| { | |
| "best_metric": 0.4649590849876404, | |
| "best_model_checkpoint": "data/Mistral-7B_task-2_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": 1.604433298110962, | |
| "learning_rate": 1.1764705882352942e-06, | |
| "loss": 0.7575, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.11764705882352941, | |
| "grad_norm": 1.6808534860610962, | |
| "learning_rate": 2.3529411764705885e-06, | |
| "loss": 0.8355, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.23529411764705882, | |
| "grad_norm": 1.9203530550003052, | |
| "learning_rate": 4.705882352941177e-06, | |
| "loss": 0.8264, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.35294117647058826, | |
| "grad_norm": 1.7332963943481445, | |
| "learning_rate": 7.058823529411765e-06, | |
| "loss": 0.769, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.47058823529411764, | |
| "grad_norm": 1.5537375211715698, | |
| "learning_rate": 9.411764705882354e-06, | |
| "loss": 0.7582, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.5882352941176471, | |
| "grad_norm": 1.2597177028656006, | |
| "learning_rate": 1.1764705882352942e-05, | |
| "loss": 0.7485, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.7058823529411765, | |
| "grad_norm": 1.1016687154769897, | |
| "learning_rate": 1.411764705882353e-05, | |
| "loss": 0.6831, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.8235294117647058, | |
| "grad_norm": 1.020566463470459, | |
| "learning_rate": 1.647058823529412e-05, | |
| "loss": 0.6637, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.9411764705882353, | |
| "grad_norm": 0.928702712059021, | |
| "learning_rate": 1.8823529411764708e-05, | |
| "loss": 0.6307, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_loss": 0.6122480034828186, | |
| "eval_runtime": 46.8114, | |
| "eval_samples_per_second": 0.769, | |
| "eval_steps_per_second": 0.769, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 1.0588235294117647, | |
| "grad_norm": 0.8042245507240295, | |
| "learning_rate": 2.1176470588235296e-05, | |
| "loss": 0.6038, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 1.1764705882352942, | |
| "grad_norm": 0.7779905796051025, | |
| "learning_rate": 2.3529411764705884e-05, | |
| "loss": 0.5814, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 1.2941176470588236, | |
| "grad_norm": 0.6743247509002686, | |
| "learning_rate": 2.5882352941176475e-05, | |
| "loss": 0.583, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 1.4117647058823528, | |
| "grad_norm": 0.7200171947479248, | |
| "learning_rate": 2.823529411764706e-05, | |
| "loss": 0.5281, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 1.5294117647058822, | |
| "grad_norm": 0.721940279006958, | |
| "learning_rate": 3.058823529411765e-05, | |
| "loss": 0.5485, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 1.6470588235294117, | |
| "grad_norm": 0.6974021792411804, | |
| "learning_rate": 3.294117647058824e-05, | |
| "loss": 0.5226, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 1.7647058823529411, | |
| "grad_norm": 0.7087254524230957, | |
| "learning_rate": 3.529411764705883e-05, | |
| "loss": 0.5171, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 1.8823529411764706, | |
| "grad_norm": 0.7080815434455872, | |
| "learning_rate": 3.7647058823529415e-05, | |
| "loss": 0.4933, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 0.7730523347854614, | |
| "learning_rate": 4e-05, | |
| "loss": 0.5229, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_loss": 0.5021528601646423, | |
| "eval_runtime": 46.8011, | |
| "eval_samples_per_second": 0.769, | |
| "eval_steps_per_second": 0.769, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 2.1176470588235294, | |
| "grad_norm": 0.8364313244819641, | |
| "learning_rate": 4.235294117647059e-05, | |
| "loss": 0.4658, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 2.235294117647059, | |
| "grad_norm": 0.6758615374565125, | |
| "learning_rate": 4.470588235294118e-05, | |
| "loss": 0.4401, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 2.3529411764705883, | |
| "grad_norm": 0.7340723276138306, | |
| "learning_rate": 4.705882352941177e-05, | |
| "loss": 0.4382, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 2.4705882352941178, | |
| "grad_norm": 0.6828162670135498, | |
| "learning_rate": 4.9411764705882355e-05, | |
| "loss": 0.4553, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 2.588235294117647, | |
| "grad_norm": 0.7578108906745911, | |
| "learning_rate": 5.176470588235295e-05, | |
| "loss": 0.4269, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 2.7058823529411766, | |
| "grad_norm": 0.8680711388587952, | |
| "learning_rate": 5.411764705882353e-05, | |
| "loss": 0.4517, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 2.8235294117647056, | |
| "grad_norm": 0.7644243836402893, | |
| "learning_rate": 5.647058823529412e-05, | |
| "loss": 0.4636, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 2.9411764705882355, | |
| "grad_norm": 0.7221245169639587, | |
| "learning_rate": 5.882352941176471e-05, | |
| "loss": 0.4431, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_loss": 0.4649590849876404, | |
| "eval_runtime": 46.8112, | |
| "eval_samples_per_second": 0.769, | |
| "eval_steps_per_second": 0.769, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 3.0588235294117645, | |
| "grad_norm": 0.6552745699882507, | |
| "learning_rate": 6.11764705882353e-05, | |
| "loss": 0.3813, | |
| "step": 52 | |
| }, | |
| { | |
| "epoch": 3.176470588235294, | |
| "grad_norm": 0.6913314461708069, | |
| "learning_rate": 6.352941176470588e-05, | |
| "loss": 0.3715, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 3.2941176470588234, | |
| "grad_norm": 0.7282626032829285, | |
| "learning_rate": 6.588235294117648e-05, | |
| "loss": 0.3842, | |
| "step": 56 | |
| }, | |
| { | |
| "epoch": 3.411764705882353, | |
| "grad_norm": 0.8809773325920105, | |
| "learning_rate": 6.823529411764707e-05, | |
| "loss": 0.3815, | |
| "step": 58 | |
| }, | |
| { | |
| "epoch": 3.5294117647058822, | |
| "grad_norm": 0.802264392375946, | |
| "learning_rate": 7.058823529411765e-05, | |
| "loss": 0.3411, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 3.6470588235294117, | |
| "grad_norm": 0.8780251145362854, | |
| "learning_rate": 7.294117647058823e-05, | |
| "loss": 0.3743, | |
| "step": 62 | |
| }, | |
| { | |
| "epoch": 3.764705882352941, | |
| "grad_norm": 0.8712494969367981, | |
| "learning_rate": 7.529411764705883e-05, | |
| "loss": 0.3493, | |
| "step": 64 | |
| }, | |
| { | |
| "epoch": 3.8823529411764706, | |
| "grad_norm": 0.9024491906166077, | |
| "learning_rate": 7.764705882352942e-05, | |
| "loss": 0.3355, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "grad_norm": 0.8319352269172668, | |
| "learning_rate": 8e-05, | |
| "loss": 0.3222, | |
| "step": 68 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_loss": 0.47462204098701477, | |
| "eval_runtime": 46.8066, | |
| "eval_samples_per_second": 0.769, | |
| "eval_steps_per_second": 0.769, | |
| "step": 68 | |
| }, | |
| { | |
| "epoch": 4.117647058823529, | |
| "grad_norm": 0.850300669670105, | |
| "learning_rate": 8.23529411764706e-05, | |
| "loss": 0.293, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 4.235294117647059, | |
| "grad_norm": 1.0695669651031494, | |
| "learning_rate": 8.470588235294118e-05, | |
| "loss": 0.2255, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 4.352941176470588, | |
| "grad_norm": 1.4161185026168823, | |
| "learning_rate": 8.705882352941177e-05, | |
| "loss": 0.2726, | |
| "step": 74 | |
| }, | |
| { | |
| "epoch": 4.470588235294118, | |
| "grad_norm": 1.1317994594573975, | |
| "learning_rate": 8.941176470588236e-05, | |
| "loss": 0.2301, | |
| "step": 76 | |
| }, | |
| { | |
| "epoch": 4.588235294117647, | |
| "grad_norm": 1.1033505201339722, | |
| "learning_rate": 9.176470588235295e-05, | |
| "loss": 0.2487, | |
| "step": 78 | |
| }, | |
| { | |
| "epoch": 4.705882352941177, | |
| "grad_norm": 0.9940263032913208, | |
| "learning_rate": 9.411764705882353e-05, | |
| "loss": 0.2427, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 4.823529411764706, | |
| "grad_norm": 1.0770084857940674, | |
| "learning_rate": 9.647058823529412e-05, | |
| "loss": 0.2379, | |
| "step": 82 | |
| }, | |
| { | |
| "epoch": 4.9411764705882355, | |
| "grad_norm": 1.088417887687683, | |
| "learning_rate": 9.882352941176471e-05, | |
| "loss": 0.2227, | |
| "step": 84 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_loss": 0.5480709671974182, | |
| "eval_runtime": 46.8144, | |
| "eval_samples_per_second": 0.769, | |
| "eval_steps_per_second": 0.769, | |
| "step": 85 | |
| }, | |
| { | |
| "epoch": 5.0588235294117645, | |
| "grad_norm": 1.105506181716919, | |
| "learning_rate": 9.99995783847866e-05, | |
| "loss": 0.171, | |
| "step": 86 | |
| }, | |
| { | |
| "epoch": 5.176470588235294, | |
| "grad_norm": 1.0123800039291382, | |
| "learning_rate": 9.999620550574153e-05, | |
| "loss": 0.1287, | |
| "step": 88 | |
| }, | |
| { | |
| "epoch": 5.294117647058823, | |
| "grad_norm": 1.8314290046691895, | |
| "learning_rate": 9.998945997517956e-05, | |
| "loss": 0.1435, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 5.411764705882353, | |
| "grad_norm": 1.3103504180908203, | |
| "learning_rate": 9.997934224814173e-05, | |
| "loss": 0.1292, | |
| "step": 92 | |
| }, | |
| { | |
| "epoch": 5.529411764705882, | |
| "grad_norm": 1.3026607036590576, | |
| "learning_rate": 9.996585300715116e-05, | |
| "loss": 0.1246, | |
| "step": 94 | |
| }, | |
| { | |
| "epoch": 5.647058823529412, | |
| "grad_norm": 1.083115577697754, | |
| "learning_rate": 9.994899316216708e-05, | |
| "loss": 0.1095, | |
| "step": 96 | |
| }, | |
| { | |
| "epoch": 5.764705882352941, | |
| "grad_norm": 1.2289258241653442, | |
| "learning_rate": 9.992876385052345e-05, | |
| "loss": 0.135, | |
| "step": 98 | |
| }, | |
| { | |
| "epoch": 5.882352941176471, | |
| "grad_norm": 1.2652201652526855, | |
| "learning_rate": 9.990516643685222e-05, | |
| "loss": 0.1443, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "grad_norm": 1.2546310424804688, | |
| "learning_rate": 9.987820251299122e-05, | |
| "loss": 0.1345, | |
| "step": 102 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_loss": 0.6141153573989868, | |
| "eval_runtime": 46.8023, | |
| "eval_samples_per_second": 0.769, | |
| "eval_steps_per_second": 0.769, | |
| "step": 102 | |
| }, | |
| { | |
| "epoch": 6.117647058823529, | |
| "grad_norm": 0.9022185802459717, | |
| "learning_rate": 9.984787389787688e-05, | |
| "loss": 0.0666, | |
| "step": 104 | |
| }, | |
| { | |
| "epoch": 6.235294117647059, | |
| "grad_norm": 1.1086210012435913, | |
| "learning_rate": 9.981418263742148e-05, | |
| "loss": 0.0751, | |
| "step": 106 | |
| }, | |
| { | |
| "epoch": 6.352941176470588, | |
| "grad_norm": 0.9500546455383301, | |
| "learning_rate": 9.977713100437509e-05, | |
| "loss": 0.0622, | |
| "step": 108 | |
| }, | |
| { | |
| "epoch": 6.470588235294118, | |
| "grad_norm": 1.2992115020751953, | |
| "learning_rate": 9.973672149817232e-05, | |
| "loss": 0.0763, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 6.588235294117647, | |
| "grad_norm": 1.0361261367797852, | |
| "learning_rate": 9.96929568447637e-05, | |
| "loss": 0.0575, | |
| "step": 112 | |
| }, | |
| { | |
| "epoch": 6.705882352941177, | |
| "grad_norm": 1.0972284078598022, | |
| "learning_rate": 9.964583999643174e-05, | |
| "loss": 0.0718, | |
| "step": 114 | |
| }, | |
| { | |
| "epoch": 6.823529411764706, | |
| "grad_norm": 1.0893183946609497, | |
| "learning_rate": 9.95953741315919e-05, | |
| "loss": 0.0758, | |
| "step": 116 | |
| }, | |
| { | |
| "epoch": 6.9411764705882355, | |
| "grad_norm": 1.0908876657485962, | |
| "learning_rate": 9.954156265457801e-05, | |
| "loss": 0.0738, | |
| "step": 118 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_loss": 0.6841809749603271, | |
| "eval_runtime": 46.7966, | |
| "eval_samples_per_second": 0.769, | |
| "eval_steps_per_second": 0.769, | |
| "step": 119 | |
| }, | |
| { | |
| "epoch": 7.0588235294117645, | |
| "grad_norm": 0.6503316760063171, | |
| "learning_rate": 9.948440919541278e-05, | |
| "loss": 0.0537, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 7.176470588235294, | |
| "grad_norm": 0.8566131591796875, | |
| "learning_rate": 9.942391760956277e-05, | |
| "loss": 0.0414, | |
| "step": 122 | |
| }, | |
| { | |
| "epoch": 7.294117647058823, | |
| "grad_norm": 0.9488412141799927, | |
| "learning_rate": 9.936009197767845e-05, | |
| "loss": 0.038, | |
| "step": 124 | |
| }, | |
| { | |
| "epoch": 7.411764705882353, | |
| "grad_norm": 1.3826652765274048, | |
| "learning_rate": 9.929293660531888e-05, | |
| "loss": 0.0442, | |
| "step": 126 | |
| }, | |
| { | |
| "epoch": 7.529411764705882, | |
| "grad_norm": 1.1183003187179565, | |
| "learning_rate": 9.922245602266118e-05, | |
| "loss": 0.0462, | |
| "step": 128 | |
| }, | |
| { | |
| "epoch": 7.647058823529412, | |
| "grad_norm": 1.0122565031051636, | |
| "learning_rate": 9.91486549841951e-05, | |
| "loss": 0.0398, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 7.764705882352941, | |
| "grad_norm": 1.112568736076355, | |
| "learning_rate": 9.90715384684021e-05, | |
| "loss": 0.04, | |
| "step": 132 | |
| }, | |
| { | |
| "epoch": 7.882352941176471, | |
| "grad_norm": 0.8670180439949036, | |
| "learning_rate": 9.899111167741966e-05, | |
| "loss": 0.0465, | |
| "step": 134 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "grad_norm": 0.8402408361434937, | |
| "learning_rate": 9.890738003669029e-05, | |
| "loss": 0.0448, | |
| "step": 136 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_loss": 0.7531328201293945, | |
| "eval_runtime": 46.7972, | |
| "eval_samples_per_second": 0.769, | |
| "eval_steps_per_second": 0.769, | |
| "step": 136 | |
| }, | |
| { | |
| "epoch": 8.117647058823529, | |
| "grad_norm": 0.562960147857666, | |
| "learning_rate": 9.882034919459555e-05, | |
| "loss": 0.0196, | |
| "step": 138 | |
| }, | |
| { | |
| "epoch": 8.235294117647058, | |
| "grad_norm": 1.4091694355010986, | |
| "learning_rate": 9.873002502207503e-05, | |
| "loss": 0.0306, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 8.352941176470589, | |
| "grad_norm": 1.0141454935073853, | |
| "learning_rate": 9.863641361223024e-05, | |
| "loss": 0.0321, | |
| "step": 142 | |
| }, | |
| { | |
| "epoch": 8.470588235294118, | |
| "grad_norm": 0.7341058254241943, | |
| "learning_rate": 9.853952127991372e-05, | |
| "loss": 0.0315, | |
| "step": 144 | |
| }, | |
| { | |
| "epoch": 8.588235294117647, | |
| "grad_norm": 0.9098450541496277, | |
| "learning_rate": 9.843935456130295e-05, | |
| "loss": 0.0296, | |
| "step": 146 | |
| }, | |
| { | |
| "epoch": 8.705882352941176, | |
| "grad_norm": 0.6944416761398315, | |
| "learning_rate": 9.833592021345937e-05, | |
| "loss": 0.0249, | |
| "step": 148 | |
| }, | |
| { | |
| "epoch": 8.823529411764707, | |
| "grad_norm": 0.7934401035308838, | |
| "learning_rate": 9.822922521387276e-05, | |
| "loss": 0.0271, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 8.941176470588236, | |
| "grad_norm": 1.0908757448196411, | |
| "learning_rate": 9.811927675999036e-05, | |
| "loss": 0.0352, | |
| "step": 152 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_loss": 0.8120158910751343, | |
| "eval_runtime": 46.8083, | |
| "eval_samples_per_second": 0.769, | |
| "eval_steps_per_second": 0.769, | |
| "step": 153 | |
| }, | |
| { | |
| "epoch": 9.058823529411764, | |
| "grad_norm": 0.6170806884765625, | |
| "learning_rate": 9.800608226873142e-05, | |
| "loss": 0.0242, | |
| "step": 154 | |
| }, | |
| { | |
| "epoch": 9.176470588235293, | |
| "grad_norm": 0.46037042140960693, | |
| "learning_rate": 9.788964937598689e-05, | |
| "loss": 0.02, | |
| "step": 156 | |
| }, | |
| { | |
| "epoch": 9.294117647058824, | |
| "grad_norm": 0.7616733908653259, | |
| "learning_rate": 9.776998593610428e-05, | |
| "loss": 0.0173, | |
| "step": 158 | |
| }, | |
| { | |
| "epoch": 9.411764705882353, | |
| "grad_norm": 0.997645378112793, | |
| "learning_rate": 9.764710002135784e-05, | |
| "loss": 0.023, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 9.529411764705882, | |
| "grad_norm": 0.733609139919281, | |
| "learning_rate": 9.752099992140399e-05, | |
| "loss": 0.0182, | |
| "step": 162 | |
| }, | |
| { | |
| "epoch": 9.647058823529411, | |
| "grad_norm": 0.9087010025978088, | |
| "learning_rate": 9.739169414272217e-05, | |
| "loss": 0.0197, | |
| "step": 164 | |
| }, | |
| { | |
| "epoch": 9.764705882352942, | |
| "grad_norm": 0.785037636756897, | |
| "learning_rate": 9.725919140804099e-05, | |
| "loss": 0.023, | |
| "step": 166 | |
| }, | |
| { | |
| "epoch": 9.882352941176471, | |
| "grad_norm": 0.8591390252113342, | |
| "learning_rate": 9.71235006557497e-05, | |
| "loss": 0.0234, | |
| "step": 168 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "grad_norm": 0.8396819829940796, | |
| "learning_rate": 9.698463103929542e-05, | |
| "loss": 0.0246, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_loss": 0.8533556461334229, | |
| "eval_runtime": 46.8108, | |
| "eval_samples_per_second": 0.769, | |
| "eval_steps_per_second": 0.769, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "step": 170, | |
| "total_flos": 1.6636987284075315e+17, | |
| "train_loss": 0.26086401211864807, | |
| "train_runtime": 6014.7432, | |
| "train_samples_per_second": 1.131, | |
| "train_steps_per_second": 0.141 | |
| } | |
| ], | |
| "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": 1.6636987284075315e+17, | |
| "train_batch_size": 1, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |