,
      "learning_rate": 9.981418263742148e-05,
      "loss": 0.239,
      "step": 106
    },
    {
      "epoch": 6.352941176470588,
      "grad_norm": 0.7977131009101868,
      "learning_rate": 9.977713100437509e-05,
      "loss": 0.1567,
      "step": 108
    },
    {
      "epoch": 6.470588235294118,
      "grad_norm": 1.3035118579864502,
      "learning_rate": 9.973672149817232e-05,
      "loss": 0.1373,
      "step": 110
    },
    {
      "epoch": 6.588235294117647,
      "grad_norm": 0.9060225486755371,
      "learning_rate": 9.96929568447637e-05,
      "loss": 0.1037,
      "step": 112
    },
    {
      "epoch": 6.705882352941177,
      "grad_norm": 0.8160073161125183,
      "learning_rate": 9.964583999643174e-05,
      "loss": 0.12,
      "step": 114
    },
    {
      "epoch": 6.823529411764706,
      "grad_norm": 1.0023682117462158,
      "learning_rate": 9.95953741315919e-05,
      "loss": 0.1666,
      "step": 116
    },
    {
      "epoch": 6.9411764705882355,
      "grad_norm": 1.2285104990005493,
      "learning_rate": 9.954156265457801e-05,
      "loss": 0.2315,
      "step": 118
    },
    {
      "epoch": 7.0,
      "eval_loss": 0.5574997067451477,
      "eval_runtime": 11.2888,
      "eval_samples_per_second": 3.189,
      "eval_steps_per_second": 3.189,
      "step": 119
    },
    {
      "epoch": 7.0588235294117645,
      "grad_norm": 1.5922584533691406,
      "learning_rate": 9.948440919541278e-05,
      "loss": 0.2191,
      "step": 120
    },
    {
      "epoch": 7.176470588235294,
      "grad_norm": 0.9241739511489868,
      "learning_rate": 9.942391760956277e-05,
      "loss": 0.1111,
      "step": 122
    },
    {
      "epoch": 7.294117647058823,
      "grad_norm": 2.565469741821289,
      "learning_rate": 9.936009197767845e-05,
      "loss": 0.0908,
      "step": 124
    },
    {
      "epoch": 7.411764705882353,
      "grad_norm": 1.435282588005066,
      "learning_rate": 9.929293660531888e-05,
      "loss": 0.0987,
      "step": 126
    },
    {
      "epoch": 7.529411764705882,
      "grad_norm": 1.7153534889221191,
      "learning_rate": 9.922245602266118e-05,
      "loss": 0.0877,
      "step": 128
    },
    {
      "epoch": 7.647058823529412,
      "grad_norm": 0.7147380113601685,
      "learning_rate": 9.91486549841951e-05,
      "loss": 0.0518,
      "step": 130
    },
    {
      "epoch": 7.764705882352941,
      "grad_norm": 1.3303829431533813,
      "learning_rate": 9.90715384684021e-05,
      "loss": 0.0636,
      "step": 132
    },
    {
      "epoch": 7.882352941176471,
      "grad_norm": 0.6465467214584351,
      "learning_rate": 9.899111167741966e-05,
      "loss": 0.094,
      "step": 134
    },
    {
      "epoch": 8.0,
      "grad_norm": 1.525161623954773,
      "learning_rate": 9.890738003669029e-05,
      "loss": 0.0902,
      "step": 136
    },
    {
      "epoch": 8.0,
      "eval_loss": 0.6358526349067688,
      "eval_runtime": 11.2879,
      "eval_samples_per_second": 3.189,
      "eval_steps_per_second": 3.189,
      "step": 136
    },
    {
      "epoch": 8.117647058823529,
      "grad_norm": 0.70051509141922,
      "learning_rate": 9.882034919459555e-05,
      "loss": 0.0567,
      "step": 138
    },
    {
      "epoch": 8.235294117647058,
      "grad_norm": 0.9678773880004883,
      "learning_rate": 9.873002502207503e-05,
      "loss": 0.053,
      "step": 140
    },
    {
      "epoch": 8.352941176470589,
      "grad_norm": 1.0803821086883545,
      "learning_rate": 9.863641361223024e-05,
      "loss": 0.0792,
      "step": 142
    },
    {
      "epoch": 8.470588235294118,
      "grad_norm": 0.001414611004292965,
      "learning_rate": 9.853952127991372e-05,
      "loss": 0.0415,
      "step": 144
    },
    {
      "epoch": 8.588235294117647,
      "grad_norm": 1.7889593839645386,
      "learning_rate": 9.843935456130295e-05,
      "loss": 0.0753,
      "step": 146
    },
    {
      "epoch": 8.705882352941176,
      "grad_norm": 0.6375395059585571,
      "learning_rate": 9.833592021345937e-05,
      "loss": 0.0169,
      "step": 148
    },
    {
      "epoch": 8.823529411764707,
      "grad_norm": 1.5927634239196777,
      "learning_rate": 9.822922521387276e-05,
      "loss": 0.0503,
      "step": 150
    },
    {
      "epoch": 8.941176470588236,
      "grad_norm": 2.6613237857818604,
      "learning_rate": 9.811927675999036e-05,
      "loss": 0.05,
      "step": 152
    },
    {
      "epoch": 9.0,
      "eval_loss": 0.8561871647834778,
      "eval_runtime": 11.3205,
      "eval_samples_per_second": 3.18,
      "eval_steps_per_second": 3.18,
      "step": 153
    },
    {
      "epoch": 9.058823529411764,
      "grad_norm": 1.165877103805542,
      "learning_rate": 9.800608226873142e-05,
      "loss": 0.035,
      "step": 154
    },
    {
      "epoch": 9.176470588235293,
      "grad_norm": 0.9223412871360779,
      "learning_rate": 9.788964937598689e-05,
      "loss": 0.025,
      "step": 156
    },
    {
      "epoch": 9.294117647058824,
      "grad_norm": 0.7580801844596863,
      "learning_rate": 9.776998593610428e-05,
      "loss": 0.0263,
      "step": 158
    },
    {
      "epoch": 9.411764705882353,
      "grad_norm": 1.0875881910324097,
      "learning_rate": 9.764710002135784e-05,
      "loss": 0.0399,
      "step": 160
    },
    {
      "epoch": 9.529411764705882,
      "grad_norm": 2.486358642578125,
      "learning_rate": 9.752099992140399e-05,
      "loss": 0.0329,
      "step": 162
    },
    {
      "epoch": 9.647058823529411,
      "grad_norm": 0.568743884563446,
      "learning_rate": 9.739169414272217e-05,
      "loss": 0.0173,
      "step": 164
    },
    {
      "epoch": 9.764705882352942,
      "grad_norm": 1.5090088844299316,
      "learning_rate": 9.725919140804099e-05,
      "loss": 0.0426,
      "step": 166
    },
    {
      "epoch": 9.882352941176471,
      "grad_norm": 0.3534574806690216,
      "learning_rate": 9.71235006557497e-05,
      "loss": 0.0237,
      "step": 168
    },
    {
      "epoch": 10.0,
      "grad_norm": 1.2611942291259766,
      "learning_rate": 9.698463103929542e-05,
      "loss": 0.0274,
      "step": 170
    },
    {
      "epoch": 10.0,
      "eval_loss": 0.8861123323440552,
      "eval_runtime": 11.3227,
      "eval_samples_per_second": 3.179,
      "eval_steps_per_second": 3.179,
      "step": 170
    },
    {
      "epoch": 10.117647058823529,
      "grad_norm": 2.7106523513793945,
      "learning_rate": 9.684259192656553e-05,
      "loss": 0.015,
      "step": 172
    },
    {
      "epoch": 10.235294117647058,
      "grad_norm": 1.1814020872116089,
      "learning_rate": 9.669739289925577e-05,
      "loss": 0.0158,
      "step": 174
    },
    {
      "epoch": 10.352941176470589,
      "grad_norm": 1.1286207437515259,
      "learning_rate": 9.654904375222385e-05,
      "loss": 0.019,
      "step": 176
    },
    {
      "epoch": 10.470588235294118,
      "grad_norm": 1.842712163925171,
      "learning_rate": 9.639755449282875e-05,
      "loss": 0.0261,
      "step": 178
    },
    {
      "epoch": 10.588235294117647,
      "grad_norm": 1.3332819938659668,
      "learning_rate": 9.62429353402556e-05,
      "loss": 0.0485,
      "step": 180
    },
    {
      "epoch": 10.705882352941176,
      "grad_norm": 0.3820013701915741,
      "learning_rate": 9.608519672482636e-05,
      "loss": 0.0182,
      "step": 182
    },
    {
      "epoch": 10.823529411764707,
      "grad_norm": 0.8928465843200684,
      "learning_rate": 9.592434928729616e-05,
      "loss": 0.0205,
      "step": 184
    },
    {
      "epoch": 10.941176470588236,
      "grad_norm": 0.4944886565208435,
      "learning_rate": 9.576040387813552e-05,
      "loss": 0.0116,
      "step": 186
    },
    {
      "epoch": 11.0,
      "eval_loss": 1.069555640220642,
      "eval_runtime": 11.3422,
      "eval_samples_per_second": 3.174,
      "eval_steps_per_second": 3.174,
      "step": 187
    },
    {
      "epoch": 11.058823529411764,
      "grad_norm": 6.004947185516357,
      "learning_rate": 9.559337155679842e-05,
      "loss": 0.0221,
      "step": 188
    },
    {
      "epoch": 11.176470588235293,
      "grad_norm": 0.796440839767456,
      "learning_rate": 9.542326359097619e-05,
      "loss": 0.0248,
      "step": 190
    },
    {
      "epoch": 11.294117647058824,
      "grad_norm": 0.6684553623199463,
      "learning_rate": 9.525009145583745e-05,
      "loss": 0.0104,
      "step": 192
    },
    {
      "epoch": 11.411764705882353,
      "grad_norm": 0.4757600426673889,
      "learning_rate": 9.507386683325404e-05,
      "loss": 0.011,
      "step": 194
    },
    {
      "epoch": 11.529411764705882,
      "grad_norm": 0.875470757484436,
      "learning_rate": 9.489460161101291e-05,
      "loss": 0.0175,
      "step": 196
    },
    {
      "epoch": 11.647058823529411,
      "grad_norm": 0.3998018801212311,
      "learning_rate": 9.471230788201429e-05,
      "loss": 0.0024,
      "step": 198
    },
    {
      "epoch": 11.764705882352942,
      "grad_norm": 0.6310029029846191,
      "learning_rate": 9.452699794345581e-05,
      "loss": 0.0182,
      "step": 200
    },
    {
      "epoch": 11.882352941176471,
      "grad_norm": 1.035658597946167,
      "learning_rate": 9.43386842960031e-05,
      "loss": 0.0089,
      "step": 202
    },
    {
      "epoch": 12.0,
      "grad_norm": 0.794379472732544,
      "learning_rate": 9.414737964294636e-05,
      "loss": 0.0158,
      "step": 204
    },
    {
      "epoch": 12.0,
      "eval_loss": 0.9895004630088806,
      "eval_runtime": 11.3224,
      "eval_samples_per_second": 3.18,
      "eval_steps_per_second": 3.18,
      "step": 204
    },
    {
      "epoch": 12.0,
      "step": 204,
      "total_flos": 4.401056504715674e+16,
      "train_loss": 0.5287151039011923,
      "train_runtime": 1850.0584,
      "train_samples_per_second": 3.676,
      "train_steps_per_second": 0.459
    }
  ],
  "logging_steps": 2,
  "max_steps": 850,
  "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": 4.401056504715674e+16,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}