Text-to-Image
Diffusers
Safetensors
UFC / MVRD /d28_lambda0.5_fullft_aug2 /trainer_state.json
ntk1507's picture
Upload folder using huggingface_hub
dd355d3 verified
Raw History Blame Contribute Delete
428 kB
{ "best_metric": null, "best_model_checkpoint": null, "epoch": 0.9997688573850065, "eval_steps": 500, "global_step": 1622, "is_hyper_param_search": false, "is_local_process_zero": true, "is_world_process_zero": true, "log_history": [ { "distill_loss": 0.9078, "epoch": 0.0, "grad_norm": 19.151357114127673, "learning_rate": 2.0408163265306121e-07, "loss": 0.7418, "lr_vision_tower": 4.081632653061224e-08, "sft_loss": 0.288, "step": 1 }, { "distill_loss": 0.9047, "epoch": 0.0, "grad_norm": 19.707155046403212, "learning_rate": 4.0816326530612243e-07, "loss": 0.7432, "lr_vision_tower": 8.163265306122448e-08, "sft_loss": 0.2908, "step": 2 }, { "distill_loss": 0.9044, "epoch": 0.0, "grad_norm": 20.608160938114064, "learning_rate": 6.122448979591837e-07, "loss": 0.7153, "lr_vision_tower": 1.2244897959183673e-07, "sft_loss": 0.2631, "step": 3 }, { "distill_loss": 0.9068, "epoch": 0.0, "grad_norm": 19.97875237942443, "learning_rate": 8.163265306122449e-07, "loss": 0.7139, "lr_vision_tower": 1.6326530612244896e-07, "sft_loss": 0.2605, "step": 4 }, { "distill_loss": 0.9045, "epoch": 0.0, "grad_norm": 16.31548300275858, "learning_rate": 1.0204081632653063e-06, "loss": 0.7107, "lr_vision_tower": 2.0408163265306121e-07, "sft_loss": 0.2585, "step": 5 }, { "distill_loss": 0.9095, "epoch": 0.0, "grad_norm": 12.295941310553859, "learning_rate": 1.2244897959183673e-06, "loss": 0.6678, "lr_vision_tower": 2.4489795918367347e-07, "sft_loss": 0.213, "step": 6 }, { "distill_loss": 0.9057, "epoch": 0.0, "grad_norm": 9.94940809093829, "learning_rate": 1.4285714285714286e-06, "loss": 0.6525, "lr_vision_tower": 2.857142857142857e-07, "sft_loss": 0.1997, "step": 7 }, { "distill_loss": 0.9034, "epoch": 0.0, "grad_norm": 2.891667866636629, "learning_rate": 1.6326530612244897e-06, "loss": 0.6037, "lr_vision_tower": 3.265306122448979e-07, "sft_loss": 0.152, "step": 8 }, { "distill_loss": 0.9043, "epoch": 0.01, "grad_norm": 2.6203634446256605, "learning_rate": 1.8367346938775512e-06, "loss": 0.6121, "lr_vision_tower": 3.673469387755102e-07, "sft_loss": 0.1599, "step": 9 }, { "distill_loss": 0.8996, "epoch": 0.01, "grad_norm": 3.0220581830416733, "learning_rate": 2.0408163265306125e-06, "loss": 0.5862, "lr_vision_tower": 4.0816326530612243e-07, "sft_loss": 0.1364, "step": 10 }, { "distill_loss": 0.8836, "epoch": 0.01, "grad_norm": 2.018651442543094, "learning_rate": 2.244897959183674e-06, "loss": 0.5854, "lr_vision_tower": 4.4897959183673465e-07, "sft_loss": 0.1436, "step": 11 }, { "distill_loss": 0.8837, "epoch": 0.01, "grad_norm": 2.204320448720915, "learning_rate": 2.4489795918367347e-06, "loss": 0.6, "lr_vision_tower": 4.897959183673469e-07, "sft_loss": 0.1582, "step": 12 }, { "distill_loss": 0.8727, "epoch": 0.01, "grad_norm": 2.044927306765267, "learning_rate": 2.6530612244897964e-06, "loss": 0.5762, "lr_vision_tower": 5.306122448979592e-07, "sft_loss": 0.1398, "step": 13 }, { "distill_loss": 0.8489, "epoch": 0.01, "grad_norm": 4.576715365757819, "learning_rate": 2.8571428571428573e-06, "loss": 0.5588, "lr_vision_tower": 5.714285714285714e-07, "sft_loss": 0.1343, "step": 14 }, { "distill_loss": 0.8121, "epoch": 0.01, "grad_norm": 2.0086693757456673, "learning_rate": 3.0612244897959185e-06, "loss": 0.5277, "lr_vision_tower": 6.122448979591837e-07, "sft_loss": 0.1216, "step": 15 }, { "distill_loss": 0.7888, "epoch": 0.01, "grad_norm": 1.594693659222105, "learning_rate": 3.2653061224489794e-06, "loss": 0.526, "lr_vision_tower": 6.530612244897958e-07, "sft_loss": 0.1316, "step": 16 }, { "distill_loss": 0.7698, "epoch": 0.01, "grad_norm": 1.4803383886716255, "learning_rate": 3.469387755102041e-06, "loss": 0.5065, "lr_vision_tower": 6.938775510204081e-07, "sft_loss": 0.1216, "step": 17 }, { "distill_loss": 0.7441, "epoch": 0.01, "grad_norm": 1.6555187466518815, "learning_rate": 3.6734693877551024e-06, "loss": 0.4973, "lr_vision_tower": 7.346938775510204e-07, "sft_loss": 0.1253, "step": 18 }, { "distill_loss": 0.7029, "epoch": 0.01, "grad_norm": 1.380837604088627, "learning_rate": 3.877551020408164e-06, "loss": 0.4773, "lr_vision_tower": 7.755102040816326e-07, "sft_loss": 0.1259, "step": 19 }, { "distill_loss": 0.6745, "epoch": 0.01, "grad_norm": 1.3448871330668835, "learning_rate": 4.081632653061225e-06, "loss": 0.46, "lr_vision_tower": 8.163265306122449e-07, "sft_loss": 0.1227, "step": 20 }, { "distill_loss": 0.6355, "epoch": 0.01, "grad_norm": 1.5034831106147426, "learning_rate": 4.2857142857142855e-06, "loss": 0.4428, "lr_vision_tower": 8.57142857142857e-07, "sft_loss": 0.1251, "step": 21 }, { "distill_loss": 0.5997, "epoch": 0.01, "grad_norm": 1.7803229479100442, "learning_rate": 4.489795918367348e-06, "loss": 0.4373, "lr_vision_tower": 8.979591836734693e-07, "sft_loss": 0.1374, "step": 22 }, { "distill_loss": 0.5617, "epoch": 0.01, "grad_norm": 1.9828469399205833, "learning_rate": 4.693877551020409e-06, "loss": 0.4036, "lr_vision_tower": 9.387755102040816e-07, "sft_loss": 0.1228, "step": 23 }, { "distill_loss": 0.5312, "epoch": 0.01, "grad_norm": 2.0513424059438004, "learning_rate": 4.897959183673469e-06, "loss": 0.4232, "lr_vision_tower": 9.795918367346939e-07, "sft_loss": 0.1576, "step": 24 }, { "distill_loss": 0.5086, "epoch": 0.02, "grad_norm": 1.1758670115994851, "learning_rate": 5.1020408163265315e-06, "loss": 0.3845, "lr_vision_tower": 1.020408163265306e-06, "sft_loss": 0.1302, "step": 25 }, { "distill_loss": 0.4904, "epoch": 0.02, "grad_norm": 1.3014176079955393, "learning_rate": 5.306122448979593e-06, "loss": 0.3915, "lr_vision_tower": 1.0612244897959184e-06, "sft_loss": 0.1463, "step": 26 }, { "distill_loss": 0.4773, "epoch": 0.02, "grad_norm": 1.4023276155472189, "learning_rate": 5.510204081632653e-06, "loss": 0.3732, "lr_vision_tower": 1.1020408163265304e-06, "sft_loss": 0.1346, "step": 27 }, { "distill_loss": 0.4518, "epoch": 0.02, "grad_norm": 1.39906481652517, "learning_rate": 5.7142857142857145e-06, "loss": 0.3581, "lr_vision_tower": 1.1428571428571428e-06, "sft_loss": 0.1322, "step": 28 }, { "distill_loss": 0.4437, "epoch": 0.02, "grad_norm": 1.9050297467056014, "learning_rate": 5.918367346938776e-06, "loss": 0.3372, "lr_vision_tower": 1.183673469387755e-06, "sft_loss": 0.1154, "step": 29 }, { "distill_loss": 0.4473, "epoch": 0.02, "grad_norm": 1.34873028864968, "learning_rate": 6.122448979591837e-06, "loss": 0.3453, "lr_vision_tower": 1.2244897959183673e-06, "sft_loss": 0.1217, "step": 30 }, { "distill_loss": 0.4396, "epoch": 0.02, "grad_norm": 1.2878819379308453, "learning_rate": 6.326530612244899e-06, "loss": 0.3671, "lr_vision_tower": 1.2653061224489795e-06, "sft_loss": 0.1473, "step": 31 }, { "distill_loss": 0.4219, "epoch": 0.02, "grad_norm": 1.193797202326762, "learning_rate": 6.530612244897959e-06, "loss": 0.3511, "lr_vision_tower": 1.3061224489795917e-06, "sft_loss": 0.1402, "step": 32 }, { "distill_loss": 0.4275, "epoch": 0.02, "grad_norm": 1.0853146846559734, "learning_rate": 6.734693877551021e-06, "loss": 0.3382, "lr_vision_tower": 1.3469387755102039e-06, "sft_loss": 0.1244, "step": 33 }, { "distill_loss": 0.4126, "epoch": 0.02, "grad_norm": 1.0032808603692989, "learning_rate": 6.938775510204082e-06, "loss": 0.3307, "lr_vision_tower": 1.3877551020408162e-06, "sft_loss": 0.1244, "step": 34 }, { "distill_loss": 0.4054, "epoch": 0.02, "grad_norm": 0.9744281742949256, "learning_rate": 7.1428571428571436e-06, "loss": 0.32, "lr_vision_tower": 1.4285714285714286e-06, "sft_loss": 0.1173, "step": 35 }, { "distill_loss": 0.4001, "epoch": 0.02, "grad_norm": 1.0224198036134944, "learning_rate": 7.346938775510205e-06, "loss": 0.3234, "lr_vision_tower": 1.4693877551020408e-06, "sft_loss": 0.1234, "step": 36 }, { "distill_loss": 0.4079, "epoch": 0.02, "grad_norm": 0.9971903906866931, "learning_rate": 7.551020408163265e-06, "loss": 0.3343, "lr_vision_tower": 1.510204081632653e-06, "sft_loss": 0.1303, "step": 37 }, { "distill_loss": 0.3918, "epoch": 0.02, "grad_norm": 0.976945819954002, "learning_rate": 7.755102040816327e-06, "loss": 0.308, "lr_vision_tower": 1.5510204081632651e-06, "sft_loss": 0.1121, "step": 38 }, { "distill_loss": 0.392, "epoch": 0.02, "grad_norm": 1.0876164119092064, "learning_rate": 7.959183673469388e-06, "loss": 0.3251, "lr_vision_tower": 1.5918367346938775e-06, "sft_loss": 0.1291, "step": 39 }, { "distill_loss": 0.3845, "epoch": 0.02, "grad_norm": 1.0898360945799417, "learning_rate": 8.16326530612245e-06, "loss": 0.3251, "lr_vision_tower": 1.6326530612244897e-06, "sft_loss": 0.1328, "step": 40 }, { "distill_loss": 0.3784, "epoch": 0.03, "grad_norm": 1.387150994565522, "learning_rate": 8.36734693877551e-06, "loss": 0.3088, "lr_vision_tower": 1.673469387755102e-06, "sft_loss": 0.1196, "step": 41 }, { "distill_loss": 0.3775, "epoch": 0.03, "grad_norm": 0.9595956313660746, "learning_rate": 8.571428571428571e-06, "loss": 0.3126, "lr_vision_tower": 1.714285714285714e-06, "sft_loss": 0.1239, "step": 42 }, { "distill_loss": 0.3792, "epoch": 0.03, "grad_norm": 1.123019975858785, "learning_rate": 8.775510204081633e-06, "loss": 0.3128, "lr_vision_tower": 1.7551020408163264e-06, "sft_loss": 0.1232, "step": 43 }, { "distill_loss": 0.3765, "epoch": 0.03, "grad_norm": 0.8777178872632262, "learning_rate": 8.979591836734695e-06, "loss": 0.3105, "lr_vision_tower": 1.7959183673469386e-06, "sft_loss": 0.1223, "step": 44 }, { "distill_loss": 0.3725, "epoch": 0.03, "grad_norm": 0.7367930302307721, "learning_rate": 9.183673469387756e-06, "loss": 0.2968, "lr_vision_tower": 1.836734693877551e-06, "sft_loss": 0.1106, "step": 45 }, { "distill_loss": 0.3662, "epoch": 0.03, "grad_norm": 1.0951187537363045, "learning_rate": 9.387755102040818e-06, "loss": 0.3118, "lr_vision_tower": 1.8775510204081632e-06, "sft_loss": 0.1287, "step": 46 }, { "distill_loss": 0.3583, "epoch": 0.03, "grad_norm": 1.0365184720866656, "learning_rate": 9.591836734693878e-06, "loss": 0.3001, "lr_vision_tower": 1.918367346938775e-06, "sft_loss": 0.121, "step": 47 }, { "distill_loss": 0.3702, "epoch": 0.03, "grad_norm": 1.0942603751965376, "learning_rate": 9.795918367346939e-06, "loss": 0.2866, "lr_vision_tower": 1.9591836734693877e-06, "sft_loss": 0.1015, "step": 48 }, { "distill_loss": 0.3576, "epoch": 0.03, "grad_norm": 1.1643001963688415, "learning_rate": 1e-05, "loss": 0.3048, "lr_vision_tower": 2e-06, "sft_loss": 0.126, "step": 49 }, { "distill_loss": 0.3526, "epoch": 0.03, "grad_norm": 0.9863003374740015, "learning_rate": 9.99999002800242e-06, "loss": 0.3254, "lr_vision_tower": 1.999998005600484e-06, "sft_loss": 0.1491, "step": 50 }, { "distill_loss": 0.3582, "epoch": 0.03, "grad_norm": 1.1563898237381633, "learning_rate": 9.999960112049457e-06, "loss": 0.3011, "lr_vision_tower": 1.999992022409891e-06, "sft_loss": 0.122, "step": 51 }, { "distill_loss": 0.3477, "epoch": 0.03, "grad_norm": 1.0091953779960545, "learning_rate": 9.999910252260435e-06, "loss": 0.2833, "lr_vision_tower": 1.9999820504520867e-06, "sft_loss": 0.1094, "step": 52 }, { "distill_loss": 0.3456, "epoch": 0.03, "grad_norm": 1.1917906980939477, "learning_rate": 9.999840448834239e-06, "loss": 0.306, "lr_vision_tower": 1.999968089766848e-06, "sft_loss": 0.1331, "step": 53 }, { "distill_loss": 0.3523, "epoch": 0.03, "grad_norm": 1.2406358809364415, "learning_rate": 9.9997507020493e-06, "loss": 0.3072, "lr_vision_tower": 1.9999501404098596e-06, "sft_loss": 0.131, "step": 54 }, { "distill_loss": 0.349, "epoch": 0.03, "grad_norm": 1.1840893804151844, "learning_rate": 9.9996410122636e-06, "loss": 0.3164, "lr_vision_tower": 1.9999282024527198e-06, "sft_loss": 0.1419, "step": 55 }, { "distill_loss": 0.3385, "epoch": 0.03, "grad_norm": 0.854048351041351, "learning_rate": 9.999511379914668e-06, "loss": 0.2829, "lr_vision_tower": 1.9999022759829335e-06, "sft_loss": 0.1137, "step": 56 }, { "distill_loss": 0.3341, "epoch": 0.04, "grad_norm": 0.8456273498847441, "learning_rate": 9.999361805519584e-06, "loss": 0.3009, "lr_vision_tower": 1.9998723611039167e-06, "sft_loss": 0.1339, "step": 57 }, { "distill_loss": 0.342, "epoch": 0.04, "grad_norm": 1.0616180083065805, "learning_rate": 9.999192289674969e-06, "loss": 0.2717, "lr_vision_tower": 1.9998384579349935e-06, "sft_loss": 0.1007, "step": 58 }, { "distill_loss": 0.3372, "epoch": 0.04, "grad_norm": 1.0607845807298224, "learning_rate": 9.999002833056986e-06, "loss": 0.2833, "lr_vision_tower": 1.999800566611397e-06, "sft_loss": 0.1147, "step": 59 }, { "distill_loss": 0.3465, "epoch": 0.04, "grad_norm": 1.3620629826067268, "learning_rate": 9.998793436421342e-06, "loss": 0.3119, "lr_vision_tower": 1.999758687284268e-06, "sft_loss": 0.1386, "step": 60 }, { "distill_loss": 0.3359, "epoch": 0.04, "grad_norm": 1.2863664729656519, "learning_rate": 9.998564100603276e-06, "loss": 0.306, "lr_vision_tower": 1.999712820120655e-06, "sft_loss": 0.138, "step": 61 }, { "distill_loss": 0.3384, "epoch": 0.04, "grad_norm": 1.3157503102619823, "learning_rate": 9.998314826517564e-06, "loss": 0.2855, "lr_vision_tower": 1.9996629653035126e-06, "sft_loss": 0.1163, "step": 62 }, { "distill_loss": 0.3149, "epoch": 0.04, "grad_norm": 1.0894162884304166, "learning_rate": 9.998045615158508e-06, "loss": 0.2694, "lr_vision_tower": 1.9996091230317015e-06, "sft_loss": 0.1119, "step": 63 }, { "distill_loss": 0.3294, "epoch": 0.04, "grad_norm": 1.1517441808207667, "learning_rate": 9.99775646759994e-06, "loss": 0.2838, "lr_vision_tower": 1.9995512935199877e-06, "sft_loss": 0.1192, "step": 64 }, { "distill_loss": 0.3195, "epoch": 0.04, "grad_norm": 1.0197202878419946, "learning_rate": 9.997447384995213e-06, "loss": 0.2768, "lr_vision_tower": 1.9994894769990424e-06, "sft_loss": 0.1171, "step": 65 }, { "distill_loss": 0.3207, "epoch": 0.04, "grad_norm": 0.9184925495124394, "learning_rate": 9.997118368577192e-06, "loss": 0.2823, "lr_vision_tower": 1.999423673715438e-06, "sft_loss": 0.1219, "step": 66 }, { "distill_loss": 0.3178, "epoch": 0.04, "grad_norm": 0.6802451707478088, "learning_rate": 9.996769419658258e-06, "loss": 0.2746, "lr_vision_tower": 1.9993538839316515e-06, "sft_loss": 0.1157, "step": 67 }, { "distill_loss": 0.3352, "epoch": 0.04, "grad_norm": 1.1760536355455045, "learning_rate": 9.996400539630301e-06, "loss": 0.2951, "lr_vision_tower": 1.99928010792606e-06, "sft_loss": 0.1275, "step": 68 }, { "distill_loss": 0.3094, "epoch": 0.04, "grad_norm": 1.151878828695408, "learning_rate": 9.996011729964707e-06, "loss": 0.2875, "lr_vision_tower": 1.999202345992941e-06, "sft_loss": 0.1327, "step": 69 }, { "distill_loss": 0.3024, "epoch": 0.04, "grad_norm": 0.8974306560088156, "learning_rate": 9.995602992212358e-06, "loss": 0.2664, "lr_vision_tower": 1.9991205984424714e-06, "sft_loss": 0.1152, "step": 70 }, { "distill_loss": 0.3241, "epoch": 0.04, "grad_norm": 1.176500878508174, "learning_rate": 9.99517432800363e-06, "loss": 0.3076, "lr_vision_tower": 1.999034865600726e-06, "sft_loss": 0.1455, "step": 71 }, { "distill_loss": 0.3318, "epoch": 0.04, "grad_norm": 1.390161464832423, "learning_rate": 9.994725739048377e-06, "loss": 0.2756, "lr_vision_tower": 1.9989451478096754e-06, "sft_loss": 0.1097, "step": 72 }, { "distill_loss": 0.3149, "epoch": 0.04, "grad_norm": 1.0577066788681024, "learning_rate": 9.994257227135931e-06, "loss": 0.2854, "lr_vision_tower": 1.998851445427186e-06, "sft_loss": 0.128, "step": 73 }, { "distill_loss": 0.3095, "epoch": 0.05, "grad_norm": 0.7659007945249396, "learning_rate": 9.99376879413509e-06, "loss": 0.2686, "lr_vision_tower": 1.998753758827018e-06, "sft_loss": 0.1139, "step": 74 }, { "distill_loss": 0.3156, "epoch": 0.05, "grad_norm": 0.9318988731578445, "learning_rate": 9.993260441994116e-06, "loss": 0.2751, "lr_vision_tower": 1.9986520883988228e-06, "sft_loss": 0.1173, "step": 75 }, { "distill_loss": 0.3043, "epoch": 0.05, "grad_norm": 1.1201587573307132, "learning_rate": 9.992732172740724e-06, "loss": 0.272, "lr_vision_tower": 1.9985464345481448e-06, "sft_loss": 0.1199, "step": 76 }, { "distill_loss": 0.3134, "epoch": 0.05, "grad_norm": 1.1019311553132318, "learning_rate": 9.992183988482073e-06, "loss": 0.2589, "lr_vision_tower": 1.9984367976964146e-06, "sft_loss": 0.1022, "step": 77 }, { "distill_loss": 0.3113, "epoch": 0.05, "grad_norm": 1.143015437149227, "learning_rate": 9.991615891404762e-06, "loss": 0.271, "lr_vision_tower": 1.9983231782809524e-06, "sft_loss": 0.1153, "step": 78 }, { "distill_loss": 0.3068, "epoch": 0.05, "grad_norm": 0.9804158513548595, "learning_rate": 9.991027883774813e-06, "loss": 0.2735, "lr_vision_tower": 1.9982055767549623e-06, "sft_loss": 0.1201, "step": 79 }, { "distill_loss": 0.3063, "epoch": 0.05, "grad_norm": 1.1092704526428467, "learning_rate": 9.990419967937672e-06, "loss": 0.269, "lr_vision_tower": 1.998083993587534e-06, "sft_loss": 0.1158, "step": 80 }, { "distill_loss": 0.3085, "epoch": 0.05, "grad_norm": 1.0915401032901195, "learning_rate": 9.989792146318193e-06, "loss": 0.2624, "lr_vision_tower": 1.9979584292636386e-06, "sft_loss": 0.1081, "step": 81 }, { "distill_loss": 0.3062, "epoch": 0.05, "grad_norm": 1.0245285766174397, "learning_rate": 9.98914442142063e-06, "loss": 0.2815, "lr_vision_tower": 1.9978288842841257e-06, "sft_loss": 0.1284, "step": 82 }, { "distill_loss": 0.2992, "epoch": 0.05, "grad_norm": 0.8666155573677307, "learning_rate": 9.988476795828629e-06, "loss": 0.2677, "lr_vision_tower": 1.997695359165725e-06, "sft_loss": 0.1181, "step": 83 }, { "distill_loss": 0.3042, "epoch": 0.05, "grad_norm": 0.8598774491066993, "learning_rate": 9.987789272205211e-06, "loss": 0.2632, "lr_vision_tower": 1.997557854441042e-06, "sft_loss": 0.1111, "step": 84 }, { "distill_loss": 0.2978, "epoch": 0.05, "grad_norm": 1.1537200228522726, "learning_rate": 9.98708185329277e-06, "loss": 0.2754, "lr_vision_tower": 1.997416370658554e-06, "sft_loss": 0.1265, "step": 85 }, { "distill_loss": 0.3142, "epoch": 0.05, "grad_norm": 1.1487421538050326, "learning_rate": 9.986354541913063e-06, "loss": 0.2841, "lr_vision_tower": 1.9972709083826122e-06, "sft_loss": 0.127, "step": 86 }, { "distill_loss": 0.2953, "epoch": 0.05, "grad_norm": 1.0489206525812866, "learning_rate": 9.985607340967183e-06, "loss": 0.2841, "lr_vision_tower": 1.9971214681934363e-06, "sft_loss": 0.1364, "step": 87 }, { "distill_loss": 0.2996, "epoch": 0.05, "grad_norm": 0.8958937385400997, "learning_rate": 9.984840253435569e-06, "loss": 0.2619, "lr_vision_tower": 1.9969680506871136e-06, "sft_loss": 0.1121, "step": 88 }, { "distill_loss": 0.2952, "epoch": 0.05, "grad_norm": 1.0334457541004252, "learning_rate": 9.984053282377974e-06, "loss": 0.2651, "lr_vision_tower": 1.9968106564755944e-06, "sft_loss": 0.1175, "step": 89 }, { "distill_loss": 0.3, "epoch": 0.06, "grad_norm": 0.9805125114909647, "learning_rate": 9.98324643093347e-06, "loss": 0.2895, "lr_vision_tower": 1.996649286186694e-06, "sft_loss": 0.1395, "step": 90 }, { "distill_loss": 0.2918, "epoch": 0.06, "grad_norm": 0.8006713933214914, "learning_rate": 9.982419702320426e-06, "loss": 0.2679, "lr_vision_tower": 1.996483940464085e-06, "sft_loss": 0.122, "step": 91 }, { "distill_loss": 0.3054, "epoch": 0.06, "grad_norm": 0.897088920356915, "learning_rate": 9.981573099836496e-06, "loss": 0.273, "lr_vision_tower": 1.996314619967299e-06, "sft_loss": 0.1203, "step": 92 }, { "distill_loss": 0.2882, "epoch": 0.06, "grad_norm": 0.8541847862850451, "learning_rate": 9.980706626858607e-06, "loss": 0.2463, "lr_vision_tower": 1.9961413253717214e-06, "sft_loss": 0.1022, "step": 93 }, { "distill_loss": 0.3032, "epoch": 0.06, "grad_norm": 1.1090918315052938, "learning_rate": 9.979820286842946e-06, "loss": 0.2703, "lr_vision_tower": 1.995964057368589e-06, "sft_loss": 0.1188, "step": 94 }, { "distill_loss": 0.2989, "epoch": 0.06, "grad_norm": 1.2477794181802067, "learning_rate": 9.978914083324942e-06, "loss": 0.2645, "lr_vision_tower": 1.995782816664988e-06, "sft_loss": 0.115, "step": 95 }, { "distill_loss": 0.2897, "epoch": 0.06, "grad_norm": 1.1413379898225342, "learning_rate": 9.977988019919263e-06, "loss": 0.2507, "lr_vision_tower": 1.9955976039838524e-06, "sft_loss": 0.1059, "step": 96 }, { "distill_loss": 0.3012, "epoch": 0.06, "grad_norm": 1.4389783833701695, "learning_rate": 9.977042100319786e-06, "loss": 0.2666, "lr_vision_tower": 1.995408420063957e-06, "sft_loss": 0.116, "step": 97 }, { "distill_loss": 0.2973, "epoch": 0.06, "grad_norm": 1.0699995580704487, "learning_rate": 9.9760763282996e-06, "loss": 0.2636, "lr_vision_tower": 1.9952152656599197e-06, "sft_loss": 0.115, "step": 98 }, { "distill_loss": 0.2896, "epoch": 0.06, "grad_norm": 1.4429871527376894, "learning_rate": 9.975090707710969e-06, "loss": 0.2567, "lr_vision_tower": 1.9950181415421934e-06, "sft_loss": 0.1119, "step": 99 }, { "distill_loss": 0.3043, "epoch": 0.06, "grad_norm": 1.1734511639678977, "learning_rate": 9.974085242485338e-06, "loss": 0.2713, "lr_vision_tower": 1.9948170484970673e-06, "sft_loss": 0.1192, "step": 100 }, { "distill_loss": 0.2946, "epoch": 0.06, "grad_norm": 0.958692521176026, "learning_rate": 9.973059936633308e-06, "loss": 0.2633, "lr_vision_tower": 1.994611987326661e-06, "sft_loss": 0.116, "step": 101 }, { "distill_loss": 0.2872, "epoch": 0.06, "grad_norm": 1.1182507270614213, "learning_rate": 9.972014794244614e-06, "loss": 0.2782, "lr_vision_tower": 1.9944029588489225e-06, "sft_loss": 0.1346, "step": 102 }, { "distill_loss": 0.2995, "epoch": 0.06, "grad_norm": 1.0710408675332443, "learning_rate": 9.970949819488122e-06, "loss": 0.2863, "lr_vision_tower": 1.994189963897624e-06, "sft_loss": 0.1366, "step": 103 }, { "distill_loss": 0.2973, "epoch": 0.06, "grad_norm": 0.9524851608567982, "learning_rate": 9.9698650166118e-06, "loss": 0.2779, "lr_vision_tower": 1.99397300332236e-06, "sft_loss": 0.1293, "step": 104 }, { "distill_loss": 0.2988, "epoch": 0.06, "grad_norm": 1.0417713806006288, "learning_rate": 9.96876038994271e-06, "loss": 0.2668, "lr_vision_tower": 1.993752077988542e-06, "sft_loss": 0.1174, "step": 105 }, { "distill_loss": 0.277, "epoch": 0.07, "grad_norm": 1.0072372126085245, "learning_rate": 9.967635943886987e-06, "loss": 0.2719, "lr_vision_tower": 1.993527188777397e-06, "sft_loss": 0.1334, "step": 106 }, { "distill_loss": 0.2769, "epoch": 0.07, "grad_norm": 0.7675538863091715, "learning_rate": 9.966491682929818e-06, "loss": 0.2562, "lr_vision_tower": 1.9932983365859634e-06, "sft_loss": 0.1178, "step": 107 }, { "distill_loss": 0.2767, "epoch": 0.07, "grad_norm": 0.6793617812813391, "learning_rate": 9.965327611635432e-06, "loss": 0.2566, "lr_vision_tower": 1.993065522327086e-06, "sft_loss": 0.1182, "step": 108 }, { "distill_loss": 0.2798, "epoch": 0.07, "grad_norm": 1.1815935919230571, "learning_rate": 9.964143734647072e-06, "loss": 0.2632, "lr_vision_tower": 1.992828746929414e-06, "sft_loss": 0.1233, "step": 109 }, { "distill_loss": 0.2788, "epoch": 0.07, "grad_norm": 1.1583178241312602, "learning_rate": 9.962940056686989e-06, "loss": 0.2669, "lr_vision_tower": 1.992588011337398e-06, "sft_loss": 0.1275, "step": 110 }, { "distill_loss": 0.2864, "epoch": 0.07, "grad_norm": 0.7620815047794848, "learning_rate": 9.961716582556412e-06, "loss": 0.2503, "lr_vision_tower": 1.992343316511282e-06, "sft_loss": 0.1071, "step": 111 }, { "distill_loss": 0.2865, "epoch": 0.07, "grad_norm": 0.8184663165266552, "learning_rate": 9.960473317135532e-06, "loss": 0.2754, "lr_vision_tower": 1.9920946634271064e-06, "sft_loss": 0.1321, "step": 112 }, { "distill_loss": 0.2784, "epoch": 0.07, "grad_norm": 0.8148012139746905, "learning_rate": 9.959210265383486e-06, "loss": 0.2579, "lr_vision_tower": 1.991842053076697e-06, "sft_loss": 0.1186, "step": 113 }, { "distill_loss": 0.2858, "epoch": 0.07, "grad_norm": 0.944148376939854, "learning_rate": 9.957927432338332e-06, "loss": 0.2707, "lr_vision_tower": 1.991585486467666e-06, "sft_loss": 0.1278, "step": 114 }, { "distill_loss": 0.2875, "epoch": 0.07, "grad_norm": 1.2600994490009574, "learning_rate": 9.956624823117036e-06, "loss": 0.2648, "lr_vision_tower": 1.9913249646234067e-06, "sft_loss": 0.121, "step": 115 }, { "distill_loss": 0.269, "epoch": 0.07, "grad_norm": 0.832597195673952, "learning_rate": 9.955302442915442e-06, "loss": 0.2522, "lr_vision_tower": 1.9910604885830883e-06, "sft_loss": 0.1177, "step": 116 }, { "distill_loss": 0.2814, "epoch": 0.07, "grad_norm": 0.9979153803297443, "learning_rate": 9.95396029700826e-06, "loss": 0.2501, "lr_vision_tower": 1.9907920594016518e-06, "sft_loss": 0.1094, "step": 117 }, { "distill_loss": 0.2798, "epoch": 0.07, "grad_norm": 0.7933019382204061, "learning_rate": 9.95259839074904e-06, "loss": 0.2582, "lr_vision_tower": 1.9905196781498074e-06, "sft_loss": 0.1183, "step": 118 }, { "distill_loss": 0.2763, "epoch": 0.07, "grad_norm": 0.8292032968192878, "learning_rate": 9.95121672957015e-06, "loss": 0.2577, "lr_vision_tower": 1.99024334591403e-06, "sft_loss": 0.1196, "step": 119 }, { "distill_loss": 0.2733, "epoch": 0.07, "grad_norm": 0.8842637326179644, "learning_rate": 9.949815318982764e-06, "loss": 0.2542, "lr_vision_tower": 1.9899630637965527e-06, "sft_loss": 0.1176, "step": 120 }, { "distill_loss": 0.2663, "epoch": 0.07, "grad_norm": 0.8273321928074939, "learning_rate": 9.948394164576823e-06, "loss": 0.2561, "lr_vision_tower": 1.9896788329153644e-06, "sft_loss": 0.1229, "step": 121 }, { "distill_loss": 0.2857, "epoch": 0.08, "grad_norm": 1.2674841345678138, "learning_rate": 9.946953272021029e-06, "loss": 0.2777, "lr_vision_tower": 1.9893906544042055e-06, "sft_loss": 0.1348, "step": 122 }, { "distill_loss": 0.2845, "epoch": 0.08, "grad_norm": 0.9082746538675867, "learning_rate": 9.945492647062808e-06, "loss": 0.26, "lr_vision_tower": 1.9890985294125616e-06, "sft_loss": 0.1177, "step": 123 }, { "distill_loss": 0.288, "epoch": 0.08, "grad_norm": 1.4091215255452085, "learning_rate": 9.944012295528307e-06, "loss": 0.2702, "lr_vision_tower": 1.988802459105661e-06, "sft_loss": 0.1262, "step": 124 }, { "distill_loss": 0.2801, "epoch": 0.08, "grad_norm": 1.0844401207476833, "learning_rate": 9.942512223322345e-06, "loss": 0.2554, "lr_vision_tower": 1.9885024446644686e-06, "sft_loss": 0.1153, "step": 125 }, { "distill_loss": 0.2834, "epoch": 0.08, "grad_norm": 0.9858483374601242, "learning_rate": 9.94099243642841e-06, "loss": 0.2654, "lr_vision_tower": 1.9881984872856818e-06, "sft_loss": 0.1237, "step": 126 }, { "distill_loss": 0.2739, "epoch": 0.08, "grad_norm": 1.0408702734967321, "learning_rate": 9.939452940908627e-06, "loss": 0.2482, "lr_vision_tower": 1.987890588181725e-06, "sft_loss": 0.1112, "step": 127 }, { "distill_loss": 0.2822, "epoch": 0.08, "grad_norm": 0.9611174796937854, "learning_rate": 9.937893742903732e-06, "loss": 0.2489, "lr_vision_tower": 1.9875787485807464e-06, "sft_loss": 0.1078, "step": 128 }, { "distill_loss": 0.2791, "epoch": 0.08, "grad_norm": 0.8907992854341656, "learning_rate": 9.936314848633057e-06, "loss": 0.2636, "lr_vision_tower": 1.9872629697266114e-06, "sft_loss": 0.1241, "step": 129 }, { "distill_loss": 0.2817, "epoch": 0.08, "grad_norm": 0.8818980860707651, "learning_rate": 9.93471626439449e-06, "loss": 0.2442, "lr_vision_tower": 1.9869432528788977e-06, "sft_loss": 0.1034, "step": 130 }, { "distill_loss": 0.272, "epoch": 0.08, "grad_norm": 0.7040818823522508, "learning_rate": 9.933097996564464e-06, "loss": 0.2441, "lr_vision_tower": 1.9866195993128925e-06, "sft_loss": 0.1081, "step": 131 }, { "distill_loss": 0.268, "epoch": 0.08, "grad_norm": 0.9982553294620407, "learning_rate": 9.931460051597925e-06, "loss": 0.249, "lr_vision_tower": 1.9862920103195847e-06, "sft_loss": 0.115, "step": 132 }, { "distill_loss": 0.276, "epoch": 0.08, "grad_norm": 0.8100516079821656, "learning_rate": 9.929802436028304e-06, "loss": 0.2563, "lr_vision_tower": 1.9859604872056605e-06, "sft_loss": 0.1183, "step": 133 }, { "distill_loss": 0.2746, "epoch": 0.08, "grad_norm": 0.8654812260827319, "learning_rate": 9.928125156467496e-06, "loss": 0.2528, "lr_vision_tower": 1.985625031293499e-06, "sft_loss": 0.1154, "step": 134 }, { "distill_loss": 0.2819, "epoch": 0.08, "grad_norm": 0.6933207147287941, "learning_rate": 9.926428219605836e-06, "loss": 0.255, "lr_vision_tower": 1.985285643921167e-06, "sft_loss": 0.1141, "step": 135 }, { "distill_loss": 0.2678, "epoch": 0.08, "grad_norm": 0.7994970477934517, "learning_rate": 9.924711632212061e-06, "loss": 0.2614, "lr_vision_tower": 1.984942326442412e-06, "sft_loss": 0.1275, "step": 136 }, { "distill_loss": 0.2708, "epoch": 0.08, "grad_norm": 0.7693149047981266, "learning_rate": 9.922975401133292e-06, "loss": 0.2509, "lr_vision_tower": 1.9845950802266584e-06, "sft_loss": 0.1155, "step": 137 }, { "distill_loss": 0.2792, "epoch": 0.09, "grad_norm": 0.8605956467190634, "learning_rate": 9.92121953329501e-06, "loss": 0.254, "lr_vision_tower": 1.9842439066590016e-06, "sft_loss": 0.1145, "step": 138 }, { "distill_loss": 0.279, "epoch": 0.09, "grad_norm": 0.9975521327722756, "learning_rate": 9.919444035701016e-06, "loss": 0.2568, "lr_vision_tower": 1.983888807140203e-06, "sft_loss": 0.1173, "step": 139 }, { "distill_loss": 0.2825, "epoch": 0.09, "grad_norm": 0.7811804255093223, "learning_rate": 9.917648915433413e-06, "loss": 0.2637, "lr_vision_tower": 1.9835297830866823e-06, "sft_loss": 0.1225, "step": 140 }, { "distill_loss": 0.2785, "epoch": 0.09, "grad_norm": 0.7163882798449384, "learning_rate": 9.915834179652577e-06, "loss": 0.2484, "lr_vision_tower": 1.983166835930515e-06, "sft_loss": 0.1091, "step": 141 }, { "distill_loss": 0.2613, "epoch": 0.09, "grad_norm": 0.7534705721897744, "learning_rate": 9.91399983559712e-06, "loss": 0.2463, "lr_vision_tower": 1.982799967119424e-06, "sft_loss": 0.1156, "step": 142 }, { "distill_loss": 0.2769, "epoch": 0.09, "grad_norm": 0.7311830875138249, "learning_rate": 9.912145890583877e-06, "loss": 0.2695, "lr_vision_tower": 1.9824291781167754e-06, "sft_loss": 0.1311, "step": 143 }, { "distill_loss": 0.256, "epoch": 0.09, "grad_norm": 0.7388103651631268, "learning_rate": 9.910272352007859e-06, "loss": 0.2301, "lr_vision_tower": 1.9820544704015714e-06, "sft_loss": 0.1021, "step": 144 }, { "distill_loss": 0.2606, "epoch": 0.09, "grad_norm": 0.9368565921318376, "learning_rate": 9.908379227342238e-06, "loss": 0.2515, "lr_vision_tower": 1.9816758454684473e-06, "sft_loss": 0.1212, "step": 145 }, { "distill_loss": 0.248, "epoch": 0.09, "grad_norm": 0.7298903617577451, "learning_rate": 9.906466524138304e-06, "loss": 0.2381, "lr_vision_tower": 1.9812933048276603e-06, "sft_loss": 0.1141, "step": 146 }, { "distill_loss": 0.2711, "epoch": 0.09, "grad_norm": 0.8285024778212924, "learning_rate": 9.904534250025447e-06, "loss": 0.2426, "lr_vision_tower": 1.980906850005089e-06, "sft_loss": 0.107, "step": 147 }, { "distill_loss": 0.2612, "epoch": 0.09, "grad_norm": 0.8014428820023963, "learning_rate": 9.90258241271112e-06, "loss": 0.2591, "lr_vision_tower": 1.9805164825422237e-06, "sft_loss": 0.1285, "step": 148 }, { "distill_loss": 0.2744, "epoch": 0.09, "grad_norm": 1.0226600828424062, "learning_rate": 9.900611019980809e-06, "loss": 0.2577, "lr_vision_tower": 1.9801222039961616e-06, "sft_loss": 0.1205, "step": 149 }, { "distill_loss": 0.2724, "epoch": 0.09, "grad_norm": 0.8483650917245883, "learning_rate": 9.898620079698006e-06, "loss": 0.252, "lr_vision_tower": 1.9797240159396008e-06, "sft_loss": 0.1158, "step": 150 }, { "distill_loss": 0.2646, "epoch": 0.09, "grad_norm": 1.1122182384136998, "learning_rate": 9.89660959980417e-06, "loss": 0.2646, "lr_vision_tower": 1.9793219199608338e-06, "sft_loss": 0.1323, "step": 151 }, { "distill_loss": 0.2689, "epoch": 0.09, "grad_norm": 0.8456510876725535, "learning_rate": 9.894579588318703e-06, "loss": 0.2553, "lr_vision_tower": 1.9789159176637404e-06, "sft_loss": 0.1209, "step": 152 }, { "distill_loss": 0.2699, "epoch": 0.09, "grad_norm": 0.7004481762375246, "learning_rate": 9.892530053338909e-06, "loss": 0.2551, "lr_vision_tower": 1.9785060106677815e-06, "sft_loss": 0.1202, "step": 153 }, { "distill_loss": 0.2592, "epoch": 0.09, "grad_norm": 0.7290373522815528, "learning_rate": 9.890461003039974e-06, "loss": 0.2477, "lr_vision_tower": 1.9780922006079946e-06, "sft_loss": 0.1181, "step": 154 }, { "distill_loss": 0.2667, "epoch": 0.1, "grad_norm": 0.7269961776309907, "learning_rate": 9.888372445674923e-06, "loss": 0.2409, "lr_vision_tower": 1.9776744891349843e-06, "sft_loss": 0.1075, "step": 155 }, { "distill_loss": 0.2771, "epoch": 0.1, "grad_norm": 0.8900198567113993, "learning_rate": 9.886264389574593e-06, "loss": 0.2616, "lr_vision_tower": 1.9772528779149186e-06, "sft_loss": 0.1231, "step": 156 }, { "distill_loss": 0.2547, "epoch": 0.1, "grad_norm": 0.8194748675179528, "learning_rate": 9.884136843147595e-06, "loss": 0.2473, "lr_vision_tower": 1.9768273686295183e-06, "sft_loss": 0.1199, "step": 157 }, { "distill_loss": 0.2698, "epoch": 0.1, "grad_norm": 0.7775674508193055, "learning_rate": 9.881989814880282e-06, "loss": 0.2477, "lr_vision_tower": 1.9763979629760562e-06, "sft_loss": 0.1128, "step": 158 }, { "distill_loss": 0.2584, "epoch": 0.1, "grad_norm": 0.7609861763483563, "learning_rate": 9.879823313336723e-06, "loss": 0.2398, "lr_vision_tower": 1.9759646626673444e-06, "sft_loss": 0.1107, "step": 159 }, { "distill_loss": 0.2653, "epoch": 0.1, "grad_norm": 0.9046954215174225, "learning_rate": 9.877637347158654e-06, "loss": 0.2501, "lr_vision_tower": 1.97552746943173e-06, "sft_loss": 0.1174, "step": 160 }, { "distill_loss": 0.2581, "epoch": 0.1, "grad_norm": 0.8575463548299735, "learning_rate": 9.875431925065454e-06, "loss": 0.2573, "lr_vision_tower": 1.975086385013091e-06, "sft_loss": 0.1283, "step": 161 }, { "distill_loss": 0.2738, "epoch": 0.1, "grad_norm": 0.9161620297394286, "learning_rate": 9.873207055854113e-06, "loss": 0.2506, "lr_vision_tower": 1.974641411170822e-06, "sft_loss": 0.1137, "step": 162 }, { "distill_loss": 0.2628, "epoch": 0.1, "grad_norm": 0.8534033793129442, "learning_rate": 9.870962748399181e-06, "loss": 0.2394, "lr_vision_tower": 1.974192549679836e-06, "sft_loss": 0.108, "step": 163 }, { "distill_loss": 0.2645, "epoch": 0.1, "grad_norm": 1.0587749465783092, "learning_rate": 9.868699011652754e-06, "loss": 0.258, "lr_vision_tower": 1.9737398023305504e-06, "sft_loss": 0.1258, "step": 164 }, { "distill_loss": 0.2592, "epoch": 0.1, "grad_norm": 0.845399201951569, "learning_rate": 9.866415854644421e-06, "loss": 0.2459, "lr_vision_tower": 1.973283170928884e-06, "sft_loss": 0.1163, "step": 165 }, { "distill_loss": 0.2692, "epoch": 0.1, "grad_norm": 0.8160334475549624, "learning_rate": 9.864113286481237e-06, "loss": 0.2474, "lr_vision_tower": 1.972822657296247e-06, "sft_loss": 0.1128, "step": 166 }, { "distill_loss": 0.262, "epoch": 0.1, "grad_norm": 0.8059520865786761, "learning_rate": 9.861791316347685e-06, "loss": 0.2497, "lr_vision_tower": 1.9723582632695367e-06, "sft_loss": 0.1187, "step": 167 }, { "distill_loss": 0.2624, "epoch": 0.1, "grad_norm": 0.8657313821792275, "learning_rate": 9.859449953505632e-06, "loss": 0.2504, "lr_vision_tower": 1.9718899907011265e-06, "sft_loss": 0.1192, "step": 168 }, { "distill_loss": 0.2658, "epoch": 0.1, "grad_norm": 0.8923802366492849, "learning_rate": 9.857089207294313e-06, "loss": 0.2514, "lr_vision_tower": 1.9714178414588624e-06, "sft_loss": 0.1184, "step": 169 }, { "distill_loss": 0.2623, "epoch": 0.1, "grad_norm": 0.6479774737741043, "learning_rate": 9.854709087130261e-06, "loss": 0.2475, "lr_vision_tower": 1.970941817426052e-06, "sft_loss": 0.1164, "step": 170 }, { "distill_loss": 0.2662, "epoch": 0.11, "grad_norm": 0.7716265190067879, "learning_rate": 9.852309602507302e-06, "loss": 0.2523, "lr_vision_tower": 1.97046192050146e-06, "sft_loss": 0.1192, "step": 171 }, { "distill_loss": 0.2716, "epoch": 0.11, "grad_norm": 0.890546536218328, "learning_rate": 9.849890762996497e-06, "loss": 0.258, "lr_vision_tower": 1.9699781525992992e-06, "sft_loss": 0.1222, "step": 172 }, { "distill_loss": 0.2642, "epoch": 0.11, "grad_norm": 0.6363482769615455, "learning_rate": 9.84745257824611e-06, "loss": 0.2467, "lr_vision_tower": 1.969490515649222e-06, "sft_loss": 0.1147, "step": 173 }, { "distill_loss": 0.2447, "epoch": 0.11, "grad_norm": 1.1640400532470805, "learning_rate": 9.84499505798157e-06, "loss": 0.2482, "lr_vision_tower": 1.9689990115963138e-06, "sft_loss": 0.1259, "step": 174 }, { "distill_loss": 0.2606, "epoch": 0.11, "grad_norm": 1.0473028508338584, "learning_rate": 9.842518212005433e-06, "loss": 0.2556, "lr_vision_tower": 1.9685036424010864e-06, "sft_loss": 0.1253, "step": 175 }, { "distill_loss": 0.2645, "epoch": 0.11, "grad_norm": 0.7504134786869423, "learning_rate": 9.840022050197338e-06, "loss": 0.2326, "lr_vision_tower": 1.9680044100394677e-06, "sft_loss": 0.1004, "step": 176 }, { "distill_loss": 0.2663, "epoch": 0.11, "grad_norm": 0.8847554011330264, "learning_rate": 9.837506582513975e-06, "loss": 0.243, "lr_vision_tower": 1.967501316502795e-06, "sft_loss": 0.1098, "step": 177 }, { "distill_loss": 0.2558, "epoch": 0.11, "grad_norm": 0.8942586194426435, "learning_rate": 9.834971818989037e-06, "loss": 0.2417, "lr_vision_tower": 1.966994363797807e-06, "sft_loss": 0.1138, "step": 178 }, { "distill_loss": 0.2485, "epoch": 0.11, "grad_norm": 0.9358129254445516, "learning_rate": 9.832417769733185e-06, "loss": 0.2473, "lr_vision_tower": 1.966483553946637e-06, "sft_loss": 0.1231, "step": 179 }, { "distill_loss": 0.26, "epoch": 0.11, "grad_norm": 0.8247139313174139, "learning_rate": 9.829844444934012e-06, "loss": 0.2314, "lr_vision_tower": 1.9659688889868024e-06, "sft_loss": 0.1014, "step": 180 }, { "distill_loss": 0.2556, "epoch": 0.11, "grad_norm": 0.9130858108182465, "learning_rate": 9.827251854855992e-06, "loss": 0.2428, "lr_vision_tower": 1.965450370971198e-06, "sft_loss": 0.115, "step": 181 }, { "distill_loss": 0.2637, "epoch": 0.11, "grad_norm": 0.9105076076786028, "learning_rate": 9.824640009840444e-06, "loss": 0.2437, "lr_vision_tower": 1.9649280019680886e-06, "sft_loss": 0.1118, "step": 182 }, { "distill_loss": 0.2486, "epoch": 0.11, "grad_norm": 0.9930390871206998, "learning_rate": 9.822008920305495e-06, "loss": 0.2439, "lr_vision_tower": 1.9644017840610985e-06, "sft_loss": 0.1196, "step": 183 }, { "distill_loss": 0.2591, "epoch": 0.11, "grad_norm": 0.7127218472835023, "learning_rate": 9.81935859674603e-06, "loss": 0.2488, "lr_vision_tower": 1.9638717193492062e-06, "sft_loss": 0.1192, "step": 184 }, { "distill_loss": 0.2592, "epoch": 0.11, "grad_norm": 0.7863844161365559, "learning_rate": 9.81668904973366e-06, "loss": 0.2542, "lr_vision_tower": 1.963337809946732e-06, "sft_loss": 0.1246, "step": 185 }, { "distill_loss": 0.2537, "epoch": 0.11, "grad_norm": 0.7436925395791913, "learning_rate": 9.814000289916672e-06, "loss": 0.2534, "lr_vision_tower": 1.9628000579833343e-06, "sft_loss": 0.1265, "step": 186 }, { "distill_loss": 0.2484, "epoch": 0.12, "grad_norm": 0.6095184186663258, "learning_rate": 9.811292328019985e-06, "loss": 0.2433, "lr_vision_tower": 1.962258465603997e-06, "sft_loss": 0.119, "step": 187 }, { "distill_loss": 0.2584, "epoch": 0.12, "grad_norm": 0.7668341423700539, "learning_rate": 9.808565174845117e-06, "loss": 0.244, "lr_vision_tower": 1.9617130349690233e-06, "sft_loss": 0.1148, "step": 188 }, { "distill_loss": 0.2382, "epoch": 0.12, "grad_norm": 0.7430414832566276, "learning_rate": 9.805818841270134e-06, "loss": 0.2395, "lr_vision_tower": 1.9611637682540267e-06, "sft_loss": 0.1204, "step": 189 }, { "distill_loss": 0.255, "epoch": 0.12, "grad_norm": 0.9821244006774078, "learning_rate": 9.803053338249606e-06, "loss": 0.2559, "lr_vision_tower": 1.960610667649921e-06, "sft_loss": 0.1284, "step": 190 }, { "distill_loss": 0.249, "epoch": 0.12, "grad_norm": 1.061897370152237, "learning_rate": 9.800268676814572e-06, "loss": 0.2518, "lr_vision_tower": 1.9600537353629143e-06, "sft_loss": 0.1273, "step": 191 }, { "distill_loss": 0.2434, "epoch": 0.12, "grad_norm": 0.8910469758261391, "learning_rate": 9.797464868072489e-06, "loss": 0.2483, "lr_vision_tower": 1.9594929736144973e-06, "sft_loss": 0.1266, "step": 192 }, { "distill_loss": 0.2465, "epoch": 0.12, "grad_norm": 0.8047807335191792, "learning_rate": 9.794641923207179e-06, "loss": 0.2454, "lr_vision_tower": 1.9589283846414357e-06, "sft_loss": 0.1222, "step": 193 }, { "distill_loss": 0.2536, "epoch": 0.12, "grad_norm": 0.757975423456102, "learning_rate": 9.791799853478808e-06, "loss": 0.2373, "lr_vision_tower": 1.9583599706957614e-06, "sft_loss": 0.1105, "step": 194 }, { "distill_loss": 0.2502, "epoch": 0.12, "grad_norm": 0.8645870472522219, "learning_rate": 9.78893867022382e-06, "loss": 0.254, "lr_vision_tower": 1.957787734044764e-06, "sft_loss": 0.1289, "step": 195 }, { "distill_loss": 0.2465, "epoch": 0.12, "grad_norm": 0.8059549534415513, "learning_rate": 9.786058384854898e-06, "loss": 0.2465, "lr_vision_tower": 1.9572116769709794e-06, "sft_loss": 0.1233, "step": 196 }, { "distill_loss": 0.2442, "epoch": 0.12, "grad_norm": 0.7605726003663524, "learning_rate": 9.783159008860923e-06, "loss": 0.2366, "lr_vision_tower": 1.9566318017721845e-06, "sft_loss": 0.1145, "step": 197 }, { "distill_loss": 0.2437, "epoch": 0.12, "grad_norm": 0.9220248296463781, "learning_rate": 9.780240553806923e-06, "loss": 0.2503, "lr_vision_tower": 1.9560481107613846e-06, "sft_loss": 0.1285, "step": 198 }, { "distill_loss": 0.269, "epoch": 0.12, "grad_norm": 0.865210112705505, "learning_rate": 9.77730303133403e-06, "loss": 0.2406, "lr_vision_tower": 1.9554606062668054e-06, "sft_loss": 0.1061, "step": 199 }, { "distill_loss": 0.2554, "epoch": 0.12, "grad_norm": 0.9170960174313143, "learning_rate": 9.774346453159426e-06, "loss": 0.2336, "lr_vision_tower": 1.9548692906318848e-06, "sft_loss": 0.1059, "step": 200 }, { "distill_loss": 0.2502, "epoch": 0.12, "grad_norm": 0.7657398677543878, "learning_rate": 9.771370831076312e-06, "loss": 0.2354, "lr_vision_tower": 1.954274166215262e-06, "sft_loss": 0.1103, "step": 201 }, { "distill_loss": 0.2579, "epoch": 0.12, "grad_norm": 1.0783940017407512, "learning_rate": 9.768376176953845e-06, "loss": 0.2452, "lr_vision_tower": 1.9536752353907686e-06, "sft_loss": 0.1163, "step": 202 }, { "distill_loss": 0.2418, "epoch": 0.13, "grad_norm": 0.8546992475313835, "learning_rate": 9.765362502737098e-06, "loss": 0.2602, "lr_vision_tower": 1.9530725005474194e-06, "sft_loss": 0.1393, "step": 203 }, { "distill_loss": 0.2436, "epoch": 0.13, "grad_norm": 0.9210730883030246, "learning_rate": 9.762329820447013e-06, "loss": 0.2292, "lr_vision_tower": 1.9524659640894024e-06, "sft_loss": 0.1074, "step": 204 }, { "distill_loss": 0.2532, "epoch": 0.13, "grad_norm": 0.7719440923275698, "learning_rate": 9.759278142180348e-06, "loss": 0.245, "lr_vision_tower": 1.9518556284360694e-06, "sft_loss": 0.1184, "step": 205 }, { "distill_loss": 0.2484, "epoch": 0.13, "grad_norm": 0.8532081293493327, "learning_rate": 9.756207480109637e-06, "loss": 0.2378, "lr_vision_tower": 1.9512414960219274e-06, "sft_loss": 0.1136, "step": 206 }, { "distill_loss": 0.2405, "epoch": 0.13, "grad_norm": 0.9262037795672229, "learning_rate": 9.753117846483133e-06, "loss": 0.2347, "lr_vision_tower": 1.9506235692966266e-06, "sft_loss": 0.1144, "step": 207 }, { "distill_loss": 0.2502, "epoch": 0.13, "grad_norm": 0.7851064025599989, "learning_rate": 9.750009253624762e-06, "loss": 0.2436, "lr_vision_tower": 1.9500018507249523e-06, "sft_loss": 0.1186, "step": 208 }, { "distill_loss": 0.253, "epoch": 0.13, "grad_norm": 0.8861548396156058, "learning_rate": 9.746881713934079e-06, "loss": 0.2421, "lr_vision_tower": 1.9493763427868155e-06, "sft_loss": 0.1156, "step": 209 }, { "distill_loss": 0.2437, "epoch": 0.13, "grad_norm": 0.7317874791142132, "learning_rate": 9.743735239886209e-06, "loss": 0.2235, "lr_vision_tower": 1.9487470479772416e-06, "sft_loss": 0.1017, "step": 210 }, { "distill_loss": 0.2426, "epoch": 0.13, "grad_norm": 0.7505614211794165, "learning_rate": 9.740569844031804e-06, "loss": 0.2341, "lr_vision_tower": 1.9481139688063604e-06, "sft_loss": 0.1128, "step": 211 }, { "distill_loss": 0.2555, "epoch": 0.13, "grad_norm": 0.7870232415879138, "learning_rate": 9.737385538996995e-06, "loss": 0.2331, "lr_vision_tower": 1.947477107799399e-06, "sft_loss": 0.1054, "step": 212 }, { "distill_loss": 0.2519, "epoch": 0.13, "grad_norm": 0.8558863410887698, "learning_rate": 9.734182337483332e-06, "loss": 0.2445, "lr_vision_tower": 1.9468364674966662e-06, "sft_loss": 0.1186, "step": 213 }, { "distill_loss": 0.2479, "epoch": 0.13, "grad_norm": 0.9475346335208955, "learning_rate": 9.730960252267744e-06, "loss": 0.2467, "lr_vision_tower": 1.9461920504535483e-06, "sft_loss": 0.1228, "step": 214 }, { "distill_loss": 0.2376, "epoch": 0.13, "grad_norm": 0.7517417036699878, "learning_rate": 9.72771929620248e-06, "loss": 0.2255, "lr_vision_tower": 1.9455438592404957e-06, "sft_loss": 0.1066, "step": 215 }, { "distill_loss": 0.2499, "epoch": 0.13, "grad_norm": 0.776130934715154, "learning_rate": 9.724459482215063e-06, "loss": 0.2356, "lr_vision_tower": 1.9448918964430122e-06, "sft_loss": 0.1107, "step": 216 }, { "distill_loss": 0.2364, "epoch": 0.13, "grad_norm": 0.9619829156110101, "learning_rate": 9.721180823308237e-06, "loss": 0.2292, "lr_vision_tower": 1.944236164661647e-06, "sft_loss": 0.111, "step": 217 }, { "distill_loss": 0.2404, "epoch": 0.13, "grad_norm": 0.7576995718540454, "learning_rate": 9.717883332559911e-06, "loss": 0.2486, "lr_vision_tower": 1.943576666511982e-06, "sft_loss": 0.1284, "step": 218 }, { "distill_loss": 0.2436, "epoch": 0.13, "grad_norm": 0.7118295624560633, "learning_rate": 9.714567023123113e-06, "loss": 0.2382, "lr_vision_tower": 1.9429134046246224e-06, "sft_loss": 0.1164, "step": 219 }, { "distill_loss": 0.2577, "epoch": 0.14, "grad_norm": 0.7742704944228296, "learning_rate": 9.71123190822594e-06, "loss": 0.2344, "lr_vision_tower": 1.9422463816451878e-06, "sft_loss": 0.1055, "step": 220 }, { "distill_loss": 0.2412, "epoch": 0.14, "grad_norm": 0.7520370911738634, "learning_rate": 9.70787800117149e-06, "loss": 0.245, "lr_vision_tower": 1.9415756002342975e-06, "sft_loss": 0.1244, "step": 221 }, { "distill_loss": 0.2293, "epoch": 0.14, "grad_norm": 0.9637744969305327, "learning_rate": 9.704505315337822e-06, "loss": 0.2178, "lr_vision_tower": 1.940901063067564e-06, "sft_loss": 0.1032, "step": 222 }, { "distill_loss": 0.2358, "epoch": 0.14, "grad_norm": 0.7995218980943921, "learning_rate": 9.701113864177909e-06, "loss": 0.2287, "lr_vision_tower": 1.9402227728355816e-06, "sft_loss": 0.1108, "step": 223 }, { "distill_loss": 0.2394, "epoch": 0.14, "grad_norm": 0.7988061397754667, "learning_rate": 9.697703661219564e-06, "loss": 0.2319, "lr_vision_tower": 1.9395407322439122e-06, "sft_loss": 0.1122, "step": 224 }, { "distill_loss": 0.2457, "epoch": 0.14, "grad_norm": 0.7786058012666396, "learning_rate": 9.6942747200654e-06, "loss": 0.248, "lr_vision_tower": 1.9388549440130797e-06, "sft_loss": 0.1251, "step": 225 }, { "distill_loss": 0.2574, "epoch": 0.14, "grad_norm": 0.78595845828903, "learning_rate": 9.690827054392777e-06, "loss": 0.2416, "lr_vision_tower": 1.9381654108785553e-06, "sft_loss": 0.1129, "step": 226 }, { "distill_loss": 0.2471, "epoch": 0.14, "grad_norm": 0.6921010398810143, "learning_rate": 9.68736067795374e-06, "loss": 0.2362, "lr_vision_tower": 1.937472135590748e-06, "sft_loss": 0.1126, "step": 227 }, { "distill_loss": 0.2357, "epoch": 0.14, "grad_norm": 0.8736845741333429, "learning_rate": 9.683875604574967e-06, "loss": 0.2375, "lr_vision_tower": 1.936775120914993e-06, "sft_loss": 0.1196, "step": 228 }, { "distill_loss": 0.2548, "epoch": 0.14, "grad_norm": 0.7158480998908788, "learning_rate": 9.680371848157716e-06, "loss": 0.2582, "lr_vision_tower": 1.9360743696315428e-06, "sft_loss": 0.1308, "step": 229 }, { "distill_loss": 0.2477, "epoch": 0.14, "grad_norm": 2.0239558507277526, "learning_rate": 9.676849422677766e-06, "loss": 0.2289, "lr_vision_tower": 1.9353698845355533e-06, "sft_loss": 0.1051, "step": 230 }, { "distill_loss": 0.2354, "epoch": 0.14, "grad_norm": 0.8287895260823903, "learning_rate": 9.673308342185366e-06, "loss": 0.2393, "lr_vision_tower": 1.934661668437073e-06, "sft_loss": 0.1216, "step": 231 }, { "distill_loss": 0.239, "epoch": 0.14, "grad_norm": 0.7324611800103574, "learning_rate": 9.669748620805173e-06, "loss": 0.2078, "lr_vision_tower": 1.9339497241610347e-06, "sft_loss": 0.0883, "step": 232 }, { "distill_loss": 0.2408, "epoch": 0.14, "grad_norm": 0.7911362477561927, "learning_rate": 9.666170272736203e-06, "loss": 0.2219, "lr_vision_tower": 1.9332340545472405e-06, "sft_loss": 0.1015, "step": 233 }, { "distill_loss": 0.2421, "epoch": 0.14, "grad_norm": 0.6009556637025324, "learning_rate": 9.662573312251765e-06, "loss": 0.2341, "lr_vision_tower": 1.9325146624503528e-06, "sft_loss": 0.113, "step": 234 }, { "distill_loss": 0.2364, "epoch": 0.14, "grad_norm": 0.6736938063914298, "learning_rate": 9.65895775369941e-06, "loss": 0.2276, "lr_vision_tower": 1.9317915507398815e-06, "sft_loss": 0.1094, "step": 235 }, { "distill_loss": 0.2331, "epoch": 0.15, "grad_norm": 0.781323202412316, "learning_rate": 9.655323611500876e-06, "loss": 0.2126, "lr_vision_tower": 1.9310647223001747e-06, "sft_loss": 0.096, "step": 236 }, { "distill_loss": 0.2371, "epoch": 0.15, "grad_norm": 0.7208278294121817, "learning_rate": 9.651670900152027e-06, "loss": 0.2199, "lr_vision_tower": 1.930334180030405e-06, "sft_loss": 0.1013, "step": 237 }, { "distill_loss": 0.2502, "epoch": 0.15, "grad_norm": 0.6949604417237473, "learning_rate": 9.647999634222792e-06, "loss": 0.2324, "lr_vision_tower": 1.9295999268445584e-06, "sft_loss": 0.1073, "step": 238 }, { "distill_loss": 0.2415, "epoch": 0.15, "grad_norm": 0.9547132535375733, "learning_rate": 9.644309828357117e-06, "loss": 0.2428, "lr_vision_tower": 1.928861965671423e-06, "sft_loss": 0.1221, "step": 239 }, { "distill_loss": 0.2425, "epoch": 0.15, "grad_norm": 1.3518908586052865, "learning_rate": 9.640601497272892e-06, "loss": 0.2293, "lr_vision_tower": 1.9281202994545783e-06, "sft_loss": 0.1081, "step": 240 }, { "distill_loss": 0.2369, "epoch": 0.15, "grad_norm": 1.047606490604499, "learning_rate": 9.636874655761908e-06, "loss": 0.2241, "lr_vision_tower": 1.9273749311523814e-06, "sft_loss": 0.1056, "step": 241 }, { "distill_loss": 0.2346, "epoch": 0.15, "grad_norm": 1.0200023030428116, "learning_rate": 9.633129318689782e-06, "loss": 0.2197, "lr_vision_tower": 1.926625863737956e-06, "sft_loss": 0.1024, "step": 242 }, { "distill_loss": 0.2483, "epoch": 0.15, "grad_norm": 0.7477983140024778, "learning_rate": 9.629365500995916e-06, "loss": 0.2346, "lr_vision_tower": 1.925873100199183e-06, "sft_loss": 0.1104, "step": 243 }, { "distill_loss": 0.2513, "epoch": 0.15, "grad_norm": 0.8226872430970978, "learning_rate": 9.625583217693419e-06, "loss": 0.2334, "lr_vision_tower": 1.9251166435386836e-06, "sft_loss": 0.1077, "step": 244 }, { "distill_loss": 0.2456, "epoch": 0.15, "grad_norm": 1.0500895389716356, "learning_rate": 9.621782483869063e-06, "loss": 0.2463, "lr_vision_tower": 1.9243564967738124e-06, "sft_loss": 0.1235, "step": 245 }, { "distill_loss": 0.2366, "epoch": 0.15, "grad_norm": 0.8141883673155764, "learning_rate": 9.617963314683205e-06, "loss": 0.2339, "lr_vision_tower": 1.9235926629366405e-06, "sft_loss": 0.1156, "step": 246 }, { "distill_loss": 0.2326, "epoch": 0.15, "grad_norm": 0.8736343647253946, "learning_rate": 9.614125725369748e-06, "loss": 0.2412, "lr_vision_tower": 1.922825145073949e-06, "sft_loss": 0.1249, "step": 247 }, { "distill_loss": 0.2453, "epoch": 0.15, "grad_norm": 0.8345434182392578, "learning_rate": 9.610269731236063e-06, "loss": 0.2552, "lr_vision_tower": 1.9220539462472123e-06, "sft_loss": 0.1325, "step": 248 }, { "distill_loss": 0.2294, "epoch": 0.15, "grad_norm": 0.8146411518835196, "learning_rate": 9.606395347662936e-06, "loss": 0.2159, "lr_vision_tower": 1.9212790695325867e-06, "sft_loss": 0.1012, "step": 249 }, { "distill_loss": 0.2433, "epoch": 0.15, "grad_norm": 0.9706095206826855, "learning_rate": 9.602502590104505e-06, "loss": 0.2369, "lr_vision_tower": 1.920500518020901e-06, "sft_loss": 0.1153, "step": 250 }, { "distill_loss": 0.2319, "epoch": 0.15, "grad_norm": 0.8304000162271418, "learning_rate": 9.598591474088197e-06, "loss": 0.2316, "lr_vision_tower": 1.9197182948176393e-06, "sft_loss": 0.1156, "step": 251 }, { "distill_loss": 0.2328, "epoch": 0.16, "grad_norm": 0.9226635791363212, "learning_rate": 9.59466201521467e-06, "loss": 0.2416, "lr_vision_tower": 1.9189324030429334e-06, "sft_loss": 0.1252, "step": 252 }, { "distill_loss": 0.2322, "epoch": 0.16, "grad_norm": 0.9748716587026257, "learning_rate": 9.59071422915774e-06, "loss": 0.2133, "lr_vision_tower": 1.9181428458315477e-06, "sft_loss": 0.0972, "step": 253 }, { "distill_loss": 0.2327, "epoch": 0.16, "grad_norm": 0.828362676566001, "learning_rate": 9.586748131664337e-06, "loss": 0.2147, "lr_vision_tower": 1.917349626332867e-06, "sft_loss": 0.0984, "step": 254 }, { "distill_loss": 0.2286, "epoch": 0.16, "grad_norm": 0.6456702768624915, "learning_rate": 9.582763738554427e-06, "loss": 0.2307, "lr_vision_tower": 1.916552747710885e-06, "sft_loss": 0.1164, "step": 255 }, { "distill_loss": 0.2378, "epoch": 0.16, "grad_norm": 0.7953979051515138, "learning_rate": 9.57876106572095e-06, "loss": 0.2273, "lr_vision_tower": 1.9157522131441897e-06, "sft_loss": 0.1084, "step": 256 }, { "distill_loss": 0.2425, "epoch": 0.16, "grad_norm": 0.862715219220027, "learning_rate": 9.574740129129767e-06, "loss": 0.2492, "lr_vision_tower": 1.914948025825953e-06, "sft_loss": 0.128, "step": 257 }, { "distill_loss": 0.2365, "epoch": 0.16, "grad_norm": 1.0333347450196473, "learning_rate": 9.570700944819584e-06, "loss": 0.2414, "lr_vision_tower": 1.9141401889639164e-06, "sft_loss": 0.1232, "step": 258 }, { "distill_loss": 0.2265, "epoch": 0.16, "grad_norm": 0.8672484767106411, "learning_rate": 9.566643528901896e-06, "loss": 0.2294, "lr_vision_tower": 1.913328705780379e-06, "sft_loss": 0.1162, "step": 259 }, { "distill_loss": 0.2357, "epoch": 0.16, "grad_norm": 0.9023261229886764, "learning_rate": 9.56256789756092e-06, "loss": 0.2332, "lr_vision_tower": 1.9125135795121836e-06, "sft_loss": 0.1154, "step": 260 }, { "distill_loss": 0.2303, "epoch": 0.16, "grad_norm": 0.919693272696708, "learning_rate": 9.558474067053531e-06, "loss": 0.2296, "lr_vision_tower": 1.911694813410706e-06, "sft_loss": 0.1145, "step": 261 }, { "distill_loss": 0.2364, "epoch": 0.16, "grad_norm": 0.8954197543797965, "learning_rate": 9.554362053709194e-06, "loss": 0.2377, "lr_vision_tower": 1.9108724107418385e-06, "sft_loss": 0.1195, "step": 262 }, { "distill_loss": 0.2294, "epoch": 0.16, "grad_norm": 0.612378729331648, "learning_rate": 9.550231873929904e-06, "loss": 0.2244, "lr_vision_tower": 1.9100463747859807e-06, "sft_loss": 0.1097, "step": 263 }, { "distill_loss": 0.2354, "epoch": 0.16, "grad_norm": 0.6918896856844189, "learning_rate": 9.54608354419012e-06, "loss": 0.2292, "lr_vision_tower": 1.909216708838024e-06, "sft_loss": 0.1115, "step": 264 }, { "distill_loss": 0.2368, "epoch": 0.16, "grad_norm": 0.7365903807864148, "learning_rate": 9.541917081036697e-06, "loss": 0.2196, "lr_vision_tower": 1.908383416207339e-06, "sft_loss": 0.1012, "step": 265 }, { "distill_loss": 0.2428, "epoch": 0.16, "grad_norm": 0.7452947465803852, "learning_rate": 9.537732501088816e-06, "loss": 0.2365, "lr_vision_tower": 1.907546500217763e-06, "sft_loss": 0.1151, "step": 266 }, { "distill_loss": 0.2372, "epoch": 0.16, "grad_norm": 0.6387632113776914, "learning_rate": 9.533529821037925e-06, "loss": 0.2416, "lr_vision_tower": 1.9067059642075848e-06, "sft_loss": 0.123, "step": 267 }, { "distill_loss": 0.2339, "epoch": 0.17, "grad_norm": 0.6229659711347473, "learning_rate": 9.529309057647673e-06, "loss": 0.2183, "lr_vision_tower": 1.9058618115295344e-06, "sft_loss": 0.1013, "step": 268 }, { "distill_loss": 0.2419, "epoch": 0.17, "grad_norm": 0.8976868900733809, "learning_rate": 9.525070227753835e-06, "loss": 0.2252, "lr_vision_tower": 1.9050140455507666e-06, "sft_loss": 0.1043, "step": 269 }, { "distill_loss": 0.2354, "epoch": 0.17, "grad_norm": 0.7572404928828642, "learning_rate": 9.520813348264252e-06, "loss": 0.2404, "lr_vision_tower": 1.9041626696528501e-06, "sft_loss": 0.1227, "step": 270 }, { "distill_loss": 0.2335, "epoch": 0.17, "grad_norm": 0.8750970641915622, "learning_rate": 9.51653843615876e-06, "loss": 0.2233, "lr_vision_tower": 1.903307687231752e-06, "sft_loss": 0.1066, "step": 271 }, { "distill_loss": 0.2351, "epoch": 0.17, "grad_norm": 0.9267182738637254, "learning_rate": 9.512245508489128e-06, "loss": 0.2442, "lr_vision_tower": 1.9024491016978252e-06, "sft_loss": 0.1266, "step": 272 }, { "distill_loss": 0.2349, "epoch": 0.17, "grad_norm": 0.8398997146979923, "learning_rate": 9.507934582378977e-06, "loss": 0.2327, "lr_vision_tower": 1.9015869164757952e-06, "sft_loss": 0.1152, "step": 273 }, { "distill_loss": 0.242, "epoch": 0.17, "grad_norm": 0.761453510435392, "learning_rate": 9.503605675023726e-06, "loss": 0.2499, "lr_vision_tower": 1.900721135004745e-06, "sft_loss": 0.1289, "step": 274 }, { "distill_loss": 0.2353, "epoch": 0.17, "grad_norm": 0.743575745809471, "learning_rate": 9.49925880369052e-06, "loss": 0.2218, "lr_vision_tower": 1.8998517607381036e-06, "sft_loss": 0.1042, "step": 275 }, { "distill_loss": 0.2162, "epoch": 0.17, "grad_norm": 0.8752589076497003, "learning_rate": 9.494893985718148e-06, "loss": 0.2283, "lr_vision_tower": 1.8989787971436295e-06, "sft_loss": 0.1201, "step": 276 }, { "distill_loss": 0.2163, "epoch": 0.17, "grad_norm": 0.7939019945714735, "learning_rate": 9.490511238516999e-06, "loss": 0.2163, "lr_vision_tower": 1.8981022477033998e-06, "sft_loss": 0.1082, "step": 277 }, { "distill_loss": 0.2375, "epoch": 0.17, "grad_norm": 0.7116803315613269, "learning_rate": 9.486110579568968e-06, "loss": 0.2228, "lr_vision_tower": 1.8972221159137934e-06, "sft_loss": 0.1041, "step": 278 }, { "distill_loss": 0.2336, "epoch": 0.17, "grad_norm": 0.8938000995104707, "learning_rate": 9.481692026427398e-06, "loss": 0.2243, "lr_vision_tower": 1.8963384052854794e-06, "sft_loss": 0.1075, "step": 279 }, { "distill_loss": 0.2315, "epoch": 0.17, "grad_norm": 0.6044067315276151, "learning_rate": 9.477255596717012e-06, "loss": 0.2305, "lr_vision_tower": 1.8954511193434021e-06, "sft_loss": 0.1148, "step": 280 }, { "distill_loss": 0.2354, "epoch": 0.17, "grad_norm": 0.8506376714680192, "learning_rate": 9.472801308133835e-06, "loss": 0.2283, "lr_vision_tower": 1.8945602616267667e-06, "sft_loss": 0.1106, "step": 281 }, { "distill_loss": 0.2329, "epoch": 0.17, "grad_norm": 0.9221347363447095, "learning_rate": 9.468329178445129e-06, "loss": 0.2401, "lr_vision_tower": 1.8936658356890253e-06, "sft_loss": 0.1237, "step": 282 }, { "distill_loss": 0.2354, "epoch": 0.17, "grad_norm": 0.7392134021282063, "learning_rate": 9.46383922548932e-06, "loss": 0.2422, "lr_vision_tower": 1.8927678450978637e-06, "sft_loss": 0.1245, "step": 283 }, { "distill_loss": 0.2254, "epoch": 0.18, "grad_norm": 0.7479922431252167, "learning_rate": 9.459331467175927e-06, "loss": 0.2067, "lr_vision_tower": 1.8918662934351853e-06, "sft_loss": 0.094, "step": 284 }, { "distill_loss": 0.2397, "epoch": 0.18, "grad_norm": 0.86330240915519, "learning_rate": 9.454805921485497e-06, "loss": 0.2271, "lr_vision_tower": 1.8909611842970991e-06, "sft_loss": 0.1073, "step": 285 }, { "distill_loss": 0.2271, "epoch": 0.18, "grad_norm": 0.7951747257585933, "learning_rate": 9.450262606469516e-06, "loss": 0.2329, "lr_vision_tower": 1.8900525212939032e-06, "sft_loss": 0.1193, "step": 286 }, { "distill_loss": 0.2327, "epoch": 0.18, "grad_norm": 0.7479422021548275, "learning_rate": 9.44570154025036e-06, "loss": 0.2345, "lr_vision_tower": 1.8891403080500717e-06, "sft_loss": 0.1181, "step": 287 }, { "distill_loss": 0.2339, "epoch": 0.18, "grad_norm": 0.7943600267671138, "learning_rate": 9.4411227410212e-06, "loss": 0.2485, "lr_vision_tower": 1.88822454820424e-06, "sft_loss": 0.1316, "step": 288 }, { "distill_loss": 0.2323, "epoch": 0.18, "grad_norm": 0.8886442326552125, "learning_rate": 9.436526227045953e-06, "loss": 0.2378, "lr_vision_tower": 1.8873052454091902e-06, "sft_loss": 0.1216, "step": 289 }, { "distill_loss": 0.2287, "epoch": 0.18, "grad_norm": 0.6163418644957444, "learning_rate": 9.431912016659183e-06, "loss": 0.2215, "lr_vision_tower": 1.8863824033318362e-06, "sft_loss": 0.1072, "step": 290 }, { "distill_loss": 0.2478, "epoch": 0.18, "grad_norm": 0.7416729505209748, "learning_rate": 9.427280128266049e-06, "loss": 0.2397, "lr_vision_tower": 1.8854560256532098e-06, "sft_loss": 0.1158, "step": 291 }, { "distill_loss": 0.2328, "epoch": 0.18, "grad_norm": 0.7305350784799234, "learning_rate": 9.422630580342227e-06, "loss": 0.2191, "lr_vision_tower": 1.8845261160684451e-06, "sft_loss": 0.1027, "step": 292 }, { "distill_loss": 0.2323, "epoch": 0.18, "grad_norm": 0.7722523867759763, "learning_rate": 9.417963391433826e-06, "loss": 0.2367, "lr_vision_tower": 1.883592678286765e-06, "sft_loss": 0.1205, "step": 293 }, { "distill_loss": 0.2338, "epoch": 0.18, "grad_norm": 0.7205483343789194, "learning_rate": 9.413278580157324e-06, "loss": 0.2453, "lr_vision_tower": 1.8826557160314646e-06, "sft_loss": 0.1284, "step": 294 }, { "distill_loss": 0.2278, "epoch": 0.18, "grad_norm": 0.9728954635232234, "learning_rate": 9.408576165199495e-06, "loss": 0.2451, "lr_vision_tower": 1.8817152330398987e-06, "sft_loss": 0.1312, "step": 295 }, { "distill_loss": 0.2333, "epoch": 0.18, "grad_norm": 0.7022373866813922, "learning_rate": 9.403856165317322e-06, "loss": 0.2329, "lr_vision_tower": 1.8807712330634642e-06, "sft_loss": 0.1163, "step": 296 }, { "distill_loss": 0.2302, "epoch": 0.18, "grad_norm": 0.8208019158701867, "learning_rate": 9.39911859933794e-06, "loss": 0.2169, "lr_vision_tower": 1.8798237198675878e-06, "sft_loss": 0.1018, "step": 297 }, { "distill_loss": 0.2265, "epoch": 0.18, "grad_norm": 0.7485207781391734, "learning_rate": 9.39436348615855e-06, "loss": 0.2284, "lr_vision_tower": 1.8788726972317095e-06, "sft_loss": 0.1151, "step": 298 }, { "distill_loss": 0.2285, "epoch": 0.18, "grad_norm": 0.6910364028878425, "learning_rate": 9.389590844746334e-06, "loss": 0.2246, "lr_vision_tower": 1.8779181689492667e-06, "sft_loss": 0.1103, "step": 299 }, { "distill_loss": 0.2278, "epoch": 0.18, "grad_norm": 0.7907219418489525, "learning_rate": 9.384800694138407e-06, "loss": 0.2242, "lr_vision_tower": 1.8769601388276812e-06, "sft_loss": 0.1102, "step": 300 }, { "distill_loss": 0.2317, "epoch": 0.19, "grad_norm": 0.8175799708338076, "learning_rate": 9.379993053441717e-06, "loss": 0.2094, "lr_vision_tower": 1.875998610688343e-06, "sft_loss": 0.0935, "step": 301 }, { "distill_loss": 0.2342, "epoch": 0.19, "grad_norm": 1.7337339526628714, "learning_rate": 9.375167941832974e-06, "loss": 0.2272, "lr_vision_tower": 1.8750335883665945e-06, "sft_loss": 0.1101, "step": 302 }, { "distill_loss": 0.2291, "epoch": 0.19, "grad_norm": 0.8947309333843646, "learning_rate": 9.370325378558578e-06, "loss": 0.2207, "lr_vision_tower": 1.8740650757117154e-06, "sft_loss": 0.1062, "step": 303 }, { "distill_loss": 0.2251, "epoch": 0.19, "grad_norm": 0.8486411704791051, "learning_rate": 9.36546538293454e-06, "loss": 0.2169, "lr_vision_tower": 1.873093076586908e-06, "sft_loss": 0.1043, "step": 304 }, { "distill_loss": 0.2228, "epoch": 0.19, "grad_norm": 0.727452098670194, "learning_rate": 9.360587974346411e-06, "loss": 0.2168, "lr_vision_tower": 1.872117594869282e-06, "sft_loss": 0.1054, "step": 305 }, { "distill_loss": 0.2358, "epoch": 0.19, "grad_norm": 0.5903689252600581, "learning_rate": 9.35569317224919e-06, "loss": 0.2331, "lr_vision_tower": 1.8711386344498377e-06, "sft_loss": 0.1152, "step": 306 }, { "distill_loss": 0.224, "epoch": 0.19, "grad_norm": 0.8370747104449365, "learning_rate": 9.350780996167257e-06, "loss": 0.2233, "lr_vision_tower": 1.8701561992334511e-06, "sft_loss": 0.1112, "step": 307 }, { "distill_loss": 0.2313, "epoch": 0.19, "grad_norm": 0.7830134806309271, "learning_rate": 9.345851465694299e-06, "loss": 0.2279, "lr_vision_tower": 1.8691702931388596e-06, "sft_loss": 0.1123, "step": 308 }, { "distill_loss": 0.2304, "epoch": 0.19, "grad_norm": 0.8901072163169664, "learning_rate": 9.34090460049322e-06, "loss": 0.2262, "lr_vision_tower": 1.8681809200986438e-06, "sft_loss": 0.1111, "step": 309 }, { "distill_loss": 0.2283, "epoch": 0.19, "grad_norm": 0.9699162447868364, "learning_rate": 9.335940420296072e-06, "loss": 0.2228, "lr_vision_tower": 1.8671880840592142e-06, "sft_loss": 0.1087, "step": 310 }, { "distill_loss": 0.2299, "epoch": 0.19, "grad_norm": 1.2556482332788368, "learning_rate": 9.330958944903973e-06, "loss": 0.239, "lr_vision_tower": 1.8661917889807944e-06, "sft_loss": 0.124, "step": 311 }, { "distill_loss": 0.2405, "epoch": 0.19, "grad_norm": 1.0733490474771539, "learning_rate": 9.325960194187025e-06, "loss": 0.2353, "lr_vision_tower": 1.8651920388374048e-06, "sft_loss": 0.115, "step": 312 }, { "distill_loss": 0.2195, "epoch": 0.19, "grad_norm": 0.8257052673152618, "learning_rate": 9.320944188084241e-06, "loss": 0.2169, "lr_vision_tower": 1.8641888376168483e-06, "sft_loss": 0.1071, "step": 313 }, { "distill_loss": 0.2306, "epoch": 0.19, "grad_norm": 0.6475881897001552, "learning_rate": 9.315910946603463e-06, "loss": 0.226, "lr_vision_tower": 1.8631821893206925e-06, "sft_loss": 0.1107, "step": 314 }, { "distill_loss": 0.2261, "epoch": 0.19, "grad_norm": 0.6289930780371628, "learning_rate": 9.310860489821279e-06, "loss": 0.2258, "lr_vision_tower": 1.8621720979642554e-06, "sft_loss": 0.1128, "step": 315 }, { "distill_loss": 0.2278, "epoch": 0.19, "grad_norm": 0.7883073241550909, "learning_rate": 9.305792837882943e-06, "loss": 0.2306, "lr_vision_tower": 1.8611585675765885e-06, "sft_loss": 0.1167, "step": 316 }, { "distill_loss": 0.2237, "epoch": 0.2, "grad_norm": 0.7921126995787641, "learning_rate": 9.300708011002303e-06, "loss": 0.2271, "lr_vision_tower": 1.8601416022004606e-06, "sft_loss": 0.1152, "step": 317 }, { "distill_loss": 0.2243, "epoch": 0.2, "grad_norm": 0.9066030160144296, "learning_rate": 9.295606029461713e-06, "loss": 0.2115, "lr_vision_tower": 1.8591212058923423e-06, "sft_loss": 0.0993, "step": 318 }, { "distill_loss": 0.2234, "epoch": 0.2, "grad_norm": 0.7534886660314661, "learning_rate": 9.29048691361195e-06, "loss": 0.2203, "lr_vision_tower": 1.8580973827223896e-06, "sft_loss": 0.1086, "step": 319 }, { "distill_loss": 0.23, "epoch": 0.2, "grad_norm": 1.2845150013125282, "learning_rate": 9.285350683872136e-06, "loss": 0.23, "lr_vision_tower": 1.857070136774427e-06, "sft_loss": 0.115, "step": 320 }, { "distill_loss": 0.2243, "epoch": 0.2, "grad_norm": 0.942149855700842, "learning_rate": 9.280197360729663e-06, "loss": 0.2177, "lr_vision_tower": 1.8560394721459325e-06, "sft_loss": 0.1055, "step": 321 }, { "distill_loss": 0.2186, "epoch": 0.2, "grad_norm": 0.9397793191136401, "learning_rate": 9.275026964740101e-06, "loss": 0.2228, "lr_vision_tower": 1.8550053929480201e-06, "sft_loss": 0.1135, "step": 322 }, { "distill_loss": 0.2345, "epoch": 0.2, "grad_norm": 0.8202326266853724, "learning_rate": 9.269839516527119e-06, "loss": 0.2217, "lr_vision_tower": 1.8539679033054237e-06, "sft_loss": 0.1044, "step": 323 }, { "distill_loss": 0.2224, "epoch": 0.2, "grad_norm": 0.9839128047459458, "learning_rate": 9.264635036782406e-06, "loss": 0.2392, "lr_vision_tower": 1.852927007356481e-06, "sft_loss": 0.128, "step": 324 }, { "distill_loss": 0.2276, "epoch": 0.2, "grad_norm": 1.1459945743355484, "learning_rate": 9.259413546265585e-06, "loss": 0.2351, "lr_vision_tower": 1.8518827092531168e-06, "sft_loss": 0.1213, "step": 325 }, { "distill_loss": 0.2211, "epoch": 0.2, "grad_norm": 0.8602936308948645, "learning_rate": 9.254175065804135e-06, "loss": 0.2242, "lr_vision_tower": 1.8508350131608266e-06, "sft_loss": 0.1137, "step": 326 }, { "distill_loss": 0.2174, "epoch": 0.2, "grad_norm": 1.0981720533002763, "learning_rate": 9.248919616293298e-06, "loss": 0.2202, "lr_vision_tower": 1.8497839232586594e-06, "sft_loss": 0.1115, "step": 327 }, { "distill_loss": 0.2294, "epoch": 0.2, "grad_norm": 1.1935242677641265, "learning_rate": 9.243647218696009e-06, "loss": 0.2301, "lr_vision_tower": 1.8487294437392014e-06, "sft_loss": 0.1154, "step": 328 }, { "distill_loss": 0.2221, "epoch": 0.2, "grad_norm": 0.8552927304797759, "learning_rate": 9.2383578940428e-06, "loss": 0.2244, "lr_vision_tower": 1.8476715788085598e-06, "sft_loss": 0.1133, "step": 329 }, { "distill_loss": 0.2212, "epoch": 0.2, "grad_norm": 0.7319527422701559, "learning_rate": 9.233051663431726e-06, "loss": 0.2239, "lr_vision_tower": 1.8466103326863451e-06, "sft_loss": 0.1133, "step": 330 }, { "distill_loss": 0.2213, "epoch": 0.2, "grad_norm": 0.6729988595357829, "learning_rate": 9.227728548028274e-06, "loss": 0.227, "lr_vision_tower": 1.8455457096056547e-06, "sft_loss": 0.1163, "step": 331 }, { "distill_loss": 0.2205, "epoch": 0.2, "grad_norm": 0.8573513383287571, "learning_rate": 9.222388569065281e-06, "loss": 0.2161, "lr_vision_tower": 1.8444777138130562e-06, "sft_loss": 0.1058, "step": 332 }, { "distill_loss": 0.2282, "epoch": 0.21, "grad_norm": 0.6067093260966088, "learning_rate": 9.217031747842851e-06, "loss": 0.2198, "lr_vision_tower": 1.84340634956857e-06, "sft_loss": 0.1057, "step": 333 }, { "distill_loss": 0.2282, "epoch": 0.21, "grad_norm": 0.9194339158084948, "learning_rate": 9.211658105728267e-06, "loss": 0.2379, "lr_vision_tower": 1.8423316211456532e-06, "sft_loss": 0.1237, "step": 334 }, { "distill_loss": 0.2266, "epoch": 0.21, "grad_norm": 0.9322242327404103, "learning_rate": 9.206267664155906e-06, "loss": 0.2265, "lr_vision_tower": 1.8412535328311812e-06, "sft_loss": 0.1132, "step": 335 }, { "distill_loss": 0.2359, "epoch": 0.21, "grad_norm": 0.9086805313901263, "learning_rate": 9.200860444627158e-06, "loss": 0.2498, "lr_vision_tower": 1.8401720889254313e-06, "sft_loss": 0.1318, "step": 336 }, { "distill_loss": 0.2236, "epoch": 0.21, "grad_norm": 0.9766256616231083, "learning_rate": 9.195436468710335e-06, "loss": 0.2337, "lr_vision_tower": 1.8390872937420668e-06, "sft_loss": 0.1219, "step": 337 }, { "distill_loss": 0.2317, "epoch": 0.21, "grad_norm": 1.1932247545761874, "learning_rate": 9.189995758040584e-06, "loss": 0.2293, "lr_vision_tower": 1.8379991516081165e-06, "sft_loss": 0.1135, "step": 338 }, { "distill_loss": 0.2245, "epoch": 0.21, "grad_norm": 0.903879932014474, "learning_rate": 9.18453833431981e-06, "loss": 0.2287, "lr_vision_tower": 1.8369076668639618e-06, "sft_loss": 0.1164, "step": 339 }, { "distill_loss": 0.2216, "epoch": 0.21, "grad_norm": 0.7593337386314678, "learning_rate": 9.179064219316578e-06, "loss": 0.2168, "lr_vision_tower": 1.8358128438633154e-06, "sft_loss": 0.106, "step": 340 }, { "distill_loss": 0.2283, "epoch": 0.21, "grad_norm": 0.7282397367403943, "learning_rate": 9.173573434866032e-06, "loss": 0.2165, "lr_vision_tower": 1.8347146869732062e-06, "sft_loss": 0.1023, "step": 341 }, { "distill_loss": 0.2225, "epoch": 0.21, "grad_norm": 0.7679713934060411, "learning_rate": 9.168066002869808e-06, "loss": 0.2202, "lr_vision_tower": 1.8336132005739615e-06, "sft_loss": 0.109, "step": 342 }, { "distill_loss": 0.2235, "epoch": 0.21, "grad_norm": 0.9506068538311745, "learning_rate": 9.162541945295948e-06, "loss": 0.2243, "lr_vision_tower": 1.8325083890591895e-06, "sft_loss": 0.1125, "step": 343 }, { "distill_loss": 0.2112, "epoch": 0.21, "grad_norm": 0.9046658006046274, "learning_rate": 9.157001284178803e-06, "loss": 0.2042, "lr_vision_tower": 1.8314002568357605e-06, "sft_loss": 0.0986, "step": 344 }, { "distill_loss": 0.2253, "epoch": 0.21, "grad_norm": 1.033040906638139, "learning_rate": 9.15144404161896e-06, "loss": 0.2249, "lr_vision_tower": 1.8302888083237918e-06, "sft_loss": 0.1122, "step": 345 }, { "distill_loss": 0.2245, "epoch": 0.21, "grad_norm": 0.9927843831377746, "learning_rate": 9.145870239783143e-06, "loss": 0.2163, "lr_vision_tower": 1.8291740479566282e-06, "sft_loss": 0.1041, "step": 346 }, { "distill_loss": 0.2335, "epoch": 0.21, "grad_norm": 1.216118619867595, "learning_rate": 9.140279900904123e-06, "loss": 0.2209, "lr_vision_tower": 1.8280559801808245e-06, "sft_loss": 0.1042, "step": 347 }, { "distill_loss": 0.2204, "epoch": 0.21, "grad_norm": 1.343150931063156, "learning_rate": 9.134673047280644e-06, "loss": 0.2229, "lr_vision_tower": 1.8269346094561286e-06, "sft_loss": 0.1127, "step": 348 }, { "distill_loss": 0.2243, "epoch": 0.22, "grad_norm": 1.5458267501839082, "learning_rate": 9.129049701277314e-06, "loss": 0.217, "lr_vision_tower": 1.8258099402554628e-06, "sft_loss": 0.1048, "step": 349 }, { "distill_loss": 0.2264, "epoch": 0.22, "grad_norm": 1.064013929904929, "learning_rate": 9.123409885324535e-06, "loss": 0.2243, "lr_vision_tower": 1.8246819770649067e-06, "sft_loss": 0.1111, "step": 350 }, { "distill_loss": 0.2227, "epoch": 0.22, "grad_norm": 1.557688420820373, "learning_rate": 9.117753621918392e-06, "loss": 0.2196, "lr_vision_tower": 1.8235507243836784e-06, "sft_loss": 0.1082, "step": 351 }, { "distill_loss": 0.2232, "epoch": 0.22, "grad_norm": 1.4481005478533517, "learning_rate": 9.112080933620591e-06, "loss": 0.2276, "lr_vision_tower": 1.8224161867241178e-06, "sft_loss": 0.116, "step": 352 }, { "distill_loss": 0.2205, "epoch": 0.22, "grad_norm": 1.0289901391122709, "learning_rate": 9.10639184305834e-06, "loss": 0.2245, "lr_vision_tower": 1.8212783686116678e-06, "sft_loss": 0.1142, "step": 353 }, { "distill_loss": 0.2299, "epoch": 0.22, "grad_norm": 0.8646658932296283, "learning_rate": 9.100686372924279e-06, "loss": 0.2168, "lr_vision_tower": 1.8201372745848556e-06, "sft_loss": 0.1019, "step": 354 }, { "distill_loss": 0.2388, "epoch": 0.22, "grad_norm": 0.8011338433327475, "learning_rate": 9.094964545976384e-06, "loss": 0.2344, "lr_vision_tower": 1.8189929091952765e-06, "sft_loss": 0.115, "step": 355 }, { "distill_loss": 0.2217, "epoch": 0.22, "grad_norm": 0.9706174835976352, "learning_rate": 9.08922638503787e-06, "loss": 0.2236, "lr_vision_tower": 1.8178452770075738e-06, "sft_loss": 0.1128, "step": 356 }, { "distill_loss": 0.22, "epoch": 0.22, "grad_norm": 0.8834272756368622, "learning_rate": 9.08347191299711e-06, "loss": 0.2169, "lr_vision_tower": 1.8166943825994216e-06, "sft_loss": 0.1069, "step": 357 }, { "distill_loss": 0.2138, "epoch": 0.22, "grad_norm": 0.9920786181091361, "learning_rate": 9.077701152807534e-06, "loss": 0.211, "lr_vision_tower": 1.8155402305615067e-06, "sft_loss": 0.1041, "step": 358 }, { "distill_loss": 0.2361, "epoch": 0.22, "grad_norm": 0.7862768121710332, "learning_rate": 9.07191412748755e-06, "loss": 0.2282, "lr_vision_tower": 1.8143828254975095e-06, "sft_loss": 0.1102, "step": 359 }, { "distill_loss": 0.2324, "epoch": 0.22, "grad_norm": 1.023693369347985, "learning_rate": 9.066110860120431e-06, "loss": 0.2241, "lr_vision_tower": 1.813222172024086e-06, "sft_loss": 0.1079, "step": 360 }, { "distill_loss": 0.2236, "epoch": 0.22, "grad_norm": 0.7103326472888367, "learning_rate": 9.060291373854252e-06, "loss": 0.2121, "lr_vision_tower": 1.81205827477085e-06, "sft_loss": 0.1003, "step": 361 }, { "distill_loss": 0.2224, "epoch": 0.22, "grad_norm": 0.797171092850119, "learning_rate": 9.054455691901768e-06, "loss": 0.2127, "lr_vision_tower": 1.8108911383803533e-06, "sft_loss": 0.1015, "step": 362 }, { "distill_loss": 0.2194, "epoch": 0.22, "grad_norm": 1.221070585006343, "learning_rate": 9.048603837540346e-06, "loss": 0.2123, "lr_vision_tower": 1.8097207675080688e-06, "sft_loss": 0.1026, "step": 363 }, { "distill_loss": 0.2305, "epoch": 0.22, "grad_norm": 0.8655150498142393, "learning_rate": 9.042735834111852e-06, "loss": 0.2207, "lr_vision_tower": 1.8085471668223704e-06, "sft_loss": 0.1054, "step": 364 }, { "distill_loss": 0.226, "epoch": 0.22, "grad_norm": 1.107533910299313, "learning_rate": 9.03685170502258e-06, "loss": 0.1997, "lr_vision_tower": 1.8073703410045159e-06, "sft_loss": 0.0867, "step": 365 }, { "distill_loss": 0.2247, "epoch": 0.23, "grad_norm": 1.1696393940900216, "learning_rate": 9.030951473743132e-06, "loss": 0.2246, "lr_vision_tower": 1.8061902947486263e-06, "sft_loss": 0.1122, "step": 366 }, { "distill_loss": 0.2191, "epoch": 0.23, "grad_norm": 1.0967545134842234, "learning_rate": 9.025035163808347e-06, "loss": 0.2153, "lr_vision_tower": 1.8050070327616691e-06, "sft_loss": 0.1057, "step": 367 }, { "distill_loss": 0.2262, "epoch": 0.23, "grad_norm": 1.146422041527609, "learning_rate": 9.019102798817196e-06, "loss": 0.2331, "lr_vision_tower": 1.803820559763439e-06, "sft_loss": 0.12, "step": 368 }, { "distill_loss": 0.2226, "epoch": 0.23, "grad_norm": 0.943571583650174, "learning_rate": 9.013154402432691e-06, "loss": 0.2121, "lr_vision_tower": 1.802630880486538e-06, "sft_loss": 0.1008, "step": 369 }, { "distill_loss": 0.2254, "epoch": 0.23, "grad_norm": 0.7207391283628878, "learning_rate": 9.007189998381791e-06, "loss": 0.2155, "lr_vision_tower": 1.801437999676358e-06, "sft_loss": 0.1028, "step": 370 }, { "distill_loss": 0.2238, "epoch": 0.23, "grad_norm": 1.2317398489966236, "learning_rate": 9.0012096104553e-06, "loss": 0.2102, "lr_vision_tower": 1.80024192209106e-06, "sft_loss": 0.0983, "step": 371 }, { "distill_loss": 0.2252, "epoch": 0.23, "grad_norm": 0.8351601209906466, "learning_rate": 8.995213262507793e-06, "loss": 0.2073, "lr_vision_tower": 1.7990426525015583e-06, "sft_loss": 0.0947, "step": 372 }, { "distill_loss": 0.219, "epoch": 0.23, "grad_norm": 0.9421251722628136, "learning_rate": 8.989200978457489e-06, "loss": 0.2187, "lr_vision_tower": 1.7978401956914976e-06, "sft_loss": 0.1092, "step": 373 }, { "distill_loss": 0.2128, "epoch": 0.23, "grad_norm": 2.7157122322128138, "learning_rate": 8.98317278228618e-06, "loss": 0.2133, "lr_vision_tower": 1.7966345564572359e-06, "sft_loss": 0.107, "step": 374 }, { "distill_loss": 0.2223, "epoch": 0.23, "grad_norm": 1.4680905381036666, "learning_rate": 8.977128698039134e-06, "loss": 0.2131, "lr_vision_tower": 1.7954257396078267e-06, "sft_loss": 0.102, "step": 375 }, { "distill_loss": 0.2154, "epoch": 0.23, "grad_norm": 1.1382786272644834, "learning_rate": 8.971068749824988e-06, "loss": 0.215, "lr_vision_tower": 1.7942137499649977e-06, "sft_loss": 0.1073, "step": 376 }, { "distill_loss": 0.2217, "epoch": 0.23, "grad_norm": 0.9964987944012388, "learning_rate": 8.964992961815655e-06, "loss": 0.2241, "lr_vision_tower": 1.792998592363131e-06, "sft_loss": 0.1132, "step": 377 }, { "distill_loss": 0.2199, "epoch": 0.23, "grad_norm": 0.9891989081513616, "learning_rate": 8.958901358246235e-06, "loss": 0.2164, "lr_vision_tower": 1.7917802716492468e-06, "sft_loss": 0.1065, "step": 378 }, { "distill_loss": 0.2162, "epoch": 0.23, "grad_norm": 2.284145771821057, "learning_rate": 8.952793963414908e-06, "loss": 0.2064, "lr_vision_tower": 1.7905587926829812e-06, "sft_loss": 0.0983, "step": 379 }, { "distill_loss": 0.2268, "epoch": 0.23, "grad_norm": 1.1403357866158617, "learning_rate": 8.946670801682844e-06, "loss": 0.217, "lr_vision_tower": 1.7893341603365685e-06, "sft_loss": 0.1036, "step": 380 }, { "distill_loss": 0.225, "epoch": 0.23, "grad_norm": 1.190233455703026, "learning_rate": 8.940531897474107e-06, "loss": 0.2133, "lr_vision_tower": 1.788106379494821e-06, "sft_loss": 0.1008, "step": 381 }, { "distill_loss": 0.2269, "epoch": 0.24, "grad_norm": 1.4618852665392612, "learning_rate": 8.93437727527555e-06, "loss": 0.2265, "lr_vision_tower": 1.7868754550551097e-06, "sft_loss": 0.113, "step": 382 }, { "distill_loss": 0.2293, "epoch": 0.24, "grad_norm": 0.9350412116859212, "learning_rate": 8.928206959636726e-06, "loss": 0.2317, "lr_vision_tower": 1.7856413919273452e-06, "sft_loss": 0.117, "step": 383 }, { "distill_loss": 0.2251, "epoch": 0.24, "grad_norm": 0.8659168169179373, "learning_rate": 8.922020975169784e-06, "loss": 0.2126, "lr_vision_tower": 1.7844041950339566e-06, "sft_loss": 0.1, "step": 384 }, { "distill_loss": 0.2187, "epoch": 0.24, "grad_norm": 0.8505525985345926, "learning_rate": 8.915819346549372e-06, "loss": 0.2117, "lr_vision_tower": 1.7831638693098743e-06, "sft_loss": 0.1023, "step": 385 }, { "distill_loss": 0.2287, "epoch": 0.24, "grad_norm": 1.0164771535393244, "learning_rate": 8.90960209851254e-06, "loss": 0.2166, "lr_vision_tower": 1.7819204197025076e-06, "sft_loss": 0.1022, "step": 386 }, { "distill_loss": 0.2186, "epoch": 0.24, "grad_norm": 0.9160170931241128, "learning_rate": 8.90336925585864e-06, "loss": 0.2157, "lr_vision_tower": 1.780673851171728e-06, "sft_loss": 0.1064, "step": 387 }, { "distill_loss": 0.2209, "epoch": 0.24, "grad_norm": 0.8597110281647385, "learning_rate": 8.897120843449232e-06, "loss": 0.2063, "lr_vision_tower": 1.7794241686898461e-06, "sft_loss": 0.0958, "step": 388 }, { "distill_loss": 0.2178, "epoch": 0.24, "grad_norm": 1.0820136516526466, "learning_rate": 8.890856886207975e-06, "loss": 0.205, "lr_vision_tower": 1.7781713772415948e-06, "sft_loss": 0.0961, "step": 389 }, { "distill_loss": 0.2197, "epoch": 0.24, "grad_norm": 0.8801723455168166, "learning_rate": 8.884577409120535e-06, "loss": 0.206, "lr_vision_tower": 1.776915481824107e-06, "sft_loss": 0.0961, "step": 390 }, { "distill_loss": 0.2236, "epoch": 0.24, "grad_norm": 1.1568483908911649, "learning_rate": 8.878282437234487e-06, "loss": 0.2163, "lr_vision_tower": 1.7756564874468971e-06, "sft_loss": 0.1045, "step": 391 }, { "distill_loss": 0.2126, "epoch": 0.24, "grad_norm": 0.748448282844724, "learning_rate": 8.871971995659206e-06, "loss": 0.2139, "lr_vision_tower": 1.7743943991318411e-06, "sft_loss": 0.1076, "step": 392 }, { "distill_loss": 0.2232, "epoch": 0.24, "grad_norm": 0.7606432315320266, "learning_rate": 8.865646109565778e-06, "loss": 0.2166, "lr_vision_tower": 1.7731292219131553e-06, "sft_loss": 0.105, "step": 393 }, { "distill_loss": 0.2275, "epoch": 0.24, "grad_norm": 1.0240955101539557, "learning_rate": 8.859304804186886e-06, "loss": 0.2097, "lr_vision_tower": 1.7718609608373772e-06, "sft_loss": 0.096, "step": 394 }, { "distill_loss": 0.2313, "epoch": 0.24, "grad_norm": 1.14694454506183, "learning_rate": 8.852948104816728e-06, "loss": 0.2321, "lr_vision_tower": 1.7705896209633455e-06, "sft_loss": 0.1165, "step": 395 }, { "distill_loss": 0.2091, "epoch": 0.24, "grad_norm": 1.0679829778369156, "learning_rate": 8.846576036810899e-06, "loss": 0.2087, "lr_vision_tower": 1.7693152073621796e-06, "sft_loss": 0.1041, "step": 396 }, { "distill_loss": 0.2194, "epoch": 0.24, "grad_norm": 0.7167074015810431, "learning_rate": 8.840188625586296e-06, "loss": 0.2122, "lr_vision_tower": 1.7680377251172592e-06, "sft_loss": 0.1025, "step": 397 }, { "distill_loss": 0.1999, "epoch": 0.25, "grad_norm": 0.8240403027498286, "learning_rate": 8.833785896621021e-06, "loss": 0.2005, "lr_vision_tower": 1.7667571793242041e-06, "sft_loss": 0.1005, "step": 398 }, { "distill_loss": 0.224, "epoch": 0.25, "grad_norm": 1.0907312761349706, "learning_rate": 8.827367875454273e-06, "loss": 0.2075, "lr_vision_tower": 1.7654735750908545e-06, "sft_loss": 0.0955, "step": 399 }, { "distill_loss": 0.2156, "epoch": 0.25, "grad_norm": 0.9419513187280193, "learning_rate": 8.820934587686247e-06, "loss": 0.2017, "lr_vision_tower": 1.764186917537249e-06, "sft_loss": 0.0939, "step": 400 }, { "distill_loss": 0.2258, "epoch": 0.25, "grad_norm": 0.8997349413538516, "learning_rate": 8.814486058978035e-06, "loss": 0.2344, "lr_vision_tower": 1.7628972117956066e-06, "sft_loss": 0.1215, "step": 401 }, { "distill_loss": 0.2163, "epoch": 0.25, "grad_norm": 0.7678089610195584, "learning_rate": 8.808022315051522e-06, "loss": 0.2181, "lr_vision_tower": 1.7616044630103044e-06, "sft_loss": 0.11, "step": 402 }, { "distill_loss": 0.2281, "epoch": 0.25, "grad_norm": 0.7898553757539757, "learning_rate": 8.801543381689287e-06, "loss": 0.2202, "lr_vision_tower": 1.760308676337857e-06, "sft_loss": 0.1062, "step": 403 }, { "distill_loss": 0.227, "epoch": 0.25, "grad_norm": 0.7214205415988401, "learning_rate": 8.795049284734488e-06, "loss": 0.1965, "lr_vision_tower": 1.7590098569468973e-06, "sft_loss": 0.083, "step": 404 }, { "distill_loss": 0.2236, "epoch": 0.25, "grad_norm": 0.8255577410020264, "learning_rate": 8.788540050090776e-06, "loss": 0.2171, "lr_vision_tower": 1.7577080100181548e-06, "sft_loss": 0.1053, "step": 405 }, { "distill_loss": 0.2132, "epoch": 0.25, "grad_norm": 1.0251250975260526, "learning_rate": 8.782015703722178e-06, "loss": 0.2178, "lr_vision_tower": 1.7564031407444354e-06, "sft_loss": 0.1112, "step": 406 }, { "distill_loss": 0.2144, "epoch": 0.25, "grad_norm": 0.9448575222628759, "learning_rate": 8.775476271653002e-06, "loss": 0.2287, "lr_vision_tower": 1.7550952543306004e-06, "sft_loss": 0.1214, "step": 407 }, { "distill_loss": 0.2251, "epoch": 0.25, "grad_norm": 0.8594246671918357, "learning_rate": 8.768921779967729e-06, "loss": 0.2164, "lr_vision_tower": 1.7537843559935457e-06, "sft_loss": 0.1039, "step": 408 }, { "distill_loss": 0.2243, "epoch": 0.25, "grad_norm": 0.9407378087837808, "learning_rate": 8.762352254810907e-06, "loss": 0.2397, "lr_vision_tower": 1.752470450962181e-06, "sft_loss": 0.1275, "step": 409 }, { "distill_loss": 0.2276, "epoch": 0.25, "grad_norm": 0.7466674800286094, "learning_rate": 8.755767722387051e-06, "loss": 0.2069, "lr_vision_tower": 1.7511535444774101e-06, "sft_loss": 0.0931, "step": 410 }, { "distill_loss": 0.2097, "epoch": 0.25, "grad_norm": 1.02158849171583, "learning_rate": 8.749168208960541e-06, "loss": 0.2004, "lr_vision_tower": 1.749833641792108e-06, "sft_loss": 0.0955, "step": 411 }, { "distill_loss": 0.2096, "epoch": 0.25, "grad_norm": 0.945258768945806, "learning_rate": 8.742553740855507e-06, "loss": 0.2068, "lr_vision_tower": 1.748510748171101e-06, "sft_loss": 0.1019, "step": 412 }, { "distill_loss": 0.2129, "epoch": 0.25, "grad_norm": 1.0970106623010876, "learning_rate": 8.735924344455732e-06, "loss": 0.2083, "lr_vision_tower": 1.7471848688911464e-06, "sft_loss": 0.1019, "step": 413 }, { "distill_loss": 0.2359, "epoch": 0.26, "grad_norm": 0.8714902231179795, "learning_rate": 8.72928004620455e-06, "loss": 0.2341, "lr_vision_tower": 1.7458560092409096e-06, "sft_loss": 0.1162, "step": 414 }, { "distill_loss": 0.2316, "epoch": 0.26, "grad_norm": 1.3521247546594524, "learning_rate": 8.722620872604728e-06, "loss": 0.2352, "lr_vision_tower": 1.7445241745209454e-06, "sft_loss": 0.1194, "step": 415 }, { "distill_loss": 0.2249, "epoch": 0.26, "grad_norm": 0.9995615164465258, "learning_rate": 8.715946850218374e-06, "loss": 0.2335, "lr_vision_tower": 1.7431893700436744e-06, "sft_loss": 0.121, "step": 416 }, { "distill_loss": 0.2118, "epoch": 0.26, "grad_norm": 0.7639893712231927, "learning_rate": 8.709258005666816e-06, "loss": 0.2163, "lr_vision_tower": 1.741851601133363e-06, "sft_loss": 0.1104, "step": 417 }, { "distill_loss": 0.2043, "epoch": 0.26, "grad_norm": 0.8556234366190011, "learning_rate": 8.702554365630518e-06, "loss": 0.189, "lr_vision_tower": 1.7405108731261034e-06, "sft_loss": 0.0868, "step": 418 }, { "distill_loss": 0.2123, "epoch": 0.26, "grad_norm": 0.8036043261633817, "learning_rate": 8.695835956848951e-06, "loss": 0.2077, "lr_vision_tower": 1.73916719136979e-06, "sft_loss": 0.1016, "step": 419 }, { "distill_loss": 0.2173, "epoch": 0.26, "grad_norm": 0.714607789102604, "learning_rate": 8.689102806120494e-06, "loss": 0.2217, "lr_vision_tower": 1.7378205612240985e-06, "sft_loss": 0.113, "step": 420 }, { "distill_loss": 0.2279, "epoch": 0.26, "grad_norm": 0.8141998118819861, "learning_rate": 8.682354940302336e-06, "loss": 0.218, "lr_vision_tower": 1.736470988060467e-06, "sft_loss": 0.1041, "step": 421 }, { "distill_loss": 0.2219, "epoch": 0.26, "grad_norm": 0.8825130656324073, "learning_rate": 8.675592386310353e-06, "loss": 0.2187, "lr_vision_tower": 1.7351184772620706e-06, "sft_loss": 0.1078, "step": 422 }, { "distill_loss": 0.2146, "epoch": 0.26, "grad_norm": 0.8504089579942838, "learning_rate": 8.66881517111902e-06, "loss": 0.2016, "lr_vision_tower": 1.7337630342238039e-06, "sft_loss": 0.0942, "step": 423 }, { "distill_loss": 0.2208, "epoch": 0.26, "grad_norm": 0.8349500587437226, "learning_rate": 8.662023321761283e-06, "loss": 0.2437, "lr_vision_tower": 1.7324046643522566e-06, "sft_loss": 0.1333, "step": 424 }, { "distill_loss": 0.202, "epoch": 0.26, "grad_norm": 1.0255690037817975, "learning_rate": 8.655216865328464e-06, "loss": 0.2084, "lr_vision_tower": 1.7310433730656926e-06, "sft_loss": 0.1074, "step": 425 }, { "distill_loss": 0.219, "epoch": 0.26, "grad_norm": 0.8506003255140733, "learning_rate": 8.64839582897015e-06, "loss": 0.2191, "lr_vision_tower": 1.72967916579403e-06, "sft_loss": 0.1096, "step": 426 }, { "distill_loss": 0.226, "epoch": 0.26, "grad_norm": 0.674764938804221, "learning_rate": 8.641560239894084e-06, "loss": 0.2101, "lr_vision_tower": 1.7283120479788167e-06, "sft_loss": 0.0971, "step": 427 }, { "distill_loss": 0.2033, "epoch": 0.26, "grad_norm": 0.7905665545309514, "learning_rate": 8.63471012536606e-06, "loss": 0.1958, "lr_vision_tower": 1.7269420250732116e-06, "sft_loss": 0.0941, "step": 428 }, { "distill_loss": 0.2186, "epoch": 0.26, "grad_norm": 0.8456330417968022, "learning_rate": 8.627845512709804e-06, "loss": 0.2077, "lr_vision_tower": 1.7255691025419605e-06, "sft_loss": 0.0984, "step": 429 }, { "distill_loss": 0.2207, "epoch": 0.27, "grad_norm": 0.8425581469817917, "learning_rate": 8.620966429306879e-06, "loss": 0.2098, "lr_vision_tower": 1.7241932858613754e-06, "sft_loss": 0.0994, "step": 430 }, { "distill_loss": 0.2198, "epoch": 0.27, "grad_norm": 1.129203605904359, "learning_rate": 8.614072902596563e-06, "loss": 0.2117, "lr_vision_tower": 1.7228145805193123e-06, "sft_loss": 0.1018, "step": 431 }, { "distill_loss": 0.2174, "epoch": 0.27, "grad_norm": 1.6075787345601067, "learning_rate": 8.607164960075753e-06, "loss": 0.2158, "lr_vision_tower": 1.7214329920151504e-06, "sft_loss": 0.1071, "step": 432 }, { "distill_loss": 0.2164, "epoch": 0.27, "grad_norm": 1.3906835552366454, "learning_rate": 8.60024262929884e-06, "loss": 0.232, "lr_vision_tower": 1.7200485258597679e-06, "sft_loss": 0.1238, "step": 433 }, { "distill_loss": 0.2159, "epoch": 0.27, "grad_norm": 0.7891599533384486, "learning_rate": 8.593305937877614e-06, "loss": 0.2073, "lr_vision_tower": 1.7186611875755225e-06, "sft_loss": 0.0993, "step": 434 }, { "distill_loss": 0.2337, "epoch": 0.27, "grad_norm": 1.3079798971520384, "learning_rate": 8.58635491348114e-06, "loss": 0.2297, "lr_vision_tower": 1.7172709826962275e-06, "sft_loss": 0.1129, "step": 435 }, { "distill_loss": 0.2173, "epoch": 0.27, "grad_norm": 0.8681854741640475, "learning_rate": 8.579389583835656e-06, "loss": 0.2087, "lr_vision_tower": 1.715877916767131e-06, "sft_loss": 0.1, "step": 436 }, { "distill_loss": 0.2367, "epoch": 0.27, "grad_norm": 0.793078205931958, "learning_rate": 8.572409976724466e-06, "loss": 0.2109, "lr_vision_tower": 1.714481995344893e-06, "sft_loss": 0.0925, "step": 437 }, { "distill_loss": 0.2091, "epoch": 0.27, "grad_norm": 0.6930946719438132, "learning_rate": 8.565416119987819e-06, "loss": 0.1976, "lr_vision_tower": 1.7130832239975634e-06, "sft_loss": 0.0931, "step": 438 }, { "distill_loss": 0.2198, "epoch": 0.27, "grad_norm": 0.8838733818464155, "learning_rate": 8.558408041522801e-06, "loss": 0.209, "lr_vision_tower": 1.71168160830456e-06, "sft_loss": 0.0991, "step": 439 }, { "distill_loss": 0.2101, "epoch": 0.27, "grad_norm": 0.8735436116722188, "learning_rate": 8.551385769283234e-06, "loss": 0.2009, "lr_vision_tower": 1.7102771538566465e-06, "sft_loss": 0.0959, "step": 440 }, { "distill_loss": 0.2154, "epoch": 0.27, "grad_norm": 1.1093692119124672, "learning_rate": 8.544349331279545e-06, "loss": 0.1916, "lr_vision_tower": 1.708869866255909e-06, "sft_loss": 0.0839, "step": 441 }, { "distill_loss": 0.2141, "epoch": 0.27, "grad_norm": 0.8393124990654838, "learning_rate": 8.537298755578676e-06, "loss": 0.218, "lr_vision_tower": 1.707459751115735e-06, "sft_loss": 0.1109, "step": 442 }, { "distill_loss": 0.214, "epoch": 0.27, "grad_norm": 1.7586287655486694, "learning_rate": 8.530234070303953e-06, "loss": 0.2034, "lr_vision_tower": 1.7060468140607903e-06, "sft_loss": 0.0964, "step": 443 }, { "distill_loss": 0.213, "epoch": 0.27, "grad_norm": 0.9096352982632909, "learning_rate": 8.523155303634986e-06, "loss": 0.2205, "lr_vision_tower": 1.7046310607269971e-06, "sft_loss": 0.114, "step": 444 }, { "distill_loss": 0.2048, "epoch": 0.27, "grad_norm": 0.8566988816951816, "learning_rate": 8.516062483807556e-06, "loss": 0.2067, "lr_vision_tower": 1.7032124967615109e-06, "sft_loss": 0.1043, "step": 445 }, { "distill_loss": 0.2076, "epoch": 0.27, "grad_norm": 1.190111093799004, "learning_rate": 8.508955639113492e-06, "loss": 0.2073, "lr_vision_tower": 1.7017911278226982e-06, "sft_loss": 0.1035, "step": 446 }, { "distill_loss": 0.2084, "epoch": 0.28, "grad_norm": 0.898428648406582, "learning_rate": 8.501834797900573e-06, "loss": 0.2112, "lr_vision_tower": 1.7003669595801142e-06, "sft_loss": 0.1071, "step": 447 }, { "distill_loss": 0.2159, "epoch": 0.28, "grad_norm": 0.9724432719694172, "learning_rate": 8.4946999885724e-06, "loss": 0.2034, "lr_vision_tower": 1.6989399977144796e-06, "sft_loss": 0.0954, "step": 448 }, { "distill_loss": 0.2101, "epoch": 0.28, "grad_norm": 0.7436390913860086, "learning_rate": 8.487551239588295e-06, "loss": 0.2131, "lr_vision_tower": 1.6975102479176586e-06, "sft_loss": 0.1081, "step": 449 }, { "distill_loss": 0.2242, "epoch": 0.28, "grad_norm": 0.8625210106471654, "learning_rate": 8.48038857946318e-06, "loss": 0.2122, "lr_vision_tower": 1.6960777158926358e-06, "sft_loss": 0.1001, "step": 450 }, { "distill_loss": 0.2028, "epoch": 0.28, "grad_norm": 0.9247830515156661, "learning_rate": 8.473212036767469e-06, "loss": 0.2173, "lr_vision_tower": 1.6946424073534936e-06, "sft_loss": 0.1159, "step": 451 }, { "distill_loss": 0.2137, "epoch": 0.28, "grad_norm": 1.0390072481724162, "learning_rate": 8.466021640126946e-06, "loss": 0.2149, "lr_vision_tower": 1.693204328025389e-06, "sft_loss": 0.1081, "step": 452 }, { "distill_loss": 0.2127, "epoch": 0.28, "grad_norm": 3.183932315354965, "learning_rate": 8.458817418222661e-06, "loss": 0.1979, "lr_vision_tower": 1.6917634836445318e-06, "sft_loss": 0.0915, "step": 453 }, { "distill_loss": 0.2144, "epoch": 0.28, "grad_norm": 0.8308589281921612, "learning_rate": 8.451599399790804e-06, "loss": 0.2098, "lr_vision_tower": 1.6903198799581606e-06, "sft_loss": 0.1026, "step": 454 }, { "distill_loss": 0.2201, "epoch": 0.28, "grad_norm": 0.7521507657524514, "learning_rate": 8.444367613622603e-06, "loss": 0.2144, "lr_vision_tower": 1.6888735227245204e-06, "sft_loss": 0.1043, "step": 455 }, { "distill_loss": 0.223, "epoch": 0.28, "grad_norm": 0.8183185074909735, "learning_rate": 8.437122088564197e-06, "loss": 0.1966, "lr_vision_tower": 1.6874244177128393e-06, "sft_loss": 0.0851, "step": 456 }, { "distill_loss": 0.2179, "epoch": 0.28, "grad_norm": 0.7246669961225403, "learning_rate": 8.429862853516534e-06, "loss": 0.2134, "lr_vision_tower": 1.6859725707033063e-06, "sft_loss": 0.1044, "step": 457 }, { "distill_loss": 0.2119, "epoch": 0.28, "grad_norm": 0.903827044943077, "learning_rate": 8.422589937435237e-06, "loss": 0.2165, "lr_vision_tower": 1.6845179874870472e-06, "sft_loss": 0.1105, "step": 458 }, { "distill_loss": 0.2002, "epoch": 0.28, "grad_norm": 0.9139588436468348, "learning_rate": 8.415303369330513e-06, "loss": 0.1882, "lr_vision_tower": 1.6830606738661022e-06, "sft_loss": 0.0881, "step": 459 }, { "distill_loss": 0.211, "epoch": 0.28, "grad_norm": 1.0178973969197451, "learning_rate": 8.408003178267011e-06, "loss": 0.2034, "lr_vision_tower": 1.681600635653402e-06, "sft_loss": 0.0979, "step": 460 }, { "distill_loss": 0.2158, "epoch": 0.28, "grad_norm": 0.8126393182315377, "learning_rate": 8.400689393363733e-06, "loss": 0.2013, "lr_vision_tower": 1.6801378786727462e-06, "sft_loss": 0.0934, "step": 461 }, { "distill_loss": 0.2128, "epoch": 0.28, "grad_norm": 1.1002493710413404, "learning_rate": 8.393362043793893e-06, "loss": 0.2234, "lr_vision_tower": 1.6786724087587782e-06, "sft_loss": 0.1171, "step": 462 }, { "distill_loss": 0.214, "epoch": 0.29, "grad_norm": 1.0639613802546144, "learning_rate": 8.386021158784816e-06, "loss": 0.2055, "lr_vision_tower": 1.677204231756963e-06, "sft_loss": 0.0984, "step": 463 }, { "distill_loss": 0.21, "epoch": 0.29, "grad_norm": 0.692900507588271, "learning_rate": 8.37866676761782e-06, "loss": 0.1932, "lr_vision_tower": 1.6757333535235638e-06, "sft_loss": 0.0882, "step": 464 }, { "distill_loss": 0.2082, "epoch": 0.29, "grad_norm": 0.96284778372826, "learning_rate": 8.371298899628091e-06, "loss": 0.2092, "lr_vision_tower": 1.674259779925618e-06, "sft_loss": 0.1052, "step": 465 }, { "distill_loss": 0.2141, "epoch": 0.29, "grad_norm": 0.9025936605986309, "learning_rate": 8.363917584204573e-06, "loss": 0.2127, "lr_vision_tower": 1.6727835168409144e-06, "sft_loss": 0.1056, "step": 466 }, { "distill_loss": 0.2156, "epoch": 0.29, "grad_norm": 0.869355473880366, "learning_rate": 8.356522850789852e-06, "loss": 0.204, "lr_vision_tower": 1.6713045701579702e-06, "sft_loss": 0.0962, "step": 467 }, { "distill_loss": 0.216, "epoch": 0.29, "grad_norm": 0.8031957833720408, "learning_rate": 8.349114728880033e-06, "loss": 0.2208, "lr_vision_tower": 1.6698229457760063e-06, "sft_loss": 0.1128, "step": 468 }, { "distill_loss": 0.2085, "epoch": 0.29, "grad_norm": 0.8141828065439758, "learning_rate": 8.341693248024623e-06, "loss": 0.2093, "lr_vision_tower": 1.6683386496049243e-06, "sft_loss": 0.1051, "step": 469 }, { "distill_loss": 0.2171, "epoch": 0.29, "grad_norm": 0.970917585496971, "learning_rate": 8.33425843782642e-06, "loss": 0.2375, "lr_vision_tower": 1.6668516875652838e-06, "sft_loss": 0.1289, "step": 470 }, { "distill_loss": 0.2182, "epoch": 0.29, "grad_norm": 0.9280046803358672, "learning_rate": 8.326810327941388e-06, "loss": 0.2151, "lr_vision_tower": 1.6653620655882774e-06, "sft_loss": 0.106, "step": 471 }, { "distill_loss": 0.2194, "epoch": 0.29, "grad_norm": 1.0900163352249725, "learning_rate": 8.319348948078538e-06, "loss": 0.2179, "lr_vision_tower": 1.6638697896157075e-06, "sft_loss": 0.1082, "step": 472 }, { "distill_loss": 0.2225, "epoch": 0.29, "grad_norm": 0.697853694080885, "learning_rate": 8.311874327999819e-06, "loss": 0.2166, "lr_vision_tower": 1.6623748655999633e-06, "sft_loss": 0.1054, "step": 473 }, { "distill_loss": 0.2222, "epoch": 0.29, "grad_norm": 0.7443749769240077, "learning_rate": 8.304386497519985e-06, "loss": 0.2239, "lr_vision_tower": 1.6608772995039969e-06, "sft_loss": 0.1129, "step": 474 }, { "distill_loss": 0.2149, "epoch": 0.29, "grad_norm": 0.7779366727368873, "learning_rate": 8.296885486506488e-06, "loss": 0.2108, "lr_vision_tower": 1.6593770973012972e-06, "sft_loss": 0.1033, "step": 475 }, { "distill_loss": 0.2097, "epoch": 0.29, "grad_norm": 1.0885642414881864, "learning_rate": 8.289371324879351e-06, "loss": 0.2199, "lr_vision_tower": 1.6578742649758701e-06, "sft_loss": 0.1151, "step": 476 }, { "distill_loss": 0.2208, "epoch": 0.29, "grad_norm": 0.9189255459470639, "learning_rate": 8.281844042611061e-06, "loss": 0.2108, "lr_vision_tower": 1.6563688085222118e-06, "sft_loss": 0.1004, "step": 477 }, { "distill_loss": 0.1981, "epoch": 0.29, "grad_norm": 1.1625081923511598, "learning_rate": 8.274303669726427e-06, "loss": 0.1777, "lr_vision_tower": 1.6548607339452852e-06, "sft_loss": 0.0786, "step": 478 }, { "distill_loss": 0.2085, "epoch": 0.3, "grad_norm": 0.9631967528353572, "learning_rate": 8.266750236302484e-06, "loss": 0.217, "lr_vision_tower": 1.6533500472604966e-06, "sft_loss": 0.1128, "step": 479 }, { "distill_loss": 0.2058, "epoch": 0.3, "grad_norm": 0.9524614122075103, "learning_rate": 8.259183772468361e-06, "loss": 0.2137, "lr_vision_tower": 1.651836754493672e-06, "sft_loss": 0.1108, "step": 480 }, { "distill_loss": 0.2088, "epoch": 0.3, "grad_norm": 1.1877741784953724, "learning_rate": 8.251604308405162e-06, "loss": 0.2014, "lr_vision_tower": 1.650320861681032e-06, "sft_loss": 0.097, "step": 481 }, { "distill_loss": 0.2045, "epoch": 0.3, "grad_norm": 0.6939714651515536, "learning_rate": 8.244011874345843e-06, "loss": 0.1893, "lr_vision_tower": 1.6488023748691684e-06, "sft_loss": 0.087, "step": 482 }, { "distill_loss": 0.2146, "epoch": 0.3, "grad_norm": 0.9197311737850303, "learning_rate": 8.2364065005751e-06, "loss": 0.2231, "lr_vision_tower": 1.64728130011502e-06, "sft_loss": 0.1157, "step": 483 }, { "distill_loss": 0.2087, "epoch": 0.3, "grad_norm": 0.8559155093103812, "learning_rate": 8.22878821742924e-06, "loss": 0.2033, "lr_vision_tower": 1.6457576434858477e-06, "sft_loss": 0.099, "step": 484 }, { "distill_loss": 0.2195, "epoch": 0.3, "grad_norm": 0.9798339222615887, "learning_rate": 8.221157055296063e-06, "loss": 0.23, "lr_vision_tower": 1.6442314110592125e-06, "sft_loss": 0.1202, "step": 485 }, { "distill_loss": 0.2111, "epoch": 0.3, "grad_norm": 0.9006282320211869, "learning_rate": 8.213513044614742e-06, "loss": 0.203, "lr_vision_tower": 1.6427026089229482e-06, "sft_loss": 0.0974, "step": 486 }, { "distill_loss": 0.2085, "epoch": 0.3, "grad_norm": 1.4518010539611907, "learning_rate": 8.205856215875698e-06, "loss": 0.1961, "lr_vision_tower": 1.6411712431751393e-06, "sft_loss": 0.0918, "step": 487 }, { "distill_loss": 0.2144, "epoch": 0.3, "grad_norm": 0.9351468434002026, "learning_rate": 8.198186599620482e-06, "loss": 0.2127, "lr_vision_tower": 1.6396373199240963e-06, "sft_loss": 0.1055, "step": 488 }, { "distill_loss": 0.1994, "epoch": 0.3, "grad_norm": 0.9621868549105707, "learning_rate": 8.190504226441654e-06, "loss": 0.1884, "lr_vision_tower": 1.6381008452883307e-06, "sft_loss": 0.0887, "step": 489 }, { "distill_loss": 0.2015, "epoch": 0.3, "grad_norm": 1.1392862565247084, "learning_rate": 8.182809126982657e-06, "loss": 0.187, "lr_vision_tower": 1.6365618253965309e-06, "sft_loss": 0.0862, "step": 490 }, { "distill_loss": 0.2079, "epoch": 0.3, "grad_norm": 0.8966572234785695, "learning_rate": 8.175101331937692e-06, "loss": 0.1855, "lr_vision_tower": 1.6350202663875384e-06, "sft_loss": 0.0816, "step": 491 }, { "distill_loss": 0.2063, "epoch": 0.3, "grad_norm": 1.0561503476257386, "learning_rate": 8.16738087205161e-06, "loss": 0.1964, "lr_vision_tower": 1.6334761744103217e-06, "sft_loss": 0.0932, "step": 492 }, { "distill_loss": 0.2104, "epoch": 0.3, "grad_norm": 1.0050430140359585, "learning_rate": 8.159647778119769e-06, "loss": 0.2239, "lr_vision_tower": 1.6319295556239534e-06, "sft_loss": 0.1188, "step": 493 }, { "distill_loss": 0.2051, "epoch": 0.3, "grad_norm": 0.9476781384291293, "learning_rate": 8.15190208098793e-06, "loss": 0.1972, "lr_vision_tower": 1.6303804161975857e-06, "sft_loss": 0.0946, "step": 494 }, { "distill_loss": 0.2151, "epoch": 0.31, "grad_norm": 1.021505269096055, "learning_rate": 8.14414381155212e-06, "loss": 0.2077, "lr_vision_tower": 1.6288287623104235e-06, "sft_loss": 0.1002, "step": 495 }, { "distill_loss": 0.2105, "epoch": 0.31, "grad_norm": 0.8474777626838097, "learning_rate": 8.136373000758518e-06, "loss": 0.2129, "lr_vision_tower": 1.6272746001517032e-06, "sft_loss": 0.1076, "step": 496 }, { "distill_loss": 0.2118, "epoch": 0.31, "grad_norm": 1.600802157534277, "learning_rate": 8.128589679603323e-06, "loss": 0.203, "lr_vision_tower": 1.6257179359206646e-06, "sft_loss": 0.0971, "step": 497 }, { "distill_loss": 0.2097, "epoch": 0.31, "grad_norm": 1.2019044331579687, "learning_rate": 8.120793879132646e-06, "loss": 0.2049, "lr_vision_tower": 1.624158775826529e-06, "sft_loss": 0.1001, "step": 498 }, { "distill_loss": 0.2205, "epoch": 0.31, "grad_norm": 1.01675495368955, "learning_rate": 8.112985630442363e-06, "loss": 0.2139, "lr_vision_tower": 1.6225971260884723e-06, "sft_loss": 0.1037, "step": 499 }, { "distill_loss": 0.2079, "epoch": 0.31, "grad_norm": 0.8930441672685282, "learning_rate": 8.105164964678009e-06, "loss": 0.2063, "lr_vision_tower": 1.6210329929356017e-06, "sft_loss": 0.1023, "step": 500 }, { "distill_loss": 0.2126, "epoch": 0.31, "grad_norm": 1.9874508118284007, "learning_rate": 8.097331913034652e-06, "loss": 0.2017, "lr_vision_tower": 1.6194663826069302e-06, "sft_loss": 0.0954, "step": 501 }, { "distill_loss": 0.2133, "epoch": 0.31, "grad_norm": 1.008950535245024, "learning_rate": 8.089486506756756e-06, "loss": 0.2214, "lr_vision_tower": 1.617897301351351e-06, "sft_loss": 0.1147, "step": 502 }, { "distill_loss": 0.2031, "epoch": 0.31, "grad_norm": 0.9997412292008583, "learning_rate": 8.081628777138073e-06, "loss": 0.2202, "lr_vision_tower": 1.6163257554276144e-06, "sft_loss": 0.1187, "step": 503 }, { "distill_loss": 0.1978, "epoch": 0.31, "grad_norm": 0.8026085139940239, "learning_rate": 8.073758755521506e-06, "loss": 0.1894, "lr_vision_tower": 1.6147517511043009e-06, "sft_loss": 0.0905, "step": 504 }, { "distill_loss": 0.2224, "epoch": 0.31, "grad_norm": 0.9325969661442544, "learning_rate": 8.06587647329899e-06, "loss": 0.2079, "lr_vision_tower": 1.6131752946597976e-06, "sft_loss": 0.0968, "step": 505 }, { "distill_loss": 0.2058, "epoch": 0.31, "grad_norm": 0.8206402620860164, "learning_rate": 8.057981961911363e-06, "loss": 0.2035, "lr_vision_tower": 1.6115963923822723e-06, "sft_loss": 0.1006, "step": 506 }, { "distill_loss": 0.1956, "epoch": 0.31, "grad_norm": 0.713762079491283, "learning_rate": 8.050075252848247e-06, "loss": 0.1971, "lr_vision_tower": 1.6100150505696494e-06, "sft_loss": 0.0993, "step": 507 }, { "distill_loss": 0.2138, "epoch": 0.31, "grad_norm": 0.8267140847307681, "learning_rate": 8.042156377647915e-06, "loss": 0.2108, "lr_vision_tower": 1.6084312755295828e-06, "sft_loss": 0.1039, "step": 508 }, { "distill_loss": 0.2037, "epoch": 0.31, "grad_norm": 0.8320301463970342, "learning_rate": 8.034225367897168e-06, "loss": 0.1924, "lr_vision_tower": 1.6068450735794334e-06, "sft_loss": 0.0905, "step": 509 }, { "distill_loss": 0.2048, "epoch": 0.31, "grad_norm": 0.876998946676875, "learning_rate": 8.02628225523121e-06, "loss": 0.1996, "lr_vision_tower": 1.6052564510462419e-06, "sft_loss": 0.0972, "step": 510 }, { "distill_loss": 0.2242, "epoch": 0.31, "grad_norm": 0.980611617891328, "learning_rate": 8.018327071333521e-06, "loss": 0.22, "lr_vision_tower": 1.603665414266704e-06, "sft_loss": 0.1079, "step": 511 }, { "distill_loss": 0.2009, "epoch": 0.32, "grad_norm": 0.8241057738713028, "learning_rate": 8.010359847935732e-06, "loss": 0.1921, "lr_vision_tower": 1.602071969587146e-06, "sft_loss": 0.0916, "step": 512 }, { "distill_loss": 0.2165, "epoch": 0.32, "grad_norm": 1.0742318374504545, "learning_rate": 8.002380616817495e-06, "loss": 0.2149, "lr_vision_tower": 1.600476123363499e-06, "sft_loss": 0.1067, "step": 513 }, { "distill_loss": 0.2118, "epoch": 0.32, "grad_norm": 0.9076715337772016, "learning_rate": 7.994389409806359e-06, "loss": 0.2034, "lr_vision_tower": 1.5988778819612714e-06, "sft_loss": 0.0975, "step": 514 }, { "distill_loss": 0.2048, "epoch": 0.32, "grad_norm": 1.1752026362559416, "learning_rate": 7.986386258777643e-06, "loss": 0.2054, "lr_vision_tower": 1.5972772517555285e-06, "sft_loss": 0.103, "step": 515 }, { "distill_loss": 0.2018, "epoch": 0.32, "grad_norm": 0.9447622324914017, "learning_rate": 7.978371195654308e-06, "loss": 0.1905, "lr_vision_tower": 1.5956742391308616e-06, "sft_loss": 0.0896, "step": 516 }, { "distill_loss": 0.2073, "epoch": 0.32, "grad_norm": 0.990760726891237, "learning_rate": 7.970344252406832e-06, "loss": 0.209, "lr_vision_tower": 1.5940688504813663e-06, "sft_loss": 0.1053, "step": 517 }, { "distill_loss": 0.2028, "epoch": 0.32, "grad_norm": 0.9421911400187183, "learning_rate": 7.962305461053076e-06, "loss": 0.2057, "lr_vision_tower": 1.592461092210615e-06, "sft_loss": 0.1042, "step": 518 }, { "distill_loss": 0.2069, "epoch": 0.32, "grad_norm": 0.9483161356402452, "learning_rate": 7.954254853658163e-06, "loss": 0.2145, "lr_vision_tower": 1.5908509707316324e-06, "sft_loss": 0.1111, "step": 519 }, { "distill_loss": 0.2075, "epoch": 0.32, "grad_norm": 1.2003483508706883, "learning_rate": 7.94619246233435e-06, "loss": 0.2178, "lr_vision_tower": 1.5892384924668698e-06, "sft_loss": 0.1141, "step": 520 }, { "distill_loss": 0.2029, "epoch": 0.32, "grad_norm": 0.7837985509062367, "learning_rate": 7.938118319240896e-06, "loss": 0.2035, "lr_vision_tower": 1.5876236638481788e-06, "sft_loss": 0.102, "step": 521 }, { "distill_loss": 0.2198, "epoch": 0.32, "grad_norm": 0.9296779426989957, "learning_rate": 7.930032456583931e-06, "loss": 0.208, "lr_vision_tower": 1.5860064913167862e-06, "sft_loss": 0.0981, "step": 522 }, { "distill_loss": 0.2022, "epoch": 0.32, "grad_norm": 0.8144136738199577, "learning_rate": 7.92193490661634e-06, "loss": 0.1956, "lr_vision_tower": 1.584386981323268e-06, "sft_loss": 0.0945, "step": 523 }, { "distill_loss": 0.2089, "epoch": 0.32, "grad_norm": 0.8515226337545817, "learning_rate": 7.913825701637623e-06, "loss": 0.2155, "lr_vision_tower": 1.5827651403275244e-06, "sft_loss": 0.111, "step": 524 }, { "distill_loss": 0.2114, "epoch": 0.32, "grad_norm": 0.8050286262777924, "learning_rate": 7.905704873993766e-06, "loss": 0.2153, "lr_vision_tower": 1.5811409747987529e-06, "sft_loss": 0.1096, "step": 525 }, { "distill_loss": 0.1986, "epoch": 0.32, "grad_norm": 1.0680046357130804, "learning_rate": 7.89757245607712e-06, "loss": 0.1965, "lr_vision_tower": 1.579514491215424e-06, "sft_loss": 0.0972, "step": 526 }, { "distill_loss": 0.2031, "epoch": 0.32, "grad_norm": 0.7893880909198114, "learning_rate": 7.889428480326266e-06, "loss": 0.1751, "lr_vision_tower": 1.577885696065253e-06, "sft_loss": 0.0736, "step": 527 }, { "distill_loss": 0.2006, "epoch": 0.33, "grad_norm": 1.9729150283169687, "learning_rate": 7.881272979225886e-06, "loss": 0.1931, "lr_vision_tower": 1.576254595845177e-06, "sft_loss": 0.0928, "step": 528 }, { "distill_loss": 0.2049, "epoch": 0.33, "grad_norm": 1.4757902574188708, "learning_rate": 7.873105985306635e-06, "loss": 0.2043, "lr_vision_tower": 1.574621197061327e-06, "sft_loss": 0.1019, "step": 529 }, { "distill_loss": 0.2152, "epoch": 0.33, "grad_norm": 0.79108610504071, "learning_rate": 7.864927531145012e-06, "loss": 0.2046, "lr_vision_tower": 1.572985506229002e-06, "sft_loss": 0.097, "step": 530 }, { "distill_loss": 0.2164, "epoch": 0.33, "grad_norm": 2.2351501195424635, "learning_rate": 7.856737649363225e-06, "loss": 0.2175, "lr_vision_tower": 1.5713475298726447e-06, "sft_loss": 0.1093, "step": 531 }, { "distill_loss": 0.1996, "epoch": 0.33, "grad_norm": 1.3755460437048448, "learning_rate": 7.848536372629065e-06, "loss": 0.2086, "lr_vision_tower": 1.5697072745258125e-06, "sft_loss": 0.1088, "step": 532 }, { "distill_loss": 0.1998, "epoch": 0.33, "grad_norm": 1.5114660291830415, "learning_rate": 7.84032373365578e-06, "loss": 0.1991, "lr_vision_tower": 1.5680647467311555e-06, "sft_loss": 0.0992, "step": 533 }, { "distill_loss": 0.2085, "epoch": 0.33, "grad_norm": 0.8170855227797522, "learning_rate": 7.832099765201934e-06, "loss": 0.1984, "lr_vision_tower": 1.5664199530403868e-06, "sft_loss": 0.0941, "step": 534 }, { "distill_loss": 0.2258, "epoch": 0.33, "grad_norm": 1.2749518839891898, "learning_rate": 7.823864500071287e-06, "loss": 0.2242, "lr_vision_tower": 1.5647729000142572e-06, "sft_loss": 0.1113, "step": 535 }, { "distill_loss": 0.2151, "epoch": 0.33, "grad_norm": 1.1640027325086661, "learning_rate": 7.815617971112657e-06, "loss": 0.2075, "lr_vision_tower": 1.5631235942225312e-06, "sft_loss": 0.0999, "step": 536 }, { "distill_loss": 0.2055, "epoch": 0.33, "grad_norm": 0.9524603415229714, "learning_rate": 7.807360211219786e-06, "loss": 0.2008, "lr_vision_tower": 1.561472042243957e-06, "sft_loss": 0.0981, "step": 537 }, { "distill_loss": 0.1984, "epoch": 0.33, "grad_norm": 0.9535254110461567, "learning_rate": 7.799091253331224e-06, "loss": 0.1937, "lr_vision_tower": 1.5598182506662446e-06, "sft_loss": 0.0945, "step": 538 }, { "distill_loss": 0.2144, "epoch": 0.33, "grad_norm": 0.9782143255220679, "learning_rate": 7.790811130430179e-06, "loss": 0.2165, "lr_vision_tower": 1.5581622260860354e-06, "sft_loss": 0.1093, "step": 539 }, { "distill_loss": 0.2059, "epoch": 0.33, "grad_norm": 0.7930812445996637, "learning_rate": 7.782519875544397e-06, "loss": 0.2081, "lr_vision_tower": 1.556503975108879e-06, "sft_loss": 0.1051, "step": 540 }, { "distill_loss": 0.1996, "epoch": 0.33, "grad_norm": 0.7730033190409091, "learning_rate": 7.774217521746026e-06, "loss": 0.1905, "lr_vision_tower": 1.5548435043492052e-06, "sft_loss": 0.0907, "step": 541 }, { "distill_loss": 0.2075, "epoch": 0.33, "grad_norm": 0.8169733202601305, "learning_rate": 7.765904102151493e-06, "loss": 0.207, "lr_vision_tower": 1.5531808204302986e-06, "sft_loss": 0.1032, "step": 542 }, { "distill_loss": 0.2068, "epoch": 0.33, "grad_norm": 1.9297642428113306, "learning_rate": 7.757579649921354e-06, "loss": 0.2207, "lr_vision_tower": 1.5515159299842707e-06, "sft_loss": 0.1173, "step": 543 }, { "distill_loss": 0.2054, "epoch": 0.34, "grad_norm": 1.2386988849959326, "learning_rate": 7.749244198260175e-06, "loss": 0.2054, "lr_vision_tower": 1.5498488396520347e-06, "sft_loss": 0.1027, "step": 544 }, { "distill_loss": 0.207, "epoch": 0.34, "grad_norm": 0.7422288494939896, "learning_rate": 7.7408977804164e-06, "loss": 0.1891, "lr_vision_tower": 1.54817955608328e-06, "sft_loss": 0.0856, "step": 545 }, { "distill_loss": 0.2067, "epoch": 0.34, "grad_norm": 0.7461018633993148, "learning_rate": 7.732540429682211e-06, "loss": 0.2104, "lr_vision_tower": 1.546508085936442e-06, "sft_loss": 0.107, "step": 546 }, { "distill_loss": 0.2106, "epoch": 0.34, "grad_norm": 0.9826378947333712, "learning_rate": 7.7241721793934e-06, "loss": 0.2007, "lr_vision_tower": 1.54483443587868e-06, "sft_loss": 0.0954, "step": 547 }, { "distill_loss": 0.2064, "epoch": 0.34, "grad_norm": 0.8816392867795807, "learning_rate": 7.715793062929238e-06, "loss": 0.1958, "lr_vision_tower": 1.5431586125858473e-06, "sft_loss": 0.0926, "step": 548 }, { "distill_loss": 0.2058, "epoch": 0.34, "grad_norm": 0.9016394621076419, "learning_rate": 7.707403113712333e-06, "loss": 0.1868, "lr_vision_tower": 1.5414806227424663e-06, "sft_loss": 0.0839, "step": 549 }, { "distill_loss": 0.2164, "epoch": 0.34, "grad_norm": 1.236010618297341, "learning_rate": 7.69900236520851e-06, "loss": 0.2131, "lr_vision_tower": 1.5398004730417017e-06, "sft_loss": 0.1049, "step": 550 }, { "distill_loss": 0.2066, "epoch": 0.34, "grad_norm": 1.2551239893955635, "learning_rate": 7.690590850926664e-06, "loss": 0.1988, "lr_vision_tower": 1.5381181701853327e-06, "sft_loss": 0.0955, "step": 551 }, { "distill_loss": 0.212, "epoch": 0.34, "grad_norm": 0.9440334812758715, "learning_rate": 7.682168604418637e-06, "loss": 0.2123, "lr_vision_tower": 1.536433720883727e-06, "sft_loss": 0.1063, "step": 552 }, { "distill_loss": 0.2103, "epoch": 0.34, "grad_norm": 0.9081271517093303, "learning_rate": 7.673735659279075e-06, "loss": 0.2055, "lr_vision_tower": 1.5347471318558148e-06, "sft_loss": 0.1004, "step": 553 }, { "distill_loss": 0.2043, "epoch": 0.34, "grad_norm": 0.8673956887584984, "learning_rate": 7.665292049145305e-06, "loss": 0.1934, "lr_vision_tower": 1.5330584098290606e-06, "sft_loss": 0.0912, "step": 554 }, { "distill_loss": 0.2034, "epoch": 0.34, "grad_norm": 1.0542109662646215, "learning_rate": 7.656837807697187e-06, "loss": 0.1928, "lr_vision_tower": 1.5313675615394372e-06, "sft_loss": 0.0911, "step": 555 }, { "distill_loss": 0.2051, "epoch": 0.34, "grad_norm": 1.3046075301339042, "learning_rate": 7.648372968656995e-06, "loss": 0.2066, "lr_vision_tower": 1.5296745937313985e-06, "sft_loss": 0.1041, "step": 556 }, { "distill_loss": 0.2106, "epoch": 0.34, "grad_norm": 1.2253103765242623, "learning_rate": 7.639897565789267e-06, "loss": 0.2036, "lr_vision_tower": 1.5279795131578532e-06, "sft_loss": 0.0983, "step": 557 }, { "distill_loss": 0.2247, "epoch": 0.34, "grad_norm": 0.8424431805883412, "learning_rate": 7.631411632900683e-06, "loss": 0.2114, "lr_vision_tower": 1.5262823265801366e-06, "sft_loss": 0.0991, "step": 558 }, { "distill_loss": 0.2099, "epoch": 0.34, "grad_norm": 1.198815219080221, "learning_rate": 7.622915203839926e-06, "loss": 0.215, "lr_vision_tower": 1.524583040767985e-06, "sft_loss": 0.1101, "step": 559 }, { "distill_loss": 0.2082, "epoch": 0.35, "grad_norm": 0.9856003068512788, "learning_rate": 7.614408312497543e-06, "loss": 0.2092, "lr_vision_tower": 1.5228816624995085e-06, "sft_loss": 0.1051, "step": 560 }, { "distill_loss": 0.203, "epoch": 0.35, "grad_norm": 0.9453472713671329, "learning_rate": 7.605890992805814e-06, "loss": 0.1963, "lr_vision_tower": 1.5211781985611626e-06, "sft_loss": 0.0948, "step": 561 }, { "distill_loss": 0.2137, "epoch": 0.35, "grad_norm": 1.0656700204964318, "learning_rate": 7.597363278738612e-06, "loss": 0.2114, "lr_vision_tower": 1.5194726557477222e-06, "sft_loss": 0.1045, "step": 562 }, { "distill_loss": 0.1981, "epoch": 0.35, "grad_norm": 0.7755959876809992, "learning_rate": 7.588825204311278e-06, "loss": 0.1932, "lr_vision_tower": 1.5177650408622555e-06, "sft_loss": 0.0941, "step": 563 }, { "distill_loss": 0.2101, "epoch": 0.35, "grad_norm": 0.7687861939476639, "learning_rate": 7.5802768035804775e-06, "loss": 0.2068, "lr_vision_tower": 1.5160553607160952e-06, "sft_loss": 0.1018, "step": 564 }, { "distill_loss": 0.2159, "epoch": 0.35, "grad_norm": 0.6859364968570225, "learning_rate": 7.571718110644058e-06, "loss": 0.212, "lr_vision_tower": 1.5143436221288116e-06, "sft_loss": 0.1041, "step": 565 }, { "distill_loss": 0.2006, "epoch": 0.35, "grad_norm": 0.9696387263420545, "learning_rate": 7.563149159640929e-06, "loss": 0.1907, "lr_vision_tower": 1.5126298319281855e-06, "sft_loss": 0.0904, "step": 566 }, { "distill_loss": 0.1993, "epoch": 0.35, "grad_norm": 0.8101170523310024, "learning_rate": 7.5545699847509125e-06, "loss": 0.1862, "lr_vision_tower": 1.5109139969501822e-06, "sft_loss": 0.0866, "step": 567 }, { "distill_loss": 0.2188, "epoch": 0.35, "grad_norm": 0.9183672656163318, "learning_rate": 7.545980620194612e-06, "loss": 0.2022, "lr_vision_tower": 1.5091961240389222e-06, "sft_loss": 0.0928, "step": 568 }, { "distill_loss": 0.1968, "epoch": 0.35, "grad_norm": 0.8244439727045669, "learning_rate": 7.5373811002332785e-06, "loss": 0.195, "lr_vision_tower": 1.5074762200466555e-06, "sft_loss": 0.0966, "step": 569 }, { "distill_loss": 0.2046, "epoch": 0.35, "grad_norm": 1.0797294087481228, "learning_rate": 7.5287714591686685e-06, "loss": 0.202, "lr_vision_tower": 1.5057542918337336e-06, "sft_loss": 0.0997, "step": 570 }, { "distill_loss": 0.2091, "epoch": 0.35, "grad_norm": 0.8577330613280998, "learning_rate": 7.520151731342909e-06, "loss": 0.1963, "lr_vision_tower": 1.5040303462685817e-06, "sft_loss": 0.0918, "step": 571 }, { "distill_loss": 0.2132, "epoch": 0.35, "grad_norm": 0.9741112340748913, "learning_rate": 7.511521951138364e-06, "loss": 0.2114, "lr_vision_tower": 1.5023043902276725e-06, "sft_loss": 0.1048, "step": 572 }, { "distill_loss": 0.1952, "epoch": 0.35, "grad_norm": 1.1236405836393422, "learning_rate": 7.502882152977491e-06, "loss": 0.2093, "lr_vision_tower": 1.500576430595498e-06, "sft_loss": 0.1117, "step": 573 }, { "distill_loss": 0.1986, "epoch": 0.35, "grad_norm": 1.0404293004904825, "learning_rate": 7.494232371322707e-06, "loss": 0.1782, "lr_vision_tower": 1.4988464742645413e-06, "sft_loss": 0.0789, "step": 574 }, { "distill_loss": 0.1938, "epoch": 0.35, "grad_norm": 1.0329756999715476, "learning_rate": 7.485572640676255e-06, "loss": 0.2035, "lr_vision_tower": 1.4971145281352508e-06, "sft_loss": 0.1066, "step": 575 }, { "distill_loss": 0.2074, "epoch": 0.36, "grad_norm": 0.8978046807888208, "learning_rate": 7.4769029955800595e-06, "loss": 0.2037, "lr_vision_tower": 1.4953805991160117e-06, "sft_loss": 0.1001, "step": 576 }, { "distill_loss": 0.2058, "epoch": 0.36, "grad_norm": 1.105019453912801, "learning_rate": 7.468223470615593e-06, "loss": 0.1805, "lr_vision_tower": 1.4936446941231185e-06, "sft_loss": 0.0776, "step": 577 }, { "distill_loss": 0.2036, "epoch": 0.36, "grad_norm": 0.88139845948242, "learning_rate": 7.459534100403736e-06, "loss": 0.1967, "lr_vision_tower": 1.491906820080747e-06, "sft_loss": 0.0949, "step": 578 }, { "distill_loss": 0.2123, "epoch": 0.36, "grad_norm": 0.9757004428807075, "learning_rate": 7.450834919604639e-06, "loss": 0.211, "lr_vision_tower": 1.4901669839209276e-06, "sft_loss": 0.1048, "step": 579 }, { "distill_loss": 0.2136, "epoch": 0.36, "grad_norm": 0.9224343444404166, "learning_rate": 7.442125962917586e-06, "loss": 0.2007, "lr_vision_tower": 1.4884251925835169e-06, "sft_loss": 0.0938, "step": 580 }, { "distill_loss": 0.2048, "epoch": 0.36, "grad_norm": 1.0368482968285915, "learning_rate": 7.433407265080857e-06, "loss": 0.2041, "lr_vision_tower": 1.4866814530161714e-06, "sft_loss": 0.1017, "step": 581 }, { "distill_loss": 0.218, "epoch": 0.36, "grad_norm": 0.8114665014549921, "learning_rate": 7.424678860871584e-06, "loss": 0.2224, "lr_vision_tower": 1.4849357721743168e-06, "sft_loss": 0.1134, "step": 582 }, { "distill_loss": 0.2015, "epoch": 0.36, "grad_norm": 0.92611841027311, "learning_rate": 7.415940785105618e-06, "loss": 0.1963, "lr_vision_tower": 1.4831881570211233e-06, "sft_loss": 0.0955, "step": 583 }, { "distill_loss": 0.1989, "epoch": 0.36, "grad_norm": 0.7485440010358154, "learning_rate": 7.407193072637384e-06, "loss": 0.2013, "lr_vision_tower": 1.4814386145274766e-06, "sft_loss": 0.1019, "step": 584 }, { "distill_loss": 0.204, "epoch": 0.36, "grad_norm": 0.6718960051397685, "learning_rate": 7.398435758359754e-06, "loss": 0.2052, "lr_vision_tower": 1.4796871516719507e-06, "sft_loss": 0.1032, "step": 585 }, { "distill_loss": 0.2056, "epoch": 0.36, "grad_norm": 1.3784762760652376, "learning_rate": 7.389668877203893e-06, "loss": 0.197, "lr_vision_tower": 1.4779337754407782e-06, "sft_loss": 0.0942, "step": 586 }, { "distill_loss": 0.1986, "epoch": 0.36, "grad_norm": 0.8631965983453119, "learning_rate": 7.380892464139126e-06, "loss": 0.2007, "lr_vision_tower": 1.4761784928278248e-06, "sft_loss": 0.1014, "step": 587 }, { "distill_loss": 0.2111, "epoch": 0.36, "grad_norm": 0.9087388614258647, "learning_rate": 7.372106554172802e-06, "loss": 0.1975, "lr_vision_tower": 1.4744213108345602e-06, "sft_loss": 0.0919, "step": 588 }, { "distill_loss": 0.2049, "epoch": 0.36, "grad_norm": 0.9954234610210507, "learning_rate": 7.363311182350151e-06, "loss": 0.2057, "lr_vision_tower": 1.4726622364700301e-06, "sft_loss": 0.1032, "step": 589 }, { "distill_loss": 0.2068, "epoch": 0.36, "grad_norm": 0.6922103965550838, "learning_rate": 7.354506383754144e-06, "loss": 0.1889, "lr_vision_tower": 1.4709012767508286e-06, "sft_loss": 0.0855, "step": 590 }, { "distill_loss": 0.2035, "epoch": 0.36, "grad_norm": 0.8716464620407636, "learning_rate": 7.345692193505352e-06, "loss": 0.2006, "lr_vision_tower": 1.4691384387010703e-06, "sft_loss": 0.0989, "step": 591 }, { "distill_loss": 0.214, "epoch": 0.36, "grad_norm": 0.8077056850637353, "learning_rate": 7.336868646761808e-06, "loss": 0.2017, "lr_vision_tower": 1.4673737293523613e-06, "sft_loss": 0.0947, "step": 592 }, { "distill_loss": 0.198, "epoch": 0.37, "grad_norm": 0.923183906417328, "learning_rate": 7.3280357787188675e-06, "loss": 0.1919, "lr_vision_tower": 1.4656071557437733e-06, "sft_loss": 0.0929, "step": 593 }, { "distill_loss": 0.1954, "epoch": 0.37, "grad_norm": 0.8844029336501031, "learning_rate": 7.3191936246090666e-06, "loss": 0.2011, "lr_vision_tower": 1.4638387249218132e-06, "sft_loss": 0.1034, "step": 594 }, { "distill_loss": 0.202, "epoch": 0.37, "grad_norm": 1.1451027947786554, "learning_rate": 7.310342219701981e-06, "loss": 0.1981, "lr_vision_tower": 1.462068443940396e-06, "sft_loss": 0.0971, "step": 595 }, { "distill_loss": 0.1981, "epoch": 0.37, "grad_norm": 0.9890462702075584, "learning_rate": 7.301481599304086e-06, "loss": 0.1892, "lr_vision_tower": 1.4602963198608168e-06, "sft_loss": 0.0901, "step": 596 }, { "distill_loss": 0.2032, "epoch": 0.37, "grad_norm": 0.81566928330756, "learning_rate": 7.292611798758613e-06, "loss": 0.2075, "lr_vision_tower": 1.4585223597517225e-06, "sft_loss": 0.1059, "step": 597 }, { "distill_loss": 0.2088, "epoch": 0.37, "grad_norm": 0.7394901180207429, "learning_rate": 7.283732853445418e-06, "loss": 0.1946, "lr_vision_tower": 1.4567465706890834e-06, "sft_loss": 0.0902, "step": 598 }, { "distill_loss": 0.1969, "epoch": 0.37, "grad_norm": 0.9125338085092892, "learning_rate": 7.274844798780826e-06, "loss": 0.1861, "lr_vision_tower": 1.454968959756165e-06, "sft_loss": 0.0876, "step": 599 }, { "distill_loss": 0.1974, "epoch": 0.37, "grad_norm": 0.749495366500317, "learning_rate": 7.265947670217503e-06, "loss": 0.1966, "lr_vision_tower": 1.4531895340435005e-06, "sft_loss": 0.0979, "step": 600 }, { "distill_loss": 0.2027, "epoch": 0.37, "grad_norm": 0.8049083215669378, "learning_rate": 7.257041503244307e-06, "loss": 0.2064, "lr_vision_tower": 1.4514083006488611e-06, "sft_loss": 0.1051, "step": 601 }, { "distill_loss": 0.2004, "epoch": 0.37, "grad_norm": 0.715003516457525, "learning_rate": 7.248126333386145e-06, "loss": 0.19, "lr_vision_tower": 1.449625266677229e-06, "sft_loss": 0.0898, "step": 602 }, { "distill_loss": 0.207, "epoch": 0.37, "grad_norm": 0.8633602886306395, "learning_rate": 7.239202196203842e-06, "loss": 0.1986, "lr_vision_tower": 1.4478404392407682e-06, "sft_loss": 0.0951, "step": 603 }, { "distill_loss": 0.202, "epoch": 0.37, "grad_norm": 0.6845976691343608, "learning_rate": 7.230269127293986e-06, "loss": 0.1901, "lr_vision_tower": 1.446053825458797e-06, "sft_loss": 0.0891, "step": 604 }, { "distill_loss": 0.2078, "epoch": 0.37, "grad_norm": 1.0847011613680662, "learning_rate": 7.221327162288792e-06, "loss": 0.2147, "lr_vision_tower": 1.4442654324577583e-06, "sft_loss": 0.1108, "step": 605 }, { "distill_loss": 0.1973, "epoch": 0.37, "grad_norm": 0.739879675607288, "learning_rate": 7.212376336855965e-06, "loss": 0.1909, "lr_vision_tower": 1.4424752673711928e-06, "sft_loss": 0.0923, "step": 606 }, { "distill_loss": 0.2071, "epoch": 0.37, "grad_norm": 0.9601273613571878, "learning_rate": 7.203416686698546e-06, "loss": 0.1948, "lr_vision_tower": 1.440683337339709e-06, "sft_loss": 0.0912, "step": 607 }, { "distill_loss": 0.2018, "epoch": 0.37, "grad_norm": 1.1187522427111884, "learning_rate": 7.19444824755478e-06, "loss": 0.1995, "lr_vision_tower": 1.4388896495109556e-06, "sft_loss": 0.0986, "step": 608 }, { "distill_loss": 0.1964, "epoch": 0.38, "grad_norm": 0.9151668146778202, "learning_rate": 7.1854710551979675e-06, "loss": 0.2067, "lr_vision_tower": 1.4370942110395933e-06, "sft_loss": 0.1085, "step": 609 }, { "distill_loss": 0.2059, "epoch": 0.38, "grad_norm": 0.7608497509332818, "learning_rate": 7.176485145436325e-06, "loss": 0.1924, "lr_vision_tower": 1.435297029087265e-06, "sft_loss": 0.0895, "step": 610 }, { "distill_loss": 0.202, "epoch": 0.38, "grad_norm": 0.8551360319303192, "learning_rate": 7.167490554112841e-06, "loss": 0.1941, "lr_vision_tower": 1.433498110822568e-06, "sft_loss": 0.0931, "step": 611 }, { "distill_loss": 0.2117, "epoch": 0.38, "grad_norm": 0.7438664371874969, "learning_rate": 7.158487317105133e-06, "loss": 0.2025, "lr_vision_tower": 1.4316974634210264e-06, "sft_loss": 0.0966, "step": 612 }, { "distill_loss": 0.2025, "epoch": 0.38, "grad_norm": 1.2661602359300455, "learning_rate": 7.149475470325303e-06, "loss": 0.2108, "lr_vision_tower": 1.4298950940650604e-06, "sft_loss": 0.1096, "step": 613 }, { "distill_loss": 0.1908, "epoch": 0.38, "grad_norm": 0.7356763608786389, "learning_rate": 7.140455049719798e-06, "loss": 0.1935, "lr_vision_tower": 1.4280910099439594e-06, "sft_loss": 0.0981, "step": 614 }, { "distill_loss": 0.1943, "epoch": 0.38, "grad_norm": 0.8920089867161672, "learning_rate": 7.131426091269262e-06, "loss": 0.2068, "lr_vision_tower": 1.4262852182538523e-06, "sft_loss": 0.1096, "step": 615 }, { "distill_loss": 0.1999, "epoch": 0.38, "grad_norm": 0.8427051545462784, "learning_rate": 7.122388630988397e-06, "loss": 0.1927, "lr_vision_tower": 1.4244777261976792e-06, "sft_loss": 0.0927, "step": 616 }, { "distill_loss": 0.1889, "epoch": 0.38, "grad_norm": 0.8300892691544267, "learning_rate": 7.113342704925813e-06, "loss": 0.1974, "lr_vision_tower": 1.4226685409851624e-06, "sft_loss": 0.1029, "step": 617 }, { "distill_loss": 0.2026, "epoch": 0.38, "grad_norm": 0.8654006914224261, "learning_rate": 7.104288349163895e-06, "loss": 0.1949, "lr_vision_tower": 1.4208576698327788e-06, "sft_loss": 0.0936, "step": 618 }, { "distill_loss": 0.2051, "epoch": 0.38, "grad_norm": 1.0470402984028886, "learning_rate": 7.095225599818645e-06, "loss": 0.2079, "lr_vision_tower": 1.4190451199637289e-06, "sft_loss": 0.1054, "step": 619 }, { "distill_loss": 0.2049, "epoch": 0.38, "grad_norm": 0.8979033085716062, "learning_rate": 7.08615449303955e-06, "loss": 0.2044, "lr_vision_tower": 1.4172308986079098e-06, "sft_loss": 0.102, "step": 620 }, { "distill_loss": 0.2028, "epoch": 0.38, "grad_norm": 0.882489645650651, "learning_rate": 7.0770750650094335e-06, "loss": 0.217, "lr_vision_tower": 1.4154150130018865e-06, "sft_loss": 0.1156, "step": 621 }, { "distill_loss": 0.2003, "epoch": 0.38, "grad_norm": 0.9642343132242768, "learning_rate": 7.067987351944307e-06, "loss": 0.1961, "lr_vision_tower": 1.4135974703888614e-06, "sft_loss": 0.0959, "step": 622 }, { "distill_loss": 0.1955, "epoch": 0.38, "grad_norm": 1.1053256004690624, "learning_rate": 7.058891390093232e-06, "loss": 0.2081, "lr_vision_tower": 1.4117782780186463e-06, "sft_loss": 0.1103, "step": 623 }, { "distill_loss": 0.2144, "epoch": 0.38, "grad_norm": 1.165643221181518, "learning_rate": 7.049787215738175e-06, "loss": 0.206, "lr_vision_tower": 1.4099574431476348e-06, "sft_loss": 0.0988, "step": 624 }, { "distill_loss": 0.2147, "epoch": 0.39, "grad_norm": 0.7970951002660133, "learning_rate": 7.040674865193855e-06, "loss": 0.2011, "lr_vision_tower": 1.4081349730387709e-06, "sft_loss": 0.0938, "step": 625 }, { "distill_loss": 0.1939, "epoch": 0.39, "grad_norm": 1.102787936637049, "learning_rate": 7.0315543748076066e-06, "loss": 0.196, "lr_vision_tower": 1.4063108749615211e-06, "sft_loss": 0.0991, "step": 626 }, { "distill_loss": 0.1979, "epoch": 0.39, "grad_norm": 0.9275414706111118, "learning_rate": 7.022425780959234e-06, "loss": 0.2102, "lr_vision_tower": 1.4044851561918466e-06, "sft_loss": 0.1112, "step": 627 }, { "distill_loss": 0.2029, "epoch": 0.39, "grad_norm": 1.0459989574413193, "learning_rate": 7.013289120060866e-06, "loss": 0.1916, "lr_vision_tower": 1.402657824012173e-06, "sft_loss": 0.0901, "step": 628 }, { "distill_loss": 0.207, "epoch": 0.39, "grad_norm": 0.9103244048363336, "learning_rate": 7.004144428556802e-06, "loss": 0.2153, "lr_vision_tower": 1.4008288857113604e-06, "sft_loss": 0.1118, "step": 629 }, { "distill_loss": 0.1948, "epoch": 0.39, "grad_norm": 0.6336964801774712, "learning_rate": 6.994991742923383e-06, "loss": 0.2122, "lr_vision_tower": 1.3989983485846764e-06, "sft_loss": 0.1148, "step": 630 }, { "distill_loss": 0.2097, "epoch": 0.39, "grad_norm": 1.140455719504122, "learning_rate": 6.9858310996688294e-06, "loss": 0.2076, "lr_vision_tower": 1.3971662199337656e-06, "sft_loss": 0.1027, "step": 631 }, { "distill_loss": 0.1988, "epoch": 0.39, "grad_norm": 0.6790942361356397, "learning_rate": 6.976662535333107e-06, "loss": 0.1979, "lr_vision_tower": 1.3953325070666213e-06, "sft_loss": 0.0986, "step": 632 }, { "distill_loss": 0.1968, "epoch": 0.39, "grad_norm": 0.8356000210000488, "learning_rate": 6.9674860864877795e-06, "loss": 0.2161, "lr_vision_tower": 1.3934972172975557e-06, "sft_loss": 0.1177, "step": 633 }, { "distill_loss": 0.1938, "epoch": 0.39, "grad_norm": 0.7312772485433465, "learning_rate": 6.958301789735853e-06, "loss": 0.1949, "lr_vision_tower": 1.3916603579471705e-06, "sft_loss": 0.098, "step": 634 }, { "distill_loss": 0.191, "epoch": 0.39, "grad_norm": 0.8222795758363839, "learning_rate": 6.949109681711641e-06, "loss": 0.182, "lr_vision_tower": 1.3898219363423281e-06, "sft_loss": 0.0865, "step": 635 }, { "distill_loss": 0.1981, "epoch": 0.39, "grad_norm": 0.8172418943756223, "learning_rate": 6.939909799080619e-06, "loss": 0.1941, "lr_vision_tower": 1.3879819598161236e-06, "sft_loss": 0.095, "step": 636 }, { "distill_loss": 0.1928, "epoch": 0.39, "grad_norm": 0.8745078610132608, "learning_rate": 6.930702178539267e-06, "loss": 0.1887, "lr_vision_tower": 1.3861404357078531e-06, "sft_loss": 0.0924, "step": 637 }, { "distill_loss": 0.2046, "epoch": 0.39, "grad_norm": 1.3558376969141133, "learning_rate": 6.921486856814934e-06, "loss": 0.2041, "lr_vision_tower": 1.3842973713629866e-06, "sft_loss": 0.1018, "step": 638 }, { "distill_loss": 0.1897, "epoch": 0.39, "grad_norm": 0.8116241116426914, "learning_rate": 6.912263870665688e-06, "loss": 0.1821, "lr_vision_tower": 1.3824527741331374e-06, "sft_loss": 0.0873, "step": 639 }, { "distill_loss": 0.1989, "epoch": 0.39, "grad_norm": 0.6659222513306564, "learning_rate": 6.903033256880164e-06, "loss": 0.2089, "lr_vision_tower": 1.3806066513760327e-06, "sft_loss": 0.1095, "step": 640 }, { "distill_loss": 0.1984, "epoch": 0.4, "grad_norm": 0.702481651393293, "learning_rate": 6.8937950522774255e-06, "loss": 0.1909, "lr_vision_tower": 1.378759010455485e-06, "sft_loss": 0.0917, "step": 641 }, { "distill_loss": 0.2001, "epoch": 0.4, "grad_norm": 0.9441619074066976, "learning_rate": 6.884549293706817e-06, "loss": 0.1885, "lr_vision_tower": 1.3769098587413631e-06, "sft_loss": 0.0885, "step": 642 }, { "distill_loss": 0.2082, "epoch": 0.4, "grad_norm": 0.9199824099510338, "learning_rate": 6.87529601804781e-06, "loss": 0.1928, "lr_vision_tower": 1.3750592036095619e-06, "sft_loss": 0.0887, "step": 643 }, { "distill_loss": 0.1952, "epoch": 0.4, "grad_norm": 1.0251121258786213, "learning_rate": 6.86603526220986e-06, "loss": 0.2133, "lr_vision_tower": 1.3732070524419718e-06, "sft_loss": 0.1158, "step": 644 }, { "distill_loss": 0.2055, "epoch": 0.4, "grad_norm": 0.8044575539183406, "learning_rate": 6.856767063132262e-06, "loss": 0.1956, "lr_vision_tower": 1.3713534126264523e-06, "sft_loss": 0.0929, "step": 645 }, { "distill_loss": 0.199, "epoch": 0.4, "grad_norm": 1.1327411648364971, "learning_rate": 6.847491457784001e-06, "loss": 0.1995, "lr_vision_tower": 1.3694982915568e-06, "sft_loss": 0.1, "step": 646 }, { "distill_loss": 0.1879, "epoch": 0.4, "grad_norm": 0.7827030535416288, "learning_rate": 6.838208483163601e-06, "loss": 0.2015, "lr_vision_tower": 1.36764169663272e-06, "sft_loss": 0.1075, "step": 647 }, { "distill_loss": 0.2084, "epoch": 0.4, "grad_norm": 0.6430098831298572, "learning_rate": 6.828918176298982e-06, "loss": 0.1916, "lr_vision_tower": 1.3657836352597962e-06, "sft_loss": 0.0874, "step": 648 }, { "distill_loss": 0.1957, "epoch": 0.4, "grad_norm": 0.7964838329991831, "learning_rate": 6.819620574247312e-06, "loss": 0.2008, "lr_vision_tower": 1.3639241148494622e-06, "sft_loss": 0.1029, "step": 649 }, { "distill_loss": 0.196, "epoch": 0.4, "grad_norm": 1.0375697250791056, "learning_rate": 6.8103157140948565e-06, "loss": 0.1937, "lr_vision_tower": 1.3620631428189711e-06, "sft_loss": 0.0957, "step": 650 }, { "distill_loss": 0.2028, "epoch": 0.4, "grad_norm": 0.8317189295014934, "learning_rate": 6.801003632956833e-06, "loss": 0.2123, "lr_vision_tower": 1.3602007265913664e-06, "sft_loss": 0.1109, "step": 651 }, { "distill_loss": 0.2113, "epoch": 0.4, "grad_norm": 0.680650542473756, "learning_rate": 6.79168436797726e-06, "loss": 0.1968, "lr_vision_tower": 1.3583368735954519e-06, "sft_loss": 0.0911, "step": 652 }, { "distill_loss": 0.1961, "epoch": 0.4, "grad_norm": 0.927139319812575, "learning_rate": 6.782357956328815e-06, "loss": 0.1914, "lr_vision_tower": 1.3564715912657628e-06, "sft_loss": 0.0934, "step": 653 }, { "distill_loss": 0.1927, "epoch": 0.4, "grad_norm": 0.8505124934130435, "learning_rate": 6.773024435212678e-06, "loss": 0.1971, "lr_vision_tower": 1.3546048870425354e-06, "sft_loss": 0.1007, "step": 654 }, { "distill_loss": 0.2066, "epoch": 0.4, "grad_norm": 0.758096117642211, "learning_rate": 6.763683841858392e-06, "loss": 0.1911, "lr_vision_tower": 1.3527367683716782e-06, "sft_loss": 0.0878, "step": 655 }, { "distill_loss": 0.2036, "epoch": 0.4, "grad_norm": 0.9616443015877203, "learning_rate": 6.7543362135237025e-06, "loss": 0.2055, "lr_vision_tower": 1.3508672427047403e-06, "sft_loss": 0.1037, "step": 656 }, { "distill_loss": 0.2014, "epoch": 0.4, "grad_norm": 0.7377293378542051, "learning_rate": 6.744981587494423e-06, "loss": 0.2094, "lr_vision_tower": 1.3489963174988846e-06, "sft_loss": 0.1087, "step": 657 }, { "distill_loss": 0.2065, "epoch": 0.41, "grad_norm": 0.7175201768265703, "learning_rate": 6.735620001084278e-06, "loss": 0.197, "lr_vision_tower": 1.3471240002168553e-06, "sft_loss": 0.0938, "step": 658 }, { "distill_loss": 0.1933, "epoch": 0.41, "grad_norm": 1.1218232314922683, "learning_rate": 6.72625149163475e-06, "loss": 0.1958, "lr_vision_tower": 1.3452502983269498e-06, "sft_loss": 0.0992, "step": 659 }, { "distill_loss": 0.2021, "epoch": 0.41, "grad_norm": 1.0227322132966554, "learning_rate": 6.716876096514944e-06, "loss": 0.1957, "lr_vision_tower": 1.3433752193029887e-06, "sft_loss": 0.0946, "step": 660 }, { "distill_loss": 0.1964, "epoch": 0.41, "grad_norm": 1.1625952154264712, "learning_rate": 6.707493853121425e-06, "loss": 0.1962, "lr_vision_tower": 1.3414987706242849e-06, "sft_loss": 0.098, "step": 661 }, { "distill_loss": 0.1862, "epoch": 0.41, "grad_norm": 1.0861831507128328, "learning_rate": 6.698104798878077e-06, "loss": 0.1887, "lr_vision_tower": 1.3396209597756151e-06, "sft_loss": 0.0956, "step": 662 }, { "distill_loss": 0.2035, "epoch": 0.41, "grad_norm": 1.113001422165158, "learning_rate": 6.688708971235952e-06, "loss": 0.2114, "lr_vision_tower": 1.3377417942471902e-06, "sft_loss": 0.1097, "step": 663 }, { "distill_loss": 0.1946, "epoch": 0.41, "grad_norm": 0.9555190769308919, "learning_rate": 6.6793064076731165e-06, "loss": 0.1942, "lr_vision_tower": 1.335861281534623e-06, "sft_loss": 0.0968, "step": 664 }, { "distill_loss": 0.2012, "epoch": 0.41, "grad_norm": 0.8157465097106148, "learning_rate": 6.669897145694507e-06, "loss": 0.1913, "lr_vision_tower": 1.3339794291389012e-06, "sft_loss": 0.0907, "step": 665 }, { "distill_loss": 0.1915, "epoch": 0.41, "grad_norm": 1.3533629582610813, "learning_rate": 6.66048122283178e-06, "loss": 0.1908, "lr_vision_tower": 1.332096244566356e-06, "sft_loss": 0.0951, "step": 666 }, { "distill_loss": 0.2043, "epoch": 0.41, "grad_norm": 0.9963694725982691, "learning_rate": 6.651058676643158e-06, "loss": 0.1937, "lr_vision_tower": 1.3302117353286315e-06, "sft_loss": 0.0916, "step": 667 }, { "distill_loss": 0.2007, "epoch": 0.41, "grad_norm": 0.8634167139264926, "learning_rate": 6.6416295447132836e-06, "loss": 0.1852, "lr_vision_tower": 1.3283259089426566e-06, "sft_loss": 0.0848, "step": 668 }, { "distill_loss": 0.1958, "epoch": 0.41, "grad_norm": 0.6502420719146875, "learning_rate": 6.632193864653071e-06, "loss": 0.1822, "lr_vision_tower": 1.326438772930614e-06, "sft_loss": 0.0843, "step": 669 }, { "distill_loss": 0.1967, "epoch": 0.41, "grad_norm": 1.2460400627237735, "learning_rate": 6.622751674099552e-06, "loss": 0.2085, "lr_vision_tower": 1.3245503348199103e-06, "sft_loss": 0.1102, "step": 670 }, { "distill_loss": 0.1956, "epoch": 0.41, "grad_norm": 0.820963352426331, "learning_rate": 6.613303010715723e-06, "loss": 0.2032, "lr_vision_tower": 1.3226606021431444e-06, "sft_loss": 0.1054, "step": 671 }, { "distill_loss": 0.1873, "epoch": 0.41, "grad_norm": 0.8435039666403726, "learning_rate": 6.603847912190408e-06, "loss": 0.2065, "lr_vision_tower": 1.3207695824380815e-06, "sft_loss": 0.1128, "step": 672 }, { "distill_loss": 0.1881, "epoch": 0.41, "grad_norm": 0.9124730229091736, "learning_rate": 6.594386416238095e-06, "loss": 0.1905, "lr_vision_tower": 1.3188772832476187e-06, "sft_loss": 0.0965, "step": 673 }, { "distill_loss": 0.1964, "epoch": 0.42, "grad_norm": 0.8938387300995939, "learning_rate": 6.584918560598784e-06, "loss": 0.1935, "lr_vision_tower": 1.3169837121197567e-06, "sft_loss": 0.0953, "step": 674 }, { "distill_loss": 0.184, "epoch": 0.42, "grad_norm": 0.7765424501379903, "learning_rate": 6.575444383037854e-06, "loss": 0.1826, "lr_vision_tower": 1.3150888766075708e-06, "sft_loss": 0.0906, "step": 675 }, { "distill_loss": 0.1896, "epoch": 0.42, "grad_norm": 0.8610673585157912, "learning_rate": 6.565963921345896e-06, "loss": 0.2, "lr_vision_tower": 1.313192784269179e-06, "sft_loss": 0.1052, "step": 676 }, { "distill_loss": 0.2025, "epoch": 0.42, "grad_norm": 1.0959624845398142, "learning_rate": 6.556477213338561e-06, "loss": 0.1913, "lr_vision_tower": 1.311295442667712e-06, "sft_loss": 0.09, "step": 677 }, { "distill_loss": 0.2003, "epoch": 0.42, "grad_norm": 0.85868744325712, "learning_rate": 6.5469842968564265e-06, "loss": 0.2059, "lr_vision_tower": 1.309396859371285e-06, "sft_loss": 0.1057, "step": 678 }, { "distill_loss": 0.1934, "epoch": 0.42, "grad_norm": 0.6851929459797714, "learning_rate": 6.537485209764825e-06, "loss": 0.1858, "lr_vision_tower": 1.3074970419529647e-06, "sft_loss": 0.0891, "step": 679 }, { "distill_loss": 0.1866, "epoch": 0.42, "grad_norm": 0.6768986224237741, "learning_rate": 6.527979989953704e-06, "loss": 0.1782, "lr_vision_tower": 1.3055959979907408e-06, "sft_loss": 0.0849, "step": 680 }, { "distill_loss": 0.1943, "epoch": 0.42, "grad_norm": 1.0121024825822211, "learning_rate": 6.518468675337481e-06, "loss": 0.1977, "lr_vision_tower": 1.3036937350674962e-06, "sft_loss": 0.1005, "step": 681 }, { "distill_loss": 0.1977, "epoch": 0.42, "grad_norm": 1.0377115088163773, "learning_rate": 6.5089513038548755e-06, "loss": 0.2007, "lr_vision_tower": 1.301790260770975e-06, "sft_loss": 0.1018, "step": 682 }, { "distill_loss": 0.1877, "epoch": 0.42, "grad_norm": 0.8671600811898509, "learning_rate": 6.499427913468769e-06, "loss": 0.198, "lr_vision_tower": 1.2998855826937537e-06, "sft_loss": 0.1041, "step": 683 }, { "distill_loss": 0.2024, "epoch": 0.42, "grad_norm": 0.8883867787317082, "learning_rate": 6.4898985421660535e-06, "loss": 0.2098, "lr_vision_tower": 1.2979797084332105e-06, "sft_loss": 0.1086, "step": 684 }, { "distill_loss": 0.1966, "epoch": 0.42, "grad_norm": 0.7555831980986011, "learning_rate": 6.480363227957474e-06, "loss": 0.1916, "lr_vision_tower": 1.2960726455914945e-06, "sft_loss": 0.0933, "step": 685 }, { "distill_loss": 0.1913, "epoch": 0.42, "grad_norm": 0.7543018365135985, "learning_rate": 6.470822008877482e-06, "loss": 0.1966, "lr_vision_tower": 1.2941644017754963e-06, "sft_loss": 0.1009, "step": 686 }, { "distill_loss": 0.1988, "epoch": 0.42, "grad_norm": 0.8567703352100344, "learning_rate": 6.461274922984087e-06, "loss": 0.1957, "lr_vision_tower": 1.2922549845968172e-06, "sft_loss": 0.0963, "step": 687 }, { "distill_loss": 0.1852, "epoch": 0.42, "grad_norm": 0.8758343746745908, "learning_rate": 6.451722008358692e-06, "loss": 0.1948, "lr_vision_tower": 1.2903444016717382e-06, "sft_loss": 0.1022, "step": 688 }, { "distill_loss": 0.2033, "epoch": 0.42, "grad_norm": 0.8989687010237055, "learning_rate": 6.442163303105956e-06, "loss": 0.1897, "lr_vision_tower": 1.288432660621191e-06, "sft_loss": 0.0881, "step": 689 }, { "distill_loss": 0.2019, "epoch": 0.43, "grad_norm": 0.8802675901247268, "learning_rate": 6.432598845353631e-06, "loss": 0.2138, "lr_vision_tower": 1.286519769070726e-06, "sft_loss": 0.1128, "step": 690 }, { "distill_loss": 0.1972, "epoch": 0.43, "grad_norm": 0.8872296721104739, "learning_rate": 6.42302867325242e-06, "loss": 0.2095, "lr_vision_tower": 1.284605734650484e-06, "sft_loss": 0.1109, "step": 691 }, { "distill_loss": 0.2022, "epoch": 0.43, "grad_norm": 0.7958277372317604, "learning_rate": 6.413452824975813e-06, "loss": 0.1884, "lr_vision_tower": 1.2826905649951626e-06, "sft_loss": 0.0873, "step": 692 }, { "distill_loss": 0.2037, "epoch": 0.43, "grad_norm": 0.9449660204014608, "learning_rate": 6.403871338719947e-06, "loss": 0.1989, "lr_vision_tower": 1.2807742677439893e-06, "sft_loss": 0.0971, "step": 693 }, { "distill_loss": 0.2008, "epoch": 0.43, "grad_norm": 1.2367810145842206, "learning_rate": 6.394284252703445e-06, "loss": 0.2099, "lr_vision_tower": 1.2788568505406888e-06, "sft_loss": 0.1095, "step": 694 }, { "distill_loss": 0.2016, "epoch": 0.43, "grad_norm": 0.8599152314349162, "learning_rate": 6.384691605167261e-06, "loss": 0.2087, "lr_vision_tower": 1.276938321033452e-06, "sft_loss": 0.1079, "step": 695 }, { "distill_loss": 0.1928, "epoch": 0.43, "grad_norm": 0.7533979062971262, "learning_rate": 6.375093434374546e-06, "loss": 0.1829, "lr_vision_tower": 1.2750186868749091e-06, "sft_loss": 0.0865, "step": 696 }, { "distill_loss": 0.1951, "epoch": 0.43, "grad_norm": 0.7336480383855496, "learning_rate": 6.36548977861047e-06, "loss": 0.182, "lr_vision_tower": 1.2730979557220937e-06, "sft_loss": 0.0845, "step": 697 }, { "distill_loss": 0.2057, "epoch": 0.43, "grad_norm": 0.6822208029987302, "learning_rate": 6.355880676182086e-06, "loss": 0.2031, "lr_vision_tower": 1.271176135236417e-06, "sft_loss": 0.1003, "step": 698 }, { "distill_loss": 0.1907, "epoch": 0.43, "grad_norm": 0.7404352200143679, "learning_rate": 6.346266165418173e-06, "loss": 0.1943, "lr_vision_tower": 1.2692532330836345e-06, "sft_loss": 0.0989, "step": 699 }, { "distill_loss": 0.1854, "epoch": 0.43, "grad_norm": 0.9257190096759348, "learning_rate": 6.336646284669083e-06, "loss": 0.217, "lr_vision_tower": 1.2673292569338165e-06, "sft_loss": 0.1243, "step": 700 }, { "distill_loss": 0.1818, "epoch": 0.43, "grad_norm": 0.8333114839113321, "learning_rate": 6.327021072306586e-06, "loss": 0.1841, "lr_vision_tower": 1.265404214461317e-06, "sft_loss": 0.0932, "step": 701 }, { "distill_loss": 0.2006, "epoch": 0.43, "grad_norm": 1.0680909541628913, "learning_rate": 6.317390566723722e-06, "loss": 0.2016, "lr_vision_tower": 1.263478113344744e-06, "sft_loss": 0.1013, "step": 702 }, { "distill_loss": 0.2014, "epoch": 0.43, "grad_norm": 0.7434946397886552, "learning_rate": 6.307754806334639e-06, "loss": 0.1937, "lr_vision_tower": 1.2615509612669274e-06, "sft_loss": 0.093, "step": 703 }, { "distill_loss": 0.2006, "epoch": 0.43, "grad_norm": 1.0166353582980177, "learning_rate": 6.298113829574448e-06, "loss": 0.1858, "lr_vision_tower": 1.2596227659148895e-06, "sft_loss": 0.0855, "step": 704 }, { "distill_loss": 0.1904, "epoch": 0.43, "grad_norm": 0.8758830989537971, "learning_rate": 6.288467674899073e-06, "loss": 0.1957, "lr_vision_tower": 1.2576935349798146e-06, "sft_loss": 0.1005, "step": 705 }, { "distill_loss": 0.1954, "epoch": 0.44, "grad_norm": 0.9854115500198026, "learning_rate": 6.278816380785082e-06, "loss": 0.1988, "lr_vision_tower": 1.2557632761570163e-06, "sft_loss": 0.1012, "step": 706 }, { "distill_loss": 0.1981, "epoch": 0.44, "grad_norm": 0.7687752047841245, "learning_rate": 6.26915998572955e-06, "loss": 0.1935, "lr_vision_tower": 1.25383199714591e-06, "sft_loss": 0.0944, "step": 707 }, { "distill_loss": 0.1992, "epoch": 0.44, "grad_norm": 0.8454332325509406, "learning_rate": 6.259498528249895e-06, "loss": 0.211, "lr_vision_tower": 1.2518997056499787e-06, "sft_loss": 0.1114, "step": 708 }, { "distill_loss": 0.2012, "epoch": 0.44, "grad_norm": 0.6772705995587684, "learning_rate": 6.249832046883729e-06, "loss": 0.1873, "lr_vision_tower": 1.2499664093767458e-06, "sft_loss": 0.0867, "step": 709 }, { "distill_loss": 0.1863, "epoch": 0.44, "grad_norm": 0.9063117479089767, "learning_rate": 6.240160580188703e-06, "loss": 0.196, "lr_vision_tower": 1.2480321160377406e-06, "sft_loss": 0.1029, "step": 710 }, { "distill_loss": 0.1873, "epoch": 0.44, "grad_norm": 0.8230892192267133, "learning_rate": 6.230484166742358e-06, "loss": 0.202, "lr_vision_tower": 1.2460968333484713e-06, "sft_loss": 0.1083, "step": 711 }, { "distill_loss": 0.193, "epoch": 0.44, "grad_norm": 0.7220571424278694, "learning_rate": 6.2208028451419575e-06, "loss": 0.1912, "lr_vision_tower": 1.2441605690283914e-06, "sft_loss": 0.0947, "step": 712 }, { "distill_loss": 0.2007, "epoch": 0.44, "grad_norm": 0.8910382524878243, "learning_rate": 6.21111665400435e-06, "loss": 0.1967, "lr_vision_tower": 1.2422233308008697e-06, "sft_loss": 0.0964, "step": 713 }, { "distill_loss": 0.1883, "epoch": 0.44, "grad_norm": 0.7164326009951378, "learning_rate": 6.201425631965805e-06, "loss": 0.1809, "lr_vision_tower": 1.2402851263931608e-06, "sft_loss": 0.0867, "step": 714 }, { "distill_loss": 0.1953, "epoch": 0.44, "grad_norm": 1.011909941652622, "learning_rate": 6.191729817681863e-06, "loss": 0.1924, "lr_vision_tower": 1.2383459635363724e-06, "sft_loss": 0.0947, "step": 715 }, { "distill_loss": 0.1919, "epoch": 0.44, "grad_norm": 1.037069363045201, "learning_rate": 6.182029249827178e-06, "loss": 0.1867, "lr_vision_tower": 1.2364058499654353e-06, "sft_loss": 0.0907, "step": 716 }, { "distill_loss": 0.1925, "epoch": 0.44, "grad_norm": 0.9194889761138951, "learning_rate": 6.172323967095364e-06, "loss": 0.2021, "lr_vision_tower": 1.2344647934190726e-06, "sft_loss": 0.1059, "step": 717 }, { "distill_loss": 0.1932, "epoch": 0.44, "grad_norm": 1.0835481425243974, "learning_rate": 6.162614008198845e-06, "loss": 0.1858, "lr_vision_tower": 1.232522801639769e-06, "sft_loss": 0.0892, "step": 718 }, { "distill_loss": 0.208, "epoch": 0.44, "grad_norm": 0.8685946419806038, "learning_rate": 6.152899411868695e-06, "loss": 0.1963, "lr_vision_tower": 1.2305798823737387e-06, "sft_loss": 0.0923, "step": 719 }, { "distill_loss": 0.1988, "epoch": 0.44, "grad_norm": 0.7373286461758755, "learning_rate": 6.143180216854488e-06, "loss": 0.1965, "lr_vision_tower": 1.2286360433708973e-06, "sft_loss": 0.0972, "step": 720 }, { "distill_loss": 0.1986, "epoch": 0.44, "grad_norm": 1.1855374226315722, "learning_rate": 6.133456461924138e-06, "loss": 0.174, "lr_vision_tower": 1.2266912923848274e-06, "sft_loss": 0.0747, "step": 721 }, { "distill_loss": 0.2138, "epoch": 0.45, "grad_norm": 0.8006586141994952, "learning_rate": 6.123728185863749e-06, "loss": 0.2209, "lr_vision_tower": 1.2247456371727496e-06, "sft_loss": 0.1141, "step": 722 }, { "distill_loss": 0.1852, "epoch": 0.45, "grad_norm": 0.7890521792504569, "learning_rate": 6.11399542747746e-06, "loss": 0.1749, "lr_vision_tower": 1.2227990854954918e-06, "sft_loss": 0.0823, "step": 723 }, { "distill_loss": 0.2041, "epoch": 0.45, "grad_norm": 0.7631502338292534, "learning_rate": 6.1042582255872875e-06, "loss": 0.189, "lr_vision_tower": 1.2208516451174573e-06, "sft_loss": 0.087, "step": 724 }, { "distill_loss": 0.183, "epoch": 0.45, "grad_norm": 0.7231190448762798, "learning_rate": 6.094516619032975e-06, "loss": 0.1818, "lr_vision_tower": 1.218903323806595e-06, "sft_loss": 0.0903, "step": 725 }, { "distill_loss": 0.1883, "epoch": 0.45, "grad_norm": 0.8118216628489985, "learning_rate": 6.08477064667183e-06, "loss": 0.193, "lr_vision_tower": 1.216954129334366e-06, "sft_loss": 0.0988, "step": 726 }, { "distill_loss": 0.186, "epoch": 0.45, "grad_norm": 1.7179334190105078, "learning_rate": 6.075020347378582e-06, "loss": 0.1919, "lr_vision_tower": 1.2150040694757162e-06, "sft_loss": 0.0989, "step": 727 }, { "distill_loss": 0.197, "epoch": 0.45, "grad_norm": 0.792348400870632, "learning_rate": 6.065265760045212e-06, "loss": 0.2003, "lr_vision_tower": 1.213053152009042e-06, "sft_loss": 0.1018, "step": 728 }, { "distill_loss": 0.2014, "epoch": 0.45, "grad_norm": 0.8917109579974977, "learning_rate": 6.055506923580808e-06, "loss": 0.2027, "lr_vision_tower": 1.2111013847161614e-06, "sft_loss": 0.1021, "step": 729 }, { "distill_loss": 0.1994, "epoch": 0.45, "grad_norm": 1.0815411882360326, "learning_rate": 6.04574387691141e-06, "loss": 0.2009, "lr_vision_tower": 1.2091487753822819e-06, "sft_loss": 0.1012, "step": 730 }, { "distill_loss": 0.1905, "epoch": 0.45, "grad_norm": 1.0777032717015338, "learning_rate": 6.035976658979846e-06, "loss": 0.1936, "lr_vision_tower": 1.207195331795969e-06, "sft_loss": 0.0983, "step": 731 }, { "distill_loss": 0.1957, "epoch": 0.45, "grad_norm": 0.8533620995756303, "learning_rate": 6.026205308745587e-06, "loss": 0.1945, "lr_vision_tower": 1.2052410617491173e-06, "sft_loss": 0.0966, "step": 732 }, { "distill_loss": 0.1918, "epoch": 0.45, "grad_norm": 0.7736324553106978, "learning_rate": 6.016429865184589e-06, "loss": 0.1827, "lr_vision_tower": 1.2032859730369175e-06, "sft_loss": 0.0868, "step": 733 }, { "distill_loss": 0.1902, "epoch": 0.45, "grad_norm": 0.8337560222539857, "learning_rate": 6.0066503672891276e-06, "loss": 0.1922, "lr_vision_tower": 1.2013300734578253e-06, "sft_loss": 0.0971, "step": 734 }, { "distill_loss": 0.1969, "epoch": 0.45, "grad_norm": 0.7794341211826351, "learning_rate": 5.996866854067654e-06, "loss": 0.1968, "lr_vision_tower": 1.1993733708135305e-06, "sft_loss": 0.0984, "step": 735 }, { "distill_loss": 0.1906, "epoch": 0.45, "grad_norm": 0.7926907717188684, "learning_rate": 5.987079364544639e-06, "loss": 0.1928, "lr_vision_tower": 1.1974158729089277e-06, "sft_loss": 0.0975, "step": 736 }, { "distill_loss": 0.2044, "epoch": 0.45, "grad_norm": 0.7735704653457495, "learning_rate": 5.9772879377604095e-06, "loss": 0.1996, "lr_vision_tower": 1.195457587552082e-06, "sft_loss": 0.0974, "step": 737 }, { "distill_loss": 0.185, "epoch": 0.45, "grad_norm": 0.8462289533961264, "learning_rate": 5.967492612770999e-06, "loss": 0.1885, "lr_vision_tower": 1.1934985225541995e-06, "sft_loss": 0.096, "step": 738 }, { "distill_loss": 0.1869, "epoch": 0.46, "grad_norm": 1.3915009676173877, "learning_rate": 5.957693428647993e-06, "loss": 0.1786, "lr_vision_tower": 1.1915386857295984e-06, "sft_loss": 0.0852, "step": 739 }, { "distill_loss": 0.194, "epoch": 0.46, "grad_norm": 0.7203143012639911, "learning_rate": 5.947890424478363e-06, "loss": 0.1913, "lr_vision_tower": 1.1895780848956725e-06, "sft_loss": 0.0943, "step": 740 }, { "distill_loss": 0.1966, "epoch": 0.46, "grad_norm": 0.7547914457959757, "learning_rate": 5.938083639364326e-06, "loss": 0.1842, "lr_vision_tower": 1.187616727872865e-06, "sft_loss": 0.0859, "step": 741 }, { "distill_loss": 0.1952, "epoch": 0.46, "grad_norm": 0.7585529201173037, "learning_rate": 5.928273112423177e-06, "loss": 0.1851, "lr_vision_tower": 1.1856546224846353e-06, "sft_loss": 0.0875, "step": 742 }, { "distill_loss": 0.1942, "epoch": 0.46, "grad_norm": 1.060230195957877, "learning_rate": 5.918458882787137e-06, "loss": 0.1829, "lr_vision_tower": 1.1836917765574273e-06, "sft_loss": 0.0858, "step": 743 }, { "distill_loss": 0.1948, "epoch": 0.46, "grad_norm": 0.7194630149135056, "learning_rate": 5.908640989603192e-06, "loss": 0.1884, "lr_vision_tower": 1.1817281979206382e-06, "sft_loss": 0.0909, "step": 744 }, { "distill_loss": 0.1978, "epoch": 0.46, "grad_norm": 0.8067767356792376, "learning_rate": 5.898819472032948e-06, "loss": 0.1837, "lr_vision_tower": 1.1797638944065896e-06, "sft_loss": 0.0848, "step": 745 }, { "distill_loss": 0.1879, "epoch": 0.46, "grad_norm": 0.8584284558647498, "learning_rate": 5.888994369252464e-06, "loss": 0.1985, "lr_vision_tower": 1.1777988738504927e-06, "sft_loss": 0.1046, "step": 746 }, { "distill_loss": 0.1966, "epoch": 0.46, "grad_norm": 1.7753261898065988, "learning_rate": 5.8791657204521e-06, "loss": 0.1957, "lr_vision_tower": 1.17583314409042e-06, "sft_loss": 0.0974, "step": 747 }, { "distill_loss": 0.2055, "epoch": 0.46, "grad_norm": 1.0142273918191138, "learning_rate": 5.869333564836361e-06, "loss": 0.2022, "lr_vision_tower": 1.173866712967272e-06, "sft_loss": 0.0994, "step": 748 }, { "distill_loss": 0.1941, "epoch": 0.46, "grad_norm": 1.037195096163142, "learning_rate": 5.859497941623739e-06, "loss": 0.2009, "lr_vision_tower": 1.1718995883247477e-06, "sft_loss": 0.1039, "step": 749 }, { "distill_loss": 0.1972, "epoch": 0.46, "grad_norm": 0.8061583760516234, "learning_rate": 5.849658890046561e-06, "loss": 0.1877, "lr_vision_tower": 1.1699317780093122e-06, "sft_loss": 0.0891, "step": 750 }, { "distill_loss": 0.1964, "epoch": 0.46, "grad_norm": 1.0215822781172204, "learning_rate": 5.839816449350824e-06, "loss": 0.1988, "lr_vision_tower": 1.1679632898701648e-06, "sft_loss": 0.1006, "step": 751 }, { "distill_loss": 0.1911, "epoch": 0.46, "grad_norm": 0.8027010594996046, "learning_rate": 5.829970658796048e-06, "loss": 0.181, "lr_vision_tower": 1.1659941317592095e-06, "sft_loss": 0.0855, "step": 752 }, { "distill_loss": 0.2018, "epoch": 0.46, "grad_norm": 1.0471245707381962, "learning_rate": 5.820121557655109e-06, "loss": 0.2007, "lr_vision_tower": 1.1640243115310217e-06, "sft_loss": 0.0998, "step": 753 }, { "distill_loss": 0.1851, "epoch": 0.46, "grad_norm": 0.7241309906954324, "learning_rate": 5.8102691852140965e-06, "loss": 0.176, "lr_vision_tower": 1.1620538370428192e-06, "sft_loss": 0.0834, "step": 754 }, { "distill_loss": 0.1929, "epoch": 0.47, "grad_norm": 0.9059869279956185, "learning_rate": 5.800413580772143e-06, "loss": 0.1925, "lr_vision_tower": 1.1600827161544284e-06, "sft_loss": 0.096, "step": 755 }, { "distill_loss": 0.195, "epoch": 0.47, "grad_norm": 0.6859312756940329, "learning_rate": 5.790554783641271e-06, "loss": 0.1985, "lr_vision_tower": 1.158110956728254e-06, "sft_loss": 0.101, "step": 756 }, { "distill_loss": 0.1931, "epoch": 0.47, "grad_norm": 0.942099408619899, "learning_rate": 5.780692833146245e-06, "loss": 0.1802, "lr_vision_tower": 1.1561385666292487e-06, "sft_loss": 0.0836, "step": 757 }, { "distill_loss": 0.1922, "epoch": 0.47, "grad_norm": 0.9466318965631267, "learning_rate": 5.7708277686243995e-06, "loss": 0.1912, "lr_vision_tower": 1.1541655537248798e-06, "sft_loss": 0.0951, "step": 758 }, { "distill_loss": 0.1844, "epoch": 0.47, "grad_norm": 0.8977196761122267, "learning_rate": 5.760959629425498e-06, "loss": 0.1775, "lr_vision_tower": 1.1521919258850994e-06, "sft_loss": 0.0853, "step": 759 }, { "distill_loss": 0.1955, "epoch": 0.47, "grad_norm": 1.0610183278778276, "learning_rate": 5.751088454911563e-06, "loss": 0.1939, "lr_vision_tower": 1.1502176909823124e-06, "sft_loss": 0.0961, "step": 760 }, { "distill_loss": 0.1852, "epoch": 0.47, "grad_norm": 1.5214882889769323, "learning_rate": 5.741214284456727e-06, "loss": 0.1733, "lr_vision_tower": 1.1482428568913452e-06, "sft_loss": 0.0807, "step": 761 }, { "distill_loss": 0.1968, "epoch": 0.47, "grad_norm": 1.375345669838741, "learning_rate": 5.731337157447068e-06, "loss": 0.1987, "lr_vision_tower": 1.1462674314894134e-06, "sft_loss": 0.1003, "step": 762 }, { "distill_loss": 0.1975, "epoch": 0.47, "grad_norm": 1.1135833435400229, "learning_rate": 5.721457113280465e-06, "loss": 0.1943, "lr_vision_tower": 1.1442914226560929e-06, "sft_loss": 0.0955, "step": 763 }, { "distill_loss": 0.1865, "epoch": 0.47, "grad_norm": 1.3428194962217213, "learning_rate": 5.711574191366427e-06, "loss": 0.186, "lr_vision_tower": 1.1423148382732853e-06, "sft_loss": 0.0927, "step": 764 }, { "distill_loss": 0.1852, "epoch": 0.47, "grad_norm": 0.9365683455058849, "learning_rate": 5.701688431125942e-06, "loss": 0.1873, "lr_vision_tower": 1.1403376862251884e-06, "sft_loss": 0.0947, "step": 765 }, { "distill_loss": 0.2006, "epoch": 0.47, "grad_norm": 0.7880008210048943, "learning_rate": 5.691799871991323e-06, "loss": 0.1972, "lr_vision_tower": 1.1383599743982644e-06, "sft_loss": 0.0969, "step": 766 }, { "distill_loss": 0.1916, "epoch": 0.47, "grad_norm": 0.9464127973202577, "learning_rate": 5.681908553406044e-06, "loss": 0.2078, "lr_vision_tower": 1.1363817106812086e-06, "sft_loss": 0.112, "step": 767 }, { "distill_loss": 0.1894, "epoch": 0.47, "grad_norm": 0.7967781017933954, "learning_rate": 5.6720145148245866e-06, "loss": 0.1892, "lr_vision_tower": 1.134402902964917e-06, "sft_loss": 0.0945, "step": 768 }, { "distill_loss": 0.2059, "epoch": 0.47, "grad_norm": 0.7057735487383826, "learning_rate": 5.662117795712283e-06, "loss": 0.1962, "lr_vision_tower": 1.1324235591424565e-06, "sft_loss": 0.0933, "step": 769 }, { "distill_loss": 0.1871, "epoch": 0.47, "grad_norm": 0.6800996765399402, "learning_rate": 5.652218435545157e-06, "loss": 0.1863, "lr_vision_tower": 1.1304436871090313e-06, "sft_loss": 0.0928, "step": 770 }, { "distill_loss": 0.1885, "epoch": 0.48, "grad_norm": 0.694024710176108, "learning_rate": 5.642316473809766e-06, "loss": 0.1877, "lr_vision_tower": 1.128463294761953e-06, "sft_loss": 0.0935, "step": 771 }, { "distill_loss": 0.1875, "epoch": 0.48, "grad_norm": 1.1365452103081484, "learning_rate": 5.632411950003047e-06, "loss": 0.1813, "lr_vision_tower": 1.1264823900006092e-06, "sft_loss": 0.0876, "step": 772 }, { "distill_loss": 0.1838, "epoch": 0.48, "grad_norm": 0.8474659594699888, "learning_rate": 5.622504903632155e-06, "loss": 0.2017, "lr_vision_tower": 1.1245009807264306e-06, "sft_loss": 0.1098, "step": 773 }, { "distill_loss": 0.1942, "epoch": 0.48, "grad_norm": 0.7877355261156028, "learning_rate": 5.6125953742143025e-06, "loss": 0.1835, "lr_vision_tower": 1.1225190748428602e-06, "sft_loss": 0.0864, "step": 774 }, { "distill_loss": 0.197, "epoch": 0.48, "grad_norm": 0.7237427325700606, "learning_rate": 5.6026834012766155e-06, "loss": 0.1907, "lr_vision_tower": 1.1205366802553228e-06, "sft_loss": 0.0922, "step": 775 }, { "distill_loss": 0.1848, "epoch": 0.48, "grad_norm": 0.979856044275378, "learning_rate": 5.592769024355962e-06, "loss": 0.1667, "lr_vision_tower": 1.1185538048711921e-06, "sft_loss": 0.0743, "step": 776 }, { "distill_loss": 0.1842, "epoch": 0.48, "grad_norm": 0.8965562021604839, "learning_rate": 5.5828522829987965e-06, "loss": 0.1859, "lr_vision_tower": 1.116570456599759e-06, "sft_loss": 0.0938, "step": 777 }, { "distill_loss": 0.1905, "epoch": 0.48, "grad_norm": 0.7864374549091181, "learning_rate": 5.57293321676101e-06, "loss": 0.1829, "lr_vision_tower": 1.1145866433522018e-06, "sft_loss": 0.0876, "step": 778 }, { "distill_loss": 0.1993, "epoch": 0.48, "grad_norm": 1.004525072917223, "learning_rate": 5.563011865207763e-06, "loss": 0.1932, "lr_vision_tower": 1.1126023730415524e-06, "sft_loss": 0.0935, "step": 779 }, { "distill_loss": 0.1985, "epoch": 0.48, "grad_norm": 0.821125672377385, "learning_rate": 5.553088267913331e-06, "loss": 0.1854, "lr_vision_tower": 1.1106176535826661e-06, "sft_loss": 0.0861, "step": 780 }, { "distill_loss": 0.1839, "epoch": 0.48, "grad_norm": 1.0366867346454114, "learning_rate": 5.543162464460953e-06, "loss": 0.1796, "lr_vision_tower": 1.1086324928921904e-06, "sft_loss": 0.0877, "step": 781 }, { "distill_loss": 0.1834, "epoch": 0.48, "grad_norm": 0.9242505931821262, "learning_rate": 5.533234494442662e-06, "loss": 0.1809, "lr_vision_tower": 1.1066468988885323e-06, "sft_loss": 0.0892, "step": 782 }, { "distill_loss": 0.1951, "epoch": 0.48, "grad_norm": 1.1786268594539728, "learning_rate": 5.523304397459135e-06, "loss": 0.1834, "lr_vision_tower": 1.1046608794918268e-06, "sft_loss": 0.0858, "step": 783 }, { "distill_loss": 0.1805, "epoch": 0.48, "grad_norm": 0.9507863582247095, "learning_rate": 5.5133722131195345e-06, "loss": 0.1992, "lr_vision_tower": 1.1026744426239068e-06, "sft_loss": 0.1089, "step": 784 }, { "distill_loss": 0.1977, "epoch": 0.48, "grad_norm": 0.9023875863482228, "learning_rate": 5.503437981041347e-06, "loss": 0.2052, "lr_vision_tower": 1.1006875962082693e-06, "sft_loss": 0.1064, "step": 785 }, { "distill_loss": 0.1909, "epoch": 0.48, "grad_norm": 0.8885731903252212, "learning_rate": 5.493501740850228e-06, "loss": 0.2034, "lr_vision_tower": 1.0987003481700454e-06, "sft_loss": 0.108, "step": 786 }, { "distill_loss": 0.1906, "epoch": 0.49, "grad_norm": 0.7647905959782842, "learning_rate": 5.4835635321798435e-06, "loss": 0.1992, "lr_vision_tower": 1.0967127064359684e-06, "sft_loss": 0.1039, "step": 787 }, { "distill_loss": 0.203, "epoch": 0.49, "grad_norm": 0.6872939728701946, "learning_rate": 5.47362339467171e-06, "loss": 0.2049, "lr_vision_tower": 1.0947246789343417e-06, "sft_loss": 0.1034, "step": 788 }, { "distill_loss": 0.1974, "epoch": 0.49, "grad_norm": 0.9448184542006406, "learning_rate": 5.4636813679750355e-06, "loss": 0.1946, "lr_vision_tower": 1.0927362735950069e-06, "sft_loss": 0.0959, "step": 789 }, { "distill_loss": 0.2015, "epoch": 0.49, "grad_norm": 0.7932951195216744, "learning_rate": 5.453737491746572e-06, "loss": 0.1879, "lr_vision_tower": 1.0907474983493144e-06, "sft_loss": 0.0872, "step": 790 }, { "distill_loss": 0.1864, "epoch": 0.49, "grad_norm": 0.6728609055391859, "learning_rate": 5.44379180565044e-06, "loss": 0.1659, "lr_vision_tower": 1.0887583611300878e-06, "sft_loss": 0.0727, "step": 791 }, { "distill_loss": 0.1907, "epoch": 0.49, "grad_norm": 0.7181982619947234, "learning_rate": 5.433844349357983e-06, "loss": 0.1756, "lr_vision_tower": 1.0867688698715964e-06, "sft_loss": 0.0802, "step": 792 }, { "distill_loss": 0.1985, "epoch": 0.49, "grad_norm": 0.8681843619700471, "learning_rate": 5.423895162547604e-06, "loss": 0.1987, "lr_vision_tower": 1.0847790325095207e-06, "sft_loss": 0.0994, "step": 793 }, { "distill_loss": 0.1827, "epoch": 0.49, "grad_norm": 0.692147936291376, "learning_rate": 5.4139442849046135e-06, "loss": 0.1759, "lr_vision_tower": 1.0827888569809224e-06, "sft_loss": 0.0845, "step": 794 }, { "distill_loss": 0.2004, "epoch": 0.49, "grad_norm": 0.8264525995477296, "learning_rate": 5.403991756121057e-06, "loss": 0.2011, "lr_vision_tower": 1.080798351224211e-06, "sft_loss": 0.1009, "step": 795 }, { "distill_loss": 0.1863, "epoch": 0.49, "grad_norm": 0.8892052977457776, "learning_rate": 5.3940376158955765e-06, "loss": 0.1947, "lr_vision_tower": 1.078807523179115e-06, "sft_loss": 0.1016, "step": 796 }, { "distill_loss": 0.1928, "epoch": 0.49, "grad_norm": 0.9072227531208005, "learning_rate": 5.384081903933235e-06, "loss": 0.193, "lr_vision_tower": 1.0768163807866468e-06, "sft_loss": 0.0966, "step": 797 }, { "distill_loss": 0.1946, "epoch": 0.49, "grad_norm": 0.7544520750569393, "learning_rate": 5.3741246599453655e-06, "loss": 0.1806, "lr_vision_tower": 1.074824931989073e-06, "sft_loss": 0.0832, "step": 798 }, { "distill_loss": 0.1839, "epoch": 0.49, "grad_norm": 0.9275931131951421, "learning_rate": 5.364165923649417e-06, "loss": 0.1926, "lr_vision_tower": 1.0728331847298832e-06, "sft_loss": 0.1006, "step": 799 }, { "distill_loss": 0.1974, "epoch": 0.49, "grad_norm": 1.2600292702854632, "learning_rate": 5.354205734768783e-06, "loss": 0.1949, "lr_vision_tower": 1.0708411469537565e-06, "sft_loss": 0.0962, "step": 800 }, { "distill_loss": 0.1799, "epoch": 0.49, "grad_norm": 1.0599471606823179, "learning_rate": 5.344244133032657e-06, "loss": 0.1922, "lr_vision_tower": 1.0688488266065312e-06, "sft_loss": 0.1022, "step": 801 }, { "distill_loss": 0.1852, "epoch": 0.49, "grad_norm": 0.7204473881696621, "learning_rate": 5.334281158175866e-06, "loss": 0.1792, "lr_vision_tower": 1.0668562316351731e-06, "sft_loss": 0.0866, "step": 802 }, { "distill_loss": 0.1845, "epoch": 0.49, "grad_norm": 1.1676690569549786, "learning_rate": 5.324316849938715e-06, "loss": 0.1987, "lr_vision_tower": 1.0648633699877429e-06, "sft_loss": 0.1064, "step": 803 }, { "distill_loss": 0.1983, "epoch": 0.5, "grad_norm": 0.8389354599988459, "learning_rate": 5.314351248066827e-06, "loss": 0.191, "lr_vision_tower": 1.0628702496133653e-06, "sft_loss": 0.0918, "step": 804 }, { "distill_loss": 0.1831, "epoch": 0.5, "grad_norm": 1.0844059316277044, "learning_rate": 5.304384392310986e-06, "loss": 0.1839, "lr_vision_tower": 1.0608768784621971e-06, "sft_loss": 0.0924, "step": 805 }, { "distill_loss": 0.1961, "epoch": 0.5, "grad_norm": 0.9417605883406137, "learning_rate": 5.294416322426975e-06, "loss": 0.1725, "lr_vision_tower": 1.058883264485395e-06, "sft_loss": 0.0744, "step": 806 }, { "distill_loss": 0.1768, "epoch": 0.5, "grad_norm": 0.6798182161117878, "learning_rate": 5.284447078175421e-06, "loss": 0.1648, "lr_vision_tower": 1.056889415635084e-06, "sft_loss": 0.0764, "step": 807 }, { "distill_loss": 0.1806, "epoch": 0.5, "grad_norm": 0.8117446534937985, "learning_rate": 5.274476699321638e-06, "loss": 0.1755, "lr_vision_tower": 1.0548953398643274e-06, "sft_loss": 0.0852, "step": 808 }, { "distill_loss": 0.1872, "epoch": 0.5, "grad_norm": 0.7869515188166679, "learning_rate": 5.264505225635463e-06, "loss": 0.1947, "lr_vision_tower": 1.0529010451270924e-06, "sft_loss": 0.1011, "step": 809 }, { "distill_loss": 0.1837, "epoch": 0.5, "grad_norm": 1.155718577043765, "learning_rate": 5.254532696891098e-06, "loss": 0.1645, "lr_vision_tower": 1.0509065393782196e-06, "sft_loss": 0.0727, "step": 810 }, { "distill_loss": 0.1988, "epoch": 0.5, "grad_norm": 0.7604795236726414, "learning_rate": 5.244559152866958e-06, "loss": 0.1896, "lr_vision_tower": 1.0489118305733915e-06, "sft_loss": 0.0902, "step": 811 }, { "distill_loss": 0.1926, "epoch": 0.5, "grad_norm": 0.7383494845231566, "learning_rate": 5.234584633345508e-06, "loss": 0.204, "lr_vision_tower": 1.0469169266691013e-06, "sft_loss": 0.1077, "step": 812 }, { "distill_loss": 0.1975, "epoch": 0.5, "grad_norm": 0.7249251403797687, "learning_rate": 5.224609178113097e-06, "loss": 0.1731, "lr_vision_tower": 1.0449218356226192e-06, "sft_loss": 0.0743, "step": 813 }, { "distill_loss": 0.1956, "epoch": 0.5, "grad_norm": 0.7186260956569978, "learning_rate": 5.214632826959816e-06, "loss": 0.1918, "lr_vision_tower": 1.042926565391963e-06, "sft_loss": 0.094, "step": 814 }, { "distill_loss": 0.1903, "epoch": 0.5, "grad_norm": 0.8268000557842075, "learning_rate": 5.204655619679322e-06, "loss": 0.1901, "lr_vision_tower": 1.0409311239358642e-06, "sft_loss": 0.095, "step": 815 }, { "distill_loss": 0.1862, "epoch": 0.5, "grad_norm": 0.9411357791385578, "learning_rate": 5.194677596068689e-06, "loss": 0.1924, "lr_vision_tower": 1.0389355192137377e-06, "sft_loss": 0.0993, "step": 816 }, { "distill_loss": 0.1891, "epoch": 0.5, "grad_norm": 2.536546320203387, "learning_rate": 5.18469879592825e-06, "loss": 0.1877, "lr_vision_tower": 1.0369397591856498e-06, "sft_loss": 0.0932, "step": 817 }, { "distill_loss": 0.1936, "epoch": 0.5, "grad_norm": 0.8190559562638263, "learning_rate": 5.174719259061434e-06, "loss": 0.1957, "lr_vision_tower": 1.0349438518122866e-06, "sft_loss": 0.0989, "step": 818 }, { "distill_loss": 0.1956, "epoch": 0.5, "grad_norm": 0.786477600682616, "learning_rate": 5.164739025274604e-06, "loss": 0.1881, "lr_vision_tower": 1.0329478050549207e-06, "sft_loss": 0.0903, "step": 819 }, { "distill_loss": 0.1827, "epoch": 0.51, "grad_norm": 0.7794317820172939, "learning_rate": 5.15475813437691e-06, "loss": 0.1805, "lr_vision_tower": 1.030951626875382e-06, "sft_loss": 0.0891, "step": 820 }, { "distill_loss": 0.1848, "epoch": 0.51, "grad_norm": 0.7746402169686504, "learning_rate": 5.1447766261801215e-06, "loss": 0.1958, "lr_vision_tower": 1.028955325236024e-06, "sft_loss": 0.1034, "step": 821 }, { "distill_loss": 0.1906, "epoch": 0.51, "grad_norm": 0.6889588444719673, "learning_rate": 5.134794540498463e-06, "loss": 0.1743, "lr_vision_tower": 1.0269589080996925e-06, "sft_loss": 0.079, "step": 822 }, { "distill_loss": 0.1918, "epoch": 0.51, "grad_norm": 1.1812471453496167, "learning_rate": 5.124811917148476e-06, "loss": 0.1926, "lr_vision_tower": 1.0249623834296949e-06, "sft_loss": 0.0967, "step": 823 }, { "distill_loss": 0.183, "epoch": 0.51, "grad_norm": 1.2122462577211872, "learning_rate": 5.114828795948833e-06, "loss": 0.1693, "lr_vision_tower": 1.0229657591897665e-06, "sft_loss": 0.0778, "step": 824 }, { "distill_loss": 0.181, "epoch": 0.51, "grad_norm": 0.8686222891175932, "learning_rate": 5.104845216720198e-06, "loss": 0.1759, "lr_vision_tower": 1.0209690433440395e-06, "sft_loss": 0.0854, "step": 825 }, { "distill_loss": 0.189, "epoch": 0.51, "grad_norm": 0.9134304534106583, "learning_rate": 5.0948612192850666e-06, "loss": 0.1992, "lr_vision_tower": 1.0189722438570132e-06, "sft_loss": 0.1047, "step": 826 }, { "distill_loss": 0.1901, "epoch": 0.51, "grad_norm": 0.8986486676688894, "learning_rate": 5.084876843467593e-06, "loss": 0.189, "lr_vision_tower": 1.0169753686935186e-06, "sft_loss": 0.094, "step": 827 }, { "distill_loss": 0.1981, "epoch": 0.51, "grad_norm": 1.6201704174271152, "learning_rate": 5.074892129093451e-06, "loss": 0.1866, "lr_vision_tower": 1.01497842581869e-06, "sft_loss": 0.0875, "step": 828 }, { "distill_loss": 0.2022, "epoch": 0.51, "grad_norm": 0.8241103389406551, "learning_rate": 5.064907115989655e-06, "loss": 0.2, "lr_vision_tower": 1.0129814231979308e-06, "sft_loss": 0.0989, "step": 829 }, { "distill_loss": 0.1941, "epoch": 0.51, "grad_norm": 0.7844111203661561, "learning_rate": 5.0549218439844185e-06, "loss": 0.193, "lr_vision_tower": 1.0109843687968835e-06, "sft_loss": 0.0959, "step": 830 }, { "distill_loss": 0.1929, "epoch": 0.51, "grad_norm": 0.8275889396937546, "learning_rate": 5.044936352906983e-06, "loss": 0.2052, "lr_vision_tower": 1.0089872705813964e-06, "sft_loss": 0.1088, "step": 831 }, { "distill_loss": 0.1905, "epoch": 0.51, "grad_norm": 0.6646644876558052, "learning_rate": 5.034950682587467e-06, "loss": 0.1979, "lr_vision_tower": 1.0069901365174934e-06, "sft_loss": 0.1027, "step": 832 }, { "distill_loss": 0.1892, "epoch": 0.51, "grad_norm": 0.7285524037010227, "learning_rate": 5.024964872856703e-06, "loss": 0.1779, "lr_vision_tower": 1.0049929745713405e-06, "sft_loss": 0.0833, "step": 833 }, { "distill_loss": 0.1928, "epoch": 0.51, "grad_norm": 0.8774007588041484, "learning_rate": 5.014978963546077e-06, "loss": 0.2156, "lr_vision_tower": 1.0029957927092153e-06, "sft_loss": 0.1192, "step": 834 }, { "distill_loss": 0.1805, "epoch": 0.51, "grad_norm": 0.7996277828935758, "learning_rate": 5.004992994487378e-06, "loss": 0.1771, "lr_vision_tower": 1.0009985988974755e-06, "sft_loss": 0.0868, "step": 835 }, { "distill_loss": 0.1862, "epoch": 0.52, "grad_norm": 0.6138681896843119, "learning_rate": 4.995007005512625e-06, "loss": 0.1857, "lr_vision_tower": 9.990014011025246e-07, "sft_loss": 0.0926, "step": 836 }, { "distill_loss": 0.182, "epoch": 0.52, "grad_norm": 0.7023491775395385, "learning_rate": 4.985021036453924e-06, "loss": 0.1901, "lr_vision_tower": 9.970042072907848e-07, "sft_loss": 0.0991, "step": 837 }, { "distill_loss": 0.1863, "epoch": 0.52, "grad_norm": 0.719239628186211, "learning_rate": 4.9750351271432975e-06, "loss": 0.1845, "lr_vision_tower": 9.950070254286594e-07, "sft_loss": 0.0913, "step": 838 }, { "distill_loss": 0.1853, "epoch": 0.52, "grad_norm": 0.6651157733676358, "learning_rate": 4.965049317412535e-06, "loss": 0.1886, "lr_vision_tower": 9.930098634825068e-07, "sft_loss": 0.096, "step": 839 }, { "distill_loss": 0.1961, "epoch": 0.52, "grad_norm": 0.9333796985727657, "learning_rate": 4.955063647093019e-06, "loss": 0.2003, "lr_vision_tower": 9.910127294186035e-07, "sft_loss": 0.1022, "step": 840 }, { "distill_loss": 0.1843, "epoch": 0.52, "grad_norm": 0.8178793351524162, "learning_rate": 4.945078156015582e-06, "loss": 0.1693, "lr_vision_tower": 9.890156312031162e-07, "sft_loss": 0.0772, "step": 841 }, { "distill_loss": 0.196, "epoch": 0.52, "grad_norm": 0.6228454819066078, "learning_rate": 4.935092884010347e-06, "loss": 0.1848, "lr_vision_tower": 9.870185768020691e-07, "sft_loss": 0.0868, "step": 842 }, { "distill_loss": 0.1918, "epoch": 0.52, "grad_norm": 0.850573969346058, "learning_rate": 4.925107870906551e-06, "loss": 0.1948, "lr_vision_tower": 9.8502157418131e-07, "sft_loss": 0.0989, "step": 843 }, { "distill_loss": 0.1825, "epoch": 0.52, "grad_norm": 0.9744708397484565, "learning_rate": 4.915123156532408e-06, "loss": 0.1784, "lr_vision_tower": 9.830246313064815e-07, "sft_loss": 0.0871, "step": 844 }, { "distill_loss": 0.1862, "epoch": 0.52, "grad_norm": 0.9149770911342321, "learning_rate": 4.905138780714935e-06, "loss": 0.184, "lr_vision_tower": 9.810277561429869e-07, "sft_loss": 0.0909, "step": 845 }, { "distill_loss": 0.206, "epoch": 0.52, "grad_norm": 0.8029818808106951, "learning_rate": 4.895154783279803e-06, "loss": 0.2112, "lr_vision_tower": 9.790309566559604e-07, "sft_loss": 0.1082, "step": 846 }, { "distill_loss": 0.1876, "epoch": 0.52, "grad_norm": 0.6693147657152346, "learning_rate": 4.885171204051169e-06, "loss": 0.195, "lr_vision_tower": 9.770342408102336e-07, "sft_loss": 0.1012, "step": 847 }, { "distill_loss": 0.1855, "epoch": 0.52, "grad_norm": 0.6342629457096771, "learning_rate": 4.875188082851527e-06, "loss": 0.174, "lr_vision_tower": 9.750376165703054e-07, "sft_loss": 0.0812, "step": 848 }, { "distill_loss": 0.1911, "epoch": 0.52, "grad_norm": 0.6559749371176598, "learning_rate": 4.8652054595015376e-06, "loss": 0.1839, "lr_vision_tower": 9.730410919003074e-07, "sft_loss": 0.0884, "step": 849 }, { "distill_loss": 0.1887, "epoch": 0.52, "grad_norm": 0.8881419511813364, "learning_rate": 4.855223373819879e-06, "loss": 0.194, "lr_vision_tower": 9.710446747639758e-07, "sft_loss": 0.0996, "step": 850 }, { "distill_loss": 0.187, "epoch": 0.52, "grad_norm": 0.830081184380593, "learning_rate": 4.845241865623091e-06, "loss": 0.1981, "lr_vision_tower": 9.69048373124618e-07, "sft_loss": 0.1046, "step": 851 }, { "distill_loss": 0.2019, "epoch": 0.53, "grad_norm": 0.665211285542916, "learning_rate": 4.835260974725397e-06, "loss": 0.1952, "lr_vision_tower": 9.670521949450794e-07, "sft_loss": 0.0943, "step": 852 }, { "distill_loss": 0.2026, "epoch": 0.53, "grad_norm": 0.8056135132290819, "learning_rate": 4.825280740938569e-06, "loss": 0.2104, "lr_vision_tower": 9.650561481877136e-07, "sft_loss": 0.1091, "step": 853 }, { "distill_loss": 0.1901, "epoch": 0.53, "grad_norm": 0.8366991190920509, "learning_rate": 4.8153012040717505e-06, "loss": 0.1794, "lr_vision_tower": 9.6306024081435e-07, "sft_loss": 0.0844, "step": 854 }, { "distill_loss": 0.1899, "epoch": 0.53, "grad_norm": 1.1795582475697919, "learning_rate": 4.805322403931312e-06, "loss": 0.2008, "lr_vision_tower": 9.610644807862623e-07, "sft_loss": 0.1058, "step": 855 }, { "distill_loss": 0.1885, "epoch": 0.53, "grad_norm": 0.7305651391526087, "learning_rate": 4.795344380320679e-06, "loss": 0.1862, "lr_vision_tower": 9.590688760641359e-07, "sft_loss": 0.0919, "step": 856 }, { "distill_loss": 0.1776, "epoch": 0.53, "grad_norm": 0.822062823514925, "learning_rate": 4.785367173040187e-06, "loss": 0.1759, "lr_vision_tower": 9.570734346080372e-07, "sft_loss": 0.0871, "step": 857 }, { "distill_loss": 0.1901, "epoch": 0.53, "grad_norm": 0.8308853465465522, "learning_rate": 4.775390821886904e-06, "loss": 0.1979, "lr_vision_tower": 9.550781643773807e-07, "sft_loss": 0.1028, "step": 858 }, { "distill_loss": 0.1871, "epoch": 0.53, "grad_norm": 0.891248453205617, "learning_rate": 4.765415366654492e-06, "loss": 0.1931, "lr_vision_tower": 9.530830733308984e-07, "sft_loss": 0.0995, "step": 859 }, { "distill_loss": 0.1935, "epoch": 0.53, "grad_norm": 0.8430367155217987, "learning_rate": 4.755440847133043e-06, "loss": 0.1719, "lr_vision_tower": 9.510881694266085e-07, "sft_loss": 0.0751, "step": 860 }, { "distill_loss": 0.1847, "epoch": 0.53, "grad_norm": 0.7572216767335602, "learning_rate": 4.745467303108903e-06, "loss": 0.1955, "lr_vision_tower": 9.490934606217806e-07, "sft_loss": 0.1031, "step": 861 }, { "distill_loss": 0.1849, "epoch": 0.53, "grad_norm": 0.8137733092188938, "learning_rate": 4.73549477436454e-06, "loss": 0.1796, "lr_vision_tower": 9.470989548729078e-07, "sft_loss": 0.0871, "step": 862 }, { "distill_loss": 0.1862, "epoch": 0.53, "grad_norm": 0.9803174706015078, "learning_rate": 4.7255233006783626e-06, "loss": 0.1909, "lr_vision_tower": 9.451046601356725e-07, "sft_loss": 0.0978, "step": 863 }, { "distill_loss": 0.1822, "epoch": 0.53, "grad_norm": 0.9394546731276939, "learning_rate": 4.715552921824581e-06, "loss": 0.1937, "lr_vision_tower": 9.43110584364916e-07, "sft_loss": 0.1026, "step": 864 }, { "distill_loss": 0.2036, "epoch": 0.53, "grad_norm": 0.9578278777151467, "learning_rate": 4.7055836775730265e-06, "loss": 0.1934, "lr_vision_tower": 9.411167355146051e-07, "sft_loss": 0.0916, "step": 865 }, { "distill_loss": 0.195, "epoch": 0.53, "grad_norm": 0.8443744469875727, "learning_rate": 4.695615607689015e-06, "loss": 0.1664, "lr_vision_tower": 9.391231215378029e-07, "sft_loss": 0.0689, "step": 866 }, { "distill_loss": 0.1809, "epoch": 0.53, "grad_norm": 0.7622622283374556, "learning_rate": 4.6856487519331745e-06, "loss": 0.1896, "lr_vision_tower": 9.371297503866348e-07, "sft_loss": 0.0991, "step": 867 }, { "distill_loss": 0.18, "epoch": 0.54, "grad_norm": 0.8629371295061887, "learning_rate": 4.6756831500612846e-06, "loss": 0.1889, "lr_vision_tower": 9.351366300122568e-07, "sft_loss": 0.0989, "step": 868 }, { "distill_loss": 0.1897, "epoch": 0.54, "grad_norm": 0.8050819283053051, "learning_rate": 4.665718841824135e-06, "loss": 0.1935, "lr_vision_tower": 9.33143768364827e-07, "sft_loss": 0.0986, "step": 869 }, { "distill_loss": 0.185, "epoch": 0.54, "grad_norm": 0.8134875340250852, "learning_rate": 4.655755866967345e-06, "loss": 0.1778, "lr_vision_tower": 9.311511733934688e-07, "sft_loss": 0.0853, "step": 870 }, { "distill_loss": 0.1949, "epoch": 0.54, "grad_norm": 0.8053048277563358, "learning_rate": 4.645794265231219e-06, "loss": 0.1916, "lr_vision_tower": 9.291588530462437e-07, "sft_loss": 0.0941, "step": 871 }, { "distill_loss": 0.1764, "epoch": 0.54, "grad_norm": 0.8574651557941073, "learning_rate": 4.6358340763505845e-06, "loss": 0.1922, "lr_vision_tower": 9.271668152701169e-07, "sft_loss": 0.104, "step": 872 }, { "distill_loss": 0.1917, "epoch": 0.54, "grad_norm": 0.9248006235678987, "learning_rate": 4.625875340054635e-06, "loss": 0.1887, "lr_vision_tower": 9.25175068010927e-07, "sft_loss": 0.0929, "step": 873 }, { "distill_loss": 0.1929, "epoch": 0.54, "grad_norm": 1.0058218669037198, "learning_rate": 4.615918096066766e-06, "loss": 0.1777, "lr_vision_tower": 9.231836192133531e-07, "sft_loss": 0.0813, "step": 874 }, { "distill_loss": 0.1922, "epoch": 0.54, "grad_norm": 0.8094250095428527, "learning_rate": 4.6059623841044234e-06, "loss": 0.1872, "lr_vision_tower": 9.211924768208846e-07, "sft_loss": 0.0912, "step": 875 }, { "distill_loss": 0.192, "epoch": 0.54, "grad_norm": 0.8626541806786739, "learning_rate": 4.596008243878944e-06, "loss": 0.207, "lr_vision_tower": 9.192016487757888e-07, "sft_loss": 0.111, "step": 876 }, { "distill_loss": 0.1841, "epoch": 0.54, "grad_norm": 0.8224258007677452, "learning_rate": 4.586055715095387e-06, "loss": 0.1972, "lr_vision_tower": 9.172111430190775e-07, "sft_loss": 0.1052, "step": 877 }, { "distill_loss": 0.1902, "epoch": 0.54, "grad_norm": 0.7717671071363646, "learning_rate": 4.5761048374523975e-06, "loss": 0.2022, "lr_vision_tower": 9.152209674904794e-07, "sft_loss": 0.1071, "step": 878 }, { "distill_loss": 0.1922, "epoch": 0.54, "grad_norm": 1.0100791840611756, "learning_rate": 4.566155650642018e-06, "loss": 0.1877, "lr_vision_tower": 9.132311301284036e-07, "sft_loss": 0.0916, "step": 879 }, { "distill_loss": 0.187, "epoch": 0.54, "grad_norm": 0.726072452175227, "learning_rate": 4.556208194349562e-06, "loss": 0.1981, "lr_vision_tower": 9.112416388699124e-07, "sft_loss": 0.1046, "step": 880 }, { "distill_loss": 0.1879, "epoch": 0.54, "grad_norm": 0.8158279815713673, "learning_rate": 4.546262508253429e-06, "loss": 0.1961, "lr_vision_tower": 9.092525016506857e-07, "sft_loss": 0.1022, "step": 881 }, { "distill_loss": 0.188, "epoch": 0.54, "grad_norm": 0.8369007595838247, "learning_rate": 4.536318632024965e-06, "loss": 0.18, "lr_vision_tower": 9.072637264049929e-07, "sft_loss": 0.086, "step": 882 }, { "distill_loss": 0.1895, "epoch": 0.54, "grad_norm": 0.817285240542518, "learning_rate": 4.526376605328292e-06, "loss": 0.1803, "lr_vision_tower": 9.052753210656583e-07, "sft_loss": 0.0855, "step": 883 }, { "distill_loss": 0.1836, "epoch": 0.54, "grad_norm": 0.9471091605235042, "learning_rate": 4.516436467820157e-06, "loss": 0.1898, "lr_vision_tower": 9.032872935640314e-07, "sft_loss": 0.098, "step": 884 }, { "distill_loss": 0.1913, "epoch": 0.55, "grad_norm": 0.8038884364243858, "learning_rate": 4.506498259149774e-06, "loss": 0.1912, "lr_vision_tower": 9.012996518299546e-07, "sft_loss": 0.0955, "step": 885 }, { "distill_loss": 0.1881, "epoch": 0.55, "grad_norm": 0.7844137200345346, "learning_rate": 4.496562018958653e-06, "loss": 0.1873, "lr_vision_tower": 8.993124037917305e-07, "sft_loss": 0.0932, "step": 886 }, { "distill_loss": 0.1847, "epoch": 0.55, "grad_norm": 1.1896010098526628, "learning_rate": 4.486627786880467e-06, "loss": 0.1944, "lr_vision_tower": 8.973255573760934e-07, "sft_loss": 0.1021, "step": 887 }, { "distill_loss": 0.1747, "epoch": 0.55, "grad_norm": 0.9890637311354146, "learning_rate": 4.476695602540867e-06, "loss": 0.1744, "lr_vision_tower": 8.953391205081731e-07, "sft_loss": 0.087, "step": 888 }, { "distill_loss": 0.1865, "epoch": 0.55, "grad_norm": 0.9869073180029521, "learning_rate": 4.46676550555734e-06, "loss": 0.1877, "lr_vision_tower": 8.933531011114679e-07, "sft_loss": 0.0944, "step": 889 }, { "distill_loss": 0.1908, "epoch": 0.55, "grad_norm": 0.9334539269617598, "learning_rate": 4.456837535539049e-06, "loss": 0.1754, "lr_vision_tower": 8.913675071078096e-07, "sft_loss": 0.08, "step": 890 }, { "distill_loss": 0.195, "epoch": 0.55, "grad_norm": 0.804769795598421, "learning_rate": 4.4469117320866705e-06, "loss": 0.1953, "lr_vision_tower": 8.89382346417334e-07, "sft_loss": 0.0978, "step": 891 }, { "distill_loss": 0.1962, "epoch": 0.55, "grad_norm": 0.7414898979885782, "learning_rate": 4.43698813479224e-06, "loss": 0.1861, "lr_vision_tower": 8.873976269584478e-07, "sft_loss": 0.0881, "step": 892 }, { "distill_loss": 0.1861, "epoch": 0.55, "grad_norm": 0.9561064506631236, "learning_rate": 4.427066783238991e-06, "loss": 0.1987, "lr_vision_tower": 8.854133566477981e-07, "sft_loss": 0.1057, "step": 893 }, { "distill_loss": 0.1836, "epoch": 0.55, "grad_norm": 0.7549010137148577, "learning_rate": 4.417147717001205e-06, "loss": 0.178, "lr_vision_tower": 8.834295434002409e-07, "sft_loss": 0.0862, "step": 894 }, { "distill_loss": 0.1898, "epoch": 0.55, "grad_norm": 0.9307226083599281, "learning_rate": 4.407230975644038e-06, "loss": 0.1927, "lr_vision_tower": 8.814461951288076e-07, "sft_loss": 0.0978, "step": 895 }, { "distill_loss": 0.1959, "epoch": 0.55, "grad_norm": 0.8837259210316948, "learning_rate": 4.397316598723385e-06, "loss": 0.1847, "lr_vision_tower": 8.79463319744677e-07, "sft_loss": 0.0868, "step": 896 }, { "distill_loss": 0.1809, "epoch": 0.55, "grad_norm": 0.6900340707827906, "learning_rate": 4.387404625785699e-06, "loss": 0.1649, "lr_vision_tower": 8.774809251571397e-07, "sft_loss": 0.0745, "step": 897 }, { "distill_loss": 0.18, "epoch": 0.55, "grad_norm": 0.9217044168218715, "learning_rate": 4.377495096367849e-06, "loss": 0.1899, "lr_vision_tower": 8.754990192735696e-07, "sft_loss": 0.0999, "step": 898 }, { "distill_loss": 0.1784, "epoch": 0.55, "grad_norm": 1.0508108375140524, "learning_rate": 4.367588049996954e-06, "loss": 0.1844, "lr_vision_tower": 8.735176099993907e-07, "sft_loss": 0.0952, "step": 899 }, { "distill_loss": 0.2047, "epoch": 0.55, "grad_norm": 0.6835080106798854, "learning_rate": 4.357683526190235e-06, "loss": 0.1993, "lr_vision_tower": 8.715367052380469e-07, "sft_loss": 0.097, "step": 900 }, { "distill_loss": 0.191, "epoch": 0.56, "grad_norm": 0.5493576941164808, "learning_rate": 4.347781564454844e-06, "loss": 0.1718, "lr_vision_tower": 8.695563128909687e-07, "sft_loss": 0.0763, "step": 901 }, { "distill_loss": 0.1888, "epoch": 0.56, "grad_norm": 0.774114524939699, "learning_rate": 4.3378822042877174e-06, "loss": 0.1668, "lr_vision_tower": 8.675764408575433e-07, "sft_loss": 0.0724, "step": 902 }, { "distill_loss": 0.1828, "epoch": 0.56, "grad_norm": 0.9140897287732278, "learning_rate": 4.327985485175415e-06, "loss": 0.1843, "lr_vision_tower": 8.65597097035083e-07, "sft_loss": 0.0929, "step": 903 }, { "distill_loss": 0.189, "epoch": 0.56, "grad_norm": 0.7853785519124149, "learning_rate": 4.318091446593956e-06, "loss": 0.1844, "lr_vision_tower": 8.636182893187913e-07, "sft_loss": 0.0899, "step": 904 }, { "distill_loss": 0.1881, "epoch": 0.56, "grad_norm": 1.1505320156202754, "learning_rate": 4.30820012800868e-06, "loss": 0.1602, "lr_vision_tower": 8.616400256017358e-07, "sft_loss": 0.0662, "step": 905 }, { "distill_loss": 0.1855, "epoch": 0.56, "grad_norm": 0.81616557747956, "learning_rate": 4.29831156887406e-06, "loss": 0.1917, "lr_vision_tower": 8.596623137748118e-07, "sft_loss": 0.0989, "step": 906 }, { "distill_loss": 0.1871, "epoch": 0.56, "grad_norm": 0.781340223065129, "learning_rate": 4.2884258086335755e-06, "loss": 0.1941, "lr_vision_tower": 8.576851617267149e-07, "sft_loss": 0.1006, "step": 907 }, { "distill_loss": 0.1835, "epoch": 0.56, "grad_norm": 0.9375173222536048, "learning_rate": 4.278542886719536e-06, "loss": 0.1759, "lr_vision_tower": 8.557085773439071e-07, "sft_loss": 0.0841, "step": 908 }, { "distill_loss": 0.1949, "epoch": 0.56, "grad_norm": 0.9885661732278007, "learning_rate": 4.268662842552933e-06, "loss": 0.1852, "lr_vision_tower": 8.537325685105865e-07, "sft_loss": 0.0877, "step": 909 }, { "distill_loss": 0.1883, "epoch": 0.56, "grad_norm": 0.8811484282653954, "learning_rate": 4.258785715543276e-06, "loss": 0.1859, "lr_vision_tower": 8.51757143108655e-07, "sft_loss": 0.0918, "step": 910 }, { "distill_loss": 0.183, "epoch": 0.56, "grad_norm": 0.9517286653517437, "learning_rate": 4.248911545088438e-06, "loss": 0.1727, "lr_vision_tower": 8.497823090176874e-07, "sft_loss": 0.0812, "step": 911 }, { "distill_loss": 0.192, "epoch": 0.56, "grad_norm": 0.9138267595366532, "learning_rate": 4.2390403705745044e-06, "loss": 0.2002, "lr_vision_tower": 8.478080741149008e-07, "sft_loss": 0.1042, "step": 912 }, { "distill_loss": 0.1835, "epoch": 0.56, "grad_norm": 0.791812113156003, "learning_rate": 4.229172231375601e-06, "loss": 0.1733, "lr_vision_tower": 8.458344462751203e-07, "sft_loss": 0.0815, "step": 913 }, { "distill_loss": 0.1806, "epoch": 0.56, "grad_norm": 0.8292846402081748, "learning_rate": 4.219307166853758e-06, "loss": 0.1827, "lr_vision_tower": 8.438614333707515e-07, "sft_loss": 0.0924, "step": 914 }, { "distill_loss": 0.1873, "epoch": 0.56, "grad_norm": 0.6321127656807779, "learning_rate": 4.20944521635873e-06, "loss": 0.1763, "lr_vision_tower": 8.41889043271746e-07, "sft_loss": 0.0827, "step": 915 }, { "distill_loss": 0.1784, "epoch": 0.56, "grad_norm": 0.7545279400257245, "learning_rate": 4.19958641922786e-06, "loss": 0.169, "lr_vision_tower": 8.399172838455718e-07, "sft_loss": 0.0798, "step": 916 }, { "distill_loss": 0.1829, "epoch": 0.57, "grad_norm": 0.7222190447571755, "learning_rate": 4.189730814785904e-06, "loss": 0.1697, "lr_vision_tower": 8.379461629571807e-07, "sft_loss": 0.0783, "step": 917 }, { "distill_loss": 0.1888, "epoch": 0.57, "grad_norm": 1.116713154842568, "learning_rate": 4.179878442344892e-06, "loss": 0.195, "lr_vision_tower": 8.359756884689783e-07, "sft_loss": 0.1006, "step": 918 }, { "distill_loss": 0.1823, "epoch": 0.57, "grad_norm": 0.7594115833527532, "learning_rate": 4.170029341203953e-06, "loss": 0.1866, "lr_vision_tower": 8.340058682407906e-07, "sft_loss": 0.0955, "step": 919 }, { "distill_loss": 0.1991, "epoch": 0.57, "grad_norm": 0.6262308145946159, "learning_rate": 4.160183550649176e-06, "loss": 0.183, "lr_vision_tower": 8.32036710129835e-07, "sft_loss": 0.0835, "step": 920 }, { "distill_loss": 0.1922, "epoch": 0.57, "grad_norm": 0.7446189576530715, "learning_rate": 4.1503411099534404e-06, "loss": 0.1738, "lr_vision_tower": 8.30068221990688e-07, "sft_loss": 0.0777, "step": 921 }, { "distill_loss": 0.182, "epoch": 0.57, "grad_norm": 0.8555641339951886, "learning_rate": 4.140502058376261e-06, "loss": 0.1923, "lr_vision_tower": 8.281004116752521e-07, "sft_loss": 0.1012, "step": 922 }, { "distill_loss": 0.1856, "epoch": 0.57, "grad_norm": 1.5866089648518757, "learning_rate": 4.130666435163641e-06, "loss": 0.2106, "lr_vision_tower": 8.261332870327281e-07, "sft_loss": 0.1177, "step": 923 }, { "distill_loss": 0.1857, "epoch": 0.57, "grad_norm": 0.6626609510360041, "learning_rate": 4.1208342795479015e-06, "loss": 0.1756, "lr_vision_tower": 8.241668559095802e-07, "sft_loss": 0.0828, "step": 924 }, { "distill_loss": 0.1889, "epoch": 0.57, "grad_norm": 0.7866663354352659, "learning_rate": 4.111005630747538e-06, "loss": 0.189, "lr_vision_tower": 8.222011261495076e-07, "sft_loss": 0.0946, "step": 925 }, { "distill_loss": 0.1812, "epoch": 0.57, "grad_norm": 0.6427478363608333, "learning_rate": 4.101180527967053e-06, "loss": 0.1739, "lr_vision_tower": 8.202361055934106e-07, "sft_loss": 0.0833, "step": 926 }, { "distill_loss": 0.1905, "epoch": 0.57, "grad_norm": 0.8495189247154903, "learning_rate": 4.091359010396808e-06, "loss": 0.2059, "lr_vision_tower": 8.182718020793616e-07, "sft_loss": 0.1106, "step": 927 }, { "distill_loss": 0.1776, "epoch": 0.57, "grad_norm": 0.8040417752395391, "learning_rate": 4.081541117212865e-06, "loss": 0.1852, "lr_vision_tower": 8.163082234425729e-07, "sft_loss": 0.0964, "step": 928 }, { "distill_loss": 0.1859, "epoch": 0.57, "grad_norm": 1.2356683414453487, "learning_rate": 4.071726887576823e-06, "loss": 0.1947, "lr_vision_tower": 8.143453775153644e-07, "sft_loss": 0.1017, "step": 929 }, { "distill_loss": 0.1843, "epoch": 0.57, "grad_norm": 0.9081211298269724, "learning_rate": 4.061916360635675e-06, "loss": 0.1723, "lr_vision_tower": 8.123832721271349e-07, "sft_loss": 0.0801, "step": 930 }, { "distill_loss": 0.1912, "epoch": 0.57, "grad_norm": 0.8758438121180456, "learning_rate": 4.052109575521639e-06, "loss": 0.1887, "lr_vision_tower": 8.104219151043276e-07, "sft_loss": 0.0931, "step": 931 }, { "distill_loss": 0.1907, "epoch": 0.57, "grad_norm": 0.8624695423950726, "learning_rate": 4.04230657135201e-06, "loss": 0.1892, "lr_vision_tower": 8.08461314270402e-07, "sft_loss": 0.0939, "step": 932 }, { "distill_loss": 0.1875, "epoch": 0.58, "grad_norm": 0.8270313757506543, "learning_rate": 4.032507387229002e-06, "loss": 0.1887, "lr_vision_tower": 8.065014774458003e-07, "sft_loss": 0.095, "step": 933 }, { "distill_loss": 0.192, "epoch": 0.58, "grad_norm": 0.888346877203517, "learning_rate": 4.022712062239593e-06, "loss": 0.1909, "lr_vision_tower": 8.045424124479185e-07, "sft_loss": 0.0949, "step": 934 }, { "distill_loss": 0.1844, "epoch": 0.58, "grad_norm": 0.8988252479148542, "learning_rate": 4.012920635455363e-06, "loss": 0.1777, "lr_vision_tower": 8.025841270910725e-07, "sft_loss": 0.0855, "step": 935 }, { "distill_loss": 0.1848, "epoch": 0.58, "grad_norm": 0.8193221560083266, "learning_rate": 4.003133145932346e-06, "loss": 0.1654, "lr_vision_tower": 8.006266291864692e-07, "sft_loss": 0.073, "step": 936 }, { "distill_loss": 0.1902, "epoch": 0.58, "grad_norm": 0.8648367088488179, "learning_rate": 3.993349632710874e-06, "loss": 0.1948, "lr_vision_tower": 7.986699265421748e-07, "sft_loss": 0.0997, "step": 937 }, { "distill_loss": 0.1874, "epoch": 0.58, "grad_norm": 0.7298175455099979, "learning_rate": 3.983570134815411e-06, "loss": 0.1776, "lr_vision_tower": 7.967140269630822e-07, "sft_loss": 0.0839, "step": 938 }, { "distill_loss": 0.1815, "epoch": 0.58, "grad_norm": 0.8823767727049635, "learning_rate": 3.973794691254413e-06, "loss": 0.1876, "lr_vision_tower": 7.947589382508826e-07, "sft_loss": 0.0968, "step": 939 }, { "distill_loss": 0.1945, "epoch": 0.58, "grad_norm": 0.9307932584066909, "learning_rate": 3.9640233410201555e-06, "loss": 0.181, "lr_vision_tower": 7.92804668204031e-07, "sft_loss": 0.0837, "step": 940 }, { "distill_loss": 0.1766, "epoch": 0.58, "grad_norm": 0.8039790339992435, "learning_rate": 3.9542561230885935e-06, "loss": 0.1729, "lr_vision_tower": 7.908512246177186e-07, "sft_loss": 0.0846, "step": 941 }, { "distill_loss": 0.1819, "epoch": 0.58, "grad_norm": 0.7808276423274312, "learning_rate": 3.944493076419194e-06, "loss": 0.1821, "lr_vision_tower": 7.888986152838388e-07, "sft_loss": 0.0912, "step": 942 }, { "distill_loss": 0.1903, "epoch": 0.58, "grad_norm": 0.8741094293401906, "learning_rate": 3.934734239954792e-06, "loss": 0.184, "lr_vision_tower": 7.869468479909582e-07, "sft_loss": 0.0888, "step": 943 }, { "distill_loss": 0.186, "epoch": 0.58, "grad_norm": 0.763380500417675, "learning_rate": 3.92497965262142e-06, "loss": 0.182, "lr_vision_tower": 7.849959305242839e-07, "sft_loss": 0.089, "step": 944 }, { "distill_loss": 0.183, "epoch": 0.58, "grad_norm": 0.7070801510019693, "learning_rate": 3.91522935332817e-06, "loss": 0.1843, "lr_vision_tower": 7.830458706656339e-07, "sft_loss": 0.0928, "step": 945 }, { "distill_loss": 0.1988, "epoch": 0.58, "grad_norm": 1.0069972643721596, "learning_rate": 3.905483380967027e-06, "loss": 0.195, "lr_vision_tower": 7.810966761934053e-07, "sft_loss": 0.0956, "step": 946 }, { "distill_loss": 0.1804, "epoch": 0.58, "grad_norm": 0.9273763568676154, "learning_rate": 3.8957417744127125e-06, "loss": 0.1819, "lr_vision_tower": 7.791483548825424e-07, "sft_loss": 0.0917, "step": 947 }, { "distill_loss": 0.1908, "epoch": 0.58, "grad_norm": 0.7696617899197015, "learning_rate": 3.886004572522543e-06, "loss": 0.1906, "lr_vision_tower": 7.772009145045085e-07, "sft_loss": 0.0952, "step": 948 }, { "distill_loss": 0.1837, "epoch": 0.58, "grad_norm": 0.8416414075429117, "learning_rate": 3.876271814136253e-06, "loss": 0.2032, "lr_vision_tower": 7.752543628272506e-07, "sft_loss": 0.1114, "step": 949 }, { "distill_loss": 0.1787, "epoch": 0.59, "grad_norm": 0.698458544513217, "learning_rate": 3.866543538075865e-06, "loss": 0.1788, "lr_vision_tower": 7.733087076151728e-07, "sft_loss": 0.0895, "step": 950 }, { "distill_loss": 0.1776, "epoch": 0.59, "grad_norm": 0.7329922466972322, "learning_rate": 3.856819783145514e-06, "loss": 0.1875, "lr_vision_tower": 7.713639566291027e-07, "sft_loss": 0.0987, "step": 951 }, { "distill_loss": 0.1846, "epoch": 0.59, "grad_norm": 0.6811933370617838, "learning_rate": 3.847100588131306e-06, "loss": 0.1837, "lr_vision_tower": 7.694201176262611e-07, "sft_loss": 0.0914, "step": 952 }, { "distill_loss": 0.1864, "epoch": 0.59, "grad_norm": 0.863898273838424, "learning_rate": 3.837385991801156e-06, "loss": 0.1901, "lr_vision_tower": 7.674771983602312e-07, "sft_loss": 0.0968, "step": 953 }, { "distill_loss": 0.1731, "epoch": 0.59, "grad_norm": 0.8787361942330266, "learning_rate": 3.827676032904637e-06, "loss": 0.1722, "lr_vision_tower": 7.655352065809273e-07, "sft_loss": 0.0856, "step": 954 }, { "distill_loss": 0.1798, "epoch": 0.59, "grad_norm": 0.7291797325246545, "learning_rate": 3.817970750172823e-06, "loss": 0.174, "lr_vision_tower": 7.635941500345646e-07, "sft_loss": 0.0841, "step": 955 }, { "distill_loss": 0.1889, "epoch": 0.59, "grad_norm": 0.9539204381105334, "learning_rate": 3.808270182318137e-06, "loss": 0.1911, "lr_vision_tower": 7.616540364636274e-07, "sft_loss": 0.0966, "step": 956 }, { "distill_loss": 0.1849, "epoch": 0.59, "grad_norm": 0.9226468566169468, "learning_rate": 3.798574368034196e-06, "loss": 0.1774, "lr_vision_tower": 7.597148736068391e-07, "sft_loss": 0.0849, "step": 957 }, { "distill_loss": 0.1834, "epoch": 0.59, "grad_norm": 0.9157685616225842, "learning_rate": 3.7888833459956507e-06, "loss": 0.1782, "lr_vision_tower": 7.577766691991301e-07, "sft_loss": 0.0865, "step": 958 }, { "distill_loss": 0.1853, "epoch": 0.59, "grad_norm": 0.9035192225537579, "learning_rate": 3.779197154858044e-06, "loss": 0.1832, "lr_vision_tower": 7.558394309716088e-07, "sft_loss": 0.0905, "step": 959 }, { "distill_loss": 0.1893, "epoch": 0.59, "grad_norm": 0.8034435401672854, "learning_rate": 3.7695158332576435e-06, "loss": 0.1766, "lr_vision_tower": 7.539031666515287e-07, "sft_loss": 0.0819, "step": 960 }, { "distill_loss": 0.1799, "epoch": 0.59, "grad_norm": 0.7706171273477707, "learning_rate": 3.7598394198112974e-06, "loss": 0.1891, "lr_vision_tower": 7.519678839622594e-07, "sft_loss": 0.0991, "step": 961 }, { "distill_loss": 0.1847, "epoch": 0.59, "grad_norm": 0.8659495166263788, "learning_rate": 3.750167953116272e-06, "loss": 0.1639, "lr_vision_tower": 7.500335906232544e-07, "sft_loss": 0.0716, "step": 962 }, { "distill_loss": 0.1669, "epoch": 0.59, "grad_norm": 0.8300970521023617, "learning_rate": 3.740501471750106e-06, "loss": 0.1728, "lr_vision_tower": 7.481002943500211e-07, "sft_loss": 0.0893, "step": 963 }, { "distill_loss": 0.1978, "epoch": 0.59, "grad_norm": 0.9407633535866251, "learning_rate": 3.7308400142704517e-06, "loss": 0.2053, "lr_vision_tower": 7.461680028540902e-07, "sft_loss": 0.1064, "step": 964 }, { "distill_loss": 0.1863, "epoch": 0.59, "grad_norm": 0.8716337511273249, "learning_rate": 3.721183619214918e-06, "loss": 0.1885, "lr_vision_tower": 7.442367238429835e-07, "sft_loss": 0.0953, "step": 965 }, { "distill_loss": 0.1856, "epoch": 0.6, "grad_norm": 0.8213117900684709, "learning_rate": 3.7115323251009293e-06, "loss": 0.18, "lr_vision_tower": 7.423064650201858e-07, "sft_loss": 0.0872, "step": 966 }, { "distill_loss": 0.1807, "epoch": 0.6, "grad_norm": 0.9665362553331115, "learning_rate": 3.701886170425553e-06, "loss": 0.1982, "lr_vision_tower": 7.403772340851104e-07, "sft_loss": 0.1078, "step": 967 }, { "distill_loss": 0.1745, "epoch": 0.6, "grad_norm": 0.7697593746039678, "learning_rate": 3.692245193665364e-06, "loss": 0.1706, "lr_vision_tower": 7.384490387330727e-07, "sft_loss": 0.0834, "step": 968 }, { "distill_loss": 0.1861, "epoch": 0.6, "grad_norm": 0.8872647059570365, "learning_rate": 3.682609433276281e-06, "loss": 0.1803, "lr_vision_tower": 7.36521886655256e-07, "sft_loss": 0.0872, "step": 969 }, { "distill_loss": 0.1833, "epoch": 0.6, "grad_norm": 0.8181424087720612, "learning_rate": 3.6729789276934156e-06, "loss": 0.1823, "lr_vision_tower": 7.345957855386831e-07, "sft_loss": 0.0906, "step": 970 }, { "distill_loss": 0.192, "epoch": 0.6, "grad_norm": 0.7234395544093316, "learning_rate": 3.6633537153309183e-06, "loss": 0.1848, "lr_vision_tower": 7.326707430661836e-07, "sft_loss": 0.0888, "step": 971 }, { "distill_loss": 0.1916, "epoch": 0.6, "grad_norm": 0.6638316624253404, "learning_rate": 3.6537338345818273e-06, "loss": 0.1847, "lr_vision_tower": 7.307467669163654e-07, "sft_loss": 0.0889, "step": 972 }, { "distill_loss": 0.1829, "epoch": 0.6, "grad_norm": 0.7831006818378543, "learning_rate": 3.6441193238179152e-06, "loss": 0.1695, "lr_vision_tower": 7.288238647635829e-07, "sft_loss": 0.078, "step": 973 }, { "distill_loss": 0.1914, "epoch": 0.6, "grad_norm": 0.900415038890171, "learning_rate": 3.634510221389531e-06, "loss": 0.1753, "lr_vision_tower": 7.269020442779061e-07, "sft_loss": 0.0796, "step": 974 }, { "distill_loss": 0.1936, "epoch": 0.6, "grad_norm": 0.7837434684367927, "learning_rate": 3.6249065656254565e-06, "loss": 0.1989, "lr_vision_tower": 7.249813131250912e-07, "sft_loss": 0.1021, "step": 975 }, { "distill_loss": 0.1882, "epoch": 0.6, "grad_norm": 0.7963952769769688, "learning_rate": 3.6153083948327398e-06, "loss": 0.1769, "lr_vision_tower": 7.230616789665479e-07, "sft_loss": 0.0828, "step": 976 }, { "distill_loss": 0.1937, "epoch": 0.6, "grad_norm": 0.7893978819052011, "learning_rate": 3.6057157472965586e-06, "loss": 0.1861, "lr_vision_tower": 7.211431494593117e-07, "sft_loss": 0.0893, "step": 977 }, { "distill_loss": 0.187, "epoch": 0.6, "grad_norm": 0.7849203761023695, "learning_rate": 3.5961286612800544e-06, "loss": 0.1756, "lr_vision_tower": 7.192257322560108e-07, "sft_loss": 0.0821, "step": 978 }, { "distill_loss": 0.1859, "epoch": 0.6, "grad_norm": 0.8241523601533186, "learning_rate": 3.586547175024187e-06, "loss": 0.1705, "lr_vision_tower": 7.173094350048373e-07, "sft_loss": 0.0776, "step": 979 }, { "distill_loss": 0.1832, "epoch": 0.6, "grad_norm": 0.8773989265550255, "learning_rate": 3.5769713267475813e-06, "loss": 0.1636, "lr_vision_tower": 7.153942653495162e-07, "sft_loss": 0.072, "step": 980 }, { "distill_loss": 0.1799, "epoch": 0.6, "grad_norm": 0.7467358994395851, "learning_rate": 3.567401154646369e-06, "loss": 0.181, "lr_vision_tower": 7.134802309292737e-07, "sft_loss": 0.091, "step": 981 }, { "distill_loss": 0.181, "epoch": 0.61, "grad_norm": 1.1475489278849351, "learning_rate": 3.557836696894045e-06, "loss": 0.1879, "lr_vision_tower": 7.11567339378809e-07, "sft_loss": 0.0974, "step": 982 }, { "distill_loss": 0.1741, "epoch": 0.61, "grad_norm": 1.0543711705770908, "learning_rate": 3.548277991641308e-06, "loss": 0.1904, "lr_vision_tower": 7.096555983282615e-07, "sft_loss": 0.1034, "step": 983 }, { "distill_loss": 0.1936, "epoch": 0.61, "grad_norm": 0.8710601699768883, "learning_rate": 3.5387250770159152e-06, "loss": 0.1878, "lr_vision_tower": 7.07745015403183e-07, "sft_loss": 0.091, "step": 984 }, { "distill_loss": 0.1785, "epoch": 0.61, "grad_norm": 0.9904513150901616, "learning_rate": 3.529177991122519e-06, "loss": 0.1696, "lr_vision_tower": 7.058355982245037e-07, "sft_loss": 0.0803, "step": 985 }, { "distill_loss": 0.1831, "epoch": 0.61, "grad_norm": 0.7467779566156182, "learning_rate": 3.519636772042529e-06, "loss": 0.1952, "lr_vision_tower": 7.039273544085057e-07, "sft_loss": 0.1036, "step": 986 }, { "distill_loss": 0.1803, "epoch": 0.61, "grad_norm": 0.8008059529837088, "learning_rate": 3.5101014578339486e-06, "loss": 0.1753, "lr_vision_tower": 7.020202915667897e-07, "sft_loss": 0.0851, "step": 987 }, { "distill_loss": 0.1797, "epoch": 0.61, "grad_norm": 0.7583641215880793, "learning_rate": 3.5005720865312313e-06, "loss": 0.1749, "lr_vision_tower": 7.001144173062462e-07, "sft_loss": 0.0851, "step": 988 }, { "distill_loss": 0.174, "epoch": 0.61, "grad_norm": 1.0837730433295152, "learning_rate": 3.4910486961451257e-06, "loss": 0.19, "lr_vision_tower": 6.98209739229025e-07, "sft_loss": 0.103, "step": 989 }, { "distill_loss": 0.1833, "epoch": 0.61, "grad_norm": 1.4834032160727046, "learning_rate": 3.481531324662519e-06, "loss": 0.1701, "lr_vision_tower": 6.963062649325036e-07, "sft_loss": 0.0784, "step": 990 }, { "distill_loss": 0.185, "epoch": 0.61, "grad_norm": 1.1631585335404642, "learning_rate": 3.4720200100462966e-06, "loss": 0.1805, "lr_vision_tower": 6.944040020092592e-07, "sft_loss": 0.088, "step": 991 }, { "distill_loss": 0.1811, "epoch": 0.61, "grad_norm": 0.9882934995416558, "learning_rate": 3.4625147902351765e-06, "loss": 0.1864, "lr_vision_tower": 6.925029580470352e-07, "sft_loss": 0.0959, "step": 992 }, { "distill_loss": 0.1748, "epoch": 0.61, "grad_norm": 0.8697792979046168, "learning_rate": 3.4530157031435764e-06, "loss": 0.1573, "lr_vision_tower": 6.906031406287151e-07, "sft_loss": 0.0699, "step": 993 }, { "distill_loss": 0.1773, "epoch": 0.61, "grad_norm": 0.7949328755420054, "learning_rate": 3.4435227866614396e-06, "loss": 0.1684, "lr_vision_tower": 6.887045573322878e-07, "sft_loss": 0.0797, "step": 994 }, { "distill_loss": 0.1825, "epoch": 0.61, "grad_norm": 1.024978270213627, "learning_rate": 3.4340360786541067e-06, "loss": 0.1806, "lr_vision_tower": 6.868072157308212e-07, "sft_loss": 0.0893, "step": 995 }, { "distill_loss": 0.1783, "epoch": 0.61, "grad_norm": 1.222517005169866, "learning_rate": 3.424555616962146e-06, "loss": 0.196, "lr_vision_tower": 6.849111233924292e-07, "sft_loss": 0.1069, "step": 996 }, { "distill_loss": 0.1714, "epoch": 0.61, "grad_norm": 1.1265400353759347, "learning_rate": 3.415081439401216e-06, "loss": 0.1777, "lr_vision_tower": 6.830162878802432e-07, "sft_loss": 0.092, "step": 997 }, { "distill_loss": 0.1744, "epoch": 0.62, "grad_norm": 0.8904211841854566, "learning_rate": 3.4056135837619077e-06, "loss": 0.1921, "lr_vision_tower": 6.811227167523814e-07, "sft_loss": 0.1049, "step": 998 }, { "distill_loss": 0.1822, "epoch": 0.62, "grad_norm": 0.7934525615710807, "learning_rate": 3.396152087809592e-06, "loss": 0.1839, "lr_vision_tower": 6.792304175619184e-07, "sft_loss": 0.0929, "step": 999 }, { "distill_loss": 0.1945, "epoch": 0.62, "grad_norm": 0.674906657916509, "learning_rate": 3.3866969892842783e-06, "loss": 0.1855, "lr_vision_tower": 6.773393978568556e-07, "sft_loss": 0.0883, "step": 1000 }, { "distill_loss": 0.1843, "epoch": 0.62, "grad_norm": 0.8699027359952425, "learning_rate": 3.3772483259004497e-06, "loss": 0.1784, "lr_vision_tower": 6.754496651800899e-07, "sft_loss": 0.0863, "step": 1001 }, { "distill_loss": 0.1835, "epoch": 0.62, "grad_norm": 0.7433968872771834, "learning_rate": 3.3678061353469304e-06, "loss": 0.1719, "lr_vision_tower": 6.73561227069386e-07, "sft_loss": 0.0801, "step": 1002 }, { "distill_loss": 0.1832, "epoch": 0.62, "grad_norm": 0.7536579679928198, "learning_rate": 3.3583704552867177e-06, "loss": 0.1688, "lr_vision_tower": 6.716740910573434e-07, "sft_loss": 0.0772, "step": 1003 }, { "distill_loss": 0.1785, "epoch": 0.62, "grad_norm": 0.7711127528411398, "learning_rate": 3.3489413233568447e-06, "loss": 0.1676, "lr_vision_tower": 6.697882646713688e-07, "sft_loss": 0.0784, "step": 1004 }, { "distill_loss": 0.1778, "epoch": 0.62, "grad_norm": 0.7069608200017388, "learning_rate": 3.3395187771682214e-06, "loss": 0.1798, "lr_vision_tower": 6.679037554336442e-07, "sft_loss": 0.0909, "step": 1005 }, { "distill_loss": 0.1721, "epoch": 0.62, "grad_norm": 0.6935975976707213, "learning_rate": 3.3301028543054935e-06, "loss": 0.1759, "lr_vision_tower": 6.660205708610986e-07, "sft_loss": 0.0898, "step": 1006 }, { "distill_loss": 0.1971, "epoch": 0.62, "grad_norm": 0.8277155506207606, "learning_rate": 3.3206935923268847e-06, "loss": 0.1985, "lr_vision_tower": 6.641387184653768e-07, "sft_loss": 0.0999, "step": 1007 }, { "distill_loss": 0.179, "epoch": 0.62, "grad_norm": 0.8028351439685514, "learning_rate": 3.3112910287640487e-06, "loss": 0.1769, "lr_vision_tower": 6.622582057528096e-07, "sft_loss": 0.0874, "step": 1008 }, { "distill_loss": 0.184, "epoch": 0.62, "grad_norm": 0.8275828936629439, "learning_rate": 3.301895201121925e-06, "loss": 0.1808, "lr_vision_tower": 6.603790402243848e-07, "sft_loss": 0.0888, "step": 1009 }, { "distill_loss": 0.1809, "epoch": 0.62, "grad_norm": 0.6447583799538498, "learning_rate": 3.292506146878576e-06, "loss": 0.1805, "lr_vision_tower": 6.585012293757151e-07, "sft_loss": 0.09, "step": 1010 }, { "distill_loss": 0.1752, "epoch": 0.62, "grad_norm": 0.7185295409816241, "learning_rate": 3.2831239034850593e-06, "loss": 0.1916, "lr_vision_tower": 6.566247806970118e-07, "sft_loss": 0.104, "step": 1011 }, { "distill_loss": 0.1839, "epoch": 0.62, "grad_norm": 1.134673299211337, "learning_rate": 3.273748508365252e-06, "loss": 0.1839, "lr_vision_tower": 6.547497016730503e-07, "sft_loss": 0.092, "step": 1012 }, { "distill_loss": 0.1938, "epoch": 0.62, "grad_norm": 1.0511319909934813, "learning_rate": 3.2643799989157253e-06, "loss": 0.1811, "lr_vision_tower": 6.52875999783145e-07, "sft_loss": 0.0842, "step": 1013 }, { "distill_loss": 0.1786, "epoch": 0.63, "grad_norm": 0.7683228170208712, "learning_rate": 3.2550184125055783e-06, "loss": 0.1619, "lr_vision_tower": 6.510036825011156e-07, "sft_loss": 0.0726, "step": 1014 }, { "distill_loss": 0.1742, "epoch": 0.63, "grad_norm": 0.7641139221586036, "learning_rate": 3.245663786476298e-06, "loss": 0.183, "lr_vision_tower": 6.491327572952595e-07, "sft_loss": 0.0959, "step": 1015 }, { "distill_loss": 0.1803, "epoch": 0.63, "grad_norm": 1.078042172934238, "learning_rate": 3.2363161581416097e-06, "loss": 0.1787, "lr_vision_tower": 6.472632316283219e-07, "sft_loss": 0.0886, "step": 1016 }, { "distill_loss": 0.1899, "epoch": 0.63, "grad_norm": 0.7343699922243181, "learning_rate": 3.226975564787322e-06, "loss": 0.1842, "lr_vision_tower": 6.453951129574643e-07, "sft_loss": 0.0892, "step": 1017 }, { "distill_loss": 0.178, "epoch": 0.63, "grad_norm": 0.8657784073584099, "learning_rate": 3.2176420436711877e-06, "loss": 0.1715, "lr_vision_tower": 6.435284087342374e-07, "sft_loss": 0.0825, "step": 1018 }, { "distill_loss": 0.1743, "epoch": 0.63, "grad_norm": 1.2974295037112547, "learning_rate": 3.2083156320227405e-06, "loss": 0.1859, "lr_vision_tower": 6.41663126404548e-07, "sft_loss": 0.0988, "step": 1019 }, { "distill_loss": 0.1863, "epoch": 0.63, "grad_norm": 0.9290788108881725, "learning_rate": 3.19899636704317e-06, "loss": 0.1798, "lr_vision_tower": 6.397992734086339e-07, "sft_loss": 0.0866, "step": 1020 }, { "distill_loss": 0.1842, "epoch": 0.63, "grad_norm": 0.8592477657641168, "learning_rate": 3.189684285905145e-06, "loss": 0.1709, "lr_vision_tower": 6.37936857181029e-07, "sft_loss": 0.0788, "step": 1021 }, { "distill_loss": 0.1682, "epoch": 0.63, "grad_norm": 0.9348035209029464, "learning_rate": 3.18037942575269e-06, "loss": 0.1885, "lr_vision_tower": 6.360758851505379e-07, "sft_loss": 0.1045, "step": 1022 }, { "distill_loss": 0.1758, "epoch": 0.63, "grad_norm": 0.5966034677164136, "learning_rate": 3.1710818237010187e-06, "loss": 0.1619, "lr_vision_tower": 6.342163647402036e-07, "sft_loss": 0.074, "step": 1023 }, { "distill_loss": 0.1823, "epoch": 0.63, "grad_norm": 0.8313698489206719, "learning_rate": 3.1617915168363994e-06, "loss": 0.1811, "lr_vision_tower": 6.323583033672798e-07, "sft_loss": 0.09, "step": 1024 }, { "distill_loss": 0.1854, "epoch": 0.63, "grad_norm": 0.8491280611392946, "learning_rate": 3.1525085422159995e-06, "loss": 0.194, "lr_vision_tower": 6.305017084431998e-07, "sft_loss": 0.1013, "step": 1025 }, { "distill_loss": 0.1933, "epoch": 0.63, "grad_norm": 1.0101538418127873, "learning_rate": 3.1432329368677373e-06, "loss": 0.1963, "lr_vision_tower": 6.286465873735474e-07, "sft_loss": 0.0997, "step": 1026 }, { "distill_loss": 0.1816, "epoch": 0.63, "grad_norm": 0.7659411071587504, "learning_rate": 3.133964737790142e-06, "loss": 0.1738, "lr_vision_tower": 6.267929475580283e-07, "sft_loss": 0.083, "step": 1027 }, { "distill_loss": 0.1957, "epoch": 0.63, "grad_norm": 0.8490678485315487, "learning_rate": 3.1247039819521907e-06, "loss": 0.1966, "lr_vision_tower": 6.24940796390438e-07, "sft_loss": 0.0987, "step": 1028 }, { "distill_loss": 0.1844, "epoch": 0.63, "grad_norm": 0.7758377581166327, "learning_rate": 3.115450706293185e-06, "loss": 0.1786, "lr_vision_tower": 6.230901412586369e-07, "sft_loss": 0.0864, "step": 1029 }, { "distill_loss": 0.1778, "epoch": 0.63, "grad_norm": 0.9866537742930455, "learning_rate": 3.1062049477225753e-06, "loss": 0.1683, "lr_vision_tower": 6.212409895445149e-07, "sft_loss": 0.0794, "step": 1030 }, { "distill_loss": 0.1805, "epoch": 0.64, "grad_norm": 0.8256754413134579, "learning_rate": 3.0969667431198386e-06, "loss": 0.1782, "lr_vision_tower": 6.193933486239676e-07, "sft_loss": 0.088, "step": 1031 }, { "distill_loss": 0.183, "epoch": 0.64, "grad_norm": 0.786110630612656, "learning_rate": 3.0877361293343143e-06, "loss": 0.182, "lr_vision_tower": 6.175472258668628e-07, "sft_loss": 0.0905, "step": 1032 }, { "distill_loss": 0.1831, "epoch": 0.64, "grad_norm": 0.6873048653889967, "learning_rate": 3.0785131431850663e-06, "loss": 0.1786, "lr_vision_tower": 6.157026286370131e-07, "sft_loss": 0.087, "step": 1033 }, { "distill_loss": 0.1745, "epoch": 0.64, "grad_norm": 0.7856573776313425, "learning_rate": 3.069297821460734e-06, "loss": 0.1766, "lr_vision_tower": 6.138595642921467e-07, "sft_loss": 0.0894, "step": 1034 }, { "distill_loss": 0.1755, "epoch": 0.64, "grad_norm": 0.7695516203435674, "learning_rate": 3.0600902009193824e-06, "loss": 0.1654, "lr_vision_tower": 6.120180401838764e-07, "sft_loss": 0.0776, "step": 1035 }, { "distill_loss": 0.1682, "epoch": 0.64, "grad_norm": 0.8372830975442493, "learning_rate": 3.050890318288361e-06, "loss": 0.1675, "lr_vision_tower": 6.101780636576721e-07, "sft_loss": 0.0835, "step": 1036 }, { "distill_loss": 0.1854, "epoch": 0.64, "grad_norm": 0.8305982227051228, "learning_rate": 3.041698210264149e-06, "loss": 0.1854, "lr_vision_tower": 6.083396420528297e-07, "sft_loss": 0.0927, "step": 1037 }, { "distill_loss": 0.1827, "epoch": 0.64, "grad_norm": 0.9543779068754304, "learning_rate": 3.0325139135122234e-06, "loss": 0.184, "lr_vision_tower": 6.065027827024446e-07, "sft_loss": 0.0926, "step": 1038 }, { "distill_loss": 0.1731, "epoch": 0.64, "grad_norm": 0.8862357106125822, "learning_rate": 3.0233374646668935e-06, "loss": 0.1749, "lr_vision_tower": 6.046674929333786e-07, "sft_loss": 0.0884, "step": 1039 }, { "distill_loss": 0.1791, "epoch": 0.64, "grad_norm": 0.7797017640736842, "learning_rate": 3.0141689003311714e-06, "loss": 0.1752, "lr_vision_tower": 6.028337800662342e-07, "sft_loss": 0.0857, "step": 1040 }, { "distill_loss": 0.1801, "epoch": 0.64, "grad_norm": 0.6977806814450191, "learning_rate": 3.0050082570766186e-06, "loss": 0.1703, "lr_vision_tower": 6.010016514153237e-07, "sft_loss": 0.0803, "step": 1041 }, { "distill_loss": 0.1771, "epoch": 0.64, "grad_norm": 0.8526672827172922, "learning_rate": 2.995855571443198e-06, "loss": 0.1753, "lr_vision_tower": 5.991711142886396e-07, "sft_loss": 0.0867, "step": 1042 }, { "distill_loss": 0.184, "epoch": 0.64, "grad_norm": 0.8553596911119389, "learning_rate": 2.986710879939136e-06, "loss": 0.1897, "lr_vision_tower": 5.973421759878271e-07, "sft_loss": 0.0977, "step": 1043 }, { "distill_loss": 0.1745, "epoch": 0.64, "grad_norm": 0.6711599273498844, "learning_rate": 2.977574219040766e-06, "loss": 0.1632, "lr_vision_tower": 5.955148438081531e-07, "sft_loss": 0.076, "step": 1044 }, { "distill_loss": 0.1729, "epoch": 0.64, "grad_norm": 0.7298703969281429, "learning_rate": 2.9684456251923955e-06, "loss": 0.1563, "lr_vision_tower": 5.93689125038479e-07, "sft_loss": 0.0699, "step": 1045 }, { "distill_loss": 0.1734, "epoch": 0.64, "grad_norm": 1.0058631200741557, "learning_rate": 2.9593251348061468e-06, "loss": 0.1751, "lr_vision_tower": 5.918650269612293e-07, "sft_loss": 0.0884, "step": 1046 }, { "distill_loss": 0.181, "epoch": 0.65, "grad_norm": 0.9907502221903642, "learning_rate": 2.950212784261828e-06, "loss": 0.1723, "lr_vision_tower": 5.900425568523655e-07, "sft_loss": 0.0818, "step": 1047 }, { "distill_loss": 0.1737, "epoch": 0.65, "grad_norm": 0.8483371573846437, "learning_rate": 2.9411086099067686e-06, "loss": 0.1695, "lr_vision_tower": 5.882217219813537e-07, "sft_loss": 0.0826, "step": 1048 }, { "distill_loss": 0.1753, "epoch": 0.65, "grad_norm": 1.0009696196081324, "learning_rate": 2.9320126480556943e-06, "loss": 0.1698, "lr_vision_tower": 5.864025296111388e-07, "sft_loss": 0.0822, "step": 1049 }, { "distill_loss": 0.1809, "epoch": 0.65, "grad_norm": 0.812669017910232, "learning_rate": 2.9229249349905686e-06, "loss": 0.1889, "lr_vision_tower": 5.845849869981136e-07, "sft_loss": 0.0984, "step": 1050 }, { "distill_loss": 0.1878, "epoch": 0.65, "grad_norm": 0.8450128301424737, "learning_rate": 2.91384550696045e-06, "loss": 0.1853, "lr_vision_tower": 5.827691013920898e-07, "sft_loss": 0.0915, "step": 1051 }, { "distill_loss": 0.1789, "epoch": 0.65, "grad_norm": 0.6694944802406769, "learning_rate": 2.904774400181357e-06, "loss": 0.1755, "lr_vision_tower": 5.809548800362714e-07, "sft_loss": 0.0861, "step": 1052 }, { "distill_loss": 0.1844, "epoch": 0.65, "grad_norm": 1.1878029839691224, "learning_rate": 2.8957116508361073e-06, "loss": 0.178, "lr_vision_tower": 5.791423301672214e-07, "sft_loss": 0.0858, "step": 1053 }, { "distill_loss": 0.1888, "epoch": 0.65, "grad_norm": 0.9082721525131889, "learning_rate": 2.886657295074189e-06, "loss": 0.182, "lr_vision_tower": 5.773314590148377e-07, "sft_loss": 0.0876, "step": 1054 }, { "distill_loss": 0.1821, "epoch": 0.65, "grad_norm": 0.9011706682029721, "learning_rate": 2.877611369011606e-06, "loss": 0.1747, "lr_vision_tower": 5.755222738023211e-07, "sft_loss": 0.0836, "step": 1055 }, { "distill_loss": 0.1761, "epoch": 0.65, "grad_norm": 0.7199493126852369, "learning_rate": 2.868573908730739e-06, "loss": 0.1739, "lr_vision_tower": 5.737147817461478e-07, "sft_loss": 0.0859, "step": 1056 }, { "distill_loss": 0.1829, "epoch": 0.65, "grad_norm": 0.8351377816450031, "learning_rate": 2.8595449502802024e-06, "loss": 0.1717, "lr_vision_tower": 5.719089900560404e-07, "sft_loss": 0.0802, "step": 1057 }, { "distill_loss": 0.1796, "epoch": 0.65, "grad_norm": 0.7801792678779993, "learning_rate": 2.8505245296746965e-06, "loss": 0.166, "lr_vision_tower": 5.701049059349393e-07, "sft_loss": 0.0762, "step": 1058 }, { "distill_loss": 0.1776, "epoch": 0.65, "grad_norm": 0.6108142448033642, "learning_rate": 2.8415126828948686e-06, "loss": 0.1761, "lr_vision_tower": 5.683025365789736e-07, "sft_loss": 0.0873, "step": 1059 }, { "distill_loss": 0.1821, "epoch": 0.65, "grad_norm": 0.605005903208194, "learning_rate": 2.83250944588716e-06, "loss": 0.1714, "lr_vision_tower": 5.665018891774319e-07, "sft_loss": 0.0803, "step": 1060 }, { "distill_loss": 0.1809, "epoch": 0.65, "grad_norm": 1.1770759909391142, "learning_rate": 2.8235148545636776e-06, "loss": 0.1826, "lr_vision_tower": 5.647029709127354e-07, "sft_loss": 0.0921, "step": 1061 }, { "distill_loss": 0.1733, "epoch": 0.65, "grad_norm": 0.7466102072767035, "learning_rate": 2.814528944802035e-06, "loss": 0.1734, "lr_vision_tower": 5.62905788960407e-07, "sft_loss": 0.0867, "step": 1062 }, { "distill_loss": 0.1794, "epoch": 0.66, "grad_norm": 1.0398621606520493, "learning_rate": 2.805551752445222e-06, "loss": 0.1779, "lr_vision_tower": 5.611103504890443e-07, "sft_loss": 0.0882, "step": 1063 }, { "distill_loss": 0.1742, "epoch": 0.66, "grad_norm": 0.802046324357632, "learning_rate": 2.796583313301455e-06, "loss": 0.1763, "lr_vision_tower": 5.59316662660291e-07, "sft_loss": 0.0892, "step": 1064 }, { "distill_loss": 0.1789, "epoch": 0.66, "grad_norm": 2.4766612599611975, "learning_rate": 2.7876236631440373e-06, "loss": 0.1861, "lr_vision_tower": 5.575247326288074e-07, "sft_loss": 0.0966, "step": 1065 }, { "distill_loss": 0.1811, "epoch": 0.66, "grad_norm": 0.7469640383586945, "learning_rate": 2.778672837711209e-06, "loss": 0.1794, "lr_vision_tower": 5.557345675422417e-07, "sft_loss": 0.0888, "step": 1066 }, { "distill_loss": 0.1818, "epoch": 0.66, "grad_norm": 1.1629886653303263, "learning_rate": 2.7697308727060156e-06, "loss": 0.1662, "lr_vision_tower": 5.53946174541203e-07, "sft_loss": 0.0753, "step": 1067 }, { "distill_loss": 0.1875, "epoch": 0.66, "grad_norm": 1.1820488886586027, "learning_rate": 2.7607978037961602e-06, "loss": 0.1718, "lr_vision_tower": 5.52159560759232e-07, "sft_loss": 0.0781, "step": 1068 }, { "distill_loss": 0.1746, "epoch": 0.66, "grad_norm": 0.8549333837139549, "learning_rate": 2.7518736666138548e-06, "loss": 0.1875, "lr_vision_tower": 5.503747333227709e-07, "sft_loss": 0.1002, "step": 1069 }, { "distill_loss": 0.1825, "epoch": 0.66, "grad_norm": 0.8311526233363922, "learning_rate": 2.7429584967556953e-06, "loss": 0.1772, "lr_vision_tower": 5.48591699351139e-07, "sft_loss": 0.086, "step": 1070 }, { "distill_loss": 0.1861, "epoch": 0.66, "grad_norm": 0.6973914615366524, "learning_rate": 2.734052329782498e-06, "loss": 0.1939, "lr_vision_tower": 5.468104659564995e-07, "sft_loss": 0.1008, "step": 1071 }, { "distill_loss": 0.1854, "epoch": 0.66, "grad_norm": 0.8735437722190986, "learning_rate": 2.7251552012191763e-06, "loss": 0.1883, "lr_vision_tower": 5.450310402438352e-07, "sft_loss": 0.0956, "step": 1072 }, { "distill_loss": 0.1818, "epoch": 0.66, "grad_norm": 0.8982640010577154, "learning_rate": 2.7162671465545844e-06, "loss": 0.1755, "lr_vision_tower": 5.432534293109168e-07, "sft_loss": 0.0846, "step": 1073 }, { "distill_loss": 0.1783, "epoch": 0.66, "grad_norm": 0.8218261263740445, "learning_rate": 2.707388201241388e-06, "loss": 0.1659, "lr_vision_tower": 5.414776402482774e-07, "sft_loss": 0.0767, "step": 1074 }, { "distill_loss": 0.1858, "epoch": 0.66, "grad_norm": 0.6726911136751501, "learning_rate": 2.6985184006959154e-06, "loss": 0.1841, "lr_vision_tower": 5.39703680139183e-07, "sft_loss": 0.0912, "step": 1075 }, { "distill_loss": 0.177, "epoch": 0.66, "grad_norm": 1.6832500036325364, "learning_rate": 2.689657780298019e-06, "loss": 0.1675, "lr_vision_tower": 5.379315560596037e-07, "sft_loss": 0.079, "step": 1076 }, { "distill_loss": 0.1902, "epoch": 0.66, "grad_norm": 1.0294760766521538, "learning_rate": 2.6808063753909343e-06, "loss": 0.1914, "lr_vision_tower": 5.361612750781867e-07, "sft_loss": 0.0963, "step": 1077 }, { "distill_loss": 0.1772, "epoch": 0.66, "grad_norm": 0.7966538226397674, "learning_rate": 2.6719642212811325e-06, "loss": 0.166, "lr_vision_tower": 5.343928442562264e-07, "sft_loss": 0.0774, "step": 1078 }, { "distill_loss": 0.1834, "epoch": 0.67, "grad_norm": 0.7278551813140686, "learning_rate": 2.6631313532381942e-06, "loss": 0.1695, "lr_vision_tower": 5.326262706476387e-07, "sft_loss": 0.0778, "step": 1079 }, { "distill_loss": 0.1739, "epoch": 0.67, "grad_norm": 0.8340468785787901, "learning_rate": 2.65430780649465e-06, "loss": 0.1877, "lr_vision_tower": 5.3086156129893e-07, "sft_loss": 0.1007, "step": 1080 }, { "distill_loss": 0.1881, "epoch": 0.67, "grad_norm": 0.67180929429592, "learning_rate": 2.6454936162458573e-06, "loss": 0.1734, "lr_vision_tower": 5.290987232491713e-07, "sft_loss": 0.0793, "step": 1081 }, { "distill_loss": 0.1714, "epoch": 0.67, "grad_norm": 0.7389772926209152, "learning_rate": 2.636688817649849e-06, "loss": 0.1723, "lr_vision_tower": 5.273377635299698e-07, "sft_loss": 0.0866, "step": 1082 }, { "distill_loss": 0.1726, "epoch": 0.67, "grad_norm": 0.7671920146667843, "learning_rate": 2.6278934458271998e-06, "loss": 0.1667, "lr_vision_tower": 5.255786891654399e-07, "sft_loss": 0.0804, "step": 1083 }, { "distill_loss": 0.1809, "epoch": 0.67, "grad_norm": 0.913198209335827, "learning_rate": 2.619107535860876e-06, "loss": 0.1867, "lr_vision_tower": 5.238215071721751e-07, "sft_loss": 0.0963, "step": 1084 }, { "distill_loss": 0.1845, "epoch": 0.67, "grad_norm": 0.9354435163439708, "learning_rate": 2.6103311227961087e-06, "loss": 0.1682, "lr_vision_tower": 5.220662245592217e-07, "sft_loss": 0.0759, "step": 1085 }, { "distill_loss": 0.1753, "epoch": 0.67, "grad_norm": 0.7995091562057636, "learning_rate": 2.6015642416402476e-06, "loss": 0.155, "lr_vision_tower": 5.203128483280494e-07, "sft_loss": 0.0673, "step": 1086 }, { "distill_loss": 0.1874, "epoch": 0.67, "grad_norm": 0.7218823961313021, "learning_rate": 2.592806927362615e-06, "loss": 0.1783, "lr_vision_tower": 5.185613854725229e-07, "sft_loss": 0.0846, "step": 1087 }, { "distill_loss": 0.1752, "epoch": 0.67, "grad_norm": 0.7532574080109804, "learning_rate": 2.5840592148943842e-06, "loss": 0.1787, "lr_vision_tower": 5.168118429788768e-07, "sft_loss": 0.0911, "step": 1088 }, { "distill_loss": 0.1712, "epoch": 0.67, "grad_norm": 0.7752137094651758, "learning_rate": 2.5753211391284172e-06, "loss": 0.171, "lr_vision_tower": 5.150642278256834e-07, "sft_loss": 0.0854, "step": 1089 }, { "distill_loss": 0.1854, "epoch": 0.67, "grad_norm": 1.3197980287339652, "learning_rate": 2.5665927349191453e-06, "loss": 0.1759, "lr_vision_tower": 5.13318546983829e-07, "sft_loss": 0.0832, "step": 1090 }, { "distill_loss": 0.1857, "epoch": 0.67, "grad_norm": 0.7903198639020548, "learning_rate": 2.5578740370824153e-06, "loss": 0.1847, "lr_vision_tower": 5.115748074164829e-07, "sft_loss": 0.0918, "step": 1091 }, { "distill_loss": 0.1831, "epoch": 0.67, "grad_norm": 1.0364693083810224, "learning_rate": 2.549165080395363e-06, "loss": 0.1788, "lr_vision_tower": 5.098330160790725e-07, "sft_loss": 0.0872, "step": 1092 }, { "distill_loss": 0.1893, "epoch": 0.67, "grad_norm": 0.8704402195299213, "learning_rate": 2.540465899596265e-06, "loss": 0.176, "lr_vision_tower": 5.08093179919253e-07, "sft_loss": 0.0814, "step": 1093 }, { "distill_loss": 0.1936, "epoch": 0.67, "grad_norm": 0.929293911389351, "learning_rate": 2.531776529384407e-06, "loss": 0.197, "lr_vision_tower": 5.063553058768813e-07, "sft_loss": 0.1001, "step": 1094 }, { "distill_loss": 0.1786, "epoch": 0.67, "grad_norm": 0.814522505947282, "learning_rate": 2.5230970044199417e-06, "loss": 0.181, "lr_vision_tower": 5.046194008839882e-07, "sft_loss": 0.0917, "step": 1095 }, { "distill_loss": 0.185, "epoch": 0.68, "grad_norm": 0.9220713583351148, "learning_rate": 2.5144273593237456e-06, "loss": 0.1867, "lr_vision_tower": 5.028854718647491e-07, "sft_loss": 0.0942, "step": 1096 }, { "distill_loss": 0.1717, "epoch": 0.68, "grad_norm": 0.7601237488095378, "learning_rate": 2.505767628677295e-06, "loss": 0.1747, "lr_vision_tower": 5.011535257354588e-07, "sft_loss": 0.0889, "step": 1097 }, { "distill_loss": 0.1792, "epoch": 0.68, "grad_norm": 0.7699140327472862, "learning_rate": 2.4971178470225114e-06, "loss": 0.1749, "lr_vision_tower": 4.994235694045023e-07, "sft_loss": 0.0854, "step": 1098 }, { "distill_loss": 0.1846, "epoch": 0.68, "grad_norm": 0.8037349936101282, "learning_rate": 2.488478048861637e-06, "loss": 0.1978, "lr_vision_tower": 4.976956097723274e-07, "sft_loss": 0.1055, "step": 1099 }, { "distill_loss": 0.1885, "epoch": 0.68, "grad_norm": 0.8365816033055938, "learning_rate": 2.4798482686570906e-06, "loss": 0.1814, "lr_vision_tower": 4.959696537314181e-07, "sft_loss": 0.0872, "step": 1100 }, { "distill_loss": 0.1725, "epoch": 0.68, "grad_norm": 0.7704234633533229, "learning_rate": 2.4712285408313314e-06, "loss": 0.1848, "lr_vision_tower": 4.942457081662662e-07, "sft_loss": 0.0985, "step": 1101 }, { "distill_loss": 0.1769, "epoch": 0.68, "grad_norm": 1.015098279961183, "learning_rate": 2.4626188997667224e-06, "loss": 0.1692, "lr_vision_tower": 4.925237799533444e-07, "sft_loss": 0.0807, "step": 1102 }, { "distill_loss": 0.1713, "epoch": 0.68, "grad_norm": 1.5981163931751712, "learning_rate": 2.4540193798053884e-06, "loss": 0.1622, "lr_vision_tower": 4.908038759610777e-07, "sft_loss": 0.0766, "step": 1103 }, { "distill_loss": 0.1753, "epoch": 0.68, "grad_norm": 0.6954988217472919, "learning_rate": 2.4454300152490904e-06, "loss": 0.1812, "lr_vision_tower": 4.890860030498181e-07, "sft_loss": 0.0936, "step": 1104 }, { "distill_loss": 0.1791, "epoch": 0.68, "grad_norm": 0.7631515720278127, "learning_rate": 2.436850840359073e-06, "loss": 0.1823, "lr_vision_tower": 4.873701680718145e-07, "sft_loss": 0.0927, "step": 1105 }, { "distill_loss": 0.183, "epoch": 0.68, "grad_norm": 0.7727478043109092, "learning_rate": 2.4282818893559427e-06, "loss": 0.1622, "lr_vision_tower": 4.856563778711885e-07, "sft_loss": 0.0707, "step": 1106 }, { "distill_loss": 0.1812, "epoch": 0.68, "grad_norm": 0.8401379150849239, "learning_rate": 2.4197231964195233e-06, "loss": 0.1648, "lr_vision_tower": 4.839446392839047e-07, "sft_loss": 0.0742, "step": 1107 }, { "distill_loss": 0.1696, "epoch": 0.68, "grad_norm": 0.8648953067145664, "learning_rate": 2.4111747956887225e-06, "loss": 0.1646, "lr_vision_tower": 4.822349591377445e-07, "sft_loss": 0.0798, "step": 1108 }, { "distill_loss": 0.1747, "epoch": 0.68, "grad_norm": 0.7106569748036413, "learning_rate": 2.4026367212613895e-06, "loss": 0.17, "lr_vision_tower": 4.805273442522778e-07, "sft_loss": 0.0827, "step": 1109 }, { "distill_loss": 0.1855, "epoch": 0.68, "grad_norm": 0.8804284032932794, "learning_rate": 2.3941090071941885e-06, "loss": 0.169, "lr_vision_tower": 4.788218014388377e-07, "sft_loss": 0.0762, "step": 1110 }, { "distill_loss": 0.1766, "epoch": 0.68, "grad_norm": 0.8223249846716819, "learning_rate": 2.3855916875024584e-06, "loss": 0.164, "lr_vision_tower": 4.771183375004915e-07, "sft_loss": 0.0757, "step": 1111 }, { "distill_loss": 0.1704, "epoch": 0.69, "grad_norm": 0.8257803461752944, "learning_rate": 2.3770847961600735e-06, "loss": 0.1527, "lr_vision_tower": 4.7541695923201464e-07, "sft_loss": 0.0675, "step": 1112 }, { "distill_loss": 0.175, "epoch": 0.69, "grad_norm": 0.8874410368697585, "learning_rate": 2.3685883670993175e-06, "loss": 0.1686, "lr_vision_tower": 4.7371767341986344e-07, "sft_loss": 0.0811, "step": 1113 }, { "distill_loss": 0.1786, "epoch": 0.69, "grad_norm": 0.7502085396694815, "learning_rate": 2.360102434210735e-06, "loss": 0.1645, "lr_vision_tower": 4.720204868421469e-07, "sft_loss": 0.0752, "step": 1114 }, { "distill_loss": 0.1787, "epoch": 0.69, "grad_norm": 0.8541581045913682, "learning_rate": 2.3516270313430085e-06, "loss": 0.1734, "lr_vision_tower": 4.7032540626860163e-07, "sft_loss": 0.0841, "step": 1115 }, { "distill_loss": 0.1807, "epoch": 0.69, "grad_norm": 0.9146472006419081, "learning_rate": 2.3431621923028146e-06, "loss": 0.1665, "lr_vision_tower": 4.686324384605629e-07, "sft_loss": 0.0762, "step": 1116 }, { "distill_loss": 0.1808, "epoch": 0.69, "grad_norm": 0.9949366485802833, "learning_rate": 2.334707950854697e-06, "loss": 0.1862, "lr_vision_tower": 4.6694159017093937e-07, "sft_loss": 0.0958, "step": 1117 }, { "distill_loss": 0.1884, "epoch": 0.69, "grad_norm": 1.1940569108921786, "learning_rate": 2.3262643407209257e-06, "loss": 0.1898, "lr_vision_tower": 4.6525286814418506e-07, "sft_loss": 0.0956, "step": 1118 }, { "distill_loss": 0.1801, "epoch": 0.69, "grad_norm": 0.7356238589256794, "learning_rate": 2.3178313955813637e-06, "loss": 0.1677, "lr_vision_tower": 4.6356627911627265e-07, "sft_loss": 0.0777, "step": 1119 }, { "distill_loss": 0.1675, "epoch": 0.69, "grad_norm": 0.7572122779496455, "learning_rate": 2.3094091490733373e-06, "loss": 0.1854, "lr_vision_tower": 4.618818298146674e-07, "sft_loss": 0.1016, "step": 1120 }, { "distill_loss": 0.1777, "epoch": 0.69, "grad_norm": 0.8107943502715718, "learning_rate": 2.300997634791491e-06, "loss": 0.1808, "lr_vision_tower": 4.601995269582981e-07, "sft_loss": 0.0919, "step": 1121 }, { "distill_loss": 0.181, "epoch": 0.69, "grad_norm": 0.9328282804129266, "learning_rate": 2.292596886287669e-06, "loss": 0.1741, "lr_vision_tower": 4.5851937725753377e-07, "sft_loss": 0.0836, "step": 1122 }, { "distill_loss": 0.1846, "epoch": 0.69, "grad_norm": 0.6899940879934611, "learning_rate": 2.2842069370707643e-06, "loss": 0.1725, "lr_vision_tower": 4.568413874141528e-07, "sft_loss": 0.0802, "step": 1123 }, { "distill_loss": 0.1696, "epoch": 0.69, "grad_norm": 0.6326629218157246, "learning_rate": 2.275827820606601e-06, "loss": 0.1534, "lr_vision_tower": 4.551655641213201e-07, "sft_loss": 0.0686, "step": 1124 }, { "distill_loss": 0.175, "epoch": 0.69, "grad_norm": 0.7402041924065775, "learning_rate": 2.26745957031779e-06, "loss": 0.1731, "lr_vision_tower": 4.534919140635579e-07, "sft_loss": 0.0856, "step": 1125 }, { "distill_loss": 0.1834, "epoch": 0.69, "grad_norm": 0.7284222629333426, "learning_rate": 2.259102219583602e-06, "loss": 0.1724, "lr_vision_tower": 4.5182044391672036e-07, "sft_loss": 0.0807, "step": 1126 }, { "distill_loss": 0.1735, "epoch": 0.69, "grad_norm": 0.9369974711942431, "learning_rate": 2.2507558017398263e-06, "loss": 0.1759, "lr_vision_tower": 4.501511603479652e-07, "sft_loss": 0.0891, "step": 1127 }, { "distill_loss": 0.1859, "epoch": 0.7, "grad_norm": 0.7769937667534744, "learning_rate": 2.2424203500786473e-06, "loss": 0.1835, "lr_vision_tower": 4.484840700157294e-07, "sft_loss": 0.0906, "step": 1128 }, { "distill_loss": 0.1707, "epoch": 0.7, "grad_norm": 0.8252368937702702, "learning_rate": 2.2340958978485074e-06, "loss": 0.1838, "lr_vision_tower": 4.468191795697014e-07, "sft_loss": 0.0985, "step": 1129 }, { "distill_loss": 0.1805, "epoch": 0.7, "grad_norm": 1.2373320936129386, "learning_rate": 2.2257824782539728e-06, "loss": 0.1892, "lr_vision_tower": 4.451564956507945e-07, "sft_loss": 0.0989, "step": 1130 }, { "distill_loss": 0.1785, "epoch": 0.7, "grad_norm": 0.835192508882424, "learning_rate": 2.217480124455606e-06, "loss": 0.1758, "lr_vision_tower": 4.4349602489112117e-07, "sft_loss": 0.0866, "step": 1131 }, { "distill_loss": 0.1801, "epoch": 0.7, "grad_norm": 1.0313244868820624, "learning_rate": 2.2091888695698238e-06, "loss": 0.1817, "lr_vision_tower": 4.418377739139647e-07, "sft_loss": 0.0916, "step": 1132 }, { "distill_loss": 0.1726, "epoch": 0.7, "grad_norm": 0.6175350020335644, "learning_rate": 2.2009087466687797e-06, "loss": 0.1614, "lr_vision_tower": 4.4018174933375587e-07, "sft_loss": 0.0751, "step": 1133 }, { "distill_loss": 0.1859, "epoch": 0.7, "grad_norm": 1.0452537232583359, "learning_rate": 2.192639788780216e-06, "loss": 0.1675, "lr_vision_tower": 4.3852795775604313e-07, "sft_loss": 0.0746, "step": 1134 }, { "distill_loss": 0.1908, "epoch": 0.7, "grad_norm": 0.8737116200304627, "learning_rate": 2.184382028887345e-06, "loss": 0.1944, "lr_vision_tower": 4.36876405777469e-07, "sft_loss": 0.099, "step": 1135 }, { "distill_loss": 0.1842, "epoch": 0.7, "grad_norm": 0.8577454085144611, "learning_rate": 2.1761354999287127e-06, "loss": 0.1853, "lr_vision_tower": 4.3522709998574245e-07, "sft_loss": 0.0932, "step": 1136 }, { "distill_loss": 0.1751, "epoch": 0.7, "grad_norm": 0.875490502518282, "learning_rate": 2.167900234798065e-06, "loss": 0.1704, "lr_vision_tower": 4.3358004695961295e-07, "sft_loss": 0.0828, "step": 1137 }, { "distill_loss": 0.1765, "epoch": 0.7, "grad_norm": 1.0408508234052312, "learning_rate": 2.159676266344222e-06, "loss": 0.1915, "lr_vision_tower": 4.3193525326884426e-07, "sft_loss": 0.1032, "step": 1138 }, { "distill_loss": 0.1858, "epoch": 0.7, "grad_norm": 0.7728577466857959, "learning_rate": 2.1514636273709364e-06, "loss": 0.1691, "lr_vision_tower": 4.3029272547418726e-07, "sft_loss": 0.0762, "step": 1139 }, { "distill_loss": 0.1789, "epoch": 0.7, "grad_norm": 2.1565025051500553, "learning_rate": 2.143262350636779e-06, "loss": 0.1758, "lr_vision_tower": 4.286524701273557e-07, "sft_loss": 0.0863, "step": 1140 }, { "distill_loss": 0.1831, "epoch": 0.7, "grad_norm": 0.6807641858710474, "learning_rate": 2.1350724688549906e-06, "loss": 0.1814, "lr_vision_tower": 4.270144937709981e-07, "sft_loss": 0.0899, "step": 1141 }, { "distill_loss": 0.1745, "epoch": 0.7, "grad_norm": 0.6478544503815301, "learning_rate": 2.126894014693366e-06, "loss": 0.1714, "lr_vision_tower": 4.2537880293867313e-07, "sft_loss": 0.0841, "step": 1142 }, { "distill_loss": 0.1911, "epoch": 0.7, "grad_norm": 0.8264021582998446, "learning_rate": 2.1187270207741143e-06, "loss": 0.1873, "lr_vision_tower": 4.237454041548229e-07, "sft_loss": 0.0917, "step": 1143 }, { "distill_loss": 0.1837, "epoch": 0.71, "grad_norm": 0.8038088279805404, "learning_rate": 2.110571519673736e-06, "loss": 0.1754, "lr_vision_tower": 4.2211430393474713e-07, "sft_loss": 0.0835, "step": 1144 }, { "distill_loss": 0.1689, "epoch": 0.71, "grad_norm": 0.7925779574347088, "learning_rate": 2.1024275439228814e-06, "loss": 0.1686, "lr_vision_tower": 4.204855087845762e-07, "sft_loss": 0.0841, "step": 1145 }, { "distill_loss": 0.183, "epoch": 0.71, "grad_norm": 0.7609429565172778, "learning_rate": 2.094295126006235e-06, "loss": 0.1827, "lr_vision_tower": 4.188590252012469e-07, "sft_loss": 0.0912, "step": 1146 }, { "distill_loss": 0.1792, "epoch": 0.71, "grad_norm": 0.6910487666706121, "learning_rate": 2.086174298362378e-06, "loss": 0.1825, "lr_vision_tower": 4.1723485967247553e-07, "sft_loss": 0.0929, "step": 1147 }, { "distill_loss": 0.1788, "epoch": 0.71, "grad_norm": 0.8763107348245328, "learning_rate": 2.0780650933836593e-06, "loss": 0.179, "lr_vision_tower": 4.1561301867673183e-07, "sft_loss": 0.0896, "step": 1148 }, { "distill_loss": 0.1778, "epoch": 0.71, "grad_norm": 0.5882235120103609, "learning_rate": 2.06996754341607e-06, "loss": 0.168, "lr_vision_tower": 4.139935086832139e-07, "sft_loss": 0.0791, "step": 1149 }, { "distill_loss": 0.1751, "epoch": 0.71, "grad_norm": 0.8129248537114786, "learning_rate": 2.0618816807591056e-06, "loss": 0.1731, "lr_vision_tower": 4.123763361518211e-07, "sft_loss": 0.0856, "step": 1150 }, { "distill_loss": 0.1839, "epoch": 0.71, "grad_norm": 0.6005288727532221, "learning_rate": 2.0538075376656517e-06, "loss": 0.1721, "lr_vision_tower": 4.1076150753313034e-07, "sft_loss": 0.0802, "step": 1151 }, { "distill_loss": 0.1724, "epoch": 0.71, "grad_norm": 0.8089885311199546, "learning_rate": 2.045745146341838e-06, "loss": 0.1667, "lr_vision_tower": 4.0914902926836757e-07, "sft_loss": 0.0805, "step": 1152 }, { "distill_loss": 0.1716, "epoch": 0.71, "grad_norm": 0.8209191416493364, "learning_rate": 2.0376945389469255e-06, "loss": 0.1739, "lr_vision_tower": 4.0753890778938506e-07, "sft_loss": 0.0881, "step": 1153 }, { "distill_loss": 0.175, "epoch": 0.71, "grad_norm": 0.7274011150781003, "learning_rate": 2.029655747593169e-06, "loss": 0.1861, "lr_vision_tower": 4.0593114951863374e-07, "sft_loss": 0.0986, "step": 1154 }, { "distill_loss": 0.1812, "epoch": 0.71, "grad_norm": 0.8885144646437683, "learning_rate": 2.0216288043456916e-06, "loss": 0.1685, "lr_vision_tower": 4.043257608691383e-07, "sft_loss": 0.0779, "step": 1155 }, { "distill_loss": 0.1809, "epoch": 0.71, "grad_norm": 0.8283331701367896, "learning_rate": 2.0136137412223587e-06, "loss": 0.1702, "lr_vision_tower": 4.027227482444717e-07, "sft_loss": 0.0797, "step": 1156 }, { "distill_loss": 0.182, "epoch": 0.71, "grad_norm": 1.0929272796565281, "learning_rate": 2.0056105901936423e-06, "loss": 0.1788, "lr_vision_tower": 4.011221180387284e-07, "sft_loss": 0.0878, "step": 1157 }, { "distill_loss": 0.1844, "epoch": 0.71, "grad_norm": 0.8799166749746606, "learning_rate": 1.997619383182508e-06, "loss": 0.1795, "lr_vision_tower": 3.9952387663650153e-07, "sft_loss": 0.0873, "step": 1158 }, { "distill_loss": 0.1716, "epoch": 0.71, "grad_norm": 0.8125352340418215, "learning_rate": 1.98964015206427e-06, "loss": 0.1687, "lr_vision_tower": 3.9792803041285393e-07, "sft_loss": 0.0829, "step": 1159 }, { "distill_loss": 0.1666, "epoch": 0.72, "grad_norm": 1.2879668899397358, "learning_rate": 1.98167292866648e-06, "loss": 0.1691, "lr_vision_tower": 3.963345857332959e-07, "sft_loss": 0.0858, "step": 1160 }, { "distill_loss": 0.1689, "epoch": 0.72, "grad_norm": 0.9433063756006003, "learning_rate": 1.973717744768791e-06, "loss": 0.1681, "lr_vision_tower": 3.9474354895375815e-07, "sft_loss": 0.0837, "step": 1161 }, { "distill_loss": 0.1747, "epoch": 0.72, "grad_norm": 0.7836150789526019, "learning_rate": 1.9657746321028325e-06, "loss": 0.1821, "lr_vision_tower": 3.931549264205665e-07, "sft_loss": 0.0947, "step": 1162 }, { "distill_loss": 0.1716, "epoch": 0.72, "grad_norm": 1.6452485624084534, "learning_rate": 1.9578436223520864e-06, "loss": 0.1692, "lr_vision_tower": 3.915687244704172e-07, "sft_loss": 0.0834, "step": 1163 }, { "distill_loss": 0.1796, "epoch": 0.72, "grad_norm": 0.756541757714199, "learning_rate": 1.9499247471517534e-06, "loss": 0.1641, "lr_vision_tower": 3.8998494943035066e-07, "sft_loss": 0.0743, "step": 1164 }, { "distill_loss": 0.1833, "epoch": 0.72, "grad_norm": 0.7316417380276786, "learning_rate": 1.9420180380886375e-06, "loss": 0.1721, "lr_vision_tower": 3.884036076177275e-07, "sft_loss": 0.0804, "step": 1165 }, { "distill_loss": 0.1696, "epoch": 0.72, "grad_norm": 0.8140082887059632, "learning_rate": 1.934123526701011e-06, "loss": 0.1607, "lr_vision_tower": 3.868247053402022e-07, "sft_loss": 0.0759, "step": 1166 }, { "distill_loss": 0.1738, "epoch": 0.72, "grad_norm": 0.766563179876473, "learning_rate": 1.926241244478496e-06, "loss": 0.1631, "lr_vision_tower": 3.8524824889569916e-07, "sft_loss": 0.0762, "step": 1167 }, { "distill_loss": 0.1736, "epoch": 0.72, "grad_norm": 0.9088808095685758, "learning_rate": 1.918371222861928e-06, "loss": 0.1682, "lr_vision_tower": 3.836742445723856e-07, "sft_loss": 0.0814, "step": 1168 }, { "distill_loss": 0.1809, "epoch": 0.72, "grad_norm": 1.059528459183986, "learning_rate": 1.910513493243246e-06, "loss": 0.1784, "lr_vision_tower": 3.821026986486492e-07, "sft_loss": 0.088, "step": 1169 }, { "distill_loss": 0.1759, "epoch": 0.72, "grad_norm": 0.7834706816075623, "learning_rate": 1.9026680869653502e-06, "loss": 0.1628, "lr_vision_tower": 3.8053361739307e-07, "sft_loss": 0.0748, "step": 1170 }, { "distill_loss": 0.1805, "epoch": 0.72, "grad_norm": 0.8266719784991254, "learning_rate": 1.8948350353219913e-06, "loss": 0.1763, "lr_vision_tower": 3.7896700706439823e-07, "sft_loss": 0.086, "step": 1171 }, { "distill_loss": 0.1725, "epoch": 0.72, "grad_norm": 0.9256940560218515, "learning_rate": 1.8870143695576382e-06, "loss": 0.1799, "lr_vision_tower": 3.774028739115276e-07, "sft_loss": 0.0936, "step": 1172 }, { "distill_loss": 0.1702, "epoch": 0.72, "grad_norm": 0.9080549690526186, "learning_rate": 1.8792061208673544e-06, "loss": 0.1715, "lr_vision_tower": 3.7584122417347087e-07, "sft_loss": 0.0864, "step": 1173 }, { "distill_loss": 0.1791, "epoch": 0.72, "grad_norm": 0.6538815749605286, "learning_rate": 1.8714103203966778e-06, "loss": 0.1644, "lr_vision_tower": 3.742820640793355e-07, "sft_loss": 0.0748, "step": 1174 }, { "distill_loss": 0.1651, "epoch": 0.72, "grad_norm": 0.9729930139800578, "learning_rate": 1.8636269992414847e-06, "loss": 0.1698, "lr_vision_tower": 3.727253998482969e-07, "sft_loss": 0.0873, "step": 1175 }, { "distill_loss": 0.1674, "epoch": 0.72, "grad_norm": 0.7634317487760693, "learning_rate": 1.8558561884478837e-06, "loss": 0.1822, "lr_vision_tower": 3.711712376895767e-07, "sft_loss": 0.0986, "step": 1176 }, { "distill_loss": 0.1877, "epoch": 0.73, "grad_norm": 0.7537268849126174, "learning_rate": 1.8480979190120735e-06, "loss": 0.1817, "lr_vision_tower": 3.6961958380241466e-07, "sft_loss": 0.0878, "step": 1177 }, { "distill_loss": 0.177, "epoch": 0.73, "grad_norm": 0.6342717740278067, "learning_rate": 1.8403522218802328e-06, "loss": 0.1747, "lr_vision_tower": 3.6807044437604653e-07, "sft_loss": 0.0862, "step": 1178 }, { "distill_loss": 0.1819, "epoch": 0.73, "grad_norm": 0.7372862894215255, "learning_rate": 1.8326191279483917e-06, "loss": 0.1891, "lr_vision_tower": 3.665238255896783e-07, "sft_loss": 0.0982, "step": 1179 }, { "distill_loss": 0.1761, "epoch": 0.73, "grad_norm": 0.6793540960532686, "learning_rate": 1.8248986680623077e-06, "loss": 0.1804, "lr_vision_tower": 3.649797336124615e-07, "sft_loss": 0.0924, "step": 1180 }, { "distill_loss": 0.1721, "epoch": 0.73, "grad_norm": 1.101599255937685, "learning_rate": 1.8171908730173455e-06, "loss": 0.1642, "lr_vision_tower": 3.63438174603469e-07, "sft_loss": 0.0782, "step": 1181 }, { "distill_loss": 0.1826, "epoch": 0.73, "grad_norm": 0.7586538513736393, "learning_rate": 1.8094957735583463e-06, "loss": 0.1904, "lr_vision_tower": 3.6189915471166923e-07, "sft_loss": 0.0991, "step": 1182 }, { "distill_loss": 0.1794, "epoch": 0.73, "grad_norm": 0.8557596130457935, "learning_rate": 1.8018134003795196e-06, "loss": 0.1888, "lr_vision_tower": 3.6036268007590385e-07, "sft_loss": 0.0991, "step": 1183 }, { "distill_loss": 0.1789, "epoch": 0.73, "grad_norm": 0.8002812281206538, "learning_rate": 1.7941437841243026e-06, "loss": 0.181, "lr_vision_tower": 3.5882875682486046e-07, "sft_loss": 0.0915, "step": 1184 }, { "distill_loss": 0.1749, "epoch": 0.73, "grad_norm": 0.8470273622457387, "learning_rate": 1.7864869553852604e-06, "loss": 0.1822, "lr_vision_tower": 3.5729739107705206e-07, "sft_loss": 0.0948, "step": 1185 }, { "distill_loss": 0.1805, "epoch": 0.73, "grad_norm": 1.0257611139479863, "learning_rate": 1.7788429447039385e-06, "loss": 0.1771, "lr_vision_tower": 3.5576858894078766e-07, "sft_loss": 0.0868, "step": 1186 }, { "distill_loss": 0.1798, "epoch": 0.73, "grad_norm": 0.7669155877491266, "learning_rate": 1.7712117825707626e-06, "loss": 0.1852, "lr_vision_tower": 3.5424235651415244e-07, "sft_loss": 0.0954, "step": 1187 }, { "distill_loss": 0.171, "epoch": 0.73, "grad_norm": 0.6392158008564098, "learning_rate": 1.7635934994249022e-06, "loss": 0.1583, "lr_vision_tower": 3.527186998849804e-07, "sft_loss": 0.0728, "step": 1188 }, { "distill_loss": 0.1702, "epoch": 0.73, "grad_norm": 0.6483009327552532, "learning_rate": 1.7559881256541583e-06, "loss": 0.1616, "lr_vision_tower": 3.5119762513083163e-07, "sft_loss": 0.0764, "step": 1189 }, { "distill_loss": 0.1745, "epoch": 0.73, "grad_norm": 0.8991785397701382, "learning_rate": 1.7483956915948397e-06, "loss": 0.1573, "lr_vision_tower": 3.496791383189679e-07, "sft_loss": 0.0701, "step": 1190 }, { "distill_loss": 0.1862, "epoch": 0.73, "grad_norm": 0.86515760053715, "learning_rate": 1.7408162275316392e-06, "loss": 0.1909, "lr_vision_tower": 3.481632455063278e-07, "sft_loss": 0.0978, "step": 1191 }, { "distill_loss": 0.1793, "epoch": 0.73, "grad_norm": 0.7092327777145084, "learning_rate": 1.7332497636975176e-06, "loss": 0.1866, "lr_vision_tower": 3.4664995273950347e-07, "sft_loss": 0.0969, "step": 1192 }, { "distill_loss": 0.1744, "epoch": 0.74, "grad_norm": 0.8551196871048818, "learning_rate": 1.7256963302735752e-06, "loss": 0.162, "lr_vision_tower": 3.45139266054715e-07, "sft_loss": 0.0748, "step": 1193 }, { "distill_loss": 0.1844, "epoch": 0.74, "grad_norm": 0.9152690822509827, "learning_rate": 1.7181559573889428e-06, "loss": 0.1811, "lr_vision_tower": 3.436311914777885e-07, "sft_loss": 0.0889, "step": 1194 }, { "distill_loss": 0.1748, "epoch": 0.74, "grad_norm": 0.8344259454959938, "learning_rate": 1.7106286751206502e-06, "loss": 0.1567, "lr_vision_tower": 3.4212573502413e-07, "sft_loss": 0.0693, "step": 1195 }, { "distill_loss": 0.1753, "epoch": 0.74, "grad_norm": 1.523090486507288, "learning_rate": 1.7031145134935145e-06, "loss": 0.1784, "lr_vision_tower": 3.4062290269870285e-07, "sft_loss": 0.0907, "step": 1196 }, { "distill_loss": 0.1727, "epoch": 0.74, "grad_norm": 0.7926667910893, "learning_rate": 1.6956135024800163e-06, "loss": 0.1674, "lr_vision_tower": 3.391227004960032e-07, "sft_loss": 0.081, "step": 1197 }, { "distill_loss": 0.1751, "epoch": 0.74, "grad_norm": 1.1034062217425902, "learning_rate": 1.6881256720001814e-06, "loss": 0.1681, "lr_vision_tower": 3.3762513440003626e-07, "sft_loss": 0.0806, "step": 1198 }, { "distill_loss": 0.1765, "epoch": 0.74, "grad_norm": 0.6945167426306443, "learning_rate": 1.6806510519214624e-06, "loss": 0.1547, "lr_vision_tower": 3.3613021038429244e-07, "sft_loss": 0.0664, "step": 1199 }, { "distill_loss": 0.1781, "epoch": 0.74, "grad_norm": 0.8118012413936914, "learning_rate": 1.6731896720586138e-06, "loss": 0.1711, "lr_vision_tower": 3.346379344117227e-07, "sft_loss": 0.082, "step": 1200 }, { "distill_loss": 0.167, "epoch": 0.74, "grad_norm": 0.8588061895617461, "learning_rate": 1.6657415621735822e-06, "loss": 0.1656, "lr_vision_tower": 3.331483124347164e-07, "sft_loss": 0.0822, "step": 1201 }, { "distill_loss": 0.1695, "epoch": 0.74, "grad_norm": 2.220495981063578, "learning_rate": 1.6583067519753776e-06, "loss": 0.1607, "lr_vision_tower": 3.3166135039507546e-07, "sft_loss": 0.0759, "step": 1202 }, { "distill_loss": 0.1794, "epoch": 0.74, "grad_norm": 0.8130065545762276, "learning_rate": 1.6508852711199697e-06, "loss": 0.1688, "lr_vision_tower": 3.301770542239939e-07, "sft_loss": 0.0791, "step": 1203 }, { "distill_loss": 0.1785, "epoch": 0.74, "grad_norm": 0.8822631484739442, "learning_rate": 1.6434771492101487e-06, "loss": 0.1802, "lr_vision_tower": 3.286954298420297e-07, "sft_loss": 0.0909, "step": 1204 }, { "distill_loss": 0.1832, "epoch": 0.74, "grad_norm": 1.0623799551438877, "learning_rate": 1.636082415795427e-06, "loss": 0.1761, "lr_vision_tower": 3.272164831590853e-07, "sft_loss": 0.0845, "step": 1205 }, { "distill_loss": 0.1754, "epoch": 0.74, "grad_norm": 0.8671885284681443, "learning_rate": 1.6287011003719105e-06, "loss": 0.1678, "lr_vision_tower": 3.2574022007438205e-07, "sft_loss": 0.0801, "step": 1206 }, { "distill_loss": 0.1768, "epoch": 0.74, "grad_norm": 0.8888647478691328, "learning_rate": 1.6213332323821807e-06, "loss": 0.1646, "lr_vision_tower": 3.242666464764361e-07, "sft_loss": 0.0762, "step": 1207 }, { "distill_loss": 0.1653, "epoch": 0.74, "grad_norm": 0.9657011438187956, "learning_rate": 1.6139788412151836e-06, "loss": 0.1693, "lr_vision_tower": 3.227957682430367e-07, "sft_loss": 0.0867, "step": 1208 }, { "distill_loss": 0.1782, "epoch": 0.75, "grad_norm": 0.9833202373570398, "learning_rate": 1.6066379562061075e-06, "loss": 0.1693, "lr_vision_tower": 3.2132759124122146e-07, "sft_loss": 0.0802, "step": 1209 }, { "distill_loss": 0.1879, "epoch": 0.75, "grad_norm": 0.7221279842153129, "learning_rate": 1.599310606636269e-06, "loss": 0.1666, "lr_vision_tower": 3.1986212132725374e-07, "sft_loss": 0.0726, "step": 1210 }, { "distill_loss": 0.1798, "epoch": 0.75, "grad_norm": 0.8630378856141436, "learning_rate": 1.5919968217329895e-06, "loss": 0.1828, "lr_vision_tower": 3.1839936434659785e-07, "sft_loss": 0.0929, "step": 1211 }, { "distill_loss": 0.1731, "epoch": 0.75, "grad_norm": 0.7157474611356492, "learning_rate": 1.5846966306694911e-06, "loss": 0.1606, "lr_vision_tower": 3.169393261338982e-07, "sft_loss": 0.074, "step": 1212 }, { "distill_loss": 0.1782, "epoch": 0.75, "grad_norm": 0.9565718086550897, "learning_rate": 1.5774100625647654e-06, "loss": 0.1755, "lr_vision_tower": 3.1548201251295305e-07, "sft_loss": 0.0865, "step": 1213 }, { "distill_loss": 0.1828, "epoch": 0.75, "grad_norm": 0.7738150629661521, "learning_rate": 1.5701371464834691e-06, "loss": 0.1741, "lr_vision_tower": 3.140274292966938e-07, "sft_loss": 0.0827, "step": 1214 }, { "distill_loss": 0.1695, "epoch": 0.75, "grad_norm": 0.8655562701602284, "learning_rate": 1.5628779114358034e-06, "loss": 0.1757, "lr_vision_tower": 3.1257558228716066e-07, "sft_loss": 0.091, "step": 1215 }, { "distill_loss": 0.1861, "epoch": 0.75, "grad_norm": 0.9014653175837968, "learning_rate": 1.555632386377398e-06, "loss": 0.1756, "lr_vision_tower": 3.1112647727547956e-07, "sft_loss": 0.0826, "step": 1216 }, { "distill_loss": 0.1775, "epoch": 0.75, "grad_norm": 0.691068052953736, "learning_rate": 1.5484006002091978e-06, "loss": 0.1719, "lr_vision_tower": 3.096801200418395e-07, "sft_loss": 0.0831, "step": 1217 }, { "distill_loss": 0.1651, "epoch": 0.75, "grad_norm": 0.7589618390191059, "learning_rate": 1.5411825817773412e-06, "loss": 0.1733, "lr_vision_tower": 3.082365163554682e-07, "sft_loss": 0.0908, "step": 1218 }, { "distill_loss": 0.1713, "epoch": 0.75, "grad_norm": 0.785418416372367, "learning_rate": 1.5339783598730568e-06, "loss": 0.1629, "lr_vision_tower": 3.067956719746113e-07, "sft_loss": 0.0772, "step": 1219 }, { "distill_loss": 0.1635, "epoch": 0.75, "grad_norm": 0.7986496359785029, "learning_rate": 1.5267879632325321e-06, "loss": 0.1602, "lr_vision_tower": 3.053575926465064e-07, "sft_loss": 0.0784, "step": 1220 }, { "distill_loss": 0.1771, "epoch": 0.75, "grad_norm": 1.040556862300299, "learning_rate": 1.5196114205368217e-06, "loss": 0.1695, "lr_vision_tower": 3.039222841073643e-07, "sft_loss": 0.081, "step": 1221 }, { "distill_loss": 0.1799, "epoch": 0.75, "grad_norm": 0.8474521696543564, "learning_rate": 1.5124487604117071e-06, "loss": 0.1708, "lr_vision_tower": 3.024897520823414e-07, "sft_loss": 0.0808, "step": 1222 }, { "distill_loss": 0.168, "epoch": 0.75, "grad_norm": 1.0018106007590388, "learning_rate": 1.5053000114276012e-06, "loss": 0.1777, "lr_vision_tower": 3.0106000228552017e-07, "sft_loss": 0.0937, "step": 1223 }, { "distill_loss": 0.177, "epoch": 0.75, "grad_norm": 0.8655207264532511, "learning_rate": 1.4981652020994293e-06, "loss": 0.1678, "lr_vision_tower": 2.996330404198858e-07, "sft_loss": 0.0793, "step": 1224 }, { "distill_loss": 0.1796, "epoch": 0.76, "grad_norm": 0.8476993353049859, "learning_rate": 1.4910443608865083e-06, "loss": 0.1675, "lr_vision_tower": 2.9820887217730163e-07, "sft_loss": 0.0777, "step": 1225 }, { "distill_loss": 0.1842, "epoch": 0.76, "grad_norm": 0.8367899157525844, "learning_rate": 1.4839375161924446e-06, "loss": 0.1646, "lr_vision_tower": 2.967875032384889e-07, "sft_loss": 0.0724, "step": 1226 }, { "distill_loss": 0.1747, "epoch": 0.76, "grad_norm": 0.8080695399480115, "learning_rate": 1.4768446963650135e-06, "loss": 0.1776, "lr_vision_tower": 2.953689392730027e-07, "sft_loss": 0.0902, "step": 1227 }, { "distill_loss": 0.1666, "epoch": 0.76, "grad_norm": 0.9636520858256987, "learning_rate": 1.4697659296960488e-06, "loss": 0.1707, "lr_vision_tower": 2.939531859392097e-07, "sft_loss": 0.0874, "step": 1228 }, { "distill_loss": 0.1667, "epoch": 0.76, "grad_norm": 1.1733905130428062, "learning_rate": 1.4627012444213257e-06, "loss": 0.1579, "lr_vision_tower": 2.9254024888426507e-07, "sft_loss": 0.0746, "step": 1229 }, { "distill_loss": 0.1772, "epoch": 0.76, "grad_norm": 0.8970721586412459, "learning_rate": 1.4556506687204569e-06, "loss": 0.1871, "lr_vision_tower": 2.9113013374409133e-07, "sft_loss": 0.0985, "step": 1230 }, { "distill_loss": 0.177, "epoch": 0.76, "grad_norm": 1.1150096584952056, "learning_rate": 1.4486142307167683e-06, "loss": 0.1805, "lr_vision_tower": 2.8972284614335363e-07, "sft_loss": 0.092, "step": 1231 }, { "distill_loss": 0.1844, "epoch": 0.76, "grad_norm": 0.8991985289933199, "learning_rate": 1.4415919584771999e-06, "loss": 0.177, "lr_vision_tower": 2.883183916954399e-07, "sft_loss": 0.0848, "step": 1232 }, { "distill_loss": 0.1689, "epoch": 0.76, "grad_norm": 0.9402709122187085, "learning_rate": 1.4345838800121824e-06, "loss": 0.169, "lr_vision_tower": 2.8691677600243647e-07, "sft_loss": 0.0845, "step": 1233 }, { "distill_loss": 0.1665, "epoch": 0.76, "grad_norm": 0.7780489187975594, "learning_rate": 1.4275900232755345e-06, "loss": 0.1669, "lr_vision_tower": 2.8551800465510687e-07, "sft_loss": 0.0836, "step": 1234 }, { "distill_loss": 0.1709, "epoch": 0.76, "grad_norm": 0.9419748511527843, "learning_rate": 1.4206104161643452e-06, "loss": 0.1817, "lr_vision_tower": 2.84122083232869e-07, "sft_loss": 0.0962, "step": 1235 }, { "distill_loss": 0.1739, "epoch": 0.76, "grad_norm": 0.7264385294132492, "learning_rate": 1.4136450865188623e-06, "loss": 0.1755, "lr_vision_tower": 2.827290173037724e-07, "sft_loss": 0.0885, "step": 1236 }, { "distill_loss": 0.1801, "epoch": 0.76, "grad_norm": 1.206191117006253, "learning_rate": 1.406694062122389e-06, "loss": 0.172, "lr_vision_tower": 2.813388124244778e-07, "sft_loss": 0.0819, "step": 1237 }, { "distill_loss": 0.1727, "epoch": 0.76, "grad_norm": 0.8419000550519812, "learning_rate": 1.3997573707011598e-06, "loss": 0.1779, "lr_vision_tower": 2.799514741402319e-07, "sft_loss": 0.0915, "step": 1238 }, { "distill_loss": 0.1781, "epoch": 0.76, "grad_norm": 1.045818289852922, "learning_rate": 1.3928350399242484e-06, "loss": 0.192, "lr_vision_tower": 2.785670079848497e-07, "sft_loss": 0.103, "step": 1239 }, { "distill_loss": 0.1828, "epoch": 0.76, "grad_norm": 0.7999803774326978, "learning_rate": 1.3859270974034378e-06, "loss": 0.1716, "lr_vision_tower": 2.771854194806875e-07, "sft_loss": 0.0802, "step": 1240 }, { "distill_loss": 0.1744, "epoch": 0.76, "grad_norm": 0.7316943311785112, "learning_rate": 1.3790335706931229e-06, "loss": 0.1577, "lr_vision_tower": 2.7580671413862454e-07, "sft_loss": 0.0705, "step": 1241 }, { "distill_loss": 0.168, "epoch": 0.77, "grad_norm": 0.7555007501015539, "learning_rate": 1.372154487290198e-06, "loss": 0.1734, "lr_vision_tower": 2.744308974580396e-07, "sft_loss": 0.0894, "step": 1242 }, { "distill_loss": 0.1681, "epoch": 0.77, "grad_norm": 0.8237932418381402, "learning_rate": 1.3652898746339417e-06, "loss": 0.1728, "lr_vision_tower": 2.730579749267883e-07, "sft_loss": 0.0888, "step": 1243 }, { "distill_loss": 0.1695, "epoch": 0.77, "grad_norm": 0.9377572102952562, "learning_rate": 1.3584397601059158e-06, "loss": 0.1719, "lr_vision_tower": 2.7168795202118313e-07, "sft_loss": 0.0871, "step": 1244 }, { "distill_loss": 0.1824, "epoch": 0.77, "grad_norm": 0.9632885506367962, "learning_rate": 1.35160417102985e-06, "loss": 0.1636, "lr_vision_tower": 2.7032083420596994e-07, "sft_loss": 0.0724, "step": 1245 }, { "distill_loss": 0.1711, "epoch": 0.77, "grad_norm": 0.9860987628558949, "learning_rate": 1.3447831346715373e-06, "loss": 0.178, "lr_vision_tower": 2.6895662693430745e-07, "sft_loss": 0.0925, "step": 1246 }, { "distill_loss": 0.1778, "epoch": 0.77, "grad_norm": 0.9593564212707714, "learning_rate": 1.3379766782387176e-06, "loss": 0.1691, "lr_vision_tower": 2.675953356477435e-07, "sft_loss": 0.0803, "step": 1247 }, { "distill_loss": 0.1752, "epoch": 0.77, "grad_norm": 0.6429870546406248, "learning_rate": 1.3311848288809815e-06, "loss": 0.166, "lr_vision_tower": 2.6623696577619625e-07, "sft_loss": 0.0784, "step": 1248 }, { "distill_loss": 0.1739, "epoch": 0.77, "grad_norm": 0.751831252075415, "learning_rate": 1.3244076136896472e-06, "loss": 0.1635, "lr_vision_tower": 2.6488152273792943e-07, "sft_loss": 0.0766, "step": 1249 }, { "distill_loss": 0.1732, "epoch": 0.77, "grad_norm": 0.8052346865743177, "learning_rate": 1.3176450596976659e-06, "loss": 0.1711, "lr_vision_tower": 2.635290119395331e-07, "sft_loss": 0.0845, "step": 1250 }, { "distill_loss": 0.1769, "epoch": 0.77, "grad_norm": 0.7174293780331271, "learning_rate": 1.3108971938795068e-06, "loss": 0.1682, "lr_vision_tower": 2.621794387759013e-07, "sft_loss": 0.0798, "step": 1251 }, { "distill_loss": 0.1706, "epoch": 0.77, "grad_norm": 0.8538703770194563, "learning_rate": 1.3041640431510506e-06, "loss": 0.1644, "lr_vision_tower": 2.6083280863021004e-07, "sft_loss": 0.0791, "step": 1252 }, { "distill_loss": 0.171, "epoch": 0.77, "grad_norm": 0.7382674098326217, "learning_rate": 1.297445634369483e-06, "loss": 0.1683, "lr_vision_tower": 2.5948912687389655e-07, "sft_loss": 0.0828, "step": 1253 }, { "distill_loss": 0.1737, "epoch": 0.77, "grad_norm": 1.2022564837522582, "learning_rate": 1.2907419943331845e-06, "loss": 0.1705, "lr_vision_tower": 2.5814839886663687e-07, "sft_loss": 0.0837, "step": 1254 }, { "distill_loss": 0.1734, "epoch": 0.77, "grad_norm": 0.8059263671162958, "learning_rate": 1.28405314978163e-06, "loss": 0.1656, "lr_vision_tower": 2.5681062995632597e-07, "sft_loss": 0.0789, "step": 1255 }, { "distill_loss": 0.1706, "epoch": 0.77, "grad_norm": 0.8166653887542358, "learning_rate": 1.277379127395274e-06, "loss": 0.1787, "lr_vision_tower": 2.554758254790548e-07, "sft_loss": 0.0934, "step": 1256 }, { "distill_loss": 0.1786, "epoch": 0.77, "grad_norm": 0.9052563304878049, "learning_rate": 1.2707199537954502e-06, "loss": 0.1835, "lr_vision_tower": 2.5414399075909e-07, "sft_loss": 0.0942, "step": 1257 }, { "distill_loss": 0.1732, "epoch": 0.78, "grad_norm": 0.9092315827935099, "learning_rate": 1.2640756555442684e-06, "loss": 0.1808, "lr_vision_tower": 2.5281513110885366e-07, "sft_loss": 0.0941, "step": 1258 }, { "distill_loss": 0.1756, "epoch": 0.78, "grad_norm": 0.7507680594520375, "learning_rate": 1.257446259144494e-06, "loss": 0.1651, "lr_vision_tower": 2.514892518288988e-07, "sft_loss": 0.0773, "step": 1259 }, { "distill_loss": 0.1803, "epoch": 0.78, "grad_norm": 0.7064268616131062, "learning_rate": 1.2508317910394613e-06, "loss": 0.1536, "lr_vision_tower": 2.501663582078922e-07, "sft_loss": 0.0634, "step": 1260 }, { "distill_loss": 0.1803, "epoch": 0.78, "grad_norm": 0.8369954848070742, "learning_rate": 1.2442322776129494e-06, "loss": 0.1935, "lr_vision_tower": 2.4884645552258985e-07, "sft_loss": 0.1034, "step": 1261 }, { "distill_loss": 0.1775, "epoch": 0.78, "grad_norm": 1.0387520785533608, "learning_rate": 1.2376477451890945e-06, "loss": 0.1807, "lr_vision_tower": 2.4752954903781886e-07, "sft_loss": 0.0919, "step": 1262 }, { "distill_loss": 0.179, "epoch": 0.78, "grad_norm": 0.8500744399398181, "learning_rate": 1.2310782200322719e-06, "loss": 0.1761, "lr_vision_tower": 2.4621564400645436e-07, "sft_loss": 0.0866, "step": 1263 }, { "distill_loss": 0.1693, "epoch": 0.78, "grad_norm": 0.7745772920727109, "learning_rate": 1.2245237283469985e-06, "loss": 0.1614, "lr_vision_tower": 2.449047456693997e-07, "sft_loss": 0.0768, "step": 1264 }, { "distill_loss": 0.1832, "epoch": 0.78, "grad_norm": 1.0308603760789798, "learning_rate": 1.2179842962778226e-06, "loss": 0.1649, "lr_vision_tower": 2.435968592555645e-07, "sft_loss": 0.0733, "step": 1265 }, { "distill_loss": 0.1771, "epoch": 0.78, "grad_norm": 0.8926703291018226, "learning_rate": 1.211459949909225e-06, "loss": 0.1904, "lr_vision_tower": 2.4229198998184496e-07, "sft_loss": 0.1018, "step": 1266 }, { "distill_loss": 0.1756, "epoch": 0.78, "grad_norm": 0.7613728469531313, "learning_rate": 1.2049507152655139e-06, "loss": 0.1656, "lr_vision_tower": 2.409901430531027e-07, "sft_loss": 0.0778, "step": 1267 }, { "distill_loss": 0.1793, "epoch": 0.78, "grad_norm": 0.9308072176065889, "learning_rate": 1.198456618310715e-06, "loss": 0.1764, "lr_vision_tower": 2.3969132366214295e-07, "sft_loss": 0.0867, "step": 1268 }, { "distill_loss": 0.1817, "epoch": 0.78, "grad_norm": 0.7894860602940512, "learning_rate": 1.1919776849484782e-06, "loss": 0.1733, "lr_vision_tower": 2.383955369896956e-07, "sft_loss": 0.0825, "step": 1269 }, { "distill_loss": 0.1677, "epoch": 0.78, "grad_norm": 1.3093943193106028, "learning_rate": 1.1855139410219657e-06, "loss": 0.167, "lr_vision_tower": 2.371027882043931e-07, "sft_loss": 0.0832, "step": 1270 }, { "distill_loss": 0.1787, "epoch": 0.78, "grad_norm": 0.7921306020291831, "learning_rate": 1.1790654123137552e-06, "loss": 0.1914, "lr_vision_tower": 2.35813082462751e-07, "sft_loss": 0.1021, "step": 1271 }, { "distill_loss": 0.1685, "epoch": 0.78, "grad_norm": 0.8932403188019405, "learning_rate": 1.1726321245457284e-06, "loss": 0.1608, "lr_vision_tower": 2.3452642490914566e-07, "sft_loss": 0.0765, "step": 1272 }, { "distill_loss": 0.1818, "epoch": 0.78, "grad_norm": 0.9552149297952655, "learning_rate": 1.1662141033789803e-06, "loss": 0.1657, "lr_vision_tower": 2.3324282067579603e-07, "sft_loss": 0.0748, "step": 1273 }, { "distill_loss": 0.1706, "epoch": 0.79, "grad_norm": 0.7947702155690823, "learning_rate": 1.1598113744137045e-06, "loss": 0.1661, "lr_vision_tower": 2.319622748827409e-07, "sft_loss": 0.0808, "step": 1274 }, { "distill_loss": 0.1801, "epoch": 0.79, "grad_norm": 0.693567741670085, "learning_rate": 1.1534239631891008e-06, "loss": 0.1697, "lr_vision_tower": 2.3068479263782014e-07, "sft_loss": 0.0796, "step": 1275 }, { "distill_loss": 0.1747, "epoch": 0.79, "grad_norm": 1.0250472429432225, "learning_rate": 1.1470518951832722e-06, "loss": 0.1738, "lr_vision_tower": 2.2941037903665438e-07, "sft_loss": 0.0864, "step": 1276 }, { "distill_loss": 0.1802, "epoch": 0.79, "grad_norm": 0.9007234353193516, "learning_rate": 1.1406951958131141e-06, "loss": 0.1758, "lr_vision_tower": 2.281390391626228e-07, "sft_loss": 0.0857, "step": 1277 }, { "distill_loss": 0.1775, "epoch": 0.79, "grad_norm": 0.7715025504496185, "learning_rate": 1.134353890434225e-06, "loss": 0.1664, "lr_vision_tower": 2.2687077808684497e-07, "sft_loss": 0.0777, "step": 1278 }, { "distill_loss": 0.1812, "epoch": 0.79, "grad_norm": 0.8413276597993495, "learning_rate": 1.1280280043407948e-06, "loss": 0.1804, "lr_vision_tower": 2.2560560086815895e-07, "sft_loss": 0.0898, "step": 1279 }, { "distill_loss": 0.1642, "epoch": 0.79, "grad_norm": 0.8650270958789995, "learning_rate": 1.1217175627655136e-06, "loss": 0.1692, "lr_vision_tower": 2.2434351255310268e-07, "sft_loss": 0.0871, "step": 1280 }, { "distill_loss": 0.1707, "epoch": 0.79, "grad_norm": 1.0169382586308402, "learning_rate": 1.115422590879464e-06, "loss": 0.1817, "lr_vision_tower": 2.230845181758928e-07, "sft_loss": 0.0964, "step": 1281 }, { "distill_loss": 0.1688, "epoch": 0.79, "grad_norm": 0.7993001735244167, "learning_rate": 1.109143113792026e-06, "loss": 0.1597, "lr_vision_tower": 2.218286227584052e-07, "sft_loss": 0.0753, "step": 1282 }, { "distill_loss": 0.1707, "epoch": 0.79, "grad_norm": 1.062295020635065, "learning_rate": 1.1028791565507684e-06, "loss": 0.1734, "lr_vision_tower": 2.2057583131015366e-07, "sft_loss": 0.0881, "step": 1283 }, { "distill_loss": 0.1886, "epoch": 0.79, "grad_norm": 0.9200920760881206, "learning_rate": 1.0966307441413598e-06, "loss": 0.1834, "lr_vision_tower": 2.1932614882827193e-07, "sft_loss": 0.0891, "step": 1284 }, { "distill_loss": 0.1791, "epoch": 0.79, "grad_norm": 0.8394585355041052, "learning_rate": 1.0903979014874623e-06, "loss": 0.1655, "lr_vision_tower": 2.1807958029749241e-07, "sft_loss": 0.0759, "step": 1285 }, { "distill_loss": 0.1779, "epoch": 0.79, "grad_norm": 0.7498784018561427, "learning_rate": 1.08418065345063e-06, "loss": 0.1755, "lr_vision_tower": 2.1683613069012596e-07, "sft_loss": 0.0865, "step": 1286 }, { "distill_loss": 0.171, "epoch": 0.79, "grad_norm": 0.9762417163207412, "learning_rate": 1.0779790248302168e-06, "loss": 0.1673, "lr_vision_tower": 2.1559580496604335e-07, "sft_loss": 0.0818, "step": 1287 }, { "distill_loss": 0.1702, "epoch": 0.79, "grad_norm": 0.6930312425968983, "learning_rate": 1.0717930403632742e-06, "loss": 0.1582, "lr_vision_tower": 2.1435860807265483e-07, "sft_loss": 0.0731, "step": 1288 }, { "distill_loss": 0.1735, "epoch": 0.79, "grad_norm": 0.8102130786765698, "learning_rate": 1.0656227247244517e-06, "loss": 0.1793, "lr_vision_tower": 2.131245449448903e-07, "sft_loss": 0.0925, "step": 1289 }, { "distill_loss": 0.1892, "epoch": 0.8, "grad_norm": 0.9515588680407151, "learning_rate": 1.0594681025258951e-06, "loss": 0.1677, "lr_vision_tower": 2.11893620505179e-07, "sft_loss": 0.0731, "step": 1290 }, { "distill_loss": 0.1763, "epoch": 0.8, "grad_norm": 0.7236713587949949, "learning_rate": 1.053329198317159e-06, "loss": 0.1821, "lr_vision_tower": 2.1066583966343177e-07, "sft_loss": 0.094, "step": 1291 }, { "distill_loss": 0.1802, "epoch": 0.8, "grad_norm": 0.8069402451016209, "learning_rate": 1.047206036585095e-06, "loss": 0.1833, "lr_vision_tower": 2.0944120731701898e-07, "sft_loss": 0.0932, "step": 1292 }, { "distill_loss": 0.1643, "epoch": 0.8, "grad_norm": 1.0216741338127908, "learning_rate": 1.0410986417537654e-06, "loss": 0.1705, "lr_vision_tower": 2.0821972835075307e-07, "sft_loss": 0.0884, "step": 1293 }, { "distill_loss": 0.1844, "epoch": 0.8, "grad_norm": 0.814303675907931, "learning_rate": 1.0350070381843452e-06, "loss": 0.1776, "lr_vision_tower": 2.0700140763686902e-07, "sft_loss": 0.0855, "step": 1294 }, { "distill_loss": 0.1656, "epoch": 0.8, "grad_norm": 0.945210392772762, "learning_rate": 1.0289312501750121e-06, "loss": 0.1629, "lr_vision_tower": 2.0578625003500239e-07, "sft_loss": 0.0801, "step": 1295 }, { "distill_loss": 0.1832, "epoch": 0.8, "grad_norm": 1.0593325367711133, "learning_rate": 1.022871301960866e-06, "loss": 0.1675, "lr_vision_tower": 2.045742603921732e-07, "sft_loss": 0.0759, "step": 1296 }, { "distill_loss": 0.179, "epoch": 0.8, "grad_norm": 0.9344369297538547, "learning_rate": 1.01682721771382e-06, "loss": 0.187, "lr_vision_tower": 2.0336544354276398e-07, "sft_loss": 0.0975, "step": 1297 }, { "distill_loss": 0.1777, "epoch": 0.8, "grad_norm": 0.7303855296623395, "learning_rate": 1.0107990215425135e-06, "loss": 0.1755, "lr_vision_tower": 2.0215980430850267e-07, "sft_loss": 0.0866, "step": 1298 }, { "distill_loss": 0.1733, "epoch": 0.8, "grad_norm": 0.7966224717138222, "learning_rate": 1.0047867374922083e-06, "loss": 0.1712, "lr_vision_tower": 2.0095734749844163e-07, "sft_loss": 0.0846, "step": 1299 }, { "distill_loss": 0.174, "epoch": 0.8, "grad_norm": 0.8226634825852562, "learning_rate": 9.987903895446994e-07, "loss": 0.1706, "lr_vision_tower": 1.9975807790893983e-07, "sft_loss": 0.0836, "step": 1300 }, { "distill_loss": 0.1709, "epoch": 0.8, "grad_norm": 1.0992901528932646, "learning_rate": 9.928100016182108e-07, "loss": 0.1664, "lr_vision_tower": 1.9856200032364213e-07, "sft_loss": 0.0809, "step": 1301 }, { "distill_loss": 0.178, "epoch": 0.8, "grad_norm": 0.7758633061228071, "learning_rate": 9.868455975673091e-07, "loss": 0.1571, "lr_vision_tower": 1.973691195134618e-07, "sft_loss": 0.0681, "step": 1302 }, { "distill_loss": 0.167, "epoch": 0.8, "grad_norm": 1.3033442902597143, "learning_rate": 9.808972011828055e-07, "loss": 0.1884, "lr_vision_tower": 1.961794402365611e-07, "sft_loss": 0.1049, "step": 1303 }, { "distill_loss": 0.1805, "epoch": 0.8, "grad_norm": 0.7444803074018599, "learning_rate": 9.749648361916542e-07, "loss": 0.1731, "lr_vision_tower": 1.9499296723833082e-07, "sft_loss": 0.0828, "step": 1304 }, { "distill_loss": 0.1671, "epoch": 0.8, "grad_norm": 0.7354783906561911, "learning_rate": 9.690485262568689e-07, "loss": 0.1652, "lr_vision_tower": 1.9380970525137376e-07, "sft_loss": 0.0816, "step": 1305 }, { "distill_loss": 0.178, "epoch": 0.8, "grad_norm": 0.9605030169444567, "learning_rate": 9.631482949774207e-07, "loss": 0.1876, "lr_vision_tower": 1.926296589954841e-07, "sft_loss": 0.0986, "step": 1306 }, { "distill_loss": 0.1716, "epoch": 0.81, "grad_norm": 1.11701259295177, "learning_rate": 9.572641658881482e-07, "loss": 0.1755, "lr_vision_tower": 1.914528331776296e-07, "sft_loss": 0.0897, "step": 1307 }, { "distill_loss": 0.166, "epoch": 0.81, "grad_norm": 0.6473674263523891, "learning_rate": 9.513961624596563e-07, "loss": 0.1641, "lr_vision_tower": 1.9027923249193122e-07, "sft_loss": 0.0811, "step": 1308 }, { "distill_loss": 0.1751, "epoch": 0.81, "grad_norm": 0.9546699474735479, "learning_rate": 9.455443080982346e-07, "loss": 0.1815, "lr_vision_tower": 1.891088616196469e-07, "sft_loss": 0.0939, "step": 1309 }, { "distill_loss": 0.1652, "epoch": 0.81, "grad_norm": 0.7119311177177755, "learning_rate": 9.397086261457511e-07, "loss": 0.1556, "lr_vision_tower": 1.879417252291502e-07, "sft_loss": 0.073, "step": 1310 }, { "distill_loss": 0.1837, "epoch": 0.81, "grad_norm": 0.8472019987908078, "learning_rate": 9.338891398795685e-07, "loss": 0.1819, "lr_vision_tower": 1.8677782797591368e-07, "sft_loss": 0.09, "step": 1311 }, { "distill_loss": 0.18, "epoch": 0.81, "grad_norm": 1.160686551761986, "learning_rate": 9.280858725124526e-07, "loss": 0.1737, "lr_vision_tower": 1.8561717450249047e-07, "sft_loss": 0.0837, "step": 1312 }, { "distill_loss": 0.1655, "epoch": 0.81, "grad_norm": 0.8657749747153409, "learning_rate": 9.22298847192466e-07, "loss": 0.1745, "lr_vision_tower": 1.8445976943849316e-07, "sft_loss": 0.0918, "step": 1313 }, { "distill_loss": 0.1733, "epoch": 0.81, "grad_norm": 1.3019747387559293, "learning_rate": 9.165280870028919e-07, "loss": 0.164, "lr_vision_tower": 1.8330561740057837e-07, "sft_loss": 0.0774, "step": 1314 }, { "distill_loss": 0.1763, "epoch": 0.81, "grad_norm": 0.9833891471635751, "learning_rate": 9.107736149621315e-07, "loss": 0.1758, "lr_vision_tower": 1.821547229924263e-07, "sft_loss": 0.0877, "step": 1315 }, { "distill_loss": 0.1701, "epoch": 0.81, "grad_norm": 0.8958111738187805, "learning_rate": 9.050354540236178e-07, "loss": 0.1685, "lr_vision_tower": 1.8100709080472353e-07, "sft_loss": 0.0835, "step": 1316 }, { "distill_loss": 0.1817, "epoch": 0.81, "grad_norm": 0.7924245144794576, "learning_rate": 8.993136270757219e-07, "loss": 0.1738, "lr_vision_tower": 1.7986272541514436e-07, "sft_loss": 0.083, "step": 1317 }, { "distill_loss": 0.1651, "epoch": 0.81, "grad_norm": 0.9471360054157333, "learning_rate": 8.93608156941661e-07, "loss": 0.1782, "lr_vision_tower": 1.787216313883322e-07, "sft_loss": 0.0957, "step": 1318 }, { "distill_loss": 0.1795, "epoch": 0.81, "grad_norm": 0.6813044001546237, "learning_rate": 8.879190663794113e-07, "loss": 0.1719, "lr_vision_tower": 1.7758381327588222e-07, "sft_loss": 0.0822, "step": 1319 }, { "distill_loss": 0.1726, "epoch": 0.81, "grad_norm": 1.295106972600921, "learning_rate": 8.822463780816082e-07, "loss": 0.1685, "lr_vision_tower": 1.7644927561632162e-07, "sft_loss": 0.0822, "step": 1320 }, { "distill_loss": 0.1705, "epoch": 0.81, "grad_norm": 0.822491296718889, "learning_rate": 8.765901146754679e-07, "loss": 0.1586, "lr_vision_tower": 1.7531802293509357e-07, "sft_loss": 0.0734, "step": 1321 }, { "distill_loss": 0.1725, "epoch": 0.81, "grad_norm": 0.6370077348411392, "learning_rate": 8.709502987226864e-07, "loss": 0.167, "lr_vision_tower": 1.7419005974453728e-07, "sft_loss": 0.0807, "step": 1322 }, { "distill_loss": 0.1708, "epoch": 0.82, "grad_norm": 0.7191095989320934, "learning_rate": 8.65326952719357e-07, "loss": 0.1694, "lr_vision_tower": 1.7306539054387137e-07, "sft_loss": 0.084, "step": 1323 }, { "distill_loss": 0.1731, "epoch": 0.82, "grad_norm": 0.8194313446894768, "learning_rate": 8.597200990958765e-07, "loss": 0.1631, "lr_vision_tower": 1.719440198191753e-07, "sft_loss": 0.0766, "step": 1324 }, { "distill_loss": 0.172, "epoch": 0.82, "grad_norm": 0.7603896582718259, "learning_rate": 8.541297602168591e-07, "loss": 0.177, "lr_vision_tower": 1.708259520433718e-07, "sft_loss": 0.091, "step": 1325 }, { "distill_loss": 0.1726, "epoch": 0.82, "grad_norm": 0.8081186088300412, "learning_rate": 8.485559583810404e-07, "loss": 0.1609, "lr_vision_tower": 1.6971119167620806e-07, "sft_loss": 0.0746, "step": 1326 }, { "distill_loss": 0.1762, "epoch": 0.82, "grad_norm": 0.91164204347661, "learning_rate": 8.429987158211972e-07, "loss": 0.1783, "lr_vision_tower": 1.685997431642394e-07, "sft_loss": 0.0902, "step": 1327 }, { "distill_loss": 0.1769, "epoch": 0.82, "grad_norm": 0.7432610437498185, "learning_rate": 8.374580547040539e-07, "loss": 0.1638, "lr_vision_tower": 1.6749161094081076e-07, "sft_loss": 0.0754, "step": 1328 }, { "distill_loss": 0.1731, "epoch": 0.82, "grad_norm": 0.9478810257628658, "learning_rate": 8.31933997130192e-07, "loss": 0.1706, "lr_vision_tower": 1.663867994260384e-07, "sft_loss": 0.084, "step": 1329 }, { "distill_loss": 0.1658, "epoch": 0.82, "grad_norm": 0.906121632067075, "learning_rate": 8.264265651339687e-07, "loss": 0.1522, "lr_vision_tower": 1.6528531302679373e-07, "sft_loss": 0.0693, "step": 1330 }, { "distill_loss": 0.1738, "epoch": 0.82, "grad_norm": 0.8007406102019475, "learning_rate": 8.209357806834229e-07, "loss": 0.1693, "lr_vision_tower": 1.6418715613668454e-07, "sft_loss": 0.0824, "step": 1331 }, { "distill_loss": 0.1799, "epoch": 0.82, "grad_norm": 1.5036923700738953, "learning_rate": 8.154616656801917e-07, "loss": 0.1799, "lr_vision_tower": 1.6309233313603831e-07, "sft_loss": 0.09, "step": 1332 }, { "distill_loss": 0.1811, "epoch": 0.82, "grad_norm": 0.9079339347117454, "learning_rate": 8.100042419594168e-07, "loss": 0.1698, "lr_vision_tower": 1.6200084839188331e-07, "sft_loss": 0.0792, "step": 1333 }, { "distill_loss": 0.1696, "epoch": 0.82, "grad_norm": 0.6754792891467513, "learning_rate": 8.045635312896676e-07, "loss": 0.1568, "lr_vision_tower": 1.6091270625793352e-07, "sft_loss": 0.072, "step": 1334 }, { "distill_loss": 0.1726, "epoch": 0.82, "grad_norm": 0.9154587012351705, "learning_rate": 7.99139555372842e-07, "loss": 0.1806, "lr_vision_tower": 1.598279110745684e-07, "sft_loss": 0.0943, "step": 1335 }, { "distill_loss": 0.186, "epoch": 0.82, "grad_norm": 0.8363859199357302, "learning_rate": 7.937323358440935e-07, "loss": 0.1807, "lr_vision_tower": 1.5874646716881868e-07, "sft_loss": 0.0877, "step": 1336 }, { "distill_loss": 0.169, "epoch": 0.82, "grad_norm": 1.2869807479531998, "learning_rate": 7.883418942717342e-07, "loss": 0.1709, "lr_vision_tower": 1.576683788543468e-07, "sft_loss": 0.0864, "step": 1337 }, { "distill_loss": 0.1749, "epoch": 0.82, "grad_norm": 0.9014021705033801, "learning_rate": 7.829682521571492e-07, "loss": 0.1827, "lr_vision_tower": 1.565936504314298e-07, "sft_loss": 0.0953, "step": 1338 }, { "distill_loss": 0.1735, "epoch": 0.83, "grad_norm": 0.7799278678559947, "learning_rate": 7.7761143093472e-07, "loss": 0.1708, "lr_vision_tower": 1.5552228618694397e-07, "sft_loss": 0.084, "step": 1339 }, { "distill_loss": 0.172, "epoch": 0.83, "grad_norm": 0.7549142934814301, "learning_rate": 7.722714519717267e-07, "loss": 0.1709, "lr_vision_tower": 1.5445429039434533e-07, "sft_loss": 0.0848, "step": 1340 }, { "distill_loss": 0.1718, "epoch": 0.83, "grad_norm": 1.2081352552008546, "learning_rate": 7.669483365682745e-07, "loss": 0.1615, "lr_vision_tower": 1.5338966731365488e-07, "sft_loss": 0.0756, "step": 1341 }, { "distill_loss": 0.1711, "epoch": 0.83, "grad_norm": 0.714401008041893, "learning_rate": 7.616421059572005e-07, "loss": 0.1633, "lr_vision_tower": 1.5232842119144006e-07, "sft_loss": 0.0778, "step": 1342 }, { "distill_loss": 0.1809, "epoch": 0.83, "grad_norm": 1.0879989653833053, "learning_rate": 7.563527813039928e-07, "loss": 0.1904, "lr_vision_tower": 1.5127055626079854e-07, "sft_loss": 0.1, "step": 1343 }, { "distill_loss": 0.1857, "epoch": 0.83, "grad_norm": 0.9730512677585926, "learning_rate": 7.510803837067032e-07, "loss": 0.1852, "lr_vision_tower": 1.5021607674134064e-07, "sft_loss": 0.0924, "step": 1344 }, { "distill_loss": 0.1717, "epoch": 0.83, "grad_norm": 1.0661866145087124, "learning_rate": 7.458249341958668e-07, "loss": 0.1667, "lr_vision_tower": 1.4916498683917333e-07, "sft_loss": 0.0808, "step": 1345 }, { "distill_loss": 0.1671, "epoch": 0.83, "grad_norm": 0.8270017922015309, "learning_rate": 7.405864537344165e-07, "loss": 0.1636, "lr_vision_tower": 1.481172907468833e-07, "sft_loss": 0.0801, "step": 1346 }, { "distill_loss": 0.1795, "epoch": 0.83, "grad_norm": 0.8295871701264163, "learning_rate": 7.353649632175957e-07, "loss": 0.1785, "lr_vision_tower": 1.4707299264351913e-07, "sft_loss": 0.0888, "step": 1347 }, { "distill_loss": 0.1774, "epoch": 0.83, "grad_norm": 0.7095527487954068, "learning_rate": 7.301604834728826e-07, "loss": 0.1584, "lr_vision_tower": 1.460320966945765e-07, "sft_loss": 0.0697, "step": 1348 }, { "distill_loss": 0.1712, "epoch": 0.83, "grad_norm": 0.8948078701825238, "learning_rate": 7.249730352599e-07, "loss": 0.1564, "lr_vision_tower": 1.4499460705197996e-07, "sft_loss": 0.0708, "step": 1349 }, { "distill_loss": 0.1807, "epoch": 0.83, "grad_norm": 0.8101597869705197, "learning_rate": 7.198026392703383e-07, "loss": 0.1624, "lr_vision_tower": 1.4396052785406764e-07, "sft_loss": 0.072, "step": 1350 }, { "distill_loss": 0.1679, "epoch": 0.83, "grad_norm": 0.8094841853223703, "learning_rate": 7.14649316127865e-07, "loss": 0.155, "lr_vision_tower": 1.4292986322557298e-07, "sft_loss": 0.071, "step": 1351 }, { "distill_loss": 0.1743, "epoch": 0.83, "grad_norm": 0.8679705384256563, "learning_rate": 7.095130863880534e-07, "loss": 0.1727, "lr_vision_tower": 1.4190261727761066e-07, "sft_loss": 0.0855, "step": 1352 }, { "distill_loss": 0.1775, "epoch": 0.83, "grad_norm": 0.9837659153852052, "learning_rate": 7.04393970538288e-07, "loss": 0.1729, "lr_vision_tower": 1.4087879410765756e-07, "sft_loss": 0.0841, "step": 1353 }, { "distill_loss": 0.1666, "epoch": 0.83, "grad_norm": 0.690500061827498, "learning_rate": 6.992919889976968e-07, "loss": 0.1604, "lr_vision_tower": 1.3985839779953935e-07, "sft_loss": 0.0771, "step": 1354 }, { "distill_loss": 0.1708, "epoch": 0.84, "grad_norm": 1.0298412291620365, "learning_rate": 6.942071621170581e-07, "loss": 0.1866, "lr_vision_tower": 1.3884143242341162e-07, "sft_loss": 0.1012, "step": 1355 }, { "distill_loss": 0.1601, "epoch": 0.84, "grad_norm": 0.9305760837135357, "learning_rate": 6.89139510178723e-07, "loss": 0.1666, "lr_vision_tower": 1.3782790203574458e-07, "sft_loss": 0.0866, "step": 1356 }, { "distill_loss": 0.165, "epoch": 0.84, "grad_norm": 0.6923641047720794, "learning_rate": 6.840890533965383e-07, "loss": 0.1549, "lr_vision_tower": 1.3681781067930764e-07, "sft_loss": 0.0724, "step": 1357 }, { "distill_loss": 0.1778, "epoch": 0.84, "grad_norm": 1.1348997774492258, "learning_rate": 6.790558119157597e-07, "loss": 0.165, "lr_vision_tower": 1.3581116238315194e-07, "sft_loss": 0.0761, "step": 1358 }, { "distill_loss": 0.1806, "epoch": 0.84, "grad_norm": 1.1156435935789628, "learning_rate": 6.74039805812976e-07, "loss": 0.1803, "lr_vision_tower": 1.348079611625952e-07, "sft_loss": 0.09, "step": 1359 }, { "distill_loss": 0.164, "epoch": 0.84, "grad_norm": 0.9155220347248181, "learning_rate": 6.69041055096028e-07, "loss": 0.1518, "lr_vision_tower": 1.3380821101920557e-07, "sft_loss": 0.0698, "step": 1360 }, { "distill_loss": 0.1673, "epoch": 0.84, "grad_norm": 0.9656875806293044, "learning_rate": 6.64059579703929e-07, "loss": 0.1618, "lr_vision_tower": 1.3281191594078578e-07, "sft_loss": 0.0782, "step": 1361 }, { "distill_loss": 0.1764, "epoch": 0.84, "grad_norm": 1.2383073828428914, "learning_rate": 6.590953995067812e-07, "loss": 0.1555, "lr_vision_tower": 1.318190799013562e-07, "sft_loss": 0.0673, "step": 1362 }, { "distill_loss": 0.1812, "epoch": 0.84, "grad_norm": 0.9695729048630427, "learning_rate": 6.541485343057019e-07, "loss": 0.1756, "lr_vision_tower": 1.3082970686114037e-07, "sft_loss": 0.085, "step": 1363 }, { "distill_loss": 0.1747, "epoch": 0.84, "grad_norm": 0.7927946815550785, "learning_rate": 6.492190038327445e-07, "loss": 0.1796, "lr_vision_tower": 1.2984380076654888e-07, "sft_loss": 0.0923, "step": 1364 }, { "distill_loss": 0.1641, "epoch": 0.84, "grad_norm": 0.8637615226676791, "learning_rate": 6.443068277508124e-07, "loss": 0.1543, "lr_vision_tower": 1.2886136555016246e-07, "sft_loss": 0.0722, "step": 1365 }, { "distill_loss": 0.1735, "epoch": 0.84, "grad_norm": 0.7574570321494416, "learning_rate": 6.394120256535902e-07, "loss": 0.1468, "lr_vision_tower": 1.27882405130718e-07, "sft_loss": 0.0601, "step": 1366 }, { "distill_loss": 0.1724, "epoch": 0.84, "grad_norm": 0.9239811092808531, "learning_rate": 6.345346170654593e-07, "loss": 0.1752, "lr_vision_tower": 1.2690692341309183e-07, "sft_loss": 0.089, "step": 1367 }, { "distill_loss": 0.1701, "epoch": 0.84, "grad_norm": 1.4328310250972083, "learning_rate": 6.296746214414246e-07, "loss": 0.1586, "lr_vision_tower": 1.259349242882849e-07, "sft_loss": 0.0736, "step": 1368 }, { "distill_loss": 0.1703, "epoch": 0.84, "grad_norm": 1.058936218906803, "learning_rate": 6.248320581670281e-07, "loss": 0.1543, "lr_vision_tower": 1.249664116334056e-07, "sft_loss": 0.0692, "step": 1369 }, { "distill_loss": 0.1745, "epoch": 0.84, "grad_norm": 0.994849747569745, "learning_rate": 6.200069465582842e-07, "loss": 0.1509, "lr_vision_tower": 1.2400138931165683e-07, "sft_loss": 0.0636, "step": 1370 }, { "distill_loss": 0.1783, "epoch": 0.85, "grad_norm": 0.7971260988860187, "learning_rate": 6.151993058615924e-07, "loss": 0.1802, "lr_vision_tower": 1.2303986117231847e-07, "sft_loss": 0.091, "step": 1371 }, { "distill_loss": 0.168, "epoch": 0.85, "grad_norm": 0.7450446282226021, "learning_rate": 6.104091552536656e-07, "loss": 0.1569, "lr_vision_tower": 1.220818310507331e-07, "sft_loss": 0.0729, "step": 1372 }, { "distill_loss": 0.1702, "epoch": 0.85, "grad_norm": 1.40599204805214, "learning_rate": 6.056365138414533e-07, "loss": 0.1756, "lr_vision_tower": 1.2112730276829063e-07, "sft_loss": 0.0905, "step": 1373 }, { "distill_loss": 0.1812, "epoch": 0.85, "grad_norm": 0.8074899220620981, "learning_rate": 6.0088140066206e-07, "loss": 0.1737, "lr_vision_tower": 1.20176280132412e-07, "sft_loss": 0.0831, "step": 1374 }, { "distill_loss": 0.1715, "epoch": 0.85, "grad_norm": 1.1220652982278547, "learning_rate": 5.961438346826792e-07, "loss": 0.1625, "lr_vision_tower": 1.1922876693653584e-07, "sft_loss": 0.0768, "step": 1375 }, { "distill_loss": 0.1744, "epoch": 0.85, "grad_norm": 0.7857360415623531, "learning_rate": 5.914238348005075e-07, "loss": 0.1633, "lr_vision_tower": 1.1828476696010148e-07, "sft_loss": 0.0761, "step": 1376 }, { "distill_loss": 0.1708, "epoch": 0.85, "grad_norm": 0.9091581171322487, "learning_rate": 5.867214198426769e-07, "loss": 0.1761, "lr_vision_tower": 1.1734428396853535e-07, "sft_loss": 0.0907, "step": 1377 }, { "distill_loss": 0.1732, "epoch": 0.85, "grad_norm": 0.8843387680044358, "learning_rate": 5.82036608566175e-07, "loss": 0.1751, "lr_vision_tower": 1.1640732171323498e-07, "sft_loss": 0.0885, "step": 1378 }, { "distill_loss": 0.1706, "epoch": 0.85, "grad_norm": 0.7853943593387063, "learning_rate": 5.773694196577734e-07, "loss": 0.1724, "lr_vision_tower": 1.1547388393155466e-07, "sft_loss": 0.0871, "step": 1379 }, { "distill_loss": 0.1643, "epoch": 0.85, "grad_norm": 0.8114155376591913, "learning_rate": 5.727198717339511e-07, "loss": 0.1557, "lr_vision_tower": 1.145439743467902e-07, "sft_loss": 0.0735, "step": 1380 }, { "distill_loss": 0.1782, "epoch": 0.85, "grad_norm": 0.7953524941567047, "learning_rate": 5.680879833408181e-07, "loss": 0.18, "lr_vision_tower": 1.1361759666816362e-07, "sft_loss": 0.0909, "step": 1381 }, { "distill_loss": 0.1778, "epoch": 0.85, "grad_norm": 0.8883556580613897, "learning_rate": 5.63473772954049e-07, "loss": 0.1764, "lr_vision_tower": 1.126947545908098e-07, "sft_loss": 0.0875, "step": 1382 }, { "distill_loss": 0.1771, "epoch": 0.85, "grad_norm": 1.2248687028397547, "learning_rate": 5.588772589788e-07, "loss": 0.1852, "lr_vision_tower": 1.1177545179575998e-07, "sft_loss": 0.0966, "step": 1383 }, { "distill_loss": 0.1696, "epoch": 0.85, "grad_norm": 0.8770838314333936, "learning_rate": 5.542984597496415e-07, "loss": 0.1618, "lr_vision_tower": 1.1085969194992828e-07, "sft_loss": 0.077, "step": 1384 }, { "distill_loss": 0.172, "epoch": 0.85, "grad_norm": 1.2158434004668894, "learning_rate": 5.497373935304839e-07, "loss": 0.1799, "lr_vision_tower": 1.0994747870609677e-07, "sft_loss": 0.0939, "step": 1385 }, { "distill_loss": 0.1794, "epoch": 0.85, "grad_norm": 1.673688182089815, "learning_rate": 5.451940785145049e-07, "loss": 0.1673, "lr_vision_tower": 1.0903881570290097e-07, "sft_loss": 0.0776, "step": 1386 }, { "distill_loss": 0.1709, "epoch": 0.85, "grad_norm": 1.3406745452436937, "learning_rate": 5.406685328240735e-07, "loss": 0.183, "lr_vision_tower": 1.0813370656481468e-07, "sft_loss": 0.0976, "step": 1387 }, { "distill_loss": 0.1669, "epoch": 0.86, "grad_norm": 1.0336674704851483, "learning_rate": 5.361607745106817e-07, "loss": 0.1512, "lr_vision_tower": 1.0723215490213632e-07, "sft_loss": 0.0678, "step": 1388 }, { "distill_loss": 0.1593, "epoch": 0.86, "grad_norm": 1.2025500501505637, "learning_rate": 5.316708215548727e-07, "loss": 0.1597, "lr_vision_tower": 1.0633416431097453e-07, "sft_loss": 0.0801, "step": 1389 }, { "distill_loss": 0.1682, "epoch": 0.86, "grad_norm": 0.7579639910580888, "learning_rate": 5.271986918661659e-07, "loss": 0.1777, "lr_vision_tower": 1.0543973837323317e-07, "sft_loss": 0.0936, "step": 1390 }, { "distill_loss": 0.1804, "epoch": 0.86, "grad_norm": 0.8045120091116746, "learning_rate": 5.227444032829887e-07, "loss": 0.1796, "lr_vision_tower": 1.0454888065659773e-07, "sft_loss": 0.0894, "step": 1391 }, { "distill_loss": 0.1732, "epoch": 0.86, "grad_norm": 1.0515349398309233, "learning_rate": 5.183079735726021e-07, "loss": 0.1856, "lr_vision_tower": 1.036615947145204e-07, "sft_loss": 0.099, "step": 1392 }, { "distill_loss": 0.1777, "epoch": 0.86, "grad_norm": 0.7193441273658059, "learning_rate": 5.138894204310335e-07, "loss": 0.1874, "lr_vision_tower": 1.0277788408620668e-07, "sft_loss": 0.0985, "step": 1393 }, { "distill_loss": 0.1866, "epoch": 0.86, "grad_norm": 0.9030548981095693, "learning_rate": 5.094887614830018e-07, "loss": 0.1889, "lr_vision_tower": 1.0189775229660036e-07, "sft_loss": 0.0956, "step": 1394 }, { "distill_loss": 0.1737, "epoch": 0.86, "grad_norm": 0.9431498568984374, "learning_rate": 5.051060142818521e-07, "loss": 0.1676, "lr_vision_tower": 1.010212028563704e-07, "sft_loss": 0.0808, "step": 1395 }, { "distill_loss": 0.1763, "epoch": 0.86, "grad_norm": 0.8803499179922124, "learning_rate": 5.007411963094822e-07, "loss": 0.1772, "lr_vision_tower": 1.0014823926189642e-07, "sft_loss": 0.0891, "step": 1396 }, { "distill_loss": 0.1684, "epoch": 0.86, "grad_norm": 1.495918116302209, "learning_rate": 4.963943249762743e-07, "loss": 0.1666, "lr_vision_tower": 9.927886499525484e-08, "sft_loss": 0.0824, "step": 1397 }, { "distill_loss": 0.1724, "epoch": 0.86, "grad_norm": 0.8579145394747144, "learning_rate": 4.920654176210249e-07, "loss": 0.1715, "lr_vision_tower": 9.841308352420497e-08, "sft_loss": 0.0853, "step": 1398 }, { "distill_loss": 0.169, "epoch": 0.86, "grad_norm": 0.9972423418172677, "learning_rate": 4.877544915108739e-07, "loss": 0.174, "lr_vision_tower": 9.755089830217478e-08, "sft_loss": 0.0895, "step": 1399 }, { "distill_loss": 0.164, "epoch": 0.86, "grad_norm": 0.6406232517237743, "learning_rate": 4.834615638412404e-07, "loss": 0.1579, "lr_vision_tower": 9.669231276824808e-08, "sft_loss": 0.0759, "step": 1400 }, { "distill_loss": 0.1764, "epoch": 0.86, "grad_norm": 0.7929681984010147, "learning_rate": 4.791866517357491e-07, "loss": 0.1677, "lr_vision_tower": 9.58373303471498e-08, "sft_loss": 0.0794, "step": 1401 }, { "distill_loss": 0.1851, "epoch": 0.86, "grad_norm": 0.9093485643044983, "learning_rate": 4.74929772246166e-07, "loss": 0.1805, "lr_vision_tower": 9.498595444923318e-08, "sft_loss": 0.088, "step": 1402 }, { "distill_loss": 0.1719, "epoch": 0.86, "grad_norm": 0.8623685901343564, "learning_rate": 4.7069094235232814e-07, "loss": 0.1806, "lr_vision_tower": 9.413818847046562e-08, "sft_loss": 0.0946, "step": 1403 }, { "distill_loss": 0.1789, "epoch": 0.87, "grad_norm": 0.7298672605172047, "learning_rate": 4.6647017896207625e-07, "loss": 0.1877, "lr_vision_tower": 9.329403579241524e-08, "sft_loss": 0.0983, "step": 1404 }, { "distill_loss": 0.1689, "epoch": 0.87, "grad_norm": 0.8030855981738159, "learning_rate": 4.6226749891118585e-07, "loss": 0.1632, "lr_vision_tower": 9.245349978223715e-08, "sft_loss": 0.0787, "step": 1405 }, { "distill_loss": 0.1726, "epoch": 0.87, "grad_norm": 0.7704266546069988, "learning_rate": 4.5808291896330403e-07, "loss": 0.1722, "lr_vision_tower": 9.161658379266079e-08, "sft_loss": 0.0859, "step": 1406 }, { "distill_loss": 0.1795, "epoch": 0.87, "grad_norm": 1.1625642076778828, "learning_rate": 4.539164558098802e-07, "loss": 0.18, "lr_vision_tower": 9.078329116197603e-08, "sft_loss": 0.0902, "step": 1407 }, { "distill_loss": 0.1831, "epoch": 0.87, "grad_norm": 0.7348807289246743, "learning_rate": 4.497681260700953e-07, "loss": 0.1721, "lr_vision_tower": 8.995362521401906e-08, "sft_loss": 0.0806, "step": 1408 }, { "distill_loss": 0.1794, "epoch": 0.87, "grad_norm": 0.7892683880899546, "learning_rate": 4.456379462908078e-07, "loss": 0.173, "lr_vision_tower": 8.912758925816155e-08, "sft_loss": 0.0833, "step": 1409 }, { "distill_loss": 0.1809, "epoch": 0.87, "grad_norm": 0.9294072881494293, "learning_rate": 4.415259329464705e-07, "loss": 0.1799, "lr_vision_tower": 8.830518658929409e-08, "sft_loss": 0.0894, "step": 1410 }, { "distill_loss": 0.1642, "epoch": 0.87, "grad_norm": 0.943019964318244, "learning_rate": 4.374321024390815e-07, "loss": 0.1625, "lr_vision_tower": 8.748642048781629e-08, "sft_loss": 0.0804, "step": 1411 }, { "distill_loss": 0.1751, "epoch": 0.87, "grad_norm": 0.7040106148125225, "learning_rate": 4.333564710981053e-07, "loss": 0.1676, "lr_vision_tower": 8.667129421962105e-08, "sft_loss": 0.08, "step": 1412 }, { "distill_loss": 0.1666, "epoch": 0.87, "grad_norm": 1.1069255247271257, "learning_rate": 4.2929905518041714e-07, "loss": 0.1683, "lr_vision_tower": 8.585981103608342e-08, "sft_loss": 0.085, "step": 1413 }, { "distill_loss": 0.1775, "epoch": 0.87, "grad_norm": 0.9312896864526329, "learning_rate": 4.2525987087023433e-07, "loss": 0.1748, "lr_vision_tower": 8.505197417404686e-08, "sft_loss": 0.086, "step": 1414 }, { "distill_loss": 0.1739, "epoch": 0.87, "grad_norm": 0.6680786526599197, "learning_rate": 4.212389342790507e-07, "loss": 0.1665, "lr_vision_tower": 8.424778685581013e-08, "sft_loss": 0.0796, "step": 1415 }, { "distill_loss": 0.1699, "epoch": 0.87, "grad_norm": 1.2545021057681673, "learning_rate": 4.1723626144557547e-07, "loss": 0.1863, "lr_vision_tower": 8.344725228911509e-08, "sft_loss": 0.1014, "step": 1416 }, { "distill_loss": 0.1744, "epoch": 0.87, "grad_norm": 0.7299847343630766, "learning_rate": 4.13251868335664e-07, "loss": 0.1604, "lr_vision_tower": 8.265037366713279e-08, "sft_loss": 0.0732, "step": 1417 }, { "distill_loss": 0.1665, "epoch": 0.87, "grad_norm": 0.7358224692028907, "learning_rate": 4.092857708422621e-07, "loss": 0.1679, "lr_vision_tower": 8.18571541684524e-08, "sft_loss": 0.0847, "step": 1418 }, { "distill_loss": 0.1851, "epoch": 0.87, "grad_norm": 0.8034717223615607, "learning_rate": 4.0533798478533337e-07, "loss": 0.1783, "lr_vision_tower": 8.106759695706667e-08, "sft_loss": 0.0857, "step": 1419 }, { "distill_loss": 0.1705, "epoch": 0.88, "grad_norm": 0.8938854905086088, "learning_rate": 4.014085259118039e-07, "loss": 0.162, "lr_vision_tower": 8.028170518236077e-08, "sft_loss": 0.0768, "step": 1420 }, { "distill_loss": 0.1737, "epoch": 0.88, "grad_norm": 0.8092601330382934, "learning_rate": 3.974974098954959e-07, "loss": 0.171, "lr_vision_tower": 7.949948197909917e-08, "sft_loss": 0.0841, "step": 1421 }, { "distill_loss": 0.182, "epoch": 0.88, "grad_norm": 0.8501021642803014, "learning_rate": 3.9360465233706566e-07, "loss": 0.1809, "lr_vision_tower": 7.872093046741313e-08, "sft_loss": 0.0899, "step": 1422 }, { "distill_loss": 0.1676, "epoch": 0.88, "grad_norm": 0.6923320616763714, "learning_rate": 3.897302687639387e-07, "loss": 0.1618, "lr_vision_tower": 7.794605375278773e-08, "sft_loss": 0.0779, "step": 1423 }, { "distill_loss": 0.1811, "epoch": 0.88, "grad_norm": 0.7402897252107119, "learning_rate": 3.858742746302535e-07, "loss": 0.1664, "lr_vision_tower": 7.717485492605069e-08, "sft_loss": 0.0759, "step": 1424 }, { "distill_loss": 0.169, "epoch": 0.88, "grad_norm": 0.8793899588017997, "learning_rate": 3.8203668531679725e-07, "loss": 0.1699, "lr_vision_tower": 7.640733706335945e-08, "sft_loss": 0.0854, "step": 1425 }, { "distill_loss": 0.1782, "epoch": 0.88, "grad_norm": 0.8405407149250937, "learning_rate": 3.7821751613093927e-07, "loss": 0.17, "lr_vision_tower": 7.564350322618785e-08, "sft_loss": 0.0809, "step": 1426 }, { "distill_loss": 0.1717, "epoch": 0.88, "grad_norm": 0.8249235585569105, "learning_rate": 3.744167823065814e-07, "loss": 0.1696, "lr_vision_tower": 7.488335646131628e-08, "sft_loss": 0.0837, "step": 1427 }, { "distill_loss": 0.1707, "epoch": 0.88, "grad_norm": 0.9721120699892216, "learning_rate": 3.7063449900408444e-07, "loss": 0.1713, "lr_vision_tower": 7.412689980081687e-08, "sft_loss": 0.0859, "step": 1428 }, { "distill_loss": 0.1667, "epoch": 0.88, "grad_norm": 1.032336770112451, "learning_rate": 3.6687068131021855e-07, "loss": 0.1745, "lr_vision_tower": 7.33741362620437e-08, "sft_loss": 0.0912, "step": 1429 }, { "distill_loss": 0.1762, "epoch": 0.88, "grad_norm": 1.0822841882417267, "learning_rate": 3.6312534423809386e-07, "loss": 0.1835, "lr_vision_tower": 7.262506884761876e-08, "sft_loss": 0.0954, "step": 1430 }, { "distill_loss": 0.168, "epoch": 0.88, "grad_norm": 0.8083639072164173, "learning_rate": 3.59398502727108e-07, "loss": 0.1643, "lr_vision_tower": 7.187970054542159e-08, "sft_loss": 0.0803, "step": 1431 }, { "distill_loss": 0.1855, "epoch": 0.88, "grad_norm": 0.8039793513883403, "learning_rate": 3.5569017164288346e-07, "loss": 0.1742, "lr_vision_tower": 7.113803432857668e-08, "sft_loss": 0.0815, "step": 1432 }, { "distill_loss": 0.1708, "epoch": 0.88, "grad_norm": 0.8198809518880921, "learning_rate": 3.5200036577720755e-07, "loss": 0.1673, "lr_vision_tower": 7.04000731554415e-08, "sft_loss": 0.082, "step": 1433 }, { "distill_loss": 0.1676, "epoch": 0.88, "grad_norm": 1.2488204862876222, "learning_rate": 3.483290998479744e-07, "loss": 0.1556, "lr_vision_tower": 6.966581996959486e-08, "sft_loss": 0.0718, "step": 1434 }, { "distill_loss": 0.1678, "epoch": 0.88, "grad_norm": 0.7531247983264878, "learning_rate": 3.44676388499125e-07, "loss": 0.1617, "lr_vision_tower": 6.893527769982499e-08, "sft_loss": 0.0778, "step": 1435 }, { "distill_loss": 0.1708, "epoch": 0.89, "grad_norm": 0.9608708170947449, "learning_rate": 3.4104224630059244e-07, "loss": 0.1693, "lr_vision_tower": 6.820844926011848e-08, "sft_loss": 0.0839, "step": 1436 }, { "distill_loss": 0.177, "epoch": 0.89, "grad_norm": 0.8489890012106226, "learning_rate": 3.3742668774823796e-07, "loss": 0.1681, "lr_vision_tower": 6.748533754964758e-08, "sft_loss": 0.0796, "step": 1437 }, { "distill_loss": 0.1685, "epoch": 0.89, "grad_norm": 1.0337720153742063, "learning_rate": 3.338297272637986e-07, "loss": 0.1741, "lr_vision_tower": 6.67659454527597e-08, "sft_loss": 0.0898, "step": 1438 }, { "distill_loss": 0.1576, "epoch": 0.89, "grad_norm": 0.6627220228941303, "learning_rate": 3.3025137919482696e-07, "loss": 0.1439, "lr_vision_tower": 6.605027583896539e-08, "sft_loss": 0.0651, "step": 1439 }, { "distill_loss": 0.1702, "epoch": 0.89, "grad_norm": 0.9320909282488761, "learning_rate": 3.26691657814634e-07, "loss": 0.1751, "lr_vision_tower": 6.533833156292679e-08, "sft_loss": 0.09, "step": 1440 }, { "distill_loss": 0.1721, "epoch": 0.89, "grad_norm": 0.9293409960599331, "learning_rate": 3.2315057732223485e-07, "loss": 0.1843, "lr_vision_tower": 6.463011546444697e-08, "sft_loss": 0.0983, "step": 1441 }, { "distill_loss": 0.1798, "epoch": 0.89, "grad_norm": 0.6985788618623183, "learning_rate": 3.1962815184228504e-07, "loss": 0.1651, "lr_vision_tower": 6.3925630368457e-08, "sft_loss": 0.0752, "step": 1442 }, { "distill_loss": 0.165, "epoch": 0.89, "grad_norm": 0.9070111328090172, "learning_rate": 3.1612439542503426e-07, "loss": 0.1586, "lr_vision_tower": 6.322487908500685e-08, "sft_loss": 0.0762, "step": 1443 }, { "distill_loss": 0.1584, "epoch": 0.89, "grad_norm": 0.7147351414478743, "learning_rate": 3.1263932204626034e-07, "loss": 0.1458, "lr_vision_tower": 6.252786440925207e-08, "sft_loss": 0.0666, "step": 1444 }, { "distill_loss": 0.1761, "epoch": 0.89, "grad_norm": 0.9272555066424408, "learning_rate": 3.0917294560722346e-07, "loss": 0.1683, "lr_vision_tower": 6.183458912144467e-08, "sft_loss": 0.0802, "step": 1445 }, { "distill_loss": 0.1722, "epoch": 0.89, "grad_norm": 0.9843253582723642, "learning_rate": 3.0572527993460054e-07, "loss": 0.1672, "lr_vision_tower": 6.11450559869201e-08, "sft_loss": 0.081, "step": 1446 }, { "distill_loss": 0.1722, "epoch": 0.89, "grad_norm": 0.8556092519887017, "learning_rate": 3.0229633878043843e-07, "loss": 0.1609, "lr_vision_tower": 6.045926775608767e-08, "sft_loss": 0.0748, "step": 1447 }, { "distill_loss": 0.1688, "epoch": 0.89, "grad_norm": 0.9833015892409109, "learning_rate": 2.9888613582209233e-07, "loss": 0.1667, "lr_vision_tower": 5.977722716441846e-08, "sft_loss": 0.0823, "step": 1448 }, { "distill_loss": 0.1752, "epoch": 0.89, "grad_norm": 0.6751178901938395, "learning_rate": 2.954946846621781e-07, "loss": 0.1658, "lr_vision_tower": 5.909893693243562e-08, "sft_loss": 0.0782, "step": 1449 }, { "distill_loss": 0.1721, "epoch": 0.89, "grad_norm": 0.8947718349125275, "learning_rate": 2.9212199882851235e-07, "loss": 0.1625, "lr_vision_tower": 5.842439976570246e-08, "sft_loss": 0.0764, "step": 1450 }, { "distill_loss": 0.1733, "epoch": 0.89, "grad_norm": 0.7828161322983453, "learning_rate": 2.887680917740615e-07, "loss": 0.1546, "lr_vision_tower": 5.7753618354812294e-08, "sft_loss": 0.068, "step": 1451 }, { "distill_loss": 0.1685, "epoch": 0.89, "grad_norm": 0.6537171879246084, "learning_rate": 2.8543297687688655e-07, "loss": 0.1471, "lr_vision_tower": 5.70865953753773e-08, "sft_loss": 0.0628, "step": 1452 }, { "distill_loss": 0.1866, "epoch": 0.9, "grad_norm": 0.7659902641351055, "learning_rate": 2.821166674400905e-07, "loss": 0.1731, "lr_vision_tower": 5.642333348801809e-08, "sft_loss": 0.0798, "step": 1453 }, { "distill_loss": 0.1766, "epoch": 0.9, "grad_norm": 0.9383742172211664, "learning_rate": 2.788191766917653e-07, "loss": 0.1786, "lr_vision_tower": 5.576383533835305e-08, "sft_loss": 0.0903, "step": 1454 }, { "distill_loss": 0.1775, "epoch": 0.9, "grad_norm": 1.0537503971902191, "learning_rate": 2.75540517784938e-07, "loss": 0.177, "lr_vision_tower": 5.510810355698759e-08, "sft_loss": 0.0883, "step": 1455 }, { "distill_loss": 0.1748, "epoch": 0.9, "grad_norm": 1.301968375882534, "learning_rate": 2.7228070379752113e-07, "loss": 0.1534, "lr_vision_tower": 5.445614075950422e-08, "sft_loss": 0.0659, "step": 1456 }, { "distill_loss": 0.1731, "epoch": 0.9, "grad_norm": 0.8867607946354287, "learning_rate": 2.6903974773225703e-07, "loss": 0.1567, "lr_vision_tower": 5.38079495464514e-08, "sft_loss": 0.0701, "step": 1457 }, { "distill_loss": 0.1706, "epoch": 0.9, "grad_norm": 0.7855216064756755, "learning_rate": 2.6581766251666863e-07, "loss": 0.1538, "lr_vision_tower": 5.316353250333372e-08, "sft_loss": 0.0686, "step": 1458 }, { "distill_loss": 0.1694, "epoch": 0.9, "grad_norm": 0.9850085469544616, "learning_rate": 2.626144610030068e-07, "loss": 0.1722, "lr_vision_tower": 5.2522892200601356e-08, "sft_loss": 0.0875, "step": 1459 }, { "distill_loss": 0.178, "epoch": 0.9, "grad_norm": 1.084062766310588, "learning_rate": 2.5943015596819664e-07, "loss": 0.1969, "lr_vision_tower": 5.188603119363932e-08, "sft_loss": 0.1079, "step": 1460 }, { "distill_loss": 0.1683, "epoch": 0.9, "grad_norm": 0.7017607270027686, "learning_rate": 2.56264760113793e-07, "loss": 0.1482, "lr_vision_tower": 5.125295202275859e-08, "sft_loss": 0.0641, "step": 1461 }, { "distill_loss": 0.1741, "epoch": 0.9, "grad_norm": 1.4693402446096975, "learning_rate": 2.5311828606592204e-07, "loss": 0.1775, "lr_vision_tower": 5.06236572131844e-08, "sft_loss": 0.0905, "step": 1462 }, { "distill_loss": 0.1744, "epoch": 0.9, "grad_norm": 0.8641383771693184, "learning_rate": 2.499907463752388e-07, "loss": 0.1691, "lr_vision_tower": 4.999814927504775e-08, "sft_loss": 0.0818, "step": 1463 }, { "distill_loss": 0.1708, "epoch": 0.9, "grad_norm": 0.8660106261179396, "learning_rate": 2.4688215351686805e-07, "loss": 0.1795, "lr_vision_tower": 4.937643070337361e-08, "sft_loss": 0.0941, "step": 1464 }, { "distill_loss": 0.1741, "epoch": 0.9, "grad_norm": 0.949765694469148, "learning_rate": 2.43792519890364e-07, "loss": 0.1634, "lr_vision_tower": 4.875850397807279e-08, "sft_loss": 0.0763, "step": 1465 }, { "distill_loss": 0.1774, "epoch": 0.9, "grad_norm": 0.7606225179858759, "learning_rate": 2.407218578196524e-07, "loss": 0.1719, "lr_vision_tower": 4.8144371563930475e-08, "sft_loss": 0.0832, "step": 1466 }, { "distill_loss": 0.1743, "epoch": 0.9, "grad_norm": 0.8230072475320834, "learning_rate": 2.3767017955298842e-07, "loss": 0.1648, "lr_vision_tower": 4.753403591059768e-08, "sft_loss": 0.0776, "step": 1467 }, { "distill_loss": 0.1689, "epoch": 0.9, "grad_norm": 1.1441755627604957, "learning_rate": 2.3463749726290287e-07, "loss": 0.1795, "lr_vision_tower": 4.692749945258057e-08, "sft_loss": 0.095, "step": 1468 }, { "distill_loss": 0.172, "epoch": 0.91, "grad_norm": 0.7154860727847241, "learning_rate": 2.3162382304615594e-07, "loss": 0.1577, "lr_vision_tower": 4.6324764609231184e-08, "sft_loss": 0.0717, "step": 1469 }, { "distill_loss": 0.1742, "epoch": 0.91, "grad_norm": 0.9029520720515354, "learning_rate": 2.2862916892368914e-07, "loss": 0.1903, "lr_vision_tower": 4.572583378473782e-08, "sft_loss": 0.1032, "step": 1470 }, { "distill_loss": 0.1807, "epoch": 0.91, "grad_norm": 0.9691045905564092, "learning_rate": 2.2565354684057518e-07, "loss": 0.182, "lr_vision_tower": 4.5130709368115025e-08, "sft_loss": 0.0917, "step": 1471 }, { "distill_loss": 0.1697, "epoch": 0.91, "grad_norm": 0.7775063386853297, "learning_rate": 2.2269696866597356e-07, "loss": 0.1728, "lr_vision_tower": 4.4539393733194706e-08, "sft_loss": 0.0879, "step": 1472 }, { "distill_loss": 0.1827, "epoch": 0.91, "grad_norm": 0.8859683471941491, "learning_rate": 2.1975944619307855e-07, "loss": 0.1612, "lr_vision_tower": 4.39518892386157e-08, "sft_loss": 0.0699, "step": 1473 }, { "distill_loss": 0.1725, "epoch": 0.91, "grad_norm": 0.6889860062869909, "learning_rate": 2.1684099113907786e-07, "loss": 0.1595, "lr_vision_tower": 4.3368198227815565e-08, "sft_loss": 0.0733, "step": 1474 }, { "distill_loss": 0.1689, "epoch": 0.91, "grad_norm": 0.8932795520207958, "learning_rate": 2.1394161514510292e-07, "loss": 0.1678, "lr_vision_tower": 4.278832302902058e-08, "sft_loss": 0.0834, "step": 1475 }, { "distill_loss": 0.1581, "epoch": 0.91, "grad_norm": 0.8011642935795185, "learning_rate": 2.1106132977618099e-07, "loss": 0.1555, "lr_vision_tower": 4.2212265955236195e-08, "sft_loss": 0.0764, "step": 1476 }, { "distill_loss": 0.1746, "epoch": 0.91, "grad_norm": 0.8627766288829848, "learning_rate": 2.0820014652119247e-07, "loss": 0.1641, "lr_vision_tower": 4.1640029304238485e-08, "sft_loss": 0.0767, "step": 1477 }, { "distill_loss": 0.1809, "epoch": 0.91, "grad_norm": 0.9310096678644908, "learning_rate": 2.053580767928215e-07, "loss": 0.1765, "lr_vision_tower": 4.107161535856429e-08, "sft_loss": 0.0861, "step": 1478 }, { "distill_loss": 0.1884, "epoch": 0.91, "grad_norm": 0.8320964941426126, "learning_rate": 2.0253513192751374e-07, "loss": 0.1802, "lr_vision_tower": 4.050702638550274e-08, "sft_loss": 0.0859, "step": 1479 }, { "distill_loss": 0.1739, "epoch": 0.91, "grad_norm": 0.7932305751850586, "learning_rate": 1.997313231854281e-07, "loss": 0.1692, "lr_vision_tower": 3.994626463708561e-08, "sft_loss": 0.0823, "step": 1480 }, { "distill_loss": 0.1797, "epoch": 0.91, "grad_norm": 0.9660200712960083, "learning_rate": 1.9694666175039402e-07, "loss": 0.1664, "lr_vision_tower": 3.93893323500788e-08, "sft_loss": 0.0765, "step": 1481 }, { "distill_loss": 0.1791, "epoch": 0.91, "grad_norm": 0.8142405759350494, "learning_rate": 1.9418115872986754e-07, "loss": 0.1558, "lr_vision_tower": 3.88362317459735e-08, "sft_loss": 0.0663, "step": 1482 }, { "distill_loss": 0.1785, "epoch": 0.91, "grad_norm": 1.673077611489029, "learning_rate": 1.9143482515488365e-07, "loss": 0.158, "lr_vision_tower": 3.8286965030976726e-08, "sft_loss": 0.0688, "step": 1483 }, { "distill_loss": 0.177, "epoch": 0.91, "grad_norm": 0.867263777653892, "learning_rate": 1.887076719800157e-07, "loss": 0.1807, "lr_vision_tower": 3.774153439600314e-08, "sft_loss": 0.0922, "step": 1484 }, { "distill_loss": 0.1676, "epoch": 0.92, "grad_norm": 0.8611759398908038, "learning_rate": 1.8599971008332886e-07, "loss": 0.1518, "lr_vision_tower": 3.719994201666577e-08, "sft_loss": 0.0681, "step": 1485 }, { "distill_loss": 0.1701, "epoch": 0.92, "grad_norm": 0.795568553436875, "learning_rate": 1.833109502663394e-07, "loss": 0.1525, "lr_vision_tower": 3.666219005326787e-08, "sft_loss": 0.0675, "step": 1486 }, { "distill_loss": 0.1664, "epoch": 0.92, "grad_norm": 1.1367422265261238, "learning_rate": 1.806414032539694e-07, "loss": 0.1676, "lr_vision_tower": 3.612828065079388e-08, "sft_loss": 0.0844, "step": 1487 }, { "distill_loss": 0.1681, "epoch": 0.92, "grad_norm": 1.2711250628937867, "learning_rate": 1.7799107969450613e-07, "loss": 0.1683, "lr_vision_tower": 3.559821593890122e-08, "sft_loss": 0.0843, "step": 1488 }, { "distill_loss": 0.1732, "epoch": 0.92, "grad_norm": 0.8661074342310299, "learning_rate": 1.7535999015955707e-07, "loss": 0.1839, "lr_vision_tower": 3.507199803191141e-08, "sft_loss": 0.0973, "step": 1489 }, { "distill_loss": 0.1788, "epoch": 0.92, "grad_norm": 0.6766747070111494, "learning_rate": 1.7274814514400995e-07, "loss": 0.1608, "lr_vision_tower": 3.4549629028801985e-08, "sft_loss": 0.0714, "step": 1490 }, { "distill_loss": 0.172, "epoch": 0.92, "grad_norm": 0.7045367743985056, "learning_rate": 1.701555550659889e-07, "loss": 0.1664, "lr_vision_tower": 3.403111101319778e-08, "sft_loss": 0.0804, "step": 1491 }, { "distill_loss": 0.1743, "epoch": 0.92, "grad_norm": 0.8166807275162989, "learning_rate": 1.6758223026681507e-07, "loss": 0.1779, "lr_vision_tower": 3.351644605336301e-08, "sft_loss": 0.0908, "step": 1492 }, { "distill_loss": 0.1649, "epoch": 0.92, "grad_norm": 0.7568932281036094, "learning_rate": 1.6502818101096496e-07, "loss": 0.1667, "lr_vision_tower": 3.3005636202192986e-08, "sft_loss": 0.0842, "step": 1493 }, { "distill_loss": 0.1791, "epoch": 0.92, "grad_norm": 0.8563367955130249, "learning_rate": 1.6249341748602597e-07, "loss": 0.1725, "lr_vision_tower": 3.249868349720519e-08, "sft_loss": 0.0829, "step": 1494 }, { "distill_loss": 0.174, "epoch": 0.92, "grad_norm": 1.006979002139693, "learning_rate": 1.5997794980266267e-07, "loss": 0.1894, "lr_vision_tower": 3.199558996053253e-08, "sft_loss": 0.1024, "step": 1495 }, { "distill_loss": 0.1739, "epoch": 0.92, "grad_norm": 0.8747578502607084, "learning_rate": 1.5748178799456782e-07, "loss": 0.1803, "lr_vision_tower": 3.1496357598913556e-08, "sft_loss": 0.0934, "step": 1496 }, { "distill_loss": 0.1715, "epoch": 0.92, "grad_norm": 1.1068075518467295, "learning_rate": 1.5500494201843075e-07, "loss": 0.1684, "lr_vision_tower": 3.1000988403686146e-08, "sft_loss": 0.0826, "step": 1497 }, { "distill_loss": 0.1796, "epoch": 0.92, "grad_norm": 0.8297454398862258, "learning_rate": 1.5254742175389137e-07, "loss": 0.157, "lr_vision_tower": 3.050948435077827e-08, "sft_loss": 0.0672, "step": 1498 }, { "distill_loss": 0.1739, "epoch": 0.92, "grad_norm": 0.9866053163486936, "learning_rate": 1.5010923700350454e-07, "loss": 0.1741, "lr_vision_tower": 3.0021847400700903e-08, "sft_loss": 0.0872, "step": 1499 }, { "distill_loss": 0.1686, "epoch": 0.92, "grad_norm": 1.0609504222754313, "learning_rate": 1.4769039749269909e-07, "loss": 0.159, "lr_vision_tower": 2.9538079498539815e-08, "sft_loss": 0.0747, "step": 1500 }, { "distill_loss": 0.172, "epoch": 0.93, "grad_norm": 0.9117839040809247, "learning_rate": 1.4529091286973994e-07, "loss": 0.1684, "lr_vision_tower": 2.9058182573947986e-08, "sft_loss": 0.0824, "step": 1501 }, { "distill_loss": 0.1789, "epoch": 0.93, "grad_norm": 0.7843691714405415, "learning_rate": 1.4291079270568942e-07, "loss": 0.166, "lr_vision_tower": 2.858215854113788e-08, "sft_loss": 0.0766, "step": 1502 }, { "distill_loss": 0.1744, "epoch": 0.93, "grad_norm": 0.7149788744911487, "learning_rate": 1.405500464943671e-07, "loss": 0.1697, "lr_vision_tower": 2.8110009298873416e-08, "sft_loss": 0.0825, "step": 1503 }, { "distill_loss": 0.1694, "epoch": 0.93, "grad_norm": 1.3956610358217896, "learning_rate": 1.3820868365231665e-07, "loss": 0.1568, "lr_vision_tower": 2.764173673046333e-08, "sft_loss": 0.0721, "step": 1504 }, { "distill_loss": 0.166, "epoch": 0.93, "grad_norm": 0.7889512973412871, "learning_rate": 1.358867135187636e-07, "loss": 0.1705, "lr_vision_tower": 2.7177342703752714e-08, "sft_loss": 0.0875, "step": 1505 }, { "distill_loss": 0.1742, "epoch": 0.93, "grad_norm": 0.9735044818843438, "learning_rate": 1.3358414535557973e-07, "loss": 0.1719, "lr_vision_tower": 2.6716829071115942e-08, "sft_loss": 0.0848, "step": 1506 }, { "distill_loss": 0.1758, "epoch": 0.93, "grad_norm": 0.8368422673432014, "learning_rate": 1.3130098834724714e-07, "loss": 0.1705, "lr_vision_tower": 2.6260197669449424e-08, "sft_loss": 0.0826, "step": 1507 }, { "distill_loss": 0.1741, "epoch": 0.93, "grad_norm": 1.0407003463249154, "learning_rate": 1.2903725160081982e-07, "loss": 0.169, "lr_vision_tower": 2.580745032016396e-08, "sft_loss": 0.0819, "step": 1508 }, { "distill_loss": 0.1687, "epoch": 0.93, "grad_norm": 1.1341231851746696, "learning_rate": 1.2679294414588927e-07, "loss": 0.1734, "lr_vision_tower": 2.535858882917785e-08, "sft_loss": 0.0891, "step": 1509 }, { "distill_loss": 0.1732, "epoch": 0.93, "grad_norm": 0.758342975749559, "learning_rate": 1.2456807493454626e-07, "loss": 0.1629, "lr_vision_tower": 2.491361498690925e-08, "sft_loss": 0.0763, "step": 1510 }, { "distill_loss": 0.1743, "epoch": 0.93, "grad_norm": 0.8309549537870278, "learning_rate": 1.223626528413474e-07, "loss": 0.1726, "lr_vision_tower": 2.4472530568269477e-08, "sft_loss": 0.0854, "step": 1511 }, { "distill_loss": 0.1795, "epoch": 0.93, "grad_norm": 0.9046798022188849, "learning_rate": 1.2017668666327752e-07, "loss": 0.1733, "lr_vision_tower": 2.4035337332655502e-08, "sft_loss": 0.0835, "step": 1512 }, { "distill_loss": 0.1717, "epoch": 0.93, "grad_norm": 0.733619248814797, "learning_rate": 1.1801018511971796e-07, "loss": 0.1718, "lr_vision_tower": 2.3602037023943587e-08, "sft_loss": 0.086, "step": 1513 }, { "distill_loss": 0.1713, "epoch": 0.93, "grad_norm": 1.0774258586343475, "learning_rate": 1.1586315685240657e-07, "loss": 0.1772, "lr_vision_tower": 2.317263137048131e-08, "sft_loss": 0.0916, "step": 1514 }, { "distill_loss": 0.1602, "epoch": 0.93, "grad_norm": 0.819744691344583, "learning_rate": 1.1373561042540837e-07, "loss": 0.1639, "lr_vision_tower": 2.274712208508167e-08, "sft_loss": 0.0838, "step": 1515 }, { "distill_loss": 0.1655, "epoch": 0.93, "grad_norm": 0.8733277749802488, "learning_rate": 1.1162755432507777e-07, "loss": 0.1792, "lr_vision_tower": 2.232551086501555e-08, "sft_loss": 0.0965, "step": 1516 }, { "distill_loss": 0.1799, "epoch": 0.94, "grad_norm": 1.0144958960918198, "learning_rate": 1.0953899696002746e-07, "loss": 0.1822, "lr_vision_tower": 2.1907799392005487e-08, "sft_loss": 0.0922, "step": 1517 }, { "distill_loss": 0.1742, "epoch": 0.94, "grad_norm": 0.8269229736182115, "learning_rate": 1.0746994666109234e-07, "loss": 0.1665, "lr_vision_tower": 2.1493989332218465e-08, "sft_loss": 0.0794, "step": 1518 }, { "distill_loss": 0.1742, "epoch": 0.94, "grad_norm": 0.8762297646920719, "learning_rate": 1.0542041168129902e-07, "loss": 0.1668, "lr_vision_tower": 2.10840823362598e-08, "sft_loss": 0.0797, "step": 1519 }, { "distill_loss": 0.1815, "epoch": 0.94, "grad_norm": 0.8409699068079924, "learning_rate": 1.033904001958308e-07, "loss": 0.1662, "lr_vision_tower": 2.0678080039166156e-08, "sft_loss": 0.0755, "step": 1520 }, { "distill_loss": 0.1688, "epoch": 0.94, "grad_norm": 0.8061200132302944, "learning_rate": 1.0137992030199495e-07, "loss": 0.1679, "lr_vision_tower": 2.027598406039899e-08, "sft_loss": 0.0836, "step": 1521 }, { "distill_loss": 0.1697, "epoch": 0.94, "grad_norm": 0.7659953772712051, "learning_rate": 9.938898001919162e-08, "loss": 0.1728, "lr_vision_tower": 1.9877796003838322e-08, "sft_loss": 0.0879, "step": 1522 }, { "distill_loss": 0.181, "epoch": 0.94, "grad_norm": 1.6623665039516913, "learning_rate": 9.741758728888218e-08, "loss": 0.1872, "lr_vision_tower": 1.9483517457776434e-08, "sft_loss": 0.0967, "step": 1523 }, { "distill_loss": 0.176, "epoch": 0.94, "grad_norm": 0.8275227591460802, "learning_rate": 9.546574997455538e-08, "loss": 0.1826, "lr_vision_tower": 1.909314999491107e-08, "sft_loss": 0.0946, "step": 1524 }, { "distill_loss": 0.1762, "epoch": 0.94, "grad_norm": 1.0032221153045742, "learning_rate": 9.353347586169847e-08, "loss": 0.1826, "lr_vision_tower": 1.870669517233969e-08, "sft_loss": 0.0946, "step": 1525 }, { "distill_loss": 0.1621, "epoch": 0.94, "grad_norm": 0.7432499639921016, "learning_rate": 9.162077265776448e-08, "loss": 0.1598, "lr_vision_tower": 1.8324154531552894e-08, "sft_loss": 0.0787, "step": 1526 }, { "distill_loss": 0.1702, "epoch": 0.94, "grad_norm": 1.0598280522751817, "learning_rate": 8.972764799214107e-08, "loss": 0.1465, "lr_vision_tower": 1.794552959842821e-08, "sft_loss": 0.0614, "step": 1527 }, { "distill_loss": 0.1707, "epoch": 0.94, "grad_norm": 0.8558409154231238, "learning_rate": 8.785410941612394e-08, "loss": 0.1765, "lr_vision_tower": 1.7570821883224784e-08, "sft_loss": 0.0912, "step": 1528 }, { "distill_loss": 0.1788, "epoch": 0.94, "grad_norm": 0.7936545457724982, "learning_rate": 8.600016440288017e-08, "loss": 0.1776, "lr_vision_tower": 1.7200032880576033e-08, "sft_loss": 0.0882, "step": 1529 }, { "distill_loss": 0.1621, "epoch": 0.94, "grad_norm": 0.9995575292360704, "learning_rate": 8.416582034742493e-08, "loss": 0.1789, "lr_vision_tower": 1.6833164069484983e-08, "sft_loss": 0.0979, "step": 1530 }, { "distill_loss": 0.1781, "epoch": 0.94, "grad_norm": 0.9455858837454386, "learning_rate": 8.235108456658814e-08, "loss": 0.192, "lr_vision_tower": 1.6470216913317624e-08, "sft_loss": 0.103, "step": 1531 }, { "distill_loss": 0.1707, "epoch": 0.94, "grad_norm": 0.8705439249215429, "learning_rate": 8.055596429898504e-08, "loss": 0.1442, "lr_vision_tower": 1.611119285979701e-08, "sft_loss": 0.0589, "step": 1532 }, { "distill_loss": 0.172, "epoch": 0.94, "grad_norm": 0.7549379984512059, "learning_rate": 7.878046670499128e-08, "loss": 0.1695, "lr_vision_tower": 1.5756093340998255e-08, "sft_loss": 0.0835, "step": 1533 }, { "distill_loss": 0.1649, "epoch": 0.95, "grad_norm": 0.8617713764141343, "learning_rate": 7.702459886670788e-08, "loss": 0.1479, "lr_vision_tower": 1.5404919773341572e-08, "sft_loss": 0.0655, "step": 1534 }, { "distill_loss": 0.1689, "epoch": 0.95, "grad_norm": 0.6991743873963433, "learning_rate": 7.528836778794013e-08, "loss": 0.1517, "lr_vision_tower": 1.5057673557588025e-08, "sft_loss": 0.0672, "step": 1535 }, { "distill_loss": 0.1693, "epoch": 0.95, "grad_norm": 0.7369799696251393, "learning_rate": 7.35717803941649e-08, "loss": 0.1602, "lr_vision_tower": 1.4714356078832979e-08, "sft_loss": 0.0756, "step": 1536 }, { "distill_loss": 0.1743, "epoch": 0.95, "grad_norm": 0.9609432681765941, "learning_rate": 7.187484353250395e-08, "loss": 0.1685, "lr_vision_tower": 1.4374968706500789e-08, "sft_loss": 0.0814, "step": 1537 }, { "distill_loss": 0.1706, "epoch": 0.95, "grad_norm": 0.9622695889199155, "learning_rate": 7.01975639716973e-08, "loss": 0.1699, "lr_vision_tower": 1.4039512794339458e-08, "sft_loss": 0.0846, "step": 1538 }, { "distill_loss": 0.1734, "epoch": 0.95, "grad_norm": 0.887018789453752, "learning_rate": 6.8539948402076e-08, "loss": 0.1743, "lr_vision_tower": 1.37079896804152e-08, "sft_loss": 0.0876, "step": 1539 }, { "distill_loss": 0.1684, "epoch": 0.95, "grad_norm": 0.7239227955245142, "learning_rate": 6.69020034355361e-08, "loss": 0.1665, "lr_vision_tower": 1.3380400687107219e-08, "sft_loss": 0.0823, "step": 1540 }, { "distill_loss": 0.1733, "epoch": 0.95, "grad_norm": 0.9718333672339291, "learning_rate": 6.528373560551027e-08, "loss": 0.1771, "lr_vision_tower": 1.3056747121102052e-08, "sft_loss": 0.0905, "step": 1541 }, { "distill_loss": 0.1708, "epoch": 0.95, "grad_norm": 0.5863393130962848, "learning_rate": 6.368515136694397e-08, "loss": 0.1579, "lr_vision_tower": 1.2737030273388794e-08, "sft_loss": 0.0725, "step": 1542 }, { "distill_loss": 0.1889, "epoch": 0.95, "grad_norm": 0.7477610978863541, "learning_rate": 6.210625709626772e-08, "loss": 0.1671, "lr_vision_tower": 1.2421251419253542e-08, "sft_loss": 0.0726, "step": 1543 }, { "distill_loss": 0.172, "epoch": 0.95, "grad_norm": 0.871388717490215, "learning_rate": 6.054705909137426e-08, "loss": 0.1623, "lr_vision_tower": 1.210941181827485e-08, "sft_loss": 0.0763, "step": 1544 }, { "distill_loss": 0.1701, "epoch": 0.95, "grad_norm": 0.9371818054240192, "learning_rate": 5.900756357159143e-08, "loss": 0.1663, "lr_vision_tower": 1.1801512714318285e-08, "sft_loss": 0.0812, "step": 1545 }, { "distill_loss": 0.1769, "epoch": 0.95, "grad_norm": 0.7567149255185033, "learning_rate": 5.748777667765659e-08, "loss": 0.1587, "lr_vision_tower": 1.1497555335531317e-08, "sft_loss": 0.0702, "step": 1546 }, { "distill_loss": 0.161, "epoch": 0.95, "grad_norm": 0.8600926213095056, "learning_rate": 5.598770447169444e-08, "loss": 0.1623, "lr_vision_tower": 1.1197540894338886e-08, "sft_loss": 0.0818, "step": 1547 }, { "distill_loss": 0.1626, "epoch": 0.95, "grad_norm": 1.0083597887946942, "learning_rate": 5.450735293719145e-08, "loss": 0.1722, "lr_vision_tower": 1.0901470587438289e-08, "sft_loss": 0.0909, "step": 1548 }, { "distill_loss": 0.1716, "epoch": 0.95, "grad_norm": 1.0487232389774679, "learning_rate": 5.3046727978973146e-08, "loss": 0.1808, "lr_vision_tower": 1.0609345595794627e-08, "sft_loss": 0.095, "step": 1549 }, { "distill_loss": 0.1653, "epoch": 0.96, "grad_norm": 0.6824072785473778, "learning_rate": 5.160583542317799e-08, "loss": 0.1578, "lr_vision_tower": 1.0321167084635595e-08, "sft_loss": 0.0751, "step": 1550 }, { "distill_loss": 0.181, "epoch": 0.96, "grad_norm": 0.7384523619346006, "learning_rate": 5.018468101723739e-08, "loss": 0.1727, "lr_vision_tower": 1.0036936203447477e-08, "sft_loss": 0.0822, "step": 1551 }, { "distill_loss": 0.167, "epoch": 0.96, "grad_norm": 0.7939619006004527, "learning_rate": 4.87832704298502e-08, "loss": 0.1561, "lr_vision_tower": 9.756654085970039e-09, "sft_loss": 0.0726, "step": 1552 }, { "distill_loss": 0.1752, "epoch": 0.96, "grad_norm": 0.8713234860196146, "learning_rate": 4.740160925096215e-08, "loss": 0.1605, "lr_vision_tower": 9.480321850192429e-09, "sft_loss": 0.0729, "step": 1553 }, { "distill_loss": 0.1792, "epoch": 0.96, "grad_norm": 0.9568948923984341, "learning_rate": 4.603970299174143e-08, "loss": 0.1793, "lr_vision_tower": 9.207940598348285e-09, "sft_loss": 0.0897, "step": 1554 }, { "distill_loss": 0.1751, "epoch": 0.96, "grad_norm": 0.8890464700051774, "learning_rate": 4.469755708455925e-08, "loss": 0.1643, "lr_vision_tower": 8.939511416911849e-09, "sft_loss": 0.0767, "step": 1555 }, { "distill_loss": 0.1705, "epoch": 0.96, "grad_norm": 0.8791020749103251, "learning_rate": 4.337517688296544e-08, "loss": 0.1626, "lr_vision_tower": 8.675035376593088e-09, "sft_loss": 0.0774, "step": 1556 }, { "distill_loss": 0.1665, "epoch": 0.96, "grad_norm": 0.7808127107007243, "learning_rate": 4.207256766166845e-08, "loss": 0.1626, "lr_vision_tower": 8.41451353233369e-09, "sft_loss": 0.0794, "step": 1557 }, { "distill_loss": 0.175, "epoch": 0.96, "grad_norm": 0.9525423382842448, "learning_rate": 4.0789734616515364e-08, "loss": 0.1642, "lr_vision_tower": 8.157946923303071e-09, "sft_loss": 0.0767, "step": 1558 }, { "distill_loss": 0.1693, "epoch": 0.96, "grad_norm": 0.7286601942406241, "learning_rate": 3.952668286446859e-08, "loss": 0.1569, "lr_vision_tower": 7.905336572893717e-09, "sft_loss": 0.0723, "step": 1559 }, { "distill_loss": 0.1677, "epoch": 0.96, "grad_norm": 0.7851422241704272, "learning_rate": 3.8283417443588656e-08, "loss": 0.164, "lr_vision_tower": 7.65668348871773e-09, "sft_loss": 0.0801, "step": 1560 }, { "distill_loss": 0.1779, "epoch": 0.96, "grad_norm": 0.8398637385264102, "learning_rate": 3.705994331301088e-08, "loss": 0.1679, "lr_vision_tower": 7.411988662602175e-09, "sft_loss": 0.079, "step": 1561 }, { "distill_loss": 0.1717, "epoch": 0.96, "grad_norm": 1.0820533027790278, "learning_rate": 3.585626535292819e-08, "loss": 0.1823, "lr_vision_tower": 7.171253070585637e-09, "sft_loss": 0.0964, "step": 1562 }, { "distill_loss": 0.1783, "epoch": 0.96, "grad_norm": 0.9961244804078322, "learning_rate": 3.467238836456943e-08, "loss": 0.1849, "lr_vision_tower": 6.934477672913885e-09, "sft_loss": 0.0958, "step": 1563 }, { "distill_loss": 0.1629, "epoch": 0.96, "grad_norm": 0.8644056862043642, "learning_rate": 3.350831707018276e-08, "loss": 0.1486, "lr_vision_tower": 6.7016634140365515e-09, "sft_loss": 0.0671, "step": 1564 }, { "distill_loss": 0.1666, "epoch": 0.96, "grad_norm": 0.9947449783777276, "learning_rate": 3.236405611301341e-08, "loss": 0.1577, "lr_vision_tower": 6.472811222602681e-09, "sft_loss": 0.0744, "step": 1565 }, { "distill_loss": 0.1666, "epoch": 0.97, "grad_norm": 0.8653656393858862, "learning_rate": 3.1239610057289816e-08, "loss": 0.16, "lr_vision_tower": 6.247922011457963e-09, "sft_loss": 0.0767, "step": 1566 }, { "distill_loss": 0.1709, "epoch": 0.97, "grad_norm": 0.7452022794317281, "learning_rate": 3.013498338820031e-08, "loss": 0.1475, "lr_vision_tower": 6.026996677640061e-09, "sft_loss": 0.062, "step": 1567 }, { "distill_loss": 0.1593, "epoch": 0.97, "grad_norm": 1.8987899980413316, "learning_rate": 2.905018051187869e-08, "loss": 0.1553, "lr_vision_tower": 5.8100361023757375e-09, "sft_loss": 0.0756, "step": 1568 }, { "distill_loss": 0.1791, "epoch": 0.97, "grad_norm": 0.6945234112093833, "learning_rate": 2.798520575538699e-08, "loss": 0.1734, "lr_vision_tower": 5.597041151077397e-09, "sft_loss": 0.0838, "step": 1569 }, { "distill_loss": 0.1632, "epoch": 0.97, "grad_norm": 0.8178609887222126, "learning_rate": 2.6940063366693303e-08, "loss": 0.1626, "lr_vision_tower": 5.38801267333866e-09, "sft_loss": 0.081, "step": 1570 }, { "distill_loss": 0.1712, "epoch": 0.97, "grad_norm": 0.7358916061852437, "learning_rate": 2.5914757514662327e-08, "loss": 0.165, "lr_vision_tower": 5.182951502932465e-09, "sft_loss": 0.0794, "step": 1571 }, { "distill_loss": 0.1748, "epoch": 0.97, "grad_norm": 0.8273383950995047, "learning_rate": 2.490929228903205e-08, "loss": 0.1563, "lr_vision_tower": 4.9818584578064094e-09, "sft_loss": 0.0689, "step": 1572 }, { "distill_loss": 0.1717, "epoch": 0.97, "grad_norm": 0.8858285050588328, "learning_rate": 2.3923671700401552e-08, "loss": 0.1725, "lr_vision_tower": 4.7847343400803095e-09, "sft_loss": 0.0867, "step": 1573 }, { "distill_loss": 0.1731, "epoch": 0.97, "grad_norm": 0.8595058676919844, "learning_rate": 2.295789968021378e-08, "loss": 0.1714, "lr_vision_tower": 4.591579936042756e-09, "sft_loss": 0.0849, "step": 1574 }, { "distill_loss": 0.1736, "epoch": 0.97, "grad_norm": 0.7636192248151898, "learning_rate": 2.201198008073835e-08, "loss": 0.1633, "lr_vision_tower": 4.402396016147669e-09, "sft_loss": 0.0765, "step": 1575 }, { "distill_loss": 0.1706, "epoch": 0.97, "grad_norm": 0.8892439169150971, "learning_rate": 2.1085916675059326e-08, "loss": 0.1663, "lr_vision_tower": 4.217183335011865e-09, "sft_loss": 0.081, "step": 1576 }, { "distill_loss": 0.1802, "epoch": 0.97, "grad_norm": 0.8386966789714614, "learning_rate": 2.0179713157056357e-08, "loss": 0.1751, "lr_vision_tower": 4.035942631411271e-09, "sft_loss": 0.085, "step": 1577 }, { "distill_loss": 0.1728, "epoch": 0.97, "grad_norm": 0.7675770140999322, "learning_rate": 1.9293373141394124e-08, "loss": 0.1645, "lr_vision_tower": 3.858674628278824e-09, "sft_loss": 0.0781, "step": 1578 }, { "distill_loss": 0.1763, "epoch": 0.97, "grad_norm": 0.9835289975284167, "learning_rate": 1.8426900163504568e-08, "loss": 0.164, "lr_vision_tower": 3.6853800327009133e-09, "sft_loss": 0.0758, "step": 1579 }, { "distill_loss": 0.1685, "epoch": 0.97, "grad_norm": 1.3063791078173723, "learning_rate": 1.7580297679574697e-08, "loss": 0.1532, "lr_vision_tower": 3.516059535914939e-09, "sft_loss": 0.069, "step": 1580 }, { "distill_loss": 0.1764, "epoch": 0.97, "grad_norm": 0.7924844857616153, "learning_rate": 1.6753569066531027e-08, "loss": 0.168, "lr_vision_tower": 3.350713813306205e-09, "sft_loss": 0.0798, "step": 1581 }, { "distill_loss": 0.1803, "epoch": 0.98, "grad_norm": 0.7883970708851163, "learning_rate": 1.5946717622027376e-08, "loss": 0.1623, "lr_vision_tower": 3.1893435244054746e-09, "sft_loss": 0.0722, "step": 1582 }, { "distill_loss": 0.1717, "epoch": 0.98, "grad_norm": 0.6419772672351395, "learning_rate": 1.51597465644332e-08, "loss": 0.1789, "lr_vision_tower": 3.031949312886639e-09, "sft_loss": 0.093, "step": 1583 }, { "distill_loss": 0.1823, "epoch": 0.98, "grad_norm": 0.7495016388416025, "learning_rate": 1.4392659032817501e-08, "loss": 0.1577, "lr_vision_tower": 2.8785318065635e-09, "sft_loss": 0.0666, "step": 1584 }, { "distill_loss": 0.1862, "epoch": 0.98, "grad_norm": 0.8891176490725273, "learning_rate": 1.364545808693829e-08, "loss": 0.1816, "lr_vision_tower": 2.7290916173876577e-09, "sft_loss": 0.0885, "step": 1585 }, { "distill_loss": 0.1669, "epoch": 0.98, "grad_norm": 1.271727565116866, "learning_rate": 1.291814670722924e-08, "loss": 0.1633, "lr_vision_tower": 2.5836293414458477e-09, "sft_loss": 0.0799, "step": 1586 }, { "distill_loss": 0.1681, "epoch": 0.98, "grad_norm": 0.753650982586318, "learning_rate": 1.2210727794790266e-08, "loss": 0.1764, "lr_vision_tower": 2.442145558958053e-09, "sft_loss": 0.0924, "step": 1587 }, { "distill_loss": 0.1713, "epoch": 0.98, "grad_norm": 0.9301293654384301, "learning_rate": 1.1523204171373093e-08, "loss": 0.1595, "lr_vision_tower": 2.304640834274618e-09, "sft_loss": 0.0738, "step": 1588 }, { "distill_loss": 0.1802, "epoch": 0.98, "grad_norm": 0.8085554656237852, "learning_rate": 1.0855578579370696e-08, "loss": 0.1694, "lr_vision_tower": 2.1711157158741387e-09, "sft_loss": 0.0793, "step": 1589 }, { "distill_loss": 0.1734, "epoch": 0.98, "grad_norm": 0.7799840081202039, "learning_rate": 1.0207853681807877e-08, "loss": 0.1608, "lr_vision_tower": 2.0415707363615754e-09, "sft_loss": 0.0741, "step": 1590 }, { "distill_loss": 0.1879, "epoch": 0.98, "grad_norm": 0.7338445315514789, "learning_rate": 9.580032062328493e-09, "loss": 0.1929, "lr_vision_tower": 1.9160064124656982e-09, "sft_loss": 0.0989, "step": 1591 }, { "distill_loss": 0.166, "epoch": 0.98, "grad_norm": 0.7633939371532082, "learning_rate": 8.972116225188232e-09, "loss": 0.1765, "lr_vision_tower": 1.7944232450376462e-09, "sft_loss": 0.0935, "step": 1592 }, { "distill_loss": 0.1668, "epoch": 0.98, "grad_norm": 0.7411467396658248, "learning_rate": 8.384108595239082e-09, "loss": 0.1669, "lr_vision_tower": 1.6768217190478163e-09, "sft_loss": 0.0835, "step": 1593 }, { "distill_loss": 0.1724, "epoch": 0.98, "grad_norm": 0.7974791105823952, "learning_rate": 7.816011517927102e-09, "loss": 0.1735, "lr_vision_tower": 1.5632023035854202e-09, "sft_loss": 0.0873, "step": 1594 }, { "distill_loss": 0.1672, "epoch": 0.98, "grad_norm": 0.7913650688065561, "learning_rate": 7.267827259276883e-09, "loss": 0.1519, "lr_vision_tower": 1.4535654518553764e-09, "sft_loss": 0.0683, "step": 1595 }, { "distill_loss": 0.1727, "epoch": 0.98, "grad_norm": 0.934133520816322, "learning_rate": 6.739558005884883e-09, "loss": 0.1547, "lr_vision_tower": 1.3479116011769765e-09, "sft_loss": 0.0683, "step": 1596 }, { "distill_loss": 0.1786, "epoch": 0.98, "grad_norm": 0.7263036844960007, "learning_rate": 6.231205864911105e-09, "loss": 0.1601, "lr_vision_tower": 1.2462411729822209e-09, "sft_loss": 0.0708, "step": 1597 }, { "distill_loss": 0.1745, "epoch": 0.98, "grad_norm": 0.8438663239364587, "learning_rate": 5.742772864070212e-09, "loss": 0.1719, "lr_vision_tower": 1.1485545728140422e-09, "sft_loss": 0.0847, "step": 1598 }, { "distill_loss": 0.1673, "epoch": 0.99, "grad_norm": 0.8275815202288196, "learning_rate": 5.2742609516232e-09, "loss": 0.1582, "lr_vision_tower": 1.0548521903246398e-09, "sft_loss": 0.0746, "step": 1599 }, { "distill_loss": 0.1755, "epoch": 0.99, "grad_norm": 2.222173289770223, "learning_rate": 4.825671996370184e-09, "loss": 0.1796, "lr_vision_tower": 9.651343992740368e-10, "sft_loss": 0.0918, "step": 1600 }, { "distill_loss": 0.1786, "epoch": 0.99, "grad_norm": 0.7184710764338789, "learning_rate": 4.397007787642071e-09, "loss": 0.1676, "lr_vision_tower": 8.794015575284141e-10, "sft_loss": 0.0783, "step": 1601 }, { "distill_loss": 0.1729, "epoch": 0.99, "grad_norm": 1.5630631856794057, "learning_rate": 3.98827003529445e-09, "loss": 0.1739, "lr_vision_tower": 7.9765400705889e-10, "sft_loss": 0.0874, "step": 1602 }, { "distill_loss": 0.1762, "epoch": 0.99, "grad_norm": 0.701951015819071, "learning_rate": 3.5994603696998253e-09, "loss": 0.1925, "lr_vision_tower": 7.19892073939965e-10, "sft_loss": 0.1044, "step": 1603 }, { "distill_loss": 0.1771, "epoch": 0.99, "grad_norm": 0.6977949168597762, "learning_rate": 3.2305803417420623e-09, "loss": 0.1718, "lr_vision_tower": 6.461160683484123e-10, "sft_loss": 0.0832, "step": 1604 }, { "distill_loss": 0.184, "epoch": 0.99, "grad_norm": 0.9392335076815718, "learning_rate": 2.8816314228091726e-09, "loss": 0.1829, "lr_vision_tower": 5.763262845618344e-10, "sft_loss": 0.0909, "step": 1605 }, { "distill_loss": 0.1705, "epoch": 0.99, "grad_norm": 1.0163746552695165, "learning_rate": 2.552615004788317e-09, "loss": 0.1707, "lr_vision_tower": 5.105230009576633e-10, "sft_loss": 0.0854, "step": 1606 }, { "distill_loss": 0.1739, "epoch": 0.99, "grad_norm": 1.1633377742493483, "learning_rate": 2.2435324000602555e-09, "loss": 0.1737, "lr_vision_tower": 4.48706480012051e-10, "sft_loss": 0.0868, "step": 1607 }, { "distill_loss": 0.1777, "epoch": 0.99, "grad_norm": 1.0206840854502959, "learning_rate": 1.9543848414926847e-09, "loss": 0.1669, "lr_vision_tower": 3.908769682985369e-10, "sft_loss": 0.078, "step": 1608 }, { "distill_loss": 0.1717, "epoch": 0.99, "grad_norm": 0.6188889579808636, "learning_rate": 1.6851734824380184e-09, "loss": 0.1605, "lr_vision_tower": 3.3703469648760363e-10, "sft_loss": 0.0746, "step": 1609 }, { "distill_loss": 0.1662, "epoch": 0.99, "grad_norm": 0.8787156772037515, "learning_rate": 1.4358993967256153e-09, "loss": 0.1698, "lr_vision_tower": 2.87179879345123e-10, "sft_loss": 0.0866, "step": 1610 }, { "distill_loss": 0.1823, "epoch": 0.99, "grad_norm": 0.9940897979765658, "learning_rate": 1.2065635786595586e-09, "loss": 0.1723, "lr_vision_tower": 2.413127157319117e-10, "sft_loss": 0.0811, "step": 1611 }, { "distill_loss": 0.1753, "epoch": 0.99, "grad_norm": 1.1247494979717556, "learning_rate": 9.971669430147712e-10, "loss": 0.1656, "lr_vision_tower": 1.9943338860295422e-10, "sft_loss": 0.0779, "step": 1612 }, { "distill_loss": 0.1751, "epoch": 0.99, "grad_norm": 1.0527297509102262, "learning_rate": 8.077103250325735e-10, "loss": 0.1829, "lr_vision_tower": 1.6154206500651468e-10, "sft_loss": 0.0954, "step": 1613 }, { "distill_loss": 0.1598, "epoch": 0.99, "grad_norm": 0.8900404192305319, "learning_rate": 6.381944804162432e-10, "loss": 0.176, "lr_vision_tower": 1.2763889608324861e-10, "sft_loss": 0.0961, "step": 1614 }, { "distill_loss": 0.1733, "epoch": 1.0, "grad_norm": 0.6808491899494402, "learning_rate": 4.886200853321255e-10, "loss": 0.1564, "lr_vision_tower": 9.772401706642508e-11, "sft_loss": 0.0698, "step": 1615 }, { "distill_loss": 0.1752, "epoch": 1.0, "grad_norm": 0.804051697094483, "learning_rate": 3.58987736400751e-10, "loss": 0.1616, "lr_vision_tower": 7.179754728015019e-11, "sft_loss": 0.074, "step": 1616 }, { "distill_loss": 0.1774, "epoch": 1.0, "grad_norm": 0.8518980322952758, "learning_rate": 2.4929795070072204e-10, "loss": 0.1731, "lr_vision_tower": 4.98595901401444e-11, "sft_loss": 0.0845, "step": 1617 }, { "distill_loss": 0.1724, "epoch": 1.0, "grad_norm": 0.7573356471418595, "learning_rate": 1.5955116576149565e-10, "loss": 0.1741, "lr_vision_tower": 3.1910233152299126e-11, "sft_loss": 0.0879, "step": 1618 }, { "distill_loss": 0.1707, "epoch": 1.0, "grad_norm": 0.9213450395860959, "learning_rate": 8.974773956560434e-11, "loss": 0.1567, "lr_vision_tower": 1.7949547913120866e-11, "sft_loss": 0.0713, "step": 1619 }, { "distill_loss": 0.1735, "epoch": 1.0, "grad_norm": 0.848410303447685, "learning_rate": 3.988795054532535e-11, "loss": 0.1686, "lr_vision_tower": 7.977590109065068e-12, "sft_loss": 0.0818, "step": 1620 }, { "distill_loss": 0.171, "epoch": 1.0, "grad_norm": 0.9817480640496691, "learning_rate": 9.971997580460191e-12, "loss": 0.1561, "lr_vision_tower": 1.994399516092038e-12, "sft_loss": 0.0706, "step": 1621 }, { "distill_loss": 0.1732, "epoch": 1.0, "grad_norm": 1.14559152949672, "learning_rate": 0.0, "loss": 0.1839, "lr_vision_tower": 0.0, "sft_loss": 0.0973, "step": 1622 }, { "epoch": 1.0, "step": 1622, "total_flos": 340300517736448.0, "train_loss": 0.2040031102799458, "train_runtime": 35929.6768, "train_samples_per_second": 5.78, "train_steps_per_second": 0.045 } ], "logging_steps": 1.0, "max_steps": 1622, "num_input_tokens_seen": 0, "num_train_epochs": 1, "save_steps": 500, "total_flos": 340300517736448.0, "train_batch_size": 1, "trial_name": null, "trial_params": null }
{
"best_metric": null,
"best_model_checkpoint": null,
"epoch": 0.9997688573850065,
"eval_steps": 500,
"global_step": 1622,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"distill_loss": 0.9078,
"epoch": 0.0,
"grad_norm": 19.151357114127673,
"learning_rate": 2.0408163265306121e-07,
"loss": 0.7418,
"lr_vision_tower": 4.081632653061224e-08,
"sft_loss": 0.288,
"step": 1
},
{
"distill_loss": 0.9047,
"epoch": 0.0,
"grad_norm": 19.707155046403212,
"learning_rate": 4.0816326530612243e-07,
"loss": 0.7432,
"lr_vision_tower": 8.163265306122448e-08,
"sft_loss": 0.2908,
"step": 2
},
{
"distill_loss": 0.9044,
"epoch": 0.0,
"grad_norm": 20.608160938114064,
"learning_rate": 6.122448979591837e-07,
"loss": 0.7153,
"lr_vision_tower": 1.2244897959183673e-07,
"sft_loss": 0.2631,
"step": 3
},
{
"distill_loss": 0.9068,
"epoch": 0.0,
"grad_norm": 19.97875237942443,
"learning_rate": 8.163265306122449e-07,
"loss": 0.7139,
"lr_vision_tower": 1.6326530612244896e-07,
"sft_loss": 0.2605,
"step": 4
},
{
"distill_loss": 0.9045,
"epoch": 0.0,
"grad_norm": 16.31548300275858,
"learning_rate": 1.0204081632653063e-06,
"loss": 0.7107,
"lr_vision_tower": 2.0408163265306121e-07,
"sft_loss": 0.2585,
"step": 5
},
{
"distill_loss": 0.9095,
"epoch": 0.0,
"grad_norm": 12.295941310553859,
"learning_rate": 1.2244897959183673e-06,
"loss": 0.6678,
"lr_vision_tower": 2.4489795918367347e-07,
"sft_loss": 0.213,
"step": 6
},
{
"distill_loss": 0.9057,
"epoch": 0.0,
"grad_norm": 9.94940809093829,
"learning_rate": 1.4285714285714286e-06,
"loss": 0.6525,
"lr_vision_tower": 2.857142857142857e-07,
"sft_loss": 0.1997,
"step": 7
},
{
"distill_loss": 0.9034,
"epoch": 0.0,
"grad_norm": 2.891667866636629,
"learning_rate": 1.6326530612244897e-06,
"loss": 0.6037,
"lr_vision_tower": 3.265306122448979e-07,
"sft_loss": 0.152,
"step": 8
},
{
"distill_loss": 0.9043,
"epoch": 0.01,
"grad_norm": 2.6203634446256605,
"learning_rate": 1.8367346938775512e-06,
"loss": 0.6121,
"lr_vision_tower": 3.673469387755102e-07,
"sft_loss": 0.1599,
"step": 9
},