| { | |
| "best_metric": 0.28800639510154724, | |
| "best_model_checkpoint": "data/Mistral-7B_task-3_180-samples_config-1_auto/checkpoint-51", | |
| "epoch": 10.0, | |
| "eval_steps": 500, | |
| "global_step": 170, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.058823529411764705, | |
| "grad_norm": 19.76529884338379, | |
| "learning_rate": 1.1764705882352942e-06, | |
| "loss": 3.7705, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.11764705882352941, | |
| "grad_norm": 9.294736862182617, | |
| "learning_rate": 2.3529411764705885e-06, | |
| "loss": 2.3037, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.23529411764705882, | |
| "grad_norm": 18.734840393066406, | |
| "learning_rate": 4.705882352941177e-06, | |
| "loss": 4.0915, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.35294117647058826, | |
| "grad_norm": 9.272418975830078, | |
| "learning_rate": 7.058823529411765e-06, | |
| "loss": 3.4703, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.47058823529411764, | |
| "grad_norm": 18.61702537536621, | |
| "learning_rate": 9.411764705882354e-06, | |
| "loss": 2.8869, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.5882352941176471, | |
| "grad_norm": 15.720806121826172, | |
| "learning_rate": 1.1764705882352942e-05, | |
| "loss": 2.6677, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.7058823529411765, | |
| "grad_norm": 8.163861274719238, | |
| "learning_rate": 1.411764705882353e-05, | |
| "loss": 1.7906, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.8235294117647058, | |
| "grad_norm": 9.789911270141602, | |
| "learning_rate": 1.647058823529412e-05, | |
| "loss": 1.9139, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.9411764705882353, | |
| "grad_norm": 3.6839351654052734, | |
| "learning_rate": 1.8823529411764708e-05, | |
| "loss": 1.384, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_loss": 1.0106091499328613, | |
| "eval_runtime": 39.2328, | |
| "eval_samples_per_second": 0.918, | |
| "eval_steps_per_second": 0.918, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 1.0588235294117647, | |
| "grad_norm": 2.842022657394409, | |
| "learning_rate": 2.1176470588235296e-05, | |
| "loss": 1.1214, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 1.1764705882352942, | |
| "grad_norm": 3.457080364227295, | |
| "learning_rate": 2.3529411764705884e-05, | |
| "loss": 0.8116, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 1.2941176470588236, | |
| "grad_norm": 1.7302788496017456, | |
| "learning_rate": 2.5882352941176475e-05, | |
| "loss": 0.6141, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 1.4117647058823528, | |
| "grad_norm": 1.9123488664627075, | |
| "learning_rate": 2.823529411764706e-05, | |
| "loss": 0.4076, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 1.5294117647058822, | |
| "grad_norm": 1.316670536994934, | |
| "learning_rate": 3.058823529411765e-05, | |
| "loss": 0.3807, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 1.6470588235294117, | |
| "grad_norm": 1.1167984008789062, | |
| "learning_rate": 3.294117647058824e-05, | |
| "loss": 0.355, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 1.7647058823529411, | |
| "grad_norm": 1.6034706830978394, | |
| "learning_rate": 3.529411764705883e-05, | |
| "loss": 0.5264, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 1.8823529411764706, | |
| "grad_norm": 1.00777006149292, | |
| "learning_rate": 3.7647058823529415e-05, | |
| "loss": 0.2688, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 0.7751888036727905, | |
| "learning_rate": 4e-05, | |
| "loss": 0.3571, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_loss": 0.33543625473976135, | |
| "eval_runtime": 39.2377, | |
| "eval_samples_per_second": 0.917, | |
| "eval_steps_per_second": 0.917, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 2.1176470588235294, | |
| "grad_norm": 0.8900798559188843, | |
| "learning_rate": 4.235294117647059e-05, | |
| "loss": 0.3145, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 2.235294117647059, | |
| "grad_norm": 1.0023255348205566, | |
| "learning_rate": 4.470588235294118e-05, | |
| "loss": 0.3117, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 2.3529411764705883, | |
| "grad_norm": 0.9155427813529968, | |
| "learning_rate": 4.705882352941177e-05, | |
| "loss": 0.3202, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 2.4705882352941178, | |
| "grad_norm": 0.7170580625534058, | |
| "learning_rate": 4.9411764705882355e-05, | |
| "loss": 0.2404, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 2.588235294117647, | |
| "grad_norm": 0.7286433577537537, | |
| "learning_rate": 5.176470588235295e-05, | |
| "loss": 0.2366, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 2.7058823529411766, | |
| "grad_norm": 0.6786469221115112, | |
| "learning_rate": 5.411764705882353e-05, | |
| "loss": 0.2502, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 2.8235294117647056, | |
| "grad_norm": 1.2029789686203003, | |
| "learning_rate": 5.647058823529412e-05, | |
| "loss": 0.2911, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 2.9411764705882355, | |
| "grad_norm": 0.571592390537262, | |
| "learning_rate": 5.882352941176471e-05, | |
| "loss": 0.2474, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_loss": 0.28800639510154724, | |
| "eval_runtime": 39.239, | |
| "eval_samples_per_second": 0.917, | |
| "eval_steps_per_second": 0.917, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 3.0588235294117645, | |
| "grad_norm": 0.6203321218490601, | |
| "learning_rate": 6.11764705882353e-05, | |
| "loss": 0.2006, | |
| "step": 52 | |
| }, | |
| { | |
| "epoch": 3.176470588235294, | |
| "grad_norm": 0.9547243118286133, | |
| "learning_rate": 6.352941176470588e-05, | |
| "loss": 0.2718, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 3.2941176470588234, | |
| "grad_norm": 0.8679971694946289, | |
| "learning_rate": 6.588235294117648e-05, | |
| "loss": 0.1981, | |
| "step": 56 | |
| }, | |
| { | |
| "epoch": 3.411764705882353, | |
| "grad_norm": 0.6779512763023376, | |
| "learning_rate": 6.823529411764707e-05, | |
| "loss": 0.2768, | |
| "step": 58 | |
| }, | |
| { | |
| "epoch": 3.5294117647058822, | |
| "grad_norm": 0.617296040058136, | |
| "learning_rate": 7.058823529411765e-05, | |
| "loss": 0.1179, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 3.6470588235294117, | |
| "grad_norm": 0.685839056968689, | |
| "learning_rate": 7.294117647058823e-05, | |
| "loss": 0.1703, | |
| "step": 62 | |
| }, | |
| { | |
| "epoch": 3.764705882352941, | |
| "grad_norm": 0.9808648228645325, | |
| "learning_rate": 7.529411764705883e-05, | |
| "loss": 0.2464, | |
| "step": 64 | |
| }, | |
| { | |
| "epoch": 3.8823529411764706, | |
| "grad_norm": 0.893202543258667, | |
| "learning_rate": 7.764705882352942e-05, | |
| "loss": 0.1521, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "grad_norm": 0.6669807434082031, | |
| "learning_rate": 8e-05, | |
| "loss": 0.1674, | |
| "step": 68 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_loss": 0.2978588342666626, | |
| "eval_runtime": 39.2468, | |
| "eval_samples_per_second": 0.917, | |
| "eval_steps_per_second": 0.917, | |
| "step": 68 | |
| }, | |
| { | |
| "epoch": 4.117647058823529, | |
| "grad_norm": 0.8498746156692505, | |
| "learning_rate": 8.23529411764706e-05, | |
| "loss": 0.1128, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 4.235294117647059, | |
| "grad_norm": 0.8267329335212708, | |
| "learning_rate": 8.470588235294118e-05, | |
| "loss": 0.1137, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 4.352941176470588, | |
| "grad_norm": 1.0202006101608276, | |
| "learning_rate": 8.705882352941177e-05, | |
| "loss": 0.1357, | |
| "step": 74 | |
| }, | |
| { | |
| "epoch": 4.470588235294118, | |
| "grad_norm": 1.0347710847854614, | |
| "learning_rate": 8.941176470588236e-05, | |
| "loss": 0.1214, | |
| "step": 76 | |
| }, | |
| { | |
| "epoch": 4.588235294117647, | |
| "grad_norm": 0.796979546546936, | |
| "learning_rate": 9.176470588235295e-05, | |
| "loss": 0.125, | |
| "step": 78 | |
| }, | |
| { | |
| "epoch": 4.705882352941177, | |
| "grad_norm": 0.9515670537948608, | |
| "learning_rate": 9.411764705882353e-05, | |
| "loss": 0.0996, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 4.823529411764706, | |
| "grad_norm": 1.0569146871566772, | |
| "learning_rate": 9.647058823529412e-05, | |
| "loss": 0.1646, | |
| "step": 82 | |
| }, | |
| { | |
| "epoch": 4.9411764705882355, | |
| "grad_norm": 1.1331194639205933, | |
| "learning_rate": 9.882352941176471e-05, | |
| "loss": 0.1058, | |
| "step": 84 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_loss": 0.3145648241043091, | |
| "eval_runtime": 39.231, | |
| "eval_samples_per_second": 0.918, | |
| "eval_steps_per_second": 0.918, | |
| "step": 85 | |
| }, | |
| { | |
| "epoch": 5.0588235294117645, | |
| "grad_norm": 0.714441180229187, | |
| "learning_rate": 9.99995783847866e-05, | |
| "loss": 0.0874, | |
| "step": 86 | |
| }, | |
| { | |
| "epoch": 5.176470588235294, | |
| "grad_norm": 0.6306532025337219, | |
| "learning_rate": 9.999620550574153e-05, | |
| "loss": 0.0779, | |
| "step": 88 | |
| }, | |
| { | |
| "epoch": 5.294117647058823, | |
| "grad_norm": 1.0911846160888672, | |
| "learning_rate": 9.998945997517956e-05, | |
| "loss": 0.0652, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 5.411764705882353, | |
| "grad_norm": 0.0012038883287459612, | |
| "learning_rate": 9.997934224814173e-05, | |
| "loss": 0.0196, | |
| "step": 92 | |
| }, | |
| { | |
| "epoch": 5.529411764705882, | |
| "grad_norm": 1.3048291206359863, | |
| "learning_rate": 9.996585300715116e-05, | |
| "loss": 0.0627, | |
| "step": 94 | |
| }, | |
| { | |
| "epoch": 5.647058823529412, | |
| "grad_norm": 1.381438970565796, | |
| "learning_rate": 9.994899316216708e-05, | |
| "loss": 0.0433, | |
| "step": 96 | |
| }, | |
| { | |
| "epoch": 5.764705882352941, | |
| "grad_norm": 1.7449076175689697, | |
| "learning_rate": 9.992876385052345e-05, | |
| "loss": 0.0645, | |
| "step": 98 | |
| }, | |
| { | |
| "epoch": 5.882352941176471, | |
| "grad_norm": 14.252650260925293, | |
| "learning_rate": 9.990516643685222e-05, | |
| "loss": 0.1015, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "grad_norm": 1.2759284973144531, | |
| "learning_rate": 9.987820251299122e-05, | |
| "loss": 0.096, | |
| "step": 102 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_loss": 0.3918328583240509, | |
| "eval_runtime": 39.2435, | |
| "eval_samples_per_second": 0.917, | |
| "eval_steps_per_second": 0.917, | |
| "step": 102 | |
| }, | |
| { | |
| "epoch": 6.117647058823529, | |
| "grad_norm": 0.6676831841468811, | |
| "learning_rate": 9.984787389787688e-05, | |
| "loss": 0.0229, | |
| "step": 104 | |
| }, | |
| { | |
| "epoch": 6.235294117647059, | |
| "grad_norm": 1.147278070449829, | |
| "learning_rate": 9.981418263742148e-05, | |
| "loss": 0.0526, | |
| "step": 106 | |
| }, | |
| { | |
| "epoch": 6.352941176470588, | |
| "grad_norm": 0.6319430470466614, | |
| "learning_rate": 9.977713100437509e-05, | |
| "loss": 0.0313, | |
| "step": 108 | |
| }, | |
| { | |
| "epoch": 6.470588235294118, | |
| "grad_norm": 0.5369738340377808, | |
| "learning_rate": 9.973672149817232e-05, | |
| "loss": 0.0286, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 6.588235294117647, | |
| "grad_norm": 0.919731080532074, | |
| "learning_rate": 9.96929568447637e-05, | |
| "loss": 0.0215, | |
| "step": 112 | |
| }, | |
| { | |
| "epoch": 6.705882352941177, | |
| "grad_norm": 0.4803030788898468, | |
| "learning_rate": 9.964583999643174e-05, | |
| "loss": 0.0129, | |
| "step": 114 | |
| }, | |
| { | |
| "epoch": 6.823529411764706, | |
| "grad_norm": 1.0743355751037598, | |
| "learning_rate": 9.95953741315919e-05, | |
| "loss": 0.0379, | |
| "step": 116 | |
| }, | |
| { | |
| "epoch": 6.9411764705882355, | |
| "grad_norm": 1.3497893810272217, | |
| "learning_rate": 9.954156265457801e-05, | |
| "loss": 0.0504, | |
| "step": 118 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_loss": 0.44687020778656006, | |
| "eval_runtime": 39.2371, | |
| "eval_samples_per_second": 0.918, | |
| "eval_steps_per_second": 0.918, | |
| "step": 119 | |
| }, | |
| { | |
| "epoch": 7.0588235294117645, | |
| "grad_norm": 0.960222601890564, | |
| "learning_rate": 9.948440919541278e-05, | |
| "loss": 0.0439, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 7.176470588235294, | |
| "grad_norm": 0.7003989815711975, | |
| "learning_rate": 9.942391760956277e-05, | |
| "loss": 0.0288, | |
| "step": 122 | |
| }, | |
| { | |
| "epoch": 7.294117647058823, | |
| "grad_norm": 0.35942715406417847, | |
| "learning_rate": 9.936009197767845e-05, | |
| "loss": 0.0159, | |
| "step": 124 | |
| }, | |
| { | |
| "epoch": 7.411764705882353, | |
| "grad_norm": 0.6450738906860352, | |
| "learning_rate": 9.929293660531888e-05, | |
| "loss": 0.0192, | |
| "step": 126 | |
| }, | |
| { | |
| "epoch": 7.529411764705882, | |
| "grad_norm": 1.1557971239089966, | |
| "learning_rate": 9.922245602266118e-05, | |
| "loss": 0.0215, | |
| "step": 128 | |
| }, | |
| { | |
| "epoch": 7.647058823529412, | |
| "grad_norm": 0.5639829635620117, | |
| "learning_rate": 9.91486549841951e-05, | |
| "loss": 0.0234, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 7.764705882352941, | |
| "grad_norm": 0.5177389979362488, | |
| "learning_rate": 9.90715384684021e-05, | |
| "loss": 0.0129, | |
| "step": 132 | |
| }, | |
| { | |
| "epoch": 7.882352941176471, | |
| "grad_norm": 0.7869104743003845, | |
| "learning_rate": 9.899111167741966e-05, | |
| "loss": 0.0357, | |
| "step": 134 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "grad_norm": 0.46523430943489075, | |
| "learning_rate": 9.890738003669029e-05, | |
| "loss": 0.0118, | |
| "step": 136 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_loss": 0.42759406566619873, | |
| "eval_runtime": 39.2372, | |
| "eval_samples_per_second": 0.917, | |
| "eval_steps_per_second": 0.917, | |
| "step": 136 | |
| }, | |
| { | |
| "epoch": 8.117647058823529, | |
| "grad_norm": 0.28435665369033813, | |
| "learning_rate": 9.882034919459555e-05, | |
| "loss": 0.0081, | |
| "step": 138 | |
| }, | |
| { | |
| "epoch": 8.235294117647058, | |
| "grad_norm": 0.5241941809654236, | |
| "learning_rate": 9.873002502207503e-05, | |
| "loss": 0.0126, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 8.352941176470589, | |
| "grad_norm": 0.5455679893493652, | |
| "learning_rate": 9.863641361223024e-05, | |
| "loss": 0.0329, | |
| "step": 142 | |
| }, | |
| { | |
| "epoch": 8.470588235294118, | |
| "grad_norm": 0.0007703916635364294, | |
| "learning_rate": 9.853952127991372e-05, | |
| "loss": 0.0184, | |
| "step": 144 | |
| }, | |
| { | |
| "epoch": 8.588235294117647, | |
| "grad_norm": 0.7606242895126343, | |
| "learning_rate": 9.843935456130295e-05, | |
| "loss": 0.0144, | |
| "step": 146 | |
| }, | |
| { | |
| "epoch": 8.705882352941176, | |
| "grad_norm": 0.34659460186958313, | |
| "learning_rate": 9.833592021345937e-05, | |
| "loss": 0.0047, | |
| "step": 148 | |
| }, | |
| { | |
| "epoch": 8.823529411764707, | |
| "grad_norm": 0.39952245354652405, | |
| "learning_rate": 9.822922521387276e-05, | |
| "loss": 0.0134, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 8.941176470588236, | |
| "grad_norm": 1.3652231693267822, | |
| "learning_rate": 9.811927675999036e-05, | |
| "loss": 0.0302, | |
| "step": 152 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_loss": 0.46331945061683655, | |
| "eval_runtime": 39.239, | |
| "eval_samples_per_second": 0.917, | |
| "eval_steps_per_second": 0.917, | |
| "step": 153 | |
| }, | |
| { | |
| "epoch": 9.058823529411764, | |
| "grad_norm": 0.2947436273097992, | |
| "learning_rate": 9.800608226873142e-05, | |
| "loss": 0.0117, | |
| "step": 154 | |
| }, | |
| { | |
| "epoch": 9.176470588235293, | |
| "grad_norm": 1.1239269971847534, | |
| "learning_rate": 9.788964937598689e-05, | |
| "loss": 0.0089, | |
| "step": 156 | |
| }, | |
| { | |
| "epoch": 9.294117647058824, | |
| "grad_norm": 0.2485365867614746, | |
| "learning_rate": 9.776998593610428e-05, | |
| "loss": 0.0088, | |
| "step": 158 | |
| }, | |
| { | |
| "epoch": 9.411764705882353, | |
| "grad_norm": 0.3209003806114197, | |
| "learning_rate": 9.764710002135784e-05, | |
| "loss": 0.0132, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 9.529411764705882, | |
| "grad_norm": 0.6713930368423462, | |
| "learning_rate": 9.752099992140399e-05, | |
| "loss": 0.0079, | |
| "step": 162 | |
| }, | |
| { | |
| "epoch": 9.647058823529411, | |
| "grad_norm": 0.42799514532089233, | |
| "learning_rate": 9.739169414272217e-05, | |
| "loss": 0.0078, | |
| "step": 164 | |
| }, | |
| { | |
| "epoch": 9.764705882352942, | |
| "grad_norm": 0.48949486017227173, | |
| "learning_rate": 9.725919140804099e-05, | |
| "loss": 0.0147, | |
| "step": 166 | |
| }, | |
| { | |
| "epoch": 9.882352941176471, | |
| "grad_norm": 0.40556517243385315, | |
| "learning_rate": 9.71235006557497e-05, | |
| "loss": 0.0101, | |
| "step": 168 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "grad_norm": 0.3796030580997467, | |
| "learning_rate": 9.698463103929542e-05, | |
| "loss": 0.0105, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_loss": 0.44934284687042236, | |
| "eval_runtime": 39.2401, | |
| "eval_samples_per_second": 0.917, | |
| "eval_steps_per_second": 0.917, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "step": 170, | |
| "total_flos": 1.3339188927240602e+17, | |
| "train_loss": 0.38110624801586657, | |
| "train_runtime": 4941.1781, | |
| "train_samples_per_second": 1.376, | |
| "train_steps_per_second": 0.172 | |
| } | |
| ], | |
| "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": 1.3339188927240602e+17, | |
| "train_batch_size": 1, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |