| { | |
| "best_metric": 0.7812196612358093, | |
| "best_model_checkpoint": "data/Mistral-7B_task-1_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": 3.0000112056732178, | |
| "learning_rate": 1.1764705882352942e-06, | |
| "loss": 1.0291, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.11764705882352941, | |
| "grad_norm": 2.661545753479004, | |
| "learning_rate": 2.3529411764705885e-06, | |
| "loss": 1.0773, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.23529411764705882, | |
| "grad_norm": 2.614375352859497, | |
| "learning_rate": 4.705882352941177e-06, | |
| "loss": 1.0342, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.35294117647058826, | |
| "grad_norm": 2.8368232250213623, | |
| "learning_rate": 7.058823529411765e-06, | |
| "loss": 1.1009, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.47058823529411764, | |
| "grad_norm": 2.8628907203674316, | |
| "learning_rate": 9.411764705882354e-06, | |
| "loss": 1.1206, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.5882352941176471, | |
| "grad_norm": 2.4519965648651123, | |
| "learning_rate": 1.1764705882352942e-05, | |
| "loss": 1.0535, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.7058823529411765, | |
| "grad_norm": 2.387321710586548, | |
| "learning_rate": 1.411764705882353e-05, | |
| "loss": 1.0246, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.8235294117647058, | |
| "grad_norm": 2.3504133224487305, | |
| "learning_rate": 1.647058823529412e-05, | |
| "loss": 1.0416, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.9411764705882353, | |
| "grad_norm": 2.2138402462005615, | |
| "learning_rate": 1.8823529411764708e-05, | |
| "loss": 0.8971, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_loss": 0.9035294055938721, | |
| "eval_runtime": 14.9925, | |
| "eval_samples_per_second": 2.401, | |
| "eval_steps_per_second": 2.401, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 1.0588235294117647, | |
| "grad_norm": 1.651427984237671, | |
| "learning_rate": 2.1176470588235296e-05, | |
| "loss": 0.9067, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 1.1764705882352942, | |
| "grad_norm": 1.7093214988708496, | |
| "learning_rate": 2.3529411764705884e-05, | |
| "loss": 0.8839, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 1.2941176470588236, | |
| "grad_norm": 1.5007519721984863, | |
| "learning_rate": 2.5882352941176475e-05, | |
| "loss": 0.7932, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 1.4117647058823528, | |
| "grad_norm": 1.4336295127868652, | |
| "learning_rate": 2.823529411764706e-05, | |
| "loss": 0.8126, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 1.5294117647058822, | |
| "grad_norm": 1.422503113746643, | |
| "learning_rate": 3.058823529411765e-05, | |
| "loss": 0.8485, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 1.6470588235294117, | |
| "grad_norm": 1.42384672164917, | |
| "learning_rate": 3.294117647058824e-05, | |
| "loss": 0.8008, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 1.7647058823529411, | |
| "grad_norm": 1.3979958295822144, | |
| "learning_rate": 3.529411764705883e-05, | |
| "loss": 0.7584, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 1.8823529411764706, | |
| "grad_norm": 1.555219054222107, | |
| "learning_rate": 3.7647058823529415e-05, | |
| "loss": 0.7696, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 1.317792296409607, | |
| "learning_rate": 4e-05, | |
| "loss": 0.7275, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_loss": 0.8006917238235474, | |
| "eval_runtime": 14.9876, | |
| "eval_samples_per_second": 2.402, | |
| "eval_steps_per_second": 2.402, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 2.1176470588235294, | |
| "grad_norm": 1.2864298820495605, | |
| "learning_rate": 4.235294117647059e-05, | |
| "loss": 0.6245, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 2.235294117647059, | |
| "grad_norm": 1.4150673151016235, | |
| "learning_rate": 4.470588235294118e-05, | |
| "loss": 0.6272, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 2.3529411764705883, | |
| "grad_norm": 1.522194743156433, | |
| "learning_rate": 4.705882352941177e-05, | |
| "loss": 0.6528, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 2.4705882352941178, | |
| "grad_norm": 1.6460087299346924, | |
| "learning_rate": 4.9411764705882355e-05, | |
| "loss": 0.6591, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 2.588235294117647, | |
| "grad_norm": 1.3955222368240356, | |
| "learning_rate": 5.176470588235295e-05, | |
| "loss": 0.6425, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 2.7058823529411766, | |
| "grad_norm": 1.6861988306045532, | |
| "learning_rate": 5.411764705882353e-05, | |
| "loss": 0.6572, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 2.8235294117647056, | |
| "grad_norm": 1.5515007972717285, | |
| "learning_rate": 5.647058823529412e-05, | |
| "loss": 0.6721, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 2.9411764705882355, | |
| "grad_norm": 1.850741982460022, | |
| "learning_rate": 5.882352941176471e-05, | |
| "loss": 0.647, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_loss": 0.7812196612358093, | |
| "eval_runtime": 14.9758, | |
| "eval_samples_per_second": 2.404, | |
| "eval_steps_per_second": 2.404, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 3.0588235294117645, | |
| "grad_norm": 1.4639921188354492, | |
| "learning_rate": 6.11764705882353e-05, | |
| "loss": 0.5502, | |
| "step": 52 | |
| }, | |
| { | |
| "epoch": 3.176470588235294, | |
| "grad_norm": 1.5166435241699219, | |
| "learning_rate": 6.352941176470588e-05, | |
| "loss": 0.4734, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 3.2941176470588234, | |
| "grad_norm": 1.6939408779144287, | |
| "learning_rate": 6.588235294117648e-05, | |
| "loss": 0.4723, | |
| "step": 56 | |
| }, | |
| { | |
| "epoch": 3.411764705882353, | |
| "grad_norm": 2.04142427444458, | |
| "learning_rate": 6.823529411764707e-05, | |
| "loss": 0.5141, | |
| "step": 58 | |
| }, | |
| { | |
| "epoch": 3.5294117647058822, | |
| "grad_norm": 2.538571834564209, | |
| "learning_rate": 7.058823529411765e-05, | |
| "loss": 0.4387, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 3.6470588235294117, | |
| "grad_norm": 2.7145118713378906, | |
| "learning_rate": 7.294117647058823e-05, | |
| "loss": 0.4559, | |
| "step": 62 | |
| }, | |
| { | |
| "epoch": 3.764705882352941, | |
| "grad_norm": 2.054623603820801, | |
| "learning_rate": 7.529411764705883e-05, | |
| "loss": 0.4012, | |
| "step": 64 | |
| }, | |
| { | |
| "epoch": 3.8823529411764706, | |
| "grad_norm": 2.1475977897644043, | |
| "learning_rate": 7.764705882352942e-05, | |
| "loss": 0.4253, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "grad_norm": 2.132322072982788, | |
| "learning_rate": 8e-05, | |
| "loss": 0.4555, | |
| "step": 68 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_loss": 0.8473093509674072, | |
| "eval_runtime": 14.9798, | |
| "eval_samples_per_second": 2.403, | |
| "eval_steps_per_second": 2.403, | |
| "step": 68 | |
| }, | |
| { | |
| "epoch": 4.117647058823529, | |
| "grad_norm": 2.356788396835327, | |
| "learning_rate": 8.23529411764706e-05, | |
| "loss": 0.3152, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 4.235294117647059, | |
| "grad_norm": 1.7438907623291016, | |
| "learning_rate": 8.470588235294118e-05, | |
| "loss": 0.2572, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 4.352941176470588, | |
| "grad_norm": 2.352015256881714, | |
| "learning_rate": 8.705882352941177e-05, | |
| "loss": 0.2326, | |
| "step": 74 | |
| }, | |
| { | |
| "epoch": 4.470588235294118, | |
| "grad_norm": 3.033743381500244, | |
| "learning_rate": 8.941176470588236e-05, | |
| "loss": 0.2204, | |
| "step": 76 | |
| }, | |
| { | |
| "epoch": 4.588235294117647, | |
| "grad_norm": 2.743682384490967, | |
| "learning_rate": 9.176470588235295e-05, | |
| "loss": 0.2725, | |
| "step": 78 | |
| }, | |
| { | |
| "epoch": 4.705882352941177, | |
| "grad_norm": 2.717099666595459, | |
| "learning_rate": 9.411764705882353e-05, | |
| "loss": 0.2415, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 4.823529411764706, | |
| "grad_norm": 3.2620532512664795, | |
| "learning_rate": 9.647058823529412e-05, | |
| "loss": 0.2948, | |
| "step": 82 | |
| }, | |
| { | |
| "epoch": 4.9411764705882355, | |
| "grad_norm": 2.446908950805664, | |
| "learning_rate": 9.882352941176471e-05, | |
| "loss": 0.2101, | |
| "step": 84 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_loss": 1.0110052824020386, | |
| "eval_runtime": 14.9964, | |
| "eval_samples_per_second": 2.401, | |
| "eval_steps_per_second": 2.401, | |
| "step": 85 | |
| }, | |
| { | |
| "epoch": 5.0588235294117645, | |
| "grad_norm": 1.7935397624969482, | |
| "learning_rate": 9.99995783847866e-05, | |
| "loss": 0.172, | |
| "step": 86 | |
| }, | |
| { | |
| "epoch": 5.176470588235294, | |
| "grad_norm": 1.4242950677871704, | |
| "learning_rate": 9.999620550574153e-05, | |
| "loss": 0.1432, | |
| "step": 88 | |
| }, | |
| { | |
| "epoch": 5.294117647058823, | |
| "grad_norm": 2.0434722900390625, | |
| "learning_rate": 9.998945997517956e-05, | |
| "loss": 0.1636, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 5.411764705882353, | |
| "grad_norm": 1.9013702869415283, | |
| "learning_rate": 9.997934224814173e-05, | |
| "loss": 0.113, | |
| "step": 92 | |
| }, | |
| { | |
| "epoch": 5.529411764705882, | |
| "grad_norm": 2.909445285797119, | |
| "learning_rate": 9.996585300715116e-05, | |
| "loss": 0.1469, | |
| "step": 94 | |
| }, | |
| { | |
| "epoch": 5.647058823529412, | |
| "grad_norm": 3.0097384452819824, | |
| "learning_rate": 9.994899316216708e-05, | |
| "loss": 0.1429, | |
| "step": 96 | |
| }, | |
| { | |
| "epoch": 5.764705882352941, | |
| "grad_norm": 2.6940877437591553, | |
| "learning_rate": 9.992876385052345e-05, | |
| "loss": 0.1463, | |
| "step": 98 | |
| }, | |
| { | |
| "epoch": 5.882352941176471, | |
| "grad_norm": 2.378119707107544, | |
| "learning_rate": 9.990516643685222e-05, | |
| "loss": 0.1581, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "grad_norm": 2.717642068862915, | |
| "learning_rate": 9.987820251299122e-05, | |
| "loss": 0.1745, | |
| "step": 102 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_loss": 1.163377285003662, | |
| "eval_runtime": 14.9789, | |
| "eval_samples_per_second": 2.403, | |
| "eval_steps_per_second": 2.403, | |
| "step": 102 | |
| }, | |
| { | |
| "epoch": 6.117647058823529, | |
| "grad_norm": 1.3047163486480713, | |
| "learning_rate": 9.984787389787688e-05, | |
| "loss": 0.0861, | |
| "step": 104 | |
| }, | |
| { | |
| "epoch": 6.235294117647059, | |
| "grad_norm": 1.8909651041030884, | |
| "learning_rate": 9.981418263742148e-05, | |
| "loss": 0.0941, | |
| "step": 106 | |
| }, | |
| { | |
| "epoch": 6.352941176470588, | |
| "grad_norm": 1.8301990032196045, | |
| "learning_rate": 9.977713100437509e-05, | |
| "loss": 0.0988, | |
| "step": 108 | |
| }, | |
| { | |
| "epoch": 6.470588235294118, | |
| "grad_norm": 1.83213472366333, | |
| "learning_rate": 9.973672149817232e-05, | |
| "loss": 0.089, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 6.588235294117647, | |
| "grad_norm": 1.5029006004333496, | |
| "learning_rate": 9.96929568447637e-05, | |
| "loss": 0.0535, | |
| "step": 112 | |
| }, | |
| { | |
| "epoch": 6.705882352941177, | |
| "grad_norm": 1.5775741338729858, | |
| "learning_rate": 9.964583999643174e-05, | |
| "loss": 0.0791, | |
| "step": 114 | |
| }, | |
| { | |
| "epoch": 6.823529411764706, | |
| "grad_norm": 2.908430814743042, | |
| "learning_rate": 9.95953741315919e-05, | |
| "loss": 0.1056, | |
| "step": 116 | |
| }, | |
| { | |
| "epoch": 6.9411764705882355, | |
| "grad_norm": 2.4293549060821533, | |
| "learning_rate": 9.954156265457801e-05, | |
| "loss": 0.0895, | |
| "step": 118 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_loss": 1.2142539024353027, | |
| "eval_runtime": 14.9804, | |
| "eval_samples_per_second": 2.403, | |
| "eval_steps_per_second": 2.403, | |
| "step": 119 | |
| }, | |
| { | |
| "epoch": 7.0588235294117645, | |
| "grad_norm": 1.420074701309204, | |
| "learning_rate": 9.948440919541278e-05, | |
| "loss": 0.0834, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 7.176470588235294, | |
| "grad_norm": 1.7070257663726807, | |
| "learning_rate": 9.942391760956277e-05, | |
| "loss": 0.0815, | |
| "step": 122 | |
| }, | |
| { | |
| "epoch": 7.294117647058823, | |
| "grad_norm": 1.1222933530807495, | |
| "learning_rate": 9.936009197767845e-05, | |
| "loss": 0.0592, | |
| "step": 124 | |
| }, | |
| { | |
| "epoch": 7.411764705882353, | |
| "grad_norm": 1.8239096403121948, | |
| "learning_rate": 9.929293660531888e-05, | |
| "loss": 0.0609, | |
| "step": 126 | |
| }, | |
| { | |
| "epoch": 7.529411764705882, | |
| "grad_norm": 2.7660880088806152, | |
| "learning_rate": 9.922245602266118e-05, | |
| "loss": 0.0876, | |
| "step": 128 | |
| }, | |
| { | |
| "epoch": 7.647058823529412, | |
| "grad_norm": 1.7250267267227173, | |
| "learning_rate": 9.91486549841951e-05, | |
| "loss": 0.0857, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 7.764705882352941, | |
| "grad_norm": 1.865657925605774, | |
| "learning_rate": 9.90715384684021e-05, | |
| "loss": 0.0565, | |
| "step": 132 | |
| }, | |
| { | |
| "epoch": 7.882352941176471, | |
| "grad_norm": 1.6430420875549316, | |
| "learning_rate": 9.899111167741966e-05, | |
| "loss": 0.0763, | |
| "step": 134 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "grad_norm": 2.06386661529541, | |
| "learning_rate": 9.890738003669029e-05, | |
| "loss": 0.0809, | |
| "step": 136 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_loss": 1.2212287187576294, | |
| "eval_runtime": 14.9765, | |
| "eval_samples_per_second": 2.404, | |
| "eval_steps_per_second": 2.404, | |
| "step": 136 | |
| }, | |
| { | |
| "epoch": 8.117647058823529, | |
| "grad_norm": 1.1084154844284058, | |
| "learning_rate": 9.882034919459555e-05, | |
| "loss": 0.0436, | |
| "step": 138 | |
| }, | |
| { | |
| "epoch": 8.235294117647058, | |
| "grad_norm": 1.1482659578323364, | |
| "learning_rate": 9.873002502207503e-05, | |
| "loss": 0.0477, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 8.352941176470589, | |
| "grad_norm": 1.9473789930343628, | |
| "learning_rate": 9.863641361223024e-05, | |
| "loss": 0.0687, | |
| "step": 142 | |
| }, | |
| { | |
| "epoch": 8.470588235294118, | |
| "grad_norm": 2.872997999191284, | |
| "learning_rate": 9.853952127991372e-05, | |
| "loss": 0.071, | |
| "step": 144 | |
| }, | |
| { | |
| "epoch": 8.588235294117647, | |
| "grad_norm": 1.7821139097213745, | |
| "learning_rate": 9.843935456130295e-05, | |
| "loss": 0.0649, | |
| "step": 146 | |
| }, | |
| { | |
| "epoch": 8.705882352941176, | |
| "grad_norm": 1.458869457244873, | |
| "learning_rate": 9.833592021345937e-05, | |
| "loss": 0.0685, | |
| "step": 148 | |
| }, | |
| { | |
| "epoch": 8.823529411764707, | |
| "grad_norm": 1.8370617628097534, | |
| "learning_rate": 9.822922521387276e-05, | |
| "loss": 0.0707, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 8.941176470588236, | |
| "grad_norm": 1.6871691942214966, | |
| "learning_rate": 9.811927675999036e-05, | |
| "loss": 0.0638, | |
| "step": 152 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_loss": 1.2625036239624023, | |
| "eval_runtime": 14.9787, | |
| "eval_samples_per_second": 2.403, | |
| "eval_steps_per_second": 2.403, | |
| "step": 153 | |
| }, | |
| { | |
| "epoch": 9.058823529411764, | |
| "grad_norm": 0.8397995829582214, | |
| "learning_rate": 9.800608226873142e-05, | |
| "loss": 0.0399, | |
| "step": 154 | |
| }, | |
| { | |
| "epoch": 9.176470588235293, | |
| "grad_norm": 1.1770823001861572, | |
| "learning_rate": 9.788964937598689e-05, | |
| "loss": 0.0437, | |
| "step": 156 | |
| }, | |
| { | |
| "epoch": 9.294117647058824, | |
| "grad_norm": 1.2451812028884888, | |
| "learning_rate": 9.776998593610428e-05, | |
| "loss": 0.0368, | |
| "step": 158 | |
| }, | |
| { | |
| "epoch": 9.411764705882353, | |
| "grad_norm": 1.8193283081054688, | |
| "learning_rate": 9.764710002135784e-05, | |
| "loss": 0.0527, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 9.529411764705882, | |
| "grad_norm": 2.604790449142456, | |
| "learning_rate": 9.752099992140399e-05, | |
| "loss": 0.0444, | |
| "step": 162 | |
| }, | |
| { | |
| "epoch": 9.647058823529411, | |
| "grad_norm": 2.581681728363037, | |
| "learning_rate": 9.739169414272217e-05, | |
| "loss": 0.0484, | |
| "step": 164 | |
| }, | |
| { | |
| "epoch": 9.764705882352942, | |
| "grad_norm": 2.257154703140259, | |
| "learning_rate": 9.725919140804099e-05, | |
| "loss": 0.0472, | |
| "step": 166 | |
| }, | |
| { | |
| "epoch": 9.882352941176471, | |
| "grad_norm": 1.5224213600158691, | |
| "learning_rate": 9.71235006557497e-05, | |
| "loss": 0.0417, | |
| "step": 168 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "grad_norm": 2.058380365371704, | |
| "learning_rate": 9.698463103929542e-05, | |
| "loss": 0.0522, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_loss": 1.4163453578948975, | |
| "eval_runtime": 14.979, | |
| "eval_samples_per_second": 2.403, | |
| "eval_steps_per_second": 2.403, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "step": 170, | |
| "total_flos": 2.289524785597645e+16, | |
| "train_loss": 0.3608771165504175, | |
| "train_runtime": 1659.2787, | |
| "train_samples_per_second": 4.098, | |
| "train_steps_per_second": 0.512 | |
| } | |
| ], | |
| "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": 2.289524785597645e+16, | |
| "train_batch_size": 1, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |