| { | |
| "best_metric": 0.48128727078437805, | |
| "best_model_checkpoint": "data/Llama-31-8B_task-3_180-samples_config-3/checkpoint-340", | |
| "epoch": 27.0, | |
| "eval_steps": 500, | |
| "global_step": 459, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.058823529411764705, | |
| "grad_norm": 2.980715036392212, | |
| "learning_rate": 3.9215686274509804e-08, | |
| "loss": 2.6305, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.11764705882352941, | |
| "grad_norm": 1.5815191268920898, | |
| "learning_rate": 7.843137254901961e-08, | |
| "loss": 2.0227, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.23529411764705882, | |
| "grad_norm": 3.0497844219207764, | |
| "learning_rate": 1.5686274509803921e-07, | |
| "loss": 2.8101, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.35294117647058826, | |
| "grad_norm": 1.5842593908309937, | |
| "learning_rate": 2.3529411764705883e-07, | |
| "loss": 2.5918, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.47058823529411764, | |
| "grad_norm": 2.9371345043182373, | |
| "learning_rate": 3.1372549019607843e-07, | |
| "loss": 2.3864, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.5882352941176471, | |
| "grad_norm": 2.9547200202941895, | |
| "learning_rate": 3.921568627450981e-07, | |
| "loss": 2.6168, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.7058823529411765, | |
| "grad_norm": 2.5302109718322754, | |
| "learning_rate": 4.7058823529411767e-07, | |
| "loss": 2.2036, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.8235294117647058, | |
| "grad_norm": 3.466365337371826, | |
| "learning_rate": 5.490196078431373e-07, | |
| "loss": 2.7139, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.9411764705882353, | |
| "grad_norm": 1.7582124471664429, | |
| "learning_rate": 6.274509803921569e-07, | |
| "loss": 2.3708, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_loss": 2.4981932640075684, | |
| "eval_runtime": 31.8694, | |
| "eval_samples_per_second": 1.13, | |
| "eval_steps_per_second": 1.13, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 1.0588235294117647, | |
| "grad_norm": 1.3897138833999634, | |
| "learning_rate": 7.058823529411766e-07, | |
| "loss": 2.2296, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 1.1764705882352942, | |
| "grad_norm": 2.683563470840454, | |
| "learning_rate": 7.843137254901962e-07, | |
| "loss": 2.5726, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 1.2941176470588236, | |
| "grad_norm": 2.160468816757202, | |
| "learning_rate": 8.627450980392157e-07, | |
| "loss": 2.4421, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 1.4117647058823528, | |
| "grad_norm": 3.092349052429199, | |
| "learning_rate": 9.411764705882353e-07, | |
| "loss": 2.7441, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 1.5294117647058822, | |
| "grad_norm": 3.0546014308929443, | |
| "learning_rate": 1.019607843137255e-06, | |
| "loss": 2.6233, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 1.6470588235294117, | |
| "grad_norm": 2.6419599056243896, | |
| "learning_rate": 1.0980392156862745e-06, | |
| "loss": 2.6529, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 1.7647058823529411, | |
| "grad_norm": 1.2905603647232056, | |
| "learning_rate": 1.1764705882352942e-06, | |
| "loss": 2.1026, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 1.8823529411764706, | |
| "grad_norm": 2.6626884937286377, | |
| "learning_rate": 1.2549019607843137e-06, | |
| "loss": 2.5973, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 3.445452928543091, | |
| "learning_rate": 1.3333333333333334e-06, | |
| "loss": 2.4065, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_loss": 2.439713954925537, | |
| "eval_runtime": 31.87, | |
| "eval_samples_per_second": 1.13, | |
| "eval_steps_per_second": 1.13, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 2.1176470588235294, | |
| "grad_norm": 2.4814882278442383, | |
| "learning_rate": 1.4117647058823531e-06, | |
| "loss": 2.3601, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 2.235294117647059, | |
| "grad_norm": 3.0113446712493896, | |
| "learning_rate": 1.4901960784313726e-06, | |
| "loss": 2.3377, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 2.3529411764705883, | |
| "grad_norm": 2.5856285095214844, | |
| "learning_rate": 1.5686274509803923e-06, | |
| "loss": 2.24, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 2.4705882352941178, | |
| "grad_norm": 3.1034915447235107, | |
| "learning_rate": 1.6470588235294118e-06, | |
| "loss": 2.5231, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 2.588235294117647, | |
| "grad_norm": 3.499271869659424, | |
| "learning_rate": 1.7254901960784315e-06, | |
| "loss": 2.4463, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 2.7058823529411766, | |
| "grad_norm": 2.990689992904663, | |
| "learning_rate": 1.8039215686274512e-06, | |
| "loss": 2.4596, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 2.8235294117647056, | |
| "grad_norm": 2.712009906768799, | |
| "learning_rate": 1.8823529411764707e-06, | |
| "loss": 2.2987, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 2.9411764705882355, | |
| "grad_norm": 3.5523183345794678, | |
| "learning_rate": 1.96078431372549e-06, | |
| "loss": 2.3549, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_loss": 2.31467604637146, | |
| "eval_runtime": 31.8743, | |
| "eval_samples_per_second": 1.129, | |
| "eval_steps_per_second": 1.129, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 3.0588235294117645, | |
| "grad_norm": 3.0498082637786865, | |
| "learning_rate": 2.03921568627451e-06, | |
| "loss": 2.4567, | |
| "step": 52 | |
| }, | |
| { | |
| "epoch": 3.176470588235294, | |
| "grad_norm": 2.239677906036377, | |
| "learning_rate": 2.1176470588235296e-06, | |
| "loss": 2.0235, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 3.2941176470588234, | |
| "grad_norm": 2.1120095252990723, | |
| "learning_rate": 2.196078431372549e-06, | |
| "loss": 2.2331, | |
| "step": 56 | |
| }, | |
| { | |
| "epoch": 3.411764705882353, | |
| "grad_norm": 3.549705743789673, | |
| "learning_rate": 2.274509803921569e-06, | |
| "loss": 2.1111, | |
| "step": 58 | |
| }, | |
| { | |
| "epoch": 3.5294117647058822, | |
| "grad_norm": 4.140646934509277, | |
| "learning_rate": 2.3529411764705885e-06, | |
| "loss": 2.4917, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 3.6470588235294117, | |
| "grad_norm": 3.1374101638793945, | |
| "learning_rate": 2.431372549019608e-06, | |
| "loss": 2.2287, | |
| "step": 62 | |
| }, | |
| { | |
| "epoch": 3.764705882352941, | |
| "grad_norm": 2.198146104812622, | |
| "learning_rate": 2.5098039215686274e-06, | |
| "loss": 1.9659, | |
| "step": 64 | |
| }, | |
| { | |
| "epoch": 3.8823529411764706, | |
| "grad_norm": 2.7118916511535645, | |
| "learning_rate": 2.5882352941176473e-06, | |
| "loss": 2.3314, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "grad_norm": 3.456117868423462, | |
| "learning_rate": 2.666666666666667e-06, | |
| "loss": 2.0578, | |
| "step": 68 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_loss": 2.085047483444214, | |
| "eval_runtime": 31.8792, | |
| "eval_samples_per_second": 1.129, | |
| "eval_steps_per_second": 1.129, | |
| "step": 68 | |
| }, | |
| { | |
| "epoch": 4.117647058823529, | |
| "grad_norm": 2.2668004035949707, | |
| "learning_rate": 2.7450980392156867e-06, | |
| "loss": 2.1099, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 4.235294117647059, | |
| "grad_norm": 2.0983450412750244, | |
| "learning_rate": 2.8235294117647062e-06, | |
| "loss": 2.0241, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 4.352941176470588, | |
| "grad_norm": 1.6723711490631104, | |
| "learning_rate": 2.901960784313726e-06, | |
| "loss": 1.8905, | |
| "step": 74 | |
| }, | |
| { | |
| "epoch": 4.470588235294118, | |
| "grad_norm": 3.0844905376434326, | |
| "learning_rate": 2.980392156862745e-06, | |
| "loss": 1.8637, | |
| "step": 76 | |
| }, | |
| { | |
| "epoch": 4.588235294117647, | |
| "grad_norm": 3.0651304721832275, | |
| "learning_rate": 3.058823529411765e-06, | |
| "loss": 1.8883, | |
| "step": 78 | |
| }, | |
| { | |
| "epoch": 4.705882352941177, | |
| "grad_norm": 2.8657171726226807, | |
| "learning_rate": 3.1372549019607846e-06, | |
| "loss": 1.8309, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 4.823529411764706, | |
| "grad_norm": 2.580749750137329, | |
| "learning_rate": 3.2156862745098045e-06, | |
| "loss": 1.7815, | |
| "step": 82 | |
| }, | |
| { | |
| "epoch": 4.9411764705882355, | |
| "grad_norm": 2.4597244262695312, | |
| "learning_rate": 3.2941176470588236e-06, | |
| "loss": 1.8089, | |
| "step": 84 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_loss": 1.7080037593841553, | |
| "eval_runtime": 31.8717, | |
| "eval_samples_per_second": 1.13, | |
| "eval_steps_per_second": 1.13, | |
| "step": 85 | |
| }, | |
| { | |
| "epoch": 5.0588235294117645, | |
| "grad_norm": 1.954211950302124, | |
| "learning_rate": 3.3725490196078435e-06, | |
| "loss": 1.6625, | |
| "step": 86 | |
| }, | |
| { | |
| "epoch": 5.176470588235294, | |
| "grad_norm": 1.9714858531951904, | |
| "learning_rate": 3.450980392156863e-06, | |
| "loss": 1.5611, | |
| "step": 88 | |
| }, | |
| { | |
| "epoch": 5.294117647058823, | |
| "grad_norm": 1.9127044677734375, | |
| "learning_rate": 3.529411764705883e-06, | |
| "loss": 1.5702, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 5.411764705882353, | |
| "grad_norm": 4.0653977394104, | |
| "learning_rate": 3.6078431372549024e-06, | |
| "loss": 1.5627, | |
| "step": 92 | |
| }, | |
| { | |
| "epoch": 5.529411764705882, | |
| "grad_norm": 1.8669339418411255, | |
| "learning_rate": 3.6862745098039223e-06, | |
| "loss": 1.4645, | |
| "step": 94 | |
| }, | |
| { | |
| "epoch": 5.647058823529412, | |
| "grad_norm": 2.6541359424591064, | |
| "learning_rate": 3.7647058823529414e-06, | |
| "loss": 1.3323, | |
| "step": 96 | |
| }, | |
| { | |
| "epoch": 5.764705882352941, | |
| "grad_norm": 1.7167555093765259, | |
| "learning_rate": 3.843137254901962e-06, | |
| "loss": 1.3833, | |
| "step": 98 | |
| }, | |
| { | |
| "epoch": 5.882352941176471, | |
| "grad_norm": 2.953925132751465, | |
| "learning_rate": 3.92156862745098e-06, | |
| "loss": 1.333, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "grad_norm": 1.2611362934112549, | |
| "learning_rate": 4.000000000000001e-06, | |
| "loss": 1.3018, | |
| "step": 102 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_loss": 1.2346911430358887, | |
| "eval_runtime": 31.8703, | |
| "eval_samples_per_second": 1.13, | |
| "eval_steps_per_second": 1.13, | |
| "step": 102 | |
| }, | |
| { | |
| "epoch": 6.117647058823529, | |
| "grad_norm": 1.9918830394744873, | |
| "learning_rate": 4.07843137254902e-06, | |
| "loss": 1.1196, | |
| "step": 104 | |
| }, | |
| { | |
| "epoch": 6.235294117647059, | |
| "grad_norm": 1.8971362113952637, | |
| "learning_rate": 4.15686274509804e-06, | |
| "loss": 1.1518, | |
| "step": 106 | |
| }, | |
| { | |
| "epoch": 6.352941176470588, | |
| "grad_norm": 2.2451581954956055, | |
| "learning_rate": 4.235294117647059e-06, | |
| "loss": 1.0684, | |
| "step": 108 | |
| }, | |
| { | |
| "epoch": 6.470588235294118, | |
| "grad_norm": 2.1816394329071045, | |
| "learning_rate": 4.313725490196079e-06, | |
| "loss": 0.9634, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 6.588235294117647, | |
| "grad_norm": 1.7623378038406372, | |
| "learning_rate": 4.392156862745098e-06, | |
| "loss": 0.9115, | |
| "step": 112 | |
| }, | |
| { | |
| "epoch": 6.705882352941177, | |
| "grad_norm": 2.0872268676757812, | |
| "learning_rate": 4.4705882352941184e-06, | |
| "loss": 0.7509, | |
| "step": 114 | |
| }, | |
| { | |
| "epoch": 6.823529411764706, | |
| "grad_norm": 1.5808390378952026, | |
| "learning_rate": 4.549019607843138e-06, | |
| "loss": 0.8618, | |
| "step": 116 | |
| }, | |
| { | |
| "epoch": 6.9411764705882355, | |
| "grad_norm": 1.1898610591888428, | |
| "learning_rate": 4.627450980392157e-06, | |
| "loss": 1.0212, | |
| "step": 118 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_loss": 0.8016352653503418, | |
| "eval_runtime": 31.8718, | |
| "eval_samples_per_second": 1.13, | |
| "eval_steps_per_second": 1.13, | |
| "step": 119 | |
| }, | |
| { | |
| "epoch": 7.0588235294117645, | |
| "grad_norm": 1.153451919555664, | |
| "learning_rate": 4.705882352941177e-06, | |
| "loss": 0.9469, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 7.176470588235294, | |
| "grad_norm": 1.1448363065719604, | |
| "learning_rate": 4.784313725490196e-06, | |
| "loss": 0.9065, | |
| "step": 122 | |
| }, | |
| { | |
| "epoch": 7.294117647058823, | |
| "grad_norm": 1.0351287126541138, | |
| "learning_rate": 4.862745098039216e-06, | |
| "loss": 0.6538, | |
| "step": 124 | |
| }, | |
| { | |
| "epoch": 7.411764705882353, | |
| "grad_norm": 0.9266816973686218, | |
| "learning_rate": 4.941176470588236e-06, | |
| "loss": 0.7569, | |
| "step": 126 | |
| }, | |
| { | |
| "epoch": 7.529411764705882, | |
| "grad_norm": 0.869621992111206, | |
| "learning_rate": 5.019607843137255e-06, | |
| "loss": 0.6096, | |
| "step": 128 | |
| }, | |
| { | |
| "epoch": 7.647058823529412, | |
| "grad_norm": 0.8324370980262756, | |
| "learning_rate": 5.098039215686274e-06, | |
| "loss": 0.4993, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 7.764705882352941, | |
| "grad_norm": 0.8301700353622437, | |
| "learning_rate": 5.176470588235295e-06, | |
| "loss": 0.4293, | |
| "step": 132 | |
| }, | |
| { | |
| "epoch": 7.882352941176471, | |
| "grad_norm": 0.3732304275035858, | |
| "learning_rate": 5.254901960784314e-06, | |
| "loss": 0.632, | |
| "step": 134 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "grad_norm": 0.5899373888969421, | |
| "learning_rate": 5.333333333333334e-06, | |
| "loss": 0.4899, | |
| "step": 136 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_loss": 0.6475194096565247, | |
| "eval_runtime": 31.8748, | |
| "eval_samples_per_second": 1.129, | |
| "eval_steps_per_second": 1.129, | |
| "step": 136 | |
| }, | |
| { | |
| "epoch": 8.117647058823529, | |
| "grad_norm": 0.6668117642402649, | |
| "learning_rate": 5.411764705882353e-06, | |
| "loss": 0.468, | |
| "step": 138 | |
| }, | |
| { | |
| "epoch": 8.235294117647058, | |
| "grad_norm": 1.0659523010253906, | |
| "learning_rate": 5.4901960784313735e-06, | |
| "loss": 0.6599, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 8.352941176470589, | |
| "grad_norm": 0.787837028503418, | |
| "learning_rate": 5.568627450980393e-06, | |
| "loss": 0.8816, | |
| "step": 142 | |
| }, | |
| { | |
| "epoch": 8.470588235294118, | |
| "grad_norm": 0.9166098237037659, | |
| "learning_rate": 5.6470588235294125e-06, | |
| "loss": 0.4608, | |
| "step": 144 | |
| }, | |
| { | |
| "epoch": 8.588235294117647, | |
| "grad_norm": 0.8993943929672241, | |
| "learning_rate": 5.725490196078431e-06, | |
| "loss": 0.6013, | |
| "step": 146 | |
| }, | |
| { | |
| "epoch": 8.705882352941176, | |
| "grad_norm": 0.5819871425628662, | |
| "learning_rate": 5.803921568627452e-06, | |
| "loss": 0.2645, | |
| "step": 148 | |
| }, | |
| { | |
| "epoch": 8.823529411764707, | |
| "grad_norm": 0.6202341318130493, | |
| "learning_rate": 5.882352941176471e-06, | |
| "loss": 0.5653, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 8.941176470588236, | |
| "grad_norm": 0.737612247467041, | |
| "learning_rate": 5.96078431372549e-06, | |
| "loss": 0.6106, | |
| "step": 152 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_loss": 0.5890491008758545, | |
| "eval_runtime": 31.8673, | |
| "eval_samples_per_second": 1.13, | |
| "eval_steps_per_second": 1.13, | |
| "step": 153 | |
| }, | |
| { | |
| "epoch": 9.058823529411764, | |
| "grad_norm": 0.8201161623001099, | |
| "learning_rate": 6.03921568627451e-06, | |
| "loss": 0.556, | |
| "step": 154 | |
| }, | |
| { | |
| "epoch": 9.176470588235293, | |
| "grad_norm": 0.5118715167045593, | |
| "learning_rate": 6.11764705882353e-06, | |
| "loss": 0.521, | |
| "step": 156 | |
| }, | |
| { | |
| "epoch": 9.294117647058824, | |
| "grad_norm": 2.472463369369507, | |
| "learning_rate": 6.19607843137255e-06, | |
| "loss": 0.5874, | |
| "step": 158 | |
| }, | |
| { | |
| "epoch": 9.411764705882353, | |
| "grad_norm": 0.6849164366722107, | |
| "learning_rate": 6.274509803921569e-06, | |
| "loss": 0.5052, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 9.529411764705882, | |
| "grad_norm": 0.5705221891403198, | |
| "learning_rate": 6.352941176470589e-06, | |
| "loss": 0.4448, | |
| "step": 162 | |
| }, | |
| { | |
| "epoch": 9.647058823529411, | |
| "grad_norm": 0.9986915588378906, | |
| "learning_rate": 6.431372549019609e-06, | |
| "loss": 0.4622, | |
| "step": 164 | |
| }, | |
| { | |
| "epoch": 9.764705882352942, | |
| "grad_norm": 0.5936263203620911, | |
| "learning_rate": 6.5098039215686285e-06, | |
| "loss": 0.5221, | |
| "step": 166 | |
| }, | |
| { | |
| "epoch": 9.882352941176471, | |
| "grad_norm": 0.4011266231536865, | |
| "learning_rate": 6.588235294117647e-06, | |
| "loss": 0.3669, | |
| "step": 168 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "grad_norm": 1.4033674001693726, | |
| "learning_rate": 6.666666666666667e-06, | |
| "loss": 0.5388, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_loss": 0.5729286670684814, | |
| "eval_runtime": 31.8736, | |
| "eval_samples_per_second": 1.129, | |
| "eval_steps_per_second": 1.129, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 10.117647058823529, | |
| "grad_norm": 0.7268795371055603, | |
| "learning_rate": 6.745098039215687e-06, | |
| "loss": 0.4123, | |
| "step": 172 | |
| }, | |
| { | |
| "epoch": 10.235294117647058, | |
| "grad_norm": 0.534287691116333, | |
| "learning_rate": 6.8235294117647065e-06, | |
| "loss": 0.3577, | |
| "step": 174 | |
| }, | |
| { | |
| "epoch": 10.352941176470589, | |
| "grad_norm": 0.46323052048683167, | |
| "learning_rate": 6.901960784313726e-06, | |
| "loss": 0.3157, | |
| "step": 176 | |
| }, | |
| { | |
| "epoch": 10.470588235294118, | |
| "grad_norm": 0.4801871180534363, | |
| "learning_rate": 6.9803921568627454e-06, | |
| "loss": 0.5767, | |
| "step": 178 | |
| }, | |
| { | |
| "epoch": 10.588235294117647, | |
| "grad_norm": 0.47244688868522644, | |
| "learning_rate": 7.058823529411766e-06, | |
| "loss": 0.5139, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 10.705882352941176, | |
| "grad_norm": 0.41083067655563354, | |
| "learning_rate": 7.137254901960785e-06, | |
| "loss": 0.5551, | |
| "step": 182 | |
| }, | |
| { | |
| "epoch": 10.823529411764707, | |
| "grad_norm": 0.6211682558059692, | |
| "learning_rate": 7.215686274509805e-06, | |
| "loss": 0.5971, | |
| "step": 184 | |
| }, | |
| { | |
| "epoch": 10.941176470588236, | |
| "grad_norm": 0.4872354567050934, | |
| "learning_rate": 7.294117647058823e-06, | |
| "loss": 0.7245, | |
| "step": 186 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "eval_loss": 0.5584969520568848, | |
| "eval_runtime": 31.8629, | |
| "eval_samples_per_second": 1.13, | |
| "eval_steps_per_second": 1.13, | |
| "step": 187 | |
| }, | |
| { | |
| "epoch": 11.058823529411764, | |
| "grad_norm": 0.41652852296829224, | |
| "learning_rate": 7.372549019607845e-06, | |
| "loss": 0.3599, | |
| "step": 188 | |
| }, | |
| { | |
| "epoch": 11.176470588235293, | |
| "grad_norm": 0.5096337795257568, | |
| "learning_rate": 7.450980392156863e-06, | |
| "loss": 0.5491, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 11.294117647058824, | |
| "grad_norm": 0.5131371021270752, | |
| "learning_rate": 7.529411764705883e-06, | |
| "loss": 0.6327, | |
| "step": 192 | |
| }, | |
| { | |
| "epoch": 11.411764705882353, | |
| "grad_norm": 0.33539438247680664, | |
| "learning_rate": 7.607843137254902e-06, | |
| "loss": 0.4431, | |
| "step": 194 | |
| }, | |
| { | |
| "epoch": 11.529411764705882, | |
| "grad_norm": 1.0883010625839233, | |
| "learning_rate": 7.686274509803923e-06, | |
| "loss": 0.5964, | |
| "step": 196 | |
| }, | |
| { | |
| "epoch": 11.647058823529411, | |
| "grad_norm": 0.13403330743312836, | |
| "learning_rate": 7.764705882352941e-06, | |
| "loss": 0.1832, | |
| "step": 198 | |
| }, | |
| { | |
| "epoch": 11.764705882352942, | |
| "grad_norm": 0.38453081250190735, | |
| "learning_rate": 7.84313725490196e-06, | |
| "loss": 0.5463, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 11.882352941176471, | |
| "grad_norm": 0.4325025677680969, | |
| "learning_rate": 7.92156862745098e-06, | |
| "loss": 0.3552, | |
| "step": 202 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "grad_norm": 0.26013830304145813, | |
| "learning_rate": 8.000000000000001e-06, | |
| "loss": 0.3568, | |
| "step": 204 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_loss": 0.5533129572868347, | |
| "eval_runtime": 31.8552, | |
| "eval_samples_per_second": 1.13, | |
| "eval_steps_per_second": 1.13, | |
| "step": 204 | |
| }, | |
| { | |
| "epoch": 12.117647058823529, | |
| "grad_norm": 0.5007168054580688, | |
| "learning_rate": 8.07843137254902e-06, | |
| "loss": 0.3615, | |
| "step": 206 | |
| }, | |
| { | |
| "epoch": 12.235294117647058, | |
| "grad_norm": 0.4056757986545563, | |
| "learning_rate": 8.15686274509804e-06, | |
| "loss": 0.7152, | |
| "step": 208 | |
| }, | |
| { | |
| "epoch": 12.352941176470589, | |
| "grad_norm": 0.2833678424358368, | |
| "learning_rate": 8.23529411764706e-06, | |
| "loss": 0.4289, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 12.470588235294118, | |
| "grad_norm": 0.3110792934894562, | |
| "learning_rate": 8.31372549019608e-06, | |
| "loss": 0.4519, | |
| "step": 212 | |
| }, | |
| { | |
| "epoch": 12.588235294117647, | |
| "grad_norm": 0.3763163387775421, | |
| "learning_rate": 8.392156862745099e-06, | |
| "loss": 0.405, | |
| "step": 214 | |
| }, | |
| { | |
| "epoch": 12.705882352941176, | |
| "grad_norm": 0.24176886677742004, | |
| "learning_rate": 8.470588235294118e-06, | |
| "loss": 0.4032, | |
| "step": 216 | |
| }, | |
| { | |
| "epoch": 12.823529411764707, | |
| "grad_norm": 0.3594723343849182, | |
| "learning_rate": 8.549019607843138e-06, | |
| "loss": 0.3982, | |
| "step": 218 | |
| }, | |
| { | |
| "epoch": 12.941176470588236, | |
| "grad_norm": 0.50929194688797, | |
| "learning_rate": 8.627450980392157e-06, | |
| "loss": 0.4165, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "eval_loss": 0.5352616310119629, | |
| "eval_runtime": 31.8489, | |
| "eval_samples_per_second": 1.13, | |
| "eval_steps_per_second": 1.13, | |
| "step": 221 | |
| }, | |
| { | |
| "epoch": 13.058823529411764, | |
| "grad_norm": 0.32427483797073364, | |
| "learning_rate": 8.705882352941177e-06, | |
| "loss": 0.49, | |
| "step": 222 | |
| }, | |
| { | |
| "epoch": 13.176470588235293, | |
| "grad_norm": 0.24497142434120178, | |
| "learning_rate": 8.784313725490196e-06, | |
| "loss": 0.3279, | |
| "step": 224 | |
| }, | |
| { | |
| "epoch": 13.294117647058824, | |
| "grad_norm": 0.34544771909713745, | |
| "learning_rate": 8.862745098039216e-06, | |
| "loss": 0.4668, | |
| "step": 226 | |
| }, | |
| { | |
| "epoch": 13.411764705882353, | |
| "grad_norm": 0.2216784805059433, | |
| "learning_rate": 8.941176470588237e-06, | |
| "loss": 0.3077, | |
| "step": 228 | |
| }, | |
| { | |
| "epoch": 13.529411764705882, | |
| "grad_norm": 0.2850175201892853, | |
| "learning_rate": 9.019607843137256e-06, | |
| "loss": 0.3767, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 13.647058823529411, | |
| "grad_norm": 0.4109225571155548, | |
| "learning_rate": 9.098039215686276e-06, | |
| "loss": 0.4402, | |
| "step": 232 | |
| }, | |
| { | |
| "epoch": 13.764705882352942, | |
| "grad_norm": 0.24095743894577026, | |
| "learning_rate": 9.176470588235294e-06, | |
| "loss": 0.3823, | |
| "step": 234 | |
| }, | |
| { | |
| "epoch": 13.882352941176471, | |
| "grad_norm": 0.5017974972724915, | |
| "learning_rate": 9.254901960784315e-06, | |
| "loss": 0.5895, | |
| "step": 236 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "grad_norm": 0.45756474137306213, | |
| "learning_rate": 9.333333333333334e-06, | |
| "loss": 0.6226, | |
| "step": 238 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "eval_loss": 0.5420221090316772, | |
| "eval_runtime": 31.849, | |
| "eval_samples_per_second": 1.13, | |
| "eval_steps_per_second": 1.13, | |
| "step": 238 | |
| }, | |
| { | |
| "epoch": 14.117647058823529, | |
| "grad_norm": 0.4417901039123535, | |
| "learning_rate": 9.411764705882354e-06, | |
| "loss": 0.545, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 14.235294117647058, | |
| "grad_norm": 0.3756580054759979, | |
| "learning_rate": 9.490196078431373e-06, | |
| "loss": 0.4739, | |
| "step": 242 | |
| }, | |
| { | |
| "epoch": 14.352941176470589, | |
| "grad_norm": 0.4785820543766022, | |
| "learning_rate": 9.568627450980393e-06, | |
| "loss": 0.4149, | |
| "step": 244 | |
| }, | |
| { | |
| "epoch": 14.470588235294118, | |
| "grad_norm": 0.444455087184906, | |
| "learning_rate": 9.647058823529412e-06, | |
| "loss": 0.508, | |
| "step": 246 | |
| }, | |
| { | |
| "epoch": 14.588235294117647, | |
| "grad_norm": 0.3459950387477875, | |
| "learning_rate": 9.725490196078432e-06, | |
| "loss": 0.3752, | |
| "step": 248 | |
| }, | |
| { | |
| "epoch": 14.705882352941176, | |
| "grad_norm": 0.276597797870636, | |
| "learning_rate": 9.803921568627451e-06, | |
| "loss": 0.3527, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 14.823529411764707, | |
| "grad_norm": 0.31315046548843384, | |
| "learning_rate": 9.882352941176472e-06, | |
| "loss": 0.396, | |
| "step": 252 | |
| }, | |
| { | |
| "epoch": 14.941176470588236, | |
| "grad_norm": 0.009531126357614994, | |
| "learning_rate": 9.960784313725492e-06, | |
| "loss": 0.3284, | |
| "step": 254 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "eval_loss": 0.5025795102119446, | |
| "eval_runtime": 31.8517, | |
| "eval_samples_per_second": 1.13, | |
| "eval_steps_per_second": 1.13, | |
| "step": 255 | |
| }, | |
| { | |
| "epoch": 15.058823529411764, | |
| "grad_norm": 0.5613787174224854, | |
| "learning_rate": 9.999995315380667e-06, | |
| "loss": 0.502, | |
| "step": 256 | |
| }, | |
| { | |
| "epoch": 15.176470588235293, | |
| "grad_norm": 0.467040091753006, | |
| "learning_rate": 9.99995783847866e-06, | |
| "loss": 0.4182, | |
| "step": 258 | |
| }, | |
| { | |
| "epoch": 15.294117647058824, | |
| "grad_norm": 0.41092485189437866, | |
| "learning_rate": 9.999882884955554e-06, | |
| "loss": 0.245, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 15.411764705882353, | |
| "grad_norm": 0.36648833751678467, | |
| "learning_rate": 9.99977045537315e-06, | |
| "loss": 0.3851, | |
| "step": 262 | |
| }, | |
| { | |
| "epoch": 15.529411764705882, | |
| "grad_norm": 0.31992214918136597, | |
| "learning_rate": 9.999620550574155e-06, | |
| "loss": 0.4584, | |
| "step": 264 | |
| }, | |
| { | |
| "epoch": 15.647058823529411, | |
| "grad_norm": 0.34673774242401123, | |
| "learning_rate": 9.999433171682158e-06, | |
| "loss": 0.3905, | |
| "step": 266 | |
| }, | |
| { | |
| "epoch": 15.764705882352942, | |
| "grad_norm": 0.46017172932624817, | |
| "learning_rate": 9.999208320101643e-06, | |
| "loss": 0.4964, | |
| "step": 268 | |
| }, | |
| { | |
| "epoch": 15.882352941176471, | |
| "grad_norm": 0.5312222838401794, | |
| "learning_rate": 9.998945997517957e-06, | |
| "loss": 0.3088, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "grad_norm": 0.5483973026275635, | |
| "learning_rate": 9.99864620589731e-06, | |
| "loss": 0.4813, | |
| "step": 272 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "eval_loss": 0.5214402079582214, | |
| "eval_runtime": 31.8457, | |
| "eval_samples_per_second": 1.13, | |
| "eval_steps_per_second": 1.13, | |
| "step": 272 | |
| }, | |
| { | |
| "epoch": 16.11764705882353, | |
| "grad_norm": 0.40592437982559204, | |
| "learning_rate": 9.998308947486753e-06, | |
| "loss": 0.274, | |
| "step": 274 | |
| }, | |
| { | |
| "epoch": 16.235294117647058, | |
| "grad_norm": 0.4894661009311676, | |
| "learning_rate": 9.997934224814173e-06, | |
| "loss": 0.361, | |
| "step": 276 | |
| }, | |
| { | |
| "epoch": 16.352941176470587, | |
| "grad_norm": 0.5340806841850281, | |
| "learning_rate": 9.997522040688258e-06, | |
| "loss": 0.4403, | |
| "step": 278 | |
| }, | |
| { | |
| "epoch": 16.470588235294116, | |
| "grad_norm": 0.6398447751998901, | |
| "learning_rate": 9.997072398198492e-06, | |
| "loss": 0.6416, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 16.58823529411765, | |
| "grad_norm": 0.4408299922943115, | |
| "learning_rate": 9.996585300715117e-06, | |
| "loss": 0.3814, | |
| "step": 282 | |
| }, | |
| { | |
| "epoch": 16.705882352941178, | |
| "grad_norm": 0.45829370617866516, | |
| "learning_rate": 9.996060751889114e-06, | |
| "loss": 0.4494, | |
| "step": 284 | |
| }, | |
| { | |
| "epoch": 16.823529411764707, | |
| "grad_norm": 0.5563424229621887, | |
| "learning_rate": 9.995498755652186e-06, | |
| "loss": 0.297, | |
| "step": 286 | |
| }, | |
| { | |
| "epoch": 16.941176470588236, | |
| "grad_norm": 0.423210084438324, | |
| "learning_rate": 9.994899316216709e-06, | |
| "loss": 0.3015, | |
| "step": 288 | |
| }, | |
| { | |
| "epoch": 17.0, | |
| "eval_loss": 0.511587381362915, | |
| "eval_runtime": 31.8593, | |
| "eval_samples_per_second": 1.13, | |
| "eval_steps_per_second": 1.13, | |
| "step": 289 | |
| }, | |
| { | |
| "epoch": 17.058823529411764, | |
| "grad_norm": 0.46335744857788086, | |
| "learning_rate": 9.994262438075713e-06, | |
| "loss": 0.4301, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 17.176470588235293, | |
| "grad_norm": 0.5282275676727295, | |
| "learning_rate": 9.993588126002848e-06, | |
| "loss": 0.3083, | |
| "step": 292 | |
| }, | |
| { | |
| "epoch": 17.294117647058822, | |
| "grad_norm": 0.6960487961769104, | |
| "learning_rate": 9.992876385052346e-06, | |
| "loss": 0.5187, | |
| "step": 294 | |
| }, | |
| { | |
| "epoch": 17.41176470588235, | |
| "grad_norm": 0.47753918170928955, | |
| "learning_rate": 9.992127220558976e-06, | |
| "loss": 0.2761, | |
| "step": 296 | |
| }, | |
| { | |
| "epoch": 17.529411764705884, | |
| "grad_norm": 0.49851253628730774, | |
| "learning_rate": 9.991340638138022e-06, | |
| "loss": 0.2777, | |
| "step": 298 | |
| }, | |
| { | |
| "epoch": 17.647058823529413, | |
| "grad_norm": 0.5102773904800415, | |
| "learning_rate": 9.990516643685222e-06, | |
| "loss": 0.4131, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 17.764705882352942, | |
| "grad_norm": 0.47633931040763855, | |
| "learning_rate": 9.98965524337673e-06, | |
| "loss": 0.5281, | |
| "step": 302 | |
| }, | |
| { | |
| "epoch": 17.88235294117647, | |
| "grad_norm": 0.33498552441596985, | |
| "learning_rate": 9.988756443669081e-06, | |
| "loss": 0.2696, | |
| "step": 304 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "grad_norm": 0.5153418183326721, | |
| "learning_rate": 9.987820251299121e-06, | |
| "loss": 0.3513, | |
| "step": 306 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "eval_loss": 0.5070950388908386, | |
| "eval_runtime": 31.8477, | |
| "eval_samples_per_second": 1.13, | |
| "eval_steps_per_second": 1.13, | |
| "step": 306 | |
| }, | |
| { | |
| "epoch": 18.11764705882353, | |
| "grad_norm": 0.3366803824901581, | |
| "learning_rate": 9.98684667328398e-06, | |
| "loss": 0.1873, | |
| "step": 308 | |
| }, | |
| { | |
| "epoch": 18.235294117647058, | |
| "grad_norm": 0.6154054403305054, | |
| "learning_rate": 9.985835716921e-06, | |
| "loss": 0.391, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 18.352941176470587, | |
| "grad_norm": 0.6151386499404907, | |
| "learning_rate": 9.984787389787689e-06, | |
| "loss": 0.5524, | |
| "step": 312 | |
| }, | |
| { | |
| "epoch": 18.470588235294116, | |
| "grad_norm": 0.34717944264411926, | |
| "learning_rate": 9.983701699741668e-06, | |
| "loss": 0.214, | |
| "step": 314 | |
| }, | |
| { | |
| "epoch": 18.58823529411765, | |
| "grad_norm": 0.5958353877067566, | |
| "learning_rate": 9.982578654920601e-06, | |
| "loss": 0.4503, | |
| "step": 316 | |
| }, | |
| { | |
| "epoch": 18.705882352941178, | |
| "grad_norm": 0.45278429985046387, | |
| "learning_rate": 9.981418263742148e-06, | |
| "loss": 0.2803, | |
| "step": 318 | |
| }, | |
| { | |
| "epoch": 18.823529411764707, | |
| "grad_norm": 1.5949257612228394, | |
| "learning_rate": 9.980220534903889e-06, | |
| "loss": 0.3574, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 18.941176470588236, | |
| "grad_norm": 0.6842171549797058, | |
| "learning_rate": 9.978985477383264e-06, | |
| "loss": 0.3638, | |
| "step": 322 | |
| }, | |
| { | |
| "epoch": 19.0, | |
| "eval_loss": 0.5485844016075134, | |
| "eval_runtime": 31.8529, | |
| "eval_samples_per_second": 1.13, | |
| "eval_steps_per_second": 1.13, | |
| "step": 323 | |
| }, | |
| { | |
| "epoch": 19.058823529411764, | |
| "grad_norm": 0.5384336113929749, | |
| "learning_rate": 9.97771310043751e-06, | |
| "loss": 0.4372, | |
| "step": 324 | |
| }, | |
| { | |
| "epoch": 19.176470588235293, | |
| "grad_norm": 18.550317764282227, | |
| "learning_rate": 9.97640341360358e-06, | |
| "loss": 0.3407, | |
| "step": 326 | |
| }, | |
| { | |
| "epoch": 19.294117647058822, | |
| "grad_norm": 2.0581817626953125, | |
| "learning_rate": 9.975056426698094e-06, | |
| "loss": 0.2623, | |
| "step": 328 | |
| }, | |
| { | |
| "epoch": 19.41176470588235, | |
| "grad_norm": 0.6005405187606812, | |
| "learning_rate": 9.973672149817232e-06, | |
| "loss": 0.258, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 19.529411764705884, | |
| "grad_norm": 0.8219459056854248, | |
| "learning_rate": 9.972250593336689e-06, | |
| "loss": 0.3986, | |
| "step": 332 | |
| }, | |
| { | |
| "epoch": 19.647058823529413, | |
| "grad_norm": 1.8439754247665405, | |
| "learning_rate": 9.970791767911581e-06, | |
| "loss": 0.2887, | |
| "step": 334 | |
| }, | |
| { | |
| "epoch": 19.764705882352942, | |
| "grad_norm": 2.892220973968506, | |
| "learning_rate": 9.96929568447637e-06, | |
| "loss": 0.3433, | |
| "step": 336 | |
| }, | |
| { | |
| "epoch": 19.88235294117647, | |
| "grad_norm": 0.6804599761962891, | |
| "learning_rate": 9.967762354244778e-06, | |
| "loss": 0.3686, | |
| "step": 338 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "grad_norm": 0.8408982753753662, | |
| "learning_rate": 9.966191788709716e-06, | |
| "loss": 0.5246, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "eval_loss": 0.48128727078437805, | |
| "eval_runtime": 31.8515, | |
| "eval_samples_per_second": 1.13, | |
| "eval_steps_per_second": 1.13, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 20.11764705882353, | |
| "grad_norm": 0.5616675019264221, | |
| "learning_rate": 9.964583999643174e-06, | |
| "loss": 0.2985, | |
| "step": 342 | |
| }, | |
| { | |
| "epoch": 20.235294117647058, | |
| "grad_norm": 0.4580283463001251, | |
| "learning_rate": 9.962938999096159e-06, | |
| "loss": 0.2537, | |
| "step": 344 | |
| }, | |
| { | |
| "epoch": 20.352941176470587, | |
| "grad_norm": 0.857369601726532, | |
| "learning_rate": 9.961256799398584e-06, | |
| "loss": 0.2804, | |
| "step": 346 | |
| }, | |
| { | |
| "epoch": 20.470588235294116, | |
| "grad_norm": 1.1619380712509155, | |
| "learning_rate": 9.95953741315919e-06, | |
| "loss": 0.519, | |
| "step": 348 | |
| }, | |
| { | |
| "epoch": 20.58823529411765, | |
| "grad_norm": 0.5293999314308167, | |
| "learning_rate": 9.957780853265441e-06, | |
| "loss": 0.2812, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 20.705882352941178, | |
| "grad_norm": 0.6322659850120544, | |
| "learning_rate": 9.955987132883435e-06, | |
| "loss": 0.2476, | |
| "step": 352 | |
| }, | |
| { | |
| "epoch": 20.823529411764707, | |
| "grad_norm": 0.6491472125053406, | |
| "learning_rate": 9.954156265457801e-06, | |
| "loss": 0.2483, | |
| "step": 354 | |
| }, | |
| { | |
| "epoch": 20.941176470588236, | |
| "grad_norm": 0.869428813457489, | |
| "learning_rate": 9.952288264711601e-06, | |
| "loss": 0.4751, | |
| "step": 356 | |
| }, | |
| { | |
| "epoch": 21.0, | |
| "eval_loss": 0.536905825138092, | |
| "eval_runtime": 31.8461, | |
| "eval_samples_per_second": 1.13, | |
| "eval_steps_per_second": 1.13, | |
| "step": 357 | |
| }, | |
| { | |
| "epoch": 21.058823529411764, | |
| "grad_norm": 0.7692728042602539, | |
| "learning_rate": 9.950383144646221e-06, | |
| "loss": 0.4331, | |
| "step": 358 | |
| }, | |
| { | |
| "epoch": 21.176470588235293, | |
| "grad_norm": 0.6134036779403687, | |
| "learning_rate": 9.948440919541277e-06, | |
| "loss": 0.3302, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 21.294117647058822, | |
| "grad_norm": 0.8637880682945251, | |
| "learning_rate": 9.946461603954499e-06, | |
| "loss": 0.5403, | |
| "step": 362 | |
| }, | |
| { | |
| "epoch": 21.41176470588235, | |
| "grad_norm": 0.7665501832962036, | |
| "learning_rate": 9.944445212721619e-06, | |
| "loss": 0.2682, | |
| "step": 364 | |
| }, | |
| { | |
| "epoch": 21.529411764705884, | |
| "grad_norm": 0.6994332671165466, | |
| "learning_rate": 9.942391760956277e-06, | |
| "loss": 0.4364, | |
| "step": 366 | |
| }, | |
| { | |
| "epoch": 21.647058823529413, | |
| "grad_norm": 0.8093287944793701, | |
| "learning_rate": 9.940301264049885e-06, | |
| "loss": 0.2887, | |
| "step": 368 | |
| }, | |
| { | |
| "epoch": 21.764705882352942, | |
| "grad_norm": 0.7530141472816467, | |
| "learning_rate": 9.938173737671531e-06, | |
| "loss": 0.2479, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 21.88235294117647, | |
| "grad_norm": 0.35253098607063293, | |
| "learning_rate": 9.936009197767847e-06, | |
| "loss": 0.1859, | |
| "step": 372 | |
| }, | |
| { | |
| "epoch": 22.0, | |
| "grad_norm": 0.8288949728012085, | |
| "learning_rate": 9.933807660562898e-06, | |
| "loss": 0.2074, | |
| "step": 374 | |
| }, | |
| { | |
| "epoch": 22.0, | |
| "eval_loss": 0.5176519751548767, | |
| "eval_runtime": 31.8467, | |
| "eval_samples_per_second": 1.13, | |
| "eval_steps_per_second": 1.13, | |
| "step": 374 | |
| }, | |
| { | |
| "epoch": 22.11764705882353, | |
| "grad_norm": 0.7651407122612, | |
| "learning_rate": 9.931569142558057e-06, | |
| "loss": 0.3298, | |
| "step": 376 | |
| }, | |
| { | |
| "epoch": 22.235294117647058, | |
| "grad_norm": 0.7411403059959412, | |
| "learning_rate": 9.929293660531889e-06, | |
| "loss": 0.3866, | |
| "step": 378 | |
| }, | |
| { | |
| "epoch": 22.352941176470587, | |
| "grad_norm": 0.8898152709007263, | |
| "learning_rate": 9.926981231540007e-06, | |
| "loss": 0.3071, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 22.470588235294116, | |
| "grad_norm": 1.235049843788147, | |
| "learning_rate": 9.924631872914967e-06, | |
| "loss": 0.3644, | |
| "step": 382 | |
| }, | |
| { | |
| "epoch": 22.58823529411765, | |
| "grad_norm": 0.9050986766815186, | |
| "learning_rate": 9.922245602266119e-06, | |
| "loss": 0.2616, | |
| "step": 384 | |
| }, | |
| { | |
| "epoch": 22.705882352941178, | |
| "grad_norm": 0.986419141292572, | |
| "learning_rate": 9.919822437479488e-06, | |
| "loss": 0.2003, | |
| "step": 386 | |
| }, | |
| { | |
| "epoch": 22.823529411764707, | |
| "grad_norm": 1.1197755336761475, | |
| "learning_rate": 9.91736239671763e-06, | |
| "loss": 0.3103, | |
| "step": 388 | |
| }, | |
| { | |
| "epoch": 22.941176470588236, | |
| "grad_norm": 1.1346862316131592, | |
| "learning_rate": 9.91486549841951e-06, | |
| "loss": 0.2513, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 23.0, | |
| "eval_loss": 0.5108680129051208, | |
| "eval_runtime": 31.8499, | |
| "eval_samples_per_second": 1.13, | |
| "eval_steps_per_second": 1.13, | |
| "step": 391 | |
| }, | |
| { | |
| "epoch": 23.058823529411764, | |
| "grad_norm": 0.3815372884273529, | |
| "learning_rate": 9.912331761300341e-06, | |
| "loss": 0.1238, | |
| "step": 392 | |
| }, | |
| { | |
| "epoch": 23.176470588235293, | |
| "grad_norm": 0.004847167991101742, | |
| "learning_rate": 9.909761204351469e-06, | |
| "loss": 0.0666, | |
| "step": 394 | |
| }, | |
| { | |
| "epoch": 23.294117647058822, | |
| "grad_norm": 0.9460958242416382, | |
| "learning_rate": 9.90715384684021e-06, | |
| "loss": 0.4942, | |
| "step": 396 | |
| }, | |
| { | |
| "epoch": 23.41176470588235, | |
| "grad_norm": 1.0769481658935547, | |
| "learning_rate": 9.904509708309723e-06, | |
| "loss": 0.2983, | |
| "step": 398 | |
| }, | |
| { | |
| "epoch": 23.529411764705884, | |
| "grad_norm": 0.9719020128250122, | |
| "learning_rate": 9.901828808578846e-06, | |
| "loss": 0.2081, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 23.647058823529413, | |
| "grad_norm": 1.7906395196914673, | |
| "learning_rate": 9.899111167741966e-06, | |
| "loss": 0.3697, | |
| "step": 402 | |
| }, | |
| { | |
| "epoch": 23.764705882352942, | |
| "grad_norm": 1.5768994092941284, | |
| "learning_rate": 9.896356806168851e-06, | |
| "loss": 0.3432, | |
| "step": 404 | |
| }, | |
| { | |
| "epoch": 23.88235294117647, | |
| "grad_norm": 0.9857346415519714, | |
| "learning_rate": 9.89356574450451e-06, | |
| "loss": 0.203, | |
| "step": 406 | |
| }, | |
| { | |
| "epoch": 24.0, | |
| "grad_norm": 1.2476699352264404, | |
| "learning_rate": 9.890738003669029e-06, | |
| "loss": 0.3019, | |
| "step": 408 | |
| }, | |
| { | |
| "epoch": 24.0, | |
| "eval_loss": 0.5099675059318542, | |
| "eval_runtime": 31.8826, | |
| "eval_samples_per_second": 1.129, | |
| "eval_steps_per_second": 1.129, | |
| "step": 408 | |
| }, | |
| { | |
| "epoch": 24.11764705882353, | |
| "grad_norm": 1.1214749813079834, | |
| "learning_rate": 9.887873604857424e-06, | |
| "loss": 0.3208, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 24.235294117647058, | |
| "grad_norm": 1.2565882205963135, | |
| "learning_rate": 9.884972569539471e-06, | |
| "loss": 0.2897, | |
| "step": 412 | |
| }, | |
| { | |
| "epoch": 24.352941176470587, | |
| "grad_norm": 0.9853368401527405, | |
| "learning_rate": 9.882034919459556e-06, | |
| "loss": 0.2316, | |
| "step": 414 | |
| }, | |
| { | |
| "epoch": 24.470588235294116, | |
| "grad_norm": 1.3978081941604614, | |
| "learning_rate": 9.879060676636502e-06, | |
| "loss": 0.3282, | |
| "step": 416 | |
| }, | |
| { | |
| "epoch": 24.58823529411765, | |
| "grad_norm": 1.6501458883285522, | |
| "learning_rate": 9.876049863363415e-06, | |
| "loss": 0.2489, | |
| "step": 418 | |
| }, | |
| { | |
| "epoch": 24.705882352941178, | |
| "grad_norm": 1.0865875482559204, | |
| "learning_rate": 9.873002502207502e-06, | |
| "loss": 0.1472, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 24.823529411764707, | |
| "grad_norm": 1.1791576147079468, | |
| "learning_rate": 9.86991861600992e-06, | |
| "loss": 0.1987, | |
| "step": 422 | |
| }, | |
| { | |
| "epoch": 24.941176470588236, | |
| "grad_norm": 1.4849443435668945, | |
| "learning_rate": 9.866798227885588e-06, | |
| "loss": 0.2039, | |
| "step": 424 | |
| }, | |
| { | |
| "epoch": 25.0, | |
| "eval_loss": 0.542856752872467, | |
| "eval_runtime": 31.8504, | |
| "eval_samples_per_second": 1.13, | |
| "eval_steps_per_second": 1.13, | |
| "step": 425 | |
| }, | |
| { | |
| "epoch": 25.058823529411764, | |
| "grad_norm": 1.064898133277893, | |
| "learning_rate": 9.863641361223025e-06, | |
| "loss": 0.2573, | |
| "step": 426 | |
| }, | |
| { | |
| "epoch": 25.176470588235293, | |
| "grad_norm": 1.3044614791870117, | |
| "learning_rate": 9.860448039684169e-06, | |
| "loss": 0.2652, | |
| "step": 428 | |
| }, | |
| { | |
| "epoch": 25.294117647058822, | |
| "grad_norm": 1.1536026000976562, | |
| "learning_rate": 9.857218287204204e-06, | |
| "loss": 0.1707, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 25.41176470588235, | |
| "grad_norm": 1.575131893157959, | |
| "learning_rate": 9.853952127991374e-06, | |
| "loss": 0.2226, | |
| "step": 432 | |
| }, | |
| { | |
| "epoch": 25.529411764705884, | |
| "grad_norm": 1.8576124906539917, | |
| "learning_rate": 9.850649586526808e-06, | |
| "loss": 0.1639, | |
| "step": 434 | |
| }, | |
| { | |
| "epoch": 25.647058823529413, | |
| "grad_norm": 1.8986772298812866, | |
| "learning_rate": 9.847310687564335e-06, | |
| "loss": 0.367, | |
| "step": 436 | |
| }, | |
| { | |
| "epoch": 25.764705882352942, | |
| "grad_norm": 1.0811642408370972, | |
| "learning_rate": 9.843935456130295e-06, | |
| "loss": 0.2656, | |
| "step": 438 | |
| }, | |
| { | |
| "epoch": 25.88235294117647, | |
| "grad_norm": 1.5638922452926636, | |
| "learning_rate": 9.840523917523354e-06, | |
| "loss": 0.1376, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 26.0, | |
| "grad_norm": 1.4766254425048828, | |
| "learning_rate": 9.83707609731432e-06, | |
| "loss": 0.228, | |
| "step": 442 | |
| }, | |
| { | |
| "epoch": 26.0, | |
| "eval_loss": 0.5160675644874573, | |
| "eval_runtime": 31.8614, | |
| "eval_samples_per_second": 1.13, | |
| "eval_steps_per_second": 1.13, | |
| "step": 442 | |
| }, | |
| { | |
| "epoch": 26.11764705882353, | |
| "grad_norm": 0.983343780040741, | |
| "learning_rate": 9.833592021345938e-06, | |
| "loss": 0.1742, | |
| "step": 444 | |
| }, | |
| { | |
| "epoch": 26.235294117647058, | |
| "grad_norm": 1.3315422534942627, | |
| "learning_rate": 9.830071715732708e-06, | |
| "loss": 0.1853, | |
| "step": 446 | |
| }, | |
| { | |
| "epoch": 26.352941176470587, | |
| "grad_norm": 0.6733668446540833, | |
| "learning_rate": 9.826515206860683e-06, | |
| "loss": 0.1493, | |
| "step": 448 | |
| }, | |
| { | |
| "epoch": 26.470588235294116, | |
| "grad_norm": 1.4821362495422363, | |
| "learning_rate": 9.822922521387277e-06, | |
| "loss": 0.2319, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 26.58823529411765, | |
| "grad_norm": 3.0113203525543213, | |
| "learning_rate": 9.819293686241057e-06, | |
| "loss": 0.253, | |
| "step": 452 | |
| }, | |
| { | |
| "epoch": 26.705882352941178, | |
| "grad_norm": 2.0522031784057617, | |
| "learning_rate": 9.81562872862155e-06, | |
| "loss": 0.1756, | |
| "step": 454 | |
| }, | |
| { | |
| "epoch": 26.823529411764707, | |
| "grad_norm": 2.2314300537109375, | |
| "learning_rate": 9.811927675999035e-06, | |
| "loss": 0.1977, | |
| "step": 456 | |
| }, | |
| { | |
| "epoch": 26.941176470588236, | |
| "grad_norm": 1.571890115737915, | |
| "learning_rate": 9.808190556114333e-06, | |
| "loss": 0.2127, | |
| "step": 458 | |
| }, | |
| { | |
| "epoch": 27.0, | |
| "eval_loss": 0.5205699801445007, | |
| "eval_runtime": 31.8543, | |
| "eval_samples_per_second": 1.13, | |
| "eval_steps_per_second": 1.13, | |
| "step": 459 | |
| }, | |
| { | |
| "epoch": 27.0, | |
| "step": 459, | |
| "total_flos": 2.8798901486092288e+17, | |
| "train_loss": 0.8010637635823689, | |
| "train_runtime": 10567.3661, | |
| "train_samples_per_second": 1.93, | |
| "train_steps_per_second": 0.241 | |
| } | |
| ], | |
| "logging_steps": 2, | |
| "max_steps": 2550, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 150, | |
| "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": 2.8798901486092288e+17, | |
| "train_batch_size": 1, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |