diff --git "a/checkpoint-3116/trainer_state.json" "b/checkpoint-3116/trainer_state.json" new file mode 100644--- /dev/null +++ "b/checkpoint-3116/trainer_state.json" @@ -0,0 +1,21845 @@ +{ + "best_metric": null, + "best_model_checkpoint": null, + "epoch": 3.9961489088575095, + "eval_steps": 500, + "global_step": 3116, + "is_hyper_param_search": false, + "is_local_process_zero": true, + "is_world_process_zero": true, + "log_history": [ + { + "epoch": 0.0012836970474967907, + "grad_norm": 0.9255548715591431, + "learning_rate": 5.000000000000001e-07, + "loss": 2.8985, + "step": 1 + }, + { + "epoch": 0.0025673940949935813, + "grad_norm": 0.7692601680755615, + "learning_rate": 1.0000000000000002e-06, + "loss": 2.9774, + "step": 2 + }, + { + "epoch": 0.0038510911424903724, + "grad_norm": 0.7884671092033386, + "learning_rate": 1.5e-06, + "loss": 2.9898, + "step": 3 + }, + { + "epoch": 0.005134788189987163, + "grad_norm": 0.8319393396377563, + "learning_rate": 2.0000000000000003e-06, + "loss": 2.9573, + "step": 4 + }, + { + "epoch": 0.006418485237483954, + "grad_norm": 0.822285532951355, + "learning_rate": 2.5e-06, + "loss": 2.9316, + "step": 5 + }, + { + "epoch": 0.007702182284980745, + "grad_norm": 0.7564650774002075, + "learning_rate": 3e-06, + "loss": 2.9537, + "step": 6 + }, + { + "epoch": 0.008985879332477536, + "grad_norm": 0.9220781922340393, + "learning_rate": 3.5000000000000004e-06, + "loss": 2.9826, + "step": 7 + }, + { + "epoch": 0.010269576379974325, + "grad_norm": 0.7977064847946167, + "learning_rate": 4.000000000000001e-06, + "loss": 2.8548, + "step": 8 + }, + { + "epoch": 0.011553273427471117, + "grad_norm": 0.6889916658401489, + "learning_rate": 4.5e-06, + "loss": 2.9936, + "step": 9 + }, + { + "epoch": 0.012836970474967908, + "grad_norm": 0.9777728915214539, + "learning_rate": 5e-06, + "loss": 2.9578, + "step": 10 + }, + { + "epoch": 0.014120667522464698, + "grad_norm": 0.8187949061393738, + "learning_rate": 5.500000000000001e-06, + "loss": 2.9442, + "step": 11 + }, + { + "epoch": 0.01540436456996149, + "grad_norm": 0.7016908526420593, + "learning_rate": 6e-06, + "loss": 2.9299, + "step": 12 + }, + { + "epoch": 0.01668806161745828, + "grad_norm": 0.6974747180938721, + "learning_rate": 6.5000000000000004e-06, + "loss": 2.9699, + "step": 13 + }, + { + "epoch": 0.01797175866495507, + "grad_norm": 0.8423139452934265, + "learning_rate": 7.000000000000001e-06, + "loss": 2.8706, + "step": 14 + }, + { + "epoch": 0.019255455712451863, + "grad_norm": 0.8017705082893372, + "learning_rate": 7.5e-06, + "loss": 2.8443, + "step": 15 + }, + { + "epoch": 0.02053915275994865, + "grad_norm": 0.9362208247184753, + "learning_rate": 8.000000000000001e-06, + "loss": 2.872, + "step": 16 + }, + { + "epoch": 0.021822849807445442, + "grad_norm": 1.013128638267517, + "learning_rate": 8.500000000000002e-06, + "loss": 2.797, + "step": 17 + }, + { + "epoch": 0.023106546854942234, + "grad_norm": 1.255325436592102, + "learning_rate": 9e-06, + "loss": 2.8109, + "step": 18 + }, + { + "epoch": 0.024390243902439025, + "grad_norm": 1.1081339120864868, + "learning_rate": 9.5e-06, + "loss": 2.6742, + "step": 19 + }, + { + "epoch": 0.025673940949935817, + "grad_norm": 0.8878622651100159, + "learning_rate": 1e-05, + "loss": 2.7251, + "step": 20 + }, + { + "epoch": 0.026957637997432605, + "grad_norm": 0.894791305065155, + "learning_rate": 1.05e-05, + "loss": 2.7127, + "step": 21 + }, + { + "epoch": 0.028241335044929396, + "grad_norm": 0.6742448806762695, + "learning_rate": 1.1000000000000001e-05, + "loss": 2.6842, + "step": 22 + }, + { + "epoch": 0.029525032092426188, + "grad_norm": 0.6250098943710327, + "learning_rate": 1.1500000000000002e-05, + "loss": 2.6476, + "step": 23 + }, + { + "epoch": 0.03080872913992298, + "grad_norm": 0.6331678032875061, + "learning_rate": 1.2e-05, + "loss": 2.6339, + "step": 24 + }, + { + "epoch": 0.03209242618741977, + "grad_norm": 0.4726584255695343, + "learning_rate": 1.25e-05, + "loss": 2.5994, + "step": 25 + }, + { + "epoch": 0.03337612323491656, + "grad_norm": 0.46077489852905273, + "learning_rate": 1.3000000000000001e-05, + "loss": 2.5545, + "step": 26 + }, + { + "epoch": 0.03465982028241335, + "grad_norm": 0.5746111273765564, + "learning_rate": 1.3500000000000001e-05, + "loss": 2.5542, + "step": 27 + }, + { + "epoch": 0.03594351732991014, + "grad_norm": 0.47136253118515015, + "learning_rate": 1.4000000000000001e-05, + "loss": 2.6138, + "step": 28 + }, + { + "epoch": 0.037227214377406934, + "grad_norm": 0.5951219797134399, + "learning_rate": 1.45e-05, + "loss": 2.5941, + "step": 29 + }, + { + "epoch": 0.038510911424903725, + "grad_norm": 0.48593708872795105, + "learning_rate": 1.5e-05, + "loss": 2.5751, + "step": 30 + }, + { + "epoch": 0.03979460847240052, + "grad_norm": 0.482264906167984, + "learning_rate": 1.55e-05, + "loss": 2.5689, + "step": 31 + }, + { + "epoch": 0.0410783055198973, + "grad_norm": 0.4937015175819397, + "learning_rate": 1.6000000000000003e-05, + "loss": 2.5559, + "step": 32 + }, + { + "epoch": 0.04236200256739409, + "grad_norm": 0.5136978030204773, + "learning_rate": 1.65e-05, + "loss": 2.5274, + "step": 33 + }, + { + "epoch": 0.043645699614890884, + "grad_norm": 0.40434661507606506, + "learning_rate": 1.7000000000000003e-05, + "loss": 2.4391, + "step": 34 + }, + { + "epoch": 0.044929396662387676, + "grad_norm": 0.49542951583862305, + "learning_rate": 1.75e-05, + "loss": 2.5738, + "step": 35 + }, + { + "epoch": 0.04621309370988447, + "grad_norm": 0.4381186366081238, + "learning_rate": 1.8e-05, + "loss": 2.5095, + "step": 36 + }, + { + "epoch": 0.04749679075738126, + "grad_norm": 0.486103892326355, + "learning_rate": 1.85e-05, + "loss": 2.6173, + "step": 37 + }, + { + "epoch": 0.04878048780487805, + "grad_norm": 0.4313197135925293, + "learning_rate": 1.9e-05, + "loss": 2.4951, + "step": 38 + }, + { + "epoch": 0.05006418485237484, + "grad_norm": 0.4211004078388214, + "learning_rate": 1.9500000000000003e-05, + "loss": 2.5469, + "step": 39 + }, + { + "epoch": 0.051347881899871634, + "grad_norm": 0.44780072569847107, + "learning_rate": 2e-05, + "loss": 2.4437, + "step": 40 + }, + { + "epoch": 0.05263157894736842, + "grad_norm": 0.44611668586730957, + "learning_rate": 2.05e-05, + "loss": 2.4795, + "step": 41 + }, + { + "epoch": 0.05391527599486521, + "grad_norm": 0.4598286747932434, + "learning_rate": 2.1e-05, + "loss": 2.4786, + "step": 42 + }, + { + "epoch": 0.055198973042362, + "grad_norm": 0.4416978061199188, + "learning_rate": 2.15e-05, + "loss": 2.453, + "step": 43 + }, + { + "epoch": 0.05648267008985879, + "grad_norm": 0.4136359989643097, + "learning_rate": 2.2000000000000003e-05, + "loss": 2.4975, + "step": 44 + }, + { + "epoch": 0.057766367137355584, + "grad_norm": 0.44032955169677734, + "learning_rate": 2.25e-05, + "loss": 2.411, + "step": 45 + }, + { + "epoch": 0.059050064184852376, + "grad_norm": 0.49505728483200073, + "learning_rate": 2.3000000000000003e-05, + "loss": 2.5049, + "step": 46 + }, + { + "epoch": 0.06033376123234917, + "grad_norm": 0.43698814511299133, + "learning_rate": 2.35e-05, + "loss": 2.5828, + "step": 47 + }, + { + "epoch": 0.06161745827984596, + "grad_norm": 0.44550350308418274, + "learning_rate": 2.4e-05, + "loss": 2.4326, + "step": 48 + }, + { + "epoch": 0.06290115532734275, + "grad_norm": 0.38959425687789917, + "learning_rate": 2.45e-05, + "loss": 2.4684, + "step": 49 + }, + { + "epoch": 0.06418485237483953, + "grad_norm": 0.4324244260787964, + "learning_rate": 2.5e-05, + "loss": 2.4496, + "step": 50 + }, + { + "epoch": 0.06546854942233633, + "grad_norm": 0.4213118553161621, + "learning_rate": 2.5500000000000003e-05, + "loss": 2.4742, + "step": 51 + }, + { + "epoch": 0.06675224646983312, + "grad_norm": 0.5279268622398376, + "learning_rate": 2.6000000000000002e-05, + "loss": 2.4222, + "step": 52 + }, + { + "epoch": 0.06803594351732992, + "grad_norm": 0.40476322174072266, + "learning_rate": 2.6500000000000004e-05, + "loss": 2.4224, + "step": 53 + }, + { + "epoch": 0.0693196405648267, + "grad_norm": 0.6938806176185608, + "learning_rate": 2.7000000000000002e-05, + "loss": 2.4194, + "step": 54 + }, + { + "epoch": 0.07060333761232349, + "grad_norm": 0.43899399042129517, + "learning_rate": 2.7500000000000004e-05, + "loss": 2.4288, + "step": 55 + }, + { + "epoch": 0.07188703465982028, + "grad_norm": 0.3968575596809387, + "learning_rate": 2.8000000000000003e-05, + "loss": 2.3968, + "step": 56 + }, + { + "epoch": 0.07317073170731707, + "grad_norm": 0.6113290786743164, + "learning_rate": 2.8499999999999998e-05, + "loss": 2.3924, + "step": 57 + }, + { + "epoch": 0.07445442875481387, + "grad_norm": 0.35704493522644043, + "learning_rate": 2.9e-05, + "loss": 2.4004, + "step": 58 + }, + { + "epoch": 0.07573812580231065, + "grad_norm": 0.3809000551700592, + "learning_rate": 2.95e-05, + "loss": 2.4303, + "step": 59 + }, + { + "epoch": 0.07702182284980745, + "grad_norm": 0.4394189715385437, + "learning_rate": 3e-05, + "loss": 2.4716, + "step": 60 + }, + { + "epoch": 0.07830551989730423, + "grad_norm": 0.4325893521308899, + "learning_rate": 3.05e-05, + "loss": 2.4715, + "step": 61 + }, + { + "epoch": 0.07958921694480103, + "grad_norm": 0.3560517728328705, + "learning_rate": 3.1e-05, + "loss": 2.4745, + "step": 62 + }, + { + "epoch": 0.08087291399229782, + "grad_norm": 0.37922918796539307, + "learning_rate": 3.15e-05, + "loss": 2.4789, + "step": 63 + }, + { + "epoch": 0.0821566110397946, + "grad_norm": 0.3665093779563904, + "learning_rate": 3.2000000000000005e-05, + "loss": 2.4829, + "step": 64 + }, + { + "epoch": 0.0834403080872914, + "grad_norm": 0.38106483221054077, + "learning_rate": 3.2500000000000004e-05, + "loss": 2.4203, + "step": 65 + }, + { + "epoch": 0.08472400513478819, + "grad_norm": 0.5111002326011658, + "learning_rate": 3.3e-05, + "loss": 2.3302, + "step": 66 + }, + { + "epoch": 0.08600770218228498, + "grad_norm": 0.37738922238349915, + "learning_rate": 3.35e-05, + "loss": 2.4991, + "step": 67 + }, + { + "epoch": 0.08729139922978177, + "grad_norm": 0.34703031182289124, + "learning_rate": 3.4000000000000007e-05, + "loss": 2.4224, + "step": 68 + }, + { + "epoch": 0.08857509627727857, + "grad_norm": 0.48427700996398926, + "learning_rate": 3.45e-05, + "loss": 2.3946, + "step": 69 + }, + { + "epoch": 0.08985879332477535, + "grad_norm": 0.360221266746521, + "learning_rate": 3.5e-05, + "loss": 2.4853, + "step": 70 + }, + { + "epoch": 0.09114249037227215, + "grad_norm": 0.4006412625312805, + "learning_rate": 3.55e-05, + "loss": 2.4654, + "step": 71 + }, + { + "epoch": 0.09242618741976893, + "grad_norm": 0.3662618398666382, + "learning_rate": 3.6e-05, + "loss": 2.434, + "step": 72 + }, + { + "epoch": 0.09370988446726572, + "grad_norm": 0.3694933354854584, + "learning_rate": 3.65e-05, + "loss": 2.4194, + "step": 73 + }, + { + "epoch": 0.09499358151476252, + "grad_norm": 0.34268808364868164, + "learning_rate": 3.7e-05, + "loss": 2.4922, + "step": 74 + }, + { + "epoch": 0.0962772785622593, + "grad_norm": 0.3664718270301819, + "learning_rate": 3.7500000000000003e-05, + "loss": 2.4083, + "step": 75 + }, + { + "epoch": 0.0975609756097561, + "grad_norm": 0.3909706473350525, + "learning_rate": 3.8e-05, + "loss": 2.4546, + "step": 76 + }, + { + "epoch": 0.09884467265725289, + "grad_norm": 0.36276674270629883, + "learning_rate": 3.85e-05, + "loss": 2.5313, + "step": 77 + }, + { + "epoch": 0.10012836970474968, + "grad_norm": 0.34822535514831543, + "learning_rate": 3.9000000000000006e-05, + "loss": 2.4833, + "step": 78 + }, + { + "epoch": 0.10141206675224647, + "grad_norm": 0.37480583786964417, + "learning_rate": 3.9500000000000005e-05, + "loss": 2.449, + "step": 79 + }, + { + "epoch": 0.10269576379974327, + "grad_norm": 0.3415388762950897, + "learning_rate": 4e-05, + "loss": 2.3953, + "step": 80 + }, + { + "epoch": 0.10397946084724005, + "grad_norm": 0.3487205505371094, + "learning_rate": 4.05e-05, + "loss": 2.4469, + "step": 81 + }, + { + "epoch": 0.10526315789473684, + "grad_norm": 0.5083756446838379, + "learning_rate": 4.1e-05, + "loss": 2.3423, + "step": 82 + }, + { + "epoch": 0.10654685494223363, + "grad_norm": 0.3747817575931549, + "learning_rate": 4.15e-05, + "loss": 2.3919, + "step": 83 + }, + { + "epoch": 0.10783055198973042, + "grad_norm": 0.39472389221191406, + "learning_rate": 4.2e-05, + "loss": 2.5431, + "step": 84 + }, + { + "epoch": 0.10911424903722722, + "grad_norm": 0.36542952060699463, + "learning_rate": 4.25e-05, + "loss": 2.3858, + "step": 85 + }, + { + "epoch": 0.110397946084724, + "grad_norm": 0.34629878401756287, + "learning_rate": 4.3e-05, + "loss": 2.3528, + "step": 86 + }, + { + "epoch": 0.1116816431322208, + "grad_norm": 0.36755290627479553, + "learning_rate": 4.35e-05, + "loss": 2.5329, + "step": 87 + }, + { + "epoch": 0.11296534017971759, + "grad_norm": 0.3599033057689667, + "learning_rate": 4.4000000000000006e-05, + "loss": 2.4971, + "step": 88 + }, + { + "epoch": 0.11424903722721438, + "grad_norm": 0.3730204701423645, + "learning_rate": 4.4500000000000004e-05, + "loss": 2.4082, + "step": 89 + }, + { + "epoch": 0.11553273427471117, + "grad_norm": 0.3773551881313324, + "learning_rate": 4.5e-05, + "loss": 2.3393, + "step": 90 + }, + { + "epoch": 0.11681643132220795, + "grad_norm": 0.36052408814430237, + "learning_rate": 4.55e-05, + "loss": 2.3799, + "step": 91 + }, + { + "epoch": 0.11810012836970475, + "grad_norm": 0.32685768604278564, + "learning_rate": 4.600000000000001e-05, + "loss": 2.4284, + "step": 92 + }, + { + "epoch": 0.11938382541720154, + "grad_norm": 0.36063292622566223, + "learning_rate": 4.6500000000000005e-05, + "loss": 2.3903, + "step": 93 + }, + { + "epoch": 0.12066752246469833, + "grad_norm": 0.3656150698661804, + "learning_rate": 4.7e-05, + "loss": 2.4378, + "step": 94 + }, + { + "epoch": 0.12195121951219512, + "grad_norm": 0.3563483655452728, + "learning_rate": 4.75e-05, + "loss": 2.4082, + "step": 95 + }, + { + "epoch": 0.12323491655969192, + "grad_norm": 0.35744163393974304, + "learning_rate": 4.8e-05, + "loss": 2.3979, + "step": 96 + }, + { + "epoch": 0.1245186136071887, + "grad_norm": 0.3400294780731201, + "learning_rate": 4.85e-05, + "loss": 2.4419, + "step": 97 + }, + { + "epoch": 0.1258023106546855, + "grad_norm": 0.3691268265247345, + "learning_rate": 4.9e-05, + "loss": 2.4671, + "step": 98 + }, + { + "epoch": 0.12708600770218229, + "grad_norm": 0.3483717739582062, + "learning_rate": 4.9500000000000004e-05, + "loss": 2.3947, + "step": 99 + }, + { + "epoch": 0.12836970474967907, + "grad_norm": 0.3494178354740143, + "learning_rate": 5e-05, + "loss": 2.4063, + "step": 100 + }, + { + "epoch": 0.12965340179717585, + "grad_norm": 0.3814634680747986, + "learning_rate": 4.9999986437272225e-05, + "loss": 2.4832, + "step": 101 + }, + { + "epoch": 0.13093709884467267, + "grad_norm": 0.378907710313797, + "learning_rate": 4.999994574910364e-05, + "loss": 2.3272, + "step": 102 + }, + { + "epoch": 0.13222079589216945, + "grad_norm": 0.3730032444000244, + "learning_rate": 4.999987793553836e-05, + "loss": 2.3965, + "step": 103 + }, + { + "epoch": 0.13350449293966624, + "grad_norm": 0.3670955002307892, + "learning_rate": 4.9999782996649994e-05, + "loss": 2.4565, + "step": 104 + }, + { + "epoch": 0.13478818998716302, + "grad_norm": 0.36450427770614624, + "learning_rate": 4.999966093254153e-05, + "loss": 2.5079, + "step": 105 + }, + { + "epoch": 0.13607188703465983, + "grad_norm": 0.38658779859542847, + "learning_rate": 4.9999511743345426e-05, + "loss": 2.4775, + "step": 106 + }, + { + "epoch": 0.13735558408215662, + "grad_norm": 0.33551573753356934, + "learning_rate": 4.999933542922354e-05, + "loss": 2.461, + "step": 107 + }, + { + "epoch": 0.1386392811296534, + "grad_norm": 0.3854399025440216, + "learning_rate": 4.999913199036719e-05, + "loss": 2.4075, + "step": 108 + }, + { + "epoch": 0.1399229781771502, + "grad_norm": 0.3932352364063263, + "learning_rate": 4.9998901426997104e-05, + "loss": 2.4311, + "step": 109 + }, + { + "epoch": 0.14120667522464697, + "grad_norm": 0.3379668593406677, + "learning_rate": 4.999864373936345e-05, + "loss": 2.5016, + "step": 110 + }, + { + "epoch": 0.14249037227214378, + "grad_norm": 0.32840296626091003, + "learning_rate": 4.9998358927745826e-05, + "loss": 2.3176, + "step": 111 + }, + { + "epoch": 0.14377406931964057, + "grad_norm": 0.3191027343273163, + "learning_rate": 4.999804699245325e-05, + "loss": 2.4007, + "step": 112 + }, + { + "epoch": 0.14505776636713735, + "grad_norm": 0.3358600437641144, + "learning_rate": 4.999770793382418e-05, + "loss": 2.3724, + "step": 113 + }, + { + "epoch": 0.14634146341463414, + "grad_norm": 0.335860937833786, + "learning_rate": 4.99973417522265e-05, + "loss": 2.3479, + "step": 114 + }, + { + "epoch": 0.14762516046213095, + "grad_norm": 0.34540730714797974, + "learning_rate": 4.999694844805753e-05, + "loss": 2.3675, + "step": 115 + }, + { + "epoch": 0.14890885750962773, + "grad_norm": 0.3298475742340088, + "learning_rate": 4.999652802174402e-05, + "loss": 2.3948, + "step": 116 + }, + { + "epoch": 0.15019255455712452, + "grad_norm": 0.3612127900123596, + "learning_rate": 4.999608047374211e-05, + "loss": 2.3855, + "step": 117 + }, + { + "epoch": 0.1514762516046213, + "grad_norm": 0.4185655415058136, + "learning_rate": 4.9995605804537426e-05, + "loss": 2.4015, + "step": 118 + }, + { + "epoch": 0.1527599486521181, + "grad_norm": 0.3759553134441376, + "learning_rate": 4.9995104014644986e-05, + "loss": 2.4483, + "step": 119 + }, + { + "epoch": 0.1540436456996149, + "grad_norm": 0.35989564657211304, + "learning_rate": 4.999457510460923e-05, + "loss": 2.4974, + "step": 120 + }, + { + "epoch": 0.15532734274711169, + "grad_norm": 0.3161202073097229, + "learning_rate": 4.999401907500405e-05, + "loss": 2.3712, + "step": 121 + }, + { + "epoch": 0.15661103979460847, + "grad_norm": 0.3105814456939697, + "learning_rate": 4.999343592643274e-05, + "loss": 2.4311, + "step": 122 + }, + { + "epoch": 0.15789473684210525, + "grad_norm": 0.3236968517303467, + "learning_rate": 4.9992825659528024e-05, + "loss": 2.5536, + "step": 123 + }, + { + "epoch": 0.15917843388960207, + "grad_norm": 0.3107609748840332, + "learning_rate": 4.9992188274952064e-05, + "loss": 2.3922, + "step": 124 + }, + { + "epoch": 0.16046213093709885, + "grad_norm": 0.39889928698539734, + "learning_rate": 4.999152377339642e-05, + "loss": 2.3488, + "step": 125 + }, + { + "epoch": 0.16174582798459564, + "grad_norm": 0.379323273897171, + "learning_rate": 4.99908321555821e-05, + "loss": 2.5278, + "step": 126 + }, + { + "epoch": 0.16302952503209242, + "grad_norm": 0.35020819306373596, + "learning_rate": 4.999011342225952e-05, + "loss": 2.3139, + "step": 127 + }, + { + "epoch": 0.1643132220795892, + "grad_norm": 0.34851884841918945, + "learning_rate": 4.998936757420851e-05, + "loss": 2.3495, + "step": 128 + }, + { + "epoch": 0.16559691912708602, + "grad_norm": 0.3453572392463684, + "learning_rate": 4.9988594612238336e-05, + "loss": 2.4128, + "step": 129 + }, + { + "epoch": 0.1668806161745828, + "grad_norm": 0.34227538108825684, + "learning_rate": 4.998779453718768e-05, + "loss": 2.3419, + "step": 130 + }, + { + "epoch": 0.1681643132220796, + "grad_norm": 0.3947238028049469, + "learning_rate": 4.998696734992462e-05, + "loss": 2.3941, + "step": 131 + }, + { + "epoch": 0.16944801026957637, + "grad_norm": 0.306533545255661, + "learning_rate": 4.998611305134669e-05, + "loss": 2.4645, + "step": 132 + }, + { + "epoch": 0.17073170731707318, + "grad_norm": 0.35172709822654724, + "learning_rate": 4.998523164238082e-05, + "loss": 2.4407, + "step": 133 + }, + { + "epoch": 0.17201540436456997, + "grad_norm": 0.34511688351631165, + "learning_rate": 4.9984323123983334e-05, + "loss": 2.3815, + "step": 134 + }, + { + "epoch": 0.17329910141206675, + "grad_norm": 0.33132505416870117, + "learning_rate": 4.9983387497140006e-05, + "loss": 2.3548, + "step": 135 + }, + { + "epoch": 0.17458279845956354, + "grad_norm": 0.32891082763671875, + "learning_rate": 4.998242476286601e-05, + "loss": 2.5308, + "step": 136 + }, + { + "epoch": 0.17586649550706032, + "grad_norm": 0.331152081489563, + "learning_rate": 4.998143492220592e-05, + "loss": 2.3858, + "step": 137 + }, + { + "epoch": 0.17715019255455713, + "grad_norm": 0.31813687086105347, + "learning_rate": 4.9980417976233735e-05, + "loss": 2.3136, + "step": 138 + }, + { + "epoch": 0.17843388960205392, + "grad_norm": 0.3268696069717407, + "learning_rate": 4.9979373926052865e-05, + "loss": 2.3133, + "step": 139 + }, + { + "epoch": 0.1797175866495507, + "grad_norm": 0.3389696180820465, + "learning_rate": 4.997830277279612e-05, + "loss": 2.3983, + "step": 140 + }, + { + "epoch": 0.1810012836970475, + "grad_norm": 0.3515508770942688, + "learning_rate": 4.997720451762572e-05, + "loss": 2.4848, + "step": 141 + }, + { + "epoch": 0.1822849807445443, + "grad_norm": 0.3302924335002899, + "learning_rate": 4.997607916173329e-05, + "loss": 2.3037, + "step": 142 + }, + { + "epoch": 0.18356867779204109, + "grad_norm": 0.3332863450050354, + "learning_rate": 4.997492670633987e-05, + "loss": 2.3563, + "step": 143 + }, + { + "epoch": 0.18485237483953787, + "grad_norm": 0.3398495614528656, + "learning_rate": 4.997374715269589e-05, + "loss": 2.4056, + "step": 144 + }, + { + "epoch": 0.18613607188703465, + "grad_norm": 0.3376169502735138, + "learning_rate": 4.9972540502081184e-05, + "loss": 2.3751, + "step": 145 + }, + { + "epoch": 0.18741976893453144, + "grad_norm": 0.33639830350875854, + "learning_rate": 4.9971306755804995e-05, + "loss": 2.432, + "step": 146 + }, + { + "epoch": 0.18870346598202825, + "grad_norm": 0.410265177488327, + "learning_rate": 4.9970045915205954e-05, + "loss": 2.3647, + "step": 147 + }, + { + "epoch": 0.18998716302952504, + "grad_norm": 0.31853362917900085, + "learning_rate": 4.99687579816521e-05, + "loss": 2.4224, + "step": 148 + }, + { + "epoch": 0.19127086007702182, + "grad_norm": 0.3495614230632782, + "learning_rate": 4.9967442956540863e-05, + "loss": 2.3961, + "step": 149 + }, + { + "epoch": 0.1925545571245186, + "grad_norm": 0.33301132917404175, + "learning_rate": 4.996610084129908e-05, + "loss": 2.359, + "step": 150 + }, + { + "epoch": 0.19383825417201542, + "grad_norm": 0.3186255395412445, + "learning_rate": 4.996473163738295e-05, + "loss": 2.4488, + "step": 151 + }, + { + "epoch": 0.1951219512195122, + "grad_norm": 0.3374113142490387, + "learning_rate": 4.996333534627809e-05, + "loss": 2.4019, + "step": 152 + }, + { + "epoch": 0.196405648267009, + "grad_norm": 0.33905014395713806, + "learning_rate": 4.996191196949952e-05, + "loss": 2.3272, + "step": 153 + }, + { + "epoch": 0.19768934531450577, + "grad_norm": 0.3546963036060333, + "learning_rate": 4.996046150859161e-05, + "loss": 2.4338, + "step": 154 + }, + { + "epoch": 0.19897304236200256, + "grad_norm": 0.35965755581855774, + "learning_rate": 4.9958983965128145e-05, + "loss": 2.3634, + "step": 155 + }, + { + "epoch": 0.20025673940949937, + "grad_norm": 0.3362196385860443, + "learning_rate": 4.995747934071229e-05, + "loss": 2.4457, + "step": 156 + }, + { + "epoch": 0.20154043645699615, + "grad_norm": 0.37214434146881104, + "learning_rate": 4.995594763697657e-05, + "loss": 2.3714, + "step": 157 + }, + { + "epoch": 0.20282413350449294, + "grad_norm": 0.3538600206375122, + "learning_rate": 4.995438885558294e-05, + "loss": 2.4338, + "step": 158 + }, + { + "epoch": 0.20410783055198972, + "grad_norm": 0.3853035569190979, + "learning_rate": 4.995280299822268e-05, + "loss": 2.4036, + "step": 159 + }, + { + "epoch": 0.20539152759948653, + "grad_norm": 0.35037294030189514, + "learning_rate": 4.9951190066616495e-05, + "loss": 2.3846, + "step": 160 + }, + { + "epoch": 0.20667522464698332, + "grad_norm": 0.3986567556858063, + "learning_rate": 4.994955006251443e-05, + "loss": 2.4405, + "step": 161 + }, + { + "epoch": 0.2079589216944801, + "grad_norm": 0.385637491941452, + "learning_rate": 4.994788298769593e-05, + "loss": 2.3237, + "step": 162 + }, + { + "epoch": 0.2092426187419769, + "grad_norm": 0.34437429904937744, + "learning_rate": 4.994618884396979e-05, + "loss": 2.3477, + "step": 163 + }, + { + "epoch": 0.21052631578947367, + "grad_norm": 0.34156107902526855, + "learning_rate": 4.99444676331742e-05, + "loss": 2.3576, + "step": 164 + }, + { + "epoch": 0.21181001283697048, + "grad_norm": 0.33888015151023865, + "learning_rate": 4.99427193571767e-05, + "loss": 2.311, + "step": 165 + }, + { + "epoch": 0.21309370988446727, + "grad_norm": 0.34711146354675293, + "learning_rate": 4.99409440178742e-05, + "loss": 2.293, + "step": 166 + }, + { + "epoch": 0.21437740693196405, + "grad_norm": 0.30586686730384827, + "learning_rate": 4.993914161719297e-05, + "loss": 2.479, + "step": 167 + }, + { + "epoch": 0.21566110397946084, + "grad_norm": 0.4145469069480896, + "learning_rate": 4.993731215708866e-05, + "loss": 2.2754, + "step": 168 + }, + { + "epoch": 0.21694480102695765, + "grad_norm": 0.6243994235992432, + "learning_rate": 4.993545563954626e-05, + "loss": 2.3751, + "step": 169 + }, + { + "epoch": 0.21822849807445444, + "grad_norm": 0.3301653265953064, + "learning_rate": 4.993357206658011e-05, + "loss": 2.6124, + "step": 170 + }, + { + "epoch": 0.21951219512195122, + "grad_norm": 0.337147980928421, + "learning_rate": 4.993166144023396e-05, + "loss": 2.3946, + "step": 171 + }, + { + "epoch": 0.220795892169448, + "grad_norm": 0.3796294033527374, + "learning_rate": 4.9929723762580835e-05, + "loss": 2.3314, + "step": 172 + }, + { + "epoch": 0.2220795892169448, + "grad_norm": 0.4021480679512024, + "learning_rate": 4.9927759035723175e-05, + "loss": 2.3035, + "step": 173 + }, + { + "epoch": 0.2233632862644416, + "grad_norm": 0.3196207582950592, + "learning_rate": 4.992576726179274e-05, + "loss": 2.4047, + "step": 174 + }, + { + "epoch": 0.2246469833119384, + "grad_norm": 0.31279096007347107, + "learning_rate": 4.992374844295064e-05, + "loss": 2.4409, + "step": 175 + }, + { + "epoch": 0.22593068035943517, + "grad_norm": 0.3068607747554779, + "learning_rate": 4.992170258138732e-05, + "loss": 2.4109, + "step": 176 + }, + { + "epoch": 0.22721437740693196, + "grad_norm": 0.35363098978996277, + "learning_rate": 4.991962967932258e-05, + "loss": 2.4083, + "step": 177 + }, + { + "epoch": 0.22849807445442877, + "grad_norm": 0.33908936381340027, + "learning_rate": 4.9917529739005574e-05, + "loss": 2.4129, + "step": 178 + }, + { + "epoch": 0.22978177150192555, + "grad_norm": 0.32510805130004883, + "learning_rate": 4.991540276271476e-05, + "loss": 2.3163, + "step": 179 + }, + { + "epoch": 0.23106546854942234, + "grad_norm": 0.32639047503471375, + "learning_rate": 4.991324875275794e-05, + "loss": 2.3567, + "step": 180 + }, + { + "epoch": 0.23234916559691912, + "grad_norm": 0.3410406708717346, + "learning_rate": 4.991106771147227e-05, + "loss": 2.4137, + "step": 181 + }, + { + "epoch": 0.2336328626444159, + "grad_norm": 0.3334997892379761, + "learning_rate": 4.990885964122421e-05, + "loss": 2.3994, + "step": 182 + }, + { + "epoch": 0.23491655969191272, + "grad_norm": 0.3203848600387573, + "learning_rate": 4.990662454440956e-05, + "loss": 2.4259, + "step": 183 + }, + { + "epoch": 0.2362002567394095, + "grad_norm": 0.34999069571495056, + "learning_rate": 4.9904362423453446e-05, + "loss": 2.413, + "step": 184 + }, + { + "epoch": 0.2374839537869063, + "grad_norm": 0.3413456678390503, + "learning_rate": 4.990207328081029e-05, + "loss": 2.3749, + "step": 185 + }, + { + "epoch": 0.23876765083440307, + "grad_norm": 0.3702191114425659, + "learning_rate": 4.989975711896388e-05, + "loss": 2.4309, + "step": 186 + }, + { + "epoch": 0.24005134788189988, + "grad_norm": 0.3365226089954376, + "learning_rate": 4.989741394042727e-05, + "loss": 2.3708, + "step": 187 + }, + { + "epoch": 0.24133504492939667, + "grad_norm": 0.36596447229385376, + "learning_rate": 4.989504374774288e-05, + "loss": 2.4387, + "step": 188 + }, + { + "epoch": 0.24261874197689345, + "grad_norm": 0.3280293941497803, + "learning_rate": 4.9892646543482377e-05, + "loss": 2.498, + "step": 189 + }, + { + "epoch": 0.24390243902439024, + "grad_norm": 0.35989147424697876, + "learning_rate": 4.989022233024681e-05, + "loss": 2.4993, + "step": 190 + }, + { + "epoch": 0.24518613607188702, + "grad_norm": 0.3124522268772125, + "learning_rate": 4.988777111066646e-05, + "loss": 2.3186, + "step": 191 + }, + { + "epoch": 0.24646983311938384, + "grad_norm": 0.32323285937309265, + "learning_rate": 4.988529288740096e-05, + "loss": 2.3859, + "step": 192 + }, + { + "epoch": 0.24775353016688062, + "grad_norm": 0.33156561851501465, + "learning_rate": 4.988278766313922e-05, + "loss": 2.3271, + "step": 193 + }, + { + "epoch": 0.2490372272143774, + "grad_norm": 0.32494014501571655, + "learning_rate": 4.9880255440599476e-05, + "loss": 2.4023, + "step": 194 + }, + { + "epoch": 0.2503209242618742, + "grad_norm": 0.3906209468841553, + "learning_rate": 4.987769622252921e-05, + "loss": 2.4245, + "step": 195 + }, + { + "epoch": 0.251604621309371, + "grad_norm": 0.3735322952270508, + "learning_rate": 4.987511001170523e-05, + "loss": 2.4883, + "step": 196 + }, + { + "epoch": 0.25288831835686776, + "grad_norm": 0.30935361981391907, + "learning_rate": 4.987249681093362e-05, + "loss": 2.5118, + "step": 197 + }, + { + "epoch": 0.25417201540436457, + "grad_norm": 0.34933099150657654, + "learning_rate": 4.986985662304976e-05, + "loss": 2.3138, + "step": 198 + }, + { + "epoch": 0.2554557124518614, + "grad_norm": 0.3328215777873993, + "learning_rate": 4.9867189450918294e-05, + "loss": 2.4475, + "step": 199 + }, + { + "epoch": 0.25673940949935814, + "grad_norm": 0.3271300494670868, + "learning_rate": 4.986449529743314e-05, + "loss": 2.4156, + "step": 200 + }, + { + "epoch": 0.25802310654685495, + "grad_norm": 0.296889990568161, + "learning_rate": 4.9861774165517536e-05, + "loss": 2.4117, + "step": 201 + }, + { + "epoch": 0.2593068035943517, + "grad_norm": 0.31897205114364624, + "learning_rate": 4.9859026058123925e-05, + "loss": 2.4637, + "step": 202 + }, + { + "epoch": 0.2605905006418485, + "grad_norm": 0.320688933134079, + "learning_rate": 4.985625097823408e-05, + "loss": 2.3865, + "step": 203 + }, + { + "epoch": 0.26187419768934533, + "grad_norm": 0.29705896973609924, + "learning_rate": 4.985344892885899e-05, + "loss": 2.365, + "step": 204 + }, + { + "epoch": 0.2631578947368421, + "grad_norm": 0.32413363456726074, + "learning_rate": 4.985061991303895e-05, + "loss": 2.4379, + "step": 205 + }, + { + "epoch": 0.2644415917843389, + "grad_norm": 0.38111138343811035, + "learning_rate": 4.984776393384348e-05, + "loss": 2.4303, + "step": 206 + }, + { + "epoch": 0.26572528883183566, + "grad_norm": 0.34943342208862305, + "learning_rate": 4.984488099437138e-05, + "loss": 2.3508, + "step": 207 + }, + { + "epoch": 0.26700898587933247, + "grad_norm": 0.34694743156433105, + "learning_rate": 4.984197109775068e-05, + "loss": 2.4488, + "step": 208 + }, + { + "epoch": 0.2682926829268293, + "grad_norm": 0.3251892626285553, + "learning_rate": 4.983903424713868e-05, + "loss": 2.4187, + "step": 209 + }, + { + "epoch": 0.26957637997432604, + "grad_norm": 0.34391969442367554, + "learning_rate": 4.9836070445721924e-05, + "loss": 2.4723, + "step": 210 + }, + { + "epoch": 0.27086007702182285, + "grad_norm": 0.32182154059410095, + "learning_rate": 4.983307969671617e-05, + "loss": 2.4282, + "step": 211 + }, + { + "epoch": 0.27214377406931967, + "grad_norm": 0.3486088514328003, + "learning_rate": 4.983006200336645e-05, + "loss": 2.4411, + "step": 212 + }, + { + "epoch": 0.2734274711168164, + "grad_norm": 0.3152271807193756, + "learning_rate": 4.9827017368947e-05, + "loss": 2.3362, + "step": 213 + }, + { + "epoch": 0.27471116816431324, + "grad_norm": 0.30361658334732056, + "learning_rate": 4.982394579676133e-05, + "loss": 2.34, + "step": 214 + }, + { + "epoch": 0.27599486521181, + "grad_norm": 0.3163071274757385, + "learning_rate": 4.9820847290142135e-05, + "loss": 2.3996, + "step": 215 + }, + { + "epoch": 0.2772785622593068, + "grad_norm": 0.319831907749176, + "learning_rate": 4.981772185245135e-05, + "loss": 2.352, + "step": 216 + }, + { + "epoch": 0.2785622593068036, + "grad_norm": 0.34143316745758057, + "learning_rate": 4.981456948708014e-05, + "loss": 2.2837, + "step": 217 + }, + { + "epoch": 0.2798459563543004, + "grad_norm": 0.3066380023956299, + "learning_rate": 4.981139019744887e-05, + "loss": 2.3848, + "step": 218 + }, + { + "epoch": 0.2811296534017972, + "grad_norm": 0.32719743251800537, + "learning_rate": 4.9808183987007136e-05, + "loss": 2.3845, + "step": 219 + }, + { + "epoch": 0.28241335044929394, + "grad_norm": 0.384798526763916, + "learning_rate": 4.980495085923372e-05, + "loss": 2.3767, + "step": 220 + }, + { + "epoch": 0.28369704749679076, + "grad_norm": 0.4263782799243927, + "learning_rate": 4.980169081763665e-05, + "loss": 2.3342, + "step": 221 + }, + { + "epoch": 0.28498074454428757, + "grad_norm": 0.3271712064743042, + "learning_rate": 4.979840386575311e-05, + "loss": 2.4539, + "step": 222 + }, + { + "epoch": 0.2862644415917843, + "grad_norm": 0.35880088806152344, + "learning_rate": 4.97950900071495e-05, + "loss": 2.3488, + "step": 223 + }, + { + "epoch": 0.28754813863928114, + "grad_norm": 0.30669230222702026, + "learning_rate": 4.9791749245421434e-05, + "loss": 2.4084, + "step": 224 + }, + { + "epoch": 0.2888318356867779, + "grad_norm": 0.34398922324180603, + "learning_rate": 4.9788381584193684e-05, + "loss": 2.3226, + "step": 225 + }, + { + "epoch": 0.2901155327342747, + "grad_norm": 0.3699093163013458, + "learning_rate": 4.9784987027120236e-05, + "loss": 2.3608, + "step": 226 + }, + { + "epoch": 0.2913992297817715, + "grad_norm": 0.3445926010608673, + "learning_rate": 4.978156557788424e-05, + "loss": 2.3771, + "step": 227 + }, + { + "epoch": 0.2926829268292683, + "grad_norm": 0.33999142050743103, + "learning_rate": 4.977811724019802e-05, + "loss": 2.3541, + "step": 228 + }, + { + "epoch": 0.2939666238767651, + "grad_norm": 0.3054107427597046, + "learning_rate": 4.9774642017803106e-05, + "loss": 2.4764, + "step": 229 + }, + { + "epoch": 0.2952503209242619, + "grad_norm": 0.3072315454483032, + "learning_rate": 4.977113991447017e-05, + "loss": 2.3055, + "step": 230 + }, + { + "epoch": 0.29653401797175866, + "grad_norm": 0.3254982531070709, + "learning_rate": 4.9767610933999055e-05, + "loss": 2.4659, + "step": 231 + }, + { + "epoch": 0.29781771501925547, + "grad_norm": 0.2930049002170563, + "learning_rate": 4.976405508021877e-05, + "loss": 2.3313, + "step": 232 + }, + { + "epoch": 0.2991014120667522, + "grad_norm": 0.30119720101356506, + "learning_rate": 4.976047235698747e-05, + "loss": 2.2981, + "step": 233 + }, + { + "epoch": 0.30038510911424904, + "grad_norm": 0.30038225650787354, + "learning_rate": 4.9756862768192504e-05, + "loss": 2.2864, + "step": 234 + }, + { + "epoch": 0.30166880616174585, + "grad_norm": 0.29573819041252136, + "learning_rate": 4.975322631775032e-05, + "loss": 2.3508, + "step": 235 + }, + { + "epoch": 0.3029525032092426, + "grad_norm": 0.316169798374176, + "learning_rate": 4.9749563009606534e-05, + "loss": 2.3648, + "step": 236 + }, + { + "epoch": 0.3042362002567394, + "grad_norm": 0.3109273910522461, + "learning_rate": 4.9745872847735894e-05, + "loss": 2.5141, + "step": 237 + }, + { + "epoch": 0.3055198973042362, + "grad_norm": 0.34613820910453796, + "learning_rate": 4.974215583614232e-05, + "loss": 2.333, + "step": 238 + }, + { + "epoch": 0.306803594351733, + "grad_norm": 0.34924328327178955, + "learning_rate": 4.9738411978858814e-05, + "loss": 2.4158, + "step": 239 + }, + { + "epoch": 0.3080872913992298, + "grad_norm": 0.2984835207462311, + "learning_rate": 4.9734641279947535e-05, + "loss": 2.3817, + "step": 240 + }, + { + "epoch": 0.30937098844672656, + "grad_norm": 0.3265876770019531, + "learning_rate": 4.9730843743499764e-05, + "loss": 2.4202, + "step": 241 + }, + { + "epoch": 0.31065468549422337, + "grad_norm": 0.31205374002456665, + "learning_rate": 4.9727019373635895e-05, + "loss": 2.3492, + "step": 242 + }, + { + "epoch": 0.3119383825417201, + "grad_norm": 0.3619811534881592, + "learning_rate": 4.972316817450544e-05, + "loss": 2.481, + "step": 243 + }, + { + "epoch": 0.31322207958921694, + "grad_norm": 0.4126303791999817, + "learning_rate": 4.9719290150287026e-05, + "loss": 2.2388, + "step": 244 + }, + { + "epoch": 0.31450577663671375, + "grad_norm": 0.34099090099334717, + "learning_rate": 4.971538530518836e-05, + "loss": 2.4, + "step": 245 + }, + { + "epoch": 0.3157894736842105, + "grad_norm": 0.32491570711135864, + "learning_rate": 4.971145364344628e-05, + "loss": 2.3854, + "step": 246 + }, + { + "epoch": 0.3170731707317073, + "grad_norm": 0.340457022190094, + "learning_rate": 4.970749516932672e-05, + "loss": 2.4011, + "step": 247 + }, + { + "epoch": 0.31835686777920413, + "grad_norm": 0.33701422810554504, + "learning_rate": 4.97035098871247e-05, + "loss": 2.4188, + "step": 248 + }, + { + "epoch": 0.3196405648267009, + "grad_norm": 0.3307265341281891, + "learning_rate": 4.969949780116431e-05, + "loss": 2.4553, + "step": 249 + }, + { + "epoch": 0.3209242618741977, + "grad_norm": 0.38966116309165955, + "learning_rate": 4.969545891579873e-05, + "loss": 2.3657, + "step": 250 + }, + { + "epoch": 0.32220795892169446, + "grad_norm": 0.36796143651008606, + "learning_rate": 4.969139323541025e-05, + "loss": 2.3044, + "step": 251 + }, + { + "epoch": 0.32349165596919127, + "grad_norm": 0.32584547996520996, + "learning_rate": 4.968730076441017e-05, + "loss": 2.4064, + "step": 252 + }, + { + "epoch": 0.3247753530166881, + "grad_norm": 0.3360714316368103, + "learning_rate": 4.968318150723893e-05, + "loss": 2.4243, + "step": 253 + }, + { + "epoch": 0.32605905006418484, + "grad_norm": 0.3058249056339264, + "learning_rate": 4.9679035468365986e-05, + "loss": 2.3598, + "step": 254 + }, + { + "epoch": 0.32734274711168165, + "grad_norm": 0.32182779908180237, + "learning_rate": 4.9674862652289865e-05, + "loss": 2.3469, + "step": 255 + }, + { + "epoch": 0.3286264441591784, + "grad_norm": 0.39011824131011963, + "learning_rate": 4.967066306353816e-05, + "loss": 2.4215, + "step": 256 + }, + { + "epoch": 0.3299101412066752, + "grad_norm": 0.3215901553630829, + "learning_rate": 4.966643670666748e-05, + "loss": 2.396, + "step": 257 + }, + { + "epoch": 0.33119383825417203, + "grad_norm": 0.3239218592643738, + "learning_rate": 4.9662183586263514e-05, + "loss": 2.4271, + "step": 258 + }, + { + "epoch": 0.3324775353016688, + "grad_norm": 0.33229637145996094, + "learning_rate": 4.965790370694097e-05, + "loss": 2.4345, + "step": 259 + }, + { + "epoch": 0.3337612323491656, + "grad_norm": 0.36766308546066284, + "learning_rate": 4.9653597073343594e-05, + "loss": 2.4257, + "step": 260 + }, + { + "epoch": 0.33504492939666236, + "grad_norm": 0.33887600898742676, + "learning_rate": 4.964926369014417e-05, + "loss": 2.4391, + "step": 261 + }, + { + "epoch": 0.3363286264441592, + "grad_norm": 0.3224686086177826, + "learning_rate": 4.964490356204449e-05, + "loss": 2.3664, + "step": 262 + }, + { + "epoch": 0.337612323491656, + "grad_norm": 0.3371565341949463, + "learning_rate": 4.964051669377538e-05, + "loss": 2.4992, + "step": 263 + }, + { + "epoch": 0.33889602053915274, + "grad_norm": 0.3373253345489502, + "learning_rate": 4.963610309009665e-05, + "loss": 2.3509, + "step": 264 + }, + { + "epoch": 0.34017971758664955, + "grad_norm": 0.34519249200820923, + "learning_rate": 4.963166275579717e-05, + "loss": 2.4388, + "step": 265 + }, + { + "epoch": 0.34146341463414637, + "grad_norm": 0.3187866806983948, + "learning_rate": 4.9627195695694774e-05, + "loss": 2.4655, + "step": 266 + }, + { + "epoch": 0.3427471116816431, + "grad_norm": 0.31844937801361084, + "learning_rate": 4.962270191463629e-05, + "loss": 2.4097, + "step": 267 + }, + { + "epoch": 0.34403080872913994, + "grad_norm": 0.359625905752182, + "learning_rate": 4.9618181417497566e-05, + "loss": 2.4364, + "step": 268 + }, + { + "epoch": 0.3453145057766367, + "grad_norm": 0.3335789740085602, + "learning_rate": 4.961363420918342e-05, + "loss": 2.2316, + "step": 269 + }, + { + "epoch": 0.3465982028241335, + "grad_norm": 0.3209823668003082, + "learning_rate": 4.960906029462766e-05, + "loss": 2.4698, + "step": 270 + }, + { + "epoch": 0.3478818998716303, + "grad_norm": 0.3264525532722473, + "learning_rate": 4.960445967879307e-05, + "loss": 2.3222, + "step": 271 + }, + { + "epoch": 0.3491655969191271, + "grad_norm": 0.29652708768844604, + "learning_rate": 4.959983236667138e-05, + "loss": 2.3258, + "step": 272 + }, + { + "epoch": 0.3504492939666239, + "grad_norm": 0.35058003664016724, + "learning_rate": 4.959517836328333e-05, + "loss": 2.3009, + "step": 273 + }, + { + "epoch": 0.35173299101412064, + "grad_norm": 0.29893478751182556, + "learning_rate": 4.959049767367859e-05, + "loss": 2.463, + "step": 274 + }, + { + "epoch": 0.35301668806161746, + "grad_norm": 0.3114057183265686, + "learning_rate": 4.95857903029358e-05, + "loss": 2.2757, + "step": 275 + }, + { + "epoch": 0.35430038510911427, + "grad_norm": 0.3837164640426636, + "learning_rate": 4.958105625616253e-05, + "loss": 2.445, + "step": 276 + }, + { + "epoch": 0.355584082156611, + "grad_norm": 0.3195387125015259, + "learning_rate": 4.957629553849532e-05, + "loss": 2.4823, + "step": 277 + }, + { + "epoch": 0.35686777920410784, + "grad_norm": 0.37014302611351013, + "learning_rate": 4.957150815509963e-05, + "loss": 2.3322, + "step": 278 + }, + { + "epoch": 0.3581514762516046, + "grad_norm": 0.28935369849205017, + "learning_rate": 4.9566694111169853e-05, + "loss": 2.4336, + "step": 279 + }, + { + "epoch": 0.3594351732991014, + "grad_norm": 0.3155868351459503, + "learning_rate": 4.956185341192933e-05, + "loss": 2.4691, + "step": 280 + }, + { + "epoch": 0.3607188703465982, + "grad_norm": 0.33008038997650146, + "learning_rate": 4.955698606263028e-05, + "loss": 2.3331, + "step": 281 + }, + { + "epoch": 0.362002567394095, + "grad_norm": 0.3546096086502075, + "learning_rate": 4.95520920685539e-05, + "loss": 2.4626, + "step": 282 + }, + { + "epoch": 0.3632862644415918, + "grad_norm": 0.3569015562534332, + "learning_rate": 4.954717143501024e-05, + "loss": 2.4165, + "step": 283 + }, + { + "epoch": 0.3645699614890886, + "grad_norm": 0.2979843318462372, + "learning_rate": 4.954222416733829e-05, + "loss": 2.3322, + "step": 284 + }, + { + "epoch": 0.36585365853658536, + "grad_norm": 0.43857890367507935, + "learning_rate": 4.953725027090591e-05, + "loss": 2.3168, + "step": 285 + }, + { + "epoch": 0.36713735558408217, + "grad_norm": 0.32624951004981995, + "learning_rate": 4.953224975110988e-05, + "loss": 2.3685, + "step": 286 + }, + { + "epoch": 0.3684210526315789, + "grad_norm": 0.3195008933544159, + "learning_rate": 4.9527222613375855e-05, + "loss": 2.4137, + "step": 287 + }, + { + "epoch": 0.36970474967907574, + "grad_norm": 0.362567663192749, + "learning_rate": 4.952216886315837e-05, + "loss": 2.2978, + "step": 288 + }, + { + "epoch": 0.37098844672657255, + "grad_norm": 0.31306540966033936, + "learning_rate": 4.951708850594083e-05, + "loss": 2.4381, + "step": 289 + }, + { + "epoch": 0.3722721437740693, + "grad_norm": 0.3275150954723358, + "learning_rate": 4.951198154723552e-05, + "loss": 2.3528, + "step": 290 + }, + { + "epoch": 0.3735558408215661, + "grad_norm": 0.32541584968566895, + "learning_rate": 4.9506847992583586e-05, + "loss": 2.5023, + "step": 291 + }, + { + "epoch": 0.3748395378690629, + "grad_norm": 0.3278298079967499, + "learning_rate": 4.9501687847555016e-05, + "loss": 2.3745, + "step": 292 + }, + { + "epoch": 0.3761232349165597, + "grad_norm": 0.3163221776485443, + "learning_rate": 4.949650111774868e-05, + "loss": 2.3451, + "step": 293 + }, + { + "epoch": 0.3774069319640565, + "grad_norm": 0.30997052788734436, + "learning_rate": 4.9491287808792265e-05, + "loss": 2.3976, + "step": 294 + }, + { + "epoch": 0.37869062901155326, + "grad_norm": 0.3496086597442627, + "learning_rate": 4.948604792634229e-05, + "loss": 2.4361, + "step": 295 + }, + { + "epoch": 0.37997432605905007, + "grad_norm": 0.3483693599700928, + "learning_rate": 4.948078147608416e-05, + "loss": 2.4022, + "step": 296 + }, + { + "epoch": 0.38125802310654683, + "grad_norm": 0.3135087490081787, + "learning_rate": 4.947548846373204e-05, + "loss": 2.3585, + "step": 297 + }, + { + "epoch": 0.38254172015404364, + "grad_norm": 0.323482871055603, + "learning_rate": 4.947016889502895e-05, + "loss": 2.3958, + "step": 298 + }, + { + "epoch": 0.38382541720154045, + "grad_norm": 0.3020581901073456, + "learning_rate": 4.946482277574673e-05, + "loss": 2.2809, + "step": 299 + }, + { + "epoch": 0.3851091142490372, + "grad_norm": 0.31288352608680725, + "learning_rate": 4.9459450111686e-05, + "loss": 2.3526, + "step": 300 + }, + { + "epoch": 0.386392811296534, + "grad_norm": 0.31495004892349243, + "learning_rate": 4.945405090867621e-05, + "loss": 2.4462, + "step": 301 + }, + { + "epoch": 0.38767650834403083, + "grad_norm": 0.3513999581336975, + "learning_rate": 4.94486251725756e-05, + "loss": 2.3795, + "step": 302 + }, + { + "epoch": 0.3889602053915276, + "grad_norm": 0.33559510111808777, + "learning_rate": 4.944317290927117e-05, + "loss": 2.3002, + "step": 303 + }, + { + "epoch": 0.3902439024390244, + "grad_norm": 0.3153408467769623, + "learning_rate": 4.943769412467875e-05, + "loss": 2.437, + "step": 304 + }, + { + "epoch": 0.39152759948652116, + "grad_norm": 0.3177226781845093, + "learning_rate": 4.943218882474291e-05, + "loss": 2.3447, + "step": 305 + }, + { + "epoch": 0.392811296534018, + "grad_norm": 0.3815303146839142, + "learning_rate": 4.9426657015436994e-05, + "loss": 2.3584, + "step": 306 + }, + { + "epoch": 0.3940949935815148, + "grad_norm": 0.34336403012275696, + "learning_rate": 4.9421098702763126e-05, + "loss": 2.2979, + "step": 307 + }, + { + "epoch": 0.39537869062901154, + "grad_norm": 0.33281055092811584, + "learning_rate": 4.941551389275217e-05, + "loss": 2.4152, + "step": 308 + }, + { + "epoch": 0.39666238767650835, + "grad_norm": 0.3309512138366699, + "learning_rate": 4.9409902591463756e-05, + "loss": 2.361, + "step": 309 + }, + { + "epoch": 0.3979460847240051, + "grad_norm": 0.3044840097427368, + "learning_rate": 4.940426480498623e-05, + "loss": 2.3172, + "step": 310 + }, + { + "epoch": 0.3992297817715019, + "grad_norm": 0.32050418853759766, + "learning_rate": 4.939860053943671e-05, + "loss": 2.374, + "step": 311 + }, + { + "epoch": 0.40051347881899874, + "grad_norm": 0.36678293347358704, + "learning_rate": 4.939290980096103e-05, + "loss": 2.3529, + "step": 312 + }, + { + "epoch": 0.4017971758664955, + "grad_norm": 0.32348355650901794, + "learning_rate": 4.9387192595733734e-05, + "loss": 2.4048, + "step": 313 + }, + { + "epoch": 0.4030808729139923, + "grad_norm": 0.3086166977882385, + "learning_rate": 4.938144892995809e-05, + "loss": 2.4091, + "step": 314 + }, + { + "epoch": 0.40436456996148906, + "grad_norm": 0.31869277358055115, + "learning_rate": 4.937567880986609e-05, + "loss": 2.3594, + "step": 315 + }, + { + "epoch": 0.4056482670089859, + "grad_norm": 0.317147821187973, + "learning_rate": 4.936988224171842e-05, + "loss": 2.2666, + "step": 316 + }, + { + "epoch": 0.4069319640564827, + "grad_norm": 0.32587262988090515, + "learning_rate": 4.936405923180446e-05, + "loss": 2.317, + "step": 317 + }, + { + "epoch": 0.40821566110397944, + "grad_norm": 0.3181535005569458, + "learning_rate": 4.935820978644228e-05, + "loss": 2.3611, + "step": 318 + }, + { + "epoch": 0.40949935815147626, + "grad_norm": 0.31699231266975403, + "learning_rate": 4.9352333911978625e-05, + "loss": 2.4676, + "step": 319 + }, + { + "epoch": 0.41078305519897307, + "grad_norm": 0.35386183857917786, + "learning_rate": 4.9346431614788945e-05, + "loss": 2.3525, + "step": 320 + }, + { + "epoch": 0.4120667522464698, + "grad_norm": 0.3134053647518158, + "learning_rate": 4.934050290127733e-05, + "loss": 2.4045, + "step": 321 + }, + { + "epoch": 0.41335044929396664, + "grad_norm": 0.3209103047847748, + "learning_rate": 4.933454777787654e-05, + "loss": 2.3606, + "step": 322 + }, + { + "epoch": 0.4146341463414634, + "grad_norm": 0.3191809356212616, + "learning_rate": 4.9328566251048e-05, + "loss": 2.4551, + "step": 323 + }, + { + "epoch": 0.4159178433889602, + "grad_norm": 0.2983877658843994, + "learning_rate": 4.9322558327281773e-05, + "loss": 2.3299, + "step": 324 + }, + { + "epoch": 0.417201540436457, + "grad_norm": 0.3376917541027069, + "learning_rate": 4.931652401309655e-05, + "loss": 2.3599, + "step": 325 + }, + { + "epoch": 0.4184852374839538, + "grad_norm": 0.36222752928733826, + "learning_rate": 4.93104633150397e-05, + "loss": 2.4432, + "step": 326 + }, + { + "epoch": 0.4197689345314506, + "grad_norm": 0.2962850034236908, + "learning_rate": 4.930437623968718e-05, + "loss": 2.3965, + "step": 327 + }, + { + "epoch": 0.42105263157894735, + "grad_norm": 0.32738175988197327, + "learning_rate": 4.929826279364357e-05, + "loss": 2.3276, + "step": 328 + }, + { + "epoch": 0.42233632862644416, + "grad_norm": 0.3119535446166992, + "learning_rate": 4.929212298354207e-05, + "loss": 2.3639, + "step": 329 + }, + { + "epoch": 0.42362002567394097, + "grad_norm": 0.3193548321723938, + "learning_rate": 4.9285956816044486e-05, + "loss": 2.4468, + "step": 330 + }, + { + "epoch": 0.4249037227214377, + "grad_norm": 0.32102808356285095, + "learning_rate": 4.927976429784124e-05, + "loss": 2.422, + "step": 331 + }, + { + "epoch": 0.42618741976893454, + "grad_norm": 0.3115790784358978, + "learning_rate": 4.92735454356513e-05, + "loss": 2.2948, + "step": 332 + }, + { + "epoch": 0.4274711168164313, + "grad_norm": 0.320351243019104, + "learning_rate": 4.926730023622227e-05, + "loss": 2.3466, + "step": 333 + }, + { + "epoch": 0.4287548138639281, + "grad_norm": 0.30213356018066406, + "learning_rate": 4.926102870633029e-05, + "loss": 2.3773, + "step": 334 + }, + { + "epoch": 0.4300385109114249, + "grad_norm": 0.3080739676952362, + "learning_rate": 4.925473085278008e-05, + "loss": 2.3822, + "step": 335 + }, + { + "epoch": 0.4313222079589217, + "grad_norm": 0.3327421545982361, + "learning_rate": 4.924840668240495e-05, + "loss": 2.4326, + "step": 336 + }, + { + "epoch": 0.4326059050064185, + "grad_norm": 0.34913370013237, + "learning_rate": 4.924205620206671e-05, + "loss": 2.4491, + "step": 337 + }, + { + "epoch": 0.4338896020539153, + "grad_norm": 0.3061310946941376, + "learning_rate": 4.923567941865577e-05, + "loss": 2.3727, + "step": 338 + }, + { + "epoch": 0.43517329910141206, + "grad_norm": 0.3259632885456085, + "learning_rate": 4.9229276339091034e-05, + "loss": 2.3775, + "step": 339 + }, + { + "epoch": 0.43645699614890887, + "grad_norm": 0.29085099697113037, + "learning_rate": 4.922284697031999e-05, + "loss": 2.3427, + "step": 340 + }, + { + "epoch": 0.43774069319640563, + "grad_norm": 0.30840590596199036, + "learning_rate": 4.921639131931859e-05, + "loss": 2.367, + "step": 341 + }, + { + "epoch": 0.43902439024390244, + "grad_norm": 0.32114389538764954, + "learning_rate": 4.920990939309135e-05, + "loss": 2.4625, + "step": 342 + }, + { + "epoch": 0.44030808729139925, + "grad_norm": 0.3091549873352051, + "learning_rate": 4.920340119867127e-05, + "loss": 2.3571, + "step": 343 + }, + { + "epoch": 0.441591784338896, + "grad_norm": 0.2978641092777252, + "learning_rate": 4.919686674311987e-05, + "loss": 2.3033, + "step": 344 + }, + { + "epoch": 0.4428754813863928, + "grad_norm": 0.33106520771980286, + "learning_rate": 4.919030603352715e-05, + "loss": 2.3276, + "step": 345 + }, + { + "epoch": 0.4441591784338896, + "grad_norm": 0.42283540964126587, + "learning_rate": 4.918371907701159e-05, + "loss": 2.3563, + "step": 346 + }, + { + "epoch": 0.4454428754813864, + "grad_norm": 0.32727953791618347, + "learning_rate": 4.9177105880720173e-05, + "loss": 2.3928, + "step": 347 + }, + { + "epoch": 0.4467265725288832, + "grad_norm": 0.3086586892604828, + "learning_rate": 4.9170466451828326e-05, + "loss": 2.2707, + "step": 348 + }, + { + "epoch": 0.44801026957637996, + "grad_norm": 0.28993648290634155, + "learning_rate": 4.916380079753995e-05, + "loss": 2.3582, + "step": 349 + }, + { + "epoch": 0.4492939666238768, + "grad_norm": 0.3154331147670746, + "learning_rate": 4.9157108925087405e-05, + "loss": 2.4003, + "step": 350 + }, + { + "epoch": 0.45057766367137353, + "grad_norm": 0.3312948942184448, + "learning_rate": 4.9150390841731485e-05, + "loss": 2.3955, + "step": 351 + }, + { + "epoch": 0.45186136071887034, + "grad_norm": 0.30381128191947937, + "learning_rate": 4.914364655476146e-05, + "loss": 2.3737, + "step": 352 + }, + { + "epoch": 0.45314505776636715, + "grad_norm": 0.30618736147880554, + "learning_rate": 4.9136876071494976e-05, + "loss": 2.3602, + "step": 353 + }, + { + "epoch": 0.4544287548138639, + "grad_norm": 0.3091343343257904, + "learning_rate": 4.913007939927814e-05, + "loss": 2.4564, + "step": 354 + }, + { + "epoch": 0.4557124518613607, + "grad_norm": 0.311162531375885, + "learning_rate": 4.912325654548546e-05, + "loss": 2.3133, + "step": 355 + }, + { + "epoch": 0.45699614890885754, + "grad_norm": 0.33965715765953064, + "learning_rate": 4.911640751751988e-05, + "loss": 2.4312, + "step": 356 + }, + { + "epoch": 0.4582798459563543, + "grad_norm": 0.3162749111652374, + "learning_rate": 4.910953232281269e-05, + "loss": 2.3438, + "step": 357 + }, + { + "epoch": 0.4595635430038511, + "grad_norm": 0.32984068989753723, + "learning_rate": 4.910263096882362e-05, + "loss": 2.3823, + "step": 358 + }, + { + "epoch": 0.46084724005134786, + "grad_norm": 0.30816513299942017, + "learning_rate": 4.909570346304076e-05, + "loss": 2.445, + "step": 359 + }, + { + "epoch": 0.4621309370988447, + "grad_norm": 0.2928940951824188, + "learning_rate": 4.908874981298057e-05, + "loss": 2.4538, + "step": 360 + }, + { + "epoch": 0.4634146341463415, + "grad_norm": 0.32563963532447815, + "learning_rate": 4.9081770026187914e-05, + "loss": 2.292, + "step": 361 + }, + { + "epoch": 0.46469833119383824, + "grad_norm": 0.29584982991218567, + "learning_rate": 4.907476411023596e-05, + "loss": 2.4046, + "step": 362 + }, + { + "epoch": 0.46598202824133506, + "grad_norm": 0.3291500210762024, + "learning_rate": 4.906773207272626e-05, + "loss": 2.3186, + "step": 363 + }, + { + "epoch": 0.4672657252888318, + "grad_norm": 0.3066290020942688, + "learning_rate": 4.9060673921288716e-05, + "loss": 2.3266, + "step": 364 + }, + { + "epoch": 0.4685494223363286, + "grad_norm": 0.3005695939064026, + "learning_rate": 4.905358966358153e-05, + "loss": 2.3712, + "step": 365 + }, + { + "epoch": 0.46983311938382544, + "grad_norm": 0.32391899824142456, + "learning_rate": 4.904647930729128e-05, + "loss": 2.3514, + "step": 366 + }, + { + "epoch": 0.4711168164313222, + "grad_norm": 0.2915700078010559, + "learning_rate": 4.903934286013281e-05, + "loss": 2.4646, + "step": 367 + }, + { + "epoch": 0.472400513478819, + "grad_norm": 0.34920045733451843, + "learning_rate": 4.90321803298493e-05, + "loss": 2.3287, + "step": 368 + }, + { + "epoch": 0.47368421052631576, + "grad_norm": 0.2955242395401001, + "learning_rate": 4.902499172421222e-05, + "loss": 2.3654, + "step": 369 + }, + { + "epoch": 0.4749679075738126, + "grad_norm": 0.3305751085281372, + "learning_rate": 4.901777705102135e-05, + "loss": 2.4419, + "step": 370 + }, + { + "epoch": 0.4762516046213094, + "grad_norm": 0.30396509170532227, + "learning_rate": 4.9010536318104734e-05, + "loss": 2.3576, + "step": 371 + }, + { + "epoch": 0.47753530166880614, + "grad_norm": 0.2961418330669403, + "learning_rate": 4.9003269533318704e-05, + "loss": 2.3039, + "step": 372 + }, + { + "epoch": 0.47881899871630296, + "grad_norm": 0.35183992981910706, + "learning_rate": 4.899597670454785e-05, + "loss": 2.3333, + "step": 373 + }, + { + "epoch": 0.48010269576379977, + "grad_norm": 0.297158420085907, + "learning_rate": 4.8988657839705024e-05, + "loss": 2.3514, + "step": 374 + }, + { + "epoch": 0.4813863928112965, + "grad_norm": 0.3379060626029968, + "learning_rate": 4.8981312946731325e-05, + "loss": 2.2811, + "step": 375 + }, + { + "epoch": 0.48267008985879334, + "grad_norm": 0.34060969948768616, + "learning_rate": 4.897394203359611e-05, + "loss": 2.4123, + "step": 376 + }, + { + "epoch": 0.4839537869062901, + "grad_norm": 0.3308861255645752, + "learning_rate": 4.896654510829694e-05, + "loss": 2.4262, + "step": 377 + }, + { + "epoch": 0.4852374839537869, + "grad_norm": 0.3043263852596283, + "learning_rate": 4.8959122178859616e-05, + "loss": 2.4291, + "step": 378 + }, + { + "epoch": 0.4865211810012837, + "grad_norm": 0.32594412565231323, + "learning_rate": 4.8951673253338156e-05, + "loss": 2.3679, + "step": 379 + }, + { + "epoch": 0.4878048780487805, + "grad_norm": 0.3090050220489502, + "learning_rate": 4.894419833981478e-05, + "loss": 2.3933, + "step": 380 + }, + { + "epoch": 0.4890885750962773, + "grad_norm": 0.3234407901763916, + "learning_rate": 4.8936697446399896e-05, + "loss": 2.3156, + "step": 381 + }, + { + "epoch": 0.49037227214377405, + "grad_norm": 0.3294239938259125, + "learning_rate": 4.892917058123212e-05, + "loss": 2.357, + "step": 382 + }, + { + "epoch": 0.49165596919127086, + "grad_norm": 0.33215636014938354, + "learning_rate": 4.8921617752478235e-05, + "loss": 2.4386, + "step": 383 + }, + { + "epoch": 0.49293966623876767, + "grad_norm": 0.30511781573295593, + "learning_rate": 4.89140389683332e-05, + "loss": 2.4283, + "step": 384 + }, + { + "epoch": 0.4942233632862644, + "grad_norm": 0.31070634722709656, + "learning_rate": 4.890643423702013e-05, + "loss": 2.3565, + "step": 385 + }, + { + "epoch": 0.49550706033376124, + "grad_norm": 0.3108975291252136, + "learning_rate": 4.8898803566790296e-05, + "loss": 2.3209, + "step": 386 + }, + { + "epoch": 0.496790757381258, + "grad_norm": 0.31348085403442383, + "learning_rate": 4.889114696592312e-05, + "loss": 2.3707, + "step": 387 + }, + { + "epoch": 0.4980744544287548, + "grad_norm": 0.3120376467704773, + "learning_rate": 4.8883464442726146e-05, + "loss": 2.3743, + "step": 388 + }, + { + "epoch": 0.4993581514762516, + "grad_norm": 0.3120869994163513, + "learning_rate": 4.887575600553506e-05, + "loss": 2.3501, + "step": 389 + }, + { + "epoch": 0.5006418485237484, + "grad_norm": 0.28657200932502747, + "learning_rate": 4.886802166271364e-05, + "loss": 2.365, + "step": 390 + }, + { + "epoch": 0.5019255455712451, + "grad_norm": 0.3038496971130371, + "learning_rate": 4.886026142265381e-05, + "loss": 2.3851, + "step": 391 + }, + { + "epoch": 0.503209242618742, + "grad_norm": 0.3245856761932373, + "learning_rate": 4.885247529377557e-05, + "loss": 2.3996, + "step": 392 + }, + { + "epoch": 0.5044929396662388, + "grad_norm": 0.2905268669128418, + "learning_rate": 4.8844663284526995e-05, + "loss": 2.4106, + "step": 393 + }, + { + "epoch": 0.5057766367137355, + "grad_norm": 0.3151610195636749, + "learning_rate": 4.883682540338428e-05, + "loss": 2.3665, + "step": 394 + }, + { + "epoch": 0.5070603337612324, + "grad_norm": 0.3045148253440857, + "learning_rate": 4.8828961658851645e-05, + "loss": 2.3244, + "step": 395 + }, + { + "epoch": 0.5083440308087291, + "grad_norm": 0.33118733763694763, + "learning_rate": 4.882107205946142e-05, + "loss": 2.3622, + "step": 396 + }, + { + "epoch": 0.5096277278562259, + "grad_norm": 0.3078731596469879, + "learning_rate": 4.881315661377393e-05, + "loss": 2.325, + "step": 397 + }, + { + "epoch": 0.5109114249037228, + "grad_norm": 0.33527669310569763, + "learning_rate": 4.880521533037762e-05, + "loss": 2.4234, + "step": 398 + }, + { + "epoch": 0.5121951219512195, + "grad_norm": 0.32107025384902954, + "learning_rate": 4.879724821788889e-05, + "loss": 2.2833, + "step": 399 + }, + { + "epoch": 0.5134788189987163, + "grad_norm": 0.31819841265678406, + "learning_rate": 4.878925528495223e-05, + "loss": 2.4178, + "step": 400 + }, + { + "epoch": 0.5147625160462131, + "grad_norm": 0.31981033086776733, + "learning_rate": 4.8781236540240106e-05, + "loss": 2.4154, + "step": 401 + }, + { + "epoch": 0.5160462130937099, + "grad_norm": 0.39689508080482483, + "learning_rate": 4.8773191992453e-05, + "loss": 2.3759, + "step": 402 + }, + { + "epoch": 0.5173299101412067, + "grad_norm": 0.3065798282623291, + "learning_rate": 4.876512165031939e-05, + "loss": 2.4322, + "step": 403 + }, + { + "epoch": 0.5186136071887034, + "grad_norm": 0.3112392723560333, + "learning_rate": 4.8757025522595756e-05, + "loss": 2.3008, + "step": 404 + }, + { + "epoch": 0.5198973042362003, + "grad_norm": 0.3170951008796692, + "learning_rate": 4.874890361806654e-05, + "loss": 2.3265, + "step": 405 + }, + { + "epoch": 0.521181001283697, + "grad_norm": 0.32137608528137207, + "learning_rate": 4.8740755945544156e-05, + "loss": 2.4448, + "step": 406 + }, + { + "epoch": 0.5224646983311938, + "grad_norm": 0.32462355494499207, + "learning_rate": 4.873258251386897e-05, + "loss": 2.3504, + "step": 407 + }, + { + "epoch": 0.5237483953786907, + "grad_norm": 0.32057517766952515, + "learning_rate": 4.872438333190931e-05, + "loss": 2.4223, + "step": 408 + }, + { + "epoch": 0.5250320924261874, + "grad_norm": 0.3375895917415619, + "learning_rate": 4.871615840856144e-05, + "loss": 2.3912, + "step": 409 + }, + { + "epoch": 0.5263157894736842, + "grad_norm": 0.29747098684310913, + "learning_rate": 4.870790775274955e-05, + "loss": 2.3684, + "step": 410 + }, + { + "epoch": 0.527599486521181, + "grad_norm": 0.31433284282684326, + "learning_rate": 4.8699631373425746e-05, + "loss": 2.4619, + "step": 411 + }, + { + "epoch": 0.5288831835686778, + "grad_norm": 0.31316566467285156, + "learning_rate": 4.869132927957007e-05, + "loss": 2.45, + "step": 412 + }, + { + "epoch": 0.5301668806161746, + "grad_norm": 0.29028958082199097, + "learning_rate": 4.8683001480190415e-05, + "loss": 2.3539, + "step": 413 + }, + { + "epoch": 0.5314505776636713, + "grad_norm": 0.32501766085624695, + "learning_rate": 4.867464798432262e-05, + "loss": 2.4104, + "step": 414 + }, + { + "epoch": 0.5327342747111682, + "grad_norm": 0.31315121054649353, + "learning_rate": 4.866626880103036e-05, + "loss": 2.4346, + "step": 415 + }, + { + "epoch": 0.5340179717586649, + "grad_norm": 0.30741140246391296, + "learning_rate": 4.865786393940522e-05, + "loss": 2.3253, + "step": 416 + }, + { + "epoch": 0.5353016688061617, + "grad_norm": 0.3042595684528351, + "learning_rate": 4.8649433408566626e-05, + "loss": 2.4107, + "step": 417 + }, + { + "epoch": 0.5365853658536586, + "grad_norm": 0.3254822790622711, + "learning_rate": 4.864097721766184e-05, + "loss": 2.368, + "step": 418 + }, + { + "epoch": 0.5378690629011553, + "grad_norm": 0.3164546489715576, + "learning_rate": 4.8632495375866004e-05, + "loss": 2.3493, + "step": 419 + }, + { + "epoch": 0.5391527599486521, + "grad_norm": 0.30588898062705994, + "learning_rate": 4.8623987892382066e-05, + "loss": 2.3751, + "step": 420 + }, + { + "epoch": 0.540436456996149, + "grad_norm": 0.3062260150909424, + "learning_rate": 4.861545477644079e-05, + "loss": 2.3137, + "step": 421 + }, + { + "epoch": 0.5417201540436457, + "grad_norm": 0.3363398611545563, + "learning_rate": 4.8606896037300776e-05, + "loss": 2.3326, + "step": 422 + }, + { + "epoch": 0.5430038510911425, + "grad_norm": 0.33340004086494446, + "learning_rate": 4.8598311684248396e-05, + "loss": 2.3511, + "step": 423 + }, + { + "epoch": 0.5442875481386393, + "grad_norm": 0.2808108329772949, + "learning_rate": 4.858970172659785e-05, + "loss": 2.4589, + "step": 424 + }, + { + "epoch": 0.5455712451861361, + "grad_norm": 0.299396276473999, + "learning_rate": 4.8581066173691074e-05, + "loss": 2.3481, + "step": 425 + }, + { + "epoch": 0.5468549422336328, + "grad_norm": 0.3083891272544861, + "learning_rate": 4.857240503489782e-05, + "loss": 2.2812, + "step": 426 + }, + { + "epoch": 0.5481386392811296, + "grad_norm": 0.3224524259567261, + "learning_rate": 4.8563718319615584e-05, + "loss": 2.2938, + "step": 427 + }, + { + "epoch": 0.5494223363286265, + "grad_norm": 0.2835102379322052, + "learning_rate": 4.8555006037269594e-05, + "loss": 2.3401, + "step": 428 + }, + { + "epoch": 0.5507060333761232, + "grad_norm": 0.28240859508514404, + "learning_rate": 4.854626819731284e-05, + "loss": 2.4238, + "step": 429 + }, + { + "epoch": 0.55198973042362, + "grad_norm": 0.28541532158851624, + "learning_rate": 4.853750480922604e-05, + "loss": 2.36, + "step": 430 + }, + { + "epoch": 0.5532734274711169, + "grad_norm": 0.3530977666378021, + "learning_rate": 4.852871588251763e-05, + "loss": 2.3936, + "step": 431 + }, + { + "epoch": 0.5545571245186136, + "grad_norm": 0.3233661949634552, + "learning_rate": 4.851990142672376e-05, + "loss": 2.3921, + "step": 432 + }, + { + "epoch": 0.5558408215661104, + "grad_norm": 0.34146299958229065, + "learning_rate": 4.851106145140827e-05, + "loss": 2.4632, + "step": 433 + }, + { + "epoch": 0.5571245186136072, + "grad_norm": 0.3273311257362366, + "learning_rate": 4.8502195966162694e-05, + "loss": 2.3328, + "step": 434 + }, + { + "epoch": 0.558408215661104, + "grad_norm": 0.29606807231903076, + "learning_rate": 4.8493304980606245e-05, + "loss": 2.3077, + "step": 435 + }, + { + "epoch": 0.5596919127086007, + "grad_norm": 0.2943183481693268, + "learning_rate": 4.8484388504385806e-05, + "loss": 2.3786, + "step": 436 + }, + { + "epoch": 0.5609756097560976, + "grad_norm": 0.29871752858161926, + "learning_rate": 4.847544654717592e-05, + "loss": 2.3262, + "step": 437 + }, + { + "epoch": 0.5622593068035944, + "grad_norm": 0.3182483911514282, + "learning_rate": 4.8466479118678766e-05, + "loss": 2.3529, + "step": 438 + }, + { + "epoch": 0.5635430038510911, + "grad_norm": 0.34097424149513245, + "learning_rate": 4.845748622862417e-05, + "loss": 2.3279, + "step": 439 + }, + { + "epoch": 0.5648267008985879, + "grad_norm": 0.295391321182251, + "learning_rate": 4.8448467886769585e-05, + "loss": 2.3059, + "step": 440 + }, + { + "epoch": 0.5661103979460848, + "grad_norm": 0.2817283868789673, + "learning_rate": 4.8439424102900064e-05, + "loss": 2.3056, + "step": 441 + }, + { + "epoch": 0.5673940949935815, + "grad_norm": 0.29399430751800537, + "learning_rate": 4.84303548868283e-05, + "loss": 2.4049, + "step": 442 + }, + { + "epoch": 0.5686777920410783, + "grad_norm": 0.32070907950401306, + "learning_rate": 4.842126024839453e-05, + "loss": 2.3548, + "step": 443 + }, + { + "epoch": 0.5699614890885751, + "grad_norm": 0.2972594201564789, + "learning_rate": 4.841214019746663e-05, + "loss": 2.4239, + "step": 444 + }, + { + "epoch": 0.5712451861360719, + "grad_norm": 0.3162088096141815, + "learning_rate": 4.840299474394e-05, + "loss": 2.4425, + "step": 445 + }, + { + "epoch": 0.5725288831835686, + "grad_norm": 0.2938075661659241, + "learning_rate": 4.8393823897737634e-05, + "loss": 2.3283, + "step": 446 + }, + { + "epoch": 0.5738125802310655, + "grad_norm": 0.2970564067363739, + "learning_rate": 4.8384627668810064e-05, + "loss": 2.3184, + "step": 447 + }, + { + "epoch": 0.5750962772785623, + "grad_norm": 0.2736704647541046, + "learning_rate": 4.837540606713538e-05, + "loss": 2.3115, + "step": 448 + }, + { + "epoch": 0.576379974326059, + "grad_norm": 0.3252422511577606, + "learning_rate": 4.8366159102719156e-05, + "loss": 2.3955, + "step": 449 + }, + { + "epoch": 0.5776636713735558, + "grad_norm": 0.3110811114311218, + "learning_rate": 4.8356886785594544e-05, + "loss": 2.3336, + "step": 450 + }, + { + "epoch": 0.5789473684210527, + "grad_norm": 0.2989579141139984, + "learning_rate": 4.834758912582217e-05, + "loss": 2.3001, + "step": 451 + }, + { + "epoch": 0.5802310654685494, + "grad_norm": 0.3055441379547119, + "learning_rate": 4.833826613349016e-05, + "loss": 2.3683, + "step": 452 + }, + { + "epoch": 0.5815147625160462, + "grad_norm": 0.30826109647750854, + "learning_rate": 4.832891781871414e-05, + "loss": 2.3054, + "step": 453 + }, + { + "epoch": 0.582798459563543, + "grad_norm": 0.32700634002685547, + "learning_rate": 4.831954419163719e-05, + "loss": 2.3865, + "step": 454 + }, + { + "epoch": 0.5840821566110398, + "grad_norm": 0.30788737535476685, + "learning_rate": 4.831014526242987e-05, + "loss": 2.3478, + "step": 455 + }, + { + "epoch": 0.5853658536585366, + "grad_norm": 0.28584685921669006, + "learning_rate": 4.8300721041290194e-05, + "loss": 2.356, + "step": 456 + }, + { + "epoch": 0.5866495507060334, + "grad_norm": 0.3295878469944, + "learning_rate": 4.829127153844361e-05, + "loss": 2.3768, + "step": 457 + }, + { + "epoch": 0.5879332477535302, + "grad_norm": 0.31243589520454407, + "learning_rate": 4.8281796764143e-05, + "loss": 2.365, + "step": 458 + }, + { + "epoch": 0.5892169448010269, + "grad_norm": 0.30623236298561096, + "learning_rate": 4.8272296728668676e-05, + "loss": 2.369, + "step": 459 + }, + { + "epoch": 0.5905006418485238, + "grad_norm": 0.2986590564250946, + "learning_rate": 4.826277144232834e-05, + "loss": 2.3042, + "step": 460 + }, + { + "epoch": 0.5917843388960206, + "grad_norm": 0.30335181951522827, + "learning_rate": 4.825322091545709e-05, + "loss": 2.2763, + "step": 461 + }, + { + "epoch": 0.5930680359435173, + "grad_norm": 0.30525484681129456, + "learning_rate": 4.824364515841745e-05, + "loss": 2.3532, + "step": 462 + }, + { + "epoch": 0.5943517329910141, + "grad_norm": 0.2878653109073639, + "learning_rate": 4.823404418159927e-05, + "loss": 2.4261, + "step": 463 + }, + { + "epoch": 0.5956354300385109, + "grad_norm": 0.29711782932281494, + "learning_rate": 4.822441799541979e-05, + "loss": 2.4149, + "step": 464 + }, + { + "epoch": 0.5969191270860077, + "grad_norm": 0.2690982520580292, + "learning_rate": 4.82147666103236e-05, + "loss": 2.3366, + "step": 465 + }, + { + "epoch": 0.5982028241335045, + "grad_norm": 0.2946816682815552, + "learning_rate": 4.8205090036782626e-05, + "loss": 2.3942, + "step": 466 + }, + { + "epoch": 0.5994865211810013, + "grad_norm": 0.3061763346195221, + "learning_rate": 4.819538828529613e-05, + "loss": 2.4006, + "step": 467 + }, + { + "epoch": 0.6007702182284981, + "grad_norm": 0.32586535811424255, + "learning_rate": 4.8185661366390676e-05, + "loss": 2.3707, + "step": 468 + }, + { + "epoch": 0.6020539152759948, + "grad_norm": 0.27984949946403503, + "learning_rate": 4.817590929062017e-05, + "loss": 2.4379, + "step": 469 + }, + { + "epoch": 0.6033376123234917, + "grad_norm": 0.329476922750473, + "learning_rate": 4.816613206856577e-05, + "loss": 2.4531, + "step": 470 + }, + { + "epoch": 0.6046213093709885, + "grad_norm": 0.31376075744628906, + "learning_rate": 4.815632971083596e-05, + "loss": 2.4197, + "step": 471 + }, + { + "epoch": 0.6059050064184852, + "grad_norm": 0.28301122784614563, + "learning_rate": 4.814650222806647e-05, + "loss": 2.2975, + "step": 472 + }, + { + "epoch": 0.6071887034659821, + "grad_norm": 0.30881014466285706, + "learning_rate": 4.813664963092029e-05, + "loss": 2.3507, + "step": 473 + }, + { + "epoch": 0.6084724005134788, + "grad_norm": 0.30735889077186584, + "learning_rate": 4.8126771930087674e-05, + "loss": 2.2855, + "step": 474 + }, + { + "epoch": 0.6097560975609756, + "grad_norm": 0.31143245100975037, + "learning_rate": 4.8116869136286106e-05, + "loss": 2.3445, + "step": 475 + }, + { + "epoch": 0.6110397946084724, + "grad_norm": 0.30091342329978943, + "learning_rate": 4.8106941260260296e-05, + "loss": 2.3587, + "step": 476 + }, + { + "epoch": 0.6123234916559692, + "grad_norm": 0.29558065533638, + "learning_rate": 4.8096988312782174e-05, + "loss": 2.4079, + "step": 477 + }, + { + "epoch": 0.613607188703466, + "grad_norm": 0.28896674513816833, + "learning_rate": 4.8087010304650866e-05, + "loss": 2.3141, + "step": 478 + }, + { + "epoch": 0.6148908857509627, + "grad_norm": 0.39745038747787476, + "learning_rate": 4.8077007246692694e-05, + "loss": 2.3588, + "step": 479 + }, + { + "epoch": 0.6161745827984596, + "grad_norm": 0.300830602645874, + "learning_rate": 4.806697914976116e-05, + "loss": 2.4217, + "step": 480 + }, + { + "epoch": 0.6174582798459564, + "grad_norm": 0.3155045211315155, + "learning_rate": 4.805692602473693e-05, + "loss": 2.3939, + "step": 481 + }, + { + "epoch": 0.6187419768934531, + "grad_norm": 0.3159199357032776, + "learning_rate": 4.8046847882527826e-05, + "loss": 2.4239, + "step": 482 + }, + { + "epoch": 0.62002567394095, + "grad_norm": 0.299836665391922, + "learning_rate": 4.8036744734068826e-05, + "loss": 2.4192, + "step": 483 + }, + { + "epoch": 0.6213093709884467, + "grad_norm": 0.3199720084667206, + "learning_rate": 4.802661659032202e-05, + "loss": 2.39, + "step": 484 + }, + { + "epoch": 0.6225930680359435, + "grad_norm": 0.29899173974990845, + "learning_rate": 4.8016463462276615e-05, + "loss": 2.3525, + "step": 485 + }, + { + "epoch": 0.6238767650834403, + "grad_norm": 0.2836560904979706, + "learning_rate": 4.8006285360948976e-05, + "loss": 2.3127, + "step": 486 + }, + { + "epoch": 0.6251604621309371, + "grad_norm": 0.31214481592178345, + "learning_rate": 4.79960822973825e-05, + "loss": 2.4241, + "step": 487 + }, + { + "epoch": 0.6264441591784339, + "grad_norm": 0.3114881217479706, + "learning_rate": 4.79858542826477e-05, + "loss": 2.2663, + "step": 488 + }, + { + "epoch": 0.6277278562259306, + "grad_norm": 0.30914467573165894, + "learning_rate": 4.7975601327842176e-05, + "loss": 2.3633, + "step": 489 + }, + { + "epoch": 0.6290115532734275, + "grad_norm": 0.29763662815093994, + "learning_rate": 4.796532344409055e-05, + "loss": 2.336, + "step": 490 + }, + { + "epoch": 0.6302952503209243, + "grad_norm": 0.2814677357673645, + "learning_rate": 4.7955020642544515e-05, + "loss": 2.4174, + "step": 491 + }, + { + "epoch": 0.631578947368421, + "grad_norm": 0.31262046098709106, + "learning_rate": 4.794469293438282e-05, + "loss": 2.3888, + "step": 492 + }, + { + "epoch": 0.6328626444159179, + "grad_norm": 0.3011872470378876, + "learning_rate": 4.7934340330811186e-05, + "loss": 2.2837, + "step": 493 + }, + { + "epoch": 0.6341463414634146, + "grad_norm": 0.3168891370296478, + "learning_rate": 4.79239628430624e-05, + "loss": 2.4802, + "step": 494 + }, + { + "epoch": 0.6354300385109114, + "grad_norm": 0.2910175621509552, + "learning_rate": 4.791356048239622e-05, + "loss": 2.3782, + "step": 495 + }, + { + "epoch": 0.6367137355584083, + "grad_norm": 0.2881637513637543, + "learning_rate": 4.7903133260099385e-05, + "loss": 2.3472, + "step": 496 + }, + { + "epoch": 0.637997432605905, + "grad_norm": 0.31721916794776917, + "learning_rate": 4.789268118748564e-05, + "loss": 2.4454, + "step": 497 + }, + { + "epoch": 0.6392811296534018, + "grad_norm": 0.2995906472206116, + "learning_rate": 4.788220427589566e-05, + "loss": 2.2958, + "step": 498 + }, + { + "epoch": 0.6405648267008985, + "grad_norm": 0.29439058899879456, + "learning_rate": 4.787170253669708e-05, + "loss": 2.3642, + "step": 499 + }, + { + "epoch": 0.6418485237483954, + "grad_norm": 0.29495733976364136, + "learning_rate": 4.78611759812845e-05, + "loss": 2.3808, + "step": 500 + }, + { + "epoch": 0.6431322207958922, + "grad_norm": 0.3079684376716614, + "learning_rate": 4.785062462107941e-05, + "loss": 2.3306, + "step": 501 + }, + { + "epoch": 0.6444159178433889, + "grad_norm": 0.3264211416244507, + "learning_rate": 4.784004846753023e-05, + "loss": 2.3811, + "step": 502 + }, + { + "epoch": 0.6456996148908858, + "grad_norm": 0.32243621349334717, + "learning_rate": 4.782944753211228e-05, + "loss": 2.3574, + "step": 503 + }, + { + "epoch": 0.6469833119383825, + "grad_norm": 0.33553004264831543, + "learning_rate": 4.781882182632776e-05, + "loss": 2.2817, + "step": 504 + }, + { + "epoch": 0.6482670089858793, + "grad_norm": 0.28932246565818787, + "learning_rate": 4.780817136170578e-05, + "loss": 2.4276, + "step": 505 + }, + { + "epoch": 0.6495507060333762, + "grad_norm": 0.30343809723854065, + "learning_rate": 4.7797496149802256e-05, + "loss": 2.4455, + "step": 506 + }, + { + "epoch": 0.6508344030808729, + "grad_norm": 0.3055610954761505, + "learning_rate": 4.77867962022e-05, + "loss": 2.318, + "step": 507 + }, + { + "epoch": 0.6521181001283697, + "grad_norm": 0.30805933475494385, + "learning_rate": 4.777607153050866e-05, + "loss": 2.4104, + "step": 508 + }, + { + "epoch": 0.6534017971758665, + "grad_norm": 0.2895534038543701, + "learning_rate": 4.7765322146364686e-05, + "loss": 2.3868, + "step": 509 + }, + { + "epoch": 0.6546854942233633, + "grad_norm": 0.3000538647174835, + "learning_rate": 4.775454806143137e-05, + "loss": 2.3504, + "step": 510 + }, + { + "epoch": 0.6559691912708601, + "grad_norm": 0.3021094799041748, + "learning_rate": 4.7743749287398776e-05, + "loss": 2.3791, + "step": 511 + }, + { + "epoch": 0.6572528883183568, + "grad_norm": 0.2954312860965729, + "learning_rate": 4.7732925835983775e-05, + "loss": 2.3347, + "step": 512 + }, + { + "epoch": 0.6585365853658537, + "grad_norm": 0.2947414815425873, + "learning_rate": 4.7722077718930014e-05, + "loss": 2.3683, + "step": 513 + }, + { + "epoch": 0.6598202824133504, + "grad_norm": 0.30434030294418335, + "learning_rate": 4.771120494800789e-05, + "loss": 2.3666, + "step": 514 + }, + { + "epoch": 0.6611039794608472, + "grad_norm": 0.2817016541957855, + "learning_rate": 4.7700307535014565e-05, + "loss": 2.3736, + "step": 515 + }, + { + "epoch": 0.6623876765083441, + "grad_norm": 0.2933237552642822, + "learning_rate": 4.768938549177393e-05, + "loss": 2.3464, + "step": 516 + }, + { + "epoch": 0.6636713735558408, + "grad_norm": 0.3133641481399536, + "learning_rate": 4.7678438830136596e-05, + "loss": 2.3103, + "step": 517 + }, + { + "epoch": 0.6649550706033376, + "grad_norm": 0.2988991439342499, + "learning_rate": 4.766746756197989e-05, + "loss": 2.4228, + "step": 518 + }, + { + "epoch": 0.6662387676508345, + "grad_norm": 0.3129512369632721, + "learning_rate": 4.765647169920785e-05, + "loss": 2.3711, + "step": 519 + }, + { + "epoch": 0.6675224646983312, + "grad_norm": 0.30164363980293274, + "learning_rate": 4.764545125375117e-05, + "loss": 2.314, + "step": 520 + }, + { + "epoch": 0.668806161745828, + "grad_norm": 0.3078322112560272, + "learning_rate": 4.7634406237567245e-05, + "loss": 2.3583, + "step": 521 + }, + { + "epoch": 0.6700898587933247, + "grad_norm": 0.29779842495918274, + "learning_rate": 4.7623336662640116e-05, + "loss": 2.3329, + "step": 522 + }, + { + "epoch": 0.6713735558408216, + "grad_norm": 0.30728939175605774, + "learning_rate": 4.761224254098048e-05, + "loss": 2.3432, + "step": 523 + }, + { + "epoch": 0.6726572528883183, + "grad_norm": 0.28595444560050964, + "learning_rate": 4.760112388462564e-05, + "loss": 2.3155, + "step": 524 + }, + { + "epoch": 0.6739409499358151, + "grad_norm": 0.32341310381889343, + "learning_rate": 4.758998070563957e-05, + "loss": 2.3919, + "step": 525 + }, + { + "epoch": 0.675224646983312, + "grad_norm": 0.33163347840309143, + "learning_rate": 4.75788130161128e-05, + "loss": 2.2703, + "step": 526 + }, + { + "epoch": 0.6765083440308087, + "grad_norm": 0.2895621061325073, + "learning_rate": 4.7567620828162486e-05, + "loss": 2.3479, + "step": 527 + }, + { + "epoch": 0.6777920410783055, + "grad_norm": 0.31375667452812195, + "learning_rate": 4.7556404153932356e-05, + "loss": 2.4763, + "step": 528 + }, + { + "epoch": 0.6790757381258024, + "grad_norm": 0.29855942726135254, + "learning_rate": 4.754516300559271e-05, + "loss": 2.4154, + "step": 529 + }, + { + "epoch": 0.6803594351732991, + "grad_norm": 0.29022201895713806, + "learning_rate": 4.7533897395340384e-05, + "loss": 2.3614, + "step": 530 + }, + { + "epoch": 0.6816431322207959, + "grad_norm": 0.280118852853775, + "learning_rate": 4.7522607335398786e-05, + "loss": 2.3306, + "step": 531 + }, + { + "epoch": 0.6829268292682927, + "grad_norm": 0.29326388239860535, + "learning_rate": 4.751129283801782e-05, + "loss": 2.3219, + "step": 532 + }, + { + "epoch": 0.6842105263157895, + "grad_norm": 0.3260889947414398, + "learning_rate": 4.7499953915473935e-05, + "loss": 2.3952, + "step": 533 + }, + { + "epoch": 0.6854942233632862, + "grad_norm": 0.2982180118560791, + "learning_rate": 4.7488590580070074e-05, + "loss": 2.3165, + "step": 534 + }, + { + "epoch": 0.686777920410783, + "grad_norm": 0.303310364484787, + "learning_rate": 4.7477202844135646e-05, + "loss": 2.2632, + "step": 535 + }, + { + "epoch": 0.6880616174582799, + "grad_norm": 0.329369455575943, + "learning_rate": 4.746579072002657e-05, + "loss": 2.328, + "step": 536 + }, + { + "epoch": 0.6893453145057766, + "grad_norm": 0.3056361973285675, + "learning_rate": 4.74543542201252e-05, + "loss": 2.4139, + "step": 537 + }, + { + "epoch": 0.6906290115532734, + "grad_norm": 0.31663402915000916, + "learning_rate": 4.744289335684034e-05, + "loss": 2.3929, + "step": 538 + }, + { + "epoch": 0.6919127086007703, + "grad_norm": 0.3166126012802124, + "learning_rate": 4.7431408142607236e-05, + "loss": 2.3308, + "step": 539 + }, + { + "epoch": 0.693196405648267, + "grad_norm": 0.3111107349395752, + "learning_rate": 4.7419898589887566e-05, + "loss": 2.3885, + "step": 540 + }, + { + "epoch": 0.6944801026957638, + "grad_norm": 0.3381554186344147, + "learning_rate": 4.74083647111694e-05, + "loss": 2.2421, + "step": 541 + }, + { + "epoch": 0.6957637997432606, + "grad_norm": 0.29884985089302063, + "learning_rate": 4.73968065189672e-05, + "loss": 2.3181, + "step": 542 + }, + { + "epoch": 0.6970474967907574, + "grad_norm": 0.2916213274002075, + "learning_rate": 4.738522402582183e-05, + "loss": 2.3626, + "step": 543 + }, + { + "epoch": 0.6983311938382541, + "grad_norm": 0.3153024911880493, + "learning_rate": 4.737361724430048e-05, + "loss": 2.4354, + "step": 544 + }, + { + "epoch": 0.699614890885751, + "grad_norm": 0.2967838644981384, + "learning_rate": 4.736198618699675e-05, + "loss": 2.4529, + "step": 545 + }, + { + "epoch": 0.7008985879332478, + "grad_norm": 0.29753878712654114, + "learning_rate": 4.7350330866530536e-05, + "loss": 2.3934, + "step": 546 + }, + { + "epoch": 0.7021822849807445, + "grad_norm": 0.2958550751209259, + "learning_rate": 4.7338651295548065e-05, + "loss": 2.2737, + "step": 547 + }, + { + "epoch": 0.7034659820282413, + "grad_norm": 0.2889609932899475, + "learning_rate": 4.7326947486721894e-05, + "loss": 2.3989, + "step": 548 + }, + { + "epoch": 0.7047496790757382, + "grad_norm": 0.2951468229293823, + "learning_rate": 4.731521945275087e-05, + "loss": 2.3909, + "step": 549 + }, + { + "epoch": 0.7060333761232349, + "grad_norm": 0.30337998270988464, + "learning_rate": 4.730346720636011e-05, + "loss": 2.3878, + "step": 550 + }, + { + "epoch": 0.7073170731707317, + "grad_norm": 0.2858850359916687, + "learning_rate": 4.7291690760301035e-05, + "loss": 2.3465, + "step": 551 + }, + { + "epoch": 0.7086007702182285, + "grad_norm": 0.2845821678638458, + "learning_rate": 4.727989012735129e-05, + "loss": 2.385, + "step": 552 + }, + { + "epoch": 0.7098844672657253, + "grad_norm": 0.2901153564453125, + "learning_rate": 4.726806532031478e-05, + "loss": 2.2939, + "step": 553 + }, + { + "epoch": 0.711168164313222, + "grad_norm": 0.27658534049987793, + "learning_rate": 4.725621635202164e-05, + "loss": 2.3292, + "step": 554 + }, + { + "epoch": 0.7124518613607189, + "grad_norm": 0.289438933134079, + "learning_rate": 4.724434323532821e-05, + "loss": 2.3481, + "step": 555 + }, + { + "epoch": 0.7137355584082157, + "grad_norm": 0.30061691999435425, + "learning_rate": 4.7232445983117045e-05, + "loss": 2.2876, + "step": 556 + }, + { + "epoch": 0.7150192554557124, + "grad_norm": 0.28911301493644714, + "learning_rate": 4.722052460829687e-05, + "loss": 2.3649, + "step": 557 + }, + { + "epoch": 0.7163029525032092, + "grad_norm": 0.32217785716056824, + "learning_rate": 4.720857912380261e-05, + "loss": 2.3052, + "step": 558 + }, + { + "epoch": 0.7175866495507061, + "grad_norm": 0.3333997428417206, + "learning_rate": 4.719660954259532e-05, + "loss": 2.3851, + "step": 559 + }, + { + "epoch": 0.7188703465982028, + "grad_norm": 0.3150656819343567, + "learning_rate": 4.718461587766221e-05, + "loss": 2.3354, + "step": 560 + }, + { + "epoch": 0.7201540436456996, + "grad_norm": 0.2966265380382538, + "learning_rate": 4.7172598142016645e-05, + "loss": 2.3681, + "step": 561 + }, + { + "epoch": 0.7214377406931964, + "grad_norm": 0.3027741014957428, + "learning_rate": 4.716055634869807e-05, + "loss": 2.3369, + "step": 562 + }, + { + "epoch": 0.7227214377406932, + "grad_norm": 0.32294961810112, + "learning_rate": 4.714849051077205e-05, + "loss": 2.4019, + "step": 563 + }, + { + "epoch": 0.72400513478819, + "grad_norm": 0.30615222454071045, + "learning_rate": 4.713640064133025e-05, + "loss": 2.4961, + "step": 564 + }, + { + "epoch": 0.7252888318356868, + "grad_norm": 0.29600149393081665, + "learning_rate": 4.712428675349039e-05, + "loss": 2.3446, + "step": 565 + }, + { + "epoch": 0.7265725288831836, + "grad_norm": 0.2971326410770416, + "learning_rate": 4.7112148860396265e-05, + "loss": 2.3006, + "step": 566 + }, + { + "epoch": 0.7278562259306803, + "grad_norm": 0.29449960589408875, + "learning_rate": 4.7099986975217704e-05, + "loss": 2.464, + "step": 567 + }, + { + "epoch": 0.7291399229781772, + "grad_norm": 0.27724847197532654, + "learning_rate": 4.708780111115057e-05, + "loss": 2.4383, + "step": 568 + }, + { + "epoch": 0.730423620025674, + "grad_norm": 0.3143709897994995, + "learning_rate": 4.7075591281416765e-05, + "loss": 2.3377, + "step": 569 + }, + { + "epoch": 0.7317073170731707, + "grad_norm": 0.32122036814689636, + "learning_rate": 4.706335749926417e-05, + "loss": 2.2914, + "step": 570 + }, + { + "epoch": 0.7329910141206675, + "grad_norm": 0.28614115715026855, + "learning_rate": 4.7051099777966645e-05, + "loss": 2.3338, + "step": 571 + }, + { + "epoch": 0.7342747111681643, + "grad_norm": 0.30070704221725464, + "learning_rate": 4.703881813082406e-05, + "loss": 2.3367, + "step": 572 + }, + { + "epoch": 0.7355584082156611, + "grad_norm": 0.2799395024776459, + "learning_rate": 4.702651257116222e-05, + "loss": 2.3604, + "step": 573 + }, + { + "epoch": 0.7368421052631579, + "grad_norm": 0.29513630270957947, + "learning_rate": 4.7014183112332874e-05, + "loss": 2.3487, + "step": 574 + }, + { + "epoch": 0.7381258023106547, + "grad_norm": 0.30071401596069336, + "learning_rate": 4.7001829767713726e-05, + "loss": 2.3422, + "step": 575 + }, + { + "epoch": 0.7394094993581515, + "grad_norm": 0.3397565186023712, + "learning_rate": 4.698945255070837e-05, + "loss": 2.3802, + "step": 576 + }, + { + "epoch": 0.7406931964056482, + "grad_norm": 0.32068637013435364, + "learning_rate": 4.697705147474631e-05, + "loss": 2.3498, + "step": 577 + }, + { + "epoch": 0.7419768934531451, + "grad_norm": 0.5262579917907715, + "learning_rate": 4.696462655328294e-05, + "loss": 2.2523, + "step": 578 + }, + { + "epoch": 0.7432605905006419, + "grad_norm": 0.3595364987850189, + "learning_rate": 4.695217779979953e-05, + "loss": 2.3578, + "step": 579 + }, + { + "epoch": 0.7445442875481386, + "grad_norm": 0.3001711964607239, + "learning_rate": 4.69397052278032e-05, + "loss": 2.2685, + "step": 580 + }, + { + "epoch": 0.7458279845956355, + "grad_norm": 0.3520994782447815, + "learning_rate": 4.6927208850826925e-05, + "loss": 2.364, + "step": 581 + }, + { + "epoch": 0.7471116816431322, + "grad_norm": 0.3192301094532013, + "learning_rate": 4.6914688682429496e-05, + "loss": 2.3454, + "step": 582 + }, + { + "epoch": 0.748395378690629, + "grad_norm": 0.28848007321357727, + "learning_rate": 4.690214473619552e-05, + "loss": 2.5347, + "step": 583 + }, + { + "epoch": 0.7496790757381258, + "grad_norm": 0.3215525448322296, + "learning_rate": 4.688957702573542e-05, + "loss": 2.3881, + "step": 584 + }, + { + "epoch": 0.7509627727856226, + "grad_norm": 0.32826587557792664, + "learning_rate": 4.687698556468538e-05, + "loss": 2.3555, + "step": 585 + }, + { + "epoch": 0.7522464698331194, + "grad_norm": 0.2755299210548401, + "learning_rate": 4.6864370366707366e-05, + "loss": 2.3278, + "step": 586 + }, + { + "epoch": 0.7535301668806161, + "grad_norm": 0.32962703704833984, + "learning_rate": 4.68517314454891e-05, + "loss": 2.3749, + "step": 587 + }, + { + "epoch": 0.754813863928113, + "grad_norm": 0.29674839973449707, + "learning_rate": 4.683906881474405e-05, + "loss": 2.3683, + "step": 588 + }, + { + "epoch": 0.7560975609756098, + "grad_norm": 0.31625881791114807, + "learning_rate": 4.682638248821138e-05, + "loss": 2.3088, + "step": 589 + }, + { + "epoch": 0.7573812580231065, + "grad_norm": 0.31702378392219543, + "learning_rate": 4.6813672479656e-05, + "loss": 2.4639, + "step": 590 + }, + { + "epoch": 0.7586649550706034, + "grad_norm": 0.3002925515174866, + "learning_rate": 4.6800938802868505e-05, + "loss": 2.3569, + "step": 591 + }, + { + "epoch": 0.7599486521181001, + "grad_norm": 0.30289632081985474, + "learning_rate": 4.6788181471665155e-05, + "loss": 2.3155, + "step": 592 + }, + { + "epoch": 0.7612323491655969, + "grad_norm": 0.2899289131164551, + "learning_rate": 4.677540049988789e-05, + "loss": 2.4306, + "step": 593 + }, + { + "epoch": 0.7625160462130937, + "grad_norm": 0.2906821370124817, + "learning_rate": 4.67625959014043e-05, + "loss": 2.3712, + "step": 594 + }, + { + "epoch": 0.7637997432605905, + "grad_norm": 0.3128328025341034, + "learning_rate": 4.674976769010761e-05, + "loss": 2.3235, + "step": 595 + }, + { + "epoch": 0.7650834403080873, + "grad_norm": 0.33928877115249634, + "learning_rate": 4.673691587991667e-05, + "loss": 2.4025, + "step": 596 + }, + { + "epoch": 0.766367137355584, + "grad_norm": 0.3034507632255554, + "learning_rate": 4.67240404847759e-05, + "loss": 2.2935, + "step": 597 + }, + { + "epoch": 0.7676508344030809, + "grad_norm": 0.29682376980781555, + "learning_rate": 4.671114151865536e-05, + "loss": 2.3951, + "step": 598 + }, + { + "epoch": 0.7689345314505777, + "grad_norm": 0.29410845041275024, + "learning_rate": 4.669821899555066e-05, + "loss": 2.3497, + "step": 599 + }, + { + "epoch": 0.7702182284980744, + "grad_norm": 0.3179001808166504, + "learning_rate": 4.668527292948298e-05, + "loss": 2.3472, + "step": 600 + }, + { + "epoch": 0.7715019255455713, + "grad_norm": 0.28910136222839355, + "learning_rate": 4.667230333449902e-05, + "loss": 2.335, + "step": 601 + }, + { + "epoch": 0.772785622593068, + "grad_norm": 0.3009548783302307, + "learning_rate": 4.665931022467105e-05, + "loss": 2.4499, + "step": 602 + }, + { + "epoch": 0.7740693196405648, + "grad_norm": 0.2947859466075897, + "learning_rate": 4.664629361409681e-05, + "loss": 2.4299, + "step": 603 + }, + { + "epoch": 0.7753530166880617, + "grad_norm": 0.30695682764053345, + "learning_rate": 4.663325351689956e-05, + "loss": 2.3051, + "step": 604 + }, + { + "epoch": 0.7766367137355584, + "grad_norm": 0.2979392111301422, + "learning_rate": 4.662018994722806e-05, + "loss": 2.4222, + "step": 605 + }, + { + "epoch": 0.7779204107830552, + "grad_norm": 0.298109769821167, + "learning_rate": 4.660710291925652e-05, + "loss": 2.4049, + "step": 606 + }, + { + "epoch": 0.7792041078305519, + "grad_norm": 0.29500192403793335, + "learning_rate": 4.6593992447184586e-05, + "loss": 2.3886, + "step": 607 + }, + { + "epoch": 0.7804878048780488, + "grad_norm": 0.2970205843448639, + "learning_rate": 4.658085854523737e-05, + "loss": 2.3958, + "step": 608 + }, + { + "epoch": 0.7817715019255456, + "grad_norm": 0.3007825016975403, + "learning_rate": 4.6567701227665394e-05, + "loss": 2.3164, + "step": 609 + }, + { + "epoch": 0.7830551989730423, + "grad_norm": 0.3098297715187073, + "learning_rate": 4.655452050874459e-05, + "loss": 2.3919, + "step": 610 + }, + { + "epoch": 0.7843388960205392, + "grad_norm": 0.29705604910850525, + "learning_rate": 4.654131640277627e-05, + "loss": 2.4343, + "step": 611 + }, + { + "epoch": 0.785622593068036, + "grad_norm": 0.2924937605857849, + "learning_rate": 4.6528088924087134e-05, + "loss": 2.2326, + "step": 612 + }, + { + "epoch": 0.7869062901155327, + "grad_norm": 0.316272109746933, + "learning_rate": 4.651483808702924e-05, + "loss": 2.3787, + "step": 613 + }, + { + "epoch": 0.7881899871630296, + "grad_norm": 0.3279688358306885, + "learning_rate": 4.650156390598e-05, + "loss": 2.4019, + "step": 614 + }, + { + "epoch": 0.7894736842105263, + "grad_norm": 0.299129456281662, + "learning_rate": 4.6488266395342105e-05, + "loss": 2.3136, + "step": 615 + }, + { + "epoch": 0.7907573812580231, + "grad_norm": 0.2993239760398865, + "learning_rate": 4.647494556954363e-05, + "loss": 2.4093, + "step": 616 + }, + { + "epoch": 0.79204107830552, + "grad_norm": 0.3059884309768677, + "learning_rate": 4.6461601443037905e-05, + "loss": 2.3847, + "step": 617 + }, + { + "epoch": 0.7933247753530167, + "grad_norm": 0.3300536274909973, + "learning_rate": 4.644823403030355e-05, + "loss": 2.3478, + "step": 618 + }, + { + "epoch": 0.7946084724005135, + "grad_norm": 0.30517104268074036, + "learning_rate": 4.643484334584445e-05, + "loss": 2.3763, + "step": 619 + }, + { + "epoch": 0.7958921694480102, + "grad_norm": 0.34745684266090393, + "learning_rate": 4.642142940418973e-05, + "loss": 2.4441, + "step": 620 + }, + { + "epoch": 0.7971758664955071, + "grad_norm": 0.2973950207233429, + "learning_rate": 4.640799221989378e-05, + "loss": 2.3866, + "step": 621 + }, + { + "epoch": 0.7984595635430038, + "grad_norm": 0.2934724688529968, + "learning_rate": 4.639453180753619e-05, + "loss": 2.3949, + "step": 622 + }, + { + "epoch": 0.7997432605905006, + "grad_norm": 0.2799731194972992, + "learning_rate": 4.6381048181721744e-05, + "loss": 2.3475, + "step": 623 + }, + { + "epoch": 0.8010269576379975, + "grad_norm": 0.30114057660102844, + "learning_rate": 4.636754135708041e-05, + "loss": 2.4434, + "step": 624 + }, + { + "epoch": 0.8023106546854942, + "grad_norm": 0.3088257610797882, + "learning_rate": 4.635401134826737e-05, + "loss": 2.3977, + "step": 625 + }, + { + "epoch": 0.803594351732991, + "grad_norm": 0.3008176386356354, + "learning_rate": 4.63404581699629e-05, + "loss": 2.3099, + "step": 626 + }, + { + "epoch": 0.8048780487804879, + "grad_norm": 0.29447734355926514, + "learning_rate": 4.632688183687246e-05, + "loss": 2.4009, + "step": 627 + }, + { + "epoch": 0.8061617458279846, + "grad_norm": 0.29392993450164795, + "learning_rate": 4.631328236372662e-05, + "loss": 2.3303, + "step": 628 + }, + { + "epoch": 0.8074454428754814, + "grad_norm": 0.28862807154655457, + "learning_rate": 4.629965976528104e-05, + "loss": 2.4822, + "step": 629 + }, + { + "epoch": 0.8087291399229781, + "grad_norm": 0.3060828447341919, + "learning_rate": 4.628601405631652e-05, + "loss": 2.4551, + "step": 630 + }, + { + "epoch": 0.810012836970475, + "grad_norm": 0.3012582063674927, + "learning_rate": 4.627234525163887e-05, + "loss": 2.3667, + "step": 631 + }, + { + "epoch": 0.8112965340179717, + "grad_norm": 0.2978503406047821, + "learning_rate": 4.625865336607901e-05, + "loss": 2.3306, + "step": 632 + }, + { + "epoch": 0.8125802310654685, + "grad_norm": 0.31029176712036133, + "learning_rate": 4.6244938414492875e-05, + "loss": 2.4326, + "step": 633 + }, + { + "epoch": 0.8138639281129654, + "grad_norm": 0.3450663983821869, + "learning_rate": 4.6231200411761444e-05, + "loss": 2.3047, + "step": 634 + }, + { + "epoch": 0.8151476251604621, + "grad_norm": 0.3168805241584778, + "learning_rate": 4.62174393727907e-05, + "loss": 2.4397, + "step": 635 + }, + { + "epoch": 0.8164313222079589, + "grad_norm": 0.31874215602874756, + "learning_rate": 4.6203655312511616e-05, + "loss": 2.5085, + "step": 636 + }, + { + "epoch": 0.8177150192554558, + "grad_norm": 0.30860742926597595, + "learning_rate": 4.618984824588016e-05, + "loss": 2.2939, + "step": 637 + }, + { + "epoch": 0.8189987163029525, + "grad_norm": 0.3643187880516052, + "learning_rate": 4.617601818787724e-05, + "loss": 2.4521, + "step": 638 + }, + { + "epoch": 0.8202824133504493, + "grad_norm": 0.29677483439445496, + "learning_rate": 4.6162165153508724e-05, + "loss": 2.4059, + "step": 639 + }, + { + "epoch": 0.8215661103979461, + "grad_norm": 0.31648752093315125, + "learning_rate": 4.6148289157805406e-05, + "loss": 2.3753, + "step": 640 + }, + { + "epoch": 0.8228498074454429, + "grad_norm": 0.30333682894706726, + "learning_rate": 4.6134390215822995e-05, + "loss": 2.367, + "step": 641 + }, + { + "epoch": 0.8241335044929397, + "grad_norm": 0.321492075920105, + "learning_rate": 4.61204683426421e-05, + "loss": 2.3, + "step": 642 + }, + { + "epoch": 0.8254172015404364, + "grad_norm": 0.2793954014778137, + "learning_rate": 4.610652355336821e-05, + "loss": 2.3858, + "step": 643 + }, + { + "epoch": 0.8267008985879333, + "grad_norm": 0.29857027530670166, + "learning_rate": 4.609255586313166e-05, + "loss": 2.3492, + "step": 644 + }, + { + "epoch": 0.82798459563543, + "grad_norm": 0.28842952847480774, + "learning_rate": 4.607856528708765e-05, + "loss": 2.3433, + "step": 645 + }, + { + "epoch": 0.8292682926829268, + "grad_norm": 0.3550688326358795, + "learning_rate": 4.606455184041622e-05, + "loss": 2.3817, + "step": 646 + }, + { + "epoch": 0.8305519897304237, + "grad_norm": 0.32418951392173767, + "learning_rate": 4.6050515538322225e-05, + "loss": 2.4612, + "step": 647 + }, + { + "epoch": 0.8318356867779204, + "grad_norm": 0.33099403977394104, + "learning_rate": 4.6036456396035294e-05, + "loss": 2.3853, + "step": 648 + }, + { + "epoch": 0.8331193838254172, + "grad_norm": 0.32067880034446716, + "learning_rate": 4.602237442880985e-05, + "loss": 2.4484, + "step": 649 + }, + { + "epoch": 0.834403080872914, + "grad_norm": 0.34124529361724854, + "learning_rate": 4.600826965192509e-05, + "loss": 2.3791, + "step": 650 + }, + { + "epoch": 0.8356867779204108, + "grad_norm": 0.2857443392276764, + "learning_rate": 4.599414208068495e-05, + "loss": 2.355, + "step": 651 + }, + { + "epoch": 0.8369704749679076, + "grad_norm": 0.324481338262558, + "learning_rate": 4.5979991730418105e-05, + "loss": 2.4303, + "step": 652 + }, + { + "epoch": 0.8382541720154044, + "grad_norm": 0.32609954476356506, + "learning_rate": 4.596581861647795e-05, + "loss": 2.3638, + "step": 653 + }, + { + "epoch": 0.8395378690629012, + "grad_norm": 0.303725004196167, + "learning_rate": 4.595162275424255e-05, + "loss": 2.3144, + "step": 654 + }, + { + "epoch": 0.8408215661103979, + "grad_norm": 0.29211246967315674, + "learning_rate": 4.59374041591147e-05, + "loss": 2.4353, + "step": 655 + }, + { + "epoch": 0.8421052631578947, + "grad_norm": 0.31560078263282776, + "learning_rate": 4.5923162846521824e-05, + "loss": 2.3982, + "step": 656 + }, + { + "epoch": 0.8433889602053916, + "grad_norm": 0.3038894534111023, + "learning_rate": 4.5908898831916e-05, + "loss": 2.2962, + "step": 657 + }, + { + "epoch": 0.8446726572528883, + "grad_norm": 0.34458431601524353, + "learning_rate": 4.589461213077395e-05, + "loss": 2.3738, + "step": 658 + }, + { + "epoch": 0.8459563543003851, + "grad_norm": 0.3022404909133911, + "learning_rate": 4.5880302758597e-05, + "loss": 2.3374, + "step": 659 + }, + { + "epoch": 0.8472400513478819, + "grad_norm": 0.30100885033607483, + "learning_rate": 4.586597073091109e-05, + "loss": 2.286, + "step": 660 + }, + { + "epoch": 0.8485237483953787, + "grad_norm": 0.31171244382858276, + "learning_rate": 4.585161606326673e-05, + "loss": 2.3892, + "step": 661 + }, + { + "epoch": 0.8498074454428755, + "grad_norm": 0.3026835322380066, + "learning_rate": 4.5837238771238975e-05, + "loss": 2.3063, + "step": 662 + }, + { + "epoch": 0.8510911424903723, + "grad_norm": 0.2819378972053528, + "learning_rate": 4.582283887042748e-05, + "loss": 2.47, + "step": 663 + }, + { + "epoch": 0.8523748395378691, + "grad_norm": 0.2845015823841095, + "learning_rate": 4.580841637645638e-05, + "loss": 2.4173, + "step": 664 + }, + { + "epoch": 0.8536585365853658, + "grad_norm": 0.3148723840713501, + "learning_rate": 4.5793971304974344e-05, + "loss": 2.3728, + "step": 665 + }, + { + "epoch": 0.8549422336328626, + "grad_norm": 0.3028092086315155, + "learning_rate": 4.5779503671654544e-05, + "loss": 2.3707, + "step": 666 + }, + { + "epoch": 0.8562259306803595, + "grad_norm": 0.2981637418270111, + "learning_rate": 4.576501349219462e-05, + "loss": 2.4751, + "step": 667 + }, + { + "epoch": 0.8575096277278562, + "grad_norm": 0.33563345670700073, + "learning_rate": 4.5750500782316684e-05, + "loss": 2.3136, + "step": 668 + }, + { + "epoch": 0.858793324775353, + "grad_norm": 0.2983044385910034, + "learning_rate": 4.57359655577673e-05, + "loss": 2.3992, + "step": 669 + }, + { + "epoch": 0.8600770218228498, + "grad_norm": 0.3396848440170288, + "learning_rate": 4.5721407834317436e-05, + "loss": 2.3828, + "step": 670 + }, + { + "epoch": 0.8613607188703466, + "grad_norm": 0.2932271659374237, + "learning_rate": 4.570682762776249e-05, + "loss": 2.3649, + "step": 671 + }, + { + "epoch": 0.8626444159178434, + "grad_norm": 0.30177557468414307, + "learning_rate": 4.5692224953922266e-05, + "loss": 2.3681, + "step": 672 + }, + { + "epoch": 0.8639281129653402, + "grad_norm": 0.3091699182987213, + "learning_rate": 4.5677599828640916e-05, + "loss": 2.3726, + "step": 673 + }, + { + "epoch": 0.865211810012837, + "grad_norm": 0.31127968430519104, + "learning_rate": 4.5662952267786974e-05, + "loss": 2.3379, + "step": 674 + }, + { + "epoch": 0.8664955070603337, + "grad_norm": 0.28275659680366516, + "learning_rate": 4.56482822872533e-05, + "loss": 2.2449, + "step": 675 + }, + { + "epoch": 0.8677792041078306, + "grad_norm": 0.31371593475341797, + "learning_rate": 4.563358990295711e-05, + "loss": 2.3419, + "step": 676 + }, + { + "epoch": 0.8690629011553274, + "grad_norm": 0.31198492646217346, + "learning_rate": 4.561887513083989e-05, + "loss": 2.372, + "step": 677 + }, + { + "epoch": 0.8703465982028241, + "grad_norm": 0.31090936064720154, + "learning_rate": 4.560413798686745e-05, + "loss": 2.3304, + "step": 678 + }, + { + "epoch": 0.8716302952503209, + "grad_norm": 0.3031321167945862, + "learning_rate": 4.5589378487029844e-05, + "loss": 2.3631, + "step": 679 + }, + { + "epoch": 0.8729139922978177, + "grad_norm": 0.27430158853530884, + "learning_rate": 4.557459664734141e-05, + "loss": 2.3155, + "step": 680 + }, + { + "epoch": 0.8741976893453145, + "grad_norm": 0.32084590196609497, + "learning_rate": 4.555979248384071e-05, + "loss": 2.4118, + "step": 681 + }, + { + "epoch": 0.8754813863928113, + "grad_norm": 0.31383204460144043, + "learning_rate": 4.554496601259054e-05, + "loss": 2.3323, + "step": 682 + }, + { + "epoch": 0.8767650834403081, + "grad_norm": 0.3000917434692383, + "learning_rate": 4.5530117249677875e-05, + "loss": 2.3542, + "step": 683 + }, + { + "epoch": 0.8780487804878049, + "grad_norm": 0.3041795492172241, + "learning_rate": 4.5515246211213906e-05, + "loss": 2.3466, + "step": 684 + }, + { + "epoch": 0.8793324775353016, + "grad_norm": 0.31530001759529114, + "learning_rate": 4.5500352913333974e-05, + "loss": 2.3783, + "step": 685 + }, + { + "epoch": 0.8806161745827985, + "grad_norm": 0.3116975724697113, + "learning_rate": 4.5485437372197584e-05, + "loss": 2.4377, + "step": 686 + }, + { + "epoch": 0.8818998716302953, + "grad_norm": 0.32885855436325073, + "learning_rate": 4.5470499603988364e-05, + "loss": 2.4546, + "step": 687 + }, + { + "epoch": 0.883183568677792, + "grad_norm": 0.30758875608444214, + "learning_rate": 4.5455539624914066e-05, + "loss": 2.3051, + "step": 688 + }, + { + "epoch": 0.8844672657252889, + "grad_norm": 0.30743828415870667, + "learning_rate": 4.544055745120655e-05, + "loss": 2.2663, + "step": 689 + }, + { + "epoch": 0.8857509627727856, + "grad_norm": 0.3043730556964874, + "learning_rate": 4.542555309912173e-05, + "loss": 2.3279, + "step": 690 + }, + { + "epoch": 0.8870346598202824, + "grad_norm": 0.3107464909553528, + "learning_rate": 4.5410526584939614e-05, + "loss": 2.3694, + "step": 691 + }, + { + "epoch": 0.8883183568677792, + "grad_norm": 0.3012385070323944, + "learning_rate": 4.539547792496424e-05, + "loss": 2.4196, + "step": 692 + }, + { + "epoch": 0.889602053915276, + "grad_norm": 0.30979493260383606, + "learning_rate": 4.5380407135523675e-05, + "loss": 2.2083, + "step": 693 + }, + { + "epoch": 0.8908857509627728, + "grad_norm": 0.29600152373313904, + "learning_rate": 4.536531423297001e-05, + "loss": 2.4051, + "step": 694 + }, + { + "epoch": 0.8921694480102695, + "grad_norm": 0.2803564965724945, + "learning_rate": 4.5350199233679316e-05, + "loss": 2.4025, + "step": 695 + }, + { + "epoch": 0.8934531450577664, + "grad_norm": 0.29657790064811707, + "learning_rate": 4.5335062154051625e-05, + "loss": 2.4621, + "step": 696 + }, + { + "epoch": 0.8947368421052632, + "grad_norm": 0.2989714443683624, + "learning_rate": 4.531990301051097e-05, + "loss": 2.5027, + "step": 697 + }, + { + "epoch": 0.8960205391527599, + "grad_norm": 0.28022336959838867, + "learning_rate": 4.530472181950528e-05, + "loss": 2.347, + "step": 698 + }, + { + "epoch": 0.8973042362002568, + "grad_norm": 0.2950182855129242, + "learning_rate": 4.528951859750643e-05, + "loss": 2.2627, + "step": 699 + }, + { + "epoch": 0.8985879332477535, + "grad_norm": 0.283989816904068, + "learning_rate": 4.52742933610102e-05, + "loss": 2.3606, + "step": 700 + }, + { + "epoch": 0.8998716302952503, + "grad_norm": 0.2783466577529907, + "learning_rate": 4.525904612653624e-05, + "loss": 2.3604, + "step": 701 + }, + { + "epoch": 0.9011553273427471, + "grad_norm": 0.3252420723438263, + "learning_rate": 4.5243776910628076e-05, + "loss": 2.3619, + "step": 702 + }, + { + "epoch": 0.9024390243902439, + "grad_norm": 0.29868730902671814, + "learning_rate": 4.5228485729853096e-05, + "loss": 2.4099, + "step": 703 + }, + { + "epoch": 0.9037227214377407, + "grad_norm": 0.30771031975746155, + "learning_rate": 4.52131726008025e-05, + "loss": 2.3787, + "step": 704 + }, + { + "epoch": 0.9050064184852374, + "grad_norm": 0.28617870807647705, + "learning_rate": 4.519783754009131e-05, + "loss": 2.3759, + "step": 705 + }, + { + "epoch": 0.9062901155327343, + "grad_norm": 0.2890000641345978, + "learning_rate": 4.5182480564358356e-05, + "loss": 2.3883, + "step": 706 + }, + { + "epoch": 0.9075738125802311, + "grad_norm": 0.30811789631843567, + "learning_rate": 4.516710169026624e-05, + "loss": 2.349, + "step": 707 + }, + { + "epoch": 0.9088575096277278, + "grad_norm": 0.29645591974258423, + "learning_rate": 4.515170093450129e-05, + "loss": 2.3139, + "step": 708 + }, + { + "epoch": 0.9101412066752247, + "grad_norm": 0.3044429421424866, + "learning_rate": 4.513627831377365e-05, + "loss": 2.4196, + "step": 709 + }, + { + "epoch": 0.9114249037227214, + "grad_norm": 0.30725550651550293, + "learning_rate": 4.5120833844817116e-05, + "loss": 2.2667, + "step": 710 + }, + { + "epoch": 0.9127086007702182, + "grad_norm": 0.29363924264907837, + "learning_rate": 4.510536754438923e-05, + "loss": 2.3684, + "step": 711 + }, + { + "epoch": 0.9139922978177151, + "grad_norm": 0.31257641315460205, + "learning_rate": 4.50898794292712e-05, + "loss": 2.4096, + "step": 712 + }, + { + "epoch": 0.9152759948652118, + "grad_norm": 0.3012520670890808, + "learning_rate": 4.507436951626792e-05, + "loss": 2.3477, + "step": 713 + }, + { + "epoch": 0.9165596919127086, + "grad_norm": 0.2953280508518219, + "learning_rate": 4.505883782220793e-05, + "loss": 2.4028, + "step": 714 + }, + { + "epoch": 0.9178433889602053, + "grad_norm": 0.3054860830307007, + "learning_rate": 4.504328436394339e-05, + "loss": 2.3845, + "step": 715 + }, + { + "epoch": 0.9191270860077022, + "grad_norm": 0.31432589888572693, + "learning_rate": 4.50277091583501e-05, + "loss": 2.3838, + "step": 716 + }, + { + "epoch": 0.920410783055199, + "grad_norm": 0.34039193391799927, + "learning_rate": 4.5012112222327434e-05, + "loss": 2.3927, + "step": 717 + }, + { + "epoch": 0.9216944801026957, + "grad_norm": 0.3081609308719635, + "learning_rate": 4.4996493572798356e-05, + "loss": 2.3497, + "step": 718 + }, + { + "epoch": 0.9229781771501926, + "grad_norm": 0.3027456998825073, + "learning_rate": 4.498085322670938e-05, + "loss": 2.2801, + "step": 719 + }, + { + "epoch": 0.9242618741976893, + "grad_norm": 0.29661089181900024, + "learning_rate": 4.496519120103057e-05, + "loss": 2.3386, + "step": 720 + }, + { + "epoch": 0.9255455712451861, + "grad_norm": 0.3082757592201233, + "learning_rate": 4.494950751275551e-05, + "loss": 2.2882, + "step": 721 + }, + { + "epoch": 0.926829268292683, + "grad_norm": 0.3266795575618744, + "learning_rate": 4.493380217890128e-05, + "loss": 2.4543, + "step": 722 + }, + { + "epoch": 0.9281129653401797, + "grad_norm": 0.30967482924461365, + "learning_rate": 4.4918075216508456e-05, + "loss": 2.3665, + "step": 723 + }, + { + "epoch": 0.9293966623876765, + "grad_norm": 0.3447597920894623, + "learning_rate": 4.4902326642641095e-05, + "loss": 2.36, + "step": 724 + }, + { + "epoch": 0.9306803594351734, + "grad_norm": 0.3120889663696289, + "learning_rate": 4.488655647438667e-05, + "loss": 2.3515, + "step": 725 + }, + { + "epoch": 0.9319640564826701, + "grad_norm": 0.29375332593917847, + "learning_rate": 4.48707647288561e-05, + "loss": 2.3017, + "step": 726 + }, + { + "epoch": 0.9332477535301669, + "grad_norm": 0.30576878786087036, + "learning_rate": 4.485495142318372e-05, + "loss": 2.4154, + "step": 727 + }, + { + "epoch": 0.9345314505776636, + "grad_norm": 0.26723507046699524, + "learning_rate": 4.483911657452726e-05, + "loss": 2.2889, + "step": 728 + }, + { + "epoch": 0.9358151476251605, + "grad_norm": 0.3043777644634247, + "learning_rate": 4.482326020006782e-05, + "loss": 2.3097, + "step": 729 + }, + { + "epoch": 0.9370988446726572, + "grad_norm": 0.28958752751350403, + "learning_rate": 4.480738231700985e-05, + "loss": 2.2452, + "step": 730 + }, + { + "epoch": 0.938382541720154, + "grad_norm": 0.2912410795688629, + "learning_rate": 4.4791482942581134e-05, + "loss": 2.2437, + "step": 731 + }, + { + "epoch": 0.9396662387676509, + "grad_norm": 0.29347458481788635, + "learning_rate": 4.4775562094032795e-05, + "loss": 2.3602, + "step": 732 + }, + { + "epoch": 0.9409499358151476, + "grad_norm": 0.29266172647476196, + "learning_rate": 4.4759619788639244e-05, + "loss": 2.3348, + "step": 733 + }, + { + "epoch": 0.9422336328626444, + "grad_norm": 0.29403918981552124, + "learning_rate": 4.474365604369817e-05, + "loss": 2.3746, + "step": 734 + }, + { + "epoch": 0.9435173299101413, + "grad_norm": 0.30329957604408264, + "learning_rate": 4.4727670876530524e-05, + "loss": 2.3238, + "step": 735 + }, + { + "epoch": 0.944801026957638, + "grad_norm": 0.2870703339576721, + "learning_rate": 4.47116643044805e-05, + "loss": 2.3218, + "step": 736 + }, + { + "epoch": 0.9460847240051348, + "grad_norm": 0.3277876377105713, + "learning_rate": 4.469563634491554e-05, + "loss": 2.2395, + "step": 737 + }, + { + "epoch": 0.9473684210526315, + "grad_norm": 0.31081393361091614, + "learning_rate": 4.4679587015226253e-05, + "loss": 2.3659, + "step": 738 + }, + { + "epoch": 0.9486521181001284, + "grad_norm": 0.3004544675350189, + "learning_rate": 4.4663516332826466e-05, + "loss": 2.3085, + "step": 739 + }, + { + "epoch": 0.9499358151476252, + "grad_norm": 0.3048396706581116, + "learning_rate": 4.464742431515315e-05, + "loss": 2.4874, + "step": 740 + }, + { + "epoch": 0.9512195121951219, + "grad_norm": 0.2936103045940399, + "learning_rate": 4.463131097966644e-05, + "loss": 2.3382, + "step": 741 + }, + { + "epoch": 0.9525032092426188, + "grad_norm": 0.3437049090862274, + "learning_rate": 4.461517634384961e-05, + "loss": 2.3761, + "step": 742 + }, + { + "epoch": 0.9537869062901155, + "grad_norm": 0.30605414509773254, + "learning_rate": 4.459902042520903e-05, + "loss": 2.3129, + "step": 743 + }, + { + "epoch": 0.9550706033376123, + "grad_norm": 0.3019481599330902, + "learning_rate": 4.458284324127415e-05, + "loss": 2.2858, + "step": 744 + }, + { + "epoch": 0.9563543003851092, + "grad_norm": 0.29308027029037476, + "learning_rate": 4.456664480959752e-05, + "loss": 2.3391, + "step": 745 + }, + { + "epoch": 0.9576379974326059, + "grad_norm": 0.30717384815216064, + "learning_rate": 4.455042514775475e-05, + "loss": 2.4426, + "step": 746 + }, + { + "epoch": 0.9589216944801027, + "grad_norm": 0.3070937395095825, + "learning_rate": 4.4534184273344445e-05, + "loss": 2.3566, + "step": 747 + }, + { + "epoch": 0.9602053915275995, + "grad_norm": 0.3065012991428375, + "learning_rate": 4.451792220398826e-05, + "loss": 2.3023, + "step": 748 + }, + { + "epoch": 0.9614890885750963, + "grad_norm": 0.32163599133491516, + "learning_rate": 4.450163895733084e-05, + "loss": 2.3961, + "step": 749 + }, + { + "epoch": 0.962772785622593, + "grad_norm": 0.4026564359664917, + "learning_rate": 4.448533455103979e-05, + "loss": 2.3264, + "step": 750 + }, + { + "epoch": 0.9640564826700898, + "grad_norm": 0.31173425912857056, + "learning_rate": 4.446900900280571e-05, + "loss": 2.4026, + "step": 751 + }, + { + "epoch": 0.9653401797175867, + "grad_norm": 0.2894265949726105, + "learning_rate": 4.445266233034209e-05, + "loss": 2.3748, + "step": 752 + }, + { + "epoch": 0.9666238767650834, + "grad_norm": 0.3028358817100525, + "learning_rate": 4.443629455138539e-05, + "loss": 2.4017, + "step": 753 + }, + { + "epoch": 0.9679075738125802, + "grad_norm": 0.31508946418762207, + "learning_rate": 4.441990568369494e-05, + "loss": 2.4487, + "step": 754 + }, + { + "epoch": 0.9691912708600771, + "grad_norm": 0.29983460903167725, + "learning_rate": 4.440349574505296e-05, + "loss": 2.3995, + "step": 755 + }, + { + "epoch": 0.9704749679075738, + "grad_norm": 0.31416934728622437, + "learning_rate": 4.438706475326453e-05, + "loss": 2.2832, + "step": 756 + }, + { + "epoch": 0.9717586649550706, + "grad_norm": 0.31629544496536255, + "learning_rate": 4.437061272615759e-05, + "loss": 2.4038, + "step": 757 + }, + { + "epoch": 0.9730423620025674, + "grad_norm": 0.31430545449256897, + "learning_rate": 4.4354139681582865e-05, + "loss": 2.3577, + "step": 758 + }, + { + "epoch": 0.9743260590500642, + "grad_norm": 0.3069208860397339, + "learning_rate": 4.4337645637413926e-05, + "loss": 2.3681, + "step": 759 + }, + { + "epoch": 0.975609756097561, + "grad_norm": 0.3285858631134033, + "learning_rate": 4.4321130611547116e-05, + "loss": 2.3895, + "step": 760 + }, + { + "epoch": 0.9768934531450578, + "grad_norm": 0.3070060610771179, + "learning_rate": 4.430459462190151e-05, + "loss": 2.297, + "step": 761 + }, + { + "epoch": 0.9781771501925546, + "grad_norm": 0.2973722219467163, + "learning_rate": 4.4288037686418995e-05, + "loss": 2.3951, + "step": 762 + }, + { + "epoch": 0.9794608472400513, + "grad_norm": 0.29620617628097534, + "learning_rate": 4.427145982306412e-05, + "loss": 2.4088, + "step": 763 + }, + { + "epoch": 0.9807445442875481, + "grad_norm": 0.30031436681747437, + "learning_rate": 4.425486104982418e-05, + "loss": 2.3678, + "step": 764 + }, + { + "epoch": 0.982028241335045, + "grad_norm": 0.312211275100708, + "learning_rate": 4.4238241384709156e-05, + "loss": 2.4162, + "step": 765 + }, + { + "epoch": 0.9833119383825417, + "grad_norm": 0.289779931306839, + "learning_rate": 4.422160084575167e-05, + "loss": 2.342, + "step": 766 + }, + { + "epoch": 0.9845956354300385, + "grad_norm": 0.3106563687324524, + "learning_rate": 4.420493945100702e-05, + "loss": 2.3842, + "step": 767 + }, + { + "epoch": 0.9858793324775353, + "grad_norm": 0.31197184324264526, + "learning_rate": 4.418825721855311e-05, + "loss": 2.353, + "step": 768 + }, + { + "epoch": 0.9871630295250321, + "grad_norm": 0.32115286588668823, + "learning_rate": 4.417155416649049e-05, + "loss": 2.3355, + "step": 769 + }, + { + "epoch": 0.9884467265725289, + "grad_norm": 0.3096913695335388, + "learning_rate": 4.415483031294225e-05, + "loss": 2.4134, + "step": 770 + }, + { + "epoch": 0.9897304236200257, + "grad_norm": 0.28306904435157776, + "learning_rate": 4.41380856760541e-05, + "loss": 2.3534, + "step": 771 + }, + { + "epoch": 0.9910141206675225, + "grad_norm": 0.2894434630870819, + "learning_rate": 4.412132027399426e-05, + "loss": 2.3527, + "step": 772 + }, + { + "epoch": 0.9922978177150192, + "grad_norm": 0.31455811858177185, + "learning_rate": 4.41045341249535e-05, + "loss": 2.4085, + "step": 773 + }, + { + "epoch": 0.993581514762516, + "grad_norm": 0.30663493275642395, + "learning_rate": 4.408772724714509e-05, + "loss": 2.3677, + "step": 774 + }, + { + "epoch": 0.9948652118100129, + "grad_norm": 0.3203311264514923, + "learning_rate": 4.407089965880482e-05, + "loss": 2.3208, + "step": 775 + }, + { + "epoch": 0.9961489088575096, + "grad_norm": 0.31689587235450745, + "learning_rate": 4.4054051378190915e-05, + "loss": 2.4329, + "step": 776 + }, + { + "epoch": 0.9974326059050064, + "grad_norm": 0.3117218017578125, + "learning_rate": 4.403718242358407e-05, + "loss": 2.4214, + "step": 777 + }, + { + "epoch": 0.9987163029525032, + "grad_norm": 0.3188619017601013, + "learning_rate": 4.40202928132874e-05, + "loss": 2.3655, + "step": 778 + }, + { + "epoch": 1.0, + "grad_norm": 0.30813828110694885, + "learning_rate": 4.400338256562645e-05, + "loss": 2.4338, + "step": 779 + }, + { + "epoch": 1.0012836970474968, + "grad_norm": 0.3029801845550537, + "learning_rate": 4.398645169894914e-05, + "loss": 2.3734, + "step": 780 + }, + { + "epoch": 1.0012836970474968, + "grad_norm": 0.2845710217952728, + "learning_rate": 4.3969500231625774e-05, + "loss": 2.2655, + "step": 781 + }, + { + "epoch": 1.0025673940949935, + "grad_norm": 0.28304269909858704, + "learning_rate": 4.3952528182049e-05, + "loss": 2.2808, + "step": 782 + }, + { + "epoch": 1.0038510911424903, + "grad_norm": 0.29494890570640564, + "learning_rate": 4.393553556863381e-05, + "loss": 2.1482, + "step": 783 + }, + { + "epoch": 1.0051347881899872, + "grad_norm": 0.27522915601730347, + "learning_rate": 4.391852240981749e-05, + "loss": 2.2518, + "step": 784 + }, + { + "epoch": 1.006418485237484, + "grad_norm": 0.3148289620876312, + "learning_rate": 4.390148872405963e-05, + "loss": 2.2193, + "step": 785 + }, + { + "epoch": 1.0077021822849808, + "grad_norm": 0.31480056047439575, + "learning_rate": 4.3884434529842076e-05, + "loss": 2.1945, + "step": 786 + }, + { + "epoch": 1.0089858793324775, + "grad_norm": 0.30653226375579834, + "learning_rate": 4.386735984566896e-05, + "loss": 2.224, + "step": 787 + }, + { + "epoch": 1.0102695763799743, + "grad_norm": 0.31125813722610474, + "learning_rate": 4.385026469006662e-05, + "loss": 2.3276, + "step": 788 + }, + { + "epoch": 1.011553273427471, + "grad_norm": 0.2961493730545044, + "learning_rate": 4.3833149081583604e-05, + "loss": 2.2199, + "step": 789 + }, + { + "epoch": 1.012836970474968, + "grad_norm": 0.298798143863678, + "learning_rate": 4.381601303879066e-05, + "loss": 2.2086, + "step": 790 + }, + { + "epoch": 1.0141206675224648, + "grad_norm": 0.336722195148468, + "learning_rate": 4.3798856580280715e-05, + "loss": 2.2231, + "step": 791 + }, + { + "epoch": 1.0154043645699615, + "grad_norm": 0.3271631896495819, + "learning_rate": 4.378167972466884e-05, + "loss": 2.2672, + "step": 792 + }, + { + "epoch": 1.0166880616174583, + "grad_norm": 0.28419965505599976, + "learning_rate": 4.376448249059222e-05, + "loss": 2.2158, + "step": 793 + }, + { + "epoch": 1.017971758664955, + "grad_norm": 0.3065083920955658, + "learning_rate": 4.3747264896710185e-05, + "loss": 2.1897, + "step": 794 + }, + { + "epoch": 1.0192554557124518, + "grad_norm": 0.31983664631843567, + "learning_rate": 4.3730026961704126e-05, + "loss": 2.2348, + "step": 795 + }, + { + "epoch": 1.0205391527599486, + "grad_norm": 0.3390856981277466, + "learning_rate": 4.371276870427753e-05, + "loss": 2.21, + "step": 796 + }, + { + "epoch": 1.0218228498074455, + "grad_norm": 0.31926363706588745, + "learning_rate": 4.36954901431559e-05, + "loss": 2.2628, + "step": 797 + }, + { + "epoch": 1.0231065468549423, + "grad_norm": 0.33794736862182617, + "learning_rate": 4.367819129708682e-05, + "loss": 2.229, + "step": 798 + }, + { + "epoch": 1.024390243902439, + "grad_norm": 0.33381831645965576, + "learning_rate": 4.3660872184839825e-05, + "loss": 2.2948, + "step": 799 + }, + { + "epoch": 1.0256739409499358, + "grad_norm": 0.32239073514938354, + "learning_rate": 4.364353282520648e-05, + "loss": 2.2511, + "step": 800 + }, + { + "epoch": 1.0269576379974326, + "grad_norm": 0.3180272877216339, + "learning_rate": 4.36261732370003e-05, + "loss": 2.1619, + "step": 801 + }, + { + "epoch": 1.0282413350449293, + "grad_norm": 0.33113232254981995, + "learning_rate": 4.360879343905676e-05, + "loss": 2.2182, + "step": 802 + }, + { + "epoch": 1.0295250320924263, + "grad_norm": 0.3161705434322357, + "learning_rate": 4.359139345023325e-05, + "loss": 2.1993, + "step": 803 + }, + { + "epoch": 1.030808729139923, + "grad_norm": 0.3182056248188019, + "learning_rate": 4.3573973289409094e-05, + "loss": 2.2228, + "step": 804 + }, + { + "epoch": 1.0320924261874198, + "grad_norm": 0.3930760324001312, + "learning_rate": 4.355653297548546e-05, + "loss": 2.2691, + "step": 805 + }, + { + "epoch": 1.0333761232349166, + "grad_norm": 0.31772395968437195, + "learning_rate": 4.353907252738543e-05, + "loss": 2.2816, + "step": 806 + }, + { + "epoch": 1.0346598202824133, + "grad_norm": 0.36085692048072815, + "learning_rate": 4.352159196405388e-05, + "loss": 2.2951, + "step": 807 + }, + { + "epoch": 1.03594351732991, + "grad_norm": 0.3501516580581665, + "learning_rate": 4.350409130445756e-05, + "loss": 2.2143, + "step": 808 + }, + { + "epoch": 1.0372272143774068, + "grad_norm": 0.3293107748031616, + "learning_rate": 4.3486570567585003e-05, + "loss": 2.2479, + "step": 809 + }, + { + "epoch": 1.0385109114249038, + "grad_norm": 0.31279006600379944, + "learning_rate": 4.3469029772446525e-05, + "loss": 2.2152, + "step": 810 + }, + { + "epoch": 1.0397946084724006, + "grad_norm": 0.34608227014541626, + "learning_rate": 4.3451468938074205e-05, + "loss": 2.173, + "step": 811 + }, + { + "epoch": 1.0410783055198973, + "grad_norm": 0.317304402589798, + "learning_rate": 4.343388808352187e-05, + "loss": 2.2071, + "step": 812 + }, + { + "epoch": 1.042362002567394, + "grad_norm": 0.3161017596721649, + "learning_rate": 4.3416287227865074e-05, + "loss": 2.2746, + "step": 813 + }, + { + "epoch": 1.0436456996148908, + "grad_norm": 0.3690197765827179, + "learning_rate": 4.339866639020106e-05, + "loss": 2.3066, + "step": 814 + }, + { + "epoch": 1.0449293966623876, + "grad_norm": 0.328646183013916, + "learning_rate": 4.338102558964876e-05, + "loss": 2.2019, + "step": 815 + }, + { + "epoch": 1.0462130937098846, + "grad_norm": 0.34105199575424194, + "learning_rate": 4.3363364845348755e-05, + "loss": 2.2957, + "step": 816 + }, + { + "epoch": 1.0474967907573813, + "grad_norm": 0.3383546769618988, + "learning_rate": 4.3345684176463286e-05, + "loss": 2.2092, + "step": 817 + }, + { + "epoch": 1.048780487804878, + "grad_norm": 0.3505076766014099, + "learning_rate": 4.3327983602176196e-05, + "loss": 2.1598, + "step": 818 + }, + { + "epoch": 1.0500641848523748, + "grad_norm": 0.35122519731521606, + "learning_rate": 4.3310263141692935e-05, + "loss": 2.174, + "step": 819 + }, + { + "epoch": 1.0513478818998716, + "grad_norm": 0.36835214495658875, + "learning_rate": 4.329252281424052e-05, + "loss": 2.2067, + "step": 820 + }, + { + "epoch": 1.0526315789473684, + "grad_norm": 0.3246287405490875, + "learning_rate": 4.327476263906753e-05, + "loss": 2.2666, + "step": 821 + }, + { + "epoch": 1.0539152759948651, + "grad_norm": 0.355728417634964, + "learning_rate": 4.325698263544409e-05, + "loss": 2.3094, + "step": 822 + }, + { + "epoch": 1.055198973042362, + "grad_norm": 0.35884207487106323, + "learning_rate": 4.3239182822661805e-05, + "loss": 2.2344, + "step": 823 + }, + { + "epoch": 1.0564826700898589, + "grad_norm": 0.37921926379203796, + "learning_rate": 4.3221363220033814e-05, + "loss": 2.2099, + "step": 824 + }, + { + "epoch": 1.0577663671373556, + "grad_norm": 0.36717137694358826, + "learning_rate": 4.320352384689471e-05, + "loss": 2.2276, + "step": 825 + }, + { + "epoch": 1.0590500641848524, + "grad_norm": 0.35223424434661865, + "learning_rate": 4.318566472260054e-05, + "loss": 2.2718, + "step": 826 + }, + { + "epoch": 1.0603337612323491, + "grad_norm": 0.321180135011673, + "learning_rate": 4.316778586652876e-05, + "loss": 2.2637, + "step": 827 + }, + { + "epoch": 1.0616174582798459, + "grad_norm": 0.36485522985458374, + "learning_rate": 4.3149887298078276e-05, + "loss": 2.2069, + "step": 828 + }, + { + "epoch": 1.0629011553273426, + "grad_norm": 0.3567776381969452, + "learning_rate": 4.313196903666935e-05, + "loss": 2.2394, + "step": 829 + }, + { + "epoch": 1.0641848523748396, + "grad_norm": 0.3673252463340759, + "learning_rate": 4.3114031101743625e-05, + "loss": 2.2388, + "step": 830 + }, + { + "epoch": 1.0654685494223364, + "grad_norm": 0.327640175819397, + "learning_rate": 4.309607351276409e-05, + "loss": 2.2847, + "step": 831 + }, + { + "epoch": 1.0667522464698331, + "grad_norm": 0.35278892517089844, + "learning_rate": 4.307809628921505e-05, + "loss": 2.1972, + "step": 832 + }, + { + "epoch": 1.0680359435173299, + "grad_norm": 0.36460593342781067, + "learning_rate": 4.306009945060212e-05, + "loss": 2.1976, + "step": 833 + }, + { + "epoch": 1.0693196405648266, + "grad_norm": 0.38041952252388, + "learning_rate": 4.3042083016452205e-05, + "loss": 2.2009, + "step": 834 + }, + { + "epoch": 1.0706033376123234, + "grad_norm": 0.3453562259674072, + "learning_rate": 4.302404700631346e-05, + "loss": 2.2788, + "step": 835 + }, + { + "epoch": 1.0718870346598204, + "grad_norm": 0.3455585539340973, + "learning_rate": 4.300599143975529e-05, + "loss": 2.2418, + "step": 836 + }, + { + "epoch": 1.0731707317073171, + "grad_norm": 0.3436455726623535, + "learning_rate": 4.298791633636831e-05, + "loss": 2.2121, + "step": 837 + }, + { + "epoch": 1.074454428754814, + "grad_norm": 0.35627031326293945, + "learning_rate": 4.2969821715764335e-05, + "loss": 2.255, + "step": 838 + }, + { + "epoch": 1.0757381258023107, + "grad_norm": 0.3669643998146057, + "learning_rate": 4.2951707597576354e-05, + "loss": 2.2815, + "step": 839 + }, + { + "epoch": 1.0770218228498074, + "grad_norm": 0.40819859504699707, + "learning_rate": 4.2933574001458524e-05, + "loss": 2.2359, + "step": 840 + }, + { + "epoch": 1.0783055198973042, + "grad_norm": 0.37338659167289734, + "learning_rate": 4.291542094708612e-05, + "loss": 2.1824, + "step": 841 + }, + { + "epoch": 1.0795892169448011, + "grad_norm": 0.33379244804382324, + "learning_rate": 4.2897248454155544e-05, + "loss": 2.2073, + "step": 842 + }, + { + "epoch": 1.080872913992298, + "grad_norm": 0.3620593845844269, + "learning_rate": 4.2879056542384274e-05, + "loss": 2.2447, + "step": 843 + }, + { + "epoch": 1.0821566110397947, + "grad_norm": 0.3987523913383484, + "learning_rate": 4.2860845231510873e-05, + "loss": 2.2799, + "step": 844 + }, + { + "epoch": 1.0834403080872914, + "grad_norm": 0.3315846621990204, + "learning_rate": 4.2842614541294935e-05, + "loss": 2.2696, + "step": 845 + }, + { + "epoch": 1.0847240051347882, + "grad_norm": 0.3277621865272522, + "learning_rate": 4.282436449151711e-05, + "loss": 2.201, + "step": 846 + }, + { + "epoch": 1.086007702182285, + "grad_norm": 0.3634556233882904, + "learning_rate": 4.2806095101979016e-05, + "loss": 2.2588, + "step": 847 + }, + { + "epoch": 1.0872913992297817, + "grad_norm": 0.3708522915840149, + "learning_rate": 4.278780639250328e-05, + "loss": 2.2732, + "step": 848 + }, + { + "epoch": 1.0885750962772787, + "grad_norm": 0.34317970275878906, + "learning_rate": 4.2769498382933484e-05, + "loss": 2.2507, + "step": 849 + }, + { + "epoch": 1.0898587933247754, + "grad_norm": 0.3626539707183838, + "learning_rate": 4.275117109313415e-05, + "loss": 2.2375, + "step": 850 + }, + { + "epoch": 1.0911424903722722, + "grad_norm": 0.34370511770248413, + "learning_rate": 4.273282454299073e-05, + "loss": 2.3254, + "step": 851 + }, + { + "epoch": 1.092426187419769, + "grad_norm": 0.3586619198322296, + "learning_rate": 4.271445875240956e-05, + "loss": 2.2233, + "step": 852 + }, + { + "epoch": 1.0937098844672657, + "grad_norm": 0.37432125210762024, + "learning_rate": 4.2696073741317854e-05, + "loss": 2.265, + "step": 853 + }, + { + "epoch": 1.0949935815147624, + "grad_norm": 0.3803035616874695, + "learning_rate": 4.267766952966369e-05, + "loss": 2.2271, + "step": 854 + }, + { + "epoch": 1.0962772785622592, + "grad_norm": 0.3703853189945221, + "learning_rate": 4.265924613741596e-05, + "loss": 2.2264, + "step": 855 + }, + { + "epoch": 1.0975609756097562, + "grad_norm": 0.32208389043807983, + "learning_rate": 4.264080358456441e-05, + "loss": 2.1602, + "step": 856 + }, + { + "epoch": 1.098844672657253, + "grad_norm": 0.34176668524742126, + "learning_rate": 4.2622341891119514e-05, + "loss": 2.2398, + "step": 857 + }, + { + "epoch": 1.1001283697047497, + "grad_norm": 0.3376818895339966, + "learning_rate": 4.260386107711256e-05, + "loss": 2.1697, + "step": 858 + }, + { + "epoch": 1.1014120667522465, + "grad_norm": 0.33973339200019836, + "learning_rate": 4.258536116259557e-05, + "loss": 2.2835, + "step": 859 + }, + { + "epoch": 1.1026957637997432, + "grad_norm": 0.3520197570323944, + "learning_rate": 4.256684216764127e-05, + "loss": 2.2117, + "step": 860 + }, + { + "epoch": 1.10397946084724, + "grad_norm": 0.36753150820732117, + "learning_rate": 4.2548304112343143e-05, + "loss": 2.1906, + "step": 861 + }, + { + "epoch": 1.1052631578947367, + "grad_norm": 0.40462440252304077, + "learning_rate": 4.252974701681528e-05, + "loss": 2.2117, + "step": 862 + }, + { + "epoch": 1.1065468549422337, + "grad_norm": 0.36666756868362427, + "learning_rate": 4.251117090119249e-05, + "loss": 2.3043, + "step": 863 + }, + { + "epoch": 1.1078305519897305, + "grad_norm": 0.34417638182640076, + "learning_rate": 4.249257578563019e-05, + "loss": 2.2083, + "step": 864 + }, + { + "epoch": 1.1091142490372272, + "grad_norm": 0.35194942355155945, + "learning_rate": 4.2473961690304415e-05, + "loss": 2.2115, + "step": 865 + }, + { + "epoch": 1.110397946084724, + "grad_norm": 0.33601894974708557, + "learning_rate": 4.24553286354118e-05, + "loss": 2.2619, + "step": 866 + }, + { + "epoch": 1.1116816431322207, + "grad_norm": 0.359961599111557, + "learning_rate": 4.243667664116956e-05, + "loss": 2.2579, + "step": 867 + }, + { + "epoch": 1.1129653401797175, + "grad_norm": 0.34904807806015015, + "learning_rate": 4.241800572781543e-05, + "loss": 2.2111, + "step": 868 + }, + { + "epoch": 1.1142490372272145, + "grad_norm": 0.3385773003101349, + "learning_rate": 4.2399315915607705e-05, + "loss": 2.2261, + "step": 869 + }, + { + "epoch": 1.1155327342747112, + "grad_norm": 0.387919157743454, + "learning_rate": 4.238060722482517e-05, + "loss": 2.2792, + "step": 870 + }, + { + "epoch": 1.116816431322208, + "grad_norm": 0.34467244148254395, + "learning_rate": 4.2361879675767094e-05, + "loss": 2.2175, + "step": 871 + }, + { + "epoch": 1.1181001283697047, + "grad_norm": 0.3811795115470886, + "learning_rate": 4.2343133288753194e-05, + "loss": 2.1043, + "step": 872 + }, + { + "epoch": 1.1193838254172015, + "grad_norm": 0.35866042971611023, + "learning_rate": 4.232436808412367e-05, + "loss": 2.1949, + "step": 873 + }, + { + "epoch": 1.1206675224646983, + "grad_norm": 0.3329986035823822, + "learning_rate": 4.230558408223909e-05, + "loss": 2.2682, + "step": 874 + }, + { + "epoch": 1.1219512195121952, + "grad_norm": 0.3897240459918976, + "learning_rate": 4.228678130348045e-05, + "loss": 2.1858, + "step": 875 + }, + { + "epoch": 1.123234916559692, + "grad_norm": 0.3876970112323761, + "learning_rate": 4.22679597682491e-05, + "loss": 2.2581, + "step": 876 + }, + { + "epoch": 1.1245186136071887, + "grad_norm": 0.3848673403263092, + "learning_rate": 4.2249119496966746e-05, + "loss": 2.2538, + "step": 877 + }, + { + "epoch": 1.1258023106546855, + "grad_norm": 0.35312554240226746, + "learning_rate": 4.223026051007544e-05, + "loss": 2.1958, + "step": 878 + }, + { + "epoch": 1.1270860077021823, + "grad_norm": 0.40287381410598755, + "learning_rate": 4.2211382828037506e-05, + "loss": 2.2334, + "step": 879 + }, + { + "epoch": 1.128369704749679, + "grad_norm": 0.3706461787223816, + "learning_rate": 4.2192486471335585e-05, + "loss": 2.1225, + "step": 880 + }, + { + "epoch": 1.1296534017971758, + "grad_norm": 0.36967286467552185, + "learning_rate": 4.2173571460472575e-05, + "loss": 2.2301, + "step": 881 + }, + { + "epoch": 1.1309370988446728, + "grad_norm": 0.3459497392177582, + "learning_rate": 4.215463781597159e-05, + "loss": 2.2297, + "step": 882 + }, + { + "epoch": 1.1322207958921695, + "grad_norm": 0.3627232313156128, + "learning_rate": 4.2135685558375994e-05, + "loss": 2.179, + "step": 883 + }, + { + "epoch": 1.1335044929396663, + "grad_norm": 0.38214111328125, + "learning_rate": 4.211671470824933e-05, + "loss": 2.3036, + "step": 884 + }, + { + "epoch": 1.134788189987163, + "grad_norm": 0.35911989212036133, + "learning_rate": 4.209772528617531e-05, + "loss": 2.1972, + "step": 885 + }, + { + "epoch": 1.1360718870346598, + "grad_norm": 0.35253652930259705, + "learning_rate": 4.207871731275781e-05, + "loss": 2.2036, + "step": 886 + }, + { + "epoch": 1.1373555840821565, + "grad_norm": 0.36945071816444397, + "learning_rate": 4.205969080862082e-05, + "loss": 2.2582, + "step": 887 + }, + { + "epoch": 1.1386392811296533, + "grad_norm": 0.36074042320251465, + "learning_rate": 4.2040645794408445e-05, + "loss": 2.2153, + "step": 888 + }, + { + "epoch": 1.1399229781771503, + "grad_norm": 0.34374964237213135, + "learning_rate": 4.202158229078488e-05, + "loss": 2.2856, + "step": 889 + }, + { + "epoch": 1.141206675224647, + "grad_norm": 0.3783753514289856, + "learning_rate": 4.200250031843436e-05, + "loss": 2.1796, + "step": 890 + }, + { + "epoch": 1.1424903722721438, + "grad_norm": 0.4086593687534332, + "learning_rate": 4.1983399898061186e-05, + "loss": 2.2755, + "step": 891 + }, + { + "epoch": 1.1437740693196405, + "grad_norm": 0.3416132926940918, + "learning_rate": 4.196428105038966e-05, + "loss": 2.2646, + "step": 892 + }, + { + "epoch": 1.1450577663671373, + "grad_norm": 0.359719455242157, + "learning_rate": 4.1945143796164076e-05, + "loss": 2.2193, + "step": 893 + }, + { + "epoch": 1.146341463414634, + "grad_norm": 0.4014301598072052, + "learning_rate": 4.19259881561487e-05, + "loss": 2.2917, + "step": 894 + }, + { + "epoch": 1.147625160462131, + "grad_norm": 0.342040091753006, + "learning_rate": 4.190681415112776e-05, + "loss": 2.2541, + "step": 895 + }, + { + "epoch": 1.1489088575096278, + "grad_norm": 0.3607075810432434, + "learning_rate": 4.1887621801905396e-05, + "loss": 2.2114, + "step": 896 + }, + { + "epoch": 1.1501925545571245, + "grad_norm": 0.34826815128326416, + "learning_rate": 4.186841112930565e-05, + "loss": 2.1932, + "step": 897 + }, + { + "epoch": 1.1514762516046213, + "grad_norm": 0.36598455905914307, + "learning_rate": 4.1849182154172475e-05, + "loss": 2.2374, + "step": 898 + }, + { + "epoch": 1.152759948652118, + "grad_norm": 0.34205228090286255, + "learning_rate": 4.1829934897369625e-05, + "loss": 2.1307, + "step": 899 + }, + { + "epoch": 1.1540436456996148, + "grad_norm": 0.3688623607158661, + "learning_rate": 4.181066937978075e-05, + "loss": 2.2279, + "step": 900 + }, + { + "epoch": 1.1553273427471118, + "grad_norm": 0.39433860778808594, + "learning_rate": 4.179138562230928e-05, + "loss": 2.2676, + "step": 901 + }, + { + "epoch": 1.1566110397946086, + "grad_norm": 0.3450692892074585, + "learning_rate": 4.1772083645878444e-05, + "loss": 2.2221, + "step": 902 + }, + { + "epoch": 1.1578947368421053, + "grad_norm": 0.3846975266933441, + "learning_rate": 4.175276347143124e-05, + "loss": 2.2659, + "step": 903 + }, + { + "epoch": 1.159178433889602, + "grad_norm": 0.3393380045890808, + "learning_rate": 4.17334251199304e-05, + "loss": 2.2868, + "step": 904 + }, + { + "epoch": 1.1604621309370988, + "grad_norm": 0.35884889960289, + "learning_rate": 4.17140686123584e-05, + "loss": 2.2106, + "step": 905 + }, + { + "epoch": 1.1617458279845956, + "grad_norm": 0.3687424063682556, + "learning_rate": 4.169469396971739e-05, + "loss": 2.2195, + "step": 906 + }, + { + "epoch": 1.1630295250320923, + "grad_norm": 0.358381450176239, + "learning_rate": 4.167530121302923e-05, + "loss": 2.2163, + "step": 907 + }, + { + "epoch": 1.1643132220795893, + "grad_norm": 0.3487226366996765, + "learning_rate": 4.1655890363335394e-05, + "loss": 2.2896, + "step": 908 + }, + { + "epoch": 1.165596919127086, + "grad_norm": 0.34116798639297485, + "learning_rate": 4.163646144169702e-05, + "loss": 2.2171, + "step": 909 + }, + { + "epoch": 1.1668806161745828, + "grad_norm": 0.34055590629577637, + "learning_rate": 4.1617014469194836e-05, + "loss": 2.242, + "step": 910 + }, + { + "epoch": 1.1681643132220796, + "grad_norm": 0.39334848523139954, + "learning_rate": 4.159754946692916e-05, + "loss": 2.1213, + "step": 911 + }, + { + "epoch": 1.1694480102695763, + "grad_norm": 0.3823232352733612, + "learning_rate": 4.157806645601988e-05, + "loss": 2.1674, + "step": 912 + }, + { + "epoch": 1.170731707317073, + "grad_norm": 0.3423396050930023, + "learning_rate": 4.1558565457606415e-05, + "loss": 2.2422, + "step": 913 + }, + { + "epoch": 1.1720154043645699, + "grad_norm": 0.3720071017742157, + "learning_rate": 4.153904649284771e-05, + "loss": 2.1096, + "step": 914 + }, + { + "epoch": 1.1732991014120668, + "grad_norm": 0.34880882501602173, + "learning_rate": 4.151950958292219e-05, + "loss": 2.2199, + "step": 915 + }, + { + "epoch": 1.1745827984595636, + "grad_norm": 0.3538772165775299, + "learning_rate": 4.149995474902776e-05, + "loss": 2.2033, + "step": 916 + }, + { + "epoch": 1.1758664955070603, + "grad_norm": 0.36488077044487, + "learning_rate": 4.1480382012381766e-05, + "loss": 2.1818, + "step": 917 + }, + { + "epoch": 1.177150192554557, + "grad_norm": 0.3882789611816406, + "learning_rate": 4.1460791394221e-05, + "loss": 2.2456, + "step": 918 + }, + { + "epoch": 1.1784338896020539, + "grad_norm": 0.4123387038707733, + "learning_rate": 4.144118291580161e-05, + "loss": 2.2155, + "step": 919 + }, + { + "epoch": 1.1797175866495506, + "grad_norm": 0.3578166663646698, + "learning_rate": 4.142155659839919e-05, + "loss": 2.3112, + "step": 920 + }, + { + "epoch": 1.1810012836970474, + "grad_norm": 0.3611357510089874, + "learning_rate": 4.1401912463308625e-05, + "loss": 2.1915, + "step": 921 + }, + { + "epoch": 1.1822849807445444, + "grad_norm": 0.3884929120540619, + "learning_rate": 4.138225053184417e-05, + "loss": 2.2782, + "step": 922 + }, + { + "epoch": 1.1835686777920411, + "grad_norm": 0.35503101348876953, + "learning_rate": 4.136257082533938e-05, + "loss": 2.2083, + "step": 923 + }, + { + "epoch": 1.1848523748395379, + "grad_norm": 0.3408321440219879, + "learning_rate": 4.1342873365147074e-05, + "loss": 2.3104, + "step": 924 + }, + { + "epoch": 1.1861360718870346, + "grad_norm": 0.3557950556278229, + "learning_rate": 4.1323158172639385e-05, + "loss": 2.1643, + "step": 925 + }, + { + "epoch": 1.1874197689345314, + "grad_norm": 0.37379875779151917, + "learning_rate": 4.130342526920765e-05, + "loss": 2.2973, + "step": 926 + }, + { + "epoch": 1.1887034659820284, + "grad_norm": 0.37019041180610657, + "learning_rate": 4.128367467626241e-05, + "loss": 2.1679, + "step": 927 + }, + { + "epoch": 1.1899871630295251, + "grad_norm": 0.3340079188346863, + "learning_rate": 4.126390641523344e-05, + "loss": 2.2195, + "step": 928 + }, + { + "epoch": 1.1912708600770219, + "grad_norm": 0.39198464155197144, + "learning_rate": 4.124412050756965e-05, + "loss": 2.2736, + "step": 929 + }, + { + "epoch": 1.1925545571245186, + "grad_norm": 0.37619346380233765, + "learning_rate": 4.122431697473912e-05, + "loss": 2.2903, + "step": 930 + }, + { + "epoch": 1.1938382541720154, + "grad_norm": 0.3682747781276703, + "learning_rate": 4.120449583822904e-05, + "loss": 2.1202, + "step": 931 + }, + { + "epoch": 1.1951219512195121, + "grad_norm": 0.37576183676719666, + "learning_rate": 4.118465711954569e-05, + "loss": 2.2263, + "step": 932 + }, + { + "epoch": 1.196405648267009, + "grad_norm": 0.3458687663078308, + "learning_rate": 4.116480084021447e-05, + "loss": 2.2303, + "step": 933 + }, + { + "epoch": 1.1976893453145059, + "grad_norm": 0.35645976662635803, + "learning_rate": 4.114492702177978e-05, + "loss": 2.3105, + "step": 934 + }, + { + "epoch": 1.1989730423620026, + "grad_norm": 0.3832726776599884, + "learning_rate": 4.1125035685805084e-05, + "loss": 2.2087, + "step": 935 + }, + { + "epoch": 1.2002567394094994, + "grad_norm": 0.3578638732433319, + "learning_rate": 4.1105126853872845e-05, + "loss": 2.2223, + "step": 936 + }, + { + "epoch": 1.2015404364569962, + "grad_norm": 0.3332241177558899, + "learning_rate": 4.108520054758451e-05, + "loss": 2.2631, + "step": 937 + }, + { + "epoch": 1.202824133504493, + "grad_norm": 0.35905030369758606, + "learning_rate": 4.106525678856049e-05, + "loss": 2.2027, + "step": 938 + }, + { + "epoch": 1.2041078305519897, + "grad_norm": 0.3650401532649994, + "learning_rate": 4.1045295598440105e-05, + "loss": 2.2894, + "step": 939 + }, + { + "epoch": 1.2053915275994864, + "grad_norm": 0.38491788506507874, + "learning_rate": 4.102531699888163e-05, + "loss": 2.2929, + "step": 940 + }, + { + "epoch": 1.2066752246469834, + "grad_norm": 0.380605012178421, + "learning_rate": 4.1005321011562206e-05, + "loss": 2.2567, + "step": 941 + }, + { + "epoch": 1.2079589216944802, + "grad_norm": 0.3601861894130707, + "learning_rate": 4.098530765817784e-05, + "loss": 2.0563, + "step": 942 + }, + { + "epoch": 1.209242618741977, + "grad_norm": 0.3581317663192749, + "learning_rate": 4.0965276960443374e-05, + "loss": 2.2161, + "step": 943 + }, + { + "epoch": 1.2105263157894737, + "grad_norm": 0.3650892972946167, + "learning_rate": 4.094522894009251e-05, + "loss": 2.2823, + "step": 944 + }, + { + "epoch": 1.2118100128369704, + "grad_norm": 0.3575249910354614, + "learning_rate": 4.0925163618877695e-05, + "loss": 2.2203, + "step": 945 + }, + { + "epoch": 1.2130937098844672, + "grad_norm": 0.38429054617881775, + "learning_rate": 4.090508101857017e-05, + "loss": 2.2489, + "step": 946 + }, + { + "epoch": 1.214377406931964, + "grad_norm": 0.3500010073184967, + "learning_rate": 4.088498116095991e-05, + "loss": 2.2303, + "step": 947 + }, + { + "epoch": 1.215661103979461, + "grad_norm": 0.36866387724876404, + "learning_rate": 4.086486406785566e-05, + "loss": 2.2372, + "step": 948 + }, + { + "epoch": 1.2169448010269577, + "grad_norm": 0.3696565330028534, + "learning_rate": 4.08447297610848e-05, + "loss": 2.2177, + "step": 949 + }, + { + "epoch": 1.2182284980744544, + "grad_norm": 0.37652555108070374, + "learning_rate": 4.082457826249343e-05, + "loss": 2.2195, + "step": 950 + }, + { + "epoch": 1.2195121951219512, + "grad_norm": 0.3684236407279968, + "learning_rate": 4.08044095939463e-05, + "loss": 2.1949, + "step": 951 + }, + { + "epoch": 1.220795892169448, + "grad_norm": 0.41459155082702637, + "learning_rate": 4.078422377732678e-05, + "loss": 2.2588, + "step": 952 + }, + { + "epoch": 1.2220795892169447, + "grad_norm": 0.3802269995212555, + "learning_rate": 4.0764020834536846e-05, + "loss": 2.1484, + "step": 953 + }, + { + "epoch": 1.2233632862644417, + "grad_norm": 0.38021379709243774, + "learning_rate": 4.074380078749707e-05, + "loss": 2.2062, + "step": 954 + }, + { + "epoch": 1.2246469833119384, + "grad_norm": 0.3608596622943878, + "learning_rate": 4.072356365814655e-05, + "loss": 2.1648, + "step": 955 + }, + { + "epoch": 1.2259306803594352, + "grad_norm": 0.36179718375205994, + "learning_rate": 4.070330946844295e-05, + "loss": 2.2361, + "step": 956 + }, + { + "epoch": 1.227214377406932, + "grad_norm": 0.39312490820884705, + "learning_rate": 4.068303824036244e-05, + "loss": 2.3051, + "step": 957 + }, + { + "epoch": 1.2284980744544287, + "grad_norm": 0.3528019189834595, + "learning_rate": 4.0662749995899666e-05, + "loss": 2.2951, + "step": 958 + }, + { + "epoch": 1.2297817715019255, + "grad_norm": 0.3852112293243408, + "learning_rate": 4.064244475706776e-05, + "loss": 2.3484, + "step": 959 + }, + { + "epoch": 1.2310654685494224, + "grad_norm": 0.3318076431751251, + "learning_rate": 4.0622122545898246e-05, + "loss": 2.2845, + "step": 960 + }, + { + "epoch": 1.2323491655969192, + "grad_norm": 0.3624611794948578, + "learning_rate": 4.0601783384441104e-05, + "loss": 2.1966, + "step": 961 + }, + { + "epoch": 1.233632862644416, + "grad_norm": 0.4128612279891968, + "learning_rate": 4.058142729476471e-05, + "loss": 2.2521, + "step": 962 + }, + { + "epoch": 1.2349165596919127, + "grad_norm": 0.3708478510379791, + "learning_rate": 4.056105429895578e-05, + "loss": 2.1304, + "step": 963 + }, + { + "epoch": 1.2362002567394095, + "grad_norm": 0.36903417110443115, + "learning_rate": 4.054066441911939e-05, + "loss": 2.1398, + "step": 964 + }, + { + "epoch": 1.2374839537869062, + "grad_norm": 0.3842657208442688, + "learning_rate": 4.052025767737893e-05, + "loss": 2.2423, + "step": 965 + }, + { + "epoch": 1.238767650834403, + "grad_norm": 0.34361815452575684, + "learning_rate": 4.049983409587608e-05, + "loss": 2.201, + "step": 966 + }, + { + "epoch": 1.2400513478819, + "grad_norm": 0.3871995508670807, + "learning_rate": 4.0479393696770805e-05, + "loss": 2.2816, + "step": 967 + }, + { + "epoch": 1.2413350449293967, + "grad_norm": 0.4010753333568573, + "learning_rate": 4.045893650224132e-05, + "loss": 2.1793, + "step": 968 + }, + { + "epoch": 1.2426187419768935, + "grad_norm": 0.364837110042572, + "learning_rate": 4.043846253448403e-05, + "loss": 2.235, + "step": 969 + }, + { + "epoch": 1.2439024390243902, + "grad_norm": 0.39040201902389526, + "learning_rate": 4.0417971815713584e-05, + "loss": 2.1954, + "step": 970 + }, + { + "epoch": 1.245186136071887, + "grad_norm": 0.3791297674179077, + "learning_rate": 4.039746436816277e-05, + "loss": 2.258, + "step": 971 + }, + { + "epoch": 1.2464698331193838, + "grad_norm": 0.37748801708221436, + "learning_rate": 4.037694021408255e-05, + "loss": 2.2057, + "step": 972 + }, + { + "epoch": 1.2477535301668805, + "grad_norm": 0.348192036151886, + "learning_rate": 4.0356399375742e-05, + "loss": 2.2646, + "step": 973 + }, + { + "epoch": 1.2490372272143775, + "grad_norm": 0.34801027178764343, + "learning_rate": 4.0335841875428324e-05, + "loss": 2.2689, + "step": 974 + }, + { + "epoch": 1.2503209242618742, + "grad_norm": 0.3744896352291107, + "learning_rate": 4.031526773544676e-05, + "loss": 2.2559, + "step": 975 + }, + { + "epoch": 1.251604621309371, + "grad_norm": 0.36844202876091003, + "learning_rate": 4.029467697812064e-05, + "loss": 2.2393, + "step": 976 + }, + { + "epoch": 1.2528883183568678, + "grad_norm": 0.3810536861419678, + "learning_rate": 4.0274069625791285e-05, + "loss": 2.2695, + "step": 977 + }, + { + "epoch": 1.2541720154043645, + "grad_norm": 0.3671133816242218, + "learning_rate": 4.0253445700818086e-05, + "loss": 2.1974, + "step": 978 + }, + { + "epoch": 1.2554557124518615, + "grad_norm": 0.3648216724395752, + "learning_rate": 4.023280522557836e-05, + "loss": 2.1598, + "step": 979 + }, + { + "epoch": 1.256739409499358, + "grad_norm": 0.3690861463546753, + "learning_rate": 4.021214822246739e-05, + "loss": 2.253, + "step": 980 + }, + { + "epoch": 1.258023106546855, + "grad_norm": 0.3612317442893982, + "learning_rate": 4.019147471389841e-05, + "loss": 2.1809, + "step": 981 + }, + { + "epoch": 1.2593068035943518, + "grad_norm": 0.3894643187522888, + "learning_rate": 4.017078472230255e-05, + "loss": 2.2193, + "step": 982 + }, + { + "epoch": 1.2605905006418485, + "grad_norm": 0.38447320461273193, + "learning_rate": 4.015007827012883e-05, + "loss": 2.2449, + "step": 983 + }, + { + "epoch": 1.2618741976893453, + "grad_norm": 0.41027238965034485, + "learning_rate": 4.012935537984414e-05, + "loss": 2.1993, + "step": 984 + }, + { + "epoch": 1.263157894736842, + "grad_norm": 0.3948719799518585, + "learning_rate": 4.0108616073933164e-05, + "loss": 2.2761, + "step": 985 + }, + { + "epoch": 1.264441591784339, + "grad_norm": 0.3726814091205597, + "learning_rate": 4.0087860374898446e-05, + "loss": 2.2068, + "step": 986 + }, + { + "epoch": 1.2657252888318355, + "grad_norm": 0.3911172151565552, + "learning_rate": 4.00670883052603e-05, + "loss": 2.1476, + "step": 987 + }, + { + "epoch": 1.2670089858793325, + "grad_norm": 0.39758938550949097, + "learning_rate": 4.00462998875568e-05, + "loss": 2.1984, + "step": 988 + }, + { + "epoch": 1.2682926829268293, + "grad_norm": 0.3571755588054657, + "learning_rate": 4.002549514434375e-05, + "loss": 2.1852, + "step": 989 + }, + { + "epoch": 1.269576379974326, + "grad_norm": 0.3718327283859253, + "learning_rate": 4.0004674098194675e-05, + "loss": 2.2064, + "step": 990 + }, + { + "epoch": 1.2708600770218228, + "grad_norm": 0.35757002234458923, + "learning_rate": 3.99838367717008e-05, + "loss": 2.2122, + "step": 991 + }, + { + "epoch": 1.2721437740693196, + "grad_norm": 0.3949007987976074, + "learning_rate": 3.9962983187470994e-05, + "loss": 2.2586, + "step": 992 + }, + { + "epoch": 1.2734274711168165, + "grad_norm": 0.3487754166126251, + "learning_rate": 3.994211336813178e-05, + "loss": 2.2183, + "step": 993 + }, + { + "epoch": 1.2747111681643133, + "grad_norm": 0.36431553959846497, + "learning_rate": 3.9921227336327293e-05, + "loss": 2.1891, + "step": 994 + }, + { + "epoch": 1.27599486521181, + "grad_norm": 0.36330950260162354, + "learning_rate": 3.990032511471926e-05, + "loss": 2.2732, + "step": 995 + }, + { + "epoch": 1.2772785622593068, + "grad_norm": 0.3715732991695404, + "learning_rate": 3.9879406725986965e-05, + "loss": 2.1761, + "step": 996 + }, + { + "epoch": 1.2785622593068036, + "grad_norm": 0.37331831455230713, + "learning_rate": 3.985847219282725e-05, + "loss": 2.3297, + "step": 997 + }, + { + "epoch": 1.2798459563543003, + "grad_norm": 0.36636462807655334, + "learning_rate": 3.983752153795445e-05, + "loss": 2.1766, + "step": 998 + }, + { + "epoch": 1.281129653401797, + "grad_norm": 0.3827669620513916, + "learning_rate": 3.981655478410043e-05, + "loss": 2.2023, + "step": 999 + }, + { + "epoch": 1.282413350449294, + "grad_norm": 0.35590195655822754, + "learning_rate": 3.979557195401447e-05, + "loss": 2.2212, + "step": 1000 + }, + { + "epoch": 1.2836970474967908, + "grad_norm": 0.35328447818756104, + "learning_rate": 3.977457307046335e-05, + "loss": 2.2325, + "step": 1001 + }, + { + "epoch": 1.2849807445442876, + "grad_norm": 0.3608861267566681, + "learning_rate": 3.975355815623124e-05, + "loss": 2.2387, + "step": 1002 + }, + { + "epoch": 1.2862644415917843, + "grad_norm": 0.3597520887851715, + "learning_rate": 3.973252723411969e-05, + "loss": 2.2482, + "step": 1003 + }, + { + "epoch": 1.287548138639281, + "grad_norm": 0.35393795371055603, + "learning_rate": 3.9711480326947636e-05, + "loss": 2.1458, + "step": 1004 + }, + { + "epoch": 1.2888318356867778, + "grad_norm": 0.33810925483703613, + "learning_rate": 3.9690417457551364e-05, + "loss": 2.2283, + "step": 1005 + }, + { + "epoch": 1.2901155327342746, + "grad_norm": 0.38078486919403076, + "learning_rate": 3.966933864878446e-05, + "loss": 2.3029, + "step": 1006 + }, + { + "epoch": 1.2913992297817716, + "grad_norm": 0.3521521985530853, + "learning_rate": 3.9648243923517836e-05, + "loss": 2.211, + "step": 1007 + }, + { + "epoch": 1.2926829268292683, + "grad_norm": 0.37834882736206055, + "learning_rate": 3.962713330463963e-05, + "loss": 2.2147, + "step": 1008 + }, + { + "epoch": 1.293966623876765, + "grad_norm": 0.3604402244091034, + "learning_rate": 3.960600681505526e-05, + "loss": 2.2126, + "step": 1009 + }, + { + "epoch": 1.2952503209242618, + "grad_norm": 0.3602832555770874, + "learning_rate": 3.958486447768736e-05, + "loss": 2.2646, + "step": 1010 + }, + { + "epoch": 1.2965340179717586, + "grad_norm": 0.368082731962204, + "learning_rate": 3.9563706315475726e-05, + "loss": 2.235, + "step": 1011 + }, + { + "epoch": 1.2978177150192556, + "grad_norm": 0.3699602484703064, + "learning_rate": 3.954253235137737e-05, + "loss": 2.1595, + "step": 1012 + }, + { + "epoch": 1.2991014120667521, + "grad_norm": 0.3776682913303375, + "learning_rate": 3.9521342608366424e-05, + "loss": 2.1841, + "step": 1013 + }, + { + "epoch": 1.300385109114249, + "grad_norm": 0.40402621030807495, + "learning_rate": 3.950013710943415e-05, + "loss": 2.3433, + "step": 1014 + }, + { + "epoch": 1.3016688061617459, + "grad_norm": 0.3616265058517456, + "learning_rate": 3.9478915877588895e-05, + "loss": 2.1792, + "step": 1015 + }, + { + "epoch": 1.3029525032092426, + "grad_norm": 0.370483934879303, + "learning_rate": 3.945767893585608e-05, + "loss": 2.2353, + "step": 1016 + }, + { + "epoch": 1.3042362002567394, + "grad_norm": 0.35736027359962463, + "learning_rate": 3.9436426307278175e-05, + "loss": 2.1984, + "step": 1017 + }, + { + "epoch": 1.3055198973042361, + "grad_norm": 0.35935720801353455, + "learning_rate": 3.941515801491468e-05, + "loss": 2.1947, + "step": 1018 + }, + { + "epoch": 1.306803594351733, + "grad_norm": 0.37101173400878906, + "learning_rate": 3.9393874081842056e-05, + "loss": 2.2611, + "step": 1019 + }, + { + "epoch": 1.3080872913992299, + "grad_norm": 0.35435688495635986, + "learning_rate": 3.937257453115378e-05, + "loss": 2.316, + "step": 1020 + }, + { + "epoch": 1.3093709884467266, + "grad_norm": 0.3522638976573944, + "learning_rate": 3.935125938596026e-05, + "loss": 2.2335, + "step": 1021 + }, + { + "epoch": 1.3106546854942234, + "grad_norm": 0.36578962206840515, + "learning_rate": 3.932992866938878e-05, + "loss": 2.1806, + "step": 1022 + }, + { + "epoch": 1.3119383825417201, + "grad_norm": 0.3987962007522583, + "learning_rate": 3.93085824045836e-05, + "loss": 2.2901, + "step": 1023 + }, + { + "epoch": 1.3132220795892169, + "grad_norm": 0.35971546173095703, + "learning_rate": 3.928722061470577e-05, + "loss": 2.2237, + "step": 1024 + }, + { + "epoch": 1.3145057766367136, + "grad_norm": 0.3714924454689026, + "learning_rate": 3.9265843322933246e-05, + "loss": 2.1631, + "step": 1025 + }, + { + "epoch": 1.3157894736842106, + "grad_norm": 0.3943617343902588, + "learning_rate": 3.924445055246076e-05, + "loss": 2.192, + "step": 1026 + }, + { + "epoch": 1.3170731707317074, + "grad_norm": 0.3856651186943054, + "learning_rate": 3.9223042326499876e-05, + "loss": 2.2358, + "step": 1027 + }, + { + "epoch": 1.3183568677792041, + "grad_norm": 0.37776678800582886, + "learning_rate": 3.920161866827889e-05, + "loss": 2.2994, + "step": 1028 + }, + { + "epoch": 1.319640564826701, + "grad_norm": 0.3672482371330261, + "learning_rate": 3.918017960104289e-05, + "loss": 2.2111, + "step": 1029 + }, + { + "epoch": 1.3209242618741976, + "grad_norm": 0.3800339698791504, + "learning_rate": 3.9158725148053624e-05, + "loss": 2.2559, + "step": 1030 + }, + { + "epoch": 1.3222079589216944, + "grad_norm": 0.3875967264175415, + "learning_rate": 3.913725533258958e-05, + "loss": 2.2545, + "step": 1031 + }, + { + "epoch": 1.3234916559691912, + "grad_norm": 0.35976120829582214, + "learning_rate": 3.91157701779459e-05, + "loss": 2.2463, + "step": 1032 + }, + { + "epoch": 1.3247753530166881, + "grad_norm": 0.36237773299217224, + "learning_rate": 3.909426970743435e-05, + "loss": 2.1812, + "step": 1033 + }, + { + "epoch": 1.326059050064185, + "grad_norm": 0.4001487195491791, + "learning_rate": 3.907275394438335e-05, + "loss": 2.2625, + "step": 1034 + }, + { + "epoch": 1.3273427471116817, + "grad_norm": 0.3896241784095764, + "learning_rate": 3.905122291213791e-05, + "loss": 2.2755, + "step": 1035 + }, + { + "epoch": 1.3286264441591784, + "grad_norm": 0.35467684268951416, + "learning_rate": 3.902967663405956e-05, + "loss": 2.2866, + "step": 1036 + }, + { + "epoch": 1.3299101412066752, + "grad_norm": 0.3871800899505615, + "learning_rate": 3.900811513352642e-05, + "loss": 2.2688, + "step": 1037 + }, + { + "epoch": 1.3311938382541721, + "grad_norm": 0.37341922521591187, + "learning_rate": 3.8986538433933116e-05, + "loss": 2.3456, + "step": 1038 + }, + { + "epoch": 1.3324775353016687, + "grad_norm": 0.3812583088874817, + "learning_rate": 3.896494655869074e-05, + "loss": 2.2648, + "step": 1039 + }, + { + "epoch": 1.3337612323491657, + "grad_norm": 0.3752239942550659, + "learning_rate": 3.8943339531226895e-05, + "loss": 2.2425, + "step": 1040 + }, + { + "epoch": 1.3350449293966624, + "grad_norm": 0.40153515338897705, + "learning_rate": 3.892171737498558e-05, + "loss": 2.2276, + "step": 1041 + }, + { + "epoch": 1.3363286264441592, + "grad_norm": 0.3683505952358246, + "learning_rate": 3.890008011342725e-05, + "loss": 2.2163, + "step": 1042 + }, + { + "epoch": 1.337612323491656, + "grad_norm": 0.38523101806640625, + "learning_rate": 3.8878427770028705e-05, + "loss": 2.2731, + "step": 1043 + }, + { + "epoch": 1.3388960205391527, + "grad_norm": 0.4340222179889679, + "learning_rate": 3.8856760368283143e-05, + "loss": 2.1979, + "step": 1044 + }, + { + "epoch": 1.3401797175866497, + "grad_norm": 0.3821227550506592, + "learning_rate": 3.8835077931700085e-05, + "loss": 2.193, + "step": 1045 + }, + { + "epoch": 1.3414634146341464, + "grad_norm": 0.3737218677997589, + "learning_rate": 3.8813380483805374e-05, + "loss": 2.305, + "step": 1046 + }, + { + "epoch": 1.3427471116816432, + "grad_norm": 0.3844810724258423, + "learning_rate": 3.879166804814114e-05, + "loss": 2.1998, + "step": 1047 + }, + { + "epoch": 1.34403080872914, + "grad_norm": 0.43873289227485657, + "learning_rate": 3.876994064826576e-05, + "loss": 2.2845, + "step": 1048 + }, + { + "epoch": 1.3453145057766367, + "grad_norm": 0.3661579489707947, + "learning_rate": 3.8748198307753874e-05, + "loss": 2.2256, + "step": 1049 + }, + { + "epoch": 1.3465982028241335, + "grad_norm": 0.3813747763633728, + "learning_rate": 3.8726441050196305e-05, + "loss": 2.1725, + "step": 1050 + }, + { + "epoch": 1.3478818998716302, + "grad_norm": 0.37020549178123474, + "learning_rate": 3.870466889920008e-05, + "loss": 2.2679, + "step": 1051 + }, + { + "epoch": 1.3491655969191272, + "grad_norm": 0.39001724123954773, + "learning_rate": 3.8682881878388376e-05, + "loss": 2.3152, + "step": 1052 + }, + { + "epoch": 1.350449293966624, + "grad_norm": 0.3887459635734558, + "learning_rate": 3.866108001140051e-05, + "loss": 2.2882, + "step": 1053 + }, + { + "epoch": 1.3517329910141207, + "grad_norm": 0.3768569827079773, + "learning_rate": 3.863926332189191e-05, + "loss": 2.2279, + "step": 1054 + }, + { + "epoch": 1.3530166880616175, + "grad_norm": 0.35536009073257446, + "learning_rate": 3.8617431833534066e-05, + "loss": 2.1596, + "step": 1055 + }, + { + "epoch": 1.3543003851091142, + "grad_norm": 0.38662397861480713, + "learning_rate": 3.859558557001456e-05, + "loss": 2.294, + "step": 1056 + }, + { + "epoch": 1.355584082156611, + "grad_norm": 0.37451159954071045, + "learning_rate": 3.857372455503697e-05, + "loss": 2.1217, + "step": 1057 + }, + { + "epoch": 1.3568677792041077, + "grad_norm": 0.3878231942653656, + "learning_rate": 3.8551848812320904e-05, + "loss": 2.235, + "step": 1058 + }, + { + "epoch": 1.3581514762516047, + "grad_norm": 0.4239773154258728, + "learning_rate": 3.8529958365601935e-05, + "loss": 2.2152, + "step": 1059 + }, + { + "epoch": 1.3594351732991015, + "grad_norm": 0.36272433400154114, + "learning_rate": 3.8508053238631614e-05, + "loss": 2.1909, + "step": 1060 + }, + { + "epoch": 1.3607188703465982, + "grad_norm": 0.3761455714702606, + "learning_rate": 3.8486133455177374e-05, + "loss": 2.2534, + "step": 1061 + }, + { + "epoch": 1.362002567394095, + "grad_norm": 0.3687390089035034, + "learning_rate": 3.8464199039022605e-05, + "loss": 2.3551, + "step": 1062 + }, + { + "epoch": 1.3632862644415917, + "grad_norm": 0.3875104784965515, + "learning_rate": 3.844225001396654e-05, + "loss": 2.1771, + "step": 1063 + }, + { + "epoch": 1.3645699614890887, + "grad_norm": 0.3739263713359833, + "learning_rate": 3.842028640382427e-05, + "loss": 2.2324, + "step": 1064 + }, + { + "epoch": 1.3658536585365852, + "grad_norm": 0.40333956480026245, + "learning_rate": 3.839830823242672e-05, + "loss": 2.2459, + "step": 1065 + }, + { + "epoch": 1.3671373555840822, + "grad_norm": 0.3819042146205902, + "learning_rate": 3.8376315523620584e-05, + "loss": 2.1794, + "step": 1066 + }, + { + "epoch": 1.368421052631579, + "grad_norm": 0.3782418668270111, + "learning_rate": 3.835430830126837e-05, + "loss": 2.2324, + "step": 1067 + }, + { + "epoch": 1.3697047496790757, + "grad_norm": 0.3628568947315216, + "learning_rate": 3.833228658924831e-05, + "loss": 2.2479, + "step": 1068 + }, + { + "epoch": 1.3709884467265725, + "grad_norm": 0.3767690658569336, + "learning_rate": 3.8310250411454373e-05, + "loss": 2.2547, + "step": 1069 + }, + { + "epoch": 1.3722721437740693, + "grad_norm": 0.4039711058139801, + "learning_rate": 3.8288199791796195e-05, + "loss": 2.2212, + "step": 1070 + }, + { + "epoch": 1.3735558408215662, + "grad_norm": 0.3665675222873688, + "learning_rate": 3.8266134754199114e-05, + "loss": 2.2553, + "step": 1071 + }, + { + "epoch": 1.3748395378690628, + "grad_norm": 0.3713792860507965, + "learning_rate": 3.82440553226041e-05, + "loss": 2.3088, + "step": 1072 + }, + { + "epoch": 1.3761232349165597, + "grad_norm": 0.3711341619491577, + "learning_rate": 3.822196152096772e-05, + "loss": 2.2099, + "step": 1073 + }, + { + "epoch": 1.3774069319640565, + "grad_norm": 0.3817347586154938, + "learning_rate": 3.8199853373262176e-05, + "loss": 2.2119, + "step": 1074 + }, + { + "epoch": 1.3786906290115533, + "grad_norm": 0.3850735127925873, + "learning_rate": 3.817773090347519e-05, + "loss": 2.2058, + "step": 1075 + }, + { + "epoch": 1.37997432605905, + "grad_norm": 0.3686407804489136, + "learning_rate": 3.8155594135610064e-05, + "loss": 2.2008, + "step": 1076 + }, + { + "epoch": 1.3812580231065468, + "grad_norm": 0.3951480984687805, + "learning_rate": 3.8133443093685585e-05, + "loss": 2.2465, + "step": 1077 + }, + { + "epoch": 1.3825417201540438, + "grad_norm": 0.4176129102706909, + "learning_rate": 3.811127780173602e-05, + "loss": 2.3168, + "step": 1078 + }, + { + "epoch": 1.3838254172015405, + "grad_norm": 0.3943917453289032, + "learning_rate": 3.808909828381115e-05, + "loss": 2.1437, + "step": 1079 + }, + { + "epoch": 1.3851091142490373, + "grad_norm": 0.37614941596984863, + "learning_rate": 3.806690456397612e-05, + "loss": 2.217, + "step": 1080 + }, + { + "epoch": 1.386392811296534, + "grad_norm": 0.38424134254455566, + "learning_rate": 3.804469666631155e-05, + "loss": 2.2226, + "step": 1081 + }, + { + "epoch": 1.3876765083440308, + "grad_norm": 0.36679860949516296, + "learning_rate": 3.802247461491341e-05, + "loss": 2.2228, + "step": 1082 + }, + { + "epoch": 1.3889602053915275, + "grad_norm": 0.3983418047428131, + "learning_rate": 3.8000238433893e-05, + "loss": 2.1754, + "step": 1083 + }, + { + "epoch": 1.3902439024390243, + "grad_norm": 0.40835443139076233, + "learning_rate": 3.7977988147377006e-05, + "loss": 2.177, + "step": 1084 + }, + { + "epoch": 1.3915275994865213, + "grad_norm": 0.3770159184932709, + "learning_rate": 3.79557237795074e-05, + "loss": 2.2697, + "step": 1085 + }, + { + "epoch": 1.392811296534018, + "grad_norm": 0.3879336416721344, + "learning_rate": 3.793344535444142e-05, + "loss": 2.3518, + "step": 1086 + }, + { + "epoch": 1.3940949935815148, + "grad_norm": 0.3921049237251282, + "learning_rate": 3.791115289635156e-05, + "loss": 2.0823, + "step": 1087 + }, + { + "epoch": 1.3953786906290115, + "grad_norm": 0.3612160384654999, + "learning_rate": 3.7888846429425546e-05, + "loss": 2.2694, + "step": 1088 + }, + { + "epoch": 1.3966623876765083, + "grad_norm": 0.36151912808418274, + "learning_rate": 3.7866525977866296e-05, + "loss": 2.2476, + "step": 1089 + }, + { + "epoch": 1.397946084724005, + "grad_norm": 0.36083683371543884, + "learning_rate": 3.784419156589192e-05, + "loss": 2.2319, + "step": 1090 + }, + { + "epoch": 1.3992297817715018, + "grad_norm": 0.3636226952075958, + "learning_rate": 3.782184321773564e-05, + "loss": 2.1486, + "step": 1091 + }, + { + "epoch": 1.4005134788189988, + "grad_norm": 0.36022767424583435, + "learning_rate": 3.779948095764584e-05, + "loss": 2.2666, + "step": 1092 + }, + { + "epoch": 1.4017971758664955, + "grad_norm": 0.36462947726249695, + "learning_rate": 3.777710480988598e-05, + "loss": 2.165, + "step": 1093 + }, + { + "epoch": 1.4030808729139923, + "grad_norm": 0.3657393157482147, + "learning_rate": 3.775471479873457e-05, + "loss": 2.2537, + "step": 1094 + }, + { + "epoch": 1.404364569961489, + "grad_norm": 0.36987847089767456, + "learning_rate": 3.773231094848519e-05, + "loss": 2.1994, + "step": 1095 + }, + { + "epoch": 1.4056482670089858, + "grad_norm": 0.3848939836025238, + "learning_rate": 3.770989328344645e-05, + "loss": 2.308, + "step": 1096 + }, + { + "epoch": 1.4069319640564828, + "grad_norm": 0.38896140456199646, + "learning_rate": 3.768746182794187e-05, + "loss": 2.1958, + "step": 1097 + }, + { + "epoch": 1.4082156611039793, + "grad_norm": 0.3703378736972809, + "learning_rate": 3.766501660631004e-05, + "loss": 2.209, + "step": 1098 + }, + { + "epoch": 1.4094993581514763, + "grad_norm": 0.373000830411911, + "learning_rate": 3.764255764290442e-05, + "loss": 2.2251, + "step": 1099 + }, + { + "epoch": 1.410783055198973, + "grad_norm": 0.3429698944091797, + "learning_rate": 3.762008496209338e-05, + "loss": 2.2303, + "step": 1100 + }, + { + "epoch": 1.4120667522464698, + "grad_norm": 0.39933037757873535, + "learning_rate": 3.7597598588260196e-05, + "loss": 2.1388, + "step": 1101 + }, + { + "epoch": 1.4133504492939666, + "grad_norm": 0.39360055327415466, + "learning_rate": 3.757509854580299e-05, + "loss": 2.2116, + "step": 1102 + }, + { + "epoch": 1.4146341463414633, + "grad_norm": 0.39372727274894714, + "learning_rate": 3.755258485913474e-05, + "loss": 2.2814, + "step": 1103 + }, + { + "epoch": 1.4159178433889603, + "grad_norm": 0.3706322908401489, + "learning_rate": 3.7530057552683175e-05, + "loss": 2.1897, + "step": 1104 + }, + { + "epoch": 1.417201540436457, + "grad_norm": 0.39840468764305115, + "learning_rate": 3.750751665089085e-05, + "loss": 2.2565, + "step": 1105 + }, + { + "epoch": 1.4184852374839538, + "grad_norm": 0.377890408039093, + "learning_rate": 3.7484962178215055e-05, + "loss": 2.2281, + "step": 1106 + }, + { + "epoch": 1.4197689345314506, + "grad_norm": 0.3951006531715393, + "learning_rate": 3.74623941591278e-05, + "loss": 2.1748, + "step": 1107 + }, + { + "epoch": 1.4210526315789473, + "grad_norm": 0.4109231233596802, + "learning_rate": 3.7439812618115795e-05, + "loss": 2.165, + "step": 1108 + }, + { + "epoch": 1.422336328626444, + "grad_norm": 0.41010499000549316, + "learning_rate": 3.7417217579680426e-05, + "loss": 2.1168, + "step": 1109 + }, + { + "epoch": 1.4236200256739409, + "grad_norm": 0.38000255823135376, + "learning_rate": 3.739460906833772e-05, + "loss": 2.2188, + "step": 1110 + }, + { + "epoch": 1.4249037227214378, + "grad_norm": 0.3960665464401245, + "learning_rate": 3.737198710861832e-05, + "loss": 2.2694, + "step": 1111 + }, + { + "epoch": 1.4261874197689346, + "grad_norm": 0.3778914213180542, + "learning_rate": 3.734935172506747e-05, + "loss": 2.3073, + "step": 1112 + }, + { + "epoch": 1.4274711168164314, + "grad_norm": 0.3883233368396759, + "learning_rate": 3.732670294224498e-05, + "loss": 2.1895, + "step": 1113 + }, + { + "epoch": 1.428754813863928, + "grad_norm": 0.38521334528923035, + "learning_rate": 3.730404078472518e-05, + "loss": 2.2923, + "step": 1114 + }, + { + "epoch": 1.4300385109114249, + "grad_norm": 0.39216870069503784, + "learning_rate": 3.728136527709694e-05, + "loss": 2.2761, + "step": 1115 + }, + { + "epoch": 1.4313222079589216, + "grad_norm": 0.3601703643798828, + "learning_rate": 3.725867644396358e-05, + "loss": 2.2398, + "step": 1116 + }, + { + "epoch": 1.4326059050064184, + "grad_norm": 0.400269091129303, + "learning_rate": 3.72359743099429e-05, + "loss": 2.2223, + "step": 1117 + }, + { + "epoch": 1.4338896020539154, + "grad_norm": 0.3815939724445343, + "learning_rate": 3.7213258899667144e-05, + "loss": 2.291, + "step": 1118 + }, + { + "epoch": 1.4351732991014121, + "grad_norm": 0.37799543142318726, + "learning_rate": 3.719053023778292e-05, + "loss": 2.2347, + "step": 1119 + }, + { + "epoch": 1.4364569961489089, + "grad_norm": 0.37301313877105713, + "learning_rate": 3.716778834895127e-05, + "loss": 2.2475, + "step": 1120 + }, + { + "epoch": 1.4377406931964056, + "grad_norm": 0.3976534307003021, + "learning_rate": 3.714503325784754e-05, + "loss": 2.2651, + "step": 1121 + }, + { + "epoch": 1.4390243902439024, + "grad_norm": 0.39929476380348206, + "learning_rate": 3.712226498916143e-05, + "loss": 2.2446, + "step": 1122 + }, + { + "epoch": 1.4403080872913994, + "grad_norm": 0.37493380904197693, + "learning_rate": 3.709948356759691e-05, + "loss": 2.1977, + "step": 1123 + }, + { + "epoch": 1.441591784338896, + "grad_norm": 0.3848809003829956, + "learning_rate": 3.7076689017872246e-05, + "loss": 2.3382, + "step": 1124 + }, + { + "epoch": 1.4428754813863929, + "grad_norm": 0.39769625663757324, + "learning_rate": 3.705388136471995e-05, + "loss": 2.3048, + "step": 1125 + }, + { + "epoch": 1.4441591784338896, + "grad_norm": 0.36819279193878174, + "learning_rate": 3.7031060632886724e-05, + "loss": 2.2417, + "step": 1126 + }, + { + "epoch": 1.4454428754813864, + "grad_norm": 0.3896733820438385, + "learning_rate": 3.700822684713349e-05, + "loss": 2.2029, + "step": 1127 + }, + { + "epoch": 1.4467265725288831, + "grad_norm": 0.36140698194503784, + "learning_rate": 3.6985380032235315e-05, + "loss": 2.149, + "step": 1128 + }, + { + "epoch": 1.44801026957638, + "grad_norm": 0.3695017993450165, + "learning_rate": 3.6962520212981405e-05, + "loss": 2.2265, + "step": 1129 + }, + { + "epoch": 1.4492939666238769, + "grad_norm": 0.3904394805431366, + "learning_rate": 3.69396474141751e-05, + "loss": 2.2479, + "step": 1130 + }, + { + "epoch": 1.4505776636713734, + "grad_norm": 0.389116495847702, + "learning_rate": 3.691676166063378e-05, + "loss": 2.2346, + "step": 1131 + }, + { + "epoch": 1.4518613607188704, + "grad_norm": 0.3761390149593353, + "learning_rate": 3.689386297718892e-05, + "loss": 2.2856, + "step": 1132 + }, + { + "epoch": 1.4531450577663672, + "grad_norm": 0.4113580882549286, + "learning_rate": 3.687095138868601e-05, + "loss": 2.1943, + "step": 1133 + }, + { + "epoch": 1.454428754813864, + "grad_norm": 0.358335018157959, + "learning_rate": 3.684802691998452e-05, + "loss": 2.1895, + "step": 1134 + }, + { + "epoch": 1.4557124518613607, + "grad_norm": 0.47489404678344727, + "learning_rate": 3.682508959595795e-05, + "loss": 2.2097, + "step": 1135 + }, + { + "epoch": 1.4569961489088574, + "grad_norm": 0.40556371212005615, + "learning_rate": 3.680213944149368e-05, + "loss": 2.2414, + "step": 1136 + }, + { + "epoch": 1.4582798459563544, + "grad_norm": 0.356882780790329, + "learning_rate": 3.677917648149307e-05, + "loss": 2.262, + "step": 1137 + }, + { + "epoch": 1.4595635430038512, + "grad_norm": 0.39578723907470703, + "learning_rate": 3.675620074087134e-05, + "loss": 2.1591, + "step": 1138 + }, + { + "epoch": 1.460847240051348, + "grad_norm": 0.3718084394931793, + "learning_rate": 3.673321224455759e-05, + "loss": 2.3229, + "step": 1139 + }, + { + "epoch": 1.4621309370988447, + "grad_norm": 0.3779333829879761, + "learning_rate": 3.671021101749476e-05, + "loss": 2.2221, + "step": 1140 + }, + { + "epoch": 1.4634146341463414, + "grad_norm": 0.3797706663608551, + "learning_rate": 3.668719708463959e-05, + "loss": 2.1386, + "step": 1141 + }, + { + "epoch": 1.4646983311938382, + "grad_norm": 0.3736761212348938, + "learning_rate": 3.666417047096262e-05, + "loss": 2.1575, + "step": 1142 + }, + { + "epoch": 1.465982028241335, + "grad_norm": 0.6424862742424011, + "learning_rate": 3.664113120144816e-05, + "loss": 2.2285, + "step": 1143 + }, + { + "epoch": 1.467265725288832, + "grad_norm": 0.36189529299736023, + "learning_rate": 3.6618079301094216e-05, + "loss": 2.1598, + "step": 1144 + }, + { + "epoch": 1.4685494223363287, + "grad_norm": 0.3950345516204834, + "learning_rate": 3.659501479491253e-05, + "loss": 2.1779, + "step": 1145 + }, + { + "epoch": 1.4698331193838254, + "grad_norm": 0.37057608366012573, + "learning_rate": 3.6571937707928524e-05, + "loss": 2.1717, + "step": 1146 + }, + { + "epoch": 1.4711168164313222, + "grad_norm": 0.3707398474216461, + "learning_rate": 3.654884806518123e-05, + "loss": 2.1828, + "step": 1147 + }, + { + "epoch": 1.472400513478819, + "grad_norm": 0.3933170735836029, + "learning_rate": 3.652574589172335e-05, + "loss": 2.2266, + "step": 1148 + }, + { + "epoch": 1.4736842105263157, + "grad_norm": 0.39545738697052, + "learning_rate": 3.650263121262117e-05, + "loss": 2.3313, + "step": 1149 + }, + { + "epoch": 1.4749679075738125, + "grad_norm": 0.3947657346725464, + "learning_rate": 3.6479504052954514e-05, + "loss": 2.2543, + "step": 1150 + }, + { + "epoch": 1.4762516046213094, + "grad_norm": 0.38942018151283264, + "learning_rate": 3.645636443781679e-05, + "loss": 2.1467, + "step": 1151 + }, + { + "epoch": 1.4775353016688062, + "grad_norm": 0.3504950702190399, + "learning_rate": 3.6433212392314905e-05, + "loss": 2.2263, + "step": 1152 + }, + { + "epoch": 1.478818998716303, + "grad_norm": 0.37644287943840027, + "learning_rate": 3.6410047941569224e-05, + "loss": 2.3145, + "step": 1153 + }, + { + "epoch": 1.4801026957637997, + "grad_norm": 0.41499248147010803, + "learning_rate": 3.638687111071363e-05, + "loss": 2.182, + "step": 1154 + }, + { + "epoch": 1.4813863928112965, + "grad_norm": 0.3680044114589691, + "learning_rate": 3.6363681924895394e-05, + "loss": 2.23, + "step": 1155 + }, + { + "epoch": 1.4826700898587934, + "grad_norm": 0.3712504804134369, + "learning_rate": 3.63404804092752e-05, + "loss": 2.3055, + "step": 1156 + }, + { + "epoch": 1.48395378690629, + "grad_norm": 0.38080593943595886, + "learning_rate": 3.6317266589027124e-05, + "loss": 2.3163, + "step": 1157 + }, + { + "epoch": 1.485237483953787, + "grad_norm": 0.3790014088153839, + "learning_rate": 3.6294040489338575e-05, + "loss": 2.2555, + "step": 1158 + }, + { + "epoch": 1.4865211810012837, + "grad_norm": 0.3629383444786072, + "learning_rate": 3.62708021354103e-05, + "loss": 2.1125, + "step": 1159 + }, + { + "epoch": 1.4878048780487805, + "grad_norm": 0.36138129234313965, + "learning_rate": 3.6247551552456336e-05, + "loss": 2.2633, + "step": 1160 + }, + { + "epoch": 1.4890885750962772, + "grad_norm": 0.35929739475250244, + "learning_rate": 3.622428876570399e-05, + "loss": 2.2202, + "step": 1161 + }, + { + "epoch": 1.490372272143774, + "grad_norm": 0.3624289333820343, + "learning_rate": 3.620101380039381e-05, + "loss": 2.2102, + "step": 1162 + }, + { + "epoch": 1.491655969191271, + "grad_norm": 0.379535049200058, + "learning_rate": 3.617772668177956e-05, + "loss": 2.2099, + "step": 1163 + }, + { + "epoch": 1.4929396662387677, + "grad_norm": 0.3564260005950928, + "learning_rate": 3.615442743512817e-05, + "loss": 2.2622, + "step": 1164 + }, + { + "epoch": 1.4942233632862645, + "grad_norm": 0.3908945918083191, + "learning_rate": 3.613111608571977e-05, + "loss": 2.2087, + "step": 1165 + }, + { + "epoch": 1.4955070603337612, + "grad_norm": 0.37525975704193115, + "learning_rate": 3.6107792658847595e-05, + "loss": 2.3016, + "step": 1166 + }, + { + "epoch": 1.496790757381258, + "grad_norm": 0.35312336683273315, + "learning_rate": 3.608445717981798e-05, + "loss": 2.2335, + "step": 1167 + }, + { + "epoch": 1.4980744544287548, + "grad_norm": 0.3973201513290405, + "learning_rate": 3.6061109673950334e-05, + "loss": 2.3283, + "step": 1168 + }, + { + "epoch": 1.4993581514762515, + "grad_norm": 0.44854387640953064, + "learning_rate": 3.603775016657715e-05, + "loss": 2.2669, + "step": 1169 + }, + { + "epoch": 1.5006418485237485, + "grad_norm": 0.40130218863487244, + "learning_rate": 3.601437868304391e-05, + "loss": 2.2615, + "step": 1170 + }, + { + "epoch": 1.501925545571245, + "grad_norm": 0.36666733026504517, + "learning_rate": 3.59909952487091e-05, + "loss": 2.1767, + "step": 1171 + }, + { + "epoch": 1.503209242618742, + "grad_norm": 0.369798481464386, + "learning_rate": 3.596759988894417e-05, + "loss": 2.2134, + "step": 1172 + }, + { + "epoch": 1.5044929396662388, + "grad_norm": 0.37616270780563354, + "learning_rate": 3.594419262913351e-05, + "loss": 2.2595, + "step": 1173 + }, + { + "epoch": 1.5057766367137355, + "grad_norm": 0.37049224972724915, + "learning_rate": 3.5920773494674434e-05, + "loss": 2.1782, + "step": 1174 + }, + { + "epoch": 1.5070603337612325, + "grad_norm": 0.38616225123405457, + "learning_rate": 3.589734251097712e-05, + "loss": 2.1387, + "step": 1175 + }, + { + "epoch": 1.508344030808729, + "grad_norm": 0.4116593599319458, + "learning_rate": 3.587389970346461e-05, + "loss": 2.2533, + "step": 1176 + }, + { + "epoch": 1.509627727856226, + "grad_norm": 0.3951702117919922, + "learning_rate": 3.585044509757278e-05, + "loss": 2.2805, + "step": 1177 + }, + { + "epoch": 1.5109114249037228, + "grad_norm": 0.3869103789329529, + "learning_rate": 3.582697871875031e-05, + "loss": 2.24, + "step": 1178 + }, + { + "epoch": 1.5121951219512195, + "grad_norm": 0.4094953238964081, + "learning_rate": 3.580350059245864e-05, + "loss": 2.2871, + "step": 1179 + }, + { + "epoch": 1.5134788189987163, + "grad_norm": 0.3869830369949341, + "learning_rate": 3.578001074417198e-05, + "loss": 2.1615, + "step": 1180 + }, + { + "epoch": 1.514762516046213, + "grad_norm": 0.3672597408294678, + "learning_rate": 3.5756509199377224e-05, + "loss": 2.2374, + "step": 1181 + }, + { + "epoch": 1.51604621309371, + "grad_norm": 0.36492928862571716, + "learning_rate": 3.573299598357398e-05, + "loss": 2.3031, + "step": 1182 + }, + { + "epoch": 1.5173299101412066, + "grad_norm": 0.3777792751789093, + "learning_rate": 3.5709471122274534e-05, + "loss": 2.2957, + "step": 1183 + }, + { + "epoch": 1.5186136071887035, + "grad_norm": 0.38662636280059814, + "learning_rate": 3.568593464100377e-05, + "loss": 2.1457, + "step": 1184 + }, + { + "epoch": 1.5198973042362003, + "grad_norm": 0.41264063119888306, + "learning_rate": 3.566238656529921e-05, + "loss": 2.2829, + "step": 1185 + }, + { + "epoch": 1.521181001283697, + "grad_norm": 0.3814603388309479, + "learning_rate": 3.5638826920710935e-05, + "loss": 2.2356, + "step": 1186 + }, + { + "epoch": 1.5224646983311938, + "grad_norm": 0.3695092797279358, + "learning_rate": 3.56152557328016e-05, + "loss": 2.1976, + "step": 1187 + }, + { + "epoch": 1.5237483953786906, + "grad_norm": 0.39186903834342957, + "learning_rate": 3.559167302714636e-05, + "loss": 2.2426, + "step": 1188 + }, + { + "epoch": 1.5250320924261875, + "grad_norm": 0.40045449137687683, + "learning_rate": 3.556807882933289e-05, + "loss": 2.2451, + "step": 1189 + }, + { + "epoch": 1.526315789473684, + "grad_norm": 0.38327574729919434, + "learning_rate": 3.5544473164961324e-05, + "loss": 2.2787, + "step": 1190 + }, + { + "epoch": 1.527599486521181, + "grad_norm": 0.36272966861724854, + "learning_rate": 3.552085605964424e-05, + "loss": 2.2206, + "step": 1191 + }, + { + "epoch": 1.5288831835686778, + "grad_norm": 0.39420098066329956, + "learning_rate": 3.5497227539006614e-05, + "loss": 2.2664, + "step": 1192 + }, + { + "epoch": 1.5301668806161746, + "grad_norm": 0.38125845789909363, + "learning_rate": 3.547358762868584e-05, + "loss": 2.2152, + "step": 1193 + }, + { + "epoch": 1.5314505776636713, + "grad_norm": 0.3611045479774475, + "learning_rate": 3.544993635433164e-05, + "loss": 2.1673, + "step": 1194 + }, + { + "epoch": 1.532734274711168, + "grad_norm": 0.38052603602409363, + "learning_rate": 3.5426273741606086e-05, + "loss": 2.276, + "step": 1195 + }, + { + "epoch": 1.534017971758665, + "grad_norm": 0.37632811069488525, + "learning_rate": 3.540259981618353e-05, + "loss": 2.2557, + "step": 1196 + }, + { + "epoch": 1.5353016688061616, + "grad_norm": 0.3829447627067566, + "learning_rate": 3.5378914603750624e-05, + "loss": 2.2649, + "step": 1197 + }, + { + "epoch": 1.5365853658536586, + "grad_norm": 0.3774879574775696, + "learning_rate": 3.5355218130006255e-05, + "loss": 2.2778, + "step": 1198 + }, + { + "epoch": 1.5378690629011553, + "grad_norm": 0.39277470111846924, + "learning_rate": 3.5331510420661525e-05, + "loss": 2.31, + "step": 1199 + }, + { + "epoch": 1.539152759948652, + "grad_norm": 0.3856041133403778, + "learning_rate": 3.530779150143973e-05, + "loss": 2.1995, + "step": 1200 + }, + { + "epoch": 1.540436456996149, + "grad_norm": 0.40345871448516846, + "learning_rate": 3.528406139807633e-05, + "loss": 2.2882, + "step": 1201 + }, + { + "epoch": 1.5417201540436456, + "grad_norm": 0.37651973962783813, + "learning_rate": 3.526032013631893e-05, + "loss": 2.3489, + "step": 1202 + }, + { + "epoch": 1.5430038510911426, + "grad_norm": 0.359997421503067, + "learning_rate": 3.523656774192721e-05, + "loss": 2.2709, + "step": 1203 + }, + { + "epoch": 1.5442875481386393, + "grad_norm": 0.4158475697040558, + "learning_rate": 3.521280424067296e-05, + "loss": 2.2083, + "step": 1204 + }, + { + "epoch": 1.545571245186136, + "grad_norm": 0.37753528356552124, + "learning_rate": 3.5189029658340025e-05, + "loss": 2.2691, + "step": 1205 + }, + { + "epoch": 1.5468549422336328, + "grad_norm": 0.3862055540084839, + "learning_rate": 3.516524402072425e-05, + "loss": 2.3055, + "step": 1206 + }, + { + "epoch": 1.5481386392811296, + "grad_norm": 0.40717101097106934, + "learning_rate": 3.5141447353633474e-05, + "loss": 2.253, + "step": 1207 + }, + { + "epoch": 1.5494223363286266, + "grad_norm": 0.4078426957130432, + "learning_rate": 3.5117639682887534e-05, + "loss": 2.1887, + "step": 1208 + }, + { + "epoch": 1.5507060333761231, + "grad_norm": 0.34787634015083313, + "learning_rate": 3.509382103431819e-05, + "loss": 2.2336, + "step": 1209 + }, + { + "epoch": 1.55198973042362, + "grad_norm": 0.38061514496803284, + "learning_rate": 3.506999143376908e-05, + "loss": 2.2623, + "step": 1210 + }, + { + "epoch": 1.5532734274711169, + "grad_norm": 0.39258530735969543, + "learning_rate": 3.5046150907095776e-05, + "loss": 2.1535, + "step": 1211 + }, + { + "epoch": 1.5545571245186136, + "grad_norm": 0.3953094184398651, + "learning_rate": 3.502229948016568e-05, + "loss": 2.1615, + "step": 1212 + }, + { + "epoch": 1.5558408215661104, + "grad_norm": 0.3790014684200287, + "learning_rate": 3.4998437178858035e-05, + "loss": 2.2498, + "step": 1213 + }, + { + "epoch": 1.5571245186136071, + "grad_norm": 0.40729162096977234, + "learning_rate": 3.497456402906385e-05, + "loss": 2.2521, + "step": 1214 + }, + { + "epoch": 1.558408215661104, + "grad_norm": 0.42329171299934387, + "learning_rate": 3.495068005668595e-05, + "loss": 2.3194, + "step": 1215 + }, + { + "epoch": 1.5596919127086006, + "grad_norm": 0.3534400463104248, + "learning_rate": 3.492678528763887e-05, + "loss": 2.2785, + "step": 1216 + }, + { + "epoch": 1.5609756097560976, + "grad_norm": 0.4265909492969513, + "learning_rate": 3.490287974784887e-05, + "loss": 2.1803, + "step": 1217 + }, + { + "epoch": 1.5622593068035944, + "grad_norm": 0.37546563148498535, + "learning_rate": 3.487896346325389e-05, + "loss": 2.2921, + "step": 1218 + }, + { + "epoch": 1.5635430038510911, + "grad_norm": 0.38932451605796814, + "learning_rate": 3.4855036459803544e-05, + "loss": 2.2686, + "step": 1219 + }, + { + "epoch": 1.5648267008985879, + "grad_norm": 0.3934752643108368, + "learning_rate": 3.4831098763459066e-05, + "loss": 2.3104, + "step": 1220 + }, + { + "epoch": 1.5661103979460846, + "grad_norm": 0.37588098645210266, + "learning_rate": 3.480715040019329e-05, + "loss": 2.2768, + "step": 1221 + }, + { + "epoch": 1.5673940949935816, + "grad_norm": 0.3763798773288727, + "learning_rate": 3.478319139599063e-05, + "loss": 2.2028, + "step": 1222 + }, + { + "epoch": 1.5686777920410782, + "grad_norm": 0.39880573749542236, + "learning_rate": 3.475922177684704e-05, + "loss": 2.2299, + "step": 1223 + }, + { + "epoch": 1.5699614890885751, + "grad_norm": 0.3821561336517334, + "learning_rate": 3.473524156876999e-05, + "loss": 2.1963, + "step": 1224 + }, + { + "epoch": 1.571245186136072, + "grad_norm": 0.39914795756340027, + "learning_rate": 3.4711250797778446e-05, + "loss": 2.2521, + "step": 1225 + }, + { + "epoch": 1.5725288831835686, + "grad_norm": 0.43123576045036316, + "learning_rate": 3.468724948990283e-05, + "loss": 2.1249, + "step": 1226 + }, + { + "epoch": 1.5738125802310656, + "grad_norm": 0.38369259238243103, + "learning_rate": 3.466323767118501e-05, + "loss": 2.2478, + "step": 1227 + }, + { + "epoch": 1.5750962772785622, + "grad_norm": 0.3837836682796478, + "learning_rate": 3.4639215367678225e-05, + "loss": 2.2308, + "step": 1228 + }, + { + "epoch": 1.5763799743260591, + "grad_norm": 0.3655678331851959, + "learning_rate": 3.461518260544713e-05, + "loss": 2.3227, + "step": 1229 + }, + { + "epoch": 1.5776636713735557, + "grad_norm": 0.3748762309551239, + "learning_rate": 3.45911394105677e-05, + "loss": 2.1733, + "step": 1230 + }, + { + "epoch": 1.5789473684210527, + "grad_norm": 0.3868464231491089, + "learning_rate": 3.456708580912725e-05, + "loss": 2.2485, + "step": 1231 + }, + { + "epoch": 1.5802310654685494, + "grad_norm": 0.37276676297187805, + "learning_rate": 3.4543021827224365e-05, + "loss": 2.2844, + "step": 1232 + }, + { + "epoch": 1.5815147625160462, + "grad_norm": 0.38882049918174744, + "learning_rate": 3.45189474909689e-05, + "loss": 2.2337, + "step": 1233 + }, + { + "epoch": 1.5827984595635431, + "grad_norm": 0.37841805815696716, + "learning_rate": 3.4494862826481966e-05, + "loss": 2.2276, + "step": 1234 + }, + { + "epoch": 1.5840821566110397, + "grad_norm": 0.3648984730243683, + "learning_rate": 3.4470767859895846e-05, + "loss": 2.1893, + "step": 1235 + }, + { + "epoch": 1.5853658536585367, + "grad_norm": 0.40236514806747437, + "learning_rate": 3.4446662617354034e-05, + "loss": 2.2086, + "step": 1236 + }, + { + "epoch": 1.5866495507060334, + "grad_norm": 0.3805043697357178, + "learning_rate": 3.442254712501115e-05, + "loss": 2.2052, + "step": 1237 + }, + { + "epoch": 1.5879332477535302, + "grad_norm": 0.3884541690349579, + "learning_rate": 3.439842140903293e-05, + "loss": 2.3047, + "step": 1238 + }, + { + "epoch": 1.589216944801027, + "grad_norm": 0.3753684461116791, + "learning_rate": 3.437428549559624e-05, + "loss": 2.258, + "step": 1239 + }, + { + "epoch": 1.5905006418485237, + "grad_norm": 0.3995250165462494, + "learning_rate": 3.4350139410888956e-05, + "loss": 2.2084, + "step": 1240 + }, + { + "epoch": 1.5917843388960207, + "grad_norm": 0.3964383602142334, + "learning_rate": 3.432598318111005e-05, + "loss": 2.1462, + "step": 1241 + }, + { + "epoch": 1.5930680359435172, + "grad_norm": 0.3621688783168793, + "learning_rate": 3.4301816832469443e-05, + "loss": 2.2277, + "step": 1242 + }, + { + "epoch": 1.5943517329910142, + "grad_norm": 0.4131968319416046, + "learning_rate": 3.4277640391188094e-05, + "loss": 2.1845, + "step": 1243 + }, + { + "epoch": 1.595635430038511, + "grad_norm": 0.41226086020469666, + "learning_rate": 3.425345388349786e-05, + "loss": 2.1758, + "step": 1244 + }, + { + "epoch": 1.5969191270860077, + "grad_norm": 0.4197986423969269, + "learning_rate": 3.422925733564155e-05, + "loss": 2.2193, + "step": 1245 + }, + { + "epoch": 1.5982028241335045, + "grad_norm": 0.4071418046951294, + "learning_rate": 3.420505077387286e-05, + "loss": 2.3172, + "step": 1246 + }, + { + "epoch": 1.5994865211810012, + "grad_norm": 0.37383535504341125, + "learning_rate": 3.418083422445635e-05, + "loss": 2.2835, + "step": 1247 + }, + { + "epoch": 1.6007702182284982, + "grad_norm": 0.40442097187042236, + "learning_rate": 3.415660771366741e-05, + "loss": 2.2389, + "step": 1248 + }, + { + "epoch": 1.6020539152759947, + "grad_norm": 0.380212664604187, + "learning_rate": 3.4132371267792264e-05, + "loss": 2.1973, + "step": 1249 + }, + { + "epoch": 1.6033376123234917, + "grad_norm": 0.3948259949684143, + "learning_rate": 3.4108124913127875e-05, + "loss": 2.254, + "step": 1250 + }, + { + "epoch": 1.6046213093709885, + "grad_norm": 0.3870978057384491, + "learning_rate": 3.4083868675981996e-05, + "loss": 2.0882, + "step": 1251 + }, + { + "epoch": 1.6059050064184852, + "grad_norm": 0.38109326362609863, + "learning_rate": 3.405960258267308e-05, + "loss": 2.2989, + "step": 1252 + }, + { + "epoch": 1.6071887034659822, + "grad_norm": 0.38316014409065247, + "learning_rate": 3.403532665953028e-05, + "loss": 2.2568, + "step": 1253 + }, + { + "epoch": 1.6084724005134787, + "grad_norm": 0.4252641201019287, + "learning_rate": 3.401104093289341e-05, + "loss": 2.289, + "step": 1254 + }, + { + "epoch": 1.6097560975609757, + "grad_norm": 0.3894971013069153, + "learning_rate": 3.3986745429112936e-05, + "loss": 2.2041, + "step": 1255 + }, + { + "epoch": 1.6110397946084722, + "grad_norm": 0.38642048835754395, + "learning_rate": 3.3962440174549924e-05, + "loss": 2.2442, + "step": 1256 + }, + { + "epoch": 1.6123234916559692, + "grad_norm": 0.38057392835617065, + "learning_rate": 3.3938125195576e-05, + "loss": 2.2573, + "step": 1257 + }, + { + "epoch": 1.613607188703466, + "grad_norm": 0.3963440954685211, + "learning_rate": 3.3913800518573385e-05, + "loss": 2.2125, + "step": 1258 + }, + { + "epoch": 1.6148908857509627, + "grad_norm": 0.37275323271751404, + "learning_rate": 3.3889466169934774e-05, + "loss": 2.1775, + "step": 1259 + }, + { + "epoch": 1.6161745827984597, + "grad_norm": 0.3996749818325043, + "learning_rate": 3.386512217606339e-05, + "loss": 2.2006, + "step": 1260 + }, + { + "epoch": 1.6174582798459562, + "grad_norm": 0.3794258236885071, + "learning_rate": 3.384076856337292e-05, + "loss": 2.2485, + "step": 1261 + }, + { + "epoch": 1.6187419768934532, + "grad_norm": 0.40182143449783325, + "learning_rate": 3.3816405358287446e-05, + "loss": 2.2404, + "step": 1262 + }, + { + "epoch": 1.62002567394095, + "grad_norm": 0.4085938036441803, + "learning_rate": 3.379203258724152e-05, + "loss": 2.2732, + "step": 1263 + }, + { + "epoch": 1.6213093709884467, + "grad_norm": 0.3972185552120209, + "learning_rate": 3.376765027668003e-05, + "loss": 2.2859, + "step": 1264 + }, + { + "epoch": 1.6225930680359435, + "grad_norm": 0.39743340015411377, + "learning_rate": 3.374325845305822e-05, + "loss": 2.2693, + "step": 1265 + }, + { + "epoch": 1.6238767650834403, + "grad_norm": 0.4192309081554413, + "learning_rate": 3.371885714284169e-05, + "loss": 2.3041, + "step": 1266 + }, + { + "epoch": 1.6251604621309372, + "grad_norm": 0.36719846725463867, + "learning_rate": 3.369444637250627e-05, + "loss": 2.2128, + "step": 1267 + }, + { + "epoch": 1.6264441591784338, + "grad_norm": 0.37783583998680115, + "learning_rate": 3.367002616853812e-05, + "loss": 2.1338, + "step": 1268 + }, + { + "epoch": 1.6277278562259307, + "grad_norm": 0.39298015832901, + "learning_rate": 3.364559655743359e-05, + "loss": 2.2804, + "step": 1269 + }, + { + "epoch": 1.6290115532734275, + "grad_norm": 0.39492067694664, + "learning_rate": 3.362115756569926e-05, + "loss": 2.1376, + "step": 1270 + }, + { + "epoch": 1.6302952503209243, + "grad_norm": 0.4016437828540802, + "learning_rate": 3.359670921985189e-05, + "loss": 2.26, + "step": 1271 + }, + { + "epoch": 1.631578947368421, + "grad_norm": 0.36699244379997253, + "learning_rate": 3.3572251546418355e-05, + "loss": 2.1975, + "step": 1272 + }, + { + "epoch": 1.6328626444159178, + "grad_norm": 0.36995914578437805, + "learning_rate": 3.3547784571935695e-05, + "loss": 2.2535, + "step": 1273 + }, + { + "epoch": 1.6341463414634148, + "grad_norm": 0.4290882349014282, + "learning_rate": 3.3523308322951016e-05, + "loss": 2.2441, + "step": 1274 + }, + { + "epoch": 1.6354300385109113, + "grad_norm": 0.37746745347976685, + "learning_rate": 3.3498822826021514e-05, + "loss": 2.1364, + "step": 1275 + }, + { + "epoch": 1.6367137355584083, + "grad_norm": 0.35144737362861633, + "learning_rate": 3.347432810771436e-05, + "loss": 2.3044, + "step": 1276 + }, + { + "epoch": 1.637997432605905, + "grad_norm": 0.3726184070110321, + "learning_rate": 3.344982419460681e-05, + "loss": 2.197, + "step": 1277 + }, + { + "epoch": 1.6392811296534018, + "grad_norm": 0.37522029876708984, + "learning_rate": 3.342531111328604e-05, + "loss": 2.1608, + "step": 1278 + }, + { + "epoch": 1.6405648267008985, + "grad_norm": 0.38890618085861206, + "learning_rate": 3.340078889034918e-05, + "loss": 2.1773, + "step": 1279 + }, + { + "epoch": 1.6418485237483953, + "grad_norm": 0.3751033544540405, + "learning_rate": 3.3376257552403314e-05, + "loss": 2.2046, + "step": 1280 + }, + { + "epoch": 1.6431322207958923, + "grad_norm": 0.3892718255519867, + "learning_rate": 3.335171712606535e-05, + "loss": 2.2177, + "step": 1281 + }, + { + "epoch": 1.6444159178433888, + "grad_norm": 0.37542441487312317, + "learning_rate": 3.3327167637962134e-05, + "loss": 2.2561, + "step": 1282 + }, + { + "epoch": 1.6456996148908858, + "grad_norm": 0.37376415729522705, + "learning_rate": 3.33026091147303e-05, + "loss": 2.1389, + "step": 1283 + }, + { + "epoch": 1.6469833119383825, + "grad_norm": 0.41433337330818176, + "learning_rate": 3.327804158301628e-05, + "loss": 2.1289, + "step": 1284 + }, + { + "epoch": 1.6482670089858793, + "grad_norm": 0.3767523765563965, + "learning_rate": 3.325346506947632e-05, + "loss": 2.2387, + "step": 1285 + }, + { + "epoch": 1.6495507060333763, + "grad_norm": 0.41720911860466003, + "learning_rate": 3.322887960077635e-05, + "loss": 2.1956, + "step": 1286 + }, + { + "epoch": 1.6508344030808728, + "grad_norm": 0.3834913372993469, + "learning_rate": 3.3204285203592076e-05, + "loss": 2.2792, + "step": 1287 + }, + { + "epoch": 1.6521181001283698, + "grad_norm": 0.37035104632377625, + "learning_rate": 3.3179681904608865e-05, + "loss": 2.12, + "step": 1288 + }, + { + "epoch": 1.6534017971758665, + "grad_norm": 0.38635921478271484, + "learning_rate": 3.3155069730521735e-05, + "loss": 2.1847, + "step": 1289 + }, + { + "epoch": 1.6546854942233633, + "grad_norm": 0.374063640832901, + "learning_rate": 3.313044870803536e-05, + "loss": 2.2127, + "step": 1290 + }, + { + "epoch": 1.65596919127086, + "grad_norm": 0.4345405101776123, + "learning_rate": 3.3105818863863966e-05, + "loss": 2.281, + "step": 1291 + }, + { + "epoch": 1.6572528883183568, + "grad_norm": 0.3770464360713959, + "learning_rate": 3.308118022473142e-05, + "loss": 2.2967, + "step": 1292 + }, + { + "epoch": 1.6585365853658538, + "grad_norm": 0.3714136481285095, + "learning_rate": 3.3056532817371065e-05, + "loss": 2.2356, + "step": 1293 + }, + { + "epoch": 1.6598202824133503, + "grad_norm": 0.3810267746448517, + "learning_rate": 3.3031876668525814e-05, + "loss": 2.171, + "step": 1294 + }, + { + "epoch": 1.6611039794608473, + "grad_norm": 0.36009058356285095, + "learning_rate": 3.3007211804948015e-05, + "loss": 2.1459, + "step": 1295 + }, + { + "epoch": 1.662387676508344, + "grad_norm": 0.417795866727829, + "learning_rate": 3.29825382533995e-05, + "loss": 2.2286, + "step": 1296 + }, + { + "epoch": 1.6636713735558408, + "grad_norm": 0.40939030051231384, + "learning_rate": 3.2957856040651535e-05, + "loss": 2.2301, + "step": 1297 + }, + { + "epoch": 1.6649550706033376, + "grad_norm": 0.4906468987464905, + "learning_rate": 3.293316519348475e-05, + "loss": 2.3213, + "step": 1298 + }, + { + "epoch": 1.6662387676508343, + "grad_norm": 0.4314740002155304, + "learning_rate": 3.290846573868918e-05, + "loss": 2.1506, + "step": 1299 + }, + { + "epoch": 1.6675224646983313, + "grad_norm": 0.40390533208847046, + "learning_rate": 3.2883757703064174e-05, + "loss": 2.1911, + "step": 1300 + }, + { + "epoch": 1.6688061617458279, + "grad_norm": 0.46770694851875305, + "learning_rate": 3.2859041113418413e-05, + "loss": 2.2219, + "step": 1301 + }, + { + "epoch": 1.6700898587933248, + "grad_norm": 0.38452643156051636, + "learning_rate": 3.2834315996569844e-05, + "loss": 2.2139, + "step": 1302 + }, + { + "epoch": 1.6713735558408216, + "grad_norm": 0.4044853150844574, + "learning_rate": 3.280958237934565e-05, + "loss": 2.2452, + "step": 1303 + }, + { + "epoch": 1.6726572528883183, + "grad_norm": 0.3696539103984833, + "learning_rate": 3.278484028858228e-05, + "loss": 2.2857, + "step": 1304 + }, + { + "epoch": 1.673940949935815, + "grad_norm": 0.4123825132846832, + "learning_rate": 3.276008975112534e-05, + "loss": 2.2101, + "step": 1305 + }, + { + "epoch": 1.6752246469833119, + "grad_norm": 0.4096415340900421, + "learning_rate": 3.273533079382962e-05, + "loss": 2.0861, + "step": 1306 + }, + { + "epoch": 1.6765083440308088, + "grad_norm": 0.3966728150844574, + "learning_rate": 3.2710563443559045e-05, + "loss": 2.2216, + "step": 1307 + }, + { + "epoch": 1.6777920410783054, + "grad_norm": 0.3775424659252167, + "learning_rate": 3.268578772718663e-05, + "loss": 2.244, + "step": 1308 + }, + { + "epoch": 1.6790757381258024, + "grad_norm": 0.411404550075531, + "learning_rate": 3.266100367159448e-05, + "loss": 2.2038, + "step": 1309 + }, + { + "epoch": 1.680359435173299, + "grad_norm": 0.36207374930381775, + "learning_rate": 3.263621130367375e-05, + "loss": 2.2385, + "step": 1310 + }, + { + "epoch": 1.6816431322207959, + "grad_norm": 0.39945024251937866, + "learning_rate": 3.2611410650324614e-05, + "loss": 2.2851, + "step": 1311 + }, + { + "epoch": 1.6829268292682928, + "grad_norm": 0.3958592712879181, + "learning_rate": 3.2586601738456226e-05, + "loss": 2.2264, + "step": 1312 + }, + { + "epoch": 1.6842105263157894, + "grad_norm": 0.386507123708725, + "learning_rate": 3.256178459498671e-05, + "loss": 2.2365, + "step": 1313 + }, + { + "epoch": 1.6854942233632864, + "grad_norm": 0.4143087565898895, + "learning_rate": 3.253695924684313e-05, + "loss": 2.1712, + "step": 1314 + }, + { + "epoch": 1.686777920410783, + "grad_norm": 0.3835086226463318, + "learning_rate": 3.251212572096143e-05, + "loss": 2.2057, + "step": 1315 + }, + { + "epoch": 1.6880616174582799, + "grad_norm": 0.41994401812553406, + "learning_rate": 3.248728404428643e-05, + "loss": 2.2044, + "step": 1316 + }, + { + "epoch": 1.6893453145057766, + "grad_norm": 0.39442333579063416, + "learning_rate": 3.246243424377182e-05, + "loss": 2.224, + "step": 1317 + }, + { + "epoch": 1.6906290115532734, + "grad_norm": 0.4174175262451172, + "learning_rate": 3.243757634638008e-05, + "loss": 2.1846, + "step": 1318 + }, + { + "epoch": 1.6919127086007704, + "grad_norm": 0.40939804911613464, + "learning_rate": 3.241271037908247e-05, + "loss": 2.2313, + "step": 1319 + }, + { + "epoch": 1.693196405648267, + "grad_norm": 0.4105866849422455, + "learning_rate": 3.238783636885903e-05, + "loss": 2.2061, + "step": 1320 + }, + { + "epoch": 1.6944801026957639, + "grad_norm": 0.37023091316223145, + "learning_rate": 3.2362954342698515e-05, + "loss": 2.2302, + "step": 1321 + }, + { + "epoch": 1.6957637997432606, + "grad_norm": 0.3909175992012024, + "learning_rate": 3.233806432759837e-05, + "loss": 2.2544, + "step": 1322 + }, + { + "epoch": 1.6970474967907574, + "grad_norm": 0.4267197847366333, + "learning_rate": 3.231316635056472e-05, + "loss": 2.2371, + "step": 1323 + }, + { + "epoch": 1.6983311938382541, + "grad_norm": 0.40300336480140686, + "learning_rate": 3.228826043861232e-05, + "loss": 2.3069, + "step": 1324 + }, + { + "epoch": 1.699614890885751, + "grad_norm": 0.3710471987724304, + "learning_rate": 3.2263346618764547e-05, + "loss": 2.1944, + "step": 1325 + }, + { + "epoch": 1.7008985879332479, + "grad_norm": 0.3692680895328522, + "learning_rate": 3.223842491805333e-05, + "loss": 2.1397, + "step": 1326 + }, + { + "epoch": 1.7021822849807444, + "grad_norm": 0.4052692949771881, + "learning_rate": 3.2213495363519184e-05, + "loss": 2.2253, + "step": 1327 + }, + { + "epoch": 1.7034659820282414, + "grad_norm": 0.41429218649864197, + "learning_rate": 3.218855798221114e-05, + "loss": 2.2482, + "step": 1328 + }, + { + "epoch": 1.7047496790757382, + "grad_norm": 0.4166804850101471, + "learning_rate": 3.2163612801186684e-05, + "loss": 2.2101, + "step": 1329 + }, + { + "epoch": 1.706033376123235, + "grad_norm": 0.38851645588874817, + "learning_rate": 3.2138659847511804e-05, + "loss": 2.1313, + "step": 1330 + }, + { + "epoch": 1.7073170731707317, + "grad_norm": 0.4157164692878723, + "learning_rate": 3.211369914826092e-05, + "loss": 2.2886, + "step": 1331 + }, + { + "epoch": 1.7086007702182284, + "grad_norm": 0.385335773229599, + "learning_rate": 3.2088730730516835e-05, + "loss": 2.2048, + "step": 1332 + }, + { + "epoch": 1.7098844672657254, + "grad_norm": 0.387554407119751, + "learning_rate": 3.206375462137074e-05, + "loss": 2.2596, + "step": 1333 + }, + { + "epoch": 1.711168164313222, + "grad_norm": 0.40404000878334045, + "learning_rate": 3.203877084792216e-05, + "loss": 2.2133, + "step": 1334 + }, + { + "epoch": 1.712451861360719, + "grad_norm": 0.40059205889701843, + "learning_rate": 3.201377943727896e-05, + "loss": 2.1791, + "step": 1335 + }, + { + "epoch": 1.7137355584082157, + "grad_norm": 0.3736129105091095, + "learning_rate": 3.198878041655727e-05, + "loss": 2.1687, + "step": 1336 + }, + { + "epoch": 1.7150192554557124, + "grad_norm": 0.3956989049911499, + "learning_rate": 3.196377381288147e-05, + "loss": 2.2798, + "step": 1337 + }, + { + "epoch": 1.7163029525032092, + "grad_norm": 0.402681827545166, + "learning_rate": 3.19387596533842e-05, + "loss": 2.1519, + "step": 1338 + }, + { + "epoch": 1.717586649550706, + "grad_norm": 0.40618735551834106, + "learning_rate": 3.191373796520627e-05, + "loss": 2.2508, + "step": 1339 + }, + { + "epoch": 1.718870346598203, + "grad_norm": 0.38825419545173645, + "learning_rate": 3.1888708775496665e-05, + "loss": 2.2578, + "step": 1340 + }, + { + "epoch": 1.7201540436456995, + "grad_norm": 0.39424875378608704, + "learning_rate": 3.186367211141252e-05, + "loss": 2.2689, + "step": 1341 + }, + { + "epoch": 1.7214377406931964, + "grad_norm": 0.3815070688724518, + "learning_rate": 3.1838628000119066e-05, + "loss": 2.2558, + "step": 1342 + }, + { + "epoch": 1.7227214377406932, + "grad_norm": 0.401339590549469, + "learning_rate": 3.1813576468789616e-05, + "loss": 2.3363, + "step": 1343 + }, + { + "epoch": 1.72400513478819, + "grad_norm": 0.39942529797554016, + "learning_rate": 3.178851754460555e-05, + "loss": 2.1929, + "step": 1344 + }, + { + "epoch": 1.725288831835687, + "grad_norm": 0.4117013216018677, + "learning_rate": 3.1763451254756246e-05, + "loss": 2.3254, + "step": 1345 + }, + { + "epoch": 1.7265725288831835, + "grad_norm": 0.37010103464126587, + "learning_rate": 3.1738377626439084e-05, + "loss": 2.1709, + "step": 1346 + }, + { + "epoch": 1.7278562259306804, + "grad_norm": 0.37751153111457825, + "learning_rate": 3.1713296686859426e-05, + "loss": 2.2218, + "step": 1347 + }, + { + "epoch": 1.7291399229781772, + "grad_norm": 0.3885636031627655, + "learning_rate": 3.168820846323053e-05, + "loss": 2.1923, + "step": 1348 + }, + { + "epoch": 1.730423620025674, + "grad_norm": 0.41108283400535583, + "learning_rate": 3.1663112982773566e-05, + "loss": 2.1655, + "step": 1349 + }, + { + "epoch": 1.7317073170731707, + "grad_norm": 0.3727162182331085, + "learning_rate": 3.1638010272717624e-05, + "loss": 2.2211, + "step": 1350 + }, + { + "epoch": 1.7329910141206675, + "grad_norm": 0.38101616501808167, + "learning_rate": 3.161290036029957e-05, + "loss": 2.286, + "step": 1351 + }, + { + "epoch": 1.7342747111681645, + "grad_norm": 0.395938903093338, + "learning_rate": 3.158778327276413e-05, + "loss": 2.1722, + "step": 1352 + }, + { + "epoch": 1.735558408215661, + "grad_norm": 0.4109671115875244, + "learning_rate": 3.15626590373638e-05, + "loss": 2.2135, + "step": 1353 + }, + { + "epoch": 1.736842105263158, + "grad_norm": 0.3670121133327484, + "learning_rate": 3.1537527681358834e-05, + "loss": 2.2307, + "step": 1354 + }, + { + "epoch": 1.7381258023106547, + "grad_norm": 0.4133658707141876, + "learning_rate": 3.15123892320172e-05, + "loss": 2.1967, + "step": 1355 + }, + { + "epoch": 1.7394094993581515, + "grad_norm": 0.4093180298805237, + "learning_rate": 3.148724371661459e-05, + "loss": 2.2776, + "step": 1356 + }, + { + "epoch": 1.7406931964056482, + "grad_norm": 0.37749165296554565, + "learning_rate": 3.1462091162434336e-05, + "loss": 2.2643, + "step": 1357 + }, + { + "epoch": 1.741976893453145, + "grad_norm": 0.38306891918182373, + "learning_rate": 3.143693159676743e-05, + "loss": 2.2403, + "step": 1358 + }, + { + "epoch": 1.743260590500642, + "grad_norm": 0.3868613541126251, + "learning_rate": 3.141176504691244e-05, + "loss": 2.2451, + "step": 1359 + }, + { + "epoch": 1.7445442875481385, + "grad_norm": 0.40217313170433044, + "learning_rate": 3.138659154017554e-05, + "loss": 2.1337, + "step": 1360 + }, + { + "epoch": 1.7458279845956355, + "grad_norm": 0.45966920256614685, + "learning_rate": 3.1361411103870455e-05, + "loss": 2.1045, + "step": 1361 + }, + { + "epoch": 1.7471116816431322, + "grad_norm": 0.37922942638397217, + "learning_rate": 3.1336223765318394e-05, + "loss": 2.2366, + "step": 1362 + }, + { + "epoch": 1.748395378690629, + "grad_norm": 0.38435661792755127, + "learning_rate": 3.1311029551848096e-05, + "loss": 2.3315, + "step": 1363 + }, + { + "epoch": 1.7496790757381258, + "grad_norm": 0.40681931376457214, + "learning_rate": 3.1285828490795746e-05, + "loss": 2.3369, + "step": 1364 + }, + { + "epoch": 1.7509627727856225, + "grad_norm": 0.4007415771484375, + "learning_rate": 3.126062060950494e-05, + "loss": 2.2209, + "step": 1365 + }, + { + "epoch": 1.7522464698331195, + "grad_norm": 0.40734994411468506, + "learning_rate": 3.123540593532668e-05, + "loss": 2.1657, + "step": 1366 + }, + { + "epoch": 1.753530166880616, + "grad_norm": 0.3657030463218689, + "learning_rate": 3.1210184495619385e-05, + "loss": 2.2239, + "step": 1367 + }, + { + "epoch": 1.754813863928113, + "grad_norm": 0.42279279232025146, + "learning_rate": 3.118495631774874e-05, + "loss": 2.1965, + "step": 1368 + }, + { + "epoch": 1.7560975609756098, + "grad_norm": 0.35997843742370605, + "learning_rate": 3.1159721429087785e-05, + "loss": 2.2264, + "step": 1369 + }, + { + "epoch": 1.7573812580231065, + "grad_norm": 0.3979395627975464, + "learning_rate": 3.1134479857016844e-05, + "loss": 2.2401, + "step": 1370 + }, + { + "epoch": 1.7586649550706035, + "grad_norm": 0.42325690388679504, + "learning_rate": 3.110923162892348e-05, + "loss": 2.2058, + "step": 1371 + }, + { + "epoch": 1.7599486521181, + "grad_norm": 0.39083489775657654, + "learning_rate": 3.1083976772202476e-05, + "loss": 2.1864, + "step": 1372 + }, + { + "epoch": 1.761232349165597, + "grad_norm": 0.37045979499816895, + "learning_rate": 3.105871531425582e-05, + "loss": 2.2315, + "step": 1373 + }, + { + "epoch": 1.7625160462130935, + "grad_norm": 0.42074283957481384, + "learning_rate": 3.1033447282492646e-05, + "loss": 2.2669, + "step": 1374 + }, + { + "epoch": 1.7637997432605905, + "grad_norm": 0.40470820665359497, + "learning_rate": 3.100817270432922e-05, + "loss": 2.2965, + "step": 1375 + }, + { + "epoch": 1.7650834403080873, + "grad_norm": 0.39896172285079956, + "learning_rate": 3.098289160718895e-05, + "loss": 2.184, + "step": 1376 + }, + { + "epoch": 1.766367137355584, + "grad_norm": 0.4028201997280121, + "learning_rate": 3.0957604018502256e-05, + "loss": 2.2784, + "step": 1377 + }, + { + "epoch": 1.767650834403081, + "grad_norm": 0.38847222924232483, + "learning_rate": 3.093230996570665e-05, + "loss": 2.2041, + "step": 1378 + }, + { + "epoch": 1.7689345314505776, + "grad_norm": 0.3881203830242157, + "learning_rate": 3.090700947624663e-05, + "loss": 2.2305, + "step": 1379 + }, + { + "epoch": 1.7702182284980745, + "grad_norm": 0.4000556468963623, + "learning_rate": 3.088170257757369e-05, + "loss": 2.1686, + "step": 1380 + }, + { + "epoch": 1.7715019255455713, + "grad_norm": 0.42821210622787476, + "learning_rate": 3.085638929714629e-05, + "loss": 2.2836, + "step": 1381 + }, + { + "epoch": 1.772785622593068, + "grad_norm": 0.3939917981624603, + "learning_rate": 3.083106966242978e-05, + "loss": 2.2193, + "step": 1382 + }, + { + "epoch": 1.7740693196405648, + "grad_norm": 0.3647666573524475, + "learning_rate": 3.080574370089643e-05, + "loss": 2.2187, + "step": 1383 + }, + { + "epoch": 1.7753530166880616, + "grad_norm": 0.34855392575263977, + "learning_rate": 3.0780411440025394e-05, + "loss": 2.2365, + "step": 1384 + }, + { + "epoch": 1.7766367137355585, + "grad_norm": 0.37924930453300476, + "learning_rate": 3.07550729073026e-05, + "loss": 2.2242, + "step": 1385 + }, + { + "epoch": 1.777920410783055, + "grad_norm": 0.40062305331230164, + "learning_rate": 3.072972813022084e-05, + "loss": 2.2718, + "step": 1386 + }, + { + "epoch": 1.779204107830552, + "grad_norm": 0.4480050504207611, + "learning_rate": 3.070437713627965e-05, + "loss": 2.2009, + "step": 1387 + }, + { + "epoch": 1.7804878048780488, + "grad_norm": 0.4204918444156647, + "learning_rate": 3.0679019952985325e-05, + "loss": 2.2574, + "step": 1388 + }, + { + "epoch": 1.7817715019255456, + "grad_norm": 0.3950962424278259, + "learning_rate": 3.065365660785088e-05, + "loss": 2.1981, + "step": 1389 + }, + { + "epoch": 1.7830551989730423, + "grad_norm": 0.4228907525539398, + "learning_rate": 3.062828712839598e-05, + "loss": 2.3068, + "step": 1390 + }, + { + "epoch": 1.784338896020539, + "grad_norm": 0.4032195806503296, + "learning_rate": 3.0602911542147004e-05, + "loss": 2.2801, + "step": 1391 + }, + { + "epoch": 1.785622593068036, + "grad_norm": 0.38753437995910645, + "learning_rate": 3.057752987663691e-05, + "loss": 2.2444, + "step": 1392 + }, + { + "epoch": 1.7869062901155326, + "grad_norm": 0.36643412709236145, + "learning_rate": 3.0552142159405255e-05, + "loss": 2.2959, + "step": 1393 + }, + { + "epoch": 1.7881899871630296, + "grad_norm": 0.38905465602874756, + "learning_rate": 3.0526748417998195e-05, + "loss": 2.2325, + "step": 1394 + }, + { + "epoch": 1.7894736842105263, + "grad_norm": 0.4012605547904968, + "learning_rate": 3.05013486799684e-05, + "loss": 2.1763, + "step": 1395 + }, + { + "epoch": 1.790757381258023, + "grad_norm": 0.39019903540611267, + "learning_rate": 3.047594297287504e-05, + "loss": 2.1153, + "step": 1396 + }, + { + "epoch": 1.79204107830552, + "grad_norm": 0.40327882766723633, + "learning_rate": 3.0450531324283766e-05, + "loss": 2.1646, + "step": 1397 + }, + { + "epoch": 1.7933247753530166, + "grad_norm": 0.3853529989719391, + "learning_rate": 3.04251137617667e-05, + "loss": 2.1396, + "step": 1398 + }, + { + "epoch": 1.7946084724005136, + "grad_norm": 0.37933430075645447, + "learning_rate": 3.0399690312902345e-05, + "loss": 2.303, + "step": 1399 + }, + { + "epoch": 1.7958921694480101, + "grad_norm": 0.4066120982170105, + "learning_rate": 3.0374261005275607e-05, + "loss": 2.2842, + "step": 1400 + }, + { + "epoch": 1.797175866495507, + "grad_norm": 0.4104210436344147, + "learning_rate": 3.034882586647775e-05, + "loss": 2.2034, + "step": 1401 + }, + { + "epoch": 1.7984595635430038, + "grad_norm": 0.388218492269516, + "learning_rate": 3.0323384924106357e-05, + "loss": 2.2327, + "step": 1402 + }, + { + "epoch": 1.7997432605905006, + "grad_norm": 0.4327325224876404, + "learning_rate": 3.0297938205765337e-05, + "loss": 2.3046, + "step": 1403 + }, + { + "epoch": 1.8010269576379976, + "grad_norm": 0.39868342876434326, + "learning_rate": 3.027248573906481e-05, + "loss": 2.1764, + "step": 1404 + }, + { + "epoch": 1.8023106546854941, + "grad_norm": 0.37082406878471375, + "learning_rate": 3.024702755162119e-05, + "loss": 2.2265, + "step": 1405 + }, + { + "epoch": 1.803594351732991, + "grad_norm": 0.3907563090324402, + "learning_rate": 3.0221563671057073e-05, + "loss": 2.2932, + "step": 1406 + }, + { + "epoch": 1.8048780487804879, + "grad_norm": 0.4111311435699463, + "learning_rate": 3.0196094125001217e-05, + "loss": 2.2655, + "step": 1407 + }, + { + "epoch": 1.8061617458279846, + "grad_norm": 0.38395512104034424, + "learning_rate": 3.0170618941088558e-05, + "loss": 2.3065, + "step": 1408 + }, + { + "epoch": 1.8074454428754814, + "grad_norm": 0.3818804621696472, + "learning_rate": 3.014513814696013e-05, + "loss": 2.2884, + "step": 1409 + }, + { + "epoch": 1.8087291399229781, + "grad_norm": 0.3713618516921997, + "learning_rate": 3.0119651770263063e-05, + "loss": 2.2115, + "step": 1410 + }, + { + "epoch": 1.810012836970475, + "grad_norm": 0.36225947737693787, + "learning_rate": 3.009415983865053e-05, + "loss": 2.1566, + "step": 1411 + }, + { + "epoch": 1.8112965340179716, + "grad_norm": 0.4227888584136963, + "learning_rate": 3.006866237978176e-05, + "loss": 2.2618, + "step": 1412 + }, + { + "epoch": 1.8125802310654686, + "grad_norm": 0.40623441338539124, + "learning_rate": 3.004315942132194e-05, + "loss": 2.3212, + "step": 1413 + }, + { + "epoch": 1.8138639281129654, + "grad_norm": 0.40290600061416626, + "learning_rate": 3.0017650990942255e-05, + "loss": 2.3071, + "step": 1414 + }, + { + "epoch": 1.8151476251604621, + "grad_norm": 0.40565595030784607, + "learning_rate": 2.9992137116319814e-05, + "loss": 2.2149, + "step": 1415 + }, + { + "epoch": 1.8164313222079589, + "grad_norm": 0.40337786078453064, + "learning_rate": 2.996661782513764e-05, + "loss": 2.247, + "step": 1416 + }, + { + "epoch": 1.8177150192554556, + "grad_norm": 0.3812531530857086, + "learning_rate": 2.994109314508462e-05, + "loss": 2.1495, + "step": 1417 + }, + { + "epoch": 1.8189987163029526, + "grad_norm": 0.4289066195487976, + "learning_rate": 2.9915563103855503e-05, + "loss": 2.248, + "step": 1418 + }, + { + "epoch": 1.8202824133504492, + "grad_norm": 0.3501010239124298, + "learning_rate": 2.9890027729150847e-05, + "loss": 2.1637, + "step": 1419 + }, + { + "epoch": 1.8215661103979461, + "grad_norm": 0.3857298791408539, + "learning_rate": 2.9864487048677015e-05, + "loss": 2.2071, + "step": 1420 + }, + { + "epoch": 1.822849807445443, + "grad_norm": 0.40616750717163086, + "learning_rate": 2.9838941090146083e-05, + "loss": 2.1312, + "step": 1421 + }, + { + "epoch": 1.8241335044929397, + "grad_norm": 0.3850340247154236, + "learning_rate": 2.98133898812759e-05, + "loss": 2.1842, + "step": 1422 + }, + { + "epoch": 1.8254172015404364, + "grad_norm": 0.3876679241657257, + "learning_rate": 2.9787833449789982e-05, + "loss": 2.2447, + "step": 1423 + }, + { + "epoch": 1.8267008985879332, + "grad_norm": 0.40384963154792786, + "learning_rate": 2.9762271823417532e-05, + "loss": 2.2854, + "step": 1424 + }, + { + "epoch": 1.8279845956354301, + "grad_norm": 0.3864003121852875, + "learning_rate": 2.973670502989338e-05, + "loss": 2.2022, + "step": 1425 + }, + { + "epoch": 1.8292682926829267, + "grad_norm": 0.38904649019241333, + "learning_rate": 2.9711133096957962e-05, + "loss": 2.3013, + "step": 1426 + }, + { + "epoch": 1.8305519897304237, + "grad_norm": 0.4106447696685791, + "learning_rate": 2.9685556052357284e-05, + "loss": 2.3069, + "step": 1427 + }, + { + "epoch": 1.8318356867779204, + "grad_norm": 0.40486323833465576, + "learning_rate": 2.9659973923842908e-05, + "loss": 2.255, + "step": 1428 + }, + { + "epoch": 1.8331193838254172, + "grad_norm": 0.4210253059864044, + "learning_rate": 2.963438673917192e-05, + "loss": 2.238, + "step": 1429 + }, + { + "epoch": 1.8344030808729141, + "grad_norm": 0.38917699456214905, + "learning_rate": 2.960879452610687e-05, + "loss": 2.2353, + "step": 1430 + }, + { + "epoch": 1.8356867779204107, + "grad_norm": 0.3805764615535736, + "learning_rate": 2.9583197312415777e-05, + "loss": 2.1832, + "step": 1431 + }, + { + "epoch": 1.8369704749679077, + "grad_norm": 0.40074852108955383, + "learning_rate": 2.9557595125872094e-05, + "loss": 2.2551, + "step": 1432 + }, + { + "epoch": 1.8382541720154044, + "grad_norm": 0.3935510814189911, + "learning_rate": 2.9531987994254638e-05, + "loss": 2.1531, + "step": 1433 + }, + { + "epoch": 1.8395378690629012, + "grad_norm": 0.3827677071094513, + "learning_rate": 2.950637594534765e-05, + "loss": 2.2708, + "step": 1434 + }, + { + "epoch": 1.840821566110398, + "grad_norm": 0.38927167654037476, + "learning_rate": 2.9480759006940634e-05, + "loss": 2.1376, + "step": 1435 + }, + { + "epoch": 1.8421052631578947, + "grad_norm": 0.39499616622924805, + "learning_rate": 2.9455137206828444e-05, + "loss": 2.1954, + "step": 1436 + }, + { + "epoch": 1.8433889602053917, + "grad_norm": 0.4076292812824249, + "learning_rate": 2.942951057281122e-05, + "loss": 2.1735, + "step": 1437 + }, + { + "epoch": 1.8446726572528882, + "grad_norm": 0.38959750533103943, + "learning_rate": 2.9403879132694295e-05, + "loss": 2.1761, + "step": 1438 + }, + { + "epoch": 1.8459563543003852, + "grad_norm": 0.44564783573150635, + "learning_rate": 2.9378242914288272e-05, + "loss": 2.2603, + "step": 1439 + }, + { + "epoch": 1.847240051347882, + "grad_norm": 0.36392900347709656, + "learning_rate": 2.935260194540891e-05, + "loss": 2.2355, + "step": 1440 + }, + { + "epoch": 1.8485237483953787, + "grad_norm": 0.39503583312034607, + "learning_rate": 2.9326956253877125e-05, + "loss": 2.214, + "step": 1441 + }, + { + "epoch": 1.8498074454428755, + "grad_norm": 0.3977903723716736, + "learning_rate": 2.9301305867518968e-05, + "loss": 2.2468, + "step": 1442 + }, + { + "epoch": 1.8510911424903722, + "grad_norm": 0.3888460695743561, + "learning_rate": 2.927565081416555e-05, + "loss": 2.2624, + "step": 1443 + }, + { + "epoch": 1.8523748395378692, + "grad_norm": 0.39346474409103394, + "learning_rate": 2.9249991121653103e-05, + "loss": 2.243, + "step": 1444 + }, + { + "epoch": 1.8536585365853657, + "grad_norm": 0.42184969782829285, + "learning_rate": 2.9224326817822845e-05, + "loss": 2.1693, + "step": 1445 + }, + { + "epoch": 1.8549422336328627, + "grad_norm": 0.4167838990688324, + "learning_rate": 2.9198657930521012e-05, + "loss": 2.2692, + "step": 1446 + }, + { + "epoch": 1.8562259306803595, + "grad_norm": 0.4268261194229126, + "learning_rate": 2.917298448759882e-05, + "loss": 2.1966, + "step": 1447 + }, + { + "epoch": 1.8575096277278562, + "grad_norm": 0.3667944371700287, + "learning_rate": 2.914730651691242e-05, + "loss": 2.246, + "step": 1448 + }, + { + "epoch": 1.858793324775353, + "grad_norm": 0.4406207501888275, + "learning_rate": 2.912162404632288e-05, + "loss": 2.1493, + "step": 1449 + }, + { + "epoch": 1.8600770218228497, + "grad_norm": 0.44947564601898193, + "learning_rate": 2.909593710369614e-05, + "loss": 2.2594, + "step": 1450 + }, + { + "epoch": 1.8613607188703467, + "grad_norm": 0.3738134801387787, + "learning_rate": 2.9070245716903016e-05, + "loss": 2.2879, + "step": 1451 + }, + { + "epoch": 1.8626444159178432, + "grad_norm": 0.3906819820404053, + "learning_rate": 2.9044549913819124e-05, + "loss": 2.2441, + "step": 1452 + }, + { + "epoch": 1.8639281129653402, + "grad_norm": 0.4164174199104309, + "learning_rate": 2.9018849722324864e-05, + "loss": 2.2237, + "step": 1453 + }, + { + "epoch": 1.865211810012837, + "grad_norm": 0.4306255877017975, + "learning_rate": 2.8993145170305442e-05, + "loss": 2.2071, + "step": 1454 + }, + { + "epoch": 1.8664955070603337, + "grad_norm": 0.4033021330833435, + "learning_rate": 2.8967436285650733e-05, + "loss": 2.2543, + "step": 1455 + }, + { + "epoch": 1.8677792041078307, + "grad_norm": 0.37400010228157043, + "learning_rate": 2.894172309625538e-05, + "loss": 2.1309, + "step": 1456 + }, + { + "epoch": 1.8690629011553272, + "grad_norm": 0.3806757628917694, + "learning_rate": 2.8916005630018623e-05, + "loss": 2.2136, + "step": 1457 + }, + { + "epoch": 1.8703465982028242, + "grad_norm": 0.402248740196228, + "learning_rate": 2.8890283914844413e-05, + "loss": 2.2479, + "step": 1458 + }, + { + "epoch": 1.8716302952503208, + "grad_norm": 0.38702309131622314, + "learning_rate": 2.886455797864127e-05, + "loss": 2.2523, + "step": 1459 + }, + { + "epoch": 1.8729139922978177, + "grad_norm": 0.4400423467159271, + "learning_rate": 2.883882784932229e-05, + "loss": 2.2503, + "step": 1460 + }, + { + "epoch": 1.8741976893453145, + "grad_norm": 0.40396174788475037, + "learning_rate": 2.8813093554805155e-05, + "loss": 2.2292, + "step": 1461 + }, + { + "epoch": 1.8754813863928113, + "grad_norm": 0.4026391804218292, + "learning_rate": 2.878735512301203e-05, + "loss": 2.2952, + "step": 1462 + }, + { + "epoch": 1.8767650834403082, + "grad_norm": 0.3762260973453522, + "learning_rate": 2.876161258186958e-05, + "loss": 2.1583, + "step": 1463 + }, + { + "epoch": 1.8780487804878048, + "grad_norm": 0.3629817068576813, + "learning_rate": 2.8735865959308938e-05, + "loss": 2.3263, + "step": 1464 + }, + { + "epoch": 1.8793324775353017, + "grad_norm": 0.3951891362667084, + "learning_rate": 2.8710115283265655e-05, + "loss": 2.235, + "step": 1465 + }, + { + "epoch": 1.8806161745827985, + "grad_norm": 0.4182018041610718, + "learning_rate": 2.8684360581679687e-05, + "loss": 2.3031, + "step": 1466 + }, + { + "epoch": 1.8818998716302953, + "grad_norm": 0.39251741766929626, + "learning_rate": 2.8658601882495357e-05, + "loss": 2.2141, + "step": 1467 + }, + { + "epoch": 1.883183568677792, + "grad_norm": 0.3830741047859192, + "learning_rate": 2.863283921366131e-05, + "loss": 2.2443, + "step": 1468 + }, + { + "epoch": 1.8844672657252888, + "grad_norm": 0.4047439694404602, + "learning_rate": 2.8607072603130524e-05, + "loss": 2.1827, + "step": 1469 + }, + { + "epoch": 1.8857509627727858, + "grad_norm": 0.38761255145072937, + "learning_rate": 2.858130207886024e-05, + "loss": 2.145, + "step": 1470 + }, + { + "epoch": 1.8870346598202823, + "grad_norm": 0.3929939866065979, + "learning_rate": 2.855552766881194e-05, + "loss": 2.204, + "step": 1471 + }, + { + "epoch": 1.8883183568677793, + "grad_norm": 0.41997402906417847, + "learning_rate": 2.852974940095133e-05, + "loss": 2.1652, + "step": 1472 + }, + { + "epoch": 1.889602053915276, + "grad_norm": 0.39648181200027466, + "learning_rate": 2.850396730324831e-05, + "loss": 2.2071, + "step": 1473 + }, + { + "epoch": 1.8908857509627728, + "grad_norm": 0.39045315980911255, + "learning_rate": 2.8478181403676908e-05, + "loss": 2.2707, + "step": 1474 + }, + { + "epoch": 1.8921694480102695, + "grad_norm": 0.4029048681259155, + "learning_rate": 2.845239173021531e-05, + "loss": 2.2134, + "step": 1475 + }, + { + "epoch": 1.8934531450577663, + "grad_norm": 0.37862762808799744, + "learning_rate": 2.8426598310845782e-05, + "loss": 2.2686, + "step": 1476 + }, + { + "epoch": 1.8947368421052633, + "grad_norm": 0.4509231448173523, + "learning_rate": 2.8400801173554635e-05, + "loss": 2.1764, + "step": 1477 + }, + { + "epoch": 1.8960205391527598, + "grad_norm": 0.3948415219783783, + "learning_rate": 2.8375000346332255e-05, + "loss": 2.2594, + "step": 1478 + }, + { + "epoch": 1.8973042362002568, + "grad_norm": 0.3953462541103363, + "learning_rate": 2.8349195857173e-05, + "loss": 2.2111, + "step": 1479 + }, + { + "epoch": 1.8985879332477535, + "grad_norm": 0.39922788739204407, + "learning_rate": 2.8323387734075213e-05, + "loss": 2.186, + "step": 1480 + }, + { + "epoch": 1.8998716302952503, + "grad_norm": 0.40049436688423157, + "learning_rate": 2.8297576005041177e-05, + "loss": 2.2176, + "step": 1481 + }, + { + "epoch": 1.901155327342747, + "grad_norm": 0.39044493436813354, + "learning_rate": 2.8271760698077087e-05, + "loss": 2.1974, + "step": 1482 + }, + { + "epoch": 1.9024390243902438, + "grad_norm": 0.3833390772342682, + "learning_rate": 2.8245941841193023e-05, + "loss": 2.1969, + "step": 1483 + }, + { + "epoch": 1.9037227214377408, + "grad_norm": 0.42505940794944763, + "learning_rate": 2.8220119462402916e-05, + "loss": 2.3153, + "step": 1484 + }, + { + "epoch": 1.9050064184852373, + "grad_norm": 0.4005906283855438, + "learning_rate": 2.8194293589724514e-05, + "loss": 2.2303, + "step": 1485 + }, + { + "epoch": 1.9062901155327343, + "grad_norm": 0.4483509957790375, + "learning_rate": 2.8168464251179365e-05, + "loss": 2.1588, + "step": 1486 + }, + { + "epoch": 1.907573812580231, + "grad_norm": 0.3799138367176056, + "learning_rate": 2.814263147479277e-05, + "loss": 2.1147, + "step": 1487 + }, + { + "epoch": 1.9088575096277278, + "grad_norm": 0.3586787283420563, + "learning_rate": 2.8116795288593755e-05, + "loss": 2.1957, + "step": 1488 + }, + { + "epoch": 1.9101412066752248, + "grad_norm": 0.3805423974990845, + "learning_rate": 2.8090955720615058e-05, + "loss": 2.1924, + "step": 1489 + }, + { + "epoch": 1.9114249037227213, + "grad_norm": 0.3956030011177063, + "learning_rate": 2.806511279889309e-05, + "loss": 2.1733, + "step": 1490 + }, + { + "epoch": 1.9127086007702183, + "grad_norm": 0.3691265285015106, + "learning_rate": 2.8039266551467873e-05, + "loss": 2.1685, + "step": 1491 + }, + { + "epoch": 1.913992297817715, + "grad_norm": 0.40167438983917236, + "learning_rate": 2.8013417006383076e-05, + "loss": 2.1618, + "step": 1492 + }, + { + "epoch": 1.9152759948652118, + "grad_norm": 0.4268973469734192, + "learning_rate": 2.7987564191685922e-05, + "loss": 2.2607, + "step": 1493 + }, + { + "epoch": 1.9165596919127086, + "grad_norm": 0.4193055331707001, + "learning_rate": 2.7961708135427167e-05, + "loss": 2.1967, + "step": 1494 + }, + { + "epoch": 1.9178433889602053, + "grad_norm": 0.37377044558525085, + "learning_rate": 2.7935848865661135e-05, + "loss": 2.1774, + "step": 1495 + }, + { + "epoch": 1.9191270860077023, + "grad_norm": 0.39957597851753235, + "learning_rate": 2.790998641044557e-05, + "loss": 2.1548, + "step": 1496 + }, + { + "epoch": 1.9204107830551989, + "grad_norm": 0.39428436756134033, + "learning_rate": 2.788412079784174e-05, + "loss": 2.2126, + "step": 1497 + }, + { + "epoch": 1.9216944801026958, + "grad_norm": 0.37812477350234985, + "learning_rate": 2.785825205591428e-05, + "loss": 2.2233, + "step": 1498 + }, + { + "epoch": 1.9229781771501926, + "grad_norm": 0.3757867217063904, + "learning_rate": 2.7832380212731264e-05, + "loss": 2.1993, + "step": 1499 + }, + { + "epoch": 1.9242618741976893, + "grad_norm": 0.4228513538837433, + "learning_rate": 2.7806505296364105e-05, + "loss": 2.1633, + "step": 1500 + }, + { + "epoch": 1.925545571245186, + "grad_norm": 0.3849712610244751, + "learning_rate": 2.778062733488756e-05, + "loss": 2.2824, + "step": 1501 + }, + { + "epoch": 1.9268292682926829, + "grad_norm": 0.36626866459846497, + "learning_rate": 2.775474635637969e-05, + "loss": 2.1655, + "step": 1502 + }, + { + "epoch": 1.9281129653401798, + "grad_norm": 0.4242492914199829, + "learning_rate": 2.772886238892183e-05, + "loss": 2.3011, + "step": 1503 + }, + { + "epoch": 1.9293966623876764, + "grad_norm": 0.44785967469215393, + "learning_rate": 2.7702975460598547e-05, + "loss": 2.1282, + "step": 1504 + }, + { + "epoch": 1.9306803594351734, + "grad_norm": 0.38648855686187744, + "learning_rate": 2.7677085599497643e-05, + "loss": 2.2709, + "step": 1505 + }, + { + "epoch": 1.9319640564826701, + "grad_norm": 0.40068304538726807, + "learning_rate": 2.765119283371007e-05, + "loss": 2.2029, + "step": 1506 + }, + { + "epoch": 1.9332477535301669, + "grad_norm": 0.3858581483364105, + "learning_rate": 2.762529719132998e-05, + "loss": 2.2835, + "step": 1507 + }, + { + "epoch": 1.9345314505776636, + "grad_norm": 0.38977059721946716, + "learning_rate": 2.7599398700454587e-05, + "loss": 2.3018, + "step": 1508 + }, + { + "epoch": 1.9358151476251604, + "grad_norm": 0.3997935354709625, + "learning_rate": 2.757349738918425e-05, + "loss": 2.2489, + "step": 1509 + }, + { + "epoch": 1.9370988446726574, + "grad_norm": 0.4249832034111023, + "learning_rate": 2.7547593285622347e-05, + "loss": 2.2626, + "step": 1510 + }, + { + "epoch": 1.938382541720154, + "grad_norm": 0.3619246780872345, + "learning_rate": 2.75216864178753e-05, + "loss": 2.2149, + "step": 1511 + }, + { + "epoch": 1.9396662387676509, + "grad_norm": 0.39956042170524597, + "learning_rate": 2.749577681405256e-05, + "loss": 2.2457, + "step": 1512 + }, + { + "epoch": 1.9409499358151476, + "grad_norm": 0.40953347086906433, + "learning_rate": 2.746986450226648e-05, + "loss": 2.1778, + "step": 1513 + }, + { + "epoch": 1.9422336328626444, + "grad_norm": 0.36535537242889404, + "learning_rate": 2.7443949510632417e-05, + "loss": 2.2318, + "step": 1514 + }, + { + "epoch": 1.9435173299101414, + "grad_norm": 0.37131211161613464, + "learning_rate": 2.7418031867268607e-05, + "loss": 2.252, + "step": 1515 + }, + { + "epoch": 1.944801026957638, + "grad_norm": 0.41581442952156067, + "learning_rate": 2.7392111600296154e-05, + "loss": 2.2889, + "step": 1516 + }, + { + "epoch": 1.9460847240051349, + "grad_norm": 0.38260316848754883, + "learning_rate": 2.7366188737839026e-05, + "loss": 2.2225, + "step": 1517 + }, + { + "epoch": 1.9473684210526314, + "grad_norm": 0.4338550269603729, + "learning_rate": 2.734026330802401e-05, + "loss": 2.2156, + "step": 1518 + }, + { + "epoch": 1.9486521181001284, + "grad_norm": 0.4523008465766907, + "learning_rate": 2.7314335338980657e-05, + "loss": 2.2107, + "step": 1519 + }, + { + "epoch": 1.9499358151476252, + "grad_norm": 0.39179036021232605, + "learning_rate": 2.7288404858841287e-05, + "loss": 2.2469, + "step": 1520 + }, + { + "epoch": 1.951219512195122, + "grad_norm": 0.6539322137832642, + "learning_rate": 2.726247189574095e-05, + "loss": 2.1483, + "step": 1521 + }, + { + "epoch": 1.9525032092426189, + "grad_norm": 0.3857417702674866, + "learning_rate": 2.7236536477817375e-05, + "loss": 2.3102, + "step": 1522 + }, + { + "epoch": 1.9537869062901154, + "grad_norm": 0.3993578255176544, + "learning_rate": 2.7210598633210977e-05, + "loss": 2.2241, + "step": 1523 + }, + { + "epoch": 1.9550706033376124, + "grad_norm": 0.4131927192211151, + "learning_rate": 2.7184658390064777e-05, + "loss": 2.3392, + "step": 1524 + }, + { + "epoch": 1.9563543003851092, + "grad_norm": 0.4235398471355438, + "learning_rate": 2.7158715776524406e-05, + "loss": 2.2559, + "step": 1525 + }, + { + "epoch": 1.957637997432606, + "grad_norm": 0.3778948187828064, + "learning_rate": 2.71327708207381e-05, + "loss": 2.3383, + "step": 1526 + }, + { + "epoch": 1.9589216944801027, + "grad_norm": 0.40795183181762695, + "learning_rate": 2.710682355085658e-05, + "loss": 2.1388, + "step": 1527 + }, + { + "epoch": 1.9602053915275994, + "grad_norm": 0.3729285001754761, + "learning_rate": 2.7080873995033116e-05, + "loss": 2.2616, + "step": 1528 + }, + { + "epoch": 1.9614890885750964, + "grad_norm": 0.41110965609550476, + "learning_rate": 2.705492218142346e-05, + "loss": 2.2194, + "step": 1529 + }, + { + "epoch": 1.962772785622593, + "grad_norm": 0.3804199695587158, + "learning_rate": 2.7028968138185782e-05, + "loss": 2.3603, + "step": 1530 + }, + { + "epoch": 1.96405648267009, + "grad_norm": 0.4288075268268585, + "learning_rate": 2.7003011893480713e-05, + "loss": 2.3144, + "step": 1531 + }, + { + "epoch": 1.9653401797175867, + "grad_norm": 0.39279666543006897, + "learning_rate": 2.6977053475471233e-05, + "loss": 2.2149, + "step": 1532 + }, + { + "epoch": 1.9666238767650834, + "grad_norm": 0.40739279985427856, + "learning_rate": 2.6951092912322713e-05, + "loss": 2.1658, + "step": 1533 + }, + { + "epoch": 1.9679075738125802, + "grad_norm": 0.3970479965209961, + "learning_rate": 2.6925130232202833e-05, + "loss": 2.0891, + "step": 1534 + }, + { + "epoch": 1.969191270860077, + "grad_norm": 0.4086737632751465, + "learning_rate": 2.6899165463281563e-05, + "loss": 2.2883, + "step": 1535 + }, + { + "epoch": 1.970474967907574, + "grad_norm": 0.41012001037597656, + "learning_rate": 2.6873198633731162e-05, + "loss": 2.26, + "step": 1536 + }, + { + "epoch": 1.9717586649550705, + "grad_norm": 0.3845211863517761, + "learning_rate": 2.6847229771726113e-05, + "loss": 2.3165, + "step": 1537 + }, + { + "epoch": 1.9730423620025674, + "grad_norm": 0.4046964943408966, + "learning_rate": 2.6821258905443105e-05, + "loss": 2.2698, + "step": 1538 + }, + { + "epoch": 1.9743260590500642, + "grad_norm": 0.47236689925193787, + "learning_rate": 2.6795286063060992e-05, + "loss": 2.2492, + "step": 1539 + }, + { + "epoch": 1.975609756097561, + "grad_norm": 0.3712567687034607, + "learning_rate": 2.6769311272760796e-05, + "loss": 2.167, + "step": 1540 + }, + { + "epoch": 1.976893453145058, + "grad_norm": 0.40271466970443726, + "learning_rate": 2.674333456272562e-05, + "loss": 2.2687, + "step": 1541 + }, + { + "epoch": 1.9781771501925545, + "grad_norm": 0.4083406627178192, + "learning_rate": 2.671735596114068e-05, + "loss": 2.2416, + "step": 1542 + }, + { + "epoch": 1.9794608472400514, + "grad_norm": 0.42068684101104736, + "learning_rate": 2.6691375496193234e-05, + "loss": 2.2504, + "step": 1543 + }, + { + "epoch": 1.980744544287548, + "grad_norm": 0.396812379360199, + "learning_rate": 2.6665393196072553e-05, + "loss": 2.2013, + "step": 1544 + }, + { + "epoch": 1.982028241335045, + "grad_norm": 0.38998210430145264, + "learning_rate": 2.6639409088969897e-05, + "loss": 2.1355, + "step": 1545 + }, + { + "epoch": 1.9833119383825417, + "grad_norm": 0.4291171431541443, + "learning_rate": 2.6613423203078525e-05, + "loss": 2.3018, + "step": 1546 + }, + { + "epoch": 1.9845956354300385, + "grad_norm": 0.4067135751247406, + "learning_rate": 2.658743556659356e-05, + "loss": 2.2752, + "step": 1547 + }, + { + "epoch": 1.9858793324775355, + "grad_norm": 0.40728282928466797, + "learning_rate": 2.6561446207712098e-05, + "loss": 2.1935, + "step": 1548 + }, + { + "epoch": 1.987163029525032, + "grad_norm": 0.406803697347641, + "learning_rate": 2.653545515463303e-05, + "loss": 2.2102, + "step": 1549 + }, + { + "epoch": 1.988446726572529, + "grad_norm": 0.3946593105792999, + "learning_rate": 2.6509462435557152e-05, + "loss": 2.1531, + "step": 1550 + }, + { + "epoch": 1.9897304236200257, + "grad_norm": 0.3815438747406006, + "learning_rate": 2.6483468078687024e-05, + "loss": 2.1454, + "step": 1551 + }, + { + "epoch": 1.9910141206675225, + "grad_norm": 0.39946454763412476, + "learning_rate": 2.6457472112227e-05, + "loss": 2.2229, + "step": 1552 + }, + { + "epoch": 1.9922978177150192, + "grad_norm": 0.409060001373291, + "learning_rate": 2.643147456438318e-05, + "loss": 2.219, + "step": 1553 + }, + { + "epoch": 1.993581514762516, + "grad_norm": 0.3979557156562805, + "learning_rate": 2.640547546336337e-05, + "loss": 2.1489, + "step": 1554 + }, + { + "epoch": 1.994865211810013, + "grad_norm": 0.3794998228549957, + "learning_rate": 2.6379474837377072e-05, + "loss": 2.2319, + "step": 1555 + }, + { + "epoch": 1.9961489088575095, + "grad_norm": 0.4063602685928345, + "learning_rate": 2.635347271463544e-05, + "loss": 2.2959, + "step": 1556 + }, + { + "epoch": 1.9974326059050065, + "grad_norm": 0.4122906029224396, + "learning_rate": 2.632746912335125e-05, + "loss": 2.2654, + "step": 1557 + }, + { + "epoch": 1.9987163029525032, + "grad_norm": 0.4008810222148895, + "learning_rate": 2.6301464091738876e-05, + "loss": 2.1928, + "step": 1558 + }, + { + "epoch": 2.0, + "grad_norm": 0.40754011273384094, + "learning_rate": 2.627545764801424e-05, + "loss": 2.2289, + "step": 1559 + }, + { + "epoch": 2.001283697047497, + "grad_norm": 0.41766345500946045, + "learning_rate": 2.6249449820394822e-05, + "loss": 2.1962, + "step": 1560 + }, + { + "epoch": 2.001283697047497, + "grad_norm": 0.3683842718601227, + "learning_rate": 2.622344063709957e-05, + "loss": 2.0411, + "step": 1561 + }, + { + "epoch": 2.0025673940949935, + "grad_norm": 0.3628217875957489, + "learning_rate": 2.6197430126348953e-05, + "loss": 2.0484, + "step": 1562 + }, + { + "epoch": 2.0038510911424905, + "grad_norm": 0.36648184061050415, + "learning_rate": 2.617141831636482e-05, + "loss": 2.0637, + "step": 1563 + }, + { + "epoch": 2.005134788189987, + "grad_norm": 0.38894155621528625, + "learning_rate": 2.6145405235370462e-05, + "loss": 2.0151, + "step": 1564 + }, + { + "epoch": 2.006418485237484, + "grad_norm": 0.4423848092556, + "learning_rate": 2.6119390911590565e-05, + "loss": 2.0706, + "step": 1565 + }, + { + "epoch": 2.0077021822849805, + "grad_norm": 0.3701459467411041, + "learning_rate": 2.609337537325112e-05, + "loss": 2.1168, + "step": 1566 + }, + { + "epoch": 2.0089858793324775, + "grad_norm": 0.3845404088497162, + "learning_rate": 2.6067358648579476e-05, + "loss": 2.0377, + "step": 1567 + }, + { + "epoch": 2.0102695763799745, + "grad_norm": 0.3896675109863281, + "learning_rate": 2.604134076580425e-05, + "loss": 2.0837, + "step": 1568 + }, + { + "epoch": 2.011553273427471, + "grad_norm": 0.40908265113830566, + "learning_rate": 2.601532175315532e-05, + "loss": 2.0675, + "step": 1569 + }, + { + "epoch": 2.012836970474968, + "grad_norm": 0.4405415654182434, + "learning_rate": 2.598930163886379e-05, + "loss": 1.9852, + "step": 1570 + }, + { + "epoch": 2.0141206675224645, + "grad_norm": 0.4286010265350342, + "learning_rate": 2.5963280451161946e-05, + "loss": 2.0891, + "step": 1571 + }, + { + "epoch": 2.0154043645699615, + "grad_norm": 0.4322686493396759, + "learning_rate": 2.5937258218283265e-05, + "loss": 2.0404, + "step": 1572 + }, + { + "epoch": 2.016688061617458, + "grad_norm": 0.4295850694179535, + "learning_rate": 2.591123496846234e-05, + "loss": 2.0121, + "step": 1573 + }, + { + "epoch": 2.017971758664955, + "grad_norm": 0.46974098682403564, + "learning_rate": 2.5885210729934872e-05, + "loss": 2.0849, + "step": 1574 + }, + { + "epoch": 2.019255455712452, + "grad_norm": 0.4593369662761688, + "learning_rate": 2.585918553093763e-05, + "loss": 2.0982, + "step": 1575 + }, + { + "epoch": 2.0205391527599486, + "grad_norm": 0.45867305994033813, + "learning_rate": 2.5833159399708427e-05, + "loss": 2.053, + "step": 1576 + }, + { + "epoch": 2.0218228498074455, + "grad_norm": 0.44415611028671265, + "learning_rate": 2.5807132364486097e-05, + "loss": 2.0538, + "step": 1577 + }, + { + "epoch": 2.023106546854942, + "grad_norm": 0.4519406259059906, + "learning_rate": 2.578110445351044e-05, + "loss": 2.0248, + "step": 1578 + }, + { + "epoch": 2.024390243902439, + "grad_norm": 0.4534233808517456, + "learning_rate": 2.5755075695022224e-05, + "loss": 1.9633, + "step": 1579 + }, + { + "epoch": 2.025673940949936, + "grad_norm": 0.4686647355556488, + "learning_rate": 2.5729046117263122e-05, + "loss": 2.0367, + "step": 1580 + }, + { + "epoch": 2.0269576379974326, + "grad_norm": 0.4543870687484741, + "learning_rate": 2.570301574847569e-05, + "loss": 1.9656, + "step": 1581 + }, + { + "epoch": 2.0282413350449295, + "grad_norm": 0.44962936639785767, + "learning_rate": 2.5676984616903367e-05, + "loss": 1.9935, + "step": 1582 + }, + { + "epoch": 2.029525032092426, + "grad_norm": 0.44485536217689514, + "learning_rate": 2.565095275079039e-05, + "loss": 1.991, + "step": 1583 + }, + { + "epoch": 2.030808729139923, + "grad_norm": 0.4554551839828491, + "learning_rate": 2.562492017838183e-05, + "loss": 2.0238, + "step": 1584 + }, + { + "epoch": 2.0320924261874196, + "grad_norm": 0.47761523723602295, + "learning_rate": 2.559888692792348e-05, + "loss": 1.9514, + "step": 1585 + }, + { + "epoch": 2.0333761232349166, + "grad_norm": 0.4294368326663971, + "learning_rate": 2.5572853027661902e-05, + "loss": 2.0058, + "step": 1586 + }, + { + "epoch": 2.0346598202824135, + "grad_norm": 0.4680323302745819, + "learning_rate": 2.5546818505844356e-05, + "loss": 2.046, + "step": 1587 + }, + { + "epoch": 2.03594351732991, + "grad_norm": 0.48906007409095764, + "learning_rate": 2.552078339071876e-05, + "loss": 1.9754, + "step": 1588 + }, + { + "epoch": 2.037227214377407, + "grad_norm": 0.45881739258766174, + "learning_rate": 2.54947477105337e-05, + "loss": 2.0218, + "step": 1589 + }, + { + "epoch": 2.0385109114249036, + "grad_norm": 0.4980340301990509, + "learning_rate": 2.5468711493538362e-05, + "loss": 2.0451, + "step": 1590 + }, + { + "epoch": 2.0397946084724006, + "grad_norm": 0.4902092516422272, + "learning_rate": 2.544267476798251e-05, + "loss": 1.9462, + "step": 1591 + }, + { + "epoch": 2.041078305519897, + "grad_norm": 0.4922800362110138, + "learning_rate": 2.541663756211647e-05, + "loss": 1.9911, + "step": 1592 + }, + { + "epoch": 2.042362002567394, + "grad_norm": 0.4797852039337158, + "learning_rate": 2.5390599904191077e-05, + "loss": 2.0808, + "step": 1593 + }, + { + "epoch": 2.043645699614891, + "grad_norm": 0.5210652351379395, + "learning_rate": 2.5364561822457676e-05, + "loss": 2.0685, + "step": 1594 + }, + { + "epoch": 2.0449293966623876, + "grad_norm": 0.4910551607608795, + "learning_rate": 2.5338523345168048e-05, + "loss": 2.0255, + "step": 1595 + }, + { + "epoch": 2.0462130937098846, + "grad_norm": 0.45405736565589905, + "learning_rate": 2.5312484500574434e-05, + "loss": 2.0551, + "step": 1596 + }, + { + "epoch": 2.047496790757381, + "grad_norm": 0.476178914308548, + "learning_rate": 2.5286445316929434e-05, + "loss": 1.9979, + "step": 1597 + }, + { + "epoch": 2.048780487804878, + "grad_norm": 0.48232659697532654, + "learning_rate": 2.5260405822486044e-05, + "loss": 1.9841, + "step": 1598 + }, + { + "epoch": 2.0500641848523746, + "grad_norm": 0.4537912905216217, + "learning_rate": 2.52343660454976e-05, + "loss": 2.0709, + "step": 1599 + }, + { + "epoch": 2.0513478818998716, + "grad_norm": 0.45053717494010925, + "learning_rate": 2.5208326014217714e-05, + "loss": 1.9786, + "step": 1600 + }, + { + "epoch": 2.0526315789473686, + "grad_norm": 0.47690239548683167, + "learning_rate": 2.5182285756900314e-05, + "loss": 2.0776, + "step": 1601 + }, + { + "epoch": 2.053915275994865, + "grad_norm": 0.4930857717990875, + "learning_rate": 2.5156245301799548e-05, + "loss": 2.034, + "step": 1602 + }, + { + "epoch": 2.055198973042362, + "grad_norm": 0.45152008533477783, + "learning_rate": 2.5130204677169783e-05, + "loss": 1.9762, + "step": 1603 + }, + { + "epoch": 2.0564826700898586, + "grad_norm": 0.45004135370254517, + "learning_rate": 2.510416391126557e-05, + "loss": 2.0055, + "step": 1604 + }, + { + "epoch": 2.0577663671373556, + "grad_norm": 0.4998362958431244, + "learning_rate": 2.507812303234161e-05, + "loss": 1.9726, + "step": 1605 + }, + { + "epoch": 2.0590500641848526, + "grad_norm": 0.52394038438797, + "learning_rate": 2.5052082068652743e-05, + "loss": 1.9919, + "step": 1606 + }, + { + "epoch": 2.060333761232349, + "grad_norm": 0.4381541311740875, + "learning_rate": 2.502604104845388e-05, + "loss": 1.9473, + "step": 1607 + }, + { + "epoch": 2.061617458279846, + "grad_norm": 0.45365461707115173, + "learning_rate": 2.5e-05, + "loss": 1.8656, + "step": 1608 + }, + { + "epoch": 2.0629011553273426, + "grad_norm": 0.46820491552352905, + "learning_rate": 2.4973958951546127e-05, + "loss": 1.9112, + "step": 1609 + }, + { + "epoch": 2.0641848523748396, + "grad_norm": 0.45921388268470764, + "learning_rate": 2.494791793134726e-05, + "loss": 1.9587, + "step": 1610 + }, + { + "epoch": 2.065468549422336, + "grad_norm": 0.4767562747001648, + "learning_rate": 2.4921876967658395e-05, + "loss": 2.0713, + "step": 1611 + }, + { + "epoch": 2.066752246469833, + "grad_norm": 0.4506584405899048, + "learning_rate": 2.4895836088734438e-05, + "loss": 2.0525, + "step": 1612 + }, + { + "epoch": 2.06803594351733, + "grad_norm": 0.4380771815776825, + "learning_rate": 2.486979532283022e-05, + "loss": 2.0481, + "step": 1613 + }, + { + "epoch": 2.0693196405648266, + "grad_norm": 0.49198731780052185, + "learning_rate": 2.4843754698200454e-05, + "loss": 2.0281, + "step": 1614 + }, + { + "epoch": 2.0706033376123236, + "grad_norm": 0.4862654209136963, + "learning_rate": 2.481771424309969e-05, + "loss": 2.0295, + "step": 1615 + }, + { + "epoch": 2.07188703465982, + "grad_norm": 0.46450483798980713, + "learning_rate": 2.4791673985782285e-05, + "loss": 2.039, + "step": 1616 + }, + { + "epoch": 2.073170731707317, + "grad_norm": 0.4759356379508972, + "learning_rate": 2.4765633954502403e-05, + "loss": 2.075, + "step": 1617 + }, + { + "epoch": 2.0744544287548137, + "grad_norm": 0.47094497084617615, + "learning_rate": 2.473959417751396e-05, + "loss": 1.9672, + "step": 1618 + }, + { + "epoch": 2.0757381258023107, + "grad_norm": 0.5017737746238708, + "learning_rate": 2.471355468307057e-05, + "loss": 2.0293, + "step": 1619 + }, + { + "epoch": 2.0770218228498076, + "grad_norm": 0.5029451251029968, + "learning_rate": 2.4687515499425572e-05, + "loss": 1.9398, + "step": 1620 + }, + { + "epoch": 2.078305519897304, + "grad_norm": 0.5085014700889587, + "learning_rate": 2.4661476654831958e-05, + "loss": 2.0163, + "step": 1621 + }, + { + "epoch": 2.079589216944801, + "grad_norm": 0.4704389274120331, + "learning_rate": 2.4635438177542336e-05, + "loss": 2.086, + "step": 1622 + }, + { + "epoch": 2.0808729139922977, + "grad_norm": 0.48211780190467834, + "learning_rate": 2.4609400095808925e-05, + "loss": 1.9601, + "step": 1623 + }, + { + "epoch": 2.0821566110397947, + "grad_norm": 0.45585399866104126, + "learning_rate": 2.458336243788354e-05, + "loss": 1.9935, + "step": 1624 + }, + { + "epoch": 2.083440308087291, + "grad_norm": 0.4696653485298157, + "learning_rate": 2.4557325232017504e-05, + "loss": 1.984, + "step": 1625 + }, + { + "epoch": 2.084724005134788, + "grad_norm": 0.4823111891746521, + "learning_rate": 2.4531288506461644e-05, + "loss": 1.9964, + "step": 1626 + }, + { + "epoch": 2.086007702182285, + "grad_norm": 0.5095885396003723, + "learning_rate": 2.4505252289466303e-05, + "loss": 1.939, + "step": 1627 + }, + { + "epoch": 2.0872913992297817, + "grad_norm": 0.5270523428916931, + "learning_rate": 2.4479216609281246e-05, + "loss": 2.0064, + "step": 1628 + }, + { + "epoch": 2.0885750962772787, + "grad_norm": 0.4640248417854309, + "learning_rate": 2.4453181494155643e-05, + "loss": 2.0692, + "step": 1629 + }, + { + "epoch": 2.089858793324775, + "grad_norm": 0.5134516358375549, + "learning_rate": 2.44271469723381e-05, + "loss": 2.1627, + "step": 1630 + }, + { + "epoch": 2.091142490372272, + "grad_norm": 0.4603593349456787, + "learning_rate": 2.4401113072076525e-05, + "loss": 2.0226, + "step": 1631 + }, + { + "epoch": 2.092426187419769, + "grad_norm": 0.45982351899147034, + "learning_rate": 2.4375079821618173e-05, + "loss": 1.9921, + "step": 1632 + }, + { + "epoch": 2.0937098844672657, + "grad_norm": 0.5020879507064819, + "learning_rate": 2.4349047249209614e-05, + "loss": 1.9718, + "step": 1633 + }, + { + "epoch": 2.0949935815147627, + "grad_norm": 0.502094566822052, + "learning_rate": 2.4323015383096643e-05, + "loss": 2.1025, + "step": 1634 + }, + { + "epoch": 2.096277278562259, + "grad_norm": 0.49221521615982056, + "learning_rate": 2.4296984251524314e-05, + "loss": 2.016, + "step": 1635 + }, + { + "epoch": 2.097560975609756, + "grad_norm": 0.47577252984046936, + "learning_rate": 2.4270953882736887e-05, + "loss": 2.0081, + "step": 1636 + }, + { + "epoch": 2.0988446726572527, + "grad_norm": 0.502552330493927, + "learning_rate": 2.4244924304977785e-05, + "loss": 1.9468, + "step": 1637 + }, + { + "epoch": 2.1001283697047497, + "grad_norm": 0.5236652493476868, + "learning_rate": 2.4218895546489563e-05, + "loss": 2.0541, + "step": 1638 + }, + { + "epoch": 2.1014120667522467, + "grad_norm": 0.5130388736724854, + "learning_rate": 2.419286763551391e-05, + "loss": 1.9744, + "step": 1639 + }, + { + "epoch": 2.102695763799743, + "grad_norm": 0.5128998756408691, + "learning_rate": 2.4166840600291585e-05, + "loss": 2.012, + "step": 1640 + }, + { + "epoch": 2.10397946084724, + "grad_norm": 0.5037839412689209, + "learning_rate": 2.414081446906238e-05, + "loss": 2.0925, + "step": 1641 + }, + { + "epoch": 2.1052631578947367, + "grad_norm": 0.4723465144634247, + "learning_rate": 2.4114789270065137e-05, + "loss": 2.0456, + "step": 1642 + }, + { + "epoch": 2.1065468549422337, + "grad_norm": 0.4586403965950012, + "learning_rate": 2.4088765031537662e-05, + "loss": 1.9783, + "step": 1643 + }, + { + "epoch": 2.1078305519897302, + "grad_norm": 0.4795694649219513, + "learning_rate": 2.4062741781716734e-05, + "loss": 2.0377, + "step": 1644 + }, + { + "epoch": 2.109114249037227, + "grad_norm": 0.489459753036499, + "learning_rate": 2.4036719548838056e-05, + "loss": 2.0133, + "step": 1645 + }, + { + "epoch": 2.110397946084724, + "grad_norm": 0.44902732968330383, + "learning_rate": 2.401069836113622e-05, + "loss": 2.0021, + "step": 1646 + }, + { + "epoch": 2.1116816431322207, + "grad_norm": 0.48210087418556213, + "learning_rate": 2.3984678246844677e-05, + "loss": 2.0542, + "step": 1647 + }, + { + "epoch": 2.1129653401797177, + "grad_norm": 0.470652312040329, + "learning_rate": 2.3958659234195753e-05, + "loss": 1.9775, + "step": 1648 + }, + { + "epoch": 2.1142490372272142, + "grad_norm": 0.48913174867630005, + "learning_rate": 2.393264135142053e-05, + "loss": 2.0404, + "step": 1649 + }, + { + "epoch": 2.1155327342747112, + "grad_norm": 0.49382442235946655, + "learning_rate": 2.3906624626748885e-05, + "loss": 1.9686, + "step": 1650 + }, + { + "epoch": 2.1168164313222078, + "grad_norm": 0.4890129566192627, + "learning_rate": 2.3880609088409438e-05, + "loss": 2.094, + "step": 1651 + }, + { + "epoch": 2.1181001283697047, + "grad_norm": 0.5227566957473755, + "learning_rate": 2.3854594764629544e-05, + "loss": 2.0911, + "step": 1652 + }, + { + "epoch": 2.1193838254172017, + "grad_norm": 0.5298448204994202, + "learning_rate": 2.382858168363519e-05, + "loss": 2.0351, + "step": 1653 + }, + { + "epoch": 2.1206675224646983, + "grad_norm": 0.5318330526351929, + "learning_rate": 2.3802569873651052e-05, + "loss": 2.0356, + "step": 1654 + }, + { + "epoch": 2.1219512195121952, + "grad_norm": 0.5014201402664185, + "learning_rate": 2.377655936290043e-05, + "loss": 2.0482, + "step": 1655 + }, + { + "epoch": 2.1232349165596918, + "grad_norm": 0.5246758460998535, + "learning_rate": 2.3750550179605187e-05, + "loss": 1.9206, + "step": 1656 + }, + { + "epoch": 2.1245186136071887, + "grad_norm": 0.5040850639343262, + "learning_rate": 2.3724542351985762e-05, + "loss": 1.9983, + "step": 1657 + }, + { + "epoch": 2.1258023106546853, + "grad_norm": 0.4852854311466217, + "learning_rate": 2.369853590826113e-05, + "loss": 2.0587, + "step": 1658 + }, + { + "epoch": 2.1270860077021823, + "grad_norm": 0.5180385708808899, + "learning_rate": 2.367253087664876e-05, + "loss": 1.9814, + "step": 1659 + }, + { + "epoch": 2.1283697047496792, + "grad_norm": 0.5086965560913086, + "learning_rate": 2.3646527285364565e-05, + "loss": 2.053, + "step": 1660 + }, + { + "epoch": 2.1296534017971758, + "grad_norm": 0.5366697311401367, + "learning_rate": 2.3620525162622933e-05, + "loss": 1.9883, + "step": 1661 + }, + { + "epoch": 2.1309370988446728, + "grad_norm": 0.48167648911476135, + "learning_rate": 2.359452453663664e-05, + "loss": 2.0622, + "step": 1662 + }, + { + "epoch": 2.1322207958921693, + "grad_norm": 0.476648211479187, + "learning_rate": 2.356852543561682e-05, + "loss": 1.9319, + "step": 1663 + }, + { + "epoch": 2.1335044929396663, + "grad_norm": 0.49231573939323425, + "learning_rate": 2.3542527887773003e-05, + "loss": 2.0435, + "step": 1664 + }, + { + "epoch": 2.1347881899871632, + "grad_norm": 0.4906701445579529, + "learning_rate": 2.351653192131298e-05, + "loss": 2.0964, + "step": 1665 + }, + { + "epoch": 2.1360718870346598, + "grad_norm": 0.5811977386474609, + "learning_rate": 2.3490537564442847e-05, + "loss": 2.0316, + "step": 1666 + }, + { + "epoch": 2.1373555840821568, + "grad_norm": 0.5391324758529663, + "learning_rate": 2.346454484536697e-05, + "loss": 1.9918, + "step": 1667 + }, + { + "epoch": 2.1386392811296533, + "grad_norm": 0.4729015529155731, + "learning_rate": 2.3438553792287915e-05, + "loss": 1.9588, + "step": 1668 + }, + { + "epoch": 2.1399229781771503, + "grad_norm": 0.48231133818626404, + "learning_rate": 2.341256443340644e-05, + "loss": 1.968, + "step": 1669 + }, + { + "epoch": 2.141206675224647, + "grad_norm": 0.503819465637207, + "learning_rate": 2.3386576796921484e-05, + "loss": 1.9917, + "step": 1670 + }, + { + "epoch": 2.142490372272144, + "grad_norm": 0.4668719172477722, + "learning_rate": 2.336059091103011e-05, + "loss": 1.9656, + "step": 1671 + }, + { + "epoch": 2.1437740693196408, + "grad_norm": 0.5099239349365234, + "learning_rate": 2.3334606803927456e-05, + "loss": 2.0223, + "step": 1672 + }, + { + "epoch": 2.1450577663671373, + "grad_norm": 0.4753022789955139, + "learning_rate": 2.3308624503806772e-05, + "loss": 2.0229, + "step": 1673 + }, + { + "epoch": 2.1463414634146343, + "grad_norm": 0.463164359331131, + "learning_rate": 2.328264403885933e-05, + "loss": 2.0003, + "step": 1674 + }, + { + "epoch": 2.147625160462131, + "grad_norm": 0.4959816336631775, + "learning_rate": 2.325666543727438e-05, + "loss": 2.059, + "step": 1675 + }, + { + "epoch": 2.148908857509628, + "grad_norm": 0.5662737488746643, + "learning_rate": 2.3230688727239214e-05, + "loss": 2.0445, + "step": 1676 + }, + { + "epoch": 2.1501925545571243, + "grad_norm": 0.4857902526855469, + "learning_rate": 2.3204713936939014e-05, + "loss": 1.8938, + "step": 1677 + }, + { + "epoch": 2.1514762516046213, + "grad_norm": 0.5183926224708557, + "learning_rate": 2.3178741094556908e-05, + "loss": 1.944, + "step": 1678 + }, + { + "epoch": 2.1527599486521183, + "grad_norm": 0.5066496133804321, + "learning_rate": 2.315277022827389e-05, + "loss": 2.0373, + "step": 1679 + }, + { + "epoch": 2.154043645699615, + "grad_norm": 0.5094098448753357, + "learning_rate": 2.312680136626884e-05, + "loss": 2.0249, + "step": 1680 + }, + { + "epoch": 2.155327342747112, + "grad_norm": 0.5131878852844238, + "learning_rate": 2.310083453671845e-05, + "loss": 2.0081, + "step": 1681 + }, + { + "epoch": 2.1566110397946083, + "grad_norm": 0.5087143182754517, + "learning_rate": 2.3074869767797172e-05, + "loss": 2.0402, + "step": 1682 + }, + { + "epoch": 2.1578947368421053, + "grad_norm": 0.5391287803649902, + "learning_rate": 2.3048907087677297e-05, + "loss": 2.0963, + "step": 1683 + }, + { + "epoch": 2.1591784338896023, + "grad_norm": 0.5140363574028015, + "learning_rate": 2.302294652452877e-05, + "loss": 1.9602, + "step": 1684 + }, + { + "epoch": 2.160462130937099, + "grad_norm": 0.5172721743583679, + "learning_rate": 2.299698810651929e-05, + "loss": 2.0234, + "step": 1685 + }, + { + "epoch": 2.161745827984596, + "grad_norm": 0.5107318162918091, + "learning_rate": 2.2971031861814223e-05, + "loss": 2.1297, + "step": 1686 + }, + { + "epoch": 2.1630295250320923, + "grad_norm": 0.4851250648498535, + "learning_rate": 2.294507781857655e-05, + "loss": 2.0589, + "step": 1687 + }, + { + "epoch": 2.1643132220795893, + "grad_norm": 0.5335860252380371, + "learning_rate": 2.2919126004966886e-05, + "loss": 2.1224, + "step": 1688 + }, + { + "epoch": 2.165596919127086, + "grad_norm": 0.5164303779602051, + "learning_rate": 2.2893176449143422e-05, + "loss": 1.9275, + "step": 1689 + }, + { + "epoch": 2.166880616174583, + "grad_norm": 0.48727819323539734, + "learning_rate": 2.2867229179261912e-05, + "loss": 2.05, + "step": 1690 + }, + { + "epoch": 2.1681643132220794, + "grad_norm": 0.48038730025291443, + "learning_rate": 2.2841284223475593e-05, + "loss": 2.0302, + "step": 1691 + }, + { + "epoch": 2.1694480102695763, + "grad_norm": 0.4774797260761261, + "learning_rate": 2.2815341609935232e-05, + "loss": 1.9629, + "step": 1692 + }, + { + "epoch": 2.1707317073170733, + "grad_norm": 0.48471665382385254, + "learning_rate": 2.2789401366789036e-05, + "loss": 1.9467, + "step": 1693 + }, + { + "epoch": 2.17201540436457, + "grad_norm": 0.5234472155570984, + "learning_rate": 2.2763463522182627e-05, + "loss": 2.0169, + "step": 1694 + }, + { + "epoch": 2.173299101412067, + "grad_norm": 0.48092490434646606, + "learning_rate": 2.2737528104259056e-05, + "loss": 1.9398, + "step": 1695 + }, + { + "epoch": 2.1745827984595634, + "grad_norm": 0.5008983612060547, + "learning_rate": 2.2711595141158716e-05, + "loss": 2.0709, + "step": 1696 + }, + { + "epoch": 2.1758664955070603, + "grad_norm": 0.4879731833934784, + "learning_rate": 2.2685664661019345e-05, + "loss": 2.0573, + "step": 1697 + }, + { + "epoch": 2.1771501925545573, + "grad_norm": 0.48747342824935913, + "learning_rate": 2.2659736691975998e-05, + "loss": 1.9135, + "step": 1698 + }, + { + "epoch": 2.178433889602054, + "grad_norm": 0.5379552245140076, + "learning_rate": 2.2633811262160977e-05, + "loss": 1.9555, + "step": 1699 + }, + { + "epoch": 2.179717586649551, + "grad_norm": 0.4753062427043915, + "learning_rate": 2.2607888399703845e-05, + "loss": 1.9494, + "step": 1700 + }, + { + "epoch": 2.1810012836970474, + "grad_norm": 0.5169556140899658, + "learning_rate": 2.2581968132731402e-05, + "loss": 2.0, + "step": 1701 + }, + { + "epoch": 2.1822849807445444, + "grad_norm": 0.4900209605693817, + "learning_rate": 2.2556050489367585e-05, + "loss": 2.0801, + "step": 1702 + }, + { + "epoch": 2.183568677792041, + "grad_norm": 0.5023864507675171, + "learning_rate": 2.2530135497733526e-05, + "loss": 1.9857, + "step": 1703 + }, + { + "epoch": 2.184852374839538, + "grad_norm": 0.4637115001678467, + "learning_rate": 2.250422318594745e-05, + "loss": 1.9309, + "step": 1704 + }, + { + "epoch": 2.186136071887035, + "grad_norm": 0.4907744824886322, + "learning_rate": 2.24783135821247e-05, + "loss": 1.8945, + "step": 1705 + }, + { + "epoch": 2.1874197689345314, + "grad_norm": 0.5133538246154785, + "learning_rate": 2.2452406714377662e-05, + "loss": 2.0137, + "step": 1706 + }, + { + "epoch": 2.1887034659820284, + "grad_norm": 0.5007182955741882, + "learning_rate": 2.242650261081575e-05, + "loss": 1.9977, + "step": 1707 + }, + { + "epoch": 2.189987163029525, + "grad_norm": 0.5040572285652161, + "learning_rate": 2.240060129954542e-05, + "loss": 1.9729, + "step": 1708 + }, + { + "epoch": 2.191270860077022, + "grad_norm": 0.5083310604095459, + "learning_rate": 2.2374702808670032e-05, + "loss": 1.9666, + "step": 1709 + }, + { + "epoch": 2.1925545571245184, + "grad_norm": 0.5064545273780823, + "learning_rate": 2.234880716628993e-05, + "loss": 2.0124, + "step": 1710 + }, + { + "epoch": 2.1938382541720154, + "grad_norm": 0.5112729072570801, + "learning_rate": 2.2322914400502366e-05, + "loss": 2.0256, + "step": 1711 + }, + { + "epoch": 2.1951219512195124, + "grad_norm": 0.48376962542533875, + "learning_rate": 2.2297024539401463e-05, + "loss": 2.018, + "step": 1712 + }, + { + "epoch": 2.196405648267009, + "grad_norm": 0.5419514179229736, + "learning_rate": 2.2271137611078176e-05, + "loss": 1.9736, + "step": 1713 + }, + { + "epoch": 2.197689345314506, + "grad_norm": 0.49089673161506653, + "learning_rate": 2.2245253643620316e-05, + "loss": 1.9843, + "step": 1714 + }, + { + "epoch": 2.1989730423620024, + "grad_norm": 0.4840286374092102, + "learning_rate": 2.221937266511245e-05, + "loss": 1.9305, + "step": 1715 + }, + { + "epoch": 2.2002567394094994, + "grad_norm": 0.4894367754459381, + "learning_rate": 2.2193494703635894e-05, + "loss": 2.0369, + "step": 1716 + }, + { + "epoch": 2.2015404364569964, + "grad_norm": 0.5115372538566589, + "learning_rate": 2.2167619787268738e-05, + "loss": 2.0096, + "step": 1717 + }, + { + "epoch": 2.202824133504493, + "grad_norm": 0.4937199056148529, + "learning_rate": 2.2141747944085722e-05, + "loss": 1.9587, + "step": 1718 + }, + { + "epoch": 2.20410783055199, + "grad_norm": 0.5416276454925537, + "learning_rate": 2.2115879202158262e-05, + "loss": 2.106, + "step": 1719 + }, + { + "epoch": 2.2053915275994864, + "grad_norm": 0.513535737991333, + "learning_rate": 2.2090013589554433e-05, + "loss": 1.9783, + "step": 1720 + }, + { + "epoch": 2.2066752246469834, + "grad_norm": 0.4994733929634094, + "learning_rate": 2.2064151134338877e-05, + "loss": 2.041, + "step": 1721 + }, + { + "epoch": 2.20795892169448, + "grad_norm": 0.5248288512229919, + "learning_rate": 2.203829186457283e-05, + "loss": 2.0543, + "step": 1722 + }, + { + "epoch": 2.209242618741977, + "grad_norm": 0.5572697520256042, + "learning_rate": 2.2012435808314084e-05, + "loss": 2.0749, + "step": 1723 + }, + { + "epoch": 2.2105263157894735, + "grad_norm": 0.5091756582260132, + "learning_rate": 2.1986582993616926e-05, + "loss": 2.0052, + "step": 1724 + }, + { + "epoch": 2.2118100128369704, + "grad_norm": 0.5291957855224609, + "learning_rate": 2.1960733448532126e-05, + "loss": 2.0403, + "step": 1725 + }, + { + "epoch": 2.2130937098844674, + "grad_norm": 0.482880175113678, + "learning_rate": 2.1934887201106916e-05, + "loss": 2.0012, + "step": 1726 + }, + { + "epoch": 2.214377406931964, + "grad_norm": 0.5281679034233093, + "learning_rate": 2.1909044279384948e-05, + "loss": 2.0411, + "step": 1727 + }, + { + "epoch": 2.215661103979461, + "grad_norm": 0.5210844874382019, + "learning_rate": 2.188320471140625e-05, + "loss": 1.9899, + "step": 1728 + }, + { + "epoch": 2.2169448010269575, + "grad_norm": 0.5443101525306702, + "learning_rate": 2.185736852520724e-05, + "loss": 1.9909, + "step": 1729 + }, + { + "epoch": 2.2182284980744544, + "grad_norm": 0.5066094994544983, + "learning_rate": 2.183153574882064e-05, + "loss": 1.9177, + "step": 1730 + }, + { + "epoch": 2.2195121951219514, + "grad_norm": 0.49774378538131714, + "learning_rate": 2.1805706410275495e-05, + "loss": 1.9658, + "step": 1731 + }, + { + "epoch": 2.220795892169448, + "grad_norm": 0.5080484747886658, + "learning_rate": 2.177988053759709e-05, + "loss": 2.0634, + "step": 1732 + }, + { + "epoch": 2.222079589216945, + "grad_norm": 0.5077035427093506, + "learning_rate": 2.1754058158806983e-05, + "loss": 1.9837, + "step": 1733 + }, + { + "epoch": 2.2233632862644415, + "grad_norm": 0.5014065504074097, + "learning_rate": 2.1728239301922925e-05, + "loss": 1.9312, + "step": 1734 + }, + { + "epoch": 2.2246469833119384, + "grad_norm": 0.5621789693832397, + "learning_rate": 2.1702423994958832e-05, + "loss": 2.0159, + "step": 1735 + }, + { + "epoch": 2.225930680359435, + "grad_norm": 0.5026384592056274, + "learning_rate": 2.1676612265924796e-05, + "loss": 2.0367, + "step": 1736 + }, + { + "epoch": 2.227214377406932, + "grad_norm": 0.5023465752601624, + "learning_rate": 2.1650804142827004e-05, + "loss": 1.9822, + "step": 1737 + }, + { + "epoch": 2.228498074454429, + "grad_norm": 0.526077926158905, + "learning_rate": 2.1624999653667747e-05, + "loss": 1.9996, + "step": 1738 + }, + { + "epoch": 2.2297817715019255, + "grad_norm": 0.5047421455383301, + "learning_rate": 2.159919882644537e-05, + "loss": 2.0887, + "step": 1739 + }, + { + "epoch": 2.2310654685494224, + "grad_norm": 0.542472779750824, + "learning_rate": 2.157340168915423e-05, + "loss": 2.0297, + "step": 1740 + }, + { + "epoch": 2.232349165596919, + "grad_norm": 0.5084927082061768, + "learning_rate": 2.1547608269784692e-05, + "loss": 1.9204, + "step": 1741 + }, + { + "epoch": 2.233632862644416, + "grad_norm": 0.5399092435836792, + "learning_rate": 2.1521818596323098e-05, + "loss": 2.0253, + "step": 1742 + }, + { + "epoch": 2.2349165596919125, + "grad_norm": 0.5358842015266418, + "learning_rate": 2.1496032696751704e-05, + "loss": 1.9783, + "step": 1743 + }, + { + "epoch": 2.2362002567394095, + "grad_norm": 0.5135545134544373, + "learning_rate": 2.1470250599048673e-05, + "loss": 1.9921, + "step": 1744 + }, + { + "epoch": 2.2374839537869065, + "grad_norm": 0.5236623287200928, + "learning_rate": 2.1444472331188064e-05, + "loss": 2.0247, + "step": 1745 + }, + { + "epoch": 2.238767650834403, + "grad_norm": 0.4750460088253021, + "learning_rate": 2.141869792113977e-05, + "loss": 2.0433, + "step": 1746 + }, + { + "epoch": 2.2400513478819, + "grad_norm": 0.4933171570301056, + "learning_rate": 2.1392927396869478e-05, + "loss": 2.0674, + "step": 1747 + }, + { + "epoch": 2.2413350449293965, + "grad_norm": 0.48384562134742737, + "learning_rate": 2.1367160786338695e-05, + "loss": 1.9916, + "step": 1748 + }, + { + "epoch": 2.2426187419768935, + "grad_norm": 0.5263171195983887, + "learning_rate": 2.134139811750466e-05, + "loss": 1.9536, + "step": 1749 + }, + { + "epoch": 2.2439024390243905, + "grad_norm": 0.5724929571151733, + "learning_rate": 2.1315639418320312e-05, + "loss": 1.9935, + "step": 1750 + }, + { + "epoch": 2.245186136071887, + "grad_norm": 0.5341793894767761, + "learning_rate": 2.1289884716734347e-05, + "loss": 2.0111, + "step": 1751 + }, + { + "epoch": 2.246469833119384, + "grad_norm": 0.49942222237586975, + "learning_rate": 2.1264134040691068e-05, + "loss": 2.0493, + "step": 1752 + }, + { + "epoch": 2.2477535301668805, + "grad_norm": 0.520527720451355, + "learning_rate": 2.1238387418130422e-05, + "loss": 1.9562, + "step": 1753 + }, + { + "epoch": 2.2490372272143775, + "grad_norm": 0.5147912502288818, + "learning_rate": 2.121264487698798e-05, + "loss": 2.049, + "step": 1754 + }, + { + "epoch": 2.250320924261874, + "grad_norm": 0.5141321420669556, + "learning_rate": 2.1186906445194854e-05, + "loss": 2.0647, + "step": 1755 + }, + { + "epoch": 2.251604621309371, + "grad_norm": 0.5374550819396973, + "learning_rate": 2.116117215067771e-05, + "loss": 2.0508, + "step": 1756 + }, + { + "epoch": 2.2528883183568675, + "grad_norm": 0.5167858600616455, + "learning_rate": 2.1135442021358738e-05, + "loss": 1.9747, + "step": 1757 + }, + { + "epoch": 2.2541720154043645, + "grad_norm": 0.4966643750667572, + "learning_rate": 2.1109716085155596e-05, + "loss": 2.0204, + "step": 1758 + }, + { + "epoch": 2.2554557124518615, + "grad_norm": 0.5275843143463135, + "learning_rate": 2.108399436998138e-05, + "loss": 2.0187, + "step": 1759 + }, + { + "epoch": 2.256739409499358, + "grad_norm": 0.5655891299247742, + "learning_rate": 2.1058276903744628e-05, + "loss": 2.0245, + "step": 1760 + }, + { + "epoch": 2.258023106546855, + "grad_norm": 0.5229381322860718, + "learning_rate": 2.103256371434927e-05, + "loss": 2.0165, + "step": 1761 + }, + { + "epoch": 2.2593068035943515, + "grad_norm": 0.5655928254127502, + "learning_rate": 2.1006854829694567e-05, + "loss": 1.9818, + "step": 1762 + }, + { + "epoch": 2.2605905006418485, + "grad_norm": 0.5208088159561157, + "learning_rate": 2.0981150277675132e-05, + "loss": 2.0291, + "step": 1763 + }, + { + "epoch": 2.2618741976893455, + "grad_norm": 0.4928179383277893, + "learning_rate": 2.0955450086180882e-05, + "loss": 2.0294, + "step": 1764 + }, + { + "epoch": 2.263157894736842, + "grad_norm": 0.5188254714012146, + "learning_rate": 2.0929754283096993e-05, + "loss": 2.0114, + "step": 1765 + }, + { + "epoch": 2.264441591784339, + "grad_norm": 0.5012592673301697, + "learning_rate": 2.0904062896303862e-05, + "loss": 1.9802, + "step": 1766 + }, + { + "epoch": 2.2657252888318355, + "grad_norm": 0.5161022543907166, + "learning_rate": 2.0878375953677128e-05, + "loss": 1.9197, + "step": 1767 + }, + { + "epoch": 2.2670089858793325, + "grad_norm": 0.5280621647834778, + "learning_rate": 2.085269348308759e-05, + "loss": 2.0209, + "step": 1768 + }, + { + "epoch": 2.2682926829268295, + "grad_norm": 0.5979487299919128, + "learning_rate": 2.0827015512401184e-05, + "loss": 2.0103, + "step": 1769 + }, + { + "epoch": 2.269576379974326, + "grad_norm": 0.4997188150882721, + "learning_rate": 2.0801342069478994e-05, + "loss": 2.0146, + "step": 1770 + }, + { + "epoch": 2.270860077021823, + "grad_norm": 0.4900396168231964, + "learning_rate": 2.0775673182177164e-05, + "loss": 1.9733, + "step": 1771 + }, + { + "epoch": 2.2721437740693196, + "grad_norm": 0.4797995686531067, + "learning_rate": 2.07500088783469e-05, + "loss": 2.0056, + "step": 1772 + }, + { + "epoch": 2.2734274711168165, + "grad_norm": 0.5041288733482361, + "learning_rate": 2.0724349185834453e-05, + "loss": 1.9609, + "step": 1773 + }, + { + "epoch": 2.274711168164313, + "grad_norm": 0.5210070610046387, + "learning_rate": 2.0698694132481044e-05, + "loss": 1.9386, + "step": 1774 + }, + { + "epoch": 2.27599486521181, + "grad_norm": 0.4784318506717682, + "learning_rate": 2.0673043746122874e-05, + "loss": 2.003, + "step": 1775 + }, + { + "epoch": 2.2772785622593066, + "grad_norm": 0.4813714027404785, + "learning_rate": 2.0647398054591093e-05, + "loss": 1.9486, + "step": 1776 + }, + { + "epoch": 2.2785622593068036, + "grad_norm": 0.4725954532623291, + "learning_rate": 2.0621757085711734e-05, + "loss": 2.0938, + "step": 1777 + }, + { + "epoch": 2.2798459563543005, + "grad_norm": 0.5049464106559753, + "learning_rate": 2.0596120867305704e-05, + "loss": 1.8625, + "step": 1778 + }, + { + "epoch": 2.281129653401797, + "grad_norm": 0.5073199272155762, + "learning_rate": 2.0570489427188787e-05, + "loss": 1.9752, + "step": 1779 + }, + { + "epoch": 2.282413350449294, + "grad_norm": 0.4938678741455078, + "learning_rate": 2.054486279317156e-05, + "loss": 2.089, + "step": 1780 + }, + { + "epoch": 2.2836970474967906, + "grad_norm": 0.5024643540382385, + "learning_rate": 2.0519240993059372e-05, + "loss": 1.9752, + "step": 1781 + }, + { + "epoch": 2.2849807445442876, + "grad_norm": 0.4869168996810913, + "learning_rate": 2.0493624054652357e-05, + "loss": 1.9786, + "step": 1782 + }, + { + "epoch": 2.2862644415917845, + "grad_norm": 0.5110926032066345, + "learning_rate": 2.0468012005745364e-05, + "loss": 2.041, + "step": 1783 + }, + { + "epoch": 2.287548138639281, + "grad_norm": 0.520343005657196, + "learning_rate": 2.044240487412792e-05, + "loss": 1.9483, + "step": 1784 + }, + { + "epoch": 2.288831835686778, + "grad_norm": 0.5444603562355042, + "learning_rate": 2.0416802687584226e-05, + "loss": 2.0078, + "step": 1785 + }, + { + "epoch": 2.2901155327342746, + "grad_norm": 0.5269668698310852, + "learning_rate": 2.0391205473893134e-05, + "loss": 2.02, + "step": 1786 + }, + { + "epoch": 2.2913992297817716, + "grad_norm": 0.5185669660568237, + "learning_rate": 2.036561326082809e-05, + "loss": 2.0973, + "step": 1787 + }, + { + "epoch": 2.292682926829268, + "grad_norm": 0.5337273478507996, + "learning_rate": 2.0340026076157094e-05, + "loss": 2.0346, + "step": 1788 + }, + { + "epoch": 2.293966623876765, + "grad_norm": 0.5509817004203796, + "learning_rate": 2.0314443947642722e-05, + "loss": 2.0896, + "step": 1789 + }, + { + "epoch": 2.295250320924262, + "grad_norm": 0.5068468451499939, + "learning_rate": 2.0288866903042054e-05, + "loss": 1.957, + "step": 1790 + }, + { + "epoch": 2.2965340179717586, + "grad_norm": 0.5349813103675842, + "learning_rate": 2.026329497010662e-05, + "loss": 2.011, + "step": 1791 + }, + { + "epoch": 2.2978177150192556, + "grad_norm": 0.48997631669044495, + "learning_rate": 2.023772817658247e-05, + "loss": 1.955, + "step": 1792 + }, + { + "epoch": 2.299101412066752, + "grad_norm": 0.4977245032787323, + "learning_rate": 2.021216655021002e-05, + "loss": 2.0107, + "step": 1793 + }, + { + "epoch": 2.300385109114249, + "grad_norm": 0.5159211158752441, + "learning_rate": 2.0186610118724104e-05, + "loss": 2.0923, + "step": 1794 + }, + { + "epoch": 2.3016688061617456, + "grad_norm": 0.5332282781600952, + "learning_rate": 2.0161058909853926e-05, + "loss": 2.1356, + "step": 1795 + }, + { + "epoch": 2.3029525032092426, + "grad_norm": 0.4944048821926117, + "learning_rate": 2.0135512951322995e-05, + "loss": 1.9775, + "step": 1796 + }, + { + "epoch": 2.3042362002567396, + "grad_norm": 0.5297811031341553, + "learning_rate": 2.0109972270849148e-05, + "loss": 1.985, + "step": 1797 + }, + { + "epoch": 2.305519897304236, + "grad_norm": 0.4751434922218323, + "learning_rate": 2.00844368961445e-05, + "loss": 2.0412, + "step": 1798 + }, + { + "epoch": 2.306803594351733, + "grad_norm": 0.5183787941932678, + "learning_rate": 2.005890685491539e-05, + "loss": 2.0507, + "step": 1799 + }, + { + "epoch": 2.3080872913992296, + "grad_norm": 0.5471688508987427, + "learning_rate": 2.003338217486237e-05, + "loss": 2.068, + "step": 1800 + }, + { + "epoch": 2.3093709884467266, + "grad_norm": 0.4964779317378998, + "learning_rate": 2.000786288368019e-05, + "loss": 2.037, + "step": 1801 + }, + { + "epoch": 2.3106546854942236, + "grad_norm": 0.5133666396141052, + "learning_rate": 1.9982349009057758e-05, + "loss": 2.0383, + "step": 1802 + }, + { + "epoch": 2.31193838254172, + "grad_norm": 0.4889377951622009, + "learning_rate": 1.995684057867806e-05, + "loss": 2.0176, + "step": 1803 + }, + { + "epoch": 2.313222079589217, + "grad_norm": 0.4990570843219757, + "learning_rate": 1.993133762021825e-05, + "loss": 2.0604, + "step": 1804 + }, + { + "epoch": 2.3145057766367136, + "grad_norm": 0.5061864852905273, + "learning_rate": 1.990584016134947e-05, + "loss": 2.0532, + "step": 1805 + }, + { + "epoch": 2.3157894736842106, + "grad_norm": 0.5334873795509338, + "learning_rate": 1.988034822973694e-05, + "loss": 2.0449, + "step": 1806 + }, + { + "epoch": 2.317073170731707, + "grad_norm": 0.551032304763794, + "learning_rate": 1.9854861853039873e-05, + "loss": 1.9786, + "step": 1807 + }, + { + "epoch": 2.318356867779204, + "grad_norm": 0.5055244565010071, + "learning_rate": 1.9829381058911448e-05, + "loss": 2.0308, + "step": 1808 + }, + { + "epoch": 2.3196405648267007, + "grad_norm": 0.5158600211143494, + "learning_rate": 1.9803905874998786e-05, + "loss": 1.9645, + "step": 1809 + }, + { + "epoch": 2.3209242618741976, + "grad_norm": 0.5720779299736023, + "learning_rate": 1.9778436328942933e-05, + "loss": 2.0479, + "step": 1810 + }, + { + "epoch": 2.3222079589216946, + "grad_norm": 0.540172815322876, + "learning_rate": 1.9752972448378814e-05, + "loss": 1.9534, + "step": 1811 + }, + { + "epoch": 2.323491655969191, + "grad_norm": 0.5196032524108887, + "learning_rate": 1.9727514260935196e-05, + "loss": 2.0752, + "step": 1812 + }, + { + "epoch": 2.324775353016688, + "grad_norm": 0.523025393486023, + "learning_rate": 1.9702061794234672e-05, + "loss": 2.0729, + "step": 1813 + }, + { + "epoch": 2.3260590500641847, + "grad_norm": 0.5281288027763367, + "learning_rate": 1.9676615075893646e-05, + "loss": 2.1114, + "step": 1814 + }, + { + "epoch": 2.3273427471116817, + "grad_norm": 0.542975127696991, + "learning_rate": 1.965117413352226e-05, + "loss": 1.9454, + "step": 1815 + }, + { + "epoch": 2.3286264441591786, + "grad_norm": 0.5025236010551453, + "learning_rate": 1.96257389947244e-05, + "loss": 2.0211, + "step": 1816 + }, + { + "epoch": 2.329910141206675, + "grad_norm": 0.4845566749572754, + "learning_rate": 1.960030968709766e-05, + "loss": 2.0555, + "step": 1817 + }, + { + "epoch": 2.331193838254172, + "grad_norm": 0.4925599992275238, + "learning_rate": 1.9574886238233308e-05, + "loss": 2.0318, + "step": 1818 + }, + { + "epoch": 2.3324775353016687, + "grad_norm": 0.5643622875213623, + "learning_rate": 1.9549468675716233e-05, + "loss": 1.9804, + "step": 1819 + }, + { + "epoch": 2.3337612323491657, + "grad_norm": 0.5365644693374634, + "learning_rate": 1.9524057027124964e-05, + "loss": 2.0194, + "step": 1820 + }, + { + "epoch": 2.335044929396662, + "grad_norm": 0.5228244066238403, + "learning_rate": 1.9498651320031612e-05, + "loss": 2.0376, + "step": 1821 + }, + { + "epoch": 2.336328626444159, + "grad_norm": 0.545239269733429, + "learning_rate": 1.9473251582001804e-05, + "loss": 2.0532, + "step": 1822 + }, + { + "epoch": 2.337612323491656, + "grad_norm": 0.5117034912109375, + "learning_rate": 1.9447857840594748e-05, + "loss": 2.0748, + "step": 1823 + }, + { + "epoch": 2.3388960205391527, + "grad_norm": 0.5135534405708313, + "learning_rate": 1.9422470123363102e-05, + "loss": 2.0135, + "step": 1824 + }, + { + "epoch": 2.3401797175866497, + "grad_norm": 0.5511173009872437, + "learning_rate": 1.9397088457852995e-05, + "loss": 2.0182, + "step": 1825 + }, + { + "epoch": 2.341463414634146, + "grad_norm": 0.5302481055259705, + "learning_rate": 1.937171287160402e-05, + "loss": 2.0243, + "step": 1826 + }, + { + "epoch": 2.342747111681643, + "grad_norm": 0.4949897527694702, + "learning_rate": 1.934634339214913e-05, + "loss": 2.0264, + "step": 1827 + }, + { + "epoch": 2.3440308087291397, + "grad_norm": 0.5643454790115356, + "learning_rate": 1.932098004701467e-05, + "loss": 2.0792, + "step": 1828 + }, + { + "epoch": 2.3453145057766367, + "grad_norm": 0.5818948745727539, + "learning_rate": 1.9295622863720356e-05, + "loss": 2.0082, + "step": 1829 + }, + { + "epoch": 2.3465982028241337, + "grad_norm": 0.5238609313964844, + "learning_rate": 1.9270271869779168e-05, + "loss": 2.0568, + "step": 1830 + }, + { + "epoch": 2.34788189987163, + "grad_norm": 0.5079338550567627, + "learning_rate": 1.9244927092697405e-05, + "loss": 1.9551, + "step": 1831 + }, + { + "epoch": 2.349165596919127, + "grad_norm": 0.49637243151664734, + "learning_rate": 1.9219588559974615e-05, + "loss": 2.0011, + "step": 1832 + }, + { + "epoch": 2.3504492939666237, + "grad_norm": 0.5095198154449463, + "learning_rate": 1.9194256299103573e-05, + "loss": 1.9731, + "step": 1833 + }, + { + "epoch": 2.3517329910141207, + "grad_norm": 0.5152751803398132, + "learning_rate": 1.9168930337570223e-05, + "loss": 1.9583, + "step": 1834 + }, + { + "epoch": 2.3530166880616177, + "grad_norm": 0.5745589137077332, + "learning_rate": 1.9143610702853718e-05, + "loss": 1.983, + "step": 1835 + }, + { + "epoch": 2.354300385109114, + "grad_norm": 0.5311444401741028, + "learning_rate": 1.9118297422426316e-05, + "loss": 2.0541, + "step": 1836 + }, + { + "epoch": 2.355584082156611, + "grad_norm": 0.5557064414024353, + "learning_rate": 1.909299052375338e-05, + "loss": 2.039, + "step": 1837 + }, + { + "epoch": 2.3568677792041077, + "grad_norm": 0.49970462918281555, + "learning_rate": 1.906769003429336e-05, + "loss": 2.0375, + "step": 1838 + }, + { + "epoch": 2.3581514762516047, + "grad_norm": 0.5127582550048828, + "learning_rate": 1.904239598149775e-05, + "loss": 2.0675, + "step": 1839 + }, + { + "epoch": 2.3594351732991012, + "grad_norm": 0.5052181482315063, + "learning_rate": 1.9017108392811065e-05, + "loss": 1.9869, + "step": 1840 + }, + { + "epoch": 2.360718870346598, + "grad_norm": 0.5562523007392883, + "learning_rate": 1.8991827295670776e-05, + "loss": 2.0538, + "step": 1841 + }, + { + "epoch": 2.3620025673940948, + "grad_norm": 0.5186200141906738, + "learning_rate": 1.8966552717507364e-05, + "loss": 1.9998, + "step": 1842 + }, + { + "epoch": 2.3632862644415917, + "grad_norm": 0.497385174036026, + "learning_rate": 1.8941284685744193e-05, + "loss": 2.019, + "step": 1843 + }, + { + "epoch": 2.3645699614890887, + "grad_norm": 0.5367319583892822, + "learning_rate": 1.891602322779752e-05, + "loss": 1.94, + "step": 1844 + }, + { + "epoch": 2.3658536585365852, + "grad_norm": 0.5452361106872559, + "learning_rate": 1.8890768371076523e-05, + "loss": 2.1033, + "step": 1845 + }, + { + "epoch": 2.3671373555840822, + "grad_norm": 0.5299851298332214, + "learning_rate": 1.886552014298316e-05, + "loss": 2.0111, + "step": 1846 + }, + { + "epoch": 2.3684210526315788, + "grad_norm": 0.5406066179275513, + "learning_rate": 1.8840278570912214e-05, + "loss": 1.9835, + "step": 1847 + }, + { + "epoch": 2.3697047496790757, + "grad_norm": 0.5409627556800842, + "learning_rate": 1.8815043682251272e-05, + "loss": 2.1716, + "step": 1848 + }, + { + "epoch": 2.3709884467265727, + "grad_norm": 0.5196877717971802, + "learning_rate": 1.8789815504380624e-05, + "loss": 2.0497, + "step": 1849 + }, + { + "epoch": 2.3722721437740693, + "grad_norm": 0.5191472172737122, + "learning_rate": 1.8764594064673314e-05, + "loss": 1.9863, + "step": 1850 + }, + { + "epoch": 2.3735558408215662, + "grad_norm": 0.4853687584400177, + "learning_rate": 1.8739379390495065e-05, + "loss": 1.9604, + "step": 1851 + }, + { + "epoch": 2.3748395378690628, + "grad_norm": 0.49184587597846985, + "learning_rate": 1.8714171509204263e-05, + "loss": 1.982, + "step": 1852 + }, + { + "epoch": 2.3761232349165597, + "grad_norm": 0.547855019569397, + "learning_rate": 1.8688970448151903e-05, + "loss": 2.1315, + "step": 1853 + }, + { + "epoch": 2.3774069319640567, + "grad_norm": 0.5392388701438904, + "learning_rate": 1.8663776234681608e-05, + "loss": 1.956, + "step": 1854 + }, + { + "epoch": 2.3786906290115533, + "grad_norm": 0.5204140543937683, + "learning_rate": 1.8638588896129557e-05, + "loss": 2.0939, + "step": 1855 + }, + { + "epoch": 2.3799743260590502, + "grad_norm": 0.5259453058242798, + "learning_rate": 1.8613408459824462e-05, + "loss": 2.0113, + "step": 1856 + }, + { + "epoch": 2.3812580231065468, + "grad_norm": 0.5509635806083679, + "learning_rate": 1.858823495308757e-05, + "loss": 1.999, + "step": 1857 + }, + { + "epoch": 2.3825417201540438, + "grad_norm": 0.6122646927833557, + "learning_rate": 1.856306840323258e-05, + "loss": 2.0128, + "step": 1858 + }, + { + "epoch": 2.3838254172015403, + "grad_norm": 0.5398964881896973, + "learning_rate": 1.8537908837565663e-05, + "loss": 1.9673, + "step": 1859 + }, + { + "epoch": 2.3851091142490373, + "grad_norm": 0.5173465609550476, + "learning_rate": 1.8512756283385416e-05, + "loss": 1.9756, + "step": 1860 + }, + { + "epoch": 2.386392811296534, + "grad_norm": 0.511570930480957, + "learning_rate": 1.8487610767982804e-05, + "loss": 2.0118, + "step": 1861 + }, + { + "epoch": 2.387676508344031, + "grad_norm": 0.5118401646614075, + "learning_rate": 1.8462472318641172e-05, + "loss": 2.0498, + "step": 1862 + }, + { + "epoch": 2.3889602053915278, + "grad_norm": 0.5387152433395386, + "learning_rate": 1.8437340962636207e-05, + "loss": 1.997, + "step": 1863 + }, + { + "epoch": 2.3902439024390243, + "grad_norm": 0.5282495617866516, + "learning_rate": 1.8412216727235875e-05, + "loss": 2.0207, + "step": 1864 + }, + { + "epoch": 2.3915275994865213, + "grad_norm": 0.5398370623588562, + "learning_rate": 1.8387099639700437e-05, + "loss": 1.9877, + "step": 1865 + }, + { + "epoch": 2.392811296534018, + "grad_norm": 0.5115079283714294, + "learning_rate": 1.836198972728238e-05, + "loss": 1.9892, + "step": 1866 + }, + { + "epoch": 2.394094993581515, + "grad_norm": 0.524516224861145, + "learning_rate": 1.8336887017226433e-05, + "loss": 2.021, + "step": 1867 + }, + { + "epoch": 2.3953786906290118, + "grad_norm": 0.5235687494277954, + "learning_rate": 1.8311791536769483e-05, + "loss": 1.9571, + "step": 1868 + }, + { + "epoch": 2.3966623876765083, + "grad_norm": 0.5616967678070068, + "learning_rate": 1.828670331314058e-05, + "loss": 2.0639, + "step": 1869 + }, + { + "epoch": 2.3979460847240053, + "grad_norm": 0.5282573699951172, + "learning_rate": 1.826162237356092e-05, + "loss": 1.97, + "step": 1870 + }, + { + "epoch": 2.399229781771502, + "grad_norm": 0.5499706864356995, + "learning_rate": 1.8236548745243763e-05, + "loss": 1.9669, + "step": 1871 + }, + { + "epoch": 2.400513478818999, + "grad_norm": 0.5184937715530396, + "learning_rate": 1.8211482455394458e-05, + "loss": 2.007, + "step": 1872 + }, + { + "epoch": 2.4017971758664953, + "grad_norm": 0.5199077725410461, + "learning_rate": 1.8186423531210383e-05, + "loss": 2.0659, + "step": 1873 + }, + { + "epoch": 2.4030808729139923, + "grad_norm": 0.5302215814590454, + "learning_rate": 1.8161371999880943e-05, + "loss": 2.0246, + "step": 1874 + }, + { + "epoch": 2.404364569961489, + "grad_norm": 0.5133960247039795, + "learning_rate": 1.8136327888587484e-05, + "loss": 2.0009, + "step": 1875 + }, + { + "epoch": 2.405648267008986, + "grad_norm": 0.5016005039215088, + "learning_rate": 1.8111291224503334e-05, + "loss": 1.9204, + "step": 1876 + }, + { + "epoch": 2.406931964056483, + "grad_norm": 0.5422957539558411, + "learning_rate": 1.8086262034793738e-05, + "loss": 2.0155, + "step": 1877 + }, + { + "epoch": 2.4082156611039793, + "grad_norm": 0.5261048078536987, + "learning_rate": 1.8061240346615798e-05, + "loss": 1.9346, + "step": 1878 + }, + { + "epoch": 2.4094993581514763, + "grad_norm": 0.5159074664115906, + "learning_rate": 1.8036226187118533e-05, + "loss": 1.9857, + "step": 1879 + }, + { + "epoch": 2.410783055198973, + "grad_norm": 0.5182771682739258, + "learning_rate": 1.801121958344274e-05, + "loss": 2.038, + "step": 1880 + }, + { + "epoch": 2.41206675224647, + "grad_norm": 0.5117934346199036, + "learning_rate": 1.798622056272104e-05, + "loss": 2.0327, + "step": 1881 + }, + { + "epoch": 2.413350449293967, + "grad_norm": 0.5302830934524536, + "learning_rate": 1.796122915207784e-05, + "loss": 2.0778, + "step": 1882 + }, + { + "epoch": 2.4146341463414633, + "grad_norm": 0.49381110072135925, + "learning_rate": 1.793624537862927e-05, + "loss": 1.9598, + "step": 1883 + }, + { + "epoch": 2.4159178433889603, + "grad_norm": 0.520897626876831, + "learning_rate": 1.7911269269483167e-05, + "loss": 2.0297, + "step": 1884 + }, + { + "epoch": 2.417201540436457, + "grad_norm": 0.49833258986473083, + "learning_rate": 1.788630085173908e-05, + "loss": 1.934, + "step": 1885 + }, + { + "epoch": 2.418485237483954, + "grad_norm": 0.5152247548103333, + "learning_rate": 1.7861340152488198e-05, + "loss": 2.0511, + "step": 1886 + }, + { + "epoch": 2.419768934531451, + "grad_norm": 0.5548563003540039, + "learning_rate": 1.783638719881332e-05, + "loss": 2.0026, + "step": 1887 + }, + { + "epoch": 2.4210526315789473, + "grad_norm": 0.5083991885185242, + "learning_rate": 1.781144201778887e-05, + "loss": 2.0631, + "step": 1888 + }, + { + "epoch": 2.4223363286264443, + "grad_norm": 0.5294684767723083, + "learning_rate": 1.7786504636480822e-05, + "loss": 2.0287, + "step": 1889 + }, + { + "epoch": 2.423620025673941, + "grad_norm": 1.3910973072052002, + "learning_rate": 1.7761575081946675e-05, + "loss": 1.9328, + "step": 1890 + }, + { + "epoch": 2.424903722721438, + "grad_norm": 0.5229716897010803, + "learning_rate": 1.7736653381235463e-05, + "loss": 1.9653, + "step": 1891 + }, + { + "epoch": 2.4261874197689344, + "grad_norm": 0.5225069522857666, + "learning_rate": 1.7711739561387683e-05, + "loss": 2.0034, + "step": 1892 + }, + { + "epoch": 2.4274711168164314, + "grad_norm": 0.5369516611099243, + "learning_rate": 1.768683364943529e-05, + "loss": 2.0218, + "step": 1893 + }, + { + "epoch": 2.428754813863928, + "grad_norm": 0.503600537776947, + "learning_rate": 1.7661935672401632e-05, + "loss": 2.0772, + "step": 1894 + }, + { + "epoch": 2.430038510911425, + "grad_norm": 0.5411602258682251, + "learning_rate": 1.7637045657301488e-05, + "loss": 2.0242, + "step": 1895 + }, + { + "epoch": 2.431322207958922, + "grad_norm": 0.5313899517059326, + "learning_rate": 1.7612163631140976e-05, + "loss": 2.0001, + "step": 1896 + }, + { + "epoch": 2.4326059050064184, + "grad_norm": 0.528943657875061, + "learning_rate": 1.7587289620917536e-05, + "loss": 2.0103, + "step": 1897 + }, + { + "epoch": 2.4338896020539154, + "grad_norm": 0.5219106078147888, + "learning_rate": 1.756242365361993e-05, + "loss": 2.0906, + "step": 1898 + }, + { + "epoch": 2.435173299101412, + "grad_norm": 0.5273393392562866, + "learning_rate": 1.7537565756228187e-05, + "loss": 1.9984, + "step": 1899 + }, + { + "epoch": 2.436456996148909, + "grad_norm": 0.5380772352218628, + "learning_rate": 1.7512715955713566e-05, + "loss": 2.0264, + "step": 1900 + }, + { + "epoch": 2.437740693196406, + "grad_norm": 0.5447054505348206, + "learning_rate": 1.748787427903858e-05, + "loss": 2.0262, + "step": 1901 + }, + { + "epoch": 2.4390243902439024, + "grad_norm": 0.5668127536773682, + "learning_rate": 1.7463040753156877e-05, + "loss": 2.0596, + "step": 1902 + }, + { + "epoch": 2.4403080872913994, + "grad_norm": 0.49480319023132324, + "learning_rate": 1.7438215405013286e-05, + "loss": 1.9197, + "step": 1903 + }, + { + "epoch": 2.441591784338896, + "grad_norm": 0.5153650641441345, + "learning_rate": 1.7413398261543777e-05, + "loss": 2.0258, + "step": 1904 + }, + { + "epoch": 2.442875481386393, + "grad_norm": 0.4807131588459015, + "learning_rate": 1.7388589349675395e-05, + "loss": 1.9846, + "step": 1905 + }, + { + "epoch": 2.4441591784338894, + "grad_norm": 0.49006974697113037, + "learning_rate": 1.7363788696326255e-05, + "loss": 1.921, + "step": 1906 + }, + { + "epoch": 2.4454428754813864, + "grad_norm": 0.49325841665267944, + "learning_rate": 1.7338996328405526e-05, + "loss": 1.9957, + "step": 1907 + }, + { + "epoch": 2.4467265725288834, + "grad_norm": 0.5443248152732849, + "learning_rate": 1.7314212272813385e-05, + "loss": 2.0333, + "step": 1908 + }, + { + "epoch": 2.44801026957638, + "grad_norm": 0.5341103076934814, + "learning_rate": 1.7289436556440964e-05, + "loss": 2.0674, + "step": 1909 + }, + { + "epoch": 2.449293966623877, + "grad_norm": 0.5169807076454163, + "learning_rate": 1.726466920617038e-05, + "loss": 2.0199, + "step": 1910 + }, + { + "epoch": 2.4505776636713734, + "grad_norm": 0.5289947390556335, + "learning_rate": 1.723991024887467e-05, + "loss": 1.9857, + "step": 1911 + }, + { + "epoch": 2.4518613607188704, + "grad_norm": 0.512602686882019, + "learning_rate": 1.721515971141772e-05, + "loss": 1.9557, + "step": 1912 + }, + { + "epoch": 2.453145057766367, + "grad_norm": 0.5398516654968262, + "learning_rate": 1.7190417620654353e-05, + "loss": 1.9857, + "step": 1913 + }, + { + "epoch": 2.454428754813864, + "grad_norm": 0.5164986252784729, + "learning_rate": 1.7165684003430165e-05, + "loss": 1.9858, + "step": 1914 + }, + { + "epoch": 2.455712451861361, + "grad_norm": 0.5139768719673157, + "learning_rate": 1.7140958886581582e-05, + "loss": 2.0377, + "step": 1915 + }, + { + "epoch": 2.4569961489088574, + "grad_norm": 0.5428260564804077, + "learning_rate": 1.7116242296935825e-05, + "loss": 1.9455, + "step": 1916 + }, + { + "epoch": 2.4582798459563544, + "grad_norm": 0.5597161054611206, + "learning_rate": 1.7091534261310823e-05, + "loss": 2.0026, + "step": 1917 + }, + { + "epoch": 2.459563543003851, + "grad_norm": 0.557461142539978, + "learning_rate": 1.706683480651526e-05, + "loss": 1.9107, + "step": 1918 + }, + { + "epoch": 2.460847240051348, + "grad_norm": 0.5398658514022827, + "learning_rate": 1.704214395934847e-05, + "loss": 1.9891, + "step": 1919 + }, + { + "epoch": 2.462130937098845, + "grad_norm": 0.5748401284217834, + "learning_rate": 1.7017461746600506e-05, + "loss": 2.0127, + "step": 1920 + }, + { + "epoch": 2.4634146341463414, + "grad_norm": 0.5319817662239075, + "learning_rate": 1.6992788195051994e-05, + "loss": 2.0391, + "step": 1921 + }, + { + "epoch": 2.4646983311938384, + "grad_norm": 0.5343374013900757, + "learning_rate": 1.696812333147419e-05, + "loss": 2.0163, + "step": 1922 + }, + { + "epoch": 2.465982028241335, + "grad_norm": 0.540174663066864, + "learning_rate": 1.694346718262894e-05, + "loss": 2.1018, + "step": 1923 + }, + { + "epoch": 2.467265725288832, + "grad_norm": 0.5344076752662659, + "learning_rate": 1.691881977526859e-05, + "loss": 2.0237, + "step": 1924 + }, + { + "epoch": 2.4685494223363285, + "grad_norm": 0.5368174314498901, + "learning_rate": 1.6894181136136033e-05, + "loss": 1.9663, + "step": 1925 + }, + { + "epoch": 2.4698331193838254, + "grad_norm": 0.5526659488677979, + "learning_rate": 1.6869551291964652e-05, + "loss": 1.9481, + "step": 1926 + }, + { + "epoch": 2.471116816431322, + "grad_norm": 0.5603521466255188, + "learning_rate": 1.6844930269478274e-05, + "loss": 2.066, + "step": 1927 + }, + { + "epoch": 2.472400513478819, + "grad_norm": 0.5779146552085876, + "learning_rate": 1.682031809539114e-05, + "loss": 1.9808, + "step": 1928 + }, + { + "epoch": 2.473684210526316, + "grad_norm": 0.4998924434185028, + "learning_rate": 1.6795714796407923e-05, + "loss": 1.9912, + "step": 1929 + }, + { + "epoch": 2.4749679075738125, + "grad_norm": 0.5300453305244446, + "learning_rate": 1.6771120399223654e-05, + "loss": 1.9618, + "step": 1930 + }, + { + "epoch": 2.4762516046213094, + "grad_norm": 0.5544107556343079, + "learning_rate": 1.6746534930523683e-05, + "loss": 1.9291, + "step": 1931 + }, + { + "epoch": 2.477535301668806, + "grad_norm": 0.5321759581565857, + "learning_rate": 1.6721958416983717e-05, + "loss": 2.0897, + "step": 1932 + }, + { + "epoch": 2.478818998716303, + "grad_norm": 0.5181989073753357, + "learning_rate": 1.6697390885269705e-05, + "loss": 2.0072, + "step": 1933 + }, + { + "epoch": 2.4801026957638, + "grad_norm": 0.5110918283462524, + "learning_rate": 1.667283236203786e-05, + "loss": 2.0152, + "step": 1934 + }, + { + "epoch": 2.4813863928112965, + "grad_norm": 0.5431555509567261, + "learning_rate": 1.6648282873934655e-05, + "loss": 1.9889, + "step": 1935 + }, + { + "epoch": 2.4826700898587934, + "grad_norm": 0.527590274810791, + "learning_rate": 1.6623742447596702e-05, + "loss": 2.0017, + "step": 1936 + }, + { + "epoch": 2.48395378690629, + "grad_norm": 0.5564057230949402, + "learning_rate": 1.659921110965082e-05, + "loss": 1.9453, + "step": 1937 + }, + { + "epoch": 2.485237483953787, + "grad_norm": 0.49485495686531067, + "learning_rate": 1.6574688886713964e-05, + "loss": 1.9909, + "step": 1938 + }, + { + "epoch": 2.486521181001284, + "grad_norm": 0.5333442687988281, + "learning_rate": 1.655017580539319e-05, + "loss": 1.93, + "step": 1939 + }, + { + "epoch": 2.4878048780487805, + "grad_norm": 0.5216184258460999, + "learning_rate": 1.6525671892285636e-05, + "loss": 2.0356, + "step": 1940 + }, + { + "epoch": 2.4890885750962775, + "grad_norm": 0.5276082754135132, + "learning_rate": 1.6501177173978495e-05, + "loss": 2.0835, + "step": 1941 + }, + { + "epoch": 2.490372272143774, + "grad_norm": 0.5176440477371216, + "learning_rate": 1.6476691677048986e-05, + "loss": 2.0003, + "step": 1942 + }, + { + "epoch": 2.491655969191271, + "grad_norm": 0.49289998412132263, + "learning_rate": 1.6452215428064308e-05, + "loss": 1.9787, + "step": 1943 + }, + { + "epoch": 2.4929396662387675, + "grad_norm": 0.5415627360343933, + "learning_rate": 1.6427748453581647e-05, + "loss": 1.9497, + "step": 1944 + }, + { + "epoch": 2.4942233632862645, + "grad_norm": 0.5466543436050415, + "learning_rate": 1.640329078014812e-05, + "loss": 2.0995, + "step": 1945 + }, + { + "epoch": 2.495507060333761, + "grad_norm": 0.5636317729949951, + "learning_rate": 1.6378842434300746e-05, + "loss": 2.064, + "step": 1946 + }, + { + "epoch": 2.496790757381258, + "grad_norm": 0.5745249390602112, + "learning_rate": 1.6354403442566413e-05, + "loss": 2.0588, + "step": 1947 + }, + { + "epoch": 2.498074454428755, + "grad_norm": 0.547034740447998, + "learning_rate": 1.6329973831461884e-05, + "loss": 1.9994, + "step": 1948 + }, + { + "epoch": 2.4993581514762515, + "grad_norm": 0.5451599359512329, + "learning_rate": 1.6305553627493737e-05, + "loss": 1.9796, + "step": 1949 + }, + { + "epoch": 2.5006418485237485, + "grad_norm": 0.5052309632301331, + "learning_rate": 1.628114285715832e-05, + "loss": 1.992, + "step": 1950 + }, + { + "epoch": 2.501925545571245, + "grad_norm": 0.5103306770324707, + "learning_rate": 1.625674154694178e-05, + "loss": 2.1255, + "step": 1951 + }, + { + "epoch": 2.503209242618742, + "grad_norm": 0.5297012329101562, + "learning_rate": 1.6232349723319978e-05, + "loss": 2.0957, + "step": 1952 + }, + { + "epoch": 2.504492939666239, + "grad_norm": 0.5590283870697021, + "learning_rate": 1.6207967412758482e-05, + "loss": 2.0199, + "step": 1953 + }, + { + "epoch": 2.5057766367137355, + "grad_norm": 0.5556820034980774, + "learning_rate": 1.6183594641712556e-05, + "loss": 2.0295, + "step": 1954 + }, + { + "epoch": 2.5070603337612325, + "grad_norm": 0.5594186782836914, + "learning_rate": 1.615923143662709e-05, + "loss": 2.0229, + "step": 1955 + }, + { + "epoch": 2.508344030808729, + "grad_norm": 0.5865979194641113, + "learning_rate": 1.613487782393661e-05, + "loss": 1.986, + "step": 1956 + }, + { + "epoch": 2.509627727856226, + "grad_norm": 0.5317298769950867, + "learning_rate": 1.6110533830065232e-05, + "loss": 1.9922, + "step": 1957 + }, + { + "epoch": 2.510911424903723, + "grad_norm": 0.5097349882125854, + "learning_rate": 1.608619948142662e-05, + "loss": 1.9618, + "step": 1958 + }, + { + "epoch": 2.5121951219512195, + "grad_norm": 0.5163970589637756, + "learning_rate": 1.6061874804424e-05, + "loss": 2.0279, + "step": 1959 + }, + { + "epoch": 2.513478818998716, + "grad_norm": 0.5088512301445007, + "learning_rate": 1.6037559825450082e-05, + "loss": 1.9988, + "step": 1960 + }, + { + "epoch": 2.514762516046213, + "grad_norm": 0.5428735017776489, + "learning_rate": 1.601325457088707e-05, + "loss": 1.9803, + "step": 1961 + }, + { + "epoch": 2.51604621309371, + "grad_norm": 0.4905744791030884, + "learning_rate": 1.5988959067106592e-05, + "loss": 2.0001, + "step": 1962 + }, + { + "epoch": 2.5173299101412066, + "grad_norm": 0.5366095304489136, + "learning_rate": 1.596467334046973e-05, + "loss": 2.0261, + "step": 1963 + }, + { + "epoch": 2.5186136071887035, + "grad_norm": 0.6065700650215149, + "learning_rate": 1.5940397417326935e-05, + "loss": 1.8876, + "step": 1964 + }, + { + "epoch": 2.5198973042362, + "grad_norm": 0.5636811852455139, + "learning_rate": 1.5916131324018007e-05, + "loss": 2.0022, + "step": 1965 + }, + { + "epoch": 2.521181001283697, + "grad_norm": 0.5613745450973511, + "learning_rate": 1.5891875086872134e-05, + "loss": 1.9776, + "step": 1966 + }, + { + "epoch": 2.522464698331194, + "grad_norm": 0.5337018966674805, + "learning_rate": 1.5867628732207745e-05, + "loss": 1.9854, + "step": 1967 + }, + { + "epoch": 2.5237483953786906, + "grad_norm": 0.5647414922714233, + "learning_rate": 1.5843392286332586e-05, + "loss": 2.0488, + "step": 1968 + }, + { + "epoch": 2.5250320924261875, + "grad_norm": 0.5213755369186401, + "learning_rate": 1.5819165775543658e-05, + "loss": 1.9618, + "step": 1969 + }, + { + "epoch": 2.526315789473684, + "grad_norm": 0.5201895236968994, + "learning_rate": 1.5794949226127145e-05, + "loss": 2.027, + "step": 1970 + }, + { + "epoch": 2.527599486521181, + "grad_norm": 0.5636970400810242, + "learning_rate": 1.577074266435845e-05, + "loss": 2.0368, + "step": 1971 + }, + { + "epoch": 2.528883183568678, + "grad_norm": 0.5626921653747559, + "learning_rate": 1.574654611650214e-05, + "loss": 1.9863, + "step": 1972 + }, + { + "epoch": 2.5301668806161746, + "grad_norm": 0.5384525656700134, + "learning_rate": 1.5722359608811912e-05, + "loss": 2.006, + "step": 1973 + }, + { + "epoch": 2.531450577663671, + "grad_norm": 0.5238264799118042, + "learning_rate": 1.5698183167530556e-05, + "loss": 1.9434, + "step": 1974 + }, + { + "epoch": 2.532734274711168, + "grad_norm": 0.5604907870292664, + "learning_rate": 1.5674016818889955e-05, + "loss": 1.9563, + "step": 1975 + }, + { + "epoch": 2.534017971758665, + "grad_norm": 0.5799750089645386, + "learning_rate": 1.5649860589111047e-05, + "loss": 1.9977, + "step": 1976 + }, + { + "epoch": 2.5353016688061616, + "grad_norm": 0.5583044290542603, + "learning_rate": 1.5625714504403773e-05, + "loss": 2.0529, + "step": 1977 + }, + { + "epoch": 2.5365853658536586, + "grad_norm": 0.5434403419494629, + "learning_rate": 1.560157859096707e-05, + "loss": 1.9488, + "step": 1978 + }, + { + "epoch": 2.537869062901155, + "grad_norm": 0.5083727240562439, + "learning_rate": 1.5577452874988856e-05, + "loss": 1.9407, + "step": 1979 + }, + { + "epoch": 2.539152759948652, + "grad_norm": 0.5344621539115906, + "learning_rate": 1.5553337382645972e-05, + "loss": 1.928, + "step": 1980 + }, + { + "epoch": 2.540436456996149, + "grad_norm": 0.5587793588638306, + "learning_rate": 1.5529232140104153e-05, + "loss": 1.9753, + "step": 1981 + }, + { + "epoch": 2.5417201540436456, + "grad_norm": 0.5375910401344299, + "learning_rate": 1.5505137173518036e-05, + "loss": 1.9366, + "step": 1982 + }, + { + "epoch": 2.5430038510911426, + "grad_norm": 0.5258891582489014, + "learning_rate": 1.548105250903111e-05, + "loss": 2.0682, + "step": 1983 + }, + { + "epoch": 2.544287548138639, + "grad_norm": 0.5473222136497498, + "learning_rate": 1.5456978172775644e-05, + "loss": 1.9827, + "step": 1984 + }, + { + "epoch": 2.545571245186136, + "grad_norm": 0.5542914867401123, + "learning_rate": 1.5432914190872757e-05, + "loss": 1.9405, + "step": 1985 + }, + { + "epoch": 2.546854942233633, + "grad_norm": 0.5510945916175842, + "learning_rate": 1.5408860589432303e-05, + "loss": 2.0563, + "step": 1986 + }, + { + "epoch": 2.5481386392811296, + "grad_norm": 0.5383923053741455, + "learning_rate": 1.538481739455287e-05, + "loss": 2.0723, + "step": 1987 + }, + { + "epoch": 2.5494223363286266, + "grad_norm": 0.5421393513679504, + "learning_rate": 1.5360784632321777e-05, + "loss": 2.1008, + "step": 1988 + }, + { + "epoch": 2.550706033376123, + "grad_norm": 0.5721552968025208, + "learning_rate": 1.5336762328815e-05, + "loss": 2.024, + "step": 1989 + }, + { + "epoch": 2.55198973042362, + "grad_norm": 0.5305389165878296, + "learning_rate": 1.531275051009717e-05, + "loss": 2.0437, + "step": 1990 + }, + { + "epoch": 2.553273427471117, + "grad_norm": 0.5392176508903503, + "learning_rate": 1.528874920222156e-05, + "loss": 1.9595, + "step": 1991 + }, + { + "epoch": 2.5545571245186136, + "grad_norm": 0.5371749401092529, + "learning_rate": 1.5264758431230015e-05, + "loss": 2.0077, + "step": 1992 + }, + { + "epoch": 2.55584082156611, + "grad_norm": 0.5438774228096008, + "learning_rate": 1.5240778223152962e-05, + "loss": 2.0482, + "step": 1993 + }, + { + "epoch": 2.557124518613607, + "grad_norm": 0.5551732182502747, + "learning_rate": 1.5216808604009369e-05, + "loss": 1.9923, + "step": 1994 + }, + { + "epoch": 2.558408215661104, + "grad_norm": 0.5298148393630981, + "learning_rate": 1.5192849599806714e-05, + "loss": 1.9784, + "step": 1995 + }, + { + "epoch": 2.5596919127086006, + "grad_norm": 0.5484992265701294, + "learning_rate": 1.5168901236540936e-05, + "loss": 2.0558, + "step": 1996 + }, + { + "epoch": 2.5609756097560976, + "grad_norm": 0.5438374876976013, + "learning_rate": 1.5144963540196458e-05, + "loss": 2.0772, + "step": 1997 + }, + { + "epoch": 2.562259306803594, + "grad_norm": 0.5419697761535645, + "learning_rate": 1.512103653674612e-05, + "loss": 2.0558, + "step": 1998 + }, + { + "epoch": 2.563543003851091, + "grad_norm": 0.5184198021888733, + "learning_rate": 1.5097120252151145e-05, + "loss": 1.9682, + "step": 1999 + }, + { + "epoch": 2.564826700898588, + "grad_norm": 0.5800809860229492, + "learning_rate": 1.507321471236114e-05, + "loss": 1.979, + "step": 2000 + }, + { + "epoch": 2.5661103979460846, + "grad_norm": 0.5533127188682556, + "learning_rate": 1.5049319943314054e-05, + "loss": 2.0394, + "step": 2001 + }, + { + "epoch": 2.5673940949935816, + "grad_norm": 0.5195207595825195, + "learning_rate": 1.5025435970936159e-05, + "loss": 2.0707, + "step": 2002 + }, + { + "epoch": 2.568677792041078, + "grad_norm": 0.5645703673362732, + "learning_rate": 1.5001562821141974e-05, + "loss": 2.0925, + "step": 2003 + }, + { + "epoch": 2.569961489088575, + "grad_norm": 0.522128164768219, + "learning_rate": 1.497770051983432e-05, + "loss": 1.963, + "step": 2004 + }, + { + "epoch": 2.571245186136072, + "grad_norm": 0.5413267612457275, + "learning_rate": 1.4953849092904232e-05, + "loss": 2.0328, + "step": 2005 + }, + { + "epoch": 2.5725288831835686, + "grad_norm": 0.5530142188072205, + "learning_rate": 1.4930008566230922e-05, + "loss": 1.9925, + "step": 2006 + }, + { + "epoch": 2.5738125802310656, + "grad_norm": 0.5327922105789185, + "learning_rate": 1.4906178965681822e-05, + "loss": 2.0725, + "step": 2007 + }, + { + "epoch": 2.575096277278562, + "grad_norm": 0.5339775085449219, + "learning_rate": 1.4882360317112465e-05, + "loss": 2.0458, + "step": 2008 + }, + { + "epoch": 2.576379974326059, + "grad_norm": 0.5579934120178223, + "learning_rate": 1.485855264636652e-05, + "loss": 2.0032, + "step": 2009 + }, + { + "epoch": 2.5776636713735557, + "grad_norm": 0.5335544943809509, + "learning_rate": 1.483475597927576e-05, + "loss": 1.9678, + "step": 2010 + }, + { + "epoch": 2.5789473684210527, + "grad_norm": 0.536522626876831, + "learning_rate": 1.481097034165998e-05, + "loss": 1.8903, + "step": 2011 + }, + { + "epoch": 2.580231065468549, + "grad_norm": 0.5415696501731873, + "learning_rate": 1.4787195759327039e-05, + "loss": 2.034, + "step": 2012 + }, + { + "epoch": 2.581514762516046, + "grad_norm": 0.5504868626594543, + "learning_rate": 1.4763432258072796e-05, + "loss": 2.0068, + "step": 2013 + }, + { + "epoch": 2.582798459563543, + "grad_norm": 0.5929387211799622, + "learning_rate": 1.4739679863681086e-05, + "loss": 2.0483, + "step": 2014 + }, + { + "epoch": 2.5840821566110397, + "grad_norm": 0.5552975535392761, + "learning_rate": 1.4715938601923674e-05, + "loss": 2.0434, + "step": 2015 + }, + { + "epoch": 2.5853658536585367, + "grad_norm": 0.5383045077323914, + "learning_rate": 1.4692208498560276e-05, + "loss": 2.01, + "step": 2016 + }, + { + "epoch": 2.586649550706033, + "grad_norm": 0.5294163823127747, + "learning_rate": 1.4668489579338484e-05, + "loss": 1.9414, + "step": 2017 + }, + { + "epoch": 2.58793324775353, + "grad_norm": 0.5274396538734436, + "learning_rate": 1.4644781869993751e-05, + "loss": 1.9508, + "step": 2018 + }, + { + "epoch": 2.589216944801027, + "grad_norm": 0.5808762907981873, + "learning_rate": 1.4621085396249379e-05, + "loss": 2.1139, + "step": 2019 + }, + { + "epoch": 2.5905006418485237, + "grad_norm": 0.5379817485809326, + "learning_rate": 1.4597400183816473e-05, + "loss": 2.0133, + "step": 2020 + }, + { + "epoch": 2.5917843388960207, + "grad_norm": 0.5073413848876953, + "learning_rate": 1.4573726258393921e-05, + "loss": 1.8789, + "step": 2021 + }, + { + "epoch": 2.593068035943517, + "grad_norm": 0.5445347428321838, + "learning_rate": 1.455006364566836e-05, + "loss": 2.0043, + "step": 2022 + }, + { + "epoch": 2.594351732991014, + "grad_norm": 0.548230767250061, + "learning_rate": 1.4526412371314163e-05, + "loss": 1.9723, + "step": 2023 + }, + { + "epoch": 2.595635430038511, + "grad_norm": 0.5219336748123169, + "learning_rate": 1.4502772460993385e-05, + "loss": 2.038, + "step": 2024 + }, + { + "epoch": 2.5969191270860077, + "grad_norm": 0.5162960886955261, + "learning_rate": 1.4479143940355769e-05, + "loss": 1.8889, + "step": 2025 + }, + { + "epoch": 2.5982028241335042, + "grad_norm": 0.536469042301178, + "learning_rate": 1.4455526835038676e-05, + "loss": 2.0286, + "step": 2026 + }, + { + "epoch": 2.599486521181001, + "grad_norm": 0.5771524310112, + "learning_rate": 1.4431921170667117e-05, + "loss": 1.9784, + "step": 2027 + }, + { + "epoch": 2.600770218228498, + "grad_norm": 0.5172984004020691, + "learning_rate": 1.4408326972853642e-05, + "loss": 1.9928, + "step": 2028 + }, + { + "epoch": 2.6020539152759947, + "grad_norm": 0.5596245527267456, + "learning_rate": 1.4384744267198414e-05, + "loss": 2.0233, + "step": 2029 + }, + { + "epoch": 2.6033376123234917, + "grad_norm": 0.5610369443893433, + "learning_rate": 1.4361173079289079e-05, + "loss": 1.949, + "step": 2030 + }, + { + "epoch": 2.6046213093709882, + "grad_norm": 0.5482391119003296, + "learning_rate": 1.43376134347008e-05, + "loss": 2.0105, + "step": 2031 + }, + { + "epoch": 2.605905006418485, + "grad_norm": 0.5363342761993408, + "learning_rate": 1.431406535899623e-05, + "loss": 2.1322, + "step": 2032 + }, + { + "epoch": 2.607188703465982, + "grad_norm": 0.6064974069595337, + "learning_rate": 1.4290528877725474e-05, + "loss": 1.8428, + "step": 2033 + }, + { + "epoch": 2.6084724005134787, + "grad_norm": 0.5704812407493591, + "learning_rate": 1.4267004016426017e-05, + "loss": 1.9904, + "step": 2034 + }, + { + "epoch": 2.6097560975609757, + "grad_norm": 0.5599377155303955, + "learning_rate": 1.4243490800622788e-05, + "loss": 2.0573, + "step": 2035 + }, + { + "epoch": 2.6110397946084722, + "grad_norm": 0.5466123223304749, + "learning_rate": 1.4219989255828029e-05, + "loss": 2.0835, + "step": 2036 + }, + { + "epoch": 2.612323491655969, + "grad_norm": 0.5415637493133545, + "learning_rate": 1.4196499407541359e-05, + "loss": 1.9669, + "step": 2037 + }, + { + "epoch": 2.613607188703466, + "grad_norm": 0.5516363978385925, + "learning_rate": 1.4173021281249696e-05, + "loss": 1.9866, + "step": 2038 + }, + { + "epoch": 2.6148908857509627, + "grad_norm": 0.562350332736969, + "learning_rate": 1.4149554902427232e-05, + "loss": 2.014, + "step": 2039 + }, + { + "epoch": 2.6161745827984597, + "grad_norm": 0.5548406839370728, + "learning_rate": 1.41261002965354e-05, + "loss": 2.0019, + "step": 2040 + }, + { + "epoch": 2.6174582798459562, + "grad_norm": 0.5459633469581604, + "learning_rate": 1.4102657489022885e-05, + "loss": 1.949, + "step": 2041 + }, + { + "epoch": 2.6187419768934532, + "grad_norm": 0.55186927318573, + "learning_rate": 1.407922650532557e-05, + "loss": 2.0362, + "step": 2042 + }, + { + "epoch": 2.62002567394095, + "grad_norm": 0.5479947328567505, + "learning_rate": 1.4055807370866485e-05, + "loss": 2.0369, + "step": 2043 + }, + { + "epoch": 2.6213093709884467, + "grad_norm": 0.5622584819793701, + "learning_rate": 1.4032400111055834e-05, + "loss": 1.9774, + "step": 2044 + }, + { + "epoch": 2.6225930680359433, + "grad_norm": 0.5559549927711487, + "learning_rate": 1.4009004751290911e-05, + "loss": 1.9956, + "step": 2045 + }, + { + "epoch": 2.6238767650834403, + "grad_norm": 0.5495298504829407, + "learning_rate": 1.3985621316956083e-05, + "loss": 2.1074, + "step": 2046 + }, + { + "epoch": 2.6251604621309372, + "grad_norm": 0.556067705154419, + "learning_rate": 1.3962249833422846e-05, + "loss": 2.0403, + "step": 2047 + }, + { + "epoch": 2.6264441591784338, + "grad_norm": 0.580852746963501, + "learning_rate": 1.3938890326049669e-05, + "loss": 1.9662, + "step": 2048 + }, + { + "epoch": 2.6277278562259307, + "grad_norm": 0.5233368873596191, + "learning_rate": 1.391554282018203e-05, + "loss": 2.0113, + "step": 2049 + }, + { + "epoch": 2.6290115532734273, + "grad_norm": 0.5601373314857483, + "learning_rate": 1.3892207341152416e-05, + "loss": 2.104, + "step": 2050 + }, + { + "epoch": 2.6302952503209243, + "grad_norm": 0.497255802154541, + "learning_rate": 1.3868883914280228e-05, + "loss": 1.9693, + "step": 2051 + }, + { + "epoch": 2.6315789473684212, + "grad_norm": 0.5294634103775024, + "learning_rate": 1.3845572564871822e-05, + "loss": 1.9905, + "step": 2052 + }, + { + "epoch": 2.6328626444159178, + "grad_norm": 0.5218788981437683, + "learning_rate": 1.3822273318220447e-05, + "loss": 1.9787, + "step": 2053 + }, + { + "epoch": 2.6341463414634148, + "grad_norm": 0.539400041103363, + "learning_rate": 1.3798986199606195e-05, + "loss": 2.0137, + "step": 2054 + }, + { + "epoch": 2.6354300385109113, + "grad_norm": 0.5297998189926147, + "learning_rate": 1.377571123429601e-05, + "loss": 1.9981, + "step": 2055 + }, + { + "epoch": 2.6367137355584083, + "grad_norm": 0.5595104694366455, + "learning_rate": 1.375244844754366e-05, + "loss": 2.0334, + "step": 2056 + }, + { + "epoch": 2.6379974326059052, + "grad_norm": 0.6274513602256775, + "learning_rate": 1.3729197864589701e-05, + "loss": 2.0055, + "step": 2057 + }, + { + "epoch": 2.639281129653402, + "grad_norm": 0.5563034415245056, + "learning_rate": 1.370595951066143e-05, + "loss": 1.9656, + "step": 2058 + }, + { + "epoch": 2.6405648267008983, + "grad_norm": 0.5607799291610718, + "learning_rate": 1.3682733410972882e-05, + "loss": 2.0541, + "step": 2059 + }, + { + "epoch": 2.6418485237483953, + "grad_norm": 0.5698713064193726, + "learning_rate": 1.3659519590724801e-05, + "loss": 2.1528, + "step": 2060 + }, + { + "epoch": 2.6431322207958923, + "grad_norm": 0.5388235449790955, + "learning_rate": 1.3636318075104612e-05, + "loss": 1.9592, + "step": 2061 + }, + { + "epoch": 2.644415917843389, + "grad_norm": 0.5151246786117554, + "learning_rate": 1.361312888928637e-05, + "loss": 2.0198, + "step": 2062 + }, + { + "epoch": 2.645699614890886, + "grad_norm": 0.5586761236190796, + "learning_rate": 1.3589952058430778e-05, + "loss": 2.0062, + "step": 2063 + }, + { + "epoch": 2.6469833119383823, + "grad_norm": 0.555968165397644, + "learning_rate": 1.3566787607685114e-05, + "loss": 1.9891, + "step": 2064 + }, + { + "epoch": 2.6482670089858793, + "grad_norm": 0.5436508655548096, + "learning_rate": 1.3543635562183215e-05, + "loss": 1.9647, + "step": 2065 + }, + { + "epoch": 2.6495507060333763, + "grad_norm": 0.5394912362098694, + "learning_rate": 1.3520495947045487e-05, + "loss": 2.0221, + "step": 2066 + }, + { + "epoch": 2.650834403080873, + "grad_norm": 0.5883912444114685, + "learning_rate": 1.3497368787378844e-05, + "loss": 1.8885, + "step": 2067 + }, + { + "epoch": 2.65211810012837, + "grad_norm": 0.5305804014205933, + "learning_rate": 1.3474254108276652e-05, + "loss": 2.0098, + "step": 2068 + }, + { + "epoch": 2.6534017971758663, + "grad_norm": 0.5512009859085083, + "learning_rate": 1.3451151934818778e-05, + "loss": 2.0474, + "step": 2069 + }, + { + "epoch": 2.6546854942233633, + "grad_norm": 0.5711780190467834, + "learning_rate": 1.3428062292071486e-05, + "loss": 1.8969, + "step": 2070 + }, + { + "epoch": 2.6559691912708603, + "grad_norm": 0.5429300665855408, + "learning_rate": 1.3404985205087466e-05, + "loss": 2.0137, + "step": 2071 + }, + { + "epoch": 2.657252888318357, + "grad_norm": 0.5195328593254089, + "learning_rate": 1.3381920698905787e-05, + "loss": 2.0082, + "step": 2072 + }, + { + "epoch": 2.658536585365854, + "grad_norm": 0.5631638169288635, + "learning_rate": 1.3358868798551855e-05, + "loss": 1.9833, + "step": 2073 + }, + { + "epoch": 2.6598202824133503, + "grad_norm": 0.5525002479553223, + "learning_rate": 1.333582952903738e-05, + "loss": 1.964, + "step": 2074 + }, + { + "epoch": 2.6611039794608473, + "grad_norm": 0.5621917843818665, + "learning_rate": 1.3312802915360411e-05, + "loss": 2.0521, + "step": 2075 + }, + { + "epoch": 2.6623876765083443, + "grad_norm": 0.6214123368263245, + "learning_rate": 1.328978898250525e-05, + "loss": 1.9617, + "step": 2076 + }, + { + "epoch": 2.663671373555841, + "grad_norm": 0.5635185837745667, + "learning_rate": 1.3266787755442412e-05, + "loss": 2.0339, + "step": 2077 + }, + { + "epoch": 2.6649550706033374, + "grad_norm": 0.5326119661331177, + "learning_rate": 1.3243799259128664e-05, + "loss": 1.9424, + "step": 2078 + }, + { + "epoch": 2.6662387676508343, + "grad_norm": 0.5824853777885437, + "learning_rate": 1.322082351850694e-05, + "loss": 1.9785, + "step": 2079 + }, + { + "epoch": 2.6675224646983313, + "grad_norm": 0.5297137498855591, + "learning_rate": 1.3197860558506322e-05, + "loss": 1.939, + "step": 2080 + }, + { + "epoch": 2.668806161745828, + "grad_norm": 0.5503960251808167, + "learning_rate": 1.3174910404042053e-05, + "loss": 2.0175, + "step": 2081 + }, + { + "epoch": 2.670089858793325, + "grad_norm": 0.5721628665924072, + "learning_rate": 1.3151973080015479e-05, + "loss": 1.9884, + "step": 2082 + }, + { + "epoch": 2.6713735558408214, + "grad_norm": 0.5558338165283203, + "learning_rate": 1.3129048611314004e-05, + "loss": 1.9269, + "step": 2083 + }, + { + "epoch": 2.6726572528883183, + "grad_norm": 0.5437233448028564, + "learning_rate": 1.3106137022811083e-05, + "loss": 1.9605, + "step": 2084 + }, + { + "epoch": 2.6739409499358153, + "grad_norm": 0.5792387127876282, + "learning_rate": 1.308323833936622e-05, + "loss": 2.1424, + "step": 2085 + }, + { + "epoch": 2.675224646983312, + "grad_norm": 0.5297888517379761, + "learning_rate": 1.3060352585824908e-05, + "loss": 1.9053, + "step": 2086 + }, + { + "epoch": 2.676508344030809, + "grad_norm": 0.5965456962585449, + "learning_rate": 1.3037479787018592e-05, + "loss": 2.0564, + "step": 2087 + }, + { + "epoch": 2.6777920410783054, + "grad_norm": 0.5408007502555847, + "learning_rate": 1.3014619967764696e-05, + "loss": 1.9581, + "step": 2088 + }, + { + "epoch": 2.6790757381258024, + "grad_norm": 0.5551177859306335, + "learning_rate": 1.2991773152866515e-05, + "loss": 2.031, + "step": 2089 + }, + { + "epoch": 2.6803594351732993, + "grad_norm": 0.6007508635520935, + "learning_rate": 1.2968939367113275e-05, + "loss": 2.0252, + "step": 2090 + }, + { + "epoch": 2.681643132220796, + "grad_norm": 0.5449510216712952, + "learning_rate": 1.2946118635280057e-05, + "loss": 2.0359, + "step": 2091 + }, + { + "epoch": 2.682926829268293, + "grad_norm": 0.6140665411949158, + "learning_rate": 1.2923310982127763e-05, + "loss": 2.0494, + "step": 2092 + }, + { + "epoch": 2.6842105263157894, + "grad_norm": 0.5340210199356079, + "learning_rate": 1.2900516432403098e-05, + "loss": 1.9674, + "step": 2093 + }, + { + "epoch": 2.6854942233632864, + "grad_norm": 0.5559483170509338, + "learning_rate": 1.2877735010838577e-05, + "loss": 2.0071, + "step": 2094 + }, + { + "epoch": 2.686777920410783, + "grad_norm": 0.547577440738678, + "learning_rate": 1.2854966742152464e-05, + "loss": 2.0624, + "step": 2095 + }, + { + "epoch": 2.68806161745828, + "grad_norm": 0.5905643105506897, + "learning_rate": 1.2832211651048731e-05, + "loss": 2.0397, + "step": 2096 + }, + { + "epoch": 2.6893453145057764, + "grad_norm": 0.5170870423316956, + "learning_rate": 1.2809469762217084e-05, + "loss": 1.9365, + "step": 2097 + }, + { + "epoch": 2.6906290115532734, + "grad_norm": 0.5267345309257507, + "learning_rate": 1.2786741100332872e-05, + "loss": 1.9793, + "step": 2098 + }, + { + "epoch": 2.6919127086007704, + "grad_norm": 0.5460230708122253, + "learning_rate": 1.2764025690057107e-05, + "loss": 1.981, + "step": 2099 + }, + { + "epoch": 2.693196405648267, + "grad_norm": 0.5553687810897827, + "learning_rate": 1.2741323556036427e-05, + "loss": 1.9858, + "step": 2100 + }, + { + "epoch": 2.694480102695764, + "grad_norm": 0.5518766641616821, + "learning_rate": 1.2718634722903073e-05, + "loss": 2.0153, + "step": 2101 + }, + { + "epoch": 2.6957637997432604, + "grad_norm": 0.5717878341674805, + "learning_rate": 1.2695959215274816e-05, + "loss": 1.9538, + "step": 2102 + }, + { + "epoch": 2.6970474967907574, + "grad_norm": 0.5203715562820435, + "learning_rate": 1.2673297057755024e-05, + "loss": 1.9726, + "step": 2103 + }, + { + "epoch": 2.6983311938382544, + "grad_norm": 0.5451764464378357, + "learning_rate": 1.2650648274932526e-05, + "loss": 1.9718, + "step": 2104 + }, + { + "epoch": 2.699614890885751, + "grad_norm": 0.5693439841270447, + "learning_rate": 1.2628012891381678e-05, + "loss": 2.0875, + "step": 2105 + }, + { + "epoch": 2.700898587933248, + "grad_norm": 0.5788887143135071, + "learning_rate": 1.2605390931662286e-05, + "loss": 2.0083, + "step": 2106 + }, + { + "epoch": 2.7021822849807444, + "grad_norm": 0.6097930669784546, + "learning_rate": 1.2582782420319587e-05, + "loss": 1.9882, + "step": 2107 + }, + { + "epoch": 2.7034659820282414, + "grad_norm": 0.5056913495063782, + "learning_rate": 1.2560187381884211e-05, + "loss": 2.043, + "step": 2108 + }, + { + "epoch": 2.7047496790757384, + "grad_norm": 0.5836002230644226, + "learning_rate": 1.2537605840872203e-05, + "loss": 2.068, + "step": 2109 + }, + { + "epoch": 2.706033376123235, + "grad_norm": 0.5466415286064148, + "learning_rate": 1.251503782178495e-05, + "loss": 1.967, + "step": 2110 + }, + { + "epoch": 2.7073170731707314, + "grad_norm": 0.5534573197364807, + "learning_rate": 1.2492483349109157e-05, + "loss": 1.9919, + "step": 2111 + }, + { + "epoch": 2.7086007702182284, + "grad_norm": 0.5439583659172058, + "learning_rate": 1.246994244731683e-05, + "loss": 1.9329, + "step": 2112 + }, + { + "epoch": 2.7098844672657254, + "grad_norm": 0.546409547328949, + "learning_rate": 1.2447415140865265e-05, + "loss": 2.0754, + "step": 2113 + }, + { + "epoch": 2.711168164313222, + "grad_norm": 0.5618013143539429, + "learning_rate": 1.2424901454197011e-05, + "loss": 1.9228, + "step": 2114 + }, + { + "epoch": 2.712451861360719, + "grad_norm": 0.5284646153450012, + "learning_rate": 1.2402401411739806e-05, + "loss": 2.0527, + "step": 2115 + }, + { + "epoch": 2.7137355584082155, + "grad_norm": 0.5196371674537659, + "learning_rate": 1.2379915037906628e-05, + "loss": 2.0839, + "step": 2116 + }, + { + "epoch": 2.7150192554557124, + "grad_norm": 0.5332878232002258, + "learning_rate": 1.2357442357095595e-05, + "loss": 1.9975, + "step": 2117 + }, + { + "epoch": 2.7163029525032094, + "grad_norm": 0.5665094256401062, + "learning_rate": 1.2334983393689962e-05, + "loss": 2.0853, + "step": 2118 + }, + { + "epoch": 2.717586649550706, + "grad_norm": 0.5371212363243103, + "learning_rate": 1.2312538172058125e-05, + "loss": 2.097, + "step": 2119 + }, + { + "epoch": 2.718870346598203, + "grad_norm": 0.5482370257377625, + "learning_rate": 1.2290106716553565e-05, + "loss": 1.9435, + "step": 2120 + }, + { + "epoch": 2.7201540436456995, + "grad_norm": 0.5669691562652588, + "learning_rate": 1.2267689051514805e-05, + "loss": 1.9736, + "step": 2121 + }, + { + "epoch": 2.7214377406931964, + "grad_norm": 0.5576844215393066, + "learning_rate": 1.2245285201265435e-05, + "loss": 2.0364, + "step": 2122 + }, + { + "epoch": 2.7227214377406934, + "grad_norm": 0.5511747598648071, + "learning_rate": 1.2222895190114025e-05, + "loss": 2.0991, + "step": 2123 + }, + { + "epoch": 2.72400513478819, + "grad_norm": 0.5507130026817322, + "learning_rate": 1.2200519042354156e-05, + "loss": 1.9673, + "step": 2124 + }, + { + "epoch": 2.725288831835687, + "grad_norm": 0.5788597464561462, + "learning_rate": 1.217815678226436e-05, + "loss": 2.0349, + "step": 2125 + }, + { + "epoch": 2.7265725288831835, + "grad_norm": 0.5715432167053223, + "learning_rate": 1.2155808434108094e-05, + "loss": 1.9314, + "step": 2126 + }, + { + "epoch": 2.7278562259306804, + "grad_norm": 0.545390248298645, + "learning_rate": 1.2133474022133707e-05, + "loss": 2.0048, + "step": 2127 + }, + { + "epoch": 2.7291399229781774, + "grad_norm": 0.5395144820213318, + "learning_rate": 1.2111153570574454e-05, + "loss": 1.9767, + "step": 2128 + }, + { + "epoch": 2.730423620025674, + "grad_norm": 0.5719544887542725, + "learning_rate": 1.2088847103648442e-05, + "loss": 1.9932, + "step": 2129 + }, + { + "epoch": 2.7317073170731705, + "grad_norm": 0.5442613363265991, + "learning_rate": 1.2066554645558578e-05, + "loss": 2.0159, + "step": 2130 + }, + { + "epoch": 2.7329910141206675, + "grad_norm": 0.5237276554107666, + "learning_rate": 1.20442762204926e-05, + "loss": 2.0023, + "step": 2131 + }, + { + "epoch": 2.7342747111681645, + "grad_norm": 0.5332069396972656, + "learning_rate": 1.2022011852623e-05, + "loss": 2.0387, + "step": 2132 + }, + { + "epoch": 2.735558408215661, + "grad_norm": 0.5519712567329407, + "learning_rate": 1.1999761566106998e-05, + "loss": 2.0386, + "step": 2133 + }, + { + "epoch": 2.736842105263158, + "grad_norm": 0.5348168611526489, + "learning_rate": 1.1977525385086603e-05, + "loss": 2.0348, + "step": 2134 + }, + { + "epoch": 2.7381258023106545, + "grad_norm": 0.5550469160079956, + "learning_rate": 1.1955303333688455e-05, + "loss": 2.0433, + "step": 2135 + }, + { + "epoch": 2.7394094993581515, + "grad_norm": 0.5633547306060791, + "learning_rate": 1.1933095436023886e-05, + "loss": 1.9942, + "step": 2136 + }, + { + "epoch": 2.7406931964056485, + "grad_norm": 0.570434033870697, + "learning_rate": 1.191090171618886e-05, + "loss": 1.9769, + "step": 2137 + }, + { + "epoch": 2.741976893453145, + "grad_norm": 0.5203814506530762, + "learning_rate": 1.1888722198263977e-05, + "loss": 2.0616, + "step": 2138 + }, + { + "epoch": 2.743260590500642, + "grad_norm": 0.5713605880737305, + "learning_rate": 1.1866556906314427e-05, + "loss": 1.9728, + "step": 2139 + }, + { + "epoch": 2.7445442875481385, + "grad_norm": 0.5360279083251953, + "learning_rate": 1.1844405864389938e-05, + "loss": 1.9946, + "step": 2140 + }, + { + "epoch": 2.7458279845956355, + "grad_norm": 0.5517716407775879, + "learning_rate": 1.1822269096524812e-05, + "loss": 2.0672, + "step": 2141 + }, + { + "epoch": 2.7471116816431325, + "grad_norm": 0.5839133262634277, + "learning_rate": 1.1800146626737826e-05, + "loss": 2.026, + "step": 2142 + }, + { + "epoch": 2.748395378690629, + "grad_norm": 0.5406709313392639, + "learning_rate": 1.1778038479032274e-05, + "loss": 2.1065, + "step": 2143 + }, + { + "epoch": 2.7496790757381255, + "grad_norm": 0.5716677904129028, + "learning_rate": 1.1755944677395905e-05, + "loss": 1.9584, + "step": 2144 + }, + { + "epoch": 2.7509627727856225, + "grad_norm": 0.5799934267997742, + "learning_rate": 1.1733865245800893e-05, + "loss": 1.9917, + "step": 2145 + }, + { + "epoch": 2.7522464698331195, + "grad_norm": 0.5244773626327515, + "learning_rate": 1.1711800208203807e-05, + "loss": 1.9952, + "step": 2146 + }, + { + "epoch": 2.753530166880616, + "grad_norm": 0.5855675935745239, + "learning_rate": 1.1689749588545629e-05, + "loss": 2.0647, + "step": 2147 + }, + { + "epoch": 2.754813863928113, + "grad_norm": 0.5698898434638977, + "learning_rate": 1.1667713410751693e-05, + "loss": 2.0831, + "step": 2148 + }, + { + "epoch": 2.7560975609756095, + "grad_norm": 0.5250542759895325, + "learning_rate": 1.1645691698731632e-05, + "loss": 2.0041, + "step": 2149 + }, + { + "epoch": 2.7573812580231065, + "grad_norm": 0.5392619967460632, + "learning_rate": 1.1623684476379424e-05, + "loss": 2.042, + "step": 2150 + }, + { + "epoch": 2.7586649550706035, + "grad_norm": 0.5124844908714294, + "learning_rate": 1.1601691767573298e-05, + "loss": 1.945, + "step": 2151 + }, + { + "epoch": 2.7599486521181, + "grad_norm": 0.6192339062690735, + "learning_rate": 1.1579713596175735e-05, + "loss": 2.0423, + "step": 2152 + }, + { + "epoch": 2.761232349165597, + "grad_norm": 0.5710389018058777, + "learning_rate": 1.1557749986033458e-05, + "loss": 2.0632, + "step": 2153 + }, + { + "epoch": 2.7625160462130935, + "grad_norm": 0.5559154152870178, + "learning_rate": 1.1535800960977397e-05, + "loss": 1.9682, + "step": 2154 + }, + { + "epoch": 2.7637997432605905, + "grad_norm": 0.5308806896209717, + "learning_rate": 1.1513866544822622e-05, + "loss": 2.0068, + "step": 2155 + }, + { + "epoch": 2.7650834403080875, + "grad_norm": 0.5691455602645874, + "learning_rate": 1.1491946761368397e-05, + "loss": 2.0152, + "step": 2156 + }, + { + "epoch": 2.766367137355584, + "grad_norm": 0.5431718230247498, + "learning_rate": 1.1470041634398063e-05, + "loss": 2.019, + "step": 2157 + }, + { + "epoch": 2.767650834403081, + "grad_norm": 0.5399120450019836, + "learning_rate": 1.1448151187679095e-05, + "loss": 1.9487, + "step": 2158 + }, + { + "epoch": 2.7689345314505776, + "grad_norm": 0.5186573266983032, + "learning_rate": 1.1426275444963034e-05, + "loss": 2.0029, + "step": 2159 + }, + { + "epoch": 2.7702182284980745, + "grad_norm": 0.5251243114471436, + "learning_rate": 1.1404414429985453e-05, + "loss": 1.9741, + "step": 2160 + }, + { + "epoch": 2.7715019255455715, + "grad_norm": 0.5473252534866333, + "learning_rate": 1.1382568166465938e-05, + "loss": 2.0089, + "step": 2161 + }, + { + "epoch": 2.772785622593068, + "grad_norm": 0.5848075151443481, + "learning_rate": 1.1360736678108092e-05, + "loss": 1.8792, + "step": 2162 + }, + { + "epoch": 2.7740693196405646, + "grad_norm": 0.5233170986175537, + "learning_rate": 1.1338919988599491e-05, + "loss": 2.0176, + "step": 2163 + }, + { + "epoch": 2.7753530166880616, + "grad_norm": 0.530375599861145, + "learning_rate": 1.131711812161163e-05, + "loss": 2.027, + "step": 2164 + }, + { + "epoch": 2.7766367137355585, + "grad_norm": 0.5231175422668457, + "learning_rate": 1.1295331100799924e-05, + "loss": 2.0015, + "step": 2165 + }, + { + "epoch": 2.777920410783055, + "grad_norm": 0.5353492498397827, + "learning_rate": 1.1273558949803704e-05, + "loss": 2.0185, + "step": 2166 + }, + { + "epoch": 2.779204107830552, + "grad_norm": 0.5592543482780457, + "learning_rate": 1.125180169224613e-05, + "loss": 1.9528, + "step": 2167 + }, + { + "epoch": 2.7804878048780486, + "grad_norm": 0.5467972159385681, + "learning_rate": 1.1230059351734237e-05, + "loss": 2.0546, + "step": 2168 + }, + { + "epoch": 2.7817715019255456, + "grad_norm": 0.5115427374839783, + "learning_rate": 1.1208331951858867e-05, + "loss": 1.8974, + "step": 2169 + }, + { + "epoch": 2.7830551989730425, + "grad_norm": 0.5372148156166077, + "learning_rate": 1.1186619516194635e-05, + "loss": 2.0312, + "step": 2170 + }, + { + "epoch": 2.784338896020539, + "grad_norm": 0.5295898914337158, + "learning_rate": 1.1164922068299925e-05, + "loss": 1.939, + "step": 2171 + }, + { + "epoch": 2.785622593068036, + "grad_norm": 0.5715629458427429, + "learning_rate": 1.1143239631716864e-05, + "loss": 2.0801, + "step": 2172 + }, + { + "epoch": 2.7869062901155326, + "grad_norm": 0.5782803297042847, + "learning_rate": 1.1121572229971306e-05, + "loss": 2.0856, + "step": 2173 + }, + { + "epoch": 2.7881899871630296, + "grad_norm": 0.5791909694671631, + "learning_rate": 1.1099919886572757e-05, + "loss": 2.0798, + "step": 2174 + }, + { + "epoch": 2.7894736842105265, + "grad_norm": 0.5609378814697266, + "learning_rate": 1.1078282625014422e-05, + "loss": 2.0374, + "step": 2175 + }, + { + "epoch": 2.790757381258023, + "grad_norm": 0.5883268713951111, + "learning_rate": 1.1056660468773108e-05, + "loss": 1.9858, + "step": 2176 + }, + { + "epoch": 2.79204107830552, + "grad_norm": 0.5559226870536804, + "learning_rate": 1.1035053441309253e-05, + "loss": 2.0092, + "step": 2177 + }, + { + "epoch": 2.7933247753530166, + "grad_norm": 0.6033201217651367, + "learning_rate": 1.1013461566066891e-05, + "loss": 1.9603, + "step": 2178 + }, + { + "epoch": 2.7946084724005136, + "grad_norm": 0.5245801210403442, + "learning_rate": 1.0991884866473587e-05, + "loss": 1.9618, + "step": 2179 + }, + { + "epoch": 2.79589216944801, + "grad_norm": 0.5792044401168823, + "learning_rate": 1.0970323365940444e-05, + "loss": 1.9944, + "step": 2180 + }, + { + "epoch": 2.797175866495507, + "grad_norm": 0.5169621706008911, + "learning_rate": 1.0948777087862092e-05, + "loss": 1.9627, + "step": 2181 + }, + { + "epoch": 2.7984595635430036, + "grad_norm": 0.5303113460540771, + "learning_rate": 1.0927246055616644e-05, + "loss": 2.0322, + "step": 2182 + }, + { + "epoch": 2.7997432605905006, + "grad_norm": 0.517459511756897, + "learning_rate": 1.0905730292565648e-05, + "loss": 1.9606, + "step": 2183 + }, + { + "epoch": 2.8010269576379976, + "grad_norm": 0.5592575669288635, + "learning_rate": 1.0884229822054113e-05, + "loss": 2.027, + "step": 2184 + }, + { + "epoch": 2.802310654685494, + "grad_norm": 0.5244090557098389, + "learning_rate": 1.0862744667410433e-05, + "loss": 2.0548, + "step": 2185 + }, + { + "epoch": 2.803594351732991, + "grad_norm": 0.5493913888931274, + "learning_rate": 1.0841274851946382e-05, + "loss": 1.9715, + "step": 2186 + }, + { + "epoch": 2.8048780487804876, + "grad_norm": 0.5771494507789612, + "learning_rate": 1.0819820398957115e-05, + "loss": 1.9779, + "step": 2187 + }, + { + "epoch": 2.8061617458279846, + "grad_norm": 0.5659067630767822, + "learning_rate": 1.0798381331721109e-05, + "loss": 1.9512, + "step": 2188 + }, + { + "epoch": 2.8074454428754816, + "grad_norm": 0.5131163001060486, + "learning_rate": 1.0776957673500134e-05, + "loss": 2.1071, + "step": 2189 + }, + { + "epoch": 2.808729139922978, + "grad_norm": 0.5305460691452026, + "learning_rate": 1.0755549447539242e-05, + "loss": 1.888, + "step": 2190 + }, + { + "epoch": 2.810012836970475, + "grad_norm": 0.5406280755996704, + "learning_rate": 1.073415667706676e-05, + "loss": 2.0052, + "step": 2191 + }, + { + "epoch": 2.8112965340179716, + "grad_norm": 0.5091846585273743, + "learning_rate": 1.0712779385294237e-05, + "loss": 1.9473, + "step": 2192 + }, + { + "epoch": 2.8125802310654686, + "grad_norm": 0.5482229590415955, + "learning_rate": 1.0691417595416407e-05, + "loss": 2.0175, + "step": 2193 + }, + { + "epoch": 2.8138639281129656, + "grad_norm": 0.6116776466369629, + "learning_rate": 1.067007133061122e-05, + "loss": 1.9647, + "step": 2194 + }, + { + "epoch": 2.815147625160462, + "grad_norm": 0.5484519600868225, + "learning_rate": 1.0648740614039748e-05, + "loss": 2.0733, + "step": 2195 + }, + { + "epoch": 2.8164313222079587, + "grad_norm": 0.5696253180503845, + "learning_rate": 1.062742546884621e-05, + "loss": 2.0131, + "step": 2196 + }, + { + "epoch": 2.8177150192554556, + "grad_norm": 0.5473034381866455, + "learning_rate": 1.0606125918157945e-05, + "loss": 1.91, + "step": 2197 + }, + { + "epoch": 2.8189987163029526, + "grad_norm": 0.5514528751373291, + "learning_rate": 1.0584841985085334e-05, + "loss": 1.9664, + "step": 2198 + }, + { + "epoch": 2.820282413350449, + "grad_norm": 0.5756730437278748, + "learning_rate": 1.056357369272183e-05, + "loss": 2.0219, + "step": 2199 + }, + { + "epoch": 2.821566110397946, + "grad_norm": 0.5806208252906799, + "learning_rate": 1.0542321064143931e-05, + "loss": 2.0091, + "step": 2200 + }, + { + "epoch": 2.8228498074454427, + "grad_norm": 0.5774882435798645, + "learning_rate": 1.0521084122411112e-05, + "loss": 1.9359, + "step": 2201 + }, + { + "epoch": 2.8241335044929397, + "grad_norm": 0.5263341665267944, + "learning_rate": 1.0499862890565853e-05, + "loss": 1.9829, + "step": 2202 + }, + { + "epoch": 2.8254172015404366, + "grad_norm": 0.5697647333145142, + "learning_rate": 1.0478657391633582e-05, + "loss": 1.9704, + "step": 2203 + }, + { + "epoch": 2.826700898587933, + "grad_norm": 0.5692322254180908, + "learning_rate": 1.0457467648622641e-05, + "loss": 2.0297, + "step": 2204 + }, + { + "epoch": 2.82798459563543, + "grad_norm": 0.6035948395729065, + "learning_rate": 1.0436293684524282e-05, + "loss": 2.0283, + "step": 2205 + }, + { + "epoch": 2.8292682926829267, + "grad_norm": 0.5711615681648254, + "learning_rate": 1.041513552231265e-05, + "loss": 2.0182, + "step": 2206 + }, + { + "epoch": 2.8305519897304237, + "grad_norm": 0.543689489364624, + "learning_rate": 1.0393993184944748e-05, + "loss": 2.0095, + "step": 2207 + }, + { + "epoch": 2.8318356867779206, + "grad_norm": 0.5345496535301208, + "learning_rate": 1.0372866695360372e-05, + "loss": 1.9479, + "step": 2208 + }, + { + "epoch": 2.833119383825417, + "grad_norm": 0.5590696930885315, + "learning_rate": 1.0351756076482175e-05, + "loss": 2.0428, + "step": 2209 + }, + { + "epoch": 2.834403080872914, + "grad_norm": 0.5923562049865723, + "learning_rate": 1.033066135121554e-05, + "loss": 1.9486, + "step": 2210 + }, + { + "epoch": 2.8356867779204107, + "grad_norm": 0.5892860889434814, + "learning_rate": 1.0309582542448637e-05, + "loss": 1.9791, + "step": 2211 + }, + { + "epoch": 2.8369704749679077, + "grad_norm": 0.5479671359062195, + "learning_rate": 1.0288519673052368e-05, + "loss": 1.952, + "step": 2212 + }, + { + "epoch": 2.8382541720154046, + "grad_norm": 0.5796504616737366, + "learning_rate": 1.0267472765880323e-05, + "loss": 2.0114, + "step": 2213 + }, + { + "epoch": 2.839537869062901, + "grad_norm": 0.6362434029579163, + "learning_rate": 1.0246441843768775e-05, + "loss": 1.8688, + "step": 2214 + }, + { + "epoch": 2.8408215661103977, + "grad_norm": 0.568681001663208, + "learning_rate": 1.0225426929536645e-05, + "loss": 2.0625, + "step": 2215 + }, + { + "epoch": 2.8421052631578947, + "grad_norm": 0.5663188099861145, + "learning_rate": 1.020442804598553e-05, + "loss": 2.0779, + "step": 2216 + }, + { + "epoch": 2.8433889602053917, + "grad_norm": 0.5323242545127869, + "learning_rate": 1.0183445215899584e-05, + "loss": 1.9739, + "step": 2217 + }, + { + "epoch": 2.844672657252888, + "grad_norm": 0.531231164932251, + "learning_rate": 1.0162478462045556e-05, + "loss": 1.9537, + "step": 2218 + }, + { + "epoch": 2.845956354300385, + "grad_norm": 0.5439385175704956, + "learning_rate": 1.0141527807172766e-05, + "loss": 2.0185, + "step": 2219 + }, + { + "epoch": 2.8472400513478817, + "grad_norm": 0.5406712293624878, + "learning_rate": 1.0120593274013041e-05, + "loss": 1.9912, + "step": 2220 + }, + { + "epoch": 2.8485237483953787, + "grad_norm": 0.5825367569923401, + "learning_rate": 1.0099674885280743e-05, + "loss": 1.8941, + "step": 2221 + }, + { + "epoch": 2.8498074454428757, + "grad_norm": 0.5363432168960571, + "learning_rate": 1.007877266367271e-05, + "loss": 2.0607, + "step": 2222 + }, + { + "epoch": 2.851091142490372, + "grad_norm": 0.5603116750717163, + "learning_rate": 1.0057886631868228e-05, + "loss": 2.025, + "step": 2223 + }, + { + "epoch": 2.852374839537869, + "grad_norm": 0.5555363297462463, + "learning_rate": 1.003701681252901e-05, + "loss": 2.0342, + "step": 2224 + }, + { + "epoch": 2.8536585365853657, + "grad_norm": 0.5991105437278748, + "learning_rate": 1.0016163228299202e-05, + "loss": 2.0278, + "step": 2225 + }, + { + "epoch": 2.8549422336328627, + "grad_norm": 0.5510726571083069, + "learning_rate": 9.99532590180533e-06, + "loss": 2.0102, + "step": 2226 + }, + { + "epoch": 2.8562259306803597, + "grad_norm": 0.5544465184211731, + "learning_rate": 9.974504855656253e-06, + "loss": 1.9193, + "step": 2227 + }, + { + "epoch": 2.857509627727856, + "grad_norm": 0.5411550998687744, + "learning_rate": 9.95370011244321e-06, + "loss": 2.0158, + "step": 2228 + }, + { + "epoch": 2.8587933247753528, + "grad_norm": 0.5146743655204773, + "learning_rate": 9.932911694739699e-06, + "loss": 2.0206, + "step": 2229 + }, + { + "epoch": 2.8600770218228497, + "grad_norm": 0.5088391304016113, + "learning_rate": 9.91213962510155e-06, + "loss": 1.9614, + "step": 2230 + }, + { + "epoch": 2.8613607188703467, + "grad_norm": 0.5583186149597168, + "learning_rate": 9.89138392606684e-06, + "loss": 1.9137, + "step": 2231 + }, + { + "epoch": 2.8626444159178432, + "grad_norm": 0.5473942756652832, + "learning_rate": 9.870644620155877e-06, + "loss": 1.9914, + "step": 2232 + }, + { + "epoch": 2.8639281129653402, + "grad_norm": 0.5835011005401611, + "learning_rate": 9.849921729871173e-06, + "loss": 2.0088, + "step": 2233 + }, + { + "epoch": 2.8652118100128368, + "grad_norm": 0.5293422937393188, + "learning_rate": 9.829215277697456e-06, + "loss": 1.9479, + "step": 2234 + }, + { + "epoch": 2.8664955070603337, + "grad_norm": 0.585949182510376, + "learning_rate": 9.808525286101596e-06, + "loss": 2.0021, + "step": 2235 + }, + { + "epoch": 2.8677792041078307, + "grad_norm": 0.5617350935935974, + "learning_rate": 9.78785177753261e-06, + "loss": 2.0325, + "step": 2236 + }, + { + "epoch": 2.8690629011553272, + "grad_norm": 0.5657144784927368, + "learning_rate": 9.767194774421648e-06, + "loss": 2.0583, + "step": 2237 + }, + { + "epoch": 2.8703465982028242, + "grad_norm": 0.5477186441421509, + "learning_rate": 9.74655429918192e-06, + "loss": 1.963, + "step": 2238 + }, + { + "epoch": 2.8716302952503208, + "grad_norm": 0.6014096736907959, + "learning_rate": 9.725930374208714e-06, + "loss": 2.0775, + "step": 2239 + }, + { + "epoch": 2.8729139922978177, + "grad_norm": 0.5369572639465332, + "learning_rate": 9.705323021879367e-06, + "loss": 2.032, + "step": 2240 + }, + { + "epoch": 2.8741976893453147, + "grad_norm": 0.5372179746627808, + "learning_rate": 9.684732264553248e-06, + "loss": 1.9709, + "step": 2241 + }, + { + "epoch": 2.8754813863928113, + "grad_norm": 0.5568748116493225, + "learning_rate": 9.664158124571687e-06, + "loss": 1.9854, + "step": 2242 + }, + { + "epoch": 2.8767650834403082, + "grad_norm": 0.5778310298919678, + "learning_rate": 9.643600624257999e-06, + "loss": 2.0342, + "step": 2243 + }, + { + "epoch": 2.8780487804878048, + "grad_norm": 0.5486297607421875, + "learning_rate": 9.623059785917452e-06, + "loss": 1.9581, + "step": 2244 + }, + { + "epoch": 2.8793324775353017, + "grad_norm": 0.5335022807121277, + "learning_rate": 9.60253563183724e-06, + "loss": 2.0121, + "step": 2245 + }, + { + "epoch": 2.8806161745827987, + "grad_norm": 0.607744038105011, + "learning_rate": 9.582028184286423e-06, + "loss": 1.9791, + "step": 2246 + }, + { + "epoch": 2.8818998716302953, + "grad_norm": 0.5861355662345886, + "learning_rate": 9.561537465515977e-06, + "loss": 2.0306, + "step": 2247 + }, + { + "epoch": 2.883183568677792, + "grad_norm": 0.5377950072288513, + "learning_rate": 9.541063497758687e-06, + "loss": 2.0159, + "step": 2248 + }, + { + "epoch": 2.8844672657252888, + "grad_norm": 0.5426115393638611, + "learning_rate": 9.52060630322919e-06, + "loss": 1.9723, + "step": 2249 + }, + { + "epoch": 2.8857509627727858, + "grad_norm": 0.5505149960517883, + "learning_rate": 9.50016590412392e-06, + "loss": 1.9265, + "step": 2250 + }, + { + "epoch": 2.8870346598202823, + "grad_norm": 0.5713925957679749, + "learning_rate": 9.479742322621077e-06, + "loss": 2.0133, + "step": 2251 + }, + { + "epoch": 2.8883183568677793, + "grad_norm": 0.5400781035423279, + "learning_rate": 9.459335580880613e-06, + "loss": 2.0647, + "step": 2252 + }, + { + "epoch": 2.889602053915276, + "grad_norm": 0.5298085808753967, + "learning_rate": 9.438945701044225e-06, + "loss": 2.0561, + "step": 2253 + }, + { + "epoch": 2.890885750962773, + "grad_norm": 0.5464864373207092, + "learning_rate": 9.418572705235293e-06, + "loss": 1.9999, + "step": 2254 + }, + { + "epoch": 2.8921694480102698, + "grad_norm": 0.5368531346321106, + "learning_rate": 9.398216615558892e-06, + "loss": 1.9648, + "step": 2255 + }, + { + "epoch": 2.8934531450577663, + "grad_norm": 0.5257331728935242, + "learning_rate": 9.377877454101764e-06, + "loss": 2.0371, + "step": 2256 + }, + { + "epoch": 2.8947368421052633, + "grad_norm": 0.617135226726532, + "learning_rate": 9.357555242932258e-06, + "loss": 1.9605, + "step": 2257 + }, + { + "epoch": 2.89602053915276, + "grad_norm": 0.619138240814209, + "learning_rate": 9.337250004100337e-06, + "loss": 2.0167, + "step": 2258 + }, + { + "epoch": 2.897304236200257, + "grad_norm": 0.5348044633865356, + "learning_rate": 9.316961759637562e-06, + "loss": 2.0107, + "step": 2259 + }, + { + "epoch": 2.8985879332477538, + "grad_norm": 0.5555276870727539, + "learning_rate": 9.296690531557057e-06, + "loss": 1.9261, + "step": 2260 + }, + { + "epoch": 2.8998716302952503, + "grad_norm": 0.5685930252075195, + "learning_rate": 9.276436341853456e-06, + "loss": 1.917, + "step": 2261 + }, + { + "epoch": 2.901155327342747, + "grad_norm": 0.553909420967102, + "learning_rate": 9.256199212502945e-06, + "loss": 2.0362, + "step": 2262 + }, + { + "epoch": 2.902439024390244, + "grad_norm": 0.5612945556640625, + "learning_rate": 9.235979165463155e-06, + "loss": 2.0065, + "step": 2263 + }, + { + "epoch": 2.903722721437741, + "grad_norm": 0.5699390769004822, + "learning_rate": 9.215776222673217e-06, + "loss": 1.9901, + "step": 2264 + }, + { + "epoch": 2.9050064184852373, + "grad_norm": 0.5504017472267151, + "learning_rate": 9.1955904060537e-06, + "loss": 1.9568, + "step": 2265 + }, + { + "epoch": 2.9062901155327343, + "grad_norm": 0.5116189122200012, + "learning_rate": 9.175421737506576e-06, + "loss": 2.0352, + "step": 2266 + }, + { + "epoch": 2.907573812580231, + "grad_norm": 0.5599797964096069, + "learning_rate": 9.155270238915206e-06, + "loss": 1.9668, + "step": 2267 + }, + { + "epoch": 2.908857509627728, + "grad_norm": 0.5884008407592773, + "learning_rate": 9.135135932144345e-06, + "loss": 1.9583, + "step": 2268 + }, + { + "epoch": 2.910141206675225, + "grad_norm": 0.5584822297096252, + "learning_rate": 9.115018839040091e-06, + "loss": 1.9956, + "step": 2269 + }, + { + "epoch": 2.9114249037227213, + "grad_norm": 0.5694496035575867, + "learning_rate": 9.094918981429845e-06, + "loss": 2.0416, + "step": 2270 + }, + { + "epoch": 2.9127086007702183, + "grad_norm": 0.5058184862136841, + "learning_rate": 9.074836381122312e-06, + "loss": 1.9496, + "step": 2271 + }, + { + "epoch": 2.913992297817715, + "grad_norm": 0.5590035915374756, + "learning_rate": 9.054771059907499e-06, + "loss": 2.0507, + "step": 2272 + }, + { + "epoch": 2.915275994865212, + "grad_norm": 0.6113227605819702, + "learning_rate": 9.034723039556625e-06, + "loss": 1.993, + "step": 2273 + }, + { + "epoch": 2.916559691912709, + "grad_norm": 0.5603733658790588, + "learning_rate": 9.014692341822165e-06, + "loss": 1.9346, + "step": 2274 + }, + { + "epoch": 2.9178433889602053, + "grad_norm": 0.5387945771217346, + "learning_rate": 8.994678988437802e-06, + "loss": 2.0688, + "step": 2275 + }, + { + "epoch": 2.9191270860077023, + "grad_norm": 0.5965076684951782, + "learning_rate": 8.97468300111838e-06, + "loss": 2.0145, + "step": 2276 + }, + { + "epoch": 2.920410783055199, + "grad_norm": 0.5389226675033569, + "learning_rate": 8.954704401559903e-06, + "loss": 2.0234, + "step": 2277 + }, + { + "epoch": 2.921694480102696, + "grad_norm": 0.5928203463554382, + "learning_rate": 8.934743211439517e-06, + "loss": 2.0118, + "step": 2278 + }, + { + "epoch": 2.922978177150193, + "grad_norm": 0.5490136742591858, + "learning_rate": 8.914799452415495e-06, + "loss": 2.0501, + "step": 2279 + }, + { + "epoch": 2.9242618741976893, + "grad_norm": 0.5735679268836975, + "learning_rate": 8.894873146127156e-06, + "loss": 1.9796, + "step": 2280 + }, + { + "epoch": 2.925545571245186, + "grad_norm": 0.5448455810546875, + "learning_rate": 8.874964314194922e-06, + "loss": 1.9267, + "step": 2281 + }, + { + "epoch": 2.926829268292683, + "grad_norm": 0.5573774576187134, + "learning_rate": 8.855072978220225e-06, + "loss": 1.9774, + "step": 2282 + }, + { + "epoch": 2.92811296534018, + "grad_norm": 0.5716160535812378, + "learning_rate": 8.835199159785532e-06, + "loss": 2.0069, + "step": 2283 + }, + { + "epoch": 2.9293966623876764, + "grad_norm": 0.5464572310447693, + "learning_rate": 8.815342880454311e-06, + "loss": 1.9789, + "step": 2284 + }, + { + "epoch": 2.9306803594351734, + "grad_norm": 0.5620017647743225, + "learning_rate": 8.795504161770974e-06, + "loss": 1.9971, + "step": 2285 + }, + { + "epoch": 2.93196405648267, + "grad_norm": 0.5787877440452576, + "learning_rate": 8.775683025260884e-06, + "loss": 1.9834, + "step": 2286 + }, + { + "epoch": 2.933247753530167, + "grad_norm": 0.5636091232299805, + "learning_rate": 8.755879492430355e-06, + "loss": 1.9932, + "step": 2287 + }, + { + "epoch": 2.934531450577664, + "grad_norm": 0.5889118313789368, + "learning_rate": 8.736093584766565e-06, + "loss": 2.0426, + "step": 2288 + }, + { + "epoch": 2.9358151476251604, + "grad_norm": 0.5796907544136047, + "learning_rate": 8.716325323737586e-06, + "loss": 1.9457, + "step": 2289 + }, + { + "epoch": 2.9370988446726574, + "grad_norm": 0.5304666757583618, + "learning_rate": 8.696574730792359e-06, + "loss": 1.9554, + "step": 2290 + }, + { + "epoch": 2.938382541720154, + "grad_norm": 0.5568850636482239, + "learning_rate": 8.67684182736062e-06, + "loss": 2.0547, + "step": 2291 + }, + { + "epoch": 2.939666238767651, + "grad_norm": 0.5534605979919434, + "learning_rate": 8.657126634852927e-06, + "loss": 1.9254, + "step": 2292 + }, + { + "epoch": 2.940949935815148, + "grad_norm": 0.5808989405632019, + "learning_rate": 8.63742917466063e-06, + "loss": 1.9938, + "step": 2293 + }, + { + "epoch": 2.9422336328626444, + "grad_norm": 0.5773752331733704, + "learning_rate": 8.61774946815584e-06, + "loss": 2.053, + "step": 2294 + }, + { + "epoch": 2.9435173299101414, + "grad_norm": 0.5459012985229492, + "learning_rate": 8.598087536691388e-06, + "loss": 2.0245, + "step": 2295 + }, + { + "epoch": 2.944801026957638, + "grad_norm": 0.5533496141433716, + "learning_rate": 8.57844340160082e-06, + "loss": 1.9366, + "step": 2296 + }, + { + "epoch": 2.946084724005135, + "grad_norm": 0.6131767630577087, + "learning_rate": 8.558817084198387e-06, + "loss": 1.9726, + "step": 2297 + }, + { + "epoch": 2.9473684210526314, + "grad_norm": 0.5860012769699097, + "learning_rate": 8.539208605779012e-06, + "loss": 2.0315, + "step": 2298 + }, + { + "epoch": 2.9486521181001284, + "grad_norm": 0.5688477158546448, + "learning_rate": 8.519617987618233e-06, + "loss": 1.9455, + "step": 2299 + }, + { + "epoch": 2.949935815147625, + "grad_norm": 0.5576227307319641, + "learning_rate": 8.50004525097225e-06, + "loss": 1.9789, + "step": 2300 + }, + { + "epoch": 2.951219512195122, + "grad_norm": 0.5461267232894897, + "learning_rate": 8.480490417077821e-06, + "loss": 1.9631, + "step": 2301 + }, + { + "epoch": 2.952503209242619, + "grad_norm": 0.5608059167861938, + "learning_rate": 8.460953507152285e-06, + "loss": 2.0173, + "step": 2302 + }, + { + "epoch": 2.9537869062901154, + "grad_norm": 0.5251237154006958, + "learning_rate": 8.441434542393584e-06, + "loss": 1.9865, + "step": 2303 + }, + { + "epoch": 2.9550706033376124, + "grad_norm": 0.5265477895736694, + "learning_rate": 8.421933543980126e-06, + "loss": 1.9613, + "step": 2304 + }, + { + "epoch": 2.956354300385109, + "grad_norm": 0.5845303535461426, + "learning_rate": 8.402450533070844e-06, + "loss": 1.9792, + "step": 2305 + }, + { + "epoch": 2.957637997432606, + "grad_norm": 0.5610324740409851, + "learning_rate": 8.382985530805177e-06, + "loss": 2.0408, + "step": 2306 + }, + { + "epoch": 2.958921694480103, + "grad_norm": 0.6044159531593323, + "learning_rate": 8.363538558302988e-06, + "loss": 2.0257, + "step": 2307 + }, + { + "epoch": 2.9602053915275994, + "grad_norm": 0.5548937320709229, + "learning_rate": 8.344109636664607e-06, + "loss": 1.9789, + "step": 2308 + }, + { + "epoch": 2.9614890885750964, + "grad_norm": 0.5614864230155945, + "learning_rate": 8.324698786970777e-06, + "loss": 2.0189, + "step": 2309 + }, + { + "epoch": 2.962772785622593, + "grad_norm": 0.5422723293304443, + "learning_rate": 8.305306030282617e-06, + "loss": 2.0011, + "step": 2310 + }, + { + "epoch": 2.96405648267009, + "grad_norm": 0.5655319094657898, + "learning_rate": 8.285931387641607e-06, + "loss": 1.9574, + "step": 2311 + }, + { + "epoch": 2.965340179717587, + "grad_norm": 0.5774760842323303, + "learning_rate": 8.266574880069603e-06, + "loss": 1.9298, + "step": 2312 + }, + { + "epoch": 2.9666238767650834, + "grad_norm": 0.6439433097839355, + "learning_rate": 8.247236528568769e-06, + "loss": 1.9681, + "step": 2313 + }, + { + "epoch": 2.96790757381258, + "grad_norm": 0.5599347352981567, + "learning_rate": 8.227916354121556e-06, + "loss": 2.0815, + "step": 2314 + }, + { + "epoch": 2.969191270860077, + "grad_norm": 0.5461101531982422, + "learning_rate": 8.208614377690724e-06, + "loss": 1.9781, + "step": 2315 + }, + { + "epoch": 2.970474967907574, + "grad_norm": 0.5638664364814758, + "learning_rate": 8.189330620219249e-06, + "loss": 1.9816, + "step": 2316 + }, + { + "epoch": 2.9717586649550705, + "grad_norm": 0.5713538527488708, + "learning_rate": 8.17006510263037e-06, + "loss": 1.9498, + "step": 2317 + }, + { + "epoch": 2.9730423620025674, + "grad_norm": 0.5505553483963013, + "learning_rate": 8.150817845827533e-06, + "loss": 2.0033, + "step": 2318 + }, + { + "epoch": 2.974326059050064, + "grad_norm": 0.5832675695419312, + "learning_rate": 8.131588870694351e-06, + "loss": 1.9819, + "step": 2319 + }, + { + "epoch": 2.975609756097561, + "grad_norm": 0.5435935854911804, + "learning_rate": 8.112378198094608e-06, + "loss": 2.018, + "step": 2320 + }, + { + "epoch": 2.976893453145058, + "grad_norm": 0.5463137030601501, + "learning_rate": 8.093185848872248e-06, + "loss": 2.0486, + "step": 2321 + }, + { + "epoch": 2.9781771501925545, + "grad_norm": 0.5777429342269897, + "learning_rate": 8.074011843851301e-06, + "loss": 2.009, + "step": 2322 + }, + { + "epoch": 2.9794608472400514, + "grad_norm": 0.5732576251029968, + "learning_rate": 8.054856203835934e-06, + "loss": 1.9873, + "step": 2323 + }, + { + "epoch": 2.980744544287548, + "grad_norm": 0.5592786073684692, + "learning_rate": 8.035718949610344e-06, + "loss": 2.109, + "step": 2324 + }, + { + "epoch": 2.982028241335045, + "grad_norm": 0.5253878235816956, + "learning_rate": 8.01660010193882e-06, + "loss": 1.9537, + "step": 2325 + }, + { + "epoch": 2.983311938382542, + "grad_norm": 0.5908203125, + "learning_rate": 7.997499681565644e-06, + "loss": 2.0497, + "step": 2326 + }, + { + "epoch": 2.9845956354300385, + "grad_norm": 0.5928517580032349, + "learning_rate": 7.978417709215125e-06, + "loss": 1.9622, + "step": 2327 + }, + { + "epoch": 2.9858793324775355, + "grad_norm": 0.5829191207885742, + "learning_rate": 7.95935420559156e-06, + "loss": 1.9907, + "step": 2328 + }, + { + "epoch": 2.987163029525032, + "grad_norm": 0.5706354379653931, + "learning_rate": 7.940309191379192e-06, + "loss": 1.9131, + "step": 2329 + }, + { + "epoch": 2.988446726572529, + "grad_norm": 0.5242451429367065, + "learning_rate": 7.921282687242198e-06, + "loss": 1.9358, + "step": 2330 + }, + { + "epoch": 2.989730423620026, + "grad_norm": 0.5542780160903931, + "learning_rate": 7.90227471382469e-06, + "loss": 1.9634, + "step": 2331 + }, + { + "epoch": 2.9910141206675225, + "grad_norm": 0.556493878364563, + "learning_rate": 7.883285291750675e-06, + "loss": 2.004, + "step": 2332 + }, + { + "epoch": 2.992297817715019, + "grad_norm": 0.5395458340644836, + "learning_rate": 7.864314441624004e-06, + "loss": 2.0374, + "step": 2333 + }, + { + "epoch": 2.993581514762516, + "grad_norm": 0.5832502245903015, + "learning_rate": 7.845362184028415e-06, + "loss": 1.9995, + "step": 2334 + }, + { + "epoch": 2.994865211810013, + "grad_norm": 0.5629451870918274, + "learning_rate": 7.82642853952744e-06, + "loss": 2.0273, + "step": 2335 + }, + { + "epoch": 2.9961489088575095, + "grad_norm": 0.5681702494621277, + "learning_rate": 7.807513528664414e-06, + "loss": 2.0371, + "step": 2336 + }, + { + "epoch": 2.9974326059050065, + "grad_norm": 0.5312502980232239, + "learning_rate": 7.788617171962501e-06, + "loss": 1.9725, + "step": 2337 + }, + { + "epoch": 2.998716302952503, + "grad_norm": 0.5660899877548218, + "learning_rate": 7.769739489924576e-06, + "loss": 1.9232, + "step": 2338 + }, + { + "epoch": 3.0, + "grad_norm": 0.5706592202186584, + "learning_rate": 7.75088050303326e-06, + "loss": 1.95, + "step": 2339 + }, + { + "epoch": 3.001283697047497, + "grad_norm": 0.5633547306060791, + "learning_rate": 7.732040231750914e-06, + "loss": 2.0436, + "step": 2340 + }, + { + "epoch": 3.001283697047497, + "grad_norm": 0.5632837414741516, + "learning_rate": 7.713218696519558e-06, + "loss": 1.8721, + "step": 2341 + }, + { + "epoch": 3.0025673940949935, + "grad_norm": 0.5506707429885864, + "learning_rate": 7.694415917760911e-06, + "loss": 1.9008, + "step": 2342 + }, + { + "epoch": 3.0038510911424905, + "grad_norm": 0.5691422820091248, + "learning_rate": 7.675631915876336e-06, + "loss": 1.8694, + "step": 2343 + }, + { + "epoch": 3.005134788189987, + "grad_norm": 0.5596054792404175, + "learning_rate": 7.656866711246813e-06, + "loss": 1.8413, + "step": 2344 + }, + { + "epoch": 3.006418485237484, + "grad_norm": 0.5364600419998169, + "learning_rate": 7.638120324232916e-06, + "loss": 1.8086, + "step": 2345 + }, + { + "epoch": 3.0077021822849805, + "grad_norm": 0.5419870018959045, + "learning_rate": 7.619392775174833e-06, + "loss": 1.8149, + "step": 2346 + }, + { + "epoch": 3.0089858793324775, + "grad_norm": 0.5442661046981812, + "learning_rate": 7.600684084392301e-06, + "loss": 1.767, + "step": 2347 + }, + { + "epoch": 3.0102695763799745, + "grad_norm": 0.5562167167663574, + "learning_rate": 7.58199427218457e-06, + "loss": 1.8505, + "step": 2348 + }, + { + "epoch": 3.011553273427471, + "grad_norm": 0.5566357970237732, + "learning_rate": 7.563323358830448e-06, + "loss": 1.8256, + "step": 2349 + }, + { + "epoch": 3.012836970474968, + "grad_norm": 0.6152045726776123, + "learning_rate": 7.544671364588199e-06, + "loss": 1.8068, + "step": 2350 + }, + { + "epoch": 3.0141206675224645, + "grad_norm": 0.5993888974189758, + "learning_rate": 7.526038309695593e-06, + "loss": 1.805, + "step": 2351 + }, + { + "epoch": 3.0154043645699615, + "grad_norm": 0.6754676699638367, + "learning_rate": 7.507424214369816e-06, + "loss": 1.8158, + "step": 2352 + }, + { + "epoch": 3.016688061617458, + "grad_norm": 0.6105367541313171, + "learning_rate": 7.488829098807518e-06, + "loss": 1.8335, + "step": 2353 + }, + { + "epoch": 3.017971758664955, + "grad_norm": 0.6178943514823914, + "learning_rate": 7.470252983184728e-06, + "loss": 1.8273, + "step": 2354 + }, + { + "epoch": 3.019255455712452, + "grad_norm": 0.5996940732002258, + "learning_rate": 7.451695887656865e-06, + "loss": 1.86, + "step": 2355 + }, + { + "epoch": 3.0205391527599486, + "grad_norm": 0.6031941771507263, + "learning_rate": 7.433157832358725e-06, + "loss": 1.9306, + "step": 2356 + }, + { + "epoch": 3.0218228498074455, + "grad_norm": 0.5989313721656799, + "learning_rate": 7.414638837404439e-06, + "loss": 1.7812, + "step": 2357 + }, + { + "epoch": 3.023106546854942, + "grad_norm": 0.6435003280639648, + "learning_rate": 7.396138922887441e-06, + "loss": 1.8923, + "step": 2358 + }, + { + "epoch": 3.024390243902439, + "grad_norm": 0.6496562957763672, + "learning_rate": 7.377658108880492e-06, + "loss": 1.9311, + "step": 2359 + }, + { + "epoch": 3.025673940949936, + "grad_norm": 0.6271670460700989, + "learning_rate": 7.359196415435596e-06, + "loss": 1.8336, + "step": 2360 + }, + { + "epoch": 3.0269576379974326, + "grad_norm": 0.6202157139778137, + "learning_rate": 7.340753862584029e-06, + "loss": 1.8547, + "step": 2361 + }, + { + "epoch": 3.0282413350449295, + "grad_norm": 0.6452667713165283, + "learning_rate": 7.3223304703363135e-06, + "loss": 1.8977, + "step": 2362 + }, + { + "epoch": 3.029525032092426, + "grad_norm": 0.6381579637527466, + "learning_rate": 7.3039262586821536e-06, + "loss": 1.7404, + "step": 2363 + }, + { + "epoch": 3.030808729139923, + "grad_norm": 0.6070159077644348, + "learning_rate": 7.285541247590444e-06, + "loss": 1.8702, + "step": 2364 + }, + { + "epoch": 3.0320924261874196, + "grad_norm": 0.6779001355171204, + "learning_rate": 7.267175457009271e-06, + "loss": 1.8486, + "step": 2365 + }, + { + "epoch": 3.0333761232349166, + "grad_norm": 0.7086068987846375, + "learning_rate": 7.248828906865854e-06, + "loss": 1.8611, + "step": 2366 + }, + { + "epoch": 3.0346598202824135, + "grad_norm": 0.6151211857795715, + "learning_rate": 7.230501617066521e-06, + "loss": 1.8753, + "step": 2367 + }, + { + "epoch": 3.03594351732991, + "grad_norm": 0.6186229586601257, + "learning_rate": 7.2121936074967284e-06, + "loss": 1.9024, + "step": 2368 + }, + { + "epoch": 3.037227214377407, + "grad_norm": 0.6652982831001282, + "learning_rate": 7.193904898020989e-06, + "loss": 1.8751, + "step": 2369 + }, + { + "epoch": 3.0385109114249036, + "grad_norm": 0.58751380443573, + "learning_rate": 7.175635508482892e-06, + "loss": 1.8302, + "step": 2370 + }, + { + "epoch": 3.0397946084724006, + "grad_norm": 0.639604389667511, + "learning_rate": 7.157385458705065e-06, + "loss": 1.8356, + "step": 2371 + }, + { + "epoch": 3.041078305519897, + "grad_norm": 0.6300159096717834, + "learning_rate": 7.1391547684891366e-06, + "loss": 1.8084, + "step": 2372 + }, + { + "epoch": 3.042362002567394, + "grad_norm": 0.638019323348999, + "learning_rate": 7.1209434576157306e-06, + "loss": 1.8283, + "step": 2373 + }, + { + "epoch": 3.043645699614891, + "grad_norm": 0.6098268032073975, + "learning_rate": 7.102751545844466e-06, + "loss": 1.7342, + "step": 2374 + }, + { + "epoch": 3.0449293966623876, + "grad_norm": 0.5811750292778015, + "learning_rate": 7.084579052913884e-06, + "loss": 1.8149, + "step": 2375 + }, + { + "epoch": 3.0462130937098846, + "grad_norm": 0.6029242873191833, + "learning_rate": 7.066425998541485e-06, + "loss": 1.893, + "step": 2376 + }, + { + "epoch": 3.047496790757381, + "grad_norm": 0.6639921069145203, + "learning_rate": 7.04829240242365e-06, + "loss": 1.7763, + "step": 2377 + }, + { + "epoch": 3.048780487804878, + "grad_norm": 0.6434364318847656, + "learning_rate": 7.030178284235675e-06, + "loss": 1.7472, + "step": 2378 + }, + { + "epoch": 3.0500641848523746, + "grad_norm": 0.6220137476921082, + "learning_rate": 7.012083663631694e-06, + "loss": 1.7797, + "step": 2379 + }, + { + "epoch": 3.0513478818998716, + "grad_norm": 0.6289108395576477, + "learning_rate": 6.994008560244708e-06, + "loss": 1.7935, + "step": 2380 + }, + { + "epoch": 3.0526315789473686, + "grad_norm": 0.6736200451850891, + "learning_rate": 6.975952993686538e-06, + "loss": 1.8409, + "step": 2381 + }, + { + "epoch": 3.053915275994865, + "grad_norm": 0.6416347026824951, + "learning_rate": 6.957916983547799e-06, + "loss": 1.8102, + "step": 2382 + }, + { + "epoch": 3.055198973042362, + "grad_norm": 0.626126229763031, + "learning_rate": 6.939900549397882e-06, + "loss": 1.7772, + "step": 2383 + }, + { + "epoch": 3.0564826700898586, + "grad_norm": 0.6948550343513489, + "learning_rate": 6.921903710784955e-06, + "loss": 1.798, + "step": 2384 + }, + { + "epoch": 3.0577663671373556, + "grad_norm": 0.6228773593902588, + "learning_rate": 6.903926487235921e-06, + "loss": 1.8332, + "step": 2385 + }, + { + "epoch": 3.0590500641848526, + "grad_norm": 0.6584948897361755, + "learning_rate": 6.8859688982563785e-06, + "loss": 1.9006, + "step": 2386 + }, + { + "epoch": 3.060333761232349, + "grad_norm": 0.6317896842956543, + "learning_rate": 6.868030963330657e-06, + "loss": 1.8598, + "step": 2387 + }, + { + "epoch": 3.061617458279846, + "grad_norm": 0.6469691395759583, + "learning_rate": 6.8501127019217346e-06, + "loss": 1.8307, + "step": 2388 + }, + { + "epoch": 3.0629011553273426, + "grad_norm": 0.6384879946708679, + "learning_rate": 6.832214133471237e-06, + "loss": 1.9037, + "step": 2389 + }, + { + "epoch": 3.0641848523748396, + "grad_norm": 0.7697003483772278, + "learning_rate": 6.814335277399467e-06, + "loss": 1.8535, + "step": 2390 + }, + { + "epoch": 3.065468549422336, + "grad_norm": 0.6824406385421753, + "learning_rate": 6.796476153105294e-06, + "loss": 1.8881, + "step": 2391 + }, + { + "epoch": 3.066752246469833, + "grad_norm": 0.6439642310142517, + "learning_rate": 6.778636779966185e-06, + "loss": 1.85, + "step": 2392 + }, + { + "epoch": 3.06803594351733, + "grad_norm": 0.6553017497062683, + "learning_rate": 6.7608171773382e-06, + "loss": 1.8328, + "step": 2393 + }, + { + "epoch": 3.0693196405648266, + "grad_norm": 0.6943056583404541, + "learning_rate": 6.743017364555917e-06, + "loss": 1.7643, + "step": 2394 + }, + { + "epoch": 3.0706033376123236, + "grad_norm": 0.6459136009216309, + "learning_rate": 6.725237360932465e-06, + "loss": 1.7932, + "step": 2395 + }, + { + "epoch": 3.07188703465982, + "grad_norm": 0.6445626020431519, + "learning_rate": 6.707477185759484e-06, + "loss": 1.8238, + "step": 2396 + }, + { + "epoch": 3.073170731707317, + "grad_norm": 0.6319563984870911, + "learning_rate": 6.689736858307072e-06, + "loss": 1.839, + "step": 2397 + }, + { + "epoch": 3.0744544287548137, + "grad_norm": 0.6121535897254944, + "learning_rate": 6.672016397823807e-06, + "loss": 1.793, + "step": 2398 + }, + { + "epoch": 3.0757381258023107, + "grad_norm": 0.686077356338501, + "learning_rate": 6.654315823536713e-06, + "loss": 1.8349, + "step": 2399 + }, + { + "epoch": 3.0770218228498076, + "grad_norm": 0.6386774182319641, + "learning_rate": 6.636635154651249e-06, + "loss": 1.8856, + "step": 2400 + }, + { + "epoch": 3.078305519897304, + "grad_norm": 0.6345734596252441, + "learning_rate": 6.618974410351247e-06, + "loss": 1.8103, + "step": 2401 + }, + { + "epoch": 3.079589216944801, + "grad_norm": 0.6236886382102966, + "learning_rate": 6.601333609798946e-06, + "loss": 1.8787, + "step": 2402 + }, + { + "epoch": 3.0808729139922977, + "grad_norm": 0.6367089748382568, + "learning_rate": 6.5837127721349285e-06, + "loss": 1.9054, + "step": 2403 + }, + { + "epoch": 3.0821566110397947, + "grad_norm": 0.618385374546051, + "learning_rate": 6.5661119164781325e-06, + "loss": 1.9124, + "step": 2404 + }, + { + "epoch": 3.083440308087291, + "grad_norm": 0.632254421710968, + "learning_rate": 6.548531061925797e-06, + "loss": 1.8721, + "step": 2405 + }, + { + "epoch": 3.084724005134788, + "grad_norm": 0.6028104424476624, + "learning_rate": 6.530970227553482e-06, + "loss": 1.7732, + "step": 2406 + }, + { + "epoch": 3.086007702182285, + "grad_norm": 0.6279903650283813, + "learning_rate": 6.513429432415005e-06, + "loss": 1.9375, + "step": 2407 + }, + { + "epoch": 3.0872913992297817, + "grad_norm": 0.6415666937828064, + "learning_rate": 6.495908695542444e-06, + "loss": 1.8564, + "step": 2408 + }, + { + "epoch": 3.0885750962772787, + "grad_norm": 0.626091480255127, + "learning_rate": 6.478408035946121e-06, + "loss": 1.8417, + "step": 2409 + }, + { + "epoch": 3.089858793324775, + "grad_norm": 0.6599276065826416, + "learning_rate": 6.460927472614583e-06, + "loss": 1.9092, + "step": 2410 + }, + { + "epoch": 3.091142490372272, + "grad_norm": 0.6787567138671875, + "learning_rate": 6.44346702451454e-06, + "loss": 1.8684, + "step": 2411 + }, + { + "epoch": 3.092426187419769, + "grad_norm": 0.6268061995506287, + "learning_rate": 6.426026710590913e-06, + "loss": 1.8127, + "step": 2412 + }, + { + "epoch": 3.0937098844672657, + "grad_norm": 0.6313393712043762, + "learning_rate": 6.408606549766749e-06, + "loss": 1.8474, + "step": 2413 + }, + { + "epoch": 3.0949935815147627, + "grad_norm": 0.6835222840309143, + "learning_rate": 6.3912065609432415e-06, + "loss": 1.893, + "step": 2414 + }, + { + "epoch": 3.096277278562259, + "grad_norm": 0.622174084186554, + "learning_rate": 6.373826762999705e-06, + "loss": 1.8401, + "step": 2415 + }, + { + "epoch": 3.097560975609756, + "grad_norm": 0.588146984577179, + "learning_rate": 6.3564671747935316e-06, + "loss": 1.6928, + "step": 2416 + }, + { + "epoch": 3.0988446726572527, + "grad_norm": 0.6589400768280029, + "learning_rate": 6.339127815160181e-06, + "loss": 1.819, + "step": 2417 + }, + { + "epoch": 3.1001283697047497, + "grad_norm": 0.6727104187011719, + "learning_rate": 6.321808702913184e-06, + "loss": 1.8272, + "step": 2418 + }, + { + "epoch": 3.1014120667522467, + "grad_norm": 0.6203311085700989, + "learning_rate": 6.304509856844096e-06, + "loss": 1.8192, + "step": 2419 + }, + { + "epoch": 3.102695763799743, + "grad_norm": 0.654226541519165, + "learning_rate": 6.28723129572247e-06, + "loss": 1.8, + "step": 2420 + }, + { + "epoch": 3.10397946084724, + "grad_norm": 0.6536027789115906, + "learning_rate": 6.269973038295876e-06, + "loss": 1.8655, + "step": 2421 + }, + { + "epoch": 3.1052631578947367, + "grad_norm": 0.6783435940742493, + "learning_rate": 6.252735103289823e-06, + "loss": 1.8466, + "step": 2422 + }, + { + "epoch": 3.1065468549422337, + "grad_norm": 0.66269850730896, + "learning_rate": 6.235517509407776e-06, + "loss": 1.8133, + "step": 2423 + }, + { + "epoch": 3.1078305519897302, + "grad_norm": 0.6523348689079285, + "learning_rate": 6.2183202753311655e-06, + "loss": 1.8375, + "step": 2424 + }, + { + "epoch": 3.109114249037227, + "grad_norm": 0.6795269846916199, + "learning_rate": 6.201143419719288e-06, + "loss": 1.8972, + "step": 2425 + }, + { + "epoch": 3.110397946084724, + "grad_norm": 0.6744492650032043, + "learning_rate": 6.18398696120934e-06, + "loss": 1.8884, + "step": 2426 + }, + { + "epoch": 3.1116816431322207, + "grad_norm": 0.6446805000305176, + "learning_rate": 6.166850918416406e-06, + "loss": 1.88, + "step": 2427 + }, + { + "epoch": 3.1129653401797177, + "grad_norm": 0.6499425172805786, + "learning_rate": 6.149735309933388e-06, + "loss": 1.9088, + "step": 2428 + }, + { + "epoch": 3.1142490372272142, + "grad_norm": 0.6045271158218384, + "learning_rate": 6.13264015433104e-06, + "loss": 1.7584, + "step": 2429 + }, + { + "epoch": 3.1155327342747112, + "grad_norm": 0.6386954188346863, + "learning_rate": 6.1155654701579274e-06, + "loss": 1.8694, + "step": 2430 + }, + { + "epoch": 3.1168164313222078, + "grad_norm": 0.7217512130737305, + "learning_rate": 6.098511275940386e-06, + "loss": 1.8029, + "step": 2431 + }, + { + "epoch": 3.1181001283697047, + "grad_norm": 0.65043705701828, + "learning_rate": 6.081477590182516e-06, + "loss": 1.9037, + "step": 2432 + }, + { + "epoch": 3.1193838254172017, + "grad_norm": 0.6597872972488403, + "learning_rate": 6.0644644313661905e-06, + "loss": 1.828, + "step": 2433 + }, + { + "epoch": 3.1206675224646983, + "grad_norm": 0.6494190692901611, + "learning_rate": 6.047471817951e-06, + "loss": 1.8891, + "step": 2434 + }, + { + "epoch": 3.1219512195121952, + "grad_norm": 0.6844896674156189, + "learning_rate": 6.030499768374231e-06, + "loss": 1.9077, + "step": 2435 + }, + { + "epoch": 3.1232349165596918, + "grad_norm": 0.6584227085113525, + "learning_rate": 6.013548301050864e-06, + "loss": 1.8379, + "step": 2436 + }, + { + "epoch": 3.1245186136071887, + "grad_norm": 0.6577264666557312, + "learning_rate": 5.996617434373552e-06, + "loss": 1.9224, + "step": 2437 + }, + { + "epoch": 3.1258023106546853, + "grad_norm": 0.6606531143188477, + "learning_rate": 5.979707186712607e-06, + "loss": 1.8955, + "step": 2438 + }, + { + "epoch": 3.1270860077021823, + "grad_norm": 0.6260818839073181, + "learning_rate": 5.962817576415936e-06, + "loss": 1.7814, + "step": 2439 + }, + { + "epoch": 3.1283697047496792, + "grad_norm": 0.7082614898681641, + "learning_rate": 5.945948621809091e-06, + "loss": 1.8656, + "step": 2440 + }, + { + "epoch": 3.1296534017971758, + "grad_norm": 0.7559731602668762, + "learning_rate": 5.929100341195188e-06, + "loss": 1.841, + "step": 2441 + }, + { + "epoch": 3.1309370988446728, + "grad_norm": 0.6229314208030701, + "learning_rate": 5.912272752854911e-06, + "loss": 1.8639, + "step": 2442 + }, + { + "epoch": 3.1322207958921693, + "grad_norm": 0.6472817659378052, + "learning_rate": 5.895465875046504e-06, + "loss": 1.8056, + "step": 2443 + }, + { + "epoch": 3.1335044929396663, + "grad_norm": 0.6459691524505615, + "learning_rate": 5.878679726005748e-06, + "loss": 1.7871, + "step": 2444 + }, + { + "epoch": 3.1347881899871632, + "grad_norm": 0.6071081757545471, + "learning_rate": 5.8619143239459055e-06, + "loss": 1.8181, + "step": 2445 + }, + { + "epoch": 3.1360718870346598, + "grad_norm": 0.6998416185379028, + "learning_rate": 5.845169687057753e-06, + "loss": 1.8194, + "step": 2446 + }, + { + "epoch": 3.1373555840821568, + "grad_norm": 0.6498885750770569, + "learning_rate": 5.828445833509516e-06, + "loss": 1.8793, + "step": 2447 + }, + { + "epoch": 3.1386392811296533, + "grad_norm": 0.632911205291748, + "learning_rate": 5.811742781446886e-06, + "loss": 1.864, + "step": 2448 + }, + { + "epoch": 3.1399229781771503, + "grad_norm": 0.6404876708984375, + "learning_rate": 5.79506054899299e-06, + "loss": 1.8207, + "step": 2449 + }, + { + "epoch": 3.141206675224647, + "grad_norm": 0.6406745314598083, + "learning_rate": 5.7783991542483404e-06, + "loss": 1.7686, + "step": 2450 + }, + { + "epoch": 3.142490372272144, + "grad_norm": 0.6381062269210815, + "learning_rate": 5.7617586152908524e-06, + "loss": 1.7445, + "step": 2451 + }, + { + "epoch": 3.1437740693196408, + "grad_norm": 0.6504518389701843, + "learning_rate": 5.745138950175818e-06, + "loss": 1.9185, + "step": 2452 + }, + { + "epoch": 3.1450577663671373, + "grad_norm": 0.6681715250015259, + "learning_rate": 5.7285401769358845e-06, + "loss": 1.7901, + "step": 2453 + }, + { + "epoch": 3.1463414634146343, + "grad_norm": 0.6318925619125366, + "learning_rate": 5.711962313581009e-06, + "loss": 1.8506, + "step": 2454 + }, + { + "epoch": 3.147625160462131, + "grad_norm": 0.636315107345581, + "learning_rate": 5.6954053780984925e-06, + "loss": 1.8915, + "step": 2455 + }, + { + "epoch": 3.148908857509628, + "grad_norm": 0.7196581363677979, + "learning_rate": 5.678869388452901e-06, + "loss": 1.7908, + "step": 2456 + }, + { + "epoch": 3.1501925545571243, + "grad_norm": 0.6940009593963623, + "learning_rate": 5.662354362586078e-06, + "loss": 1.9142, + "step": 2457 + }, + { + "epoch": 3.1514762516046213, + "grad_norm": 0.6408509016036987, + "learning_rate": 5.645860318417137e-06, + "loss": 1.9797, + "step": 2458 + }, + { + "epoch": 3.1527599486521183, + "grad_norm": 0.6157658100128174, + "learning_rate": 5.629387273842421e-06, + "loss": 1.8644, + "step": 2459 + }, + { + "epoch": 3.154043645699615, + "grad_norm": 0.7050480842590332, + "learning_rate": 5.612935246735476e-06, + "loss": 1.8271, + "step": 2460 + }, + { + "epoch": 3.155327342747112, + "grad_norm": 0.6329376697540283, + "learning_rate": 5.5965042549470465e-06, + "loss": 1.8351, + "step": 2461 + }, + { + "epoch": 3.1566110397946083, + "grad_norm": 0.6419767141342163, + "learning_rate": 5.580094316305063e-06, + "loss": 1.9005, + "step": 2462 + }, + { + "epoch": 3.1578947368421053, + "grad_norm": 0.6756376028060913, + "learning_rate": 5.563705448614615e-06, + "loss": 1.9276, + "step": 2463 + }, + { + "epoch": 3.1591784338896023, + "grad_norm": 0.6476380825042725, + "learning_rate": 5.547337669657912e-06, + "loss": 1.8448, + "step": 2464 + }, + { + "epoch": 3.160462130937099, + "grad_norm": 0.6600066423416138, + "learning_rate": 5.530990997194302e-06, + "loss": 1.8759, + "step": 2465 + }, + { + "epoch": 3.161745827984596, + "grad_norm": 0.6828904151916504, + "learning_rate": 5.51466544896021e-06, + "loss": 1.8338, + "step": 2466 + }, + { + "epoch": 3.1630295250320923, + "grad_norm": 0.6354855298995972, + "learning_rate": 5.498361042669162e-06, + "loss": 1.8443, + "step": 2467 + }, + { + "epoch": 3.1643132220795893, + "grad_norm": 0.6827327609062195, + "learning_rate": 5.482077796011742e-06, + "loss": 1.7586, + "step": 2468 + }, + { + "epoch": 3.165596919127086, + "grad_norm": 0.6475961208343506, + "learning_rate": 5.46581572665556e-06, + "loss": 1.7795, + "step": 2469 + }, + { + "epoch": 3.166880616174583, + "grad_norm": 0.661024272441864, + "learning_rate": 5.449574852245254e-06, + "loss": 1.8766, + "step": 2470 + }, + { + "epoch": 3.1681643132220794, + "grad_norm": 0.6311275362968445, + "learning_rate": 5.433355190402473e-06, + "loss": 1.7328, + "step": 2471 + }, + { + "epoch": 3.1694480102695763, + "grad_norm": 0.7183083295822144, + "learning_rate": 5.417156758725855e-06, + "loss": 1.8477, + "step": 2472 + }, + { + "epoch": 3.1707317073170733, + "grad_norm": 0.6651450395584106, + "learning_rate": 5.400979574790977e-06, + "loss": 1.8005, + "step": 2473 + }, + { + "epoch": 3.17201540436457, + "grad_norm": 0.6510912179946899, + "learning_rate": 5.384823656150392e-06, + "loss": 1.8683, + "step": 2474 + }, + { + "epoch": 3.173299101412067, + "grad_norm": 0.6630425453186035, + "learning_rate": 5.3686890203335645e-06, + "loss": 1.7495, + "step": 2475 + }, + { + "epoch": 3.1745827984595634, + "grad_norm": 0.6187940239906311, + "learning_rate": 5.352575684846856e-06, + "loss": 1.8652, + "step": 2476 + }, + { + "epoch": 3.1758664955070603, + "grad_norm": 0.6903923153877258, + "learning_rate": 5.3364836671735365e-06, + "loss": 1.8444, + "step": 2477 + }, + { + "epoch": 3.1771501925545573, + "grad_norm": 0.5970506072044373, + "learning_rate": 5.320412984773748e-06, + "loss": 1.8473, + "step": 2478 + }, + { + "epoch": 3.178433889602054, + "grad_norm": 0.7086596488952637, + "learning_rate": 5.30436365508446e-06, + "loss": 1.8704, + "step": 2479 + }, + { + "epoch": 3.179717586649551, + "grad_norm": 0.6811697483062744, + "learning_rate": 5.288335695519497e-06, + "loss": 1.81, + "step": 2480 + }, + { + "epoch": 3.1810012836970474, + "grad_norm": 0.6327681541442871, + "learning_rate": 5.2723291234694775e-06, + "loss": 1.7599, + "step": 2481 + }, + { + "epoch": 3.1822849807445444, + "grad_norm": 0.6731870174407959, + "learning_rate": 5.256343956301829e-06, + "loss": 1.9096, + "step": 2482 + }, + { + "epoch": 3.183568677792041, + "grad_norm": 0.6829713582992554, + "learning_rate": 5.2403802113607566e-06, + "loss": 1.7646, + "step": 2483 + }, + { + "epoch": 3.184852374839538, + "grad_norm": 0.6533412933349609, + "learning_rate": 5.224437905967208e-06, + "loss": 1.8265, + "step": 2484 + }, + { + "epoch": 3.186136071887035, + "grad_norm": 0.6502362489700317, + "learning_rate": 5.2085170574188685e-06, + "loss": 1.8348, + "step": 2485 + }, + { + "epoch": 3.1874197689345314, + "grad_norm": 0.6658063530921936, + "learning_rate": 5.192617682990156e-06, + "loss": 1.7652, + "step": 2486 + }, + { + "epoch": 3.1887034659820284, + "grad_norm": 0.7827169895172119, + "learning_rate": 5.176739799932184e-06, + "loss": 1.8159, + "step": 2487 + }, + { + "epoch": 3.189987163029525, + "grad_norm": 0.6694787740707397, + "learning_rate": 5.160883425472743e-06, + "loss": 1.8454, + "step": 2488 + }, + { + "epoch": 3.191270860077022, + "grad_norm": 0.6306465268135071, + "learning_rate": 5.1450485768162805e-06, + "loss": 1.8307, + "step": 2489 + }, + { + "epoch": 3.1925545571245184, + "grad_norm": 0.6069169640541077, + "learning_rate": 5.1292352711439e-06, + "loss": 1.8525, + "step": 2490 + }, + { + "epoch": 3.1938382541720154, + "grad_norm": 0.6383316516876221, + "learning_rate": 5.113443525613334e-06, + "loss": 1.8147, + "step": 2491 + }, + { + "epoch": 3.1951219512195124, + "grad_norm": 0.6666647791862488, + "learning_rate": 5.097673357358907e-06, + "loss": 1.8862, + "step": 2492 + }, + { + "epoch": 3.196405648267009, + "grad_norm": 0.6443461775779724, + "learning_rate": 5.081924783491543e-06, + "loss": 1.6981, + "step": 2493 + }, + { + "epoch": 3.197689345314506, + "grad_norm": 0.6898028254508972, + "learning_rate": 5.066197821098728e-06, + "loss": 1.8163, + "step": 2494 + }, + { + "epoch": 3.1989730423620024, + "grad_norm": 0.6226739883422852, + "learning_rate": 5.050492487244498e-06, + "loss": 1.8452, + "step": 2495 + }, + { + "epoch": 3.2002567394094994, + "grad_norm": 0.6201604604721069, + "learning_rate": 5.034808798969434e-06, + "loss": 1.7734, + "step": 2496 + }, + { + "epoch": 3.2015404364569964, + "grad_norm": 0.6525123119354248, + "learning_rate": 5.0191467732906256e-06, + "loss": 1.91, + "step": 2497 + }, + { + "epoch": 3.202824133504493, + "grad_norm": 0.6529130935668945, + "learning_rate": 5.003506427201648e-06, + "loss": 1.8439, + "step": 2498 + }, + { + "epoch": 3.20410783055199, + "grad_norm": 0.6540563106536865, + "learning_rate": 4.987887777672571e-06, + "loss": 1.8104, + "step": 2499 + }, + { + "epoch": 3.2053915275994864, + "grad_norm": 0.6934303641319275, + "learning_rate": 4.972290841649901e-06, + "loss": 1.7921, + "step": 2500 + }, + { + "epoch": 3.2066752246469834, + "grad_norm": 0.6331066489219666, + "learning_rate": 4.956715636056608e-06, + "loss": 1.9148, + "step": 2501 + }, + { + "epoch": 3.20795892169448, + "grad_norm": 0.6602702140808105, + "learning_rate": 4.941162177792077e-06, + "loss": 1.8512, + "step": 2502 + }, + { + "epoch": 3.209242618741977, + "grad_norm": 0.6286439299583435, + "learning_rate": 4.9256304837320885e-06, + "loss": 1.8532, + "step": 2503 + }, + { + "epoch": 3.2105263157894735, + "grad_norm": 0.6425065398216248, + "learning_rate": 4.910120570728805e-06, + "loss": 1.8592, + "step": 2504 + }, + { + "epoch": 3.2118100128369704, + "grad_norm": 0.6924281716346741, + "learning_rate": 4.894632455610773e-06, + "loss": 1.7432, + "step": 2505 + }, + { + "epoch": 3.2130937098844674, + "grad_norm": 0.6882502436637878, + "learning_rate": 4.879166155182888e-06, + "loss": 1.8532, + "step": 2506 + }, + { + "epoch": 3.214377406931964, + "grad_norm": 0.6845598220825195, + "learning_rate": 4.86372168622635e-06, + "loss": 1.841, + "step": 2507 + }, + { + "epoch": 3.215661103979461, + "grad_norm": 0.6545780301094055, + "learning_rate": 4.848299065498707e-06, + "loss": 1.8505, + "step": 2508 + }, + { + "epoch": 3.2169448010269575, + "grad_norm": 0.69523686170578, + "learning_rate": 4.832898309733774e-06, + "loss": 1.7954, + "step": 2509 + }, + { + "epoch": 3.2182284980744544, + "grad_norm": 0.6519680619239807, + "learning_rate": 4.8175194356416405e-06, + "loss": 1.8696, + "step": 2510 + }, + { + "epoch": 3.2195121951219514, + "grad_norm": 0.7183489203453064, + "learning_rate": 4.8021624599086875e-06, + "loss": 1.8155, + "step": 2511 + }, + { + "epoch": 3.220795892169448, + "grad_norm": 0.6638036966323853, + "learning_rate": 4.786827399197505e-06, + "loss": 1.8436, + "step": 2512 + }, + { + "epoch": 3.222079589216945, + "grad_norm": 0.6539146304130554, + "learning_rate": 4.771514270146912e-06, + "loss": 1.8146, + "step": 2513 + }, + { + "epoch": 3.2233632862644415, + "grad_norm": 0.6592859625816345, + "learning_rate": 4.756223089371925e-06, + "loss": 1.9218, + "step": 2514 + }, + { + "epoch": 3.2246469833119384, + "grad_norm": 0.6947745084762573, + "learning_rate": 4.7409538734637606e-06, + "loss": 1.7609, + "step": 2515 + }, + { + "epoch": 3.225930680359435, + "grad_norm": 0.7831366658210754, + "learning_rate": 4.725706638989805e-06, + "loss": 1.8954, + "step": 2516 + }, + { + "epoch": 3.227214377406932, + "grad_norm": 0.6695125699043274, + "learning_rate": 4.710481402493569e-06, + "loss": 1.8394, + "step": 2517 + }, + { + "epoch": 3.228498074454429, + "grad_norm": 0.7059369683265686, + "learning_rate": 4.695278180494725e-06, + "loss": 1.762, + "step": 2518 + }, + { + "epoch": 3.2297817715019255, + "grad_norm": 0.6574482321739197, + "learning_rate": 4.680096989489036e-06, + "loss": 1.8136, + "step": 2519 + }, + { + "epoch": 3.2310654685494224, + "grad_norm": 0.6736265420913696, + "learning_rate": 4.664937845948375e-06, + "loss": 1.8689, + "step": 2520 + }, + { + "epoch": 3.232349165596919, + "grad_norm": 0.6809471249580383, + "learning_rate": 4.649800766320695e-06, + "loss": 1.8112, + "step": 2521 + }, + { + "epoch": 3.233632862644416, + "grad_norm": 0.6577104926109314, + "learning_rate": 4.634685767029997e-06, + "loss": 1.8869, + "step": 2522 + }, + { + "epoch": 3.2349165596919125, + "grad_norm": 0.686516523361206, + "learning_rate": 4.619592864476324e-06, + "loss": 1.8001, + "step": 2523 + }, + { + "epoch": 3.2362002567394095, + "grad_norm": 0.6891453862190247, + "learning_rate": 4.604522075035761e-06, + "loss": 1.9459, + "step": 2524 + }, + { + "epoch": 3.2374839537869065, + "grad_norm": 0.6619938015937805, + "learning_rate": 4.589473415060389e-06, + "loss": 1.8013, + "step": 2525 + }, + { + "epoch": 3.238767650834403, + "grad_norm": 0.6440871357917786, + "learning_rate": 4.574446900878271e-06, + "loss": 1.753, + "step": 2526 + }, + { + "epoch": 3.2400513478819, + "grad_norm": 0.6430378556251526, + "learning_rate": 4.559442548793455e-06, + "loss": 1.797, + "step": 2527 + }, + { + "epoch": 3.2413350449293965, + "grad_norm": 0.6333809494972229, + "learning_rate": 4.544460375085938e-06, + "loss": 1.8128, + "step": 2528 + }, + { + "epoch": 3.2426187419768935, + "grad_norm": 0.7178351879119873, + "learning_rate": 4.5295003960116405e-06, + "loss": 1.7443, + "step": 2529 + }, + { + "epoch": 3.2439024390243905, + "grad_norm": 0.6922745108604431, + "learning_rate": 4.514562627802418e-06, + "loss": 1.899, + "step": 2530 + }, + { + "epoch": 3.245186136071887, + "grad_norm": 0.6388792395591736, + "learning_rate": 4.499647086666029e-06, + "loss": 1.8496, + "step": 2531 + }, + { + "epoch": 3.246469833119384, + "grad_norm": 0.6731453537940979, + "learning_rate": 4.484753788786094e-06, + "loss": 1.8498, + "step": 2532 + }, + { + "epoch": 3.2477535301668805, + "grad_norm": 0.6593973636627197, + "learning_rate": 4.469882750322127e-06, + "loss": 1.8905, + "step": 2533 + }, + { + "epoch": 3.2490372272143775, + "grad_norm": 0.6302469968795776, + "learning_rate": 4.455033987409463e-06, + "loss": 1.8119, + "step": 2534 + }, + { + "epoch": 3.250320924261874, + "grad_norm": 0.6799659132957458, + "learning_rate": 4.440207516159284e-06, + "loss": 1.8207, + "step": 2535 + }, + { + "epoch": 3.251604621309371, + "grad_norm": 0.6262779235839844, + "learning_rate": 4.425403352658591e-06, + "loss": 1.8271, + "step": 2536 + }, + { + "epoch": 3.2528883183568675, + "grad_norm": 0.6541080474853516, + "learning_rate": 4.410621512970162e-06, + "loss": 1.8577, + "step": 2537 + }, + { + "epoch": 3.2541720154043645, + "grad_norm": 0.6848664283752441, + "learning_rate": 4.395862013132557e-06, + "loss": 1.8253, + "step": 2538 + }, + { + "epoch": 3.2554557124518615, + "grad_norm": 0.6990413665771484, + "learning_rate": 4.381124869160111e-06, + "loss": 1.8913, + "step": 2539 + }, + { + "epoch": 3.256739409499358, + "grad_norm": 0.6671677231788635, + "learning_rate": 4.366410097042897e-06, + "loss": 1.902, + "step": 2540 + }, + { + "epoch": 3.258023106546855, + "grad_norm": 0.6806551218032837, + "learning_rate": 4.351717712746703e-06, + "loss": 1.8234, + "step": 2541 + }, + { + "epoch": 3.2593068035943515, + "grad_norm": 0.6618762016296387, + "learning_rate": 4.337047732213034e-06, + "loss": 1.802, + "step": 2542 + }, + { + "epoch": 3.2605905006418485, + "grad_norm": 0.6978524327278137, + "learning_rate": 4.322400171359092e-06, + "loss": 1.8022, + "step": 2543 + }, + { + "epoch": 3.2618741976893455, + "grad_norm": 0.6757627129554749, + "learning_rate": 4.307775046077739e-06, + "loss": 1.8731, + "step": 2544 + }, + { + "epoch": 3.263157894736842, + "grad_norm": 0.6596835851669312, + "learning_rate": 4.2931723722375055e-06, + "loss": 1.8383, + "step": 2545 + }, + { + "epoch": 3.264441591784339, + "grad_norm": 0.6343722939491272, + "learning_rate": 4.278592165682568e-06, + "loss": 1.7625, + "step": 2546 + }, + { + "epoch": 3.2657252888318355, + "grad_norm": 0.6410669684410095, + "learning_rate": 4.264034442232709e-06, + "loss": 1.779, + "step": 2547 + }, + { + "epoch": 3.2670089858793325, + "grad_norm": 0.6475167870521545, + "learning_rate": 4.249499217683317e-06, + "loss": 1.7807, + "step": 2548 + }, + { + "epoch": 3.2682926829268295, + "grad_norm": 0.6346061825752258, + "learning_rate": 4.2349865078053805e-06, + "loss": 1.8682, + "step": 2549 + }, + { + "epoch": 3.269576379974326, + "grad_norm": 0.6768635511398315, + "learning_rate": 4.220496328345463e-06, + "loss": 1.8102, + "step": 2550 + }, + { + "epoch": 3.270860077021823, + "grad_norm": 0.6839442849159241, + "learning_rate": 4.20602869502566e-06, + "loss": 1.8731, + "step": 2551 + }, + { + "epoch": 3.2721437740693196, + "grad_norm": 0.7137340903282166, + "learning_rate": 4.191583623543629e-06, + "loss": 1.8751, + "step": 2552 + }, + { + "epoch": 3.2734274711168165, + "grad_norm": 0.7194527387619019, + "learning_rate": 4.1771611295725235e-06, + "loss": 1.8277, + "step": 2553 + }, + { + "epoch": 3.274711168164313, + "grad_norm": 0.6361044645309448, + "learning_rate": 4.16276122876102e-06, + "loss": 1.8588, + "step": 2554 + }, + { + "epoch": 3.27599486521181, + "grad_norm": 0.6724072098731995, + "learning_rate": 4.148383936733277e-06, + "loss": 1.8798, + "step": 2555 + }, + { + "epoch": 3.2772785622593066, + "grad_norm": 0.6768350005149841, + "learning_rate": 4.134029269088913e-06, + "loss": 1.7995, + "step": 2556 + }, + { + "epoch": 3.2785622593068036, + "grad_norm": 0.6531705260276794, + "learning_rate": 4.119697241402998e-06, + "loss": 1.7824, + "step": 2557 + }, + { + "epoch": 3.2798459563543005, + "grad_norm": 0.7098740935325623, + "learning_rate": 4.1053878692260515e-06, + "loss": 1.7457, + "step": 2558 + }, + { + "epoch": 3.281129653401797, + "grad_norm": 0.7518153786659241, + "learning_rate": 4.091101168084005e-06, + "loss": 1.8691, + "step": 2559 + }, + { + "epoch": 3.282413350449294, + "grad_norm": 0.6463475227355957, + "learning_rate": 4.07683715347818e-06, + "loss": 1.8697, + "step": 2560 + }, + { + "epoch": 3.2836970474967906, + "grad_norm": 0.6746723651885986, + "learning_rate": 4.062595840885303e-06, + "loss": 1.8927, + "step": 2561 + }, + { + "epoch": 3.2849807445442876, + "grad_norm": 0.6478902697563171, + "learning_rate": 4.048377245757454e-06, + "loss": 1.8447, + "step": 2562 + }, + { + "epoch": 3.2862644415917845, + "grad_norm": 0.6705543994903564, + "learning_rate": 4.03418138352206e-06, + "loss": 1.8241, + "step": 2563 + }, + { + "epoch": 3.287548138639281, + "grad_norm": 0.6572802662849426, + "learning_rate": 4.020008269581895e-06, + "loss": 1.8119, + "step": 2564 + }, + { + "epoch": 3.288831835686778, + "grad_norm": 0.6686354279518127, + "learning_rate": 4.0058579193150535e-06, + "loss": 1.8595, + "step": 2565 + }, + { + "epoch": 3.2901155327342746, + "grad_norm": 0.7033422589302063, + "learning_rate": 3.991730348074921e-06, + "loss": 1.8594, + "step": 2566 + }, + { + "epoch": 3.2913992297817716, + "grad_norm": 0.6615843772888184, + "learning_rate": 3.977625571190155e-06, + "loss": 1.8002, + "step": 2567 + }, + { + "epoch": 3.292682926829268, + "grad_norm": 0.6344552636146545, + "learning_rate": 3.963543603964709e-06, + "loss": 1.8295, + "step": 2568 + }, + { + "epoch": 3.293966623876765, + "grad_norm": 0.6962928175926208, + "learning_rate": 3.9494844616777775e-06, + "loss": 1.9257, + "step": 2569 + }, + { + "epoch": 3.295250320924262, + "grad_norm": 0.670727550983429, + "learning_rate": 3.935448159583774e-06, + "loss": 1.7982, + "step": 2570 + }, + { + "epoch": 3.2965340179717586, + "grad_norm": 0.6550553441047668, + "learning_rate": 3.921434712912353e-06, + "loss": 1.8412, + "step": 2571 + }, + { + "epoch": 3.2978177150192556, + "grad_norm": 0.6494854688644409, + "learning_rate": 3.907444136868349e-06, + "loss": 1.8002, + "step": 2572 + }, + { + "epoch": 3.299101412066752, + "grad_norm": 0.6607004404067993, + "learning_rate": 3.893476446631797e-06, + "loss": 1.852, + "step": 2573 + }, + { + "epoch": 3.300385109114249, + "grad_norm": 0.6911225318908691, + "learning_rate": 3.879531657357902e-06, + "loss": 1.8873, + "step": 2574 + }, + { + "epoch": 3.3016688061617456, + "grad_norm": 0.6940852999687195, + "learning_rate": 3.865609784177008e-06, + "loss": 1.8919, + "step": 2575 + }, + { + "epoch": 3.3029525032092426, + "grad_norm": 0.6637902855873108, + "learning_rate": 3.851710842194595e-06, + "loss": 1.7805, + "step": 2576 + }, + { + "epoch": 3.3042362002567396, + "grad_norm": 0.6169712543487549, + "learning_rate": 3.837834846491284e-06, + "loss": 1.8922, + "step": 2577 + }, + { + "epoch": 3.305519897304236, + "grad_norm": 0.645349383354187, + "learning_rate": 3.823981812122765e-06, + "loss": 1.9721, + "step": 2578 + }, + { + "epoch": 3.306803594351733, + "grad_norm": 0.7103525996208191, + "learning_rate": 3.810151754119842e-06, + "loss": 1.8048, + "step": 2579 + }, + { + "epoch": 3.3080872913992296, + "grad_norm": 0.6252644062042236, + "learning_rate": 3.7963446874883848e-06, + "loss": 1.8385, + "step": 2580 + }, + { + "epoch": 3.3093709884467266, + "grad_norm": 0.6849333643913269, + "learning_rate": 3.782560627209308e-06, + "loss": 1.8185, + "step": 2581 + }, + { + "epoch": 3.3106546854942236, + "grad_norm": 0.6445103287696838, + "learning_rate": 3.7687995882385613e-06, + "loss": 1.8281, + "step": 2582 + }, + { + "epoch": 3.31193838254172, + "grad_norm": 0.6480722427368164, + "learning_rate": 3.7550615855071277e-06, + "loss": 1.8721, + "step": 2583 + }, + { + "epoch": 3.313222079589217, + "grad_norm": 0.6748453974723816, + "learning_rate": 3.7413466339210002e-06, + "loss": 1.906, + "step": 2584 + }, + { + "epoch": 3.3145057766367136, + "grad_norm": 0.6606651544570923, + "learning_rate": 3.727654748361134e-06, + "loss": 1.8296, + "step": 2585 + }, + { + "epoch": 3.3157894736842106, + "grad_norm": 0.6556921601295471, + "learning_rate": 3.7139859436834873e-06, + "loss": 1.9271, + "step": 2586 + }, + { + "epoch": 3.317073170731707, + "grad_norm": 0.6581631898880005, + "learning_rate": 3.700340234718955e-06, + "loss": 1.8146, + "step": 2587 + }, + { + "epoch": 3.318356867779204, + "grad_norm": 0.6664019227027893, + "learning_rate": 3.686717636273382e-06, + "loss": 1.7949, + "step": 2588 + }, + { + "epoch": 3.3196405648267007, + "grad_norm": 0.6999354362487793, + "learning_rate": 3.6731181631275424e-06, + "loss": 1.8269, + "step": 2589 + }, + { + "epoch": 3.3209242618741976, + "grad_norm": 0.6236140131950378, + "learning_rate": 3.6595418300371053e-06, + "loss": 1.75, + "step": 2590 + }, + { + "epoch": 3.3222079589216946, + "grad_norm": 0.6586688756942749, + "learning_rate": 3.645988651732643e-06, + "loss": 1.8923, + "step": 2591 + }, + { + "epoch": 3.323491655969191, + "grad_norm": 0.6219316720962524, + "learning_rate": 3.632458642919584e-06, + "loss": 1.8029, + "step": 2592 + }, + { + "epoch": 3.324775353016688, + "grad_norm": 0.6433475613594055, + "learning_rate": 3.618951818278263e-06, + "loss": 1.8805, + "step": 2593 + }, + { + "epoch": 3.3260590500641847, + "grad_norm": 0.6489351987838745, + "learning_rate": 3.605468192463815e-06, + "loss": 1.8195, + "step": 2594 + }, + { + "epoch": 3.3273427471116817, + "grad_norm": 0.6364544630050659, + "learning_rate": 3.5920077801062187e-06, + "loss": 1.8263, + "step": 2595 + }, + { + "epoch": 3.3286264441591786, + "grad_norm": 0.6636047959327698, + "learning_rate": 3.578570595810274e-06, + "loss": 1.8486, + "step": 2596 + }, + { + "epoch": 3.329910141206675, + "grad_norm": 0.6690197587013245, + "learning_rate": 3.5651566541555604e-06, + "loss": 1.8295, + "step": 2597 + }, + { + "epoch": 3.331193838254172, + "grad_norm": 0.6748347878456116, + "learning_rate": 3.551765969696455e-06, + "loss": 1.8692, + "step": 2598 + }, + { + "epoch": 3.3324775353016687, + "grad_norm": 0.6832334995269775, + "learning_rate": 3.5383985569621e-06, + "loss": 1.8633, + "step": 2599 + }, + { + "epoch": 3.3337612323491657, + "grad_norm": 0.6405670046806335, + "learning_rate": 3.525054430456376e-06, + "loss": 1.8337, + "step": 2600 + }, + { + "epoch": 3.335044929396662, + "grad_norm": 0.6735053062438965, + "learning_rate": 3.511733604657899e-06, + "loss": 1.7649, + "step": 2601 + }, + { + "epoch": 3.336328626444159, + "grad_norm": 0.6594917178153992, + "learning_rate": 3.4984360940200134e-06, + "loss": 1.7817, + "step": 2602 + }, + { + "epoch": 3.337612323491656, + "grad_norm": 0.6529385447502136, + "learning_rate": 3.4851619129707637e-06, + "loss": 1.7919, + "step": 2603 + }, + { + "epoch": 3.3388960205391527, + "grad_norm": 0.686526894569397, + "learning_rate": 3.4719110759128675e-06, + "loss": 1.9141, + "step": 2604 + }, + { + "epoch": 3.3401797175866497, + "grad_norm": 0.6494244933128357, + "learning_rate": 3.4586835972237384e-06, + "loss": 1.8117, + "step": 2605 + }, + { + "epoch": 3.341463414634146, + "grad_norm": 0.7029903531074524, + "learning_rate": 3.4454794912554183e-06, + "loss": 1.8476, + "step": 2606 + }, + { + "epoch": 3.342747111681643, + "grad_norm": 0.6315235495567322, + "learning_rate": 3.432298772334608e-06, + "loss": 1.8368, + "step": 2607 + }, + { + "epoch": 3.3440308087291397, + "grad_norm": 0.7127735018730164, + "learning_rate": 3.419141454762634e-06, + "loss": 1.8627, + "step": 2608 + }, + { + "epoch": 3.3453145057766367, + "grad_norm": 0.6757027506828308, + "learning_rate": 3.406007552815421e-06, + "loss": 1.8164, + "step": 2609 + }, + { + "epoch": 3.3465982028241337, + "grad_norm": 0.6804640889167786, + "learning_rate": 3.3928970807434846e-06, + "loss": 1.7726, + "step": 2610 + }, + { + "epoch": 3.34788189987163, + "grad_norm": 0.6786353588104248, + "learning_rate": 3.3798100527719395e-06, + "loss": 1.9359, + "step": 2611 + }, + { + "epoch": 3.349165596919127, + "grad_norm": 0.6764732003211975, + "learning_rate": 3.3667464831004357e-06, + "loss": 2.0082, + "step": 2612 + }, + { + "epoch": 3.3504492939666237, + "grad_norm": 0.671834409236908, + "learning_rate": 3.3537063859031937e-06, + "loss": 1.8453, + "step": 2613 + }, + { + "epoch": 3.3517329910141207, + "grad_norm": 0.717833936214447, + "learning_rate": 3.340689775328956e-06, + "loss": 1.8891, + "step": 2614 + }, + { + "epoch": 3.3530166880616177, + "grad_norm": 0.6400378942489624, + "learning_rate": 3.327696665500982e-06, + "loss": 1.8714, + "step": 2615 + }, + { + "epoch": 3.354300385109114, + "grad_norm": 0.6316032409667969, + "learning_rate": 3.3147270705170263e-06, + "loss": 1.8419, + "step": 2616 + }, + { + "epoch": 3.355584082156611, + "grad_norm": 0.7087571620941162, + "learning_rate": 3.301781004449339e-06, + "loss": 1.8284, + "step": 2617 + }, + { + "epoch": 3.3568677792041077, + "grad_norm": 0.674679696559906, + "learning_rate": 3.288858481344645e-06, + "loss": 1.8953, + "step": 2618 + }, + { + "epoch": 3.3581514762516047, + "grad_norm": 0.7128342390060425, + "learning_rate": 3.2759595152241087e-06, + "loss": 1.8231, + "step": 2619 + }, + { + "epoch": 3.3594351732991012, + "grad_norm": 0.6478931903839111, + "learning_rate": 3.2630841200833427e-06, + "loss": 1.8583, + "step": 2620 + }, + { + "epoch": 3.360718870346598, + "grad_norm": 0.6589947938919067, + "learning_rate": 3.25023230989239e-06, + "loss": 1.8601, + "step": 2621 + }, + { + "epoch": 3.3620025673940948, + "grad_norm": 0.6713588237762451, + "learning_rate": 3.2374040985957004e-06, + "loss": 1.9216, + "step": 2622 + }, + { + "epoch": 3.3632862644415917, + "grad_norm": 0.6780170798301697, + "learning_rate": 3.2245995001121106e-06, + "loss": 1.8351, + "step": 2623 + }, + { + "epoch": 3.3645699614890887, + "grad_norm": 0.6822700500488281, + "learning_rate": 3.2118185283348505e-06, + "loss": 1.8976, + "step": 2624 + }, + { + "epoch": 3.3658536585365852, + "grad_norm": 0.6591815948486328, + "learning_rate": 3.1990611971315003e-06, + "loss": 1.8135, + "step": 2625 + }, + { + "epoch": 3.3671373555840822, + "grad_norm": 0.6575180888175964, + "learning_rate": 3.1863275203439985e-06, + "loss": 1.9122, + "step": 2626 + }, + { + "epoch": 3.3684210526315788, + "grad_norm": 0.6749311089515686, + "learning_rate": 3.1736175117886233e-06, + "loss": 1.832, + "step": 2627 + }, + { + "epoch": 3.3697047496790757, + "grad_norm": 0.6874610781669617, + "learning_rate": 3.1609311852559607e-06, + "loss": 1.8221, + "step": 2628 + }, + { + "epoch": 3.3709884467265727, + "grad_norm": 0.7745242118835449, + "learning_rate": 3.1482685545109e-06, + "loss": 1.8703, + "step": 2629 + }, + { + "epoch": 3.3722721437740693, + "grad_norm": 0.6648337244987488, + "learning_rate": 3.1356296332926388e-06, + "loss": 1.8152, + "step": 2630 + }, + { + "epoch": 3.3735558408215662, + "grad_norm": 0.734294056892395, + "learning_rate": 3.1230144353146223e-06, + "loss": 1.7616, + "step": 2631 + }, + { + "epoch": 3.3748395378690628, + "grad_norm": 0.6978136301040649, + "learning_rate": 3.1104229742645792e-06, + "loss": 1.7525, + "step": 2632 + }, + { + "epoch": 3.3761232349165597, + "grad_norm": 0.6303237676620483, + "learning_rate": 3.0978552638044776e-06, + "loss": 1.9576, + "step": 2633 + }, + { + "epoch": 3.3774069319640567, + "grad_norm": 0.6415073871612549, + "learning_rate": 3.0853113175705108e-06, + "loss": 1.8497, + "step": 2634 + }, + { + "epoch": 3.3786906290115533, + "grad_norm": 0.6855865716934204, + "learning_rate": 3.0727911491730764e-06, + "loss": 1.8144, + "step": 2635 + }, + { + "epoch": 3.3799743260590502, + "grad_norm": 0.6594429612159729, + "learning_rate": 3.0602947721967978e-06, + "loss": 1.8633, + "step": 2636 + }, + { + "epoch": 3.3812580231065468, + "grad_norm": 0.6884453892707825, + "learning_rate": 3.047822200200473e-06, + "loss": 1.882, + "step": 2637 + }, + { + "epoch": 3.3825417201540438, + "grad_norm": 0.6963464021682739, + "learning_rate": 3.03537344671706e-06, + "loss": 1.9252, + "step": 2638 + }, + { + "epoch": 3.3838254172015403, + "grad_norm": 0.6457254886627197, + "learning_rate": 3.0229485252536966e-06, + "loss": 1.7307, + "step": 2639 + }, + { + "epoch": 3.3851091142490373, + "grad_norm": 0.6833950281143188, + "learning_rate": 3.010547449291634e-06, + "loss": 1.8267, + "step": 2640 + }, + { + "epoch": 3.386392811296534, + "grad_norm": 0.6227461695671082, + "learning_rate": 2.998170232286274e-06, + "loss": 1.8181, + "step": 2641 + }, + { + "epoch": 3.387676508344031, + "grad_norm": 0.6513006687164307, + "learning_rate": 2.985816887667123e-06, + "loss": 1.8647, + "step": 2642 + }, + { + "epoch": 3.3889602053915278, + "grad_norm": 0.649646520614624, + "learning_rate": 2.9734874288377873e-06, + "loss": 1.7668, + "step": 2643 + }, + { + "epoch": 3.3902439024390243, + "grad_norm": 0.6144006848335266, + "learning_rate": 2.961181869175944e-06, + "loss": 1.8887, + "step": 2644 + }, + { + "epoch": 3.3915275994865213, + "grad_norm": 0.6675272583961487, + "learning_rate": 2.9489002220333545e-06, + "loss": 1.7511, + "step": 2645 + }, + { + "epoch": 3.392811296534018, + "grad_norm": 0.6306872963905334, + "learning_rate": 2.9366425007358395e-06, + "loss": 1.8314, + "step": 2646 + }, + { + "epoch": 3.394094993581515, + "grad_norm": 0.6446864604949951, + "learning_rate": 2.924408718583238e-06, + "loss": 1.8283, + "step": 2647 + }, + { + "epoch": 3.3953786906290118, + "grad_norm": 0.7040843963623047, + "learning_rate": 2.9121988888494297e-06, + "loss": 1.9371, + "step": 2648 + }, + { + "epoch": 3.3966623876765083, + "grad_norm": 0.72823566198349, + "learning_rate": 2.9000130247823052e-06, + "loss": 1.9235, + "step": 2649 + }, + { + "epoch": 3.3979460847240053, + "grad_norm": 0.6722999215126038, + "learning_rate": 2.887851139603742e-06, + "loss": 1.8111, + "step": 2650 + }, + { + "epoch": 3.399229781771502, + "grad_norm": 0.6660619378089905, + "learning_rate": 2.875713246509612e-06, + "loss": 1.9032, + "step": 2651 + }, + { + "epoch": 3.400513478818999, + "grad_norm": 0.661228358745575, + "learning_rate": 2.8635993586697553e-06, + "loss": 1.851, + "step": 2652 + }, + { + "epoch": 3.4017971758664953, + "grad_norm": 0.6408766508102417, + "learning_rate": 2.851509489227955e-06, + "loss": 1.8168, + "step": 2653 + }, + { + "epoch": 3.4030808729139923, + "grad_norm": 0.6745179295539856, + "learning_rate": 2.8394436513019375e-06, + "loss": 1.8262, + "step": 2654 + }, + { + "epoch": 3.404364569961489, + "grad_norm": 0.6842418313026428, + "learning_rate": 2.8274018579833583e-06, + "loss": 1.9423, + "step": 2655 + }, + { + "epoch": 3.405648267008986, + "grad_norm": 0.6807554364204407, + "learning_rate": 2.815384122337791e-06, + "loss": 1.7621, + "step": 2656 + }, + { + "epoch": 3.406931964056483, + "grad_norm": 0.6784489154815674, + "learning_rate": 2.8033904574046837e-06, + "loss": 1.7761, + "step": 2657 + }, + { + "epoch": 3.4082156611039793, + "grad_norm": 0.6785680651664734, + "learning_rate": 2.7914208761973955e-06, + "loss": 1.8613, + "step": 2658 + }, + { + "epoch": 3.4094993581514763, + "grad_norm": 0.678587019443512, + "learning_rate": 2.7794753917031293e-06, + "loss": 1.8291, + "step": 2659 + }, + { + "epoch": 3.410783055198973, + "grad_norm": 0.652815580368042, + "learning_rate": 2.767554016882956e-06, + "loss": 1.8917, + "step": 2660 + }, + { + "epoch": 3.41206675224647, + "grad_norm": 0.6575085520744324, + "learning_rate": 2.7556567646717907e-06, + "loss": 1.8515, + "step": 2661 + }, + { + "epoch": 3.413350449293967, + "grad_norm": 0.7121369242668152, + "learning_rate": 2.7437836479783654e-06, + "loss": 1.7924, + "step": 2662 + }, + { + "epoch": 3.4146341463414633, + "grad_norm": 0.7083568572998047, + "learning_rate": 2.731934679685222e-06, + "loss": 1.8132, + "step": 2663 + }, + { + "epoch": 3.4159178433889603, + "grad_norm": 0.6503782272338867, + "learning_rate": 2.7201098726487157e-06, + "loss": 1.8174, + "step": 2664 + }, + { + "epoch": 3.417201540436457, + "grad_norm": 0.6855242848396301, + "learning_rate": 2.7083092396989696e-06, + "loss": 1.8873, + "step": 2665 + }, + { + "epoch": 3.418485237483954, + "grad_norm": 0.6668589115142822, + "learning_rate": 2.6965327936398887e-06, + "loss": 1.7637, + "step": 2666 + }, + { + "epoch": 3.419768934531451, + "grad_norm": 0.7007141709327698, + "learning_rate": 2.6847805472491356e-06, + "loss": 1.8823, + "step": 2667 + }, + { + "epoch": 3.4210526315789473, + "grad_norm": 0.6616640090942383, + "learning_rate": 2.6730525132781126e-06, + "loss": 1.819, + "step": 2668 + }, + { + "epoch": 3.4223363286264443, + "grad_norm": 0.6326522827148438, + "learning_rate": 2.661348704451938e-06, + "loss": 1.8063, + "step": 2669 + }, + { + "epoch": 3.423620025673941, + "grad_norm": 0.6677194833755493, + "learning_rate": 2.649669133469468e-06, + "loss": 1.8924, + "step": 2670 + }, + { + "epoch": 3.424903722721438, + "grad_norm": 0.638127326965332, + "learning_rate": 2.638013813003251e-06, + "loss": 1.8208, + "step": 2671 + }, + { + "epoch": 3.4261874197689344, + "grad_norm": 0.6437703967094421, + "learning_rate": 2.626382755699522e-06, + "loss": 1.8065, + "step": 2672 + }, + { + "epoch": 3.4274711168164314, + "grad_norm": 0.6569207906723022, + "learning_rate": 2.6147759741781795e-06, + "loss": 1.7795, + "step": 2673 + }, + { + "epoch": 3.428754813863928, + "grad_norm": 0.6713672876358032, + "learning_rate": 2.603193481032801e-06, + "loss": 1.7629, + "step": 2674 + }, + { + "epoch": 3.430038510911425, + "grad_norm": 0.6681112051010132, + "learning_rate": 2.5916352888306073e-06, + "loss": 1.9494, + "step": 2675 + }, + { + "epoch": 3.431322207958922, + "grad_norm": 0.6852450966835022, + "learning_rate": 2.580101410112437e-06, + "loss": 1.9241, + "step": 2676 + }, + { + "epoch": 3.4326059050064184, + "grad_norm": 0.6879110932350159, + "learning_rate": 2.5685918573927686e-06, + "loss": 1.7415, + "step": 2677 + }, + { + "epoch": 3.4338896020539154, + "grad_norm": 0.6720344424247742, + "learning_rate": 2.5571066431596734e-06, + "loss": 1.8235, + "step": 2678 + }, + { + "epoch": 3.435173299101412, + "grad_norm": 0.6403012275695801, + "learning_rate": 2.545645779874806e-06, + "loss": 1.8131, + "step": 2679 + }, + { + "epoch": 3.436456996148909, + "grad_norm": 0.7156141400337219, + "learning_rate": 2.534209279973432e-06, + "loss": 1.7997, + "step": 2680 + }, + { + "epoch": 3.437740693196406, + "grad_norm": 0.6833835244178772, + "learning_rate": 2.5227971558643537e-06, + "loss": 1.8253, + "step": 2681 + }, + { + "epoch": 3.4390243902439024, + "grad_norm": 0.6780205965042114, + "learning_rate": 2.5114094199299276e-06, + "loss": 1.8243, + "step": 2682 + }, + { + "epoch": 3.4403080872913994, + "grad_norm": 0.7018178701400757, + "learning_rate": 2.5000460845260643e-06, + "loss": 1.7879, + "step": 2683 + }, + { + "epoch": 3.441591784338896, + "grad_norm": 0.6158751845359802, + "learning_rate": 2.4887071619821817e-06, + "loss": 1.8971, + "step": 2684 + }, + { + "epoch": 3.442875481386393, + "grad_norm": 0.6514267921447754, + "learning_rate": 2.477392664601219e-06, + "loss": 1.8301, + "step": 2685 + }, + { + "epoch": 3.4441591784338894, + "grad_norm": 0.7314130663871765, + "learning_rate": 2.4661026046596202e-06, + "loss": 1.8217, + "step": 2686 + }, + { + "epoch": 3.4454428754813864, + "grad_norm": 0.6833764314651489, + "learning_rate": 2.4548369944073004e-06, + "loss": 1.8198, + "step": 2687 + }, + { + "epoch": 3.4467265725288834, + "grad_norm": 0.6534621715545654, + "learning_rate": 2.443595846067645e-06, + "loss": 1.9071, + "step": 2688 + }, + { + "epoch": 3.44801026957638, + "grad_norm": 0.6934870481491089, + "learning_rate": 2.432379171837512e-06, + "loss": 1.9708, + "step": 2689 + }, + { + "epoch": 3.449293966623877, + "grad_norm": 0.720779299736023, + "learning_rate": 2.4211869838872027e-06, + "loss": 1.8333, + "step": 2690 + }, + { + "epoch": 3.4505776636713734, + "grad_norm": 0.679433286190033, + "learning_rate": 2.4100192943604325e-06, + "loss": 1.821, + "step": 2691 + }, + { + "epoch": 3.4518613607188704, + "grad_norm": 0.689063549041748, + "learning_rate": 2.39887611537436e-06, + "loss": 1.859, + "step": 2692 + }, + { + "epoch": 3.453145057766367, + "grad_norm": 0.6851860880851746, + "learning_rate": 2.387757459019527e-06, + "loss": 1.7796, + "step": 2693 + }, + { + "epoch": 3.454428754813864, + "grad_norm": 0.6356110572814941, + "learning_rate": 2.376663337359883e-06, + "loss": 1.7431, + "step": 2694 + }, + { + "epoch": 3.455712451861361, + "grad_norm": 0.6671273112297058, + "learning_rate": 2.365593762432758e-06, + "loss": 1.8488, + "step": 2695 + }, + { + "epoch": 3.4569961489088574, + "grad_norm": 0.6804895997047424, + "learning_rate": 2.354548746248836e-06, + "loss": 1.8167, + "step": 2696 + }, + { + "epoch": 3.4582798459563544, + "grad_norm": 0.6547261476516724, + "learning_rate": 2.3435283007921556e-06, + "loss": 1.893, + "step": 2697 + }, + { + "epoch": 3.459563543003851, + "grad_norm": 0.7350782155990601, + "learning_rate": 2.332532438020113e-06, + "loss": 1.7925, + "step": 2698 + }, + { + "epoch": 3.460847240051348, + "grad_norm": 0.6886659860610962, + "learning_rate": 2.3215611698634065e-06, + "loss": 1.8415, + "step": 2699 + }, + { + "epoch": 3.462130937098845, + "grad_norm": 0.6594906449317932, + "learning_rate": 2.310614508226078e-06, + "loss": 1.8048, + "step": 2700 + }, + { + "epoch": 3.4634146341463414, + "grad_norm": 0.681807816028595, + "learning_rate": 2.299692464985437e-06, + "loss": 1.7946, + "step": 2701 + }, + { + "epoch": 3.4646983311938384, + "grad_norm": 0.6862046718597412, + "learning_rate": 2.2887950519921152e-06, + "loss": 1.8455, + "step": 2702 + }, + { + "epoch": 3.465982028241335, + "grad_norm": 0.7029485106468201, + "learning_rate": 2.2779222810699913e-06, + "loss": 1.7994, + "step": 2703 + }, + { + "epoch": 3.467265725288832, + "grad_norm": 0.6883523464202881, + "learning_rate": 2.267074164016228e-06, + "loss": 1.8207, + "step": 2704 + }, + { + "epoch": 3.4685494223363285, + "grad_norm": 0.642486035823822, + "learning_rate": 2.2562507126012323e-06, + "loss": 1.7513, + "step": 2705 + }, + { + "epoch": 3.4698331193838254, + "grad_norm": 0.6693221926689148, + "learning_rate": 2.2454519385686406e-06, + "loss": 1.7834, + "step": 2706 + }, + { + "epoch": 3.471116816431322, + "grad_norm": 0.6846239566802979, + "learning_rate": 2.234677853635317e-06, + "loss": 1.7737, + "step": 2707 + }, + { + "epoch": 3.472400513478819, + "grad_norm": 0.6837488412857056, + "learning_rate": 2.2239284694913433e-06, + "loss": 1.7561, + "step": 2708 + }, + { + "epoch": 3.473684210526316, + "grad_norm": 0.710226833820343, + "learning_rate": 2.213203797800001e-06, + "loss": 1.7867, + "step": 2709 + }, + { + "epoch": 3.4749679075738125, + "grad_norm": 0.7029847502708435, + "learning_rate": 2.2025038501977486e-06, + "loss": 1.7792, + "step": 2710 + }, + { + "epoch": 3.4762516046213094, + "grad_norm": 0.6867928504943848, + "learning_rate": 2.19182863829423e-06, + "loss": 1.8149, + "step": 2711 + }, + { + "epoch": 3.477535301668806, + "grad_norm": 0.6788585186004639, + "learning_rate": 2.181178173672241e-06, + "loss": 1.7818, + "step": 2712 + }, + { + "epoch": 3.478818998716303, + "grad_norm": 0.6661455631256104, + "learning_rate": 2.170552467887721e-06, + "loss": 1.8117, + "step": 2713 + }, + { + "epoch": 3.4801026957638, + "grad_norm": 0.6487402319908142, + "learning_rate": 2.15995153246977e-06, + "loss": 1.8868, + "step": 2714 + }, + { + "epoch": 3.4813863928112965, + "grad_norm": 0.6551908254623413, + "learning_rate": 2.1493753789205926e-06, + "loss": 1.8799, + "step": 2715 + }, + { + "epoch": 3.4826700898587934, + "grad_norm": 0.6740906238555908, + "learning_rate": 2.138824018715499e-06, + "loss": 1.8732, + "step": 2716 + }, + { + "epoch": 3.48395378690629, + "grad_norm": 0.7061806917190552, + "learning_rate": 2.1282974633029196e-06, + "loss": 1.8408, + "step": 2717 + }, + { + "epoch": 3.485237483953787, + "grad_norm": 0.6638944149017334, + "learning_rate": 2.117795724104346e-06, + "loss": 1.8559, + "step": 2718 + }, + { + "epoch": 3.486521181001284, + "grad_norm": 0.694407045841217, + "learning_rate": 2.107318812514364e-06, + "loss": 1.8178, + "step": 2719 + }, + { + "epoch": 3.4878048780487805, + "grad_norm": 0.6622540354728699, + "learning_rate": 2.0968667399006158e-06, + "loss": 1.8414, + "step": 2720 + }, + { + "epoch": 3.4890885750962775, + "grad_norm": 0.683251142501831, + "learning_rate": 2.086439517603789e-06, + "loss": 1.7489, + "step": 2721 + }, + { + "epoch": 3.490372272143774, + "grad_norm": 0.6841902136802673, + "learning_rate": 2.0760371569376017e-06, + "loss": 1.9183, + "step": 2722 + }, + { + "epoch": 3.491655969191271, + "grad_norm": 0.6584630012512207, + "learning_rate": 2.065659669188813e-06, + "loss": 1.7601, + "step": 2723 + }, + { + "epoch": 3.4929396662387675, + "grad_norm": 0.6510747075080872, + "learning_rate": 2.0553070656171873e-06, + "loss": 1.7219, + "step": 2724 + }, + { + "epoch": 3.4942233632862645, + "grad_norm": 0.6634497046470642, + "learning_rate": 2.0449793574554842e-06, + "loss": 1.7809, + "step": 2725 + }, + { + "epoch": 3.495507060333761, + "grad_norm": 0.7119478583335876, + "learning_rate": 2.0346765559094567e-06, + "loss": 1.8456, + "step": 2726 + }, + { + "epoch": 3.496790757381258, + "grad_norm": 0.7097324728965759, + "learning_rate": 2.024398672157829e-06, + "loss": 1.9038, + "step": 2727 + }, + { + "epoch": 3.498074454428755, + "grad_norm": 0.6528558135032654, + "learning_rate": 2.0141457173523005e-06, + "loss": 1.8047, + "step": 2728 + }, + { + "epoch": 3.4993581514762515, + "grad_norm": 0.6872973442077637, + "learning_rate": 2.0039177026175053e-06, + "loss": 1.7995, + "step": 2729 + }, + { + "epoch": 3.5006418485237485, + "grad_norm": 0.6781868934631348, + "learning_rate": 1.9937146390510307e-06, + "loss": 1.8211, + "step": 2730 + }, + { + "epoch": 3.501925545571245, + "grad_norm": 0.6640962362289429, + "learning_rate": 1.9835365377233878e-06, + "loss": 1.7747, + "step": 2731 + }, + { + "epoch": 3.503209242618742, + "grad_norm": 0.6422116756439209, + "learning_rate": 1.9733834096779926e-06, + "loss": 1.9304, + "step": 2732 + }, + { + "epoch": 3.504492939666239, + "grad_norm": 0.697226881980896, + "learning_rate": 1.96325526593118e-06, + "loss": 1.8982, + "step": 2733 + }, + { + "epoch": 3.5057766367137355, + "grad_norm": 0.7131691575050354, + "learning_rate": 1.953152117472176e-06, + "loss": 1.9915, + "step": 2734 + }, + { + "epoch": 3.5070603337612325, + "grad_norm": 0.7035425901412964, + "learning_rate": 1.943073975263074e-06, + "loss": 1.8454, + "step": 2735 + }, + { + "epoch": 3.508344030808729, + "grad_norm": 0.6722110509872437, + "learning_rate": 1.933020850238845e-06, + "loss": 1.8848, + "step": 2736 + }, + { + "epoch": 3.509627727856226, + "grad_norm": 0.697409987449646, + "learning_rate": 1.9229927533073094e-06, + "loss": 1.8976, + "step": 2737 + }, + { + "epoch": 3.510911424903723, + "grad_norm": 0.6401705741882324, + "learning_rate": 1.912989695349138e-06, + "loss": 1.7914, + "step": 2738 + }, + { + "epoch": 3.5121951219512195, + "grad_norm": 0.7147530317306519, + "learning_rate": 1.9030116872178316e-06, + "loss": 1.8388, + "step": 2739 + }, + { + "epoch": 3.513478818998716, + "grad_norm": 0.6942839622497559, + "learning_rate": 1.8930587397397114e-06, + "loss": 1.8171, + "step": 2740 + }, + { + "epoch": 3.514762516046213, + "grad_norm": 0.6534262895584106, + "learning_rate": 1.8831308637139e-06, + "loss": 1.8734, + "step": 2741 + }, + { + "epoch": 3.51604621309371, + "grad_norm": 0.6813603043556213, + "learning_rate": 1.8732280699123305e-06, + "loss": 1.7927, + "step": 2742 + }, + { + "epoch": 3.5173299101412066, + "grad_norm": 0.6672655344009399, + "learning_rate": 1.8633503690797167e-06, + "loss": 1.8408, + "step": 2743 + }, + { + "epoch": 3.5186136071887035, + "grad_norm": 0.6549263000488281, + "learning_rate": 1.8534977719335351e-06, + "loss": 1.8183, + "step": 2744 + }, + { + "epoch": 3.5198973042362, + "grad_norm": 0.6611000895500183, + "learning_rate": 1.8436702891640401e-06, + "loss": 1.8982, + "step": 2745 + }, + { + "epoch": 3.521181001283697, + "grad_norm": 0.6557934284210205, + "learning_rate": 1.8338679314342272e-06, + "loss": 1.8413, + "step": 2746 + }, + { + "epoch": 3.522464698331194, + "grad_norm": 0.6668517589569092, + "learning_rate": 1.8240907093798332e-06, + "loss": 1.8796, + "step": 2747 + }, + { + "epoch": 3.5237483953786906, + "grad_norm": 0.6254555583000183, + "learning_rate": 1.8143386336093248e-06, + "loss": 1.8753, + "step": 2748 + }, + { + "epoch": 3.5250320924261875, + "grad_norm": 0.6814380288124084, + "learning_rate": 1.80461171470388e-06, + "loss": 1.7818, + "step": 2749 + }, + { + "epoch": 3.526315789473684, + "grad_norm": 0.6406975388526917, + "learning_rate": 1.7949099632173783e-06, + "loss": 1.8044, + "step": 2750 + }, + { + "epoch": 3.527599486521181, + "grad_norm": 0.7635696530342102, + "learning_rate": 1.7852333896764079e-06, + "loss": 1.7932, + "step": 2751 + }, + { + "epoch": 3.528883183568678, + "grad_norm": 0.6396098136901855, + "learning_rate": 1.7755820045802145e-06, + "loss": 1.7891, + "step": 2752 + }, + { + "epoch": 3.5301668806161746, + "grad_norm": 0.7031031250953674, + "learning_rate": 1.7659558184007386e-06, + "loss": 1.8236, + "step": 2753 + }, + { + "epoch": 3.531450577663671, + "grad_norm": 0.6767717003822327, + "learning_rate": 1.7563548415825554e-06, + "loss": 1.8881, + "step": 2754 + }, + { + "epoch": 3.532734274711168, + "grad_norm": 0.6459426879882812, + "learning_rate": 1.7467790845429126e-06, + "loss": 1.8017, + "step": 2755 + }, + { + "epoch": 3.534017971758665, + "grad_norm": 0.6687619090080261, + "learning_rate": 1.7372285576716717e-06, + "loss": 1.8613, + "step": 2756 + }, + { + "epoch": 3.5353016688061616, + "grad_norm": 0.6281467080116272, + "learning_rate": 1.7277032713313295e-06, + "loss": 1.8091, + "step": 2757 + }, + { + "epoch": 3.5365853658536586, + "grad_norm": 0.6684985160827637, + "learning_rate": 1.7182032358570022e-06, + "loss": 1.8329, + "step": 2758 + }, + { + "epoch": 3.537869062901155, + "grad_norm": 0.6524038314819336, + "learning_rate": 1.708728461556397e-06, + "loss": 1.8656, + "step": 2759 + }, + { + "epoch": 3.539152759948652, + "grad_norm": 0.6794835329055786, + "learning_rate": 1.6992789587098129e-06, + "loss": 1.7846, + "step": 2760 + }, + { + "epoch": 3.540436456996149, + "grad_norm": 0.6298161149024963, + "learning_rate": 1.6898547375701346e-06, + "loss": 1.9021, + "step": 2761 + }, + { + "epoch": 3.5417201540436456, + "grad_norm": 0.7223351001739502, + "learning_rate": 1.6804558083628186e-06, + "loss": 1.8203, + "step": 2762 + }, + { + "epoch": 3.5430038510911426, + "grad_norm": 0.6623315811157227, + "learning_rate": 1.6710821812858656e-06, + "loss": 1.8658, + "step": 2763 + }, + { + "epoch": 3.544287548138639, + "grad_norm": 0.7143245339393616, + "learning_rate": 1.661733866509843e-06, + "loss": 1.8473, + "step": 2764 + }, + { + "epoch": 3.545571245186136, + "grad_norm": 0.704624593257904, + "learning_rate": 1.6524108741778372e-06, + "loss": 1.8253, + "step": 2765 + }, + { + "epoch": 3.546854942233633, + "grad_norm": 0.6446236968040466, + "learning_rate": 1.643113214405456e-06, + "loss": 1.7553, + "step": 2766 + }, + { + "epoch": 3.5481386392811296, + "grad_norm": 0.6443824172019958, + "learning_rate": 1.6338408972808473e-06, + "loss": 1.8019, + "step": 2767 + }, + { + "epoch": 3.5494223363286266, + "grad_norm": 0.641877293586731, + "learning_rate": 1.624593932864632e-06, + "loss": 1.849, + "step": 2768 + }, + { + "epoch": 3.550706033376123, + "grad_norm": 0.6798678040504456, + "learning_rate": 1.615372331189935e-06, + "loss": 1.908, + "step": 2769 + }, + { + "epoch": 3.55198973042362, + "grad_norm": 0.6474749445915222, + "learning_rate": 1.606176102262369e-06, + "loss": 1.7798, + "step": 2770 + }, + { + "epoch": 3.553273427471117, + "grad_norm": 0.6328116059303284, + "learning_rate": 1.5970052560599997e-06, + "loss": 1.8755, + "step": 2771 + }, + { + "epoch": 3.5545571245186136, + "grad_norm": 0.6635220050811768, + "learning_rate": 1.587859802533373e-06, + "loss": 1.8087, + "step": 2772 + }, + { + "epoch": 3.55584082156611, + "grad_norm": 0.6252384781837463, + "learning_rate": 1.578739751605468e-06, + "loss": 1.7724, + "step": 2773 + }, + { + "epoch": 3.557124518613607, + "grad_norm": 0.6702415943145752, + "learning_rate": 1.5696451131717082e-06, + "loss": 1.7883, + "step": 2774 + }, + { + "epoch": 3.558408215661104, + "grad_norm": 0.6448876261711121, + "learning_rate": 1.5605758970999362e-06, + "loss": 1.7617, + "step": 2775 + }, + { + "epoch": 3.5596919127086006, + "grad_norm": 0.6451162695884705, + "learning_rate": 1.5515321132304194e-06, + "loss": 1.8379, + "step": 2776 + }, + { + "epoch": 3.5609756097560976, + "grad_norm": 0.7123709917068481, + "learning_rate": 1.5425137713758337e-06, + "loss": 1.9099, + "step": 2777 + }, + { + "epoch": 3.562259306803594, + "grad_norm": 0.6250616908073425, + "learning_rate": 1.5335208813212375e-06, + "loss": 1.8362, + "step": 2778 + }, + { + "epoch": 3.563543003851091, + "grad_norm": 0.6640663146972656, + "learning_rate": 1.5245534528240846e-06, + "loss": 1.8416, + "step": 2779 + }, + { + "epoch": 3.564826700898588, + "grad_norm": 0.6926436424255371, + "learning_rate": 1.5156114956141942e-06, + "loss": 1.8774, + "step": 2780 + }, + { + "epoch": 3.5661103979460846, + "grad_norm": 0.6474257111549377, + "learning_rate": 1.5066950193937585e-06, + "loss": 1.8891, + "step": 2781 + }, + { + "epoch": 3.5673940949935816, + "grad_norm": 0.6635065674781799, + "learning_rate": 1.4978040338373077e-06, + "loss": 1.8382, + "step": 2782 + }, + { + "epoch": 3.568677792041078, + "grad_norm": 0.6670807003974915, + "learning_rate": 1.488938548591734e-06, + "loss": 1.8365, + "step": 2783 + }, + { + "epoch": 3.569961489088575, + "grad_norm": 0.6843037009239197, + "learning_rate": 1.4800985732762452e-06, + "loss": 1.7375, + "step": 2784 + }, + { + "epoch": 3.571245186136072, + "grad_norm": 0.6189756989479065, + "learning_rate": 1.4712841174823716e-06, + "loss": 1.8769, + "step": 2785 + }, + { + "epoch": 3.5725288831835686, + "grad_norm": 0.7124006748199463, + "learning_rate": 1.4624951907739631e-06, + "loss": 1.7842, + "step": 2786 + }, + { + "epoch": 3.5738125802310656, + "grad_norm": 0.7035257816314697, + "learning_rate": 1.4537318026871694e-06, + "loss": 1.8423, + "step": 2787 + }, + { + "epoch": 3.575096277278562, + "grad_norm": 0.6559979915618896, + "learning_rate": 1.4449939627304148e-06, + "loss": 1.8534, + "step": 2788 + }, + { + "epoch": 3.576379974326059, + "grad_norm": 0.7083045244216919, + "learning_rate": 1.436281680384427e-06, + "loss": 1.794, + "step": 2789 + }, + { + "epoch": 3.5776636713735557, + "grad_norm": 0.675186038017273, + "learning_rate": 1.4275949651021803e-06, + "loss": 1.8944, + "step": 2790 + }, + { + "epoch": 3.5789473684210527, + "grad_norm": 0.6736690998077393, + "learning_rate": 1.4189338263089241e-06, + "loss": 1.8598, + "step": 2791 + }, + { + "epoch": 3.580231065468549, + "grad_norm": 0.6566684246063232, + "learning_rate": 1.410298273402158e-06, + "loss": 1.8749, + "step": 2792 + }, + { + "epoch": 3.581514762516046, + "grad_norm": 0.6833084225654602, + "learning_rate": 1.4016883157516064e-06, + "loss": 1.7784, + "step": 2793 + }, + { + "epoch": 3.582798459563543, + "grad_norm": 0.7400025725364685, + "learning_rate": 1.3931039626992297e-06, + "loss": 1.8944, + "step": 2794 + }, + { + "epoch": 3.5840821566110397, + "grad_norm": 0.6538530588150024, + "learning_rate": 1.3845452235592083e-06, + "loss": 1.8075, + "step": 2795 + }, + { + "epoch": 3.5853658536585367, + "grad_norm": 0.6660887002944946, + "learning_rate": 1.3760121076179383e-06, + "loss": 1.9065, + "step": 2796 + }, + { + "epoch": 3.586649550706033, + "grad_norm": 0.6582172513008118, + "learning_rate": 1.3675046241339918e-06, + "loss": 1.8322, + "step": 2797 + }, + { + "epoch": 3.58793324775353, + "grad_norm": 0.6685164570808411, + "learning_rate": 1.3590227823381568e-06, + "loss": 1.767, + "step": 2798 + }, + { + "epoch": 3.589216944801027, + "grad_norm": 0.7221265435218811, + "learning_rate": 1.3505665914333805e-06, + "loss": 1.8029, + "step": 2799 + }, + { + "epoch": 3.5905006418485237, + "grad_norm": 0.6410666108131409, + "learning_rate": 1.3421360605947757e-06, + "loss": 1.7677, + "step": 2800 + }, + { + "epoch": 3.5917843388960207, + "grad_norm": 0.6532052159309387, + "learning_rate": 1.33373119896964e-06, + "loss": 1.8786, + "step": 2801 + }, + { + "epoch": 3.593068035943517, + "grad_norm": 0.6443427801132202, + "learning_rate": 1.3253520156773891e-06, + "loss": 1.7784, + "step": 2802 + }, + { + "epoch": 3.594351732991014, + "grad_norm": 0.6705851554870605, + "learning_rate": 1.3169985198095897e-06, + "loss": 1.7793, + "step": 2803 + }, + { + "epoch": 3.595635430038511, + "grad_norm": 0.7572922706604004, + "learning_rate": 1.3086707204299414e-06, + "loss": 1.8356, + "step": 2804 + }, + { + "epoch": 3.5969191270860077, + "grad_norm": 0.6533336043357849, + "learning_rate": 1.3003686265742525e-06, + "loss": 1.8044, + "step": 2805 + }, + { + "epoch": 3.5982028241335042, + "grad_norm": 0.6709038615226746, + "learning_rate": 1.292092247250451e-06, + "loss": 1.8291, + "step": 2806 + }, + { + "epoch": 3.599486521181001, + "grad_norm": 0.6841021180152893, + "learning_rate": 1.2838415914385616e-06, + "loss": 1.8653, + "step": 2807 + }, + { + "epoch": 3.600770218228498, + "grad_norm": 0.644106924533844, + "learning_rate": 1.275616668090693e-06, + "loss": 1.8094, + "step": 2808 + }, + { + "epoch": 3.6020539152759947, + "grad_norm": 0.707465648651123, + "learning_rate": 1.267417486131034e-06, + "loss": 1.9545, + "step": 2809 + }, + { + "epoch": 3.6033376123234917, + "grad_norm": 0.7002444863319397, + "learning_rate": 1.2592440544558493e-06, + "loss": 1.7835, + "step": 2810 + }, + { + "epoch": 3.6046213093709882, + "grad_norm": 0.6951387524604797, + "learning_rate": 1.2510963819334615e-06, + "loss": 1.8666, + "step": 2811 + }, + { + "epoch": 3.605905006418485, + "grad_norm": 0.6568358540534973, + "learning_rate": 1.2429744774042463e-06, + "loss": 1.9194, + "step": 2812 + }, + { + "epoch": 3.607188703465982, + "grad_norm": 0.7096032500267029, + "learning_rate": 1.23487834968061e-06, + "loss": 1.7837, + "step": 2813 + }, + { + "epoch": 3.6084724005134787, + "grad_norm": 0.7023941874504089, + "learning_rate": 1.2268080075470035e-06, + "loss": 1.8923, + "step": 2814 + }, + { + "epoch": 3.6097560975609757, + "grad_norm": 0.6400555968284607, + "learning_rate": 1.2187634597599001e-06, + "loss": 1.8305, + "step": 2815 + }, + { + "epoch": 3.6110397946084722, + "grad_norm": 0.8000922203063965, + "learning_rate": 1.2107447150477735e-06, + "loss": 1.8297, + "step": 2816 + }, + { + "epoch": 3.612323491655969, + "grad_norm": 0.6877830624580383, + "learning_rate": 1.2027517821111112e-06, + "loss": 1.7882, + "step": 2817 + }, + { + "epoch": 3.613607188703466, + "grad_norm": 0.6941173076629639, + "learning_rate": 1.1947846696223896e-06, + "loss": 1.8609, + "step": 2818 + }, + { + "epoch": 3.6148908857509627, + "grad_norm": 0.7106480002403259, + "learning_rate": 1.1868433862260719e-06, + "loss": 1.8372, + "step": 2819 + }, + { + "epoch": 3.6161745827984597, + "grad_norm": 0.6833379864692688, + "learning_rate": 1.1789279405385905e-06, + "loss": 1.8525, + "step": 2820 + }, + { + "epoch": 3.6174582798459562, + "grad_norm": 0.6946989893913269, + "learning_rate": 1.1710383411483588e-06, + "loss": 1.8425, + "step": 2821 + }, + { + "epoch": 3.6187419768934532, + "grad_norm": 0.6517040133476257, + "learning_rate": 1.1631745966157266e-06, + "loss": 1.7653, + "step": 2822 + }, + { + "epoch": 3.62002567394095, + "grad_norm": 0.6516625285148621, + "learning_rate": 1.1553367154730044e-06, + "loss": 1.8009, + "step": 2823 + }, + { + "epoch": 3.6213093709884467, + "grad_norm": 0.6811094284057617, + "learning_rate": 1.1475247062244344e-06, + "loss": 1.8407, + "step": 2824 + }, + { + "epoch": 3.6225930680359433, + "grad_norm": 0.7301627993583679, + "learning_rate": 1.1397385773461861e-06, + "loss": 1.8078, + "step": 2825 + }, + { + "epoch": 3.6238767650834403, + "grad_norm": 0.662013590335846, + "learning_rate": 1.1319783372863602e-06, + "loss": 1.8568, + "step": 2826 + }, + { + "epoch": 3.6251604621309372, + "grad_norm": 0.6447609663009644, + "learning_rate": 1.124243994464952e-06, + "loss": 1.8007, + "step": 2827 + }, + { + "epoch": 3.6264441591784338, + "grad_norm": 0.6389984488487244, + "learning_rate": 1.1165355572738622e-06, + "loss": 1.7853, + "step": 2828 + }, + { + "epoch": 3.6277278562259307, + "grad_norm": 0.7189795970916748, + "learning_rate": 1.108853034076887e-06, + "loss": 1.8461, + "step": 2829 + }, + { + "epoch": 3.6290115532734273, + "grad_norm": 0.6491907835006714, + "learning_rate": 1.1011964332097114e-06, + "loss": 1.8163, + "step": 2830 + }, + { + "epoch": 3.6302952503209243, + "grad_norm": 0.6397409439086914, + "learning_rate": 1.0935657629798758e-06, + "loss": 1.7216, + "step": 2831 + }, + { + "epoch": 3.6315789473684212, + "grad_norm": 0.6603519916534424, + "learning_rate": 1.0859610316668073e-06, + "loss": 1.7593, + "step": 2832 + }, + { + "epoch": 3.6328626444159178, + "grad_norm": 0.6851031184196472, + "learning_rate": 1.0783822475217698e-06, + "loss": 1.8953, + "step": 2833 + }, + { + "epoch": 3.6341463414634148, + "grad_norm": 0.6895718574523926, + "learning_rate": 1.0708294187678826e-06, + "loss": 1.8338, + "step": 2834 + }, + { + "epoch": 3.6354300385109113, + "grad_norm": 0.6721134185791016, + "learning_rate": 1.0633025536001069e-06, + "loss": 1.774, + "step": 2835 + }, + { + "epoch": 3.6367137355584083, + "grad_norm": 0.7149147391319275, + "learning_rate": 1.0558016601852266e-06, + "loss": 1.8334, + "step": 2836 + }, + { + "epoch": 3.6379974326059052, + "grad_norm": 0.679861843585968, + "learning_rate": 1.0483267466618485e-06, + "loss": 1.836, + "step": 2837 + }, + { + "epoch": 3.639281129653402, + "grad_norm": 0.6761167645454407, + "learning_rate": 1.0408778211403847e-06, + "loss": 1.9264, + "step": 2838 + }, + { + "epoch": 3.6405648267008983, + "grad_norm": 0.725320041179657, + "learning_rate": 1.0334548917030618e-06, + "loss": 1.8512, + "step": 2839 + }, + { + "epoch": 3.6418485237483953, + "grad_norm": 0.6159487366676331, + "learning_rate": 1.0260579664038928e-06, + "loss": 1.7774, + "step": 2840 + }, + { + "epoch": 3.6431322207958923, + "grad_norm": 0.6713334321975708, + "learning_rate": 1.0186870532686742e-06, + "loss": 1.8322, + "step": 2841 + }, + { + "epoch": 3.644415917843389, + "grad_norm": 0.7280192971229553, + "learning_rate": 1.0113421602949813e-06, + "loss": 1.8075, + "step": 2842 + }, + { + "epoch": 3.645699614890886, + "grad_norm": 0.6649914979934692, + "learning_rate": 1.0040232954521557e-06, + "loss": 1.7793, + "step": 2843 + }, + { + "epoch": 3.6469833119383823, + "grad_norm": 0.7348213195800781, + "learning_rate": 9.967304666813005e-07, + "loss": 1.9435, + "step": 2844 + }, + { + "epoch": 3.6482670089858793, + "grad_norm": 0.6966539025306702, + "learning_rate": 9.894636818952697e-07, + "loss": 1.7796, + "step": 2845 + }, + { + "epoch": 3.6495507060333763, + "grad_norm": 0.673272967338562, + "learning_rate": 9.822229489786588e-07, + "loss": 1.8737, + "step": 2846 + }, + { + "epoch": 3.650834403080873, + "grad_norm": 0.6700506806373596, + "learning_rate": 9.750082757877837e-07, + "loss": 1.7015, + "step": 2847 + }, + { + "epoch": 3.65211810012837, + "grad_norm": 0.6828154921531677, + "learning_rate": 9.678196701507076e-07, + "loss": 1.8316, + "step": 2848 + }, + { + "epoch": 3.6534017971758663, + "grad_norm": 0.6462768316268921, + "learning_rate": 9.606571398671966e-07, + "loss": 1.7308, + "step": 2849 + }, + { + "epoch": 3.6546854942233633, + "grad_norm": 0.744316816329956, + "learning_rate": 9.535206927087231e-07, + "loss": 1.8945, + "step": 2850 + }, + { + "epoch": 3.6559691912708603, + "grad_norm": 0.713550865650177, + "learning_rate": 9.464103364184657e-07, + "loss": 1.7941, + "step": 2851 + }, + { + "epoch": 3.657252888318357, + "grad_norm": 0.6467413902282715, + "learning_rate": 9.393260787112918e-07, + "loss": 1.7632, + "step": 2852 + }, + { + "epoch": 3.658536585365854, + "grad_norm": 0.7298699617385864, + "learning_rate": 9.322679272737416e-07, + "loss": 1.8637, + "step": 2853 + }, + { + "epoch": 3.6598202824133503, + "grad_norm": 0.7114826440811157, + "learning_rate": 9.252358897640451e-07, + "loss": 1.7621, + "step": 2854 + }, + { + "epoch": 3.6611039794608473, + "grad_norm": 0.6848855018615723, + "learning_rate": 9.182299738120931e-07, + "loss": 1.8408, + "step": 2855 + }, + { + "epoch": 3.6623876765083443, + "grad_norm": 0.703012228012085, + "learning_rate": 9.112501870194273e-07, + "loss": 1.8154, + "step": 2856 + }, + { + "epoch": 3.663671373555841, + "grad_norm": 0.6660111546516418, + "learning_rate": 9.042965369592454e-07, + "loss": 1.7996, + "step": 2857 + }, + { + "epoch": 3.6649550706033374, + "grad_norm": 0.7060369253158569, + "learning_rate": 8.973690311763816e-07, + "loss": 1.717, + "step": 2858 + }, + { + "epoch": 3.6662387676508343, + "grad_norm": 0.7304860949516296, + "learning_rate": 8.904676771873094e-07, + "loss": 1.9723, + "step": 2859 + }, + { + "epoch": 3.6675224646983313, + "grad_norm": 0.7360041737556458, + "learning_rate": 8.83592482480125e-07, + "loss": 1.8617, + "step": 2860 + }, + { + "epoch": 3.668806161745828, + "grad_norm": 0.7998003363609314, + "learning_rate": 8.767434545145364e-07, + "loss": 1.8342, + "step": 2861 + }, + { + "epoch": 3.670089858793325, + "grad_norm": 0.6786078214645386, + "learning_rate": 8.699206007218657e-07, + "loss": 1.7917, + "step": 2862 + }, + { + "epoch": 3.6713735558408214, + "grad_norm": 0.7264149188995361, + "learning_rate": 8.631239285050302e-07, + "loss": 1.854, + "step": 2863 + }, + { + "epoch": 3.6726572528883183, + "grad_norm": 0.6646893620491028, + "learning_rate": 8.563534452385474e-07, + "loss": 1.7923, + "step": 2864 + }, + { + "epoch": 3.6739409499358153, + "grad_norm": 0.6921852827072144, + "learning_rate": 8.496091582685162e-07, + "loss": 1.8372, + "step": 2865 + }, + { + "epoch": 3.675224646983312, + "grad_norm": 0.6729761362075806, + "learning_rate": 8.428910749126023e-07, + "loss": 1.861, + "step": 2866 + }, + { + "epoch": 3.676508344030809, + "grad_norm": 0.7467935681343079, + "learning_rate": 8.361992024600557e-07, + "loss": 1.7632, + "step": 2867 + }, + { + "epoch": 3.6777920410783054, + "grad_norm": 0.7036812901496887, + "learning_rate": 8.29533548171682e-07, + "loss": 1.8148, + "step": 2868 + }, + { + "epoch": 3.6790757381258024, + "grad_norm": 0.6987660527229309, + "learning_rate": 8.228941192798323e-07, + "loss": 1.8086, + "step": 2869 + }, + { + "epoch": 3.6803594351732993, + "grad_norm": 0.7198107838630676, + "learning_rate": 8.162809229884105e-07, + "loss": 1.8596, + "step": 2870 + }, + { + "epoch": 3.681643132220796, + "grad_norm": 0.7107940316200256, + "learning_rate": 8.096939664728547e-07, + "loss": 1.8165, + "step": 2871 + }, + { + "epoch": 3.682926829268293, + "grad_norm": 0.6722975969314575, + "learning_rate": 8.031332568801287e-07, + "loss": 1.7402, + "step": 2872 + }, + { + "epoch": 3.6842105263157894, + "grad_norm": 0.6462680697441101, + "learning_rate": 7.965988013287267e-07, + "loss": 1.8718, + "step": 2873 + }, + { + "epoch": 3.6854942233632864, + "grad_norm": 0.6885252594947815, + "learning_rate": 7.900906069086528e-07, + "loss": 1.8498, + "step": 2874 + }, + { + "epoch": 3.686777920410783, + "grad_norm": 0.6483273506164551, + "learning_rate": 7.836086806814108e-07, + "loss": 1.6875, + "step": 2875 + }, + { + "epoch": 3.68806161745828, + "grad_norm": 0.7190673351287842, + "learning_rate": 7.771530296800167e-07, + "loss": 1.7809, + "step": 2876 + }, + { + "epoch": 3.6893453145057764, + "grad_norm": 0.6884718537330627, + "learning_rate": 7.707236609089647e-07, + "loss": 1.7848, + "step": 2877 + }, + { + "epoch": 3.6906290115532734, + "grad_norm": 0.609427809715271, + "learning_rate": 7.643205813442356e-07, + "loss": 1.8092, + "step": 2878 + }, + { + "epoch": 3.6919127086007704, + "grad_norm": 0.6686990261077881, + "learning_rate": 7.579437979332915e-07, + "loss": 1.7761, + "step": 2879 + }, + { + "epoch": 3.693196405648267, + "grad_norm": 0.6748306155204773, + "learning_rate": 7.51593317595059e-07, + "loss": 1.8133, + "step": 2880 + }, + { + "epoch": 3.694480102695764, + "grad_norm": 0.6457929611206055, + "learning_rate": 7.45269147219918e-07, + "loss": 1.7673, + "step": 2881 + }, + { + "epoch": 3.6957637997432604, + "grad_norm": 0.7032188773155212, + "learning_rate": 7.389712936697129e-07, + "loss": 1.7109, + "step": 2882 + }, + { + "epoch": 3.6970474967907574, + "grad_norm": 0.7396710515022278, + "learning_rate": 7.326997637777333e-07, + "loss": 1.7983, + "step": 2883 + }, + { + "epoch": 3.6983311938382544, + "grad_norm": 0.6629140973091125, + "learning_rate": 7.264545643486997e-07, + "loss": 1.8262, + "step": 2884 + }, + { + "epoch": 3.699614890885751, + "grad_norm": 0.6856799721717834, + "learning_rate": 7.202357021587669e-07, + "loss": 1.8761, + "step": 2885 + }, + { + "epoch": 3.700898587933248, + "grad_norm": 0.6479711532592773, + "learning_rate": 7.14043183955515e-07, + "loss": 1.7488, + "step": 2886 + }, + { + "epoch": 3.7021822849807444, + "grad_norm": 0.6708347201347351, + "learning_rate": 7.078770164579335e-07, + "loss": 1.8556, + "step": 2887 + }, + { + "epoch": 3.7034659820282414, + "grad_norm": 0.7006207704544067, + "learning_rate": 7.01737206356437e-07, + "loss": 1.8954, + "step": 2888 + }, + { + "epoch": 3.7047496790757384, + "grad_norm": 0.8129990100860596, + "learning_rate": 6.956237603128274e-07, + "loss": 1.8819, + "step": 2889 + }, + { + "epoch": 3.706033376123235, + "grad_norm": 0.6833080053329468, + "learning_rate": 6.895366849603014e-07, + "loss": 1.7669, + "step": 2890 + }, + { + "epoch": 3.7073170731707314, + "grad_norm": 0.6700402498245239, + "learning_rate": 6.834759869034452e-07, + "loss": 1.8312, + "step": 2891 + }, + { + "epoch": 3.7086007702182284, + "grad_norm": 0.6441825032234192, + "learning_rate": 6.77441672718232e-07, + "loss": 1.8106, + "step": 2892 + }, + { + "epoch": 3.7098844672657254, + "grad_norm": 0.6286627054214478, + "learning_rate": 6.714337489520023e-07, + "loss": 1.848, + "step": 2893 + }, + { + "epoch": 3.711168164313222, + "grad_norm": 0.7012709379196167, + "learning_rate": 6.65452222123461e-07, + "loss": 1.8267, + "step": 2894 + }, + { + "epoch": 3.712451861360719, + "grad_norm": 0.6759612560272217, + "learning_rate": 6.59497098722675e-07, + "loss": 1.8105, + "step": 2895 + }, + { + "epoch": 3.7137355584082155, + "grad_norm": 0.6712021827697754, + "learning_rate": 6.535683852110563e-07, + "loss": 1.7602, + "step": 2896 + }, + { + "epoch": 3.7150192554557124, + "grad_norm": 0.6660760641098022, + "learning_rate": 6.476660880213758e-07, + "loss": 1.8573, + "step": 2897 + }, + { + "epoch": 3.7163029525032094, + "grad_norm": 0.713329553604126, + "learning_rate": 6.417902135577275e-07, + "loss": 1.8619, + "step": 2898 + }, + { + "epoch": 3.717586649550706, + "grad_norm": 0.6891172528266907, + "learning_rate": 6.359407681955476e-07, + "loss": 1.8114, + "step": 2899 + }, + { + "epoch": 3.718870346598203, + "grad_norm": 0.6934365034103394, + "learning_rate": 6.301177582815814e-07, + "loss": 1.8197, + "step": 2900 + }, + { + "epoch": 3.7201540436456995, + "grad_norm": 0.7213584780693054, + "learning_rate": 6.243211901339086e-07, + "loss": 1.8469, + "step": 2901 + }, + { + "epoch": 3.7214377406931964, + "grad_norm": 0.647114098072052, + "learning_rate": 6.185510700419117e-07, + "loss": 1.8637, + "step": 2902 + }, + { + "epoch": 3.7227214377406934, + "grad_norm": 0.7151072025299072, + "learning_rate": 6.128074042662718e-07, + "loss": 1.8257, + "step": 2903 + }, + { + "epoch": 3.72400513478819, + "grad_norm": 0.6614383459091187, + "learning_rate": 6.070901990389733e-07, + "loss": 1.7997, + "step": 2904 + }, + { + "epoch": 3.725288831835687, + "grad_norm": 0.7178487181663513, + "learning_rate": 6.0139946056329e-07, + "loss": 1.7377, + "step": 2905 + }, + { + "epoch": 3.7265725288831835, + "grad_norm": 0.7142190933227539, + "learning_rate": 5.957351950137719e-07, + "loss": 1.8802, + "step": 2906 + }, + { + "epoch": 3.7278562259306804, + "grad_norm": 0.6852755546569824, + "learning_rate": 5.900974085362499e-07, + "loss": 1.8491, + "step": 2907 + }, + { + "epoch": 3.7291399229781774, + "grad_norm": 0.642707884311676, + "learning_rate": 5.844861072478336e-07, + "loss": 1.7999, + "step": 2908 + }, + { + "epoch": 3.730423620025674, + "grad_norm": 0.695456862449646, + "learning_rate": 5.789012972368779e-07, + "loss": 1.7905, + "step": 2909 + }, + { + "epoch": 3.7317073170731705, + "grad_norm": 0.6972944736480713, + "learning_rate": 5.733429845630106e-07, + "loss": 1.8205, + "step": 2910 + }, + { + "epoch": 3.7329910141206675, + "grad_norm": 0.7139705419540405, + "learning_rate": 5.678111752570964e-07, + "loss": 1.8791, + "step": 2911 + }, + { + "epoch": 3.7342747111681645, + "grad_norm": 0.6532818078994751, + "learning_rate": 5.623058753212507e-07, + "loss": 1.785, + "step": 2912 + }, + { + "epoch": 3.735558408215661, + "grad_norm": 0.6599393486976624, + "learning_rate": 5.568270907288287e-07, + "loss": 1.7929, + "step": 2913 + }, + { + "epoch": 3.736842105263158, + "grad_norm": 0.6802878379821777, + "learning_rate": 5.513748274244057e-07, + "loss": 1.842, + "step": 2914 + }, + { + "epoch": 3.7381258023106545, + "grad_norm": 0.661422848701477, + "learning_rate": 5.459490913237885e-07, + "loss": 1.8576, + "step": 2915 + }, + { + "epoch": 3.7394094993581515, + "grad_norm": 0.6969189047813416, + "learning_rate": 5.405498883139987e-07, + "loss": 1.8253, + "step": 2916 + }, + { + "epoch": 3.7406931964056485, + "grad_norm": 0.6817802786827087, + "learning_rate": 5.351772242532749e-07, + "loss": 1.7906, + "step": 2917 + }, + { + "epoch": 3.741976893453145, + "grad_norm": 0.6672261953353882, + "learning_rate": 5.298311049710513e-07, + "loss": 1.8563, + "step": 2918 + }, + { + "epoch": 3.743260590500642, + "grad_norm": 0.6878708004951477, + "learning_rate": 5.245115362679632e-07, + "loss": 1.8739, + "step": 2919 + }, + { + "epoch": 3.7445442875481385, + "grad_norm": 0.7043143510818481, + "learning_rate": 5.192185239158459e-07, + "loss": 1.8452, + "step": 2920 + }, + { + "epoch": 3.7458279845956355, + "grad_norm": 0.6927399635314941, + "learning_rate": 5.139520736577058e-07, + "loss": 1.8027, + "step": 2921 + }, + { + "epoch": 3.7471116816431325, + "grad_norm": 0.6661304235458374, + "learning_rate": 5.087121912077414e-07, + "loss": 1.8155, + "step": 2922 + }, + { + "epoch": 3.748395378690629, + "grad_norm": 0.6643577814102173, + "learning_rate": 5.034988822513242e-07, + "loss": 1.8205, + "step": 2923 + }, + { + "epoch": 3.7496790757381255, + "grad_norm": 0.7294153571128845, + "learning_rate": 4.983121524449852e-07, + "loss": 1.8592, + "step": 2924 + }, + { + "epoch": 3.7509627727856225, + "grad_norm": 0.6864620447158813, + "learning_rate": 4.931520074164225e-07, + "loss": 1.8288, + "step": 2925 + }, + { + "epoch": 3.7522464698331195, + "grad_norm": 0.6785827875137329, + "learning_rate": 4.880184527644826e-07, + "loss": 1.861, + "step": 2926 + }, + { + "epoch": 3.753530166880616, + "grad_norm": 0.6758741140365601, + "learning_rate": 4.829114940591739e-07, + "loss": 1.8077, + "step": 2927 + }, + { + "epoch": 3.754813863928113, + "grad_norm": 0.6769395470619202, + "learning_rate": 4.778311368416333e-07, + "loss": 1.8647, + "step": 2928 + }, + { + "epoch": 3.7560975609756095, + "grad_norm": 0.6743032932281494, + "learning_rate": 4.7277738662414583e-07, + "loss": 1.7623, + "step": 2929 + }, + { + "epoch": 3.7573812580231065, + "grad_norm": 0.6283568739891052, + "learning_rate": 4.6775024889011976e-07, + "loss": 1.7081, + "step": 2930 + }, + { + "epoch": 3.7586649550706035, + "grad_norm": 0.6856492757797241, + "learning_rate": 4.627497290940891e-07, + "loss": 1.785, + "step": 2931 + }, + { + "epoch": 3.7599486521181, + "grad_norm": 0.7218705415725708, + "learning_rate": 4.5777583266171377e-07, + "loss": 1.8922, + "step": 2932 + }, + { + "epoch": 3.761232349165597, + "grad_norm": 0.6684859395027161, + "learning_rate": 4.528285649897601e-07, + "loss": 1.8258, + "step": 2933 + }, + { + "epoch": 3.7625160462130935, + "grad_norm": 0.6406568884849548, + "learning_rate": 4.4790793144610097e-07, + "loss": 1.8005, + "step": 2934 + }, + { + "epoch": 3.7637997432605905, + "grad_norm": 0.6361000537872314, + "learning_rate": 4.430139373697156e-07, + "loss": 1.8394, + "step": 2935 + }, + { + "epoch": 3.7650834403080875, + "grad_norm": 0.6674690842628479, + "learning_rate": 4.3814658807067865e-07, + "loss": 1.8521, + "step": 2936 + }, + { + "epoch": 3.766367137355584, + "grad_norm": 0.6956868767738342, + "learning_rate": 4.333058888301489e-07, + "loss": 1.8303, + "step": 2937 + }, + { + "epoch": 3.767650834403081, + "grad_norm": 0.6924374103546143, + "learning_rate": 4.2849184490037497e-07, + "loss": 1.9095, + "step": 2938 + }, + { + "epoch": 3.7689345314505776, + "grad_norm": 0.7128576040267944, + "learning_rate": 4.2370446150468425e-07, + "loss": 1.8723, + "step": 2939 + }, + { + "epoch": 3.7702182284980745, + "grad_norm": 0.6443933844566345, + "learning_rate": 4.189437438374716e-07, + "loss": 1.9024, + "step": 2940 + }, + { + "epoch": 3.7715019255455715, + "grad_norm": 0.6834497451782227, + "learning_rate": 4.142096970642051e-07, + "loss": 1.8135, + "step": 2941 + }, + { + "epoch": 3.772785622593068, + "grad_norm": 0.6395919919013977, + "learning_rate": 4.095023263214121e-07, + "loss": 1.8437, + "step": 2942 + }, + { + "epoch": 3.7740693196405646, + "grad_norm": 0.7005466818809509, + "learning_rate": 4.048216367166735e-07, + "loss": 1.8464, + "step": 2943 + }, + { + "epoch": 3.7753530166880616, + "grad_norm": 0.7825756072998047, + "learning_rate": 4.001676333286214e-07, + "loss": 1.7796, + "step": 2944 + }, + { + "epoch": 3.7766367137355585, + "grad_norm": 0.675746500492096, + "learning_rate": 3.9554032120693587e-07, + "loss": 1.832, + "step": 2945 + }, + { + "epoch": 3.777920410783055, + "grad_norm": 0.6461715698242188, + "learning_rate": 3.909397053723396e-07, + "loss": 1.8698, + "step": 2946 + }, + { + "epoch": 3.779204107830552, + "grad_norm": 0.6578251719474792, + "learning_rate": 3.8636579081657577e-07, + "loss": 1.8958, + "step": 2947 + }, + { + "epoch": 3.7804878048780486, + "grad_norm": 0.7041683793067932, + "learning_rate": 3.8181858250243284e-07, + "loss": 1.8088, + "step": 2948 + }, + { + "epoch": 3.7817715019255456, + "grad_norm": 0.6478399038314819, + "learning_rate": 3.772980853637087e-07, + "loss": 1.8627, + "step": 2949 + }, + { + "epoch": 3.7830551989730425, + "grad_norm": 0.7004360556602478, + "learning_rate": 3.7280430430522986e-07, + "loss": 1.8681, + "step": 2950 + }, + { + "epoch": 3.784338896020539, + "grad_norm": 0.6904712319374084, + "learning_rate": 3.6833724420282933e-07, + "loss": 1.9395, + "step": 2951 + }, + { + "epoch": 3.785622593068036, + "grad_norm": 0.6531252264976501, + "learning_rate": 3.6389690990334955e-07, + "loss": 1.8185, + "step": 2952 + }, + { + "epoch": 3.7869062901155326, + "grad_norm": 0.6477494835853577, + "learning_rate": 3.5948330622462834e-07, + "loss": 1.8141, + "step": 2953 + }, + { + "epoch": 3.7881899871630296, + "grad_norm": 0.6874672174453735, + "learning_rate": 3.550964379555155e-07, + "loss": 1.8747, + "step": 2954 + }, + { + "epoch": 3.7894736842105265, + "grad_norm": 0.7073059678077698, + "learning_rate": 3.507363098558342e-07, + "loss": 1.8164, + "step": 2955 + }, + { + "epoch": 3.790757381258023, + "grad_norm": 0.6805193424224854, + "learning_rate": 3.464029266564056e-07, + "loss": 1.8473, + "step": 2956 + }, + { + "epoch": 3.79204107830552, + "grad_norm": 0.6755671501159668, + "learning_rate": 3.4209629305903547e-07, + "loss": 1.8077, + "step": 2957 + }, + { + "epoch": 3.7933247753530166, + "grad_norm": 0.6598567366600037, + "learning_rate": 3.3781641373649144e-07, + "loss": 1.8231, + "step": 2958 + }, + { + "epoch": 3.7946084724005136, + "grad_norm": 0.6560249328613281, + "learning_rate": 3.335632933325228e-07, + "loss": 1.7871, + "step": 2959 + }, + { + "epoch": 3.79589216944801, + "grad_norm": 0.6740313768386841, + "learning_rate": 3.293369364618465e-07, + "loss": 1.8317, + "step": 2960 + }, + { + "epoch": 3.797175866495507, + "grad_norm": 0.7260071635246277, + "learning_rate": 3.251373477101333e-07, + "loss": 1.868, + "step": 2961 + }, + { + "epoch": 3.7984595635430036, + "grad_norm": 0.6838186383247375, + "learning_rate": 3.2096453163401317e-07, + "loss": 1.8674, + "step": 2962 + }, + { + "epoch": 3.7997432605905006, + "grad_norm": 0.666966438293457, + "learning_rate": 3.168184927610701e-07, + "loss": 1.8629, + "step": 2963 + }, + { + "epoch": 3.8010269576379976, + "grad_norm": 0.687440037727356, + "learning_rate": 3.126992355898306e-07, + "loss": 1.8695, + "step": 2964 + }, + { + "epoch": 3.802310654685494, + "grad_norm": 0.7029922604560852, + "learning_rate": 3.086067645897611e-07, + "loss": 1.7466, + "step": 2965 + }, + { + "epoch": 3.803594351732991, + "grad_norm": 0.6689265370368958, + "learning_rate": 3.0454108420127357e-07, + "loss": 1.7655, + "step": 2966 + }, + { + "epoch": 3.8048780487804876, + "grad_norm": 0.6436645984649658, + "learning_rate": 3.005021988357004e-07, + "loss": 1.7989, + "step": 2967 + }, + { + "epoch": 3.8061617458279846, + "grad_norm": 0.6715465188026428, + "learning_rate": 2.964901128753056e-07, + "loss": 1.8158, + "step": 2968 + }, + { + "epoch": 3.8074454428754816, + "grad_norm": 0.6582645773887634, + "learning_rate": 2.9250483067327637e-07, + "loss": 1.8651, + "step": 2969 + }, + { + "epoch": 3.808729139922978, + "grad_norm": 0.6608855128288269, + "learning_rate": 2.8854635655371776e-07, + "loss": 1.8361, + "step": 2970 + }, + { + "epoch": 3.810012836970475, + "grad_norm": 0.6625649929046631, + "learning_rate": 2.846146948116468e-07, + "loss": 1.8706, + "step": 2971 + }, + { + "epoch": 3.8112965340179716, + "grad_norm": 0.665543258190155, + "learning_rate": 2.8070984971298454e-07, + "loss": 1.806, + "step": 2972 + }, + { + "epoch": 3.8125802310654686, + "grad_norm": 0.6705846190452576, + "learning_rate": 2.7683182549456123e-07, + "loss": 1.8768, + "step": 2973 + }, + { + "epoch": 3.8138639281129656, + "grad_norm": 0.6723948121070862, + "learning_rate": 2.7298062636410547e-07, + "loss": 1.788, + "step": 2974 + }, + { + "epoch": 3.815147625160462, + "grad_norm": 0.737170934677124, + "learning_rate": 2.691562565002359e-07, + "loss": 1.8055, + "step": 2975 + }, + { + "epoch": 3.8164313222079587, + "grad_norm": 0.6787354350090027, + "learning_rate": 2.6535872005246374e-07, + "loss": 1.734, + "step": 2976 + }, + { + "epoch": 3.8177150192554556, + "grad_norm": 0.6879172325134277, + "learning_rate": 2.615880211411875e-07, + "loss": 1.7762, + "step": 2977 + }, + { + "epoch": 3.8189987163029526, + "grad_norm": 0.6681011319160461, + "learning_rate": 2.5784416385768174e-07, + "loss": 1.8336, + "step": 2978 + }, + { + "epoch": 3.820282413350449, + "grad_norm": 0.6756552457809448, + "learning_rate": 2.5412715226410255e-07, + "loss": 1.8594, + "step": 2979 + }, + { + "epoch": 3.821566110397946, + "grad_norm": 0.6793838739395142, + "learning_rate": 2.504369903934739e-07, + "loss": 1.8277, + "step": 2980 + }, + { + "epoch": 3.8228498074454427, + "grad_norm": 0.6522098779678345, + "learning_rate": 2.4677368224968466e-07, + "loss": 1.8335, + "step": 2981 + }, + { + "epoch": 3.8241335044929397, + "grad_norm": 0.6828100085258484, + "learning_rate": 2.4313723180749706e-07, + "loss": 1.8625, + "step": 2982 + }, + { + "epoch": 3.8254172015404366, + "grad_norm": 0.6997185349464417, + "learning_rate": 2.395276430125243e-07, + "loss": 1.7832, + "step": 2983 + }, + { + "epoch": 3.826700898587933, + "grad_norm": 0.6843633651733398, + "learning_rate": 2.3594491978123356e-07, + "loss": 1.8715, + "step": 2984 + }, + { + "epoch": 3.82798459563543, + "grad_norm": 0.7087182998657227, + "learning_rate": 2.3238906600094868e-07, + "loss": 1.7939, + "step": 2985 + }, + { + "epoch": 3.8292682926829267, + "grad_norm": 0.6925849318504333, + "learning_rate": 2.288600855298306e-07, + "loss": 1.7967, + "step": 2986 + }, + { + "epoch": 3.8305519897304237, + "grad_norm": 0.6413093209266663, + "learning_rate": 2.2535798219689152e-07, + "loss": 1.8225, + "step": 2987 + }, + { + "epoch": 3.8318356867779206, + "grad_norm": 0.6723337769508362, + "learning_rate": 2.2188275980197792e-07, + "loss": 1.7955, + "step": 2988 + }, + { + "epoch": 3.833119383825417, + "grad_norm": 0.6955317854881287, + "learning_rate": 2.1843442211576802e-07, + "loss": 1.8711, + "step": 2989 + }, + { + "epoch": 3.834403080872914, + "grad_norm": 0.683172345161438, + "learning_rate": 2.1501297287976897e-07, + "loss": 1.8932, + "step": 2990 + }, + { + "epoch": 3.8356867779204107, + "grad_norm": 0.6905835866928101, + "learning_rate": 2.1161841580631947e-07, + "loss": 1.7761, + "step": 2991 + }, + { + "epoch": 3.8369704749679077, + "grad_norm": 0.6678906083106995, + "learning_rate": 2.0825075457857336e-07, + "loss": 1.8705, + "step": 2992 + }, + { + "epoch": 3.8382541720154046, + "grad_norm": 0.7068698406219482, + "learning_rate": 2.0490999285050215e-07, + "loss": 1.816, + "step": 2993 + }, + { + "epoch": 3.839537869062901, + "grad_norm": 0.6478477716445923, + "learning_rate": 2.0159613424689804e-07, + "loss": 1.907, + "step": 2994 + }, + { + "epoch": 3.8408215661103977, + "grad_norm": 0.6875187158584595, + "learning_rate": 1.9830918236335705e-07, + "loss": 1.8255, + "step": 2995 + }, + { + "epoch": 3.8421052631578947, + "grad_norm": 0.6499404311180115, + "learning_rate": 1.9504914076627912e-07, + "loss": 1.8322, + "step": 2996 + }, + { + "epoch": 3.8433889602053917, + "grad_norm": 0.7307029366493225, + "learning_rate": 1.918160129928681e-07, + "loss": 1.8224, + "step": 2997 + }, + { + "epoch": 3.844672657252888, + "grad_norm": 0.7088853716850281, + "learning_rate": 1.886098025511318e-07, + "loss": 1.824, + "step": 2998 + }, + { + "epoch": 3.845956354300385, + "grad_norm": 0.858163058757782, + "learning_rate": 1.8543051291986247e-07, + "loss": 1.825, + "step": 2999 + }, + { + "epoch": 3.8472400513478817, + "grad_norm": 0.6919712424278259, + "learning_rate": 1.8227814754865068e-07, + "loss": 1.8508, + "step": 3000 + }, + { + "epoch": 3.8485237483953787, + "grad_norm": 0.6605830192565918, + "learning_rate": 1.791527098578688e-07, + "loss": 1.7699, + "step": 3001 + }, + { + "epoch": 3.8498074454428757, + "grad_norm": 0.6294465065002441, + "learning_rate": 1.7605420323867085e-07, + "loss": 1.8656, + "step": 3002 + }, + { + "epoch": 3.851091142490372, + "grad_norm": 0.6383609175682068, + "learning_rate": 1.7298263105299527e-07, + "loss": 1.7608, + "step": 3003 + }, + { + "epoch": 3.852374839537869, + "grad_norm": 0.6814265251159668, + "learning_rate": 1.6993799663355402e-07, + "loss": 1.9599, + "step": 3004 + }, + { + "epoch": 3.8536585365853657, + "grad_norm": 0.7011674046516418, + "learning_rate": 1.6692030328383235e-07, + "loss": 1.7906, + "step": 3005 + }, + { + "epoch": 3.8549422336328627, + "grad_norm": 0.6612873673439026, + "learning_rate": 1.6392955427808056e-07, + "loss": 1.8064, + "step": 3006 + }, + { + "epoch": 3.8562259306803597, + "grad_norm": 0.6918783783912659, + "learning_rate": 1.6096575286131677e-07, + "loss": 1.8556, + "step": 3007 + }, + { + "epoch": 3.857509627727856, + "grad_norm": 0.6483519673347473, + "learning_rate": 1.5802890224931865e-07, + "loss": 1.7576, + "step": 3008 + }, + { + "epoch": 3.8587933247753528, + "grad_norm": 0.6942324042320251, + "learning_rate": 1.5511900562862057e-07, + "loss": 1.9055, + "step": 3009 + }, + { + "epoch": 3.8600770218228497, + "grad_norm": 0.7148955464363098, + "learning_rate": 1.522360661565192e-07, + "loss": 1.9011, + "step": 3010 + }, + { + "epoch": 3.8613607188703467, + "grad_norm": 0.6807995438575745, + "learning_rate": 1.4938008696105398e-07, + "loss": 1.7935, + "step": 3011 + }, + { + "epoch": 3.8626444159178432, + "grad_norm": 0.6975430846214294, + "learning_rate": 1.4655107114101007e-07, + "loss": 1.7777, + "step": 3012 + }, + { + "epoch": 3.8639281129653402, + "grad_norm": 0.6569209098815918, + "learning_rate": 1.4374902176592652e-07, + "loss": 1.7767, + "step": 3013 + }, + { + "epoch": 3.8652118100128368, + "grad_norm": 0.7519148588180542, + "learning_rate": 1.4097394187607416e-07, + "loss": 1.8522, + "step": 3014 + }, + { + "epoch": 3.8664955070603337, + "grad_norm": 0.6771644353866577, + "learning_rate": 1.382258344824694e-07, + "loss": 1.9117, + "step": 3015 + }, + { + "epoch": 3.8677792041078307, + "grad_norm": 0.7255068421363831, + "learning_rate": 1.3550470256685488e-07, + "loss": 1.8474, + "step": 3016 + }, + { + "epoch": 3.8690629011553272, + "grad_norm": 0.7167854309082031, + "learning_rate": 1.3281054908171053e-07, + "loss": 1.85, + "step": 3017 + }, + { + "epoch": 3.8703465982028242, + "grad_norm": 0.6731257438659668, + "learning_rate": 1.301433769502397e-07, + "loss": 1.7916, + "step": 3018 + }, + { + "epoch": 3.8716302952503208, + "grad_norm": 0.6870619654655457, + "learning_rate": 1.2750318906637747e-07, + "loss": 1.8691, + "step": 3019 + }, + { + "epoch": 3.8729139922978177, + "grad_norm": 0.6972445249557495, + "learning_rate": 1.248899882947713e-07, + "loss": 1.7962, + "step": 3020 + }, + { + "epoch": 3.8741976893453147, + "grad_norm": 0.6904467344284058, + "learning_rate": 1.223037774707947e-07, + "loss": 1.8391, + "step": 3021 + }, + { + "epoch": 3.8754813863928113, + "grad_norm": 0.704420804977417, + "learning_rate": 1.1974455940052809e-07, + "loss": 1.8622, + "step": 3022 + }, + { + "epoch": 3.8767650834403082, + "grad_norm": 0.7310137152671814, + "learning_rate": 1.1721233686077803e-07, + "loss": 1.875, + "step": 3023 + }, + { + "epoch": 3.8780487804878048, + "grad_norm": 0.6643840074539185, + "learning_rate": 1.1470711259904399e-07, + "loss": 1.8323, + "step": 3024 + }, + { + "epoch": 3.8793324775353017, + "grad_norm": 0.6591954231262207, + "learning_rate": 1.1222888933354602e-07, + "loss": 1.8701, + "step": 3025 + }, + { + "epoch": 3.8806161745827987, + "grad_norm": 0.698911190032959, + "learning_rate": 1.0977766975319992e-07, + "loss": 1.8552, + "step": 3026 + }, + { + "epoch": 3.8818998716302953, + "grad_norm": 0.7447785139083862, + "learning_rate": 1.0735345651761986e-07, + "loss": 1.7124, + "step": 3027 + }, + { + "epoch": 3.883183568677792, + "grad_norm": 0.6927556395530701, + "learning_rate": 1.0495625225712402e-07, + "loss": 1.8409, + "step": 3028 + }, + { + "epoch": 3.8844672657252888, + "grad_norm": 0.7274348139762878, + "learning_rate": 1.0258605957272627e-07, + "loss": 1.7755, + "step": 3029 + }, + { + "epoch": 3.8857509627727858, + "grad_norm": 0.6909927129745483, + "learning_rate": 1.00242881036125e-07, + "loss": 1.7936, + "step": 3030 + }, + { + "epoch": 3.8870346598202823, + "grad_norm": 0.649178683757782, + "learning_rate": 9.792671918970875e-08, + "loss": 1.9058, + "step": 3031 + }, + { + "epoch": 3.8883183568677793, + "grad_norm": 0.7014222145080566, + "learning_rate": 9.563757654656169e-08, + "loss": 1.9214, + "step": 3032 + }, + { + "epoch": 3.889602053915276, + "grad_norm": 0.717758059501648, + "learning_rate": 9.337545559044148e-08, + "loss": 1.8723, + "step": 3033 + }, + { + "epoch": 3.890885750962773, + "grad_norm": 0.7004956007003784, + "learning_rate": 9.114035877579308e-08, + "loss": 1.8471, + "step": 3034 + }, + { + "epoch": 3.8921694480102698, + "grad_norm": 0.6657041907310486, + "learning_rate": 8.893228852773494e-08, + "loss": 1.8324, + "step": 3035 + }, + { + "epoch": 3.8934531450577663, + "grad_norm": 0.6509225964546204, + "learning_rate": 8.675124724206174e-08, + "loss": 1.8639, + "step": 3036 + }, + { + "epoch": 3.8947368421052633, + "grad_norm": 0.6598386764526367, + "learning_rate": 8.459723728524716e-08, + "loss": 1.8194, + "step": 3037 + }, + { + "epoch": 3.89602053915276, + "grad_norm": 0.7055901288986206, + "learning_rate": 8.247026099443277e-08, + "loss": 1.88, + "step": 3038 + }, + { + "epoch": 3.897304236200257, + "grad_norm": 0.6470000743865967, + "learning_rate": 8.037032067741978e-08, + "loss": 1.8489, + "step": 3039 + }, + { + "epoch": 3.8985879332477538, + "grad_norm": 0.72780841588974, + "learning_rate": 7.829741861268558e-08, + "loss": 1.9117, + "step": 3040 + }, + { + "epoch": 3.8998716302952503, + "grad_norm": 0.6857308149337769, + "learning_rate": 7.625155704936715e-08, + "loss": 1.9137, + "step": 3041 + }, + { + "epoch": 3.901155327342747, + "grad_norm": 0.6991900205612183, + "learning_rate": 7.423273820726106e-08, + "loss": 1.8185, + "step": 3042 + }, + { + "epoch": 3.902439024390244, + "grad_norm": 0.6952182054519653, + "learning_rate": 7.22409642768207e-08, + "loss": 1.7844, + "step": 3043 + }, + { + "epoch": 3.903722721437741, + "grad_norm": 0.6540995240211487, + "learning_rate": 7.027623741916178e-08, + "loss": 1.9256, + "step": 3044 + }, + { + "epoch": 3.9050064184852373, + "grad_norm": 0.6949002146720886, + "learning_rate": 6.833855976604575e-08, + "loss": 1.8613, + "step": 3045 + }, + { + "epoch": 3.9062901155327343, + "grad_norm": 0.6699240207672119, + "learning_rate": 6.64279334198853e-08, + "loss": 1.8299, + "step": 3046 + }, + { + "epoch": 3.907573812580231, + "grad_norm": 0.6777414083480835, + "learning_rate": 6.454436045374713e-08, + "loss": 1.8244, + "step": 3047 + }, + { + "epoch": 3.908857509627728, + "grad_norm": 0.6643387079238892, + "learning_rate": 6.268784291134643e-08, + "loss": 1.9072, + "step": 3048 + }, + { + "epoch": 3.910141206675225, + "grad_norm": 0.6651031374931335, + "learning_rate": 6.085838280703304e-08, + "loss": 1.8545, + "step": 3049 + }, + { + "epoch": 3.9114249037227213, + "grad_norm": 0.6464062333106995, + "learning_rate": 5.905598212580521e-08, + "loss": 1.9241, + "step": 3050 + }, + { + "epoch": 3.9127086007702183, + "grad_norm": 0.6788432598114014, + "learning_rate": 5.728064282330137e-08, + "loss": 1.8655, + "step": 3051 + }, + { + "epoch": 3.913992297817715, + "grad_norm": 0.7156592011451721, + "learning_rate": 5.5532366825800096e-08, + "loss": 1.7796, + "step": 3052 + }, + { + "epoch": 3.915275994865212, + "grad_norm": 0.6611512303352356, + "learning_rate": 5.3811156030209034e-08, + "loss": 1.8998, + "step": 3053 + }, + { + "epoch": 3.916559691912709, + "grad_norm": 0.7166811227798462, + "learning_rate": 5.2117012304073175e-08, + "loss": 1.8733, + "step": 3054 + }, + { + "epoch": 3.9178433889602053, + "grad_norm": 0.6279780268669128, + "learning_rate": 5.044993748557214e-08, + "loss": 1.8764, + "step": 3055 + }, + { + "epoch": 3.9191270860077023, + "grad_norm": 0.6971473097801208, + "learning_rate": 4.880993338350626e-08, + "loss": 1.8422, + "step": 3056 + }, + { + "epoch": 3.920410783055199, + "grad_norm": 0.677911639213562, + "learning_rate": 4.7197001777318787e-08, + "loss": 1.806, + "step": 3057 + }, + { + "epoch": 3.921694480102696, + "grad_norm": 0.6541029214859009, + "learning_rate": 4.56111444170626e-08, + "loss": 1.9138, + "step": 3058 + }, + { + "epoch": 3.922978177150193, + "grad_norm": 0.7557929158210754, + "learning_rate": 4.405236302342519e-08, + "loss": 1.7849, + "step": 3059 + }, + { + "epoch": 3.9242618741976893, + "grad_norm": 0.6749855279922485, + "learning_rate": 4.252065928771476e-08, + "loss": 1.859, + "step": 3060 + }, + { + "epoch": 3.925545571245186, + "grad_norm": 0.7163145542144775, + "learning_rate": 4.1016034871851905e-08, + "loss": 1.8613, + "step": 3061 + }, + { + "epoch": 3.926829268292683, + "grad_norm": 0.6910869479179382, + "learning_rate": 3.953849140838628e-08, + "loss": 1.8443, + "step": 3062 + }, + { + "epoch": 3.92811296534018, + "grad_norm": 0.7175049781799316, + "learning_rate": 3.808803050047716e-08, + "loss": 1.9049, + "step": 3063 + }, + { + "epoch": 3.9293966623876764, + "grad_norm": 0.6840599775314331, + "learning_rate": 3.666465372190453e-08, + "loss": 1.7828, + "step": 3064 + }, + { + "epoch": 3.9306803594351734, + "grad_norm": 0.6794530749320984, + "learning_rate": 3.526836261705246e-08, + "loss": 1.8262, + "step": 3065 + }, + { + "epoch": 3.93196405648267, + "grad_norm": 0.6638824343681335, + "learning_rate": 3.389915870092575e-08, + "loss": 1.8133, + "step": 3066 + }, + { + "epoch": 3.933247753530167, + "grad_norm": 0.6415587067604065, + "learning_rate": 3.255704345913602e-08, + "loss": 1.8844, + "step": 3067 + }, + { + "epoch": 3.934531450577664, + "grad_norm": 0.6261206269264221, + "learning_rate": 3.1242018347901745e-08, + "loss": 1.8204, + "step": 3068 + }, + { + "epoch": 3.9358151476251604, + "grad_norm": 0.6884837746620178, + "learning_rate": 2.995408479404826e-08, + "loss": 1.8661, + "step": 3069 + }, + { + "epoch": 3.9370988446726574, + "grad_norm": 0.6606943607330322, + "learning_rate": 2.8693244195010515e-08, + "loss": 1.856, + "step": 3070 + }, + { + "epoch": 3.938382541720154, + "grad_norm": 0.6605387330055237, + "learning_rate": 2.745949791881919e-08, + "loss": 1.8105, + "step": 3071 + }, + { + "epoch": 3.939666238767651, + "grad_norm": 0.6659144759178162, + "learning_rate": 2.6252847304111815e-08, + "loss": 1.8085, + "step": 3072 + }, + { + "epoch": 3.940949935815148, + "grad_norm": 0.6942153573036194, + "learning_rate": 2.5073293660130005e-08, + "loss": 1.7524, + "step": 3073 + }, + { + "epoch": 3.9422336328626444, + "grad_norm": 0.6719822883605957, + "learning_rate": 2.3920838266708322e-08, + "loss": 1.8729, + "step": 3074 + }, + { + "epoch": 3.9435173299101414, + "grad_norm": 0.6301265358924866, + "learning_rate": 2.279548237428264e-08, + "loss": 1.8172, + "step": 3075 + }, + { + "epoch": 3.944801026957638, + "grad_norm": 0.6782194375991821, + "learning_rate": 2.169722720388179e-08, + "loss": 1.8349, + "step": 3076 + }, + { + "epoch": 3.946084724005135, + "grad_norm": 0.6993538737297058, + "learning_rate": 2.0626073947138668e-08, + "loss": 1.7943, + "step": 3077 + }, + { + "epoch": 3.9473684210526314, + "grad_norm": 0.6600044369697571, + "learning_rate": 1.9582023766268055e-08, + "loss": 1.8533, + "step": 3078 + }, + { + "epoch": 3.9486521181001284, + "grad_norm": 0.6593739986419678, + "learning_rate": 1.8565077794086006e-08, + "loss": 1.8273, + "step": 3079 + }, + { + "epoch": 3.949935815147625, + "grad_norm": 0.6571626663208008, + "learning_rate": 1.7575237133996003e-08, + "loss": 1.7737, + "step": 3080 + }, + { + "epoch": 3.951219512195122, + "grad_norm": 0.7249687910079956, + "learning_rate": 1.661250285999727e-08, + "loss": 1.8103, + "step": 3081 + }, + { + "epoch": 3.952503209242619, + "grad_norm": 0.6333588361740112, + "learning_rate": 1.567687601666812e-08, + "loss": 1.7787, + "step": 3082 + }, + { + "epoch": 3.9537869062901154, + "grad_norm": 0.7097930908203125, + "learning_rate": 1.4768357619188155e-08, + "loss": 1.827, + "step": 3083 + }, + { + "epoch": 3.9550706033376124, + "grad_norm": 0.6591911315917969, + "learning_rate": 1.3886948653307751e-08, + "loss": 1.8892, + "step": 3084 + }, + { + "epoch": 3.956354300385109, + "grad_norm": 0.6825366020202637, + "learning_rate": 1.3032650075378571e-08, + "loss": 1.8188, + "step": 3085 + }, + { + "epoch": 3.957637997432606, + "grad_norm": 0.6867747902870178, + "learning_rate": 1.220546281232582e-08, + "loss": 1.7528, + "step": 3086 + }, + { + "epoch": 3.958921694480103, + "grad_norm": 0.7803851962089539, + "learning_rate": 1.1405387761664887e-08, + "loss": 1.8538, + "step": 3087 + }, + { + "epoch": 3.9602053915275994, + "grad_norm": 0.639132559299469, + "learning_rate": 1.0632425791493039e-08, + "loss": 1.7348, + "step": 3088 + }, + { + "epoch": 3.9614890885750964, + "grad_norm": 0.6911587715148926, + "learning_rate": 9.886577740486625e-09, + "loss": 1.8073, + "step": 3089 + }, + { + "epoch": 3.962772785622593, + "grad_norm": 0.6958117485046387, + "learning_rate": 9.167844417901083e-09, + "loss": 1.834, + "step": 3090 + }, + { + "epoch": 3.96405648267009, + "grad_norm": 0.692919909954071, + "learning_rate": 8.476226603579274e-09, + "loss": 1.8767, + "step": 3091 + }, + { + "epoch": 3.965340179717587, + "grad_norm": 0.6917592883110046, + "learning_rate": 7.811725047940367e-09, + "loss": 1.8936, + "step": 3092 + }, + { + "epoch": 3.9666238767650834, + "grad_norm": 0.7046009302139282, + "learning_rate": 7.17434047197707e-09, + "loss": 1.7454, + "step": 3093 + }, + { + "epoch": 3.96790757381258, + "grad_norm": 0.6525453329086304, + "learning_rate": 6.564073567263962e-09, + "loss": 1.8685, + "step": 3094 + }, + { + "epoch": 3.969191270860077, + "grad_norm": 0.656565248966217, + "learning_rate": 5.980924995951931e-09, + "loss": 1.8418, + "step": 3095 + }, + { + "epoch": 3.970474967907574, + "grad_norm": 0.6744842529296875, + "learning_rate": 5.424895390770957e-09, + "loss": 1.8681, + "step": 3096 + }, + { + "epoch": 3.9717586649550705, + "grad_norm": 0.6641948819160461, + "learning_rate": 4.895985355019006e-09, + "loss": 1.7845, + "step": 3097 + }, + { + "epoch": 3.9730423620025674, + "grad_norm": 0.6951947212219238, + "learning_rate": 4.394195462575912e-09, + "loss": 1.8054, + "step": 3098 + }, + { + "epoch": 3.974326059050064, + "grad_norm": 0.7356829047203064, + "learning_rate": 3.919526257889494e-09, + "loss": 1.8289, + "step": 3099 + }, + { + "epoch": 3.975609756097561, + "grad_norm": 0.7113009095191956, + "learning_rate": 3.471978255986663e-09, + "loss": 1.7631, + "step": 3100 + }, + { + "epoch": 3.976893453145058, + "grad_norm": 0.6887560486793518, + "learning_rate": 3.0515519424678675e-09, + "loss": 1.8306, + "step": 3101 + }, + { + "epoch": 3.9781771501925545, + "grad_norm": 0.6676644682884216, + "learning_rate": 2.658247773498768e-09, + "loss": 1.9038, + "step": 3102 + }, + { + "epoch": 3.9794608472400514, + "grad_norm": 0.6779635548591614, + "learning_rate": 2.292066175821339e-09, + "loss": 1.8595, + "step": 3103 + }, + { + "epoch": 3.980744544287548, + "grad_norm": 0.6806266903877258, + "learning_rate": 1.9530075467538712e-09, + "loss": 1.8296, + "step": 3104 + }, + { + "epoch": 3.982028241335045, + "grad_norm": 0.7207275629043579, + "learning_rate": 1.6410722541798651e-09, + "loss": 1.9041, + "step": 3105 + }, + { + "epoch": 3.983311938382542, + "grad_norm": 0.7071326375007629, + "learning_rate": 1.35626063655081e-09, + "loss": 1.8545, + "step": 3106 + }, + { + "epoch": 3.9845956354300385, + "grad_norm": 0.6699494123458862, + "learning_rate": 1.098573002897285e-09, + "loss": 1.7776, + "step": 3107 + }, + { + "epoch": 3.9858793324775355, + "grad_norm": 0.6723994016647339, + "learning_rate": 8.680096328123055e-10, + "loss": 1.7331, + "step": 3108 + }, + { + "epoch": 3.987163029525032, + "grad_norm": 0.7223527431488037, + "learning_rate": 6.645707764596498e-10, + "loss": 1.8352, + "step": 3109 + }, + { + "epoch": 3.988446726572529, + "grad_norm": 0.7247270345687866, + "learning_rate": 4.882566545794109e-10, + "loss": 1.8508, + "step": 3110 + }, + { + "epoch": 3.989730423620026, + "grad_norm": 0.6417693495750427, + "learning_rate": 3.390674584713427e-10, + "loss": 1.8192, + "step": 3111 + }, + { + "epoch": 3.9910141206675225, + "grad_norm": 0.6637742519378662, + "learning_rate": 2.1700335001151317e-10, + "loss": 1.827, + "step": 3112 + }, + { + "epoch": 3.992297817715019, + "grad_norm": 0.6469768285751343, + "learning_rate": 1.22064461638427e-10, + "loss": 1.8962, + "step": 3113 + }, + { + "epoch": 3.993581514762516, + "grad_norm": 0.7709643244743347, + "learning_rate": 5.4250896366903057e-11, + "loss": 1.8561, + "step": 3114 + }, + { + "epoch": 3.994865211810013, + "grad_norm": 0.6629145741462708, + "learning_rate": 1.3562727771421201e-11, + "loss": 1.8455, + "step": 3115 + }, + { + "epoch": 3.9961489088575095, + "grad_norm": 0.7026376724243164, + "learning_rate": 0.0, + "loss": 1.7111, + "step": 3116 + } + ], + "logging_steps": 1, + "max_steps": 3116, + "num_input_tokens_seen": 0, + "num_train_epochs": 4, + "save_steps": 195, + "stateful_callbacks": { + "TrainerControl": { + "args": { + "should_epoch_stop": false, + "should_evaluate": false, + "should_log": false, + "should_save": true, + "should_training_stop": true + }, + "attributes": {} + } + }, + "total_flos": 3.7819112526658404e+19, + "train_batch_size": 4, + "trial_name": null, + "trial_params": null +}