| { | |
| "best_metric": 1.0008424520492554, | |
| "best_model_checkpoint": "data/Llama-31-8B_task-2_180-samples_config-2_full_auto/checkpoint-85", | |
| "epoch": 16.941176470588236, | |
| "eval_steps": 500, | |
| "global_step": 144, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.11764705882352941, | |
| "grad_norm": 0.4826774597167969, | |
| "learning_rate": 2.5e-06, | |
| "loss": 1.5282, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.23529411764705882, | |
| "grad_norm": 0.4908909797668457, | |
| "learning_rate": 5e-06, | |
| "loss": 1.5413, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.47058823529411764, | |
| "grad_norm": 0.41724953055381775, | |
| "learning_rate": 1e-05, | |
| "loss": 1.5287, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.7058823529411765, | |
| "grad_norm": 0.40612688660621643, | |
| "learning_rate": 1.5e-05, | |
| "loss": 1.4941, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.9411764705882353, | |
| "grad_norm": 0.40694111585617065, | |
| "learning_rate": 2e-05, | |
| "loss": 1.4945, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.9411764705882353, | |
| "eval_loss": 1.4963806867599487, | |
| "eval_runtime": 38.0148, | |
| "eval_samples_per_second": 0.947, | |
| "eval_steps_per_second": 0.947, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 1.1764705882352942, | |
| "grad_norm": 0.4671599268913269, | |
| "learning_rate": 2.5e-05, | |
| "loss": 1.4818, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 1.4117647058823528, | |
| "grad_norm": 0.34635108709335327, | |
| "learning_rate": 3e-05, | |
| "loss": 1.44, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 1.6470588235294117, | |
| "grad_norm": 0.34106263518333435, | |
| "learning_rate": 3.5e-05, | |
| "loss": 1.4156, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 1.8823529411764706, | |
| "grad_norm": 0.3254939913749695, | |
| "learning_rate": 4e-05, | |
| "loss": 1.3936, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_loss": 1.3699593544006348, | |
| "eval_runtime": 38.0152, | |
| "eval_samples_per_second": 0.947, | |
| "eval_steps_per_second": 0.947, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 2.1176470588235294, | |
| "grad_norm": 0.25313565135002136, | |
| "learning_rate": 4.5e-05, | |
| "loss": 1.3737, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 2.3529411764705883, | |
| "grad_norm": 0.2182977795600891, | |
| "learning_rate": 5e-05, | |
| "loss": 1.3106, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 2.588235294117647, | |
| "grad_norm": 0.23302167654037476, | |
| "learning_rate": 5.500000000000001e-05, | |
| "loss": 1.2835, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 2.8235294117647056, | |
| "grad_norm": 0.24765853583812714, | |
| "learning_rate": 6e-05, | |
| "loss": 1.2937, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 2.9411764705882355, | |
| "eval_loss": 1.2535487413406372, | |
| "eval_runtime": 38.0071, | |
| "eval_samples_per_second": 0.947, | |
| "eval_steps_per_second": 0.947, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 3.0588235294117645, | |
| "grad_norm": 0.2522370219230652, | |
| "learning_rate": 6.500000000000001e-05, | |
| "loss": 1.2284, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 3.2941176470588234, | |
| "grad_norm": 0.20119093358516693, | |
| "learning_rate": 7e-05, | |
| "loss": 1.1927, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 3.5294117647058822, | |
| "grad_norm": 0.2230207473039627, | |
| "learning_rate": 7.500000000000001e-05, | |
| "loss": 1.1363, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 3.764705882352941, | |
| "grad_norm": 0.26864516735076904, | |
| "learning_rate": 8e-05, | |
| "loss": 1.1482, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "grad_norm": 0.23896686732769012, | |
| "learning_rate": 8.5e-05, | |
| "loss": 1.0479, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_loss": 1.1012284755706787, | |
| "eval_runtime": 38.004, | |
| "eval_samples_per_second": 0.947, | |
| "eval_steps_per_second": 0.947, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 4.235294117647059, | |
| "grad_norm": 0.2105560153722763, | |
| "learning_rate": 9e-05, | |
| "loss": 1.0413, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 4.470588235294118, | |
| "grad_norm": 0.1726371794939041, | |
| "learning_rate": 9.5e-05, | |
| "loss": 1.0196, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 4.705882352941177, | |
| "grad_norm": 0.14054733514785767, | |
| "learning_rate": 0.0001, | |
| "loss": 1.0248, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 4.9411764705882355, | |
| "grad_norm": 0.13392707705497742, | |
| "learning_rate": 9.999238475781957e-05, | |
| "loss": 0.9934, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 4.9411764705882355, | |
| "eval_loss": 1.0543981790542603, | |
| "eval_runtime": 37.9991, | |
| "eval_samples_per_second": 0.947, | |
| "eval_steps_per_second": 0.947, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 5.176470588235294, | |
| "grad_norm": 0.1387275904417038, | |
| "learning_rate": 9.99695413509548e-05, | |
| "loss": 0.9566, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 5.411764705882353, | |
| "grad_norm": 0.13243956863880157, | |
| "learning_rate": 9.99314767377287e-05, | |
| "loss": 1.0016, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 5.647058823529412, | |
| "grad_norm": 0.138446643948555, | |
| "learning_rate": 9.987820251299122e-05, | |
| "loss": 0.9691, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 5.882352941176471, | |
| "grad_norm": 0.1291918009519577, | |
| "learning_rate": 9.980973490458728e-05, | |
| "loss": 0.9927, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_loss": 1.0315124988555908, | |
| "eval_runtime": 37.9948, | |
| "eval_samples_per_second": 0.947, | |
| "eval_steps_per_second": 0.947, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 6.117647058823529, | |
| "grad_norm": 0.13240665197372437, | |
| "learning_rate": 9.972609476841367e-05, | |
| "loss": 0.9476, | |
| "step": 52 | |
| }, | |
| { | |
| "epoch": 6.352941176470588, | |
| "grad_norm": 0.12557284533977509, | |
| "learning_rate": 9.962730758206611e-05, | |
| "loss": 0.923, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 6.588235294117647, | |
| "grad_norm": 0.13065005838871002, | |
| "learning_rate": 9.951340343707852e-05, | |
| "loss": 0.952, | |
| "step": 56 | |
| }, | |
| { | |
| "epoch": 6.823529411764706, | |
| "grad_norm": 0.12568555772304535, | |
| "learning_rate": 9.938441702975689e-05, | |
| "loss": 0.9944, | |
| "step": 58 | |
| }, | |
| { | |
| "epoch": 6.9411764705882355, | |
| "eval_loss": 1.017928123474121, | |
| "eval_runtime": 37.9997, | |
| "eval_samples_per_second": 0.947, | |
| "eval_steps_per_second": 0.947, | |
| "step": 59 | |
| }, | |
| { | |
| "epoch": 7.0588235294117645, | |
| "grad_norm": 0.13367407023906708, | |
| "learning_rate": 9.924038765061042e-05, | |
| "loss": 0.9565, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 7.294117647058823, | |
| "grad_norm": 0.13120189309120178, | |
| "learning_rate": 9.908135917238321e-05, | |
| "loss": 0.9356, | |
| "step": 62 | |
| }, | |
| { | |
| "epoch": 7.529411764705882, | |
| "grad_norm": 0.1361704021692276, | |
| "learning_rate": 9.890738003669029e-05, | |
| "loss": 0.908, | |
| "step": 64 | |
| }, | |
| { | |
| "epoch": 7.764705882352941, | |
| "grad_norm": 0.13371413946151733, | |
| "learning_rate": 9.871850323926177e-05, | |
| "loss": 0.9102, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "grad_norm": 0.15247705578804016, | |
| "learning_rate": 9.851478631379982e-05, | |
| "loss": 0.9293, | |
| "step": 68 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_loss": 1.008510708808899, | |
| "eval_runtime": 38.0017, | |
| "eval_samples_per_second": 0.947, | |
| "eval_steps_per_second": 0.947, | |
| "step": 68 | |
| }, | |
| { | |
| "epoch": 8.235294117647058, | |
| "grad_norm": 0.14097601175308228, | |
| "learning_rate": 9.829629131445342e-05, | |
| "loss": 0.8935, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 8.470588235294118, | |
| "grad_norm": 0.16155090928077698, | |
| "learning_rate": 9.806308479691595e-05, | |
| "loss": 0.9291, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 8.705882352941176, | |
| "grad_norm": 0.17741553485393524, | |
| "learning_rate": 9.781523779815179e-05, | |
| "loss": 0.9116, | |
| "step": 74 | |
| }, | |
| { | |
| "epoch": 8.941176470588236, | |
| "grad_norm": 0.1956615149974823, | |
| "learning_rate": 9.755282581475769e-05, | |
| "loss": 0.8712, | |
| "step": 76 | |
| }, | |
| { | |
| "epoch": 8.941176470588236, | |
| "eval_loss": 1.0033499002456665, | |
| "eval_runtime": 37.9932, | |
| "eval_samples_per_second": 0.948, | |
| "eval_steps_per_second": 0.948, | |
| "step": 76 | |
| }, | |
| { | |
| "epoch": 9.176470588235293, | |
| "grad_norm": 0.18102702498435974, | |
| "learning_rate": 9.727592877996585e-05, | |
| "loss": 0.88, | |
| "step": 78 | |
| }, | |
| { | |
| "epoch": 9.411764705882353, | |
| "grad_norm": 0.15716667473316193, | |
| "learning_rate": 9.698463103929542e-05, | |
| "loss": 0.8875, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 9.647058823529411, | |
| "grad_norm": 0.17173871397972107, | |
| "learning_rate": 9.667902132486009e-05, | |
| "loss": 0.8779, | |
| "step": 82 | |
| }, | |
| { | |
| "epoch": 9.882352941176471, | |
| "grad_norm": 0.18177638947963715, | |
| "learning_rate": 9.635919272833938e-05, | |
| "loss": 0.8604, | |
| "step": 84 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_loss": 1.0008424520492554, | |
| "eval_runtime": 37.9975, | |
| "eval_samples_per_second": 0.947, | |
| "eval_steps_per_second": 0.947, | |
| "step": 85 | |
| }, | |
| { | |
| "epoch": 10.117647058823529, | |
| "grad_norm": 0.1904071718454361, | |
| "learning_rate": 9.602524267262203e-05, | |
| "loss": 0.8384, | |
| "step": 86 | |
| }, | |
| { | |
| "epoch": 10.352941176470589, | |
| "grad_norm": 0.19200074672698975, | |
| "learning_rate": 9.567727288213005e-05, | |
| "loss": 0.8651, | |
| "step": 88 | |
| }, | |
| { | |
| "epoch": 10.588235294117647, | |
| "grad_norm": 0.19696439802646637, | |
| "learning_rate": 9.53153893518325e-05, | |
| "loss": 0.8231, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 10.823529411764707, | |
| "grad_norm": 0.23459963500499725, | |
| "learning_rate": 9.493970231495835e-05, | |
| "loss": 0.8278, | |
| "step": 92 | |
| }, | |
| { | |
| "epoch": 10.941176470588236, | |
| "eval_loss": 1.0047024488449097, | |
| "eval_runtime": 37.9999, | |
| "eval_samples_per_second": 0.947, | |
| "eval_steps_per_second": 0.947, | |
| "step": 93 | |
| }, | |
| { | |
| "epoch": 11.058823529411764, | |
| "grad_norm": 0.21655581891536713, | |
| "learning_rate": 9.45503262094184e-05, | |
| "loss": 0.8636, | |
| "step": 94 | |
| }, | |
| { | |
| "epoch": 11.294117647058824, | |
| "grad_norm": 0.2166631668806076, | |
| "learning_rate": 9.414737964294636e-05, | |
| "loss": 0.8597, | |
| "step": 96 | |
| }, | |
| { | |
| "epoch": 11.529411764705882, | |
| "grad_norm": 0.26026663184165955, | |
| "learning_rate": 9.373098535696979e-05, | |
| "loss": 0.7708, | |
| "step": 98 | |
| }, | |
| { | |
| "epoch": 11.764705882352942, | |
| "grad_norm": 0.2587176561355591, | |
| "learning_rate": 9.330127018922194e-05, | |
| "loss": 0.8096, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "grad_norm": 0.24491576850414276, | |
| "learning_rate": 9.285836503510562e-05, | |
| "loss": 0.7609, | |
| "step": 102 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_loss": 1.0211098194122314, | |
| "eval_runtime": 37.9854, | |
| "eval_samples_per_second": 0.948, | |
| "eval_steps_per_second": 0.948, | |
| "step": 102 | |
| }, | |
| { | |
| "epoch": 12.235294117647058, | |
| "grad_norm": 0.31898877024650574, | |
| "learning_rate": 9.24024048078213e-05, | |
| "loss": 0.7488, | |
| "step": 104 | |
| }, | |
| { | |
| "epoch": 12.470588235294118, | |
| "grad_norm": 0.3386828899383545, | |
| "learning_rate": 9.193352839727121e-05, | |
| "loss": 0.7499, | |
| "step": 106 | |
| }, | |
| { | |
| "epoch": 12.705882352941176, | |
| "grad_norm": 0.29767414927482605, | |
| "learning_rate": 9.145187862775209e-05, | |
| "loss": 0.7744, | |
| "step": 108 | |
| }, | |
| { | |
| "epoch": 12.941176470588236, | |
| "grad_norm": 0.3418169617652893, | |
| "learning_rate": 9.09576022144496e-05, | |
| "loss": 0.7599, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 12.941176470588236, | |
| "eval_loss": 1.0358421802520752, | |
| "eval_runtime": 38.002, | |
| "eval_samples_per_second": 0.947, | |
| "eval_steps_per_second": 0.947, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 13.176470588235293, | |
| "grad_norm": 0.3283166289329529, | |
| "learning_rate": 9.045084971874738e-05, | |
| "loss": 0.7414, | |
| "step": 112 | |
| }, | |
| { | |
| "epoch": 13.411764705882353, | |
| "grad_norm": 0.47032415866851807, | |
| "learning_rate": 8.993177550236464e-05, | |
| "loss": 0.6713, | |
| "step": 114 | |
| }, | |
| { | |
| "epoch": 13.647058823529411, | |
| "grad_norm": 0.43123728036880493, | |
| "learning_rate": 8.940053768033609e-05, | |
| "loss": 0.7097, | |
| "step": 116 | |
| }, | |
| { | |
| "epoch": 13.882352941176471, | |
| "grad_norm": 0.3955651819705963, | |
| "learning_rate": 8.885729807284856e-05, | |
| "loss": 0.7215, | |
| "step": 118 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "eval_loss": 1.064591646194458, | |
| "eval_runtime": 37.9998, | |
| "eval_samples_per_second": 0.947, | |
| "eval_steps_per_second": 0.947, | |
| "step": 119 | |
| }, | |
| { | |
| "epoch": 14.117647058823529, | |
| "grad_norm": 0.44894838333129883, | |
| "learning_rate": 8.83022221559489e-05, | |
| "loss": 0.7134, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 14.352941176470589, | |
| "grad_norm": 0.5536584258079529, | |
| "learning_rate": 8.773547901113862e-05, | |
| "loss": 0.5967, | |
| "step": 122 | |
| }, | |
| { | |
| "epoch": 14.588235294117647, | |
| "grad_norm": 0.52504962682724, | |
| "learning_rate": 8.715724127386972e-05, | |
| "loss": 0.6813, | |
| "step": 124 | |
| }, | |
| { | |
| "epoch": 14.823529411764707, | |
| "grad_norm": 0.6435345411300659, | |
| "learning_rate": 8.656768508095853e-05, | |
| "loss": 0.6195, | |
| "step": 126 | |
| }, | |
| { | |
| "epoch": 14.941176470588236, | |
| "eval_loss": 1.086135983467102, | |
| "eval_runtime": 38.0003, | |
| "eval_samples_per_second": 0.947, | |
| "eval_steps_per_second": 0.947, | |
| "step": 127 | |
| }, | |
| { | |
| "epoch": 15.058823529411764, | |
| "grad_norm": 0.5894057750701904, | |
| "learning_rate": 8.596699001693255e-05, | |
| "loss": 0.6674, | |
| "step": 128 | |
| }, | |
| { | |
| "epoch": 15.294117647058824, | |
| "grad_norm": 0.6027148365974426, | |
| "learning_rate": 8.535533905932738e-05, | |
| "loss": 0.5952, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 15.529411764705882, | |
| "grad_norm": 0.6572749614715576, | |
| "learning_rate": 8.473291852294987e-05, | |
| "loss": 0.6192, | |
| "step": 132 | |
| }, | |
| { | |
| "epoch": 15.764705882352942, | |
| "grad_norm": 0.6183807253837585, | |
| "learning_rate": 8.409991800312493e-05, | |
| "loss": 0.6169, | |
| "step": 134 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "grad_norm": 0.5846746563911438, | |
| "learning_rate": 8.345653031794292e-05, | |
| "loss": 0.5401, | |
| "step": 136 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "eval_loss": 1.1416957378387451, | |
| "eval_runtime": 38.0002, | |
| "eval_samples_per_second": 0.947, | |
| "eval_steps_per_second": 0.947, | |
| "step": 136 | |
| }, | |
| { | |
| "epoch": 16.235294117647058, | |
| "grad_norm": 0.5969430208206177, | |
| "learning_rate": 8.280295144952536e-05, | |
| "loss": 0.4974, | |
| "step": 138 | |
| }, | |
| { | |
| "epoch": 16.470588235294116, | |
| "grad_norm": 0.5740926265716553, | |
| "learning_rate": 8.213938048432697e-05, | |
| "loss": 0.5811, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 16.705882352941178, | |
| "grad_norm": 0.6392034888267517, | |
| "learning_rate": 8.146601955249188e-05, | |
| "loss": 0.5123, | |
| "step": 142 | |
| }, | |
| { | |
| "epoch": 16.941176470588236, | |
| "grad_norm": 0.8722455501556396, | |
| "learning_rate": 8.07830737662829e-05, | |
| "loss": 0.5303, | |
| "step": 144 | |
| }, | |
| { | |
| "epoch": 16.941176470588236, | |
| "eval_loss": 1.1947134733200073, | |
| "eval_runtime": 37.9914, | |
| "eval_samples_per_second": 0.948, | |
| "eval_steps_per_second": 0.948, | |
| "step": 144 | |
| }, | |
| { | |
| "epoch": 16.941176470588236, | |
| "step": 144, | |
| "total_flos": 2.2953699906368307e+17, | |
| "train_loss": 0.9383543334487412, | |
| "train_runtime": 8156.4882, | |
| "train_samples_per_second": 0.834, | |
| "train_steps_per_second": 0.049 | |
| } | |
| ], | |
| "logging_steps": 2, | |
| "max_steps": 400, | |
| "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": 2.2953699906368307e+17, | |
| "train_batch_size": 1, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |