{
  "best_metric": 0.37464410066604614,
  "best_model_checkpoint": "data/Gemma-2-2B_task-3_180-samples_config-1_auto/checkpoint-68",
  "epoch": 11.0,
  "eval_steps": 500,
  "global_step": 187,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.058823529411764705,
      "grad_norm": 7.757319450378418,
      "learning_rate": 1.1764705882352942e-06,
      "loss": 3.5517,
      "step": 1
    },
    {
      "epoch": 0.11764705882352941,
      "grad_norm": 3.4050068855285645,
      "learning_rate": 2.3529411764705885e-06,
      "loss": 2.2986,
      "step": 2
    },
    {
      "epoch": 0.23529411764705882,
      "grad_norm": 7.462129592895508,
      "learning_rate": 4.705882352941177e-06,
      "loss": 3.7524,
      "step": 4
    },
    {
      "epoch": 0.35294117647058826,
      "grad_norm": 3.522613763809204,
      "learning_rate": 7.058823529411765e-06,
      "loss": 3.3844,
      "step": 6
    },
    {
      "epoch": 0.47058823529411764,
      "grad_norm": 7.847003936767578,
      "learning_rate": 9.411764705882354e-06,
      "loss": 3.0978,
      "step": 8
    },
    {
      "epoch": 0.5882352941176471,
      "grad_norm": 7.462653160095215,
      "learning_rate": 1.1764705882352942e-05,
      "loss": 3.2537,
      "step": 10
    },
    {
      "epoch": 0.7058823529411765,
      "grad_norm": 6.174149513244629,
      "learning_rate": 1.411764705882353e-05,
      "loss": 2.499,
      "step": 12
    },
    {
      "epoch": 0.8235294117647058,
      "grad_norm": 8.465475082397461,
      "learning_rate": 1.647058823529412e-05,
      "loss": 3.0587,
      "step": 14
    },
    {
      "epoch": 0.9411764705882353,
      "grad_norm": 3.148521900177002,
      "learning_rate": 1.8823529411764708e-05,
      "loss": 2.3428,
      "step": 16
    },
    {
      "epoch": 1.0,
      "eval_loss": 2.0249428749084473,
      "eval_runtime": 11.3419,
      "eval_samples_per_second": 3.174,
      "eval_steps_per_second": 3.174,
      "step": 17
    },
    {
      "epoch": 1.0588235294117647,
      "grad_norm": 2.112591505050659,
      "learning_rate": 2.1176470588235296e-05,
      "loss": 2.1465,
      "step": 18
    },
    {
      "epoch": 1.1764705882352942,
      "grad_norm": 3.6208736896514893,
      "learning_rate": 2.3529411764705884e-05,
      "loss": 1.8433,
      "step": 20
    },
    {
      "epoch": 1.2941176470588236,
      "grad_norm": 2.6201634407043457,
      "learning_rate": 2.5882352941176475e-05,
      "loss": 1.5223,
      "step": 22
    },
    {
      "epoch": 1.4117647058823528,
      "grad_norm": 3.5025103092193604,
      "learning_rate": 2.823529411764706e-05,
      "loss": 1.3362,
      "step": 24
    },
    {
      "epoch": 1.5294117647058822,
      "grad_norm": 2.9054648876190186,
      "learning_rate": 3.058823529411765e-05,
      "loss": 1.1043,
      "step": 26
    },
    {
      "epoch": 1.6470588235294117,
      "grad_norm": 2.0959935188293457,
      "learning_rate": 3.294117647058824e-05,
      "loss": 1.0914,
      "step": 28
    },
    {
      "epoch": 1.7647058823529411,
      "grad_norm": 1.236687421798706,
      "learning_rate": 3.529411764705883e-05,
      "loss": 0.8952,
      "step": 30
    },
    {
      "epoch": 1.8823529411764706,
      "grad_norm": 1.1082926988601685,
      "learning_rate": 3.7647058823529415e-05,
      "loss": 0.499,
      "step": 32
    },
    {
      "epoch": 2.0,
      "grad_norm": 0.9132806062698364,
      "learning_rate": 4e-05,
      "loss": 0.6095,
      "step": 34
    },
    {
      "epoch": 2.0,
      "eval_loss": 0.5500472784042358,
      "eval_runtime": 11.3569,
      "eval_samples_per_second": 3.17,
      "eval_steps_per_second": 3.17,
      "step": 34
    },
    {
      "epoch": 2.1176470588235294,
      "grad_norm": 0.7398793697357178,
      "learning_rate": 4.235294117647059e-05,
      "loss": 0.5307,
      "step": 36
    },
    {
      "epoch": 2.235294117647059,
      "grad_norm": 0.5786101818084717,
      "learning_rate": 4.470588235294118e-05,
      "loss": 0.521,
      "step": 38
    },
    {
      "epoch": 2.3529411764705883,
      "grad_norm": 0.8229206204414368,
      "learning_rate": 4.705882352941177e-05,
      "loss": 0.5108,
      "step": 40
    },
    {
      "epoch": 2.4705882352941178,
      "grad_norm": 0.48744791746139526,
      "learning_rate": 4.9411764705882355e-05,
      "loss": 0.5632,
      "step": 42
    },
    {
      "epoch": 2.588235294117647,
      "grad_norm": 0.40403854846954346,
      "learning_rate": 5.176470588235295e-05,
      "loss": 0.3901,
      "step": 44
    },
    {
      "epoch": 2.7058823529411766,
      "grad_norm": 0.4570540487766266,
      "learning_rate": 5.411764705882353e-05,
      "loss": 0.4198,
      "step": 46
    },
    {
      "epoch": 2.8235294117647056,
      "grad_norm": 0.6631785035133362,
      "learning_rate": 5.647058823529412e-05,
      "loss": 0.4318,
      "step": 48
    },
    {
      "epoch": 2.9411764705882355,
      "grad_norm": 0.8489178419113159,
      "learning_rate": 5.882352941176471e-05,
      "loss": 0.3433,
      "step": 50
    },
    {
      "epoch": 3.0,
      "eval_loss": 0.41212335228919983,
      "eval_runtime": 11.3477,
      "eval_samples_per_second": 3.172,
      "eval_steps_per_second": 3.172,
      "step": 51
    },
    {
      "epoch": 3.0588235294117645,
      "grad_norm": 0.4572676420211792,
      "learning_rate": 6.11764705882353e-05,
      "loss": 0.3047,
      "step": 52
    },
    {
      "epoch": 3.176470588235294,
      "grad_norm": 0.677977979183197,
      "learning_rate": 6.352941176470588e-05,
      "loss": 0.4538,
      "step": 54
    },
    {
      "epoch": 3.2941176470588234,
      "grad_norm": 0.6195520758628845,
      "learning_rate": 6.588235294117648e-05,
      "loss": 0.3407,
      "step": 56
    },
    {
      "epoch": 3.411764705882353,
      "grad_norm": 1.0320086479187012,
      "learning_rate": 6.823529411764707e-05,
      "loss": 0.5925,
      "step": 58
    },
    {
      "epoch": 3.5294117647058822,
      "grad_norm": 0.33734291791915894,
      "learning_rate": 7.058823529411765e-05,
      "loss": 0.193,
      "step": 60
    },
    {
      "epoch": 3.6470588235294117,
      "grad_norm": 0.49442484974861145,
      "learning_rate": 7.294117647058823e-05,
      "loss": 0.3146,
      "step": 62
    },
    {
      "epoch": 3.764705882352941,
      "grad_norm": 0.5530934929847717,
      "learning_rate": 7.529411764705883e-05,
      "loss": 0.4063,
      "step": 64
    },
    {
      "epoch": 3.8823529411764706,
      "grad_norm": 0.5266151428222656,
      "learning_rate": 7.764705882352942e-05,
      "loss": 0.252,
      "step": 66
    },
    {
      "epoch": 4.0,
      "grad_norm": 0.4030764698982239,
      "learning_rate": 8e-05,
      "loss": 0.3036,
      "step": 68
    },
    {
      "epoch": 4.0,
      "eval_loss": 0.37464410066604614,
      "eval_runtime": 11.3472,
      "eval_samples_per_second": 3.173,
      "eval_steps_per_second": 3.173,
      "step": 68
    },
    {
      "epoch": 4.117647058823529,
      "grad_norm": 0.7355620861053467,
      "learning_rate": 8.23529411764706e-05,
      "loss": 0.3745,
      "step": 70
    },
    {
      "epoch": 4.235294117647059,
      "grad_norm": 0.6046448349952698,
      "learning_rate": 8.470588235294118e-05,
      "loss": 0.2459,
      "step": 72
    },
    {
      "epoch": 4.352941176470588,
      "grad_norm": 0.6981219053268433,
      "learning_rate": 8.705882352941177e-05,
      "loss": 0.3181,
      "step": 74
    },
    {
      "epoch": 4.470588235294118,
      "grad_norm": 0.7978318333625793,
      "learning_rate": 8.941176470588236e-05,
      "loss": 0.3047,
      "step": 76
    },
    {
      "epoch": 4.588235294117647,
      "grad_norm": 0.44372427463531494,
      "learning_rate": 9.176470588235295e-05,
      "loss": 0.3103,
      "step": 78
    },
    {
      "epoch": 4.705882352941177,
      "grad_norm": 0.6354032754898071,
      "learning_rate": 9.411764705882353e-05,
      "loss": 0.2416,
      "step": 80
    },
    {
      "epoch": 4.823529411764706,
      "grad_norm": 0.6499518156051636,
      "learning_rate": 9.647058823529412e-05,
      "loss": 0.3272,
      "step": 82
    },
    {
      "epoch": 4.9411764705882355,
      "grad_norm": 0.7763482928276062,
      "learning_rate": 9.882352941176471e-05,
      "loss": 0.2007,
      "step": 84
    },
    {
      "epoch": 5.0,
      "eval_loss": 0.3968093991279602,
      "eval_runtime": 11.3409,
      "eval_samples_per_second": 3.174,
      "eval_steps_per_second": 3.174,
      "step": 85
    },
    {
      "epoch": 5.0588235294117645,
      "grad_norm": 2.083740472793579,
      "learning_rate": 9.99995783847866e-05,
      "loss": 0.2216,
      "step": 86
    },
    {
      "epoch": 5.176470588235294,
      "grad_norm": 0.5484854578971863,
      "learning_rate": 9.999620550574153e-05,
      "loss": 0.2563,
      "step": 88
    },
    {
      "epoch": 5.294117647058823,
      "grad_norm": 0.7849002480506897,
      "learning_rate": 9.998945997517956e-05,
      "loss": 0.2076,
      "step": 90
    },
    {
      "epoch": 5.411764705882353,
      "grad_norm": 0.011317969299852848,
      "learning_rate": 9.997934224814173e-05,
      "loss": 0.0623,
      "step": 92
    },
    {
      "epoch": 5.529411764705882,
      "grad_norm": 2.8171374797821045,
      "learning_rate": 9.996585300715116e-05,
      "loss": 0.1977,
      "step": 94
    },
    {
      "epoch": 5.647058823529412,
      "grad_norm": 0.6430180072784424,
      "learning_rate": 9.994899316216708e-05,
      "loss": 0.1825,
      "step": 96
    },
    {
      "epoch": 5.764705882352941,
      "grad_norm": 0.8138560652732849,
      "learning_rate": 9.992876385052345e-05,
      "loss": 0.2163,
      "step": 98
    },
    {
      "epoch": 5.882352941176471,
      "grad_norm": 0.8699930310249329,
      "learning_rate": 9.990516643685222e-05,
      "loss": 0.2828,
      "step": 100
    },
    {
      "epoch": 6.0,
      "grad_norm": 1.1116987466812134,
      "learning_rate": 9.987820251299122e-05,
      "loss": 0.3084,
      "step": 102
    },
    {
      "epoch": 6.0,
      "eval_loss": 0.4083389937877655,
      "eval_runtime": 11.3469,
      "eval_samples_per_second": 3.173,
      "eval_steps_per_second": 3.173,
      "step": 102
    },
    {
      "epoch": 6.117647058823529,
      "grad_norm": 0.6466872692108154,
      "learning_rate": 9.984787389787688e-05,
      "loss": 0.0989,
      "step": 104
    },
    {
      "epoch": 6.235294117647059,
      "grad_norm": 2.04292631149292,
      "learning_rate": 9.981418263742148e-05,
      "loss": 0.2486,
      "step": 106
    },
    {
      "epoch": 6.352941176470588,
      "grad_norm": 0.4618268311023712,
      "learning_rate": 9.977713100437509e-05,
      "loss": 0.1556,
      "step": 108
    },
    {
      "epoch": 6.470588235294118,
      "grad_norm": 0.8857285976409912,
      "learning_rate": 9.973672149817232e-05,
      "loss": 0.1179,
      "step": 110
    },
    {
      "epoch": 6.588235294117647,
      "grad_norm": 0.6350628137588501,
      "learning_rate": 9.96929568447637e-05,
      "loss": 0.1024,
      "step": 112
    },
    {
      "epoch": 6.705882352941177,
      "grad_norm": 0.5955278277397156,
      "learning_rate": 9.964583999643174e-05,
      "loss": 0.0832,
      "step": 114
    },
    {
      "epoch": 6.823529411764706,
      "grad_norm": 1.1298725605010986,
      "learning_rate": 9.95953741315919e-05,
      "loss": 0.1823,
      "step": 116
    },
    {
      "epoch": 6.9411764705882355,
      "grad_norm": 1.0891544818878174,
      "learning_rate": 9.954156265457801e-05,
      "loss": 0.1851,
      "step": 118
    },
    {
      "epoch": 7.0,
      "eval_loss": 0.4679950177669525,
      "eval_runtime": 11.3412,
      "eval_samples_per_second": 3.174,
      "eval_steps_per_second": 3.174,
      "step": 119
    },
    {
      "epoch": 7.0588235294117645,
      "grad_norm": 1.528273105621338,
      "learning_rate": 9.948440919541278e-05,
      "loss": 0.2052,
      "step": 120
    },
    {
      "epoch": 7.176470588235294,
      "grad_norm": 0.7731708288192749,
      "learning_rate": 9.942391760956277e-05,
      "loss": 0.1102,
      "step": 122
    },
    {
      "epoch": 7.294117647058823,
      "grad_norm": 0.9216702580451965,
      "learning_rate": 9.936009197767845e-05,
      "loss": 0.0904,
      "step": 124
    },
    {
      "epoch": 7.411764705882353,
      "grad_norm": 0.9811257123947144,
      "learning_rate": 9.929293660531888e-05,
      "loss": 0.0858,
      "step": 126
    },
    {
      "epoch": 7.529411764705882,
      "grad_norm": 0.7818095684051514,
      "learning_rate": 9.922245602266118e-05,
      "loss": 0.0688,
      "step": 128
    },
    {
      "epoch": 7.647058823529412,
      "grad_norm": 0.6925339698791504,
      "learning_rate": 9.91486549841951e-05,
      "loss": 0.0752,
      "step": 130
    },
    {
      "epoch": 7.764705882352941,
      "grad_norm": 1.2245044708251953,
      "learning_rate": 9.90715384684021e-05,
      "loss": 0.0549,
      "step": 132
    },
    {
      "epoch": 7.882352941176471,
      "grad_norm": 0.8782073259353638,
      "learning_rate": 9.899111167741966e-05,
      "loss": 0.1141,
      "step": 134
    },
    {
      "epoch": 8.0,
      "grad_norm": 0.7497034072875977,
      "learning_rate": 9.890738003669029e-05,
      "loss": 0.0652,
      "step": 136
    },
    {
      "epoch": 8.0,
      "eval_loss": 0.5924408435821533,
      "eval_runtime": 11.3501,
      "eval_samples_per_second": 3.172,
      "eval_steps_per_second": 3.172,
      "step": 136
    },
    {
      "epoch": 8.117647058823529,
      "grad_norm": 0.5516383647918701,
      "learning_rate": 9.882034919459555e-05,
      "loss": 0.0337,
      "step": 138
    },
    {
      "epoch": 8.235294117647058,
      "grad_norm": 0.8810736536979675,
      "learning_rate": 9.873002502207503e-05,
      "loss": 0.0526,
      "step": 140
    },
    {
      "epoch": 8.352941176470589,
      "grad_norm": 0.9770943522453308,
      "learning_rate": 9.863641361223024e-05,
      "loss": 0.0673,
      "step": 142
    },
    {
      "epoch": 8.470588235294118,
      "grad_norm": 0.001721722655929625,
      "learning_rate": 9.853952127991372e-05,
      "loss": 0.0437,
      "step": 144
    },
    {
      "epoch": 8.588235294117647,
      "grad_norm": 0.9866031408309937,
      "learning_rate": 9.843935456130295e-05,
      "loss": 0.0563,
      "step": 146
    },
    {
      "epoch": 8.705882352941176,
      "grad_norm": 0.7083449959754944,
      "learning_rate": 9.833592021345937e-05,
      "loss": 0.0262,
      "step": 148
    },
    {
      "epoch": 8.823529411764707,
      "grad_norm": 0.7849775552749634,
      "learning_rate": 9.822922521387276e-05,
      "loss": 0.0443,
      "step": 150
    },
    {
      "epoch": 8.941176470588236,
      "grad_norm": 2.4482357501983643,
      "learning_rate": 9.811927675999036e-05,
      "loss": 0.0799,
      "step": 152
    },
    {
      "epoch": 9.0,
      "eval_loss": 0.7859164476394653,
      "eval_runtime": 11.3554,
      "eval_samples_per_second": 3.17,
      "eval_steps_per_second": 3.17,
      "step": 153
    },
    {
      "epoch": 9.058823529411764,
      "grad_norm": 1.439387321472168,
      "learning_rate": 9.800608226873142e-05,
      "loss": 0.0444,
      "step": 154
    },
    {
      "epoch": 9.176470588235293,
      "grad_norm": 0.6562706828117371,
      "learning_rate": 9.788964937598689e-05,
      "loss": 0.03,
      "step": 156
    },
    {
      "epoch": 9.294117647058824,
      "grad_norm": 0.5121655464172363,
      "learning_rate": 9.776998593610428e-05,
      "loss": 0.024,
      "step": 158
    },
    {
      "epoch": 9.411764705882353,
      "grad_norm": 1.0863844156265259,
      "learning_rate": 9.764710002135784e-05,
      "loss": 0.0425,
      "step": 160
    },
    {
      "epoch": 9.529411764705882,
      "grad_norm": 0.4708317518234253,
      "learning_rate": 9.752099992140399e-05,
      "loss": 0.0152,
      "step": 162
    },
    {
      "epoch": 9.647058823529411,
      "grad_norm": 0.798844039440155,
      "learning_rate": 9.739169414272217e-05,
      "loss": 0.0288,
      "step": 164
    },
    {
      "epoch": 9.764705882352942,
      "grad_norm": 0.6964065432548523,
      "learning_rate": 9.725919140804099e-05,
      "loss": 0.0318,
      "step": 166
    },
    {
      "epoch": 9.882352941176471,
      "grad_norm": 0.2773822844028473,
      "learning_rate": 9.71235006557497e-05,
      "loss": 0.0198,
      "step": 168
    },
    {
      "epoch": 10.0,
      "grad_norm": 0.9961733222007751,
      "learning_rate": 9.698463103929542e-05,
      "loss": 0.0279,
      "step": 170
    },
    {
      "epoch": 10.0,
      "eval_loss": 0.7901482582092285,
      "eval_runtime": 11.3379,
      "eval_samples_per_second": 3.175,
      "eval_steps_per_second": 3.175,
      "step": 170
    },
    {
      "epoch": 10.117647058823529,
      "grad_norm": 1.3345593214035034,
      "learning_rate": 9.684259192656553e-05,
      "loss": 0.0199,
      "step": 172
    },
    {
      "epoch": 10.235294117647058,
      "grad_norm": 0.46965789794921875,
      "learning_rate": 9.669739289925577e-05,
      "loss": 0.0095,
      "step": 174
    },
    {
      "epoch": 10.352941176470589,
      "grad_norm": 1.2512540817260742,
      "learning_rate": 9.654904375222385e-05,
      "loss": 0.0191,
      "step": 176
    },
    {
      "epoch": 10.470588235294118,
      "grad_norm": 0.9788813591003418,
      "learning_rate": 9.639755449282875e-05,
      "loss": 0.025,
      "step": 178
    },
    {
      "epoch": 10.588235294117647,
      "grad_norm": 1.7676632404327393,
      "learning_rate": 9.62429353402556e-05,
      "loss": 0.0392,
      "step": 180
    },
    {
      "epoch": 10.705882352941176,
      "grad_norm": 0.6700814962387085,
      "learning_rate": 9.608519672482636e-05,
      "loss": 0.0266,
      "step": 182
    },
    {
      "epoch": 10.823529411764707,
      "grad_norm": 0.6903836727142334,
      "learning_rate": 9.592434928729616e-05,
      "loss": 0.0254,
      "step": 184
    },
    {
      "epoch": 10.941176470588236,
      "grad_norm": 0.9346083402633667,
      "learning_rate": 9.576040387813552e-05,
      "loss": 0.0289,
      "step": 186
    },
    {
      "epoch": 11.0,
      "eval_loss": 0.8221306204795837,
      "eval_runtime": 11.3362,
      "eval_samples_per_second": 3.176,
      "eval_steps_per_second": 3.176,
      "step": 187
    },
    {
      "epoch": 11.0,
      "step": 187,
      "total_flos": 4.05670994116608e+16,
      "train_loss": 0.5283502815789319,
      "train_runtime": 1702.1671,
      "train_samples_per_second": 3.995,
      "train_steps_per_second": 0.499
    }
  ],
  "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": 4.05670994116608e+16,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}