| { | |
| "best_metric": 1.014785885810852, | |
| "best_model_checkpoint": "data/Gemma-2-2B_task-1_180-samples_config-1_full/checkpoint-102", | |
| "epoch": 13.0, | |
| "eval_steps": 500, | |
| "global_step": 221, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.058823529411764705, | |
| "grad_norm": 1.3523170948028564, | |
| "learning_rate": 1.1764705882352942e-06, | |
| "loss": 2.5087, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.11764705882352941, | |
| "grad_norm": 1.309058427810669, | |
| "learning_rate": 2.3529411764705885e-06, | |
| "loss": 2.5155, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.23529411764705882, | |
| "grad_norm": 1.3226087093353271, | |
| "learning_rate": 4.705882352941177e-06, | |
| "loss": 2.4301, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.35294117647058826, | |
| "grad_norm": 1.306518316268921, | |
| "learning_rate": 7.058823529411765e-06, | |
| "loss": 2.4884, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.47058823529411764, | |
| "grad_norm": 1.5081236362457275, | |
| "learning_rate": 9.411764705882354e-06, | |
| "loss": 2.4818, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.5882352941176471, | |
| "grad_norm": 1.346767544746399, | |
| "learning_rate": 1.1764705882352942e-05, | |
| "loss": 2.4643, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.7058823529411765, | |
| "grad_norm": 1.4245949983596802, | |
| "learning_rate": 1.411764705882353e-05, | |
| "loss": 2.5415, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.8235294117647058, | |
| "grad_norm": 1.400976300239563, | |
| "learning_rate": 1.647058823529412e-05, | |
| "loss": 2.4819, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.9411764705882353, | |
| "grad_norm": 1.347977638244629, | |
| "learning_rate": 1.8823529411764708e-05, | |
| "loss": 2.367, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_loss": 2.2712149620056152, | |
| "eval_runtime": 3.2211, | |
| "eval_samples_per_second": 11.176, | |
| "eval_steps_per_second": 11.176, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 1.0588235294117647, | |
| "grad_norm": 1.230087399482727, | |
| "learning_rate": 2.1176470588235296e-05, | |
| "loss": 2.2496, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 1.1764705882352942, | |
| "grad_norm": 1.1536972522735596, | |
| "learning_rate": 2.3529411764705884e-05, | |
| "loss": 2.1844, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 1.2941176470588236, | |
| "grad_norm": 1.1565961837768555, | |
| "learning_rate": 2.5882352941176475e-05, | |
| "loss": 2.1498, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 1.4117647058823528, | |
| "grad_norm": 1.0430412292480469, | |
| "learning_rate": 2.823529411764706e-05, | |
| "loss": 1.9951, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 1.5294117647058822, | |
| "grad_norm": 1.0497277975082397, | |
| "learning_rate": 3.058823529411765e-05, | |
| "loss": 1.9594, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 1.6470588235294117, | |
| "grad_norm": 1.070556402206421, | |
| "learning_rate": 3.294117647058824e-05, | |
| "loss": 1.8072, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 1.7647058823529411, | |
| "grad_norm": 0.9399714469909668, | |
| "learning_rate": 3.529411764705883e-05, | |
| "loss": 1.7482, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 1.8823529411764706, | |
| "grad_norm": 0.943767249584198, | |
| "learning_rate": 3.7647058823529415e-05, | |
| "loss": 1.5977, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 0.8990877866744995, | |
| "learning_rate": 4e-05, | |
| "loss": 1.5938, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_loss": 1.5140504837036133, | |
| "eval_runtime": 3.147, | |
| "eval_samples_per_second": 11.44, | |
| "eval_steps_per_second": 11.44, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 2.1176470588235294, | |
| "grad_norm": 0.9269075989723206, | |
| "learning_rate": 4.235294117647059e-05, | |
| "loss": 1.4948, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 2.235294117647059, | |
| "grad_norm": 0.8362197875976562, | |
| "learning_rate": 4.470588235294118e-05, | |
| "loss": 1.418, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 2.3529411764705883, | |
| "grad_norm": 0.8460010290145874, | |
| "learning_rate": 4.705882352941177e-05, | |
| "loss": 1.2589, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 2.4705882352941178, | |
| "grad_norm": 0.726374626159668, | |
| "learning_rate": 4.9411764705882355e-05, | |
| "loss": 1.3223, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 2.588235294117647, | |
| "grad_norm": 0.6924421191215515, | |
| "learning_rate": 5.176470588235295e-05, | |
| "loss": 1.2024, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 2.7058823529411766, | |
| "grad_norm": 0.592928946018219, | |
| "learning_rate": 5.411764705882353e-05, | |
| "loss": 1.2822, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 2.8235294117647056, | |
| "grad_norm": 0.6075505018234253, | |
| "learning_rate": 5.647058823529412e-05, | |
| "loss": 1.1869, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 2.9411764705882355, | |
| "grad_norm": 0.5907753705978394, | |
| "learning_rate": 5.882352941176471e-05, | |
| "loss": 1.1165, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_loss": 1.1451517343521118, | |
| "eval_runtime": 3.1854, | |
| "eval_samples_per_second": 11.301, | |
| "eval_steps_per_second": 11.301, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 3.0588235294117645, | |
| "grad_norm": 0.5991860628128052, | |
| "learning_rate": 6.11764705882353e-05, | |
| "loss": 1.0539, | |
| "step": 52 | |
| }, | |
| { | |
| "epoch": 3.176470588235294, | |
| "grad_norm": 0.5755998492240906, | |
| "learning_rate": 6.352941176470588e-05, | |
| "loss": 1.1236, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 3.2941176470588234, | |
| "grad_norm": 0.573941171169281, | |
| "learning_rate": 6.588235294117648e-05, | |
| "loss": 1.0975, | |
| "step": 56 | |
| }, | |
| { | |
| "epoch": 3.411764705882353, | |
| "grad_norm": 0.5601457953453064, | |
| "learning_rate": 6.823529411764707e-05, | |
| "loss": 1.102, | |
| "step": 58 | |
| }, | |
| { | |
| "epoch": 3.5294117647058822, | |
| "grad_norm": 0.5785244107246399, | |
| "learning_rate": 7.058823529411765e-05, | |
| "loss": 1.0358, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 3.6470588235294117, | |
| "grad_norm": 0.6545214653015137, | |
| "learning_rate": 7.294117647058823e-05, | |
| "loss": 1.0396, | |
| "step": 62 | |
| }, | |
| { | |
| "epoch": 3.764705882352941, | |
| "grad_norm": 0.6020636558532715, | |
| "learning_rate": 7.529411764705883e-05, | |
| "loss": 1.0214, | |
| "step": 64 | |
| }, | |
| { | |
| "epoch": 3.8823529411764706, | |
| "grad_norm": 0.5745831727981567, | |
| "learning_rate": 7.764705882352942e-05, | |
| "loss": 1.0284, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "grad_norm": 0.6231544017791748, | |
| "learning_rate": 8e-05, | |
| "loss": 0.957, | |
| "step": 68 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_loss": 1.0573265552520752, | |
| "eval_runtime": 3.2336, | |
| "eval_samples_per_second": 11.133, | |
| "eval_steps_per_second": 11.133, | |
| "step": 68 | |
| }, | |
| { | |
| "epoch": 4.117647058823529, | |
| "grad_norm": 0.6013826131820679, | |
| "learning_rate": 8.23529411764706e-05, | |
| "loss": 0.9615, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 4.235294117647059, | |
| "grad_norm": 0.56843101978302, | |
| "learning_rate": 8.470588235294118e-05, | |
| "loss": 0.9463, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 4.352941176470588, | |
| "grad_norm": 0.6903876662254333, | |
| "learning_rate": 8.705882352941177e-05, | |
| "loss": 1.0033, | |
| "step": 74 | |
| }, | |
| { | |
| "epoch": 4.470588235294118, | |
| "grad_norm": 0.7048185467720032, | |
| "learning_rate": 8.941176470588236e-05, | |
| "loss": 0.9152, | |
| "step": 76 | |
| }, | |
| { | |
| "epoch": 4.588235294117647, | |
| "grad_norm": 0.710409939289093, | |
| "learning_rate": 9.176470588235295e-05, | |
| "loss": 0.9552, | |
| "step": 78 | |
| }, | |
| { | |
| "epoch": 4.705882352941177, | |
| "grad_norm": 0.6932670474052429, | |
| "learning_rate": 9.411764705882353e-05, | |
| "loss": 0.8956, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 4.823529411764706, | |
| "grad_norm": 0.6886160373687744, | |
| "learning_rate": 9.647058823529412e-05, | |
| "loss": 0.9172, | |
| "step": 82 | |
| }, | |
| { | |
| "epoch": 4.9411764705882355, | |
| "grad_norm": 0.7828485369682312, | |
| "learning_rate": 9.882352941176471e-05, | |
| "loss": 0.9136, | |
| "step": 84 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_loss": 1.023837685585022, | |
| "eval_runtime": 3.139, | |
| "eval_samples_per_second": 11.469, | |
| "eval_steps_per_second": 11.469, | |
| "step": 85 | |
| }, | |
| { | |
| "epoch": 5.0588235294117645, | |
| "grad_norm": 0.7136608958244324, | |
| "learning_rate": 9.99995783847866e-05, | |
| "loss": 0.8661, | |
| "step": 86 | |
| }, | |
| { | |
| "epoch": 5.176470588235294, | |
| "grad_norm": 0.7053413987159729, | |
| "learning_rate": 9.999620550574153e-05, | |
| "loss": 0.8454, | |
| "step": 88 | |
| }, | |
| { | |
| "epoch": 5.294117647058823, | |
| "grad_norm": 0.7243173718452454, | |
| "learning_rate": 9.998945997517956e-05, | |
| "loss": 0.8188, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 5.411764705882353, | |
| "grad_norm": 0.8558421730995178, | |
| "learning_rate": 9.997934224814173e-05, | |
| "loss": 0.8438, | |
| "step": 92 | |
| }, | |
| { | |
| "epoch": 5.529411764705882, | |
| "grad_norm": 0.9421637654304504, | |
| "learning_rate": 9.996585300715116e-05, | |
| "loss": 0.7601, | |
| "step": 94 | |
| }, | |
| { | |
| "epoch": 5.647058823529412, | |
| "grad_norm": 0.9478726387023926, | |
| "learning_rate": 9.994899316216708e-05, | |
| "loss": 0.7368, | |
| "step": 96 | |
| }, | |
| { | |
| "epoch": 5.764705882352941, | |
| "grad_norm": 0.9122819900512695, | |
| "learning_rate": 9.992876385052345e-05, | |
| "loss": 0.8189, | |
| "step": 98 | |
| }, | |
| { | |
| "epoch": 5.882352941176471, | |
| "grad_norm": 0.9140463471412659, | |
| "learning_rate": 9.990516643685222e-05, | |
| "loss": 0.8439, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "grad_norm": 0.8590845465660095, | |
| "learning_rate": 9.987820251299122e-05, | |
| "loss": 0.7999, | |
| "step": 102 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_loss": 1.014785885810852, | |
| "eval_runtime": 3.1532, | |
| "eval_samples_per_second": 11.417, | |
| "eval_steps_per_second": 11.417, | |
| "step": 102 | |
| }, | |
| { | |
| "epoch": 6.117647058823529, | |
| "grad_norm": 0.891880214214325, | |
| "learning_rate": 9.984787389787688e-05, | |
| "loss": 0.6276, | |
| "step": 104 | |
| }, | |
| { | |
| "epoch": 6.235294117647059, | |
| "grad_norm": 0.9322286248207092, | |
| "learning_rate": 9.981418263742148e-05, | |
| "loss": 0.6816, | |
| "step": 106 | |
| }, | |
| { | |
| "epoch": 6.352941176470588, | |
| "grad_norm": 0.9799051880836487, | |
| "learning_rate": 9.977713100437509e-05, | |
| "loss": 0.6547, | |
| "step": 108 | |
| }, | |
| { | |
| "epoch": 6.470588235294118, | |
| "grad_norm": 1.251034140586853, | |
| "learning_rate": 9.973672149817232e-05, | |
| "loss": 0.6866, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 6.588235294117647, | |
| "grad_norm": 1.1815893650054932, | |
| "learning_rate": 9.96929568447637e-05, | |
| "loss": 0.671, | |
| "step": 112 | |
| }, | |
| { | |
| "epoch": 6.705882352941177, | |
| "grad_norm": 1.3927816152572632, | |
| "learning_rate": 9.964583999643174e-05, | |
| "loss": 0.7147, | |
| "step": 114 | |
| }, | |
| { | |
| "epoch": 6.823529411764706, | |
| "grad_norm": 1.1381196975708008, | |
| "learning_rate": 9.95953741315919e-05, | |
| "loss": 0.5951, | |
| "step": 116 | |
| }, | |
| { | |
| "epoch": 6.9411764705882355, | |
| "grad_norm": 1.1752654314041138, | |
| "learning_rate": 9.954156265457801e-05, | |
| "loss": 0.6625, | |
| "step": 118 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_loss": 1.0485187768936157, | |
| "eval_runtime": 3.1242, | |
| "eval_samples_per_second": 11.523, | |
| "eval_steps_per_second": 11.523, | |
| "step": 119 | |
| }, | |
| { | |
| "epoch": 7.0588235294117645, | |
| "grad_norm": 1.03155517578125, | |
| "learning_rate": 9.948440919541278e-05, | |
| "loss": 0.6323, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 7.176470588235294, | |
| "grad_norm": 1.0549365282058716, | |
| "learning_rate": 9.942391760956277e-05, | |
| "loss": 0.5145, | |
| "step": 122 | |
| }, | |
| { | |
| "epoch": 7.294117647058823, | |
| "grad_norm": 1.3340102434158325, | |
| "learning_rate": 9.936009197767845e-05, | |
| "loss": 0.5033, | |
| "step": 124 | |
| }, | |
| { | |
| "epoch": 7.411764705882353, | |
| "grad_norm": 1.6354776620864868, | |
| "learning_rate": 9.929293660531888e-05, | |
| "loss": 0.5158, | |
| "step": 126 | |
| }, | |
| { | |
| "epoch": 7.529411764705882, | |
| "grad_norm": 1.549373745918274, | |
| "learning_rate": 9.922245602266118e-05, | |
| "loss": 0.4814, | |
| "step": 128 | |
| }, | |
| { | |
| "epoch": 7.647058823529412, | |
| "grad_norm": 1.3779453039169312, | |
| "learning_rate": 9.91486549841951e-05, | |
| "loss": 0.4581, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 7.764705882352941, | |
| "grad_norm": 1.5334746837615967, | |
| "learning_rate": 9.90715384684021e-05, | |
| "loss": 0.4759, | |
| "step": 132 | |
| }, | |
| { | |
| "epoch": 7.882352941176471, | |
| "grad_norm": 1.37034273147583, | |
| "learning_rate": 9.899111167741966e-05, | |
| "loss": 0.5054, | |
| "step": 134 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "grad_norm": 1.4103301763534546, | |
| "learning_rate": 9.890738003669029e-05, | |
| "loss": 0.5317, | |
| "step": 136 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_loss": 1.184562087059021, | |
| "eval_runtime": 3.132, | |
| "eval_samples_per_second": 11.494, | |
| "eval_steps_per_second": 11.494, | |
| "step": 136 | |
| }, | |
| { | |
| "epoch": 8.117647058823529, | |
| "grad_norm": 1.4115442037582397, | |
| "learning_rate": 9.882034919459555e-05, | |
| "loss": 0.3821, | |
| "step": 138 | |
| }, | |
| { | |
| "epoch": 8.235294117647058, | |
| "grad_norm": 1.294102668762207, | |
| "learning_rate": 9.873002502207503e-05, | |
| "loss": 0.3555, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 8.352941176470589, | |
| "grad_norm": 1.4196933507919312, | |
| "learning_rate": 9.863641361223024e-05, | |
| "loss": 0.3326, | |
| "step": 142 | |
| }, | |
| { | |
| "epoch": 8.470588235294118, | |
| "grad_norm": 1.8555864095687866, | |
| "learning_rate": 9.853952127991372e-05, | |
| "loss": 0.4028, | |
| "step": 144 | |
| }, | |
| { | |
| "epoch": 8.588235294117647, | |
| "grad_norm": 2.0621461868286133, | |
| "learning_rate": 9.843935456130295e-05, | |
| "loss": 0.3618, | |
| "step": 146 | |
| }, | |
| { | |
| "epoch": 8.705882352941176, | |
| "grad_norm": 1.7417529821395874, | |
| "learning_rate": 9.833592021345937e-05, | |
| "loss": 0.3031, | |
| "step": 148 | |
| }, | |
| { | |
| "epoch": 8.823529411764707, | |
| "grad_norm": 1.5859607458114624, | |
| "learning_rate": 9.822922521387276e-05, | |
| "loss": 0.378, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 8.941176470588236, | |
| "grad_norm": 1.9726958274841309, | |
| "learning_rate": 9.811927675999036e-05, | |
| "loss": 0.3318, | |
| "step": 152 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_loss": 1.402917742729187, | |
| "eval_runtime": 3.2062, | |
| "eval_samples_per_second": 11.228, | |
| "eval_steps_per_second": 11.228, | |
| "step": 153 | |
| }, | |
| { | |
| "epoch": 9.058823529411764, | |
| "grad_norm": 1.4162489175796509, | |
| "learning_rate": 9.800608226873142e-05, | |
| "loss": 0.283, | |
| "step": 154 | |
| }, | |
| { | |
| "epoch": 9.176470588235293, | |
| "grad_norm": 1.3106006383895874, | |
| "learning_rate": 9.788964937598689e-05, | |
| "loss": 0.2585, | |
| "step": 156 | |
| }, | |
| { | |
| "epoch": 9.294117647058824, | |
| "grad_norm": 1.5103355646133423, | |
| "learning_rate": 9.776998593610428e-05, | |
| "loss": 0.2068, | |
| "step": 158 | |
| }, | |
| { | |
| "epoch": 9.411764705882353, | |
| "grad_norm": 2.0582549571990967, | |
| "learning_rate": 9.764710002135784e-05, | |
| "loss": 0.2427, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 9.529411764705882, | |
| "grad_norm": 2.148146152496338, | |
| "learning_rate": 9.752099992140399e-05, | |
| "loss": 0.2307, | |
| "step": 162 | |
| }, | |
| { | |
| "epoch": 9.647058823529411, | |
| "grad_norm": 1.9384764432907104, | |
| "learning_rate": 9.739169414272217e-05, | |
| "loss": 0.2503, | |
| "step": 164 | |
| }, | |
| { | |
| "epoch": 9.764705882352942, | |
| "grad_norm": 1.8676784038543701, | |
| "learning_rate": 9.725919140804099e-05, | |
| "loss": 0.2532, | |
| "step": 166 | |
| }, | |
| { | |
| "epoch": 9.882352941176471, | |
| "grad_norm": 1.8864047527313232, | |
| "learning_rate": 9.71235006557497e-05, | |
| "loss": 0.2398, | |
| "step": 168 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "grad_norm": 1.8081555366516113, | |
| "learning_rate": 9.698463103929542e-05, | |
| "loss": 0.2543, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_loss": 1.6158816814422607, | |
| "eval_runtime": 3.2, | |
| "eval_samples_per_second": 11.25, | |
| "eval_steps_per_second": 11.25, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 10.117647058823529, | |
| "grad_norm": 1.5666933059692383, | |
| "learning_rate": 9.684259192656553e-05, | |
| "loss": 0.1716, | |
| "step": 172 | |
| }, | |
| { | |
| "epoch": 10.235294117647058, | |
| "grad_norm": 1.7216593027114868, | |
| "learning_rate": 9.669739289925577e-05, | |
| "loss": 0.1506, | |
| "step": 174 | |
| }, | |
| { | |
| "epoch": 10.352941176470589, | |
| "grad_norm": 1.6455700397491455, | |
| "learning_rate": 9.654904375222385e-05, | |
| "loss": 0.1577, | |
| "step": 176 | |
| }, | |
| { | |
| "epoch": 10.470588235294118, | |
| "grad_norm": 1.653693675994873, | |
| "learning_rate": 9.639755449282875e-05, | |
| "loss": 0.1595, | |
| "step": 178 | |
| }, | |
| { | |
| "epoch": 10.588235294117647, | |
| "grad_norm": 1.9750311374664307, | |
| "learning_rate": 9.62429353402556e-05, | |
| "loss": 0.1757, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 10.705882352941176, | |
| "grad_norm": 2.0074963569641113, | |
| "learning_rate": 9.608519672482636e-05, | |
| "loss": 0.1985, | |
| "step": 182 | |
| }, | |
| { | |
| "epoch": 10.823529411764707, | |
| "grad_norm": 1.8556835651397705, | |
| "learning_rate": 9.592434928729616e-05, | |
| "loss": 0.1731, | |
| "step": 184 | |
| }, | |
| { | |
| "epoch": 10.941176470588236, | |
| "grad_norm": 2.081277847290039, | |
| "learning_rate": 9.576040387813552e-05, | |
| "loss": 0.183, | |
| "step": 186 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "eval_loss": 1.8190038204193115, | |
| "eval_runtime": 3.1325, | |
| "eval_samples_per_second": 11.493, | |
| "eval_steps_per_second": 11.493, | |
| "step": 187 | |
| }, | |
| { | |
| "epoch": 11.058823529411764, | |
| "grad_norm": 1.533440113067627, | |
| "learning_rate": 9.559337155679842e-05, | |
| "loss": 0.1486, | |
| "step": 188 | |
| }, | |
| { | |
| "epoch": 11.176470588235293, | |
| "grad_norm": 1.4578951597213745, | |
| "learning_rate": 9.542326359097619e-05, | |
| "loss": 0.1188, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 11.294117647058824, | |
| "grad_norm": 1.2918648719787598, | |
| "learning_rate": 9.525009145583745e-05, | |
| "loss": 0.1172, | |
| "step": 192 | |
| }, | |
| { | |
| "epoch": 11.411764705882353, | |
| "grad_norm": 1.8749967813491821, | |
| "learning_rate": 9.507386683325404e-05, | |
| "loss": 0.1304, | |
| "step": 194 | |
| }, | |
| { | |
| "epoch": 11.529411764705882, | |
| "grad_norm": 1.5770972967147827, | |
| "learning_rate": 9.489460161101291e-05, | |
| "loss": 0.1265, | |
| "step": 196 | |
| }, | |
| { | |
| "epoch": 11.647058823529411, | |
| "grad_norm": 1.6993085145950317, | |
| "learning_rate": 9.471230788201429e-05, | |
| "loss": 0.1211, | |
| "step": 198 | |
| }, | |
| { | |
| "epoch": 11.764705882352942, | |
| "grad_norm": 1.740233302116394, | |
| "learning_rate": 9.452699794345581e-05, | |
| "loss": 0.127, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 11.882352941176471, | |
| "grad_norm": 1.9086484909057617, | |
| "learning_rate": 9.43386842960031e-05, | |
| "loss": 0.1279, | |
| "step": 202 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "grad_norm": 1.5930449962615967, | |
| "learning_rate": 9.414737964294636e-05, | |
| "loss": 0.1305, | |
| "step": 204 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_loss": 1.9710564613342285, | |
| "eval_runtime": 3.1289, | |
| "eval_samples_per_second": 11.506, | |
| "eval_steps_per_second": 11.506, | |
| "step": 204 | |
| }, | |
| { | |
| "epoch": 12.117647058823529, | |
| "grad_norm": 1.2723948955535889, | |
| "learning_rate": 9.395309688934351e-05, | |
| "loss": 0.0873, | |
| "step": 206 | |
| }, | |
| { | |
| "epoch": 12.235294117647058, | |
| "grad_norm": 1.2322949171066284, | |
| "learning_rate": 9.375584914114963e-05, | |
| "loss": 0.0923, | |
| "step": 208 | |
| }, | |
| { | |
| "epoch": 12.352941176470589, | |
| "grad_norm": 1.4487413167953491, | |
| "learning_rate": 9.355564970433288e-05, | |
| "loss": 0.1005, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 12.470588235294118, | |
| "grad_norm": 1.4588546752929688, | |
| "learning_rate": 9.335251208397684e-05, | |
| "loss": 0.0924, | |
| "step": 212 | |
| }, | |
| { | |
| "epoch": 12.588235294117647, | |
| "grad_norm": 1.450233817100525, | |
| "learning_rate": 9.314644998336949e-05, | |
| "loss": 0.0979, | |
| "step": 214 | |
| }, | |
| { | |
| "epoch": 12.705882352941176, | |
| "grad_norm": 1.538287878036499, | |
| "learning_rate": 9.293747730307889e-05, | |
| "loss": 0.1086, | |
| "step": 216 | |
| }, | |
| { | |
| "epoch": 12.823529411764707, | |
| "grad_norm": 1.1530896425247192, | |
| "learning_rate": 9.272560814001539e-05, | |
| "loss": 0.0925, | |
| "step": 218 | |
| }, | |
| { | |
| "epoch": 12.941176470588236, | |
| "grad_norm": 1.597603440284729, | |
| "learning_rate": 9.251085678648072e-05, | |
| "loss": 0.1007, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "eval_loss": 2.0467374324798584, | |
| "eval_runtime": 3.1236, | |
| "eval_samples_per_second": 11.525, | |
| "eval_steps_per_second": 11.525, | |
| "step": 221 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "step": 221, | |
| "total_flos": 7344862460379136.0, | |
| "train_loss": 0.8140930119413057, | |
| "train_runtime": 748.2379, | |
| "train_samples_per_second": 9.088, | |
| "train_steps_per_second": 1.136 | |
| } | |
| ], | |
| "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": 7344862460379136.0, | |
| "train_batch_size": 1, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |