{ "best_global_step": null, "best_metric": null, "best_model_checkpoint": null, "epoch": 0.16071033970148055, "eval_steps": 200, "global_step": 2000, "is_hyper_param_search": false, "is_local_process_zero": true, "is_world_process_zero": true, "log_history": [ { "epoch": 8.035516985074027e-05, "grad_norm": 10.267086029052734, "learning_rate": 0.0, "loss": 0.4708889126777649, "num_input_tokens_seen": 2822, "step": 1, "train_runtime": 8.3411, "train_tokens_per_second": 338.326 }, { "epoch": 0.00016071033970148054, "grad_norm": 10.736565589904785, "learning_rate": 4.000000000000001e-06, "loss": 0.5914044380187988, "num_input_tokens_seen": 4940, "step": 2, "train_runtime": 10.803, "train_tokens_per_second": 457.28 }, { "epoch": 0.0002410655095522208, "grad_norm": 7.805853843688965, "learning_rate": 8.000000000000001e-06, "loss": 0.4639473557472229, "num_input_tokens_seen": 7812, "step": 3, "train_runtime": 13.5017, "train_tokens_per_second": 578.593 }, { "epoch": 0.0003214206794029611, "grad_norm": 5.533205986022949, "learning_rate": 1.2e-05, "loss": 0.4620698094367981, "num_input_tokens_seen": 10572, "step": 4, "train_runtime": 16.1136, "train_tokens_per_second": 656.091 }, { "epoch": 0.0004017758492537014, "grad_norm": 5.112827301025391, "learning_rate": 1.6000000000000003e-05, "loss": 0.33940941095352173, "num_input_tokens_seen": 11966, "step": 5, "train_runtime": 18.4878, "train_tokens_per_second": 647.236 }, { "epoch": 0.0004821310191044416, "grad_norm": 5.489980697631836, "learning_rate": 2e-05, "loss": 0.2592361271381378, "num_input_tokens_seen": 15664, "step": 6, "train_runtime": 21.3397, "train_tokens_per_second": 734.031 }, { "epoch": 0.0005624861889551819, "grad_norm": 4.013363838195801, "learning_rate": 1.998997493734336e-05, "loss": 0.39688190817832947, "num_input_tokens_seen": 18540, "step": 7, "train_runtime": 23.9723, "train_tokens_per_second": 773.392 }, { "epoch": 0.0006428413588059222, "grad_norm": 3.0903775691986084, "learning_rate": 1.997994987468672e-05, "loss": 0.26538965106010437, "num_input_tokens_seen": 23576, "step": 8, "train_runtime": 27.1449, "train_tokens_per_second": 868.526 }, { "epoch": 0.0007231965286566625, "grad_norm": 4.888548374176025, "learning_rate": 1.9969924812030075e-05, "loss": 0.3179853856563568, "num_input_tokens_seen": 27178, "step": 9, "train_runtime": 30.1333, "train_tokens_per_second": 901.925 }, { "epoch": 0.0008035516985074027, "grad_norm": 4.030289649963379, "learning_rate": 1.9959899749373436e-05, "loss": 0.3284183740615845, "num_input_tokens_seen": 29798, "step": 10, "train_runtime": 32.7286, "train_tokens_per_second": 910.458 }, { "epoch": 0.000883906868358143, "grad_norm": 3.7689971923828125, "learning_rate": 1.9949874686716792e-05, "loss": 0.29121366143226624, "num_input_tokens_seen": 31756, "step": 11, "train_runtime": 35.0745, "train_tokens_per_second": 905.387 }, { "epoch": 0.0009642620382088832, "grad_norm": 4.130695819854736, "learning_rate": 1.9939849624060153e-05, "loss": 0.3157326281070709, "num_input_tokens_seen": 33724, "step": 12, "train_runtime": 37.5484, "train_tokens_per_second": 898.147 }, { "epoch": 0.0010446172080596236, "grad_norm": 3.3276731967926025, "learning_rate": 1.992982456140351e-05, "loss": 0.3226930797100067, "num_input_tokens_seen": 36654, "step": 13, "train_runtime": 40.2781, "train_tokens_per_second": 910.023 }, { "epoch": 0.0011249723779103637, "grad_norm": 3.4712131023406982, "learning_rate": 1.991979949874687e-05, "loss": 0.2731683850288391, "num_input_tokens_seen": 39044, "step": 14, "train_runtime": 42.7636, "train_tokens_per_second": 913.02 }, { "epoch": 0.001205327547761104, "grad_norm": 2.964151382446289, "learning_rate": 1.9909774436090226e-05, "loss": 0.34246891736984253, "num_input_tokens_seen": 45640, "step": 15, "train_runtime": 46.3101, "train_tokens_per_second": 985.53 }, { "epoch": 0.0012856827176118443, "grad_norm": 3.8719708919525146, "learning_rate": 1.9899749373433587e-05, "loss": 0.4222320318222046, "num_input_tokens_seen": 48780, "step": 16, "train_runtime": 48.944, "train_tokens_per_second": 996.649 }, { "epoch": 0.0013660378874625846, "grad_norm": 2.856308937072754, "learning_rate": 1.9889724310776943e-05, "loss": 0.2585548162460327, "num_input_tokens_seen": 51850, "step": 17, "train_runtime": 51.5368, "train_tokens_per_second": 1006.078 }, { "epoch": 0.001446393057313325, "grad_norm": 3.759415864944458, "learning_rate": 1.9879699248120303e-05, "loss": 0.3500666618347168, "num_input_tokens_seen": 53792, "step": 18, "train_runtime": 53.8421, "train_tokens_per_second": 999.07 }, { "epoch": 0.0015267482271640652, "grad_norm": 2.9645113945007324, "learning_rate": 1.986967418546366e-05, "loss": 0.3153449594974518, "num_input_tokens_seen": 56558, "step": 19, "train_runtime": 56.3916, "train_tokens_per_second": 1002.95 }, { "epoch": 0.0016071033970148055, "grad_norm": 3.7543790340423584, "learning_rate": 1.9859649122807017e-05, "loss": 0.34221383929252625, "num_input_tokens_seen": 58172, "step": 20, "train_runtime": 58.6526, "train_tokens_per_second": 991.806 }, { "epoch": 0.0016874585668655458, "grad_norm": 4.238717555999756, "learning_rate": 1.9849624060150377e-05, "loss": 0.3234458863735199, "num_input_tokens_seen": 60110, "step": 21, "train_runtime": 60.9999, "train_tokens_per_second": 985.411 }, { "epoch": 0.001767813736716286, "grad_norm": 3.4829607009887695, "learning_rate": 1.9839598997493737e-05, "loss": 0.3707271218299866, "num_input_tokens_seen": 62466, "step": 22, "train_runtime": 63.4376, "train_tokens_per_second": 984.683 }, { "epoch": 0.0018481689065670262, "grad_norm": 3.8435935974121094, "learning_rate": 1.9829573934837094e-05, "loss": 0.2531720995903015, "num_input_tokens_seen": 64476, "step": 23, "train_runtime": 65.8015, "train_tokens_per_second": 979.857 }, { "epoch": 0.0019285240764177665, "grad_norm": 3.686642646789551, "learning_rate": 1.981954887218045e-05, "loss": 0.41092386841773987, "num_input_tokens_seen": 67956, "step": 24, "train_runtime": 68.5433, "train_tokens_per_second": 991.432 }, { "epoch": 0.0020088792462685068, "grad_norm": 2.8219215869903564, "learning_rate": 1.980952380952381e-05, "loss": 0.30358248949050903, "num_input_tokens_seen": 70274, "step": 25, "train_runtime": 71.064, "train_tokens_per_second": 988.883 }, { "epoch": 0.0020892344161192473, "grad_norm": 3.530162811279297, "learning_rate": 1.9799498746867168e-05, "loss": 0.34581345319747925, "num_input_tokens_seen": 73744, "step": 26, "train_runtime": 73.8044, "train_tokens_per_second": 999.182 }, { "epoch": 0.0021695895859699874, "grad_norm": 3.5129871368408203, "learning_rate": 1.9789473684210528e-05, "loss": 0.40038084983825684, "num_input_tokens_seen": 77042, "step": 27, "train_runtime": 76.4968, "train_tokens_per_second": 1007.127 }, { "epoch": 0.0022499447558207274, "grad_norm": 4.039092063903809, "learning_rate": 1.977944862155389e-05, "loss": 0.38728469610214233, "num_input_tokens_seen": 78480, "step": 28, "train_runtime": 78.7623, "train_tokens_per_second": 996.416 }, { "epoch": 0.002330299925671468, "grad_norm": 3.4911768436431885, "learning_rate": 1.9769423558897245e-05, "loss": 0.3402363061904907, "num_input_tokens_seen": 82624, "step": 29, "train_runtime": 81.7282, "train_tokens_per_second": 1010.96 }, { "epoch": 0.002410655095522208, "grad_norm": 3.762068510055542, "learning_rate": 1.9759398496240602e-05, "loss": 0.36101511120796204, "num_input_tokens_seen": 84966, "step": 30, "train_runtime": 84.2855, "train_tokens_per_second": 1008.074 }, { "epoch": 0.0024910102653729485, "grad_norm": 3.143077850341797, "learning_rate": 1.974937343358396e-05, "loss": 0.3193308413028717, "num_input_tokens_seen": 86812, "step": 31, "train_runtime": 99.4653, "train_tokens_per_second": 872.787 }, { "epoch": 0.0025713654352236886, "grad_norm": 3.737004041671753, "learning_rate": 1.973934837092732e-05, "loss": 0.4077346920967102, "num_input_tokens_seen": 88936, "step": 32, "train_runtime": 101.8813, "train_tokens_per_second": 872.937 }, { "epoch": 0.002651720605074429, "grad_norm": 1.27534019947052, "learning_rate": 1.972932330827068e-05, "loss": 0.15198269486427307, "num_input_tokens_seen": 95730, "step": 33, "train_runtime": 105.5383, "train_tokens_per_second": 907.064 }, { "epoch": 0.002732075774925169, "grad_norm": 3.5800557136535645, "learning_rate": 1.9719298245614036e-05, "loss": 0.341566264629364, "num_input_tokens_seen": 99292, "step": 34, "train_runtime": 108.2804, "train_tokens_per_second": 916.989 }, { "epoch": 0.0028124309447759097, "grad_norm": 4.609138011932373, "learning_rate": 1.9709273182957396e-05, "loss": 0.372038871049881, "num_input_tokens_seen": 107158, "step": 35, "train_runtime": 112.1973, "train_tokens_per_second": 955.085 }, { "epoch": 0.00289278611462665, "grad_norm": 3.6133768558502197, "learning_rate": 1.9699248120300753e-05, "loss": 0.2687448561191559, "num_input_tokens_seen": 110232, "step": 36, "train_runtime": 114.8832, "train_tokens_per_second": 959.514 }, { "epoch": 0.00297314128447739, "grad_norm": 3.505882501602173, "learning_rate": 1.968922305764411e-05, "loss": 0.3442843556404114, "num_input_tokens_seen": 112086, "step": 37, "train_runtime": 117.2356, "train_tokens_per_second": 956.075 }, { "epoch": 0.0030534964543281304, "grad_norm": 3.0436816215515137, "learning_rate": 1.967919799498747e-05, "loss": 0.4329058527946472, "num_input_tokens_seen": 115980, "step": 38, "train_runtime": 120.0926, "train_tokens_per_second": 965.755 }, { "epoch": 0.0031338516241788705, "grad_norm": 3.1545121669769287, "learning_rate": 1.966917293233083e-05, "loss": 0.33709442615509033, "num_input_tokens_seen": 119938, "step": 39, "train_runtime": 123.0487, "train_tokens_per_second": 974.72 }, { "epoch": 0.003214206794029611, "grad_norm": 2.73949933052063, "learning_rate": 1.9659147869674187e-05, "loss": 0.2894617021083832, "num_input_tokens_seen": 122798, "step": 40, "train_runtime": 125.8104, "train_tokens_per_second": 976.056 }, { "epoch": 0.003294561963880351, "grad_norm": 2.921114444732666, "learning_rate": 1.9649122807017544e-05, "loss": 0.2873356342315674, "num_input_tokens_seen": 125676, "step": 41, "train_runtime": 128.4759, "train_tokens_per_second": 978.206 }, { "epoch": 0.0033749171337310916, "grad_norm": 3.1436357498168945, "learning_rate": 1.9639097744360904e-05, "loss": 0.28181493282318115, "num_input_tokens_seen": 128612, "step": 42, "train_runtime": 131.0784, "train_tokens_per_second": 981.183 }, { "epoch": 0.0034552723035818317, "grad_norm": 3.2524266242980957, "learning_rate": 1.962907268170426e-05, "loss": 0.31523528695106506, "num_input_tokens_seen": 132146, "step": 43, "train_runtime": 133.7953, "train_tokens_per_second": 987.673 }, { "epoch": 0.003535627473432572, "grad_norm": 2.538588285446167, "learning_rate": 1.961904761904762e-05, "loss": 0.19783711433410645, "num_input_tokens_seen": 134470, "step": 44, "train_runtime": 136.2329, "train_tokens_per_second": 987.06 }, { "epoch": 0.0036159826432833123, "grad_norm": 3.0886895656585693, "learning_rate": 1.960902255639098e-05, "loss": 0.25540030002593994, "num_input_tokens_seen": 136706, "step": 45, "train_runtime": 138.7092, "train_tokens_per_second": 985.558 }, { "epoch": 0.0036963378131340523, "grad_norm": 2.9666295051574707, "learning_rate": 1.9598997493734338e-05, "loss": 0.2856194078922272, "num_input_tokens_seen": 139706, "step": 46, "train_runtime": 141.3792, "train_tokens_per_second": 988.165 }, { "epoch": 0.003776692982984793, "grad_norm": 2.8323373794555664, "learning_rate": 1.9588972431077695e-05, "loss": 0.2751491963863373, "num_input_tokens_seen": 142534, "step": 47, "train_runtime": 143.999, "train_tokens_per_second": 989.827 }, { "epoch": 0.003857048152835533, "grad_norm": 3.4154772758483887, "learning_rate": 1.9578947368421055e-05, "loss": 0.3031178414821625, "num_input_tokens_seen": 145630, "step": 48, "train_runtime": 146.6097, "train_tokens_per_second": 993.318 }, { "epoch": 0.0039374033226862734, "grad_norm": 3.2098400592803955, "learning_rate": 1.956892230576441e-05, "loss": 0.31351014971733093, "num_input_tokens_seen": 148360, "step": 49, "train_runtime": 149.1445, "train_tokens_per_second": 994.74 }, { "epoch": 0.0040177584925370135, "grad_norm": 3.455293893814087, "learning_rate": 1.9558897243107772e-05, "loss": 0.30366986989974976, "num_input_tokens_seen": 150374, "step": 50, "train_runtime": 151.5088, "train_tokens_per_second": 992.51 }, { "epoch": 0.004098113662387754, "grad_norm": 3.5996172428131104, "learning_rate": 1.954887218045113e-05, "loss": 0.35624831914901733, "num_input_tokens_seen": 152262, "step": 51, "train_runtime": 153.8458, "train_tokens_per_second": 989.705 }, { "epoch": 0.0041784688322384946, "grad_norm": 3.212676763534546, "learning_rate": 1.953884711779449e-05, "loss": 0.2639515995979309, "num_input_tokens_seen": 154574, "step": 52, "train_runtime": 156.301, "train_tokens_per_second": 988.951 }, { "epoch": 0.004258824002089235, "grad_norm": 3.0706729888916016, "learning_rate": 1.9528822055137845e-05, "loss": 0.28738051652908325, "num_input_tokens_seen": 157530, "step": 53, "train_runtime": 158.8884, "train_tokens_per_second": 991.451 }, { "epoch": 0.004339179171939975, "grad_norm": 3.4318594932556152, "learning_rate": 1.9518796992481202e-05, "loss": 0.3200945556163788, "num_input_tokens_seen": 159162, "step": 54, "train_runtime": 161.1637, "train_tokens_per_second": 987.58 }, { "epoch": 0.004419534341790715, "grad_norm": 3.712693214416504, "learning_rate": 1.9508771929824562e-05, "loss": 0.3589724600315094, "num_input_tokens_seen": 160720, "step": 55, "train_runtime": 163.4359, "train_tokens_per_second": 983.382 }, { "epoch": 0.004499889511641455, "grad_norm": 3.985139846801758, "learning_rate": 1.9498746867167923e-05, "loss": 0.32290276885032654, "num_input_tokens_seen": 163258, "step": 56, "train_runtime": 165.9031, "train_tokens_per_second": 984.056 }, { "epoch": 0.004580244681492196, "grad_norm": 3.0357258319854736, "learning_rate": 1.948872180451128e-05, "loss": 0.3405299782752991, "num_input_tokens_seen": 164968, "step": 57, "train_runtime": 168.1456, "train_tokens_per_second": 981.102 }, { "epoch": 0.004660599851342936, "grad_norm": 2.9525301456451416, "learning_rate": 1.947869674185464e-05, "loss": 0.3019135594367981, "num_input_tokens_seen": 169658, "step": 58, "train_runtime": 171.149, "train_tokens_per_second": 991.288 }, { "epoch": 0.004740955021193676, "grad_norm": 3.2162818908691406, "learning_rate": 1.9468671679197996e-05, "loss": 0.3009644150733948, "num_input_tokens_seen": 172196, "step": 59, "train_runtime": 173.6475, "train_tokens_per_second": 991.641 }, { "epoch": 0.004821310191044416, "grad_norm": 3.1514010429382324, "learning_rate": 1.9458646616541353e-05, "loss": 0.2679328918457031, "num_input_tokens_seen": 174778, "step": 60, "train_runtime": 176.1812, "train_tokens_per_second": 992.036 }, { "epoch": 0.004901665360895157, "grad_norm": 3.1374621391296387, "learning_rate": 1.9448621553884713e-05, "loss": 0.37104442715644836, "num_input_tokens_seen": 178284, "step": 61, "train_runtime": 190.8222, "train_tokens_per_second": 934.294 }, { "epoch": 0.004982020530745897, "grad_norm": 3.6638741493225098, "learning_rate": 1.9438596491228074e-05, "loss": 0.36299383640289307, "num_input_tokens_seen": 181958, "step": 62, "train_runtime": 193.6282, "train_tokens_per_second": 939.729 }, { "epoch": 0.005062375700596637, "grad_norm": 3.1998159885406494, "learning_rate": 1.942857142857143e-05, "loss": 0.34475719928741455, "num_input_tokens_seen": 185142, "step": 63, "train_runtime": 196.3399, "train_tokens_per_second": 942.967 }, { "epoch": 0.005142730870447377, "grad_norm": 3.673114538192749, "learning_rate": 1.9418546365914787e-05, "loss": 0.3713194727897644, "num_input_tokens_seen": 187328, "step": 64, "train_runtime": 198.8321, "train_tokens_per_second": 942.142 }, { "epoch": 0.005223086040298117, "grad_norm": 3.516136407852173, "learning_rate": 1.9408521303258147e-05, "loss": 0.36788931488990784, "num_input_tokens_seen": 189348, "step": 65, "train_runtime": 201.2305, "train_tokens_per_second": 940.951 }, { "epoch": 0.005303441210148858, "grad_norm": 3.6462159156799316, "learning_rate": 1.9398496240601504e-05, "loss": 0.35305100679397583, "num_input_tokens_seen": 191928, "step": 66, "train_runtime": 203.7343, "train_tokens_per_second": 942.05 }, { "epoch": 0.005383796379999598, "grad_norm": 4.176733016967773, "learning_rate": 1.9388471177944864e-05, "loss": 0.34327852725982666, "num_input_tokens_seen": 194596, "step": 67, "train_runtime": 206.2736, "train_tokens_per_second": 943.388 }, { "epoch": 0.005464151549850338, "grad_norm": 3.2090394496917725, "learning_rate": 1.9378446115288224e-05, "loss": 0.3244638741016388, "num_input_tokens_seen": 197120, "step": 68, "train_runtime": 208.7683, "train_tokens_per_second": 944.205 }, { "epoch": 0.0055445067197010785, "grad_norm": 3.1368675231933594, "learning_rate": 1.936842105263158e-05, "loss": 0.2839701771736145, "num_input_tokens_seen": 200682, "step": 69, "train_runtime": 211.5237, "train_tokens_per_second": 948.745 }, { "epoch": 0.0056248618895518195, "grad_norm": 2.9911038875579834, "learning_rate": 1.9358395989974938e-05, "loss": 0.2502208352088928, "num_input_tokens_seen": 204614, "step": 70, "train_runtime": 214.4615, "train_tokens_per_second": 954.082 }, { "epoch": 0.0057052170594025595, "grad_norm": 3.3676953315734863, "learning_rate": 1.9348370927318295e-05, "loss": 0.3486398458480835, "num_input_tokens_seen": 206430, "step": 71, "train_runtime": 216.7906, "train_tokens_per_second": 952.209 }, { "epoch": 0.0057855722292533, "grad_norm": 3.650989055633545, "learning_rate": 1.9338345864661655e-05, "loss": 0.36962807178497314, "num_input_tokens_seen": 208714, "step": 72, "train_runtime": 219.2243, "train_tokens_per_second": 952.057 }, { "epoch": 0.00586592739910404, "grad_norm": 2.902194023132324, "learning_rate": 1.9328320802005015e-05, "loss": 0.3146677613258362, "num_input_tokens_seen": 211176, "step": 73, "train_runtime": 221.7005, "train_tokens_per_second": 952.528 }, { "epoch": 0.00594628256895478, "grad_norm": 4.142306327819824, "learning_rate": 1.9318295739348372e-05, "loss": 0.3711496889591217, "num_input_tokens_seen": 215008, "step": 74, "train_runtime": 224.4884, "train_tokens_per_second": 957.769 }, { "epoch": 0.006026637738805521, "grad_norm": 3.4705777168273926, "learning_rate": 1.9308270676691732e-05, "loss": 0.35939905047416687, "num_input_tokens_seen": 217532, "step": 75, "train_runtime": 227.0355, "train_tokens_per_second": 958.141 }, { "epoch": 0.006106992908656261, "grad_norm": 2.7905337810516357, "learning_rate": 1.929824561403509e-05, "loss": 0.32816433906555176, "num_input_tokens_seen": 220410, "step": 76, "train_runtime": 229.6387, "train_tokens_per_second": 959.812 }, { "epoch": 0.006187348078507001, "grad_norm": 2.5559539794921875, "learning_rate": 1.9288220551378446e-05, "loss": 0.2607327401638031, "num_input_tokens_seen": 224342, "step": 77, "train_runtime": 232.4905, "train_tokens_per_second": 964.951 }, { "epoch": 0.006267703248357741, "grad_norm": 3.0967226028442383, "learning_rate": 1.9278195488721806e-05, "loss": 0.2515665888786316, "num_input_tokens_seen": 226328, "step": 78, "train_runtime": 234.8649, "train_tokens_per_second": 963.652 }, { "epoch": 0.006348058418208482, "grad_norm": 2.7118916511535645, "learning_rate": 1.9268170426065166e-05, "loss": 0.2946966290473938, "num_input_tokens_seen": 229720, "step": 79, "train_runtime": 237.7187, "train_tokens_per_second": 966.352 }, { "epoch": 0.006428413588059222, "grad_norm": 3.1221718788146973, "learning_rate": 1.9258145363408523e-05, "loss": 0.30246037244796753, "num_input_tokens_seen": 231458, "step": 80, "train_runtime": 240.0373, "train_tokens_per_second": 964.259 }, { "epoch": 0.006508768757909962, "grad_norm": 3.5483312606811523, "learning_rate": 1.924812030075188e-05, "loss": 0.3221529722213745, "num_input_tokens_seen": 233346, "step": 81, "train_runtime": 242.4818, "train_tokens_per_second": 962.324 }, { "epoch": 0.006589123927760702, "grad_norm": 3.2886037826538086, "learning_rate": 1.923809523809524e-05, "loss": 0.2958499491214752, "num_input_tokens_seen": 235828, "step": 82, "train_runtime": 244.986, "train_tokens_per_second": 962.618 }, { "epoch": 0.006669479097611442, "grad_norm": 3.265742301940918, "learning_rate": 1.9228070175438597e-05, "loss": 0.32420074939727783, "num_input_tokens_seen": 238092, "step": 83, "train_runtime": 247.4285, "train_tokens_per_second": 962.266 }, { "epoch": 0.006749834267462183, "grad_norm": 3.408738851547241, "learning_rate": 1.9218045112781957e-05, "loss": 0.3219103515148163, "num_input_tokens_seen": 240922, "step": 84, "train_runtime": 250.0085, "train_tokens_per_second": 963.655 }, { "epoch": 0.006830189437312923, "grad_norm": 3.9614155292510986, "learning_rate": 1.9208020050125317e-05, "loss": 0.46318304538726807, "num_input_tokens_seen": 243884, "step": 85, "train_runtime": 252.6135, "train_tokens_per_second": 965.443 }, { "epoch": 0.006910544607163663, "grad_norm": 3.6170296669006348, "learning_rate": 1.9197994987468674e-05, "loss": 0.35194239020347595, "num_input_tokens_seen": 246216, "step": 86, "train_runtime": 255.0475, "train_tokens_per_second": 965.373 }, { "epoch": 0.006990899777014403, "grad_norm": 3.0530805587768555, "learning_rate": 1.918796992481203e-05, "loss": 0.41397809982299805, "num_input_tokens_seen": 251602, "step": 87, "train_runtime": 258.2524, "train_tokens_per_second": 974.249 }, { "epoch": 0.007071254946865144, "grad_norm": 3.1978719234466553, "learning_rate": 1.917794486215539e-05, "loss": 0.2865869700908661, "num_input_tokens_seen": 254184, "step": 88, "train_runtime": 260.7828, "train_tokens_per_second": 974.696 }, { "epoch": 0.0071516101167158844, "grad_norm": 3.1670546531677246, "learning_rate": 1.9167919799498748e-05, "loss": 0.28800493478775024, "num_input_tokens_seen": 258414, "step": 89, "train_runtime": 263.6866, "train_tokens_per_second": 980.004 }, { "epoch": 0.0072319652865666245, "grad_norm": 3.514169692993164, "learning_rate": 1.9157894736842108e-05, "loss": 0.2796398997306824, "num_input_tokens_seen": 260526, "step": 90, "train_runtime": 266.0642, "train_tokens_per_second": 979.185 }, { "epoch": 0.007312320456417365, "grad_norm": 3.867985725402832, "learning_rate": 1.9147869674185465e-05, "loss": 0.32255056500434875, "num_input_tokens_seen": 262442, "step": 91, "train_runtime": 280.0226, "train_tokens_per_second": 937.217 }, { "epoch": 0.007392675626268105, "grad_norm": 3.6463613510131836, "learning_rate": 1.9137844611528825e-05, "loss": 0.3378679156303406, "num_input_tokens_seen": 264318, "step": 92, "train_runtime": 282.5745, "train_tokens_per_second": 935.392 }, { "epoch": 0.007473030796118846, "grad_norm": 3.333885908126831, "learning_rate": 1.912781954887218e-05, "loss": 0.3246857225894928, "num_input_tokens_seen": 266054, "step": 93, "train_runtime": 284.9646, "train_tokens_per_second": 933.639 }, { "epoch": 0.007553385965969586, "grad_norm": 3.541717052459717, "learning_rate": 1.911779448621554e-05, "loss": 0.3612404763698578, "num_input_tokens_seen": 269346, "step": 94, "train_runtime": 287.6684, "train_tokens_per_second": 936.307 }, { "epoch": 0.007633741135820326, "grad_norm": 2.642096757888794, "learning_rate": 1.91077694235589e-05, "loss": 0.1868799924850464, "num_input_tokens_seen": 271172, "step": 95, "train_runtime": 289.9498, "train_tokens_per_second": 935.238 }, { "epoch": 0.007714096305671066, "grad_norm": 2.7246742248535156, "learning_rate": 1.909774436090226e-05, "loss": 0.2878131568431854, "num_input_tokens_seen": 273134, "step": 96, "train_runtime": 292.3761, "train_tokens_per_second": 934.187 }, { "epoch": 0.007794451475521807, "grad_norm": 2.916330337524414, "learning_rate": 1.9087719298245616e-05, "loss": 0.30803316831588745, "num_input_tokens_seen": 275832, "step": 97, "train_runtime": 294.999, "train_tokens_per_second": 935.027 }, { "epoch": 0.007874806645372547, "grad_norm": 3.045095682144165, "learning_rate": 1.9077694235588976e-05, "loss": 0.22688502073287964, "num_input_tokens_seen": 277816, "step": 98, "train_runtime": 297.4487, "train_tokens_per_second": 933.996 }, { "epoch": 0.007955161815223287, "grad_norm": 3.4464447498321533, "learning_rate": 1.9067669172932333e-05, "loss": 0.3309476673603058, "num_input_tokens_seen": 280236, "step": 99, "train_runtime": 300.0289, "train_tokens_per_second": 934.03 }, { "epoch": 0.008035516985074027, "grad_norm": 3.2731871604919434, "learning_rate": 1.905764411027569e-05, "loss": 0.3433924913406372, "num_input_tokens_seen": 285570, "step": 100, "train_runtime": 303.3553, "train_tokens_per_second": 941.371 }, { "epoch": 0.008115872154924767, "grad_norm": 3.1046175956726074, "learning_rate": 1.904761904761905e-05, "loss": 0.31572702527046204, "num_input_tokens_seen": 288588, "step": 101, "train_runtime": 306.0421, "train_tokens_per_second": 942.968 }, { "epoch": 0.008196227324775507, "grad_norm": 3.2183034420013428, "learning_rate": 1.903759398496241e-05, "loss": 0.30764836072921753, "num_input_tokens_seen": 292002, "step": 102, "train_runtime": 309.0193, "train_tokens_per_second": 944.931 }, { "epoch": 0.008276582494626247, "grad_norm": 2.9037692546844482, "learning_rate": 1.9027568922305766e-05, "loss": 0.3522126078605652, "num_input_tokens_seen": 294522, "step": 103, "train_runtime": 311.593, "train_tokens_per_second": 945.214 }, { "epoch": 0.008356937664476989, "grad_norm": 3.2714290618896484, "learning_rate": 1.9017543859649123e-05, "loss": 0.3230966329574585, "num_input_tokens_seen": 297246, "step": 104, "train_runtime": 314.1951, "train_tokens_per_second": 946.056 }, { "epoch": 0.00843729283432773, "grad_norm": 3.6126554012298584, "learning_rate": 1.9007518796992483e-05, "loss": 0.3347752094268799, "num_input_tokens_seen": 299016, "step": 105, "train_runtime": 316.5504, "train_tokens_per_second": 944.608 }, { "epoch": 0.00851764800417847, "grad_norm": 3.9401164054870605, "learning_rate": 1.899749373433584e-05, "loss": 0.3375231623649597, "num_input_tokens_seen": 302906, "step": 106, "train_runtime": 319.4026, "train_tokens_per_second": 948.352 }, { "epoch": 0.00859800317402921, "grad_norm": 3.15846586227417, "learning_rate": 1.89874686716792e-05, "loss": 0.38462358713150024, "num_input_tokens_seen": 305334, "step": 107, "train_runtime": 321.903, "train_tokens_per_second": 948.528 }, { "epoch": 0.00867835834387995, "grad_norm": 3.519613027572632, "learning_rate": 1.8977443609022557e-05, "loss": 0.44392335414886475, "num_input_tokens_seen": 307528, "step": 108, "train_runtime": 324.3116, "train_tokens_per_second": 948.249 }, { "epoch": 0.00875871351373069, "grad_norm": 2.890005588531494, "learning_rate": 1.8967418546365917e-05, "loss": 0.2595115602016449, "num_input_tokens_seen": 310466, "step": 109, "train_runtime": 326.8675, "train_tokens_per_second": 949.822 }, { "epoch": 0.00883906868358143, "grad_norm": 3.033766746520996, "learning_rate": 1.8957393483709274e-05, "loss": 0.22816655039787292, "num_input_tokens_seen": 313058, "step": 110, "train_runtime": 329.3192, "train_tokens_per_second": 950.622 }, { "epoch": 0.00891942385343217, "grad_norm": 3.4524760246276855, "learning_rate": 1.894736842105263e-05, "loss": 0.33903974294662476, "num_input_tokens_seen": 315904, "step": 111, "train_runtime": 331.9117, "train_tokens_per_second": 951.771 }, { "epoch": 0.00899977902328291, "grad_norm": 3.3936610221862793, "learning_rate": 1.893734335839599e-05, "loss": 0.27746936678886414, "num_input_tokens_seen": 318634, "step": 112, "train_runtime": 334.3932, "train_tokens_per_second": 952.872 }, { "epoch": 0.009080134193133652, "grad_norm": 3.1677498817443848, "learning_rate": 1.892731829573935e-05, "loss": 0.3661844730377197, "num_input_tokens_seen": 322506, "step": 113, "train_runtime": 337.2312, "train_tokens_per_second": 956.335 }, { "epoch": 0.009160489362984392, "grad_norm": 3.9176061153411865, "learning_rate": 1.8917293233082708e-05, "loss": 0.36736249923706055, "num_input_tokens_seen": 324862, "step": 114, "train_runtime": 339.6919, "train_tokens_per_second": 956.343 }, { "epoch": 0.009240844532835132, "grad_norm": 2.6952080726623535, "learning_rate": 1.890726817042607e-05, "loss": 0.28575974702835083, "num_input_tokens_seen": 329088, "step": 115, "train_runtime": 342.7247, "train_tokens_per_second": 960.211 }, { "epoch": 0.009321199702685872, "grad_norm": 3.4457948207855225, "learning_rate": 1.8897243107769425e-05, "loss": 0.3582334518432617, "num_input_tokens_seen": 331290, "step": 116, "train_runtime": 345.1649, "train_tokens_per_second": 959.802 }, { "epoch": 0.009401554872536612, "grad_norm": 3.1988351345062256, "learning_rate": 1.8887218045112782e-05, "loss": 0.299877792596817, "num_input_tokens_seen": 333874, "step": 117, "train_runtime": 347.6554, "train_tokens_per_second": 960.359 }, { "epoch": 0.009481910042387352, "grad_norm": 3.783158302307129, "learning_rate": 1.8877192982456142e-05, "loss": 0.36389392614364624, "num_input_tokens_seen": 335806, "step": 118, "train_runtime": 349.9727, "train_tokens_per_second": 959.521 }, { "epoch": 0.009562265212238092, "grad_norm": 3.7515358924865723, "learning_rate": 1.88671679197995e-05, "loss": 0.3550061583518982, "num_input_tokens_seen": 338378, "step": 119, "train_runtime": 352.4745, "train_tokens_per_second": 960.007 }, { "epoch": 0.009642620382088832, "grad_norm": 3.009211778640747, "learning_rate": 1.885714285714286e-05, "loss": 0.29806363582611084, "num_input_tokens_seen": 340922, "step": 120, "train_runtime": 354.9705, "train_tokens_per_second": 960.424 }, { "epoch": 0.009722975551939572, "grad_norm": 3.4795329570770264, "learning_rate": 1.8847117794486216e-05, "loss": 0.29093825817108154, "num_input_tokens_seen": 342626, "step": 121, "train_runtime": 368.6756, "train_tokens_per_second": 929.343 }, { "epoch": 0.009803330721790314, "grad_norm": 4.135226726531982, "learning_rate": 1.8837092731829576e-05, "loss": 0.3835040032863617, "num_input_tokens_seen": 344490, "step": 122, "train_runtime": 371.1672, "train_tokens_per_second": 928.126 }, { "epoch": 0.009883685891641054, "grad_norm": 3.2283291816711426, "learning_rate": 1.8827067669172933e-05, "loss": 0.3511234521865845, "num_input_tokens_seen": 347710, "step": 123, "train_runtime": 373.8199, "train_tokens_per_second": 930.154 }, { "epoch": 0.009964041061491794, "grad_norm": 4.0754923820495605, "learning_rate": 1.8817042606516293e-05, "loss": 0.35088321566581726, "num_input_tokens_seen": 349898, "step": 124, "train_runtime": 376.2251, "train_tokens_per_second": 930.023 }, { "epoch": 0.010044396231342534, "grad_norm": 2.5573019981384277, "learning_rate": 1.880701754385965e-05, "loss": 0.2523025870323181, "num_input_tokens_seen": 352324, "step": 125, "train_runtime": 378.688, "train_tokens_per_second": 930.381 }, { "epoch": 0.010124751401193274, "grad_norm": 3.235337734222412, "learning_rate": 1.879699248120301e-05, "loss": 0.37001633644104004, "num_input_tokens_seen": 355216, "step": 126, "train_runtime": 381.3584, "train_tokens_per_second": 931.449 }, { "epoch": 0.010205106571044014, "grad_norm": 3.5324137210845947, "learning_rate": 1.8786967418546367e-05, "loss": 0.4304981827735901, "num_input_tokens_seen": 358336, "step": 127, "train_runtime": 384.1306, "train_tokens_per_second": 932.849 }, { "epoch": 0.010285461740894754, "grad_norm": 3.339933156967163, "learning_rate": 1.8776942355889727e-05, "loss": 0.25890880823135376, "num_input_tokens_seen": 359980, "step": 128, "train_runtime": 386.5679, "train_tokens_per_second": 931.221 }, { "epoch": 0.010365816910745495, "grad_norm": 3.2004621028900146, "learning_rate": 1.8766917293233084e-05, "loss": 0.2821299731731415, "num_input_tokens_seen": 362594, "step": 129, "train_runtime": 389.1974, "train_tokens_per_second": 931.645 }, { "epoch": 0.010446172080596235, "grad_norm": 3.3605291843414307, "learning_rate": 1.8756892230576444e-05, "loss": 0.38272252678871155, "num_input_tokens_seen": 364876, "step": 130, "train_runtime": 391.6761, "train_tokens_per_second": 931.576 }, { "epoch": 0.010526527250446976, "grad_norm": 3.6350884437561035, "learning_rate": 1.87468671679198e-05, "loss": 0.2878875732421875, "num_input_tokens_seen": 368372, "step": 131, "train_runtime": 394.3933, "train_tokens_per_second": 934.022 }, { "epoch": 0.010606882420297717, "grad_norm": 4.783580303192139, "learning_rate": 1.873684210526316e-05, "loss": 0.27603620290756226, "num_input_tokens_seen": 370650, "step": 132, "train_runtime": 396.8822, "train_tokens_per_second": 933.904 }, { "epoch": 0.010687237590148457, "grad_norm": 3.4207003116607666, "learning_rate": 1.8726817042606518e-05, "loss": 0.2751544117927551, "num_input_tokens_seen": 371918, "step": 133, "train_runtime": 399.173, "train_tokens_per_second": 931.721 }, { "epoch": 0.010767592759999197, "grad_norm": 3.4688777923583984, "learning_rate": 1.8716791979949875e-05, "loss": 0.31182169914245605, "num_input_tokens_seen": 373712, "step": 134, "train_runtime": 401.5159, "train_tokens_per_second": 930.753 }, { "epoch": 0.010847947929849937, "grad_norm": 3.8189005851745605, "learning_rate": 1.8706766917293235e-05, "loss": 0.2983708679676056, "num_input_tokens_seen": 375978, "step": 135, "train_runtime": 403.9783, "train_tokens_per_second": 930.689 }, { "epoch": 0.010928303099700677, "grad_norm": 3.310286045074463, "learning_rate": 1.869674185463659e-05, "loss": 0.2808976173400879, "num_input_tokens_seen": 377336, "step": 136, "train_runtime": 406.2387, "train_tokens_per_second": 928.853 }, { "epoch": 0.011008658269551417, "grad_norm": 3.8449008464813232, "learning_rate": 1.868671679197995e-05, "loss": 0.28999945521354675, "num_input_tokens_seen": 384726, "step": 137, "train_runtime": 410.0803, "train_tokens_per_second": 938.172 }, { "epoch": 0.011089013439402157, "grad_norm": 2.5358874797821045, "learning_rate": 1.8676691729323312e-05, "loss": 0.24205231666564941, "num_input_tokens_seen": 386840, "step": 138, "train_runtime": 412.5353, "train_tokens_per_second": 937.714 }, { "epoch": 0.011169368609252897, "grad_norm": 3.34633731842041, "learning_rate": 1.866666666666667e-05, "loss": 0.21847444772720337, "num_input_tokens_seen": 388456, "step": 139, "train_runtime": 414.7971, "train_tokens_per_second": 936.496 }, { "epoch": 0.011249723779103639, "grad_norm": 3.2042272090911865, "learning_rate": 1.8656641604010025e-05, "loss": 0.33320945501327515, "num_input_tokens_seen": 390618, "step": 140, "train_runtime": 417.1969, "train_tokens_per_second": 936.292 }, { "epoch": 0.011330078948954379, "grad_norm": 3.511211633682251, "learning_rate": 1.8646616541353386e-05, "loss": 0.31219303607940674, "num_input_tokens_seen": 392626, "step": 141, "train_runtime": 419.5754, "train_tokens_per_second": 935.77 }, { "epoch": 0.011410434118805119, "grad_norm": 3.544792652130127, "learning_rate": 1.8636591478696742e-05, "loss": 0.309020459651947, "num_input_tokens_seen": 394652, "step": 142, "train_runtime": 422.0132, "train_tokens_per_second": 935.165 }, { "epoch": 0.01149078928865586, "grad_norm": 3.712691068649292, "learning_rate": 1.8626566416040103e-05, "loss": 0.3304578363895416, "num_input_tokens_seen": 396644, "step": 143, "train_runtime": 424.4241, "train_tokens_per_second": 934.546 }, { "epoch": 0.0115711444585066, "grad_norm": 2.4781103134155273, "learning_rate": 1.861654135338346e-05, "loss": 0.2742910087108612, "num_input_tokens_seen": 400900, "step": 144, "train_runtime": 427.3255, "train_tokens_per_second": 938.161 }, { "epoch": 0.01165149962835734, "grad_norm": 3.332270622253418, "learning_rate": 1.860651629072682e-05, "loss": 0.2767263948917389, "num_input_tokens_seen": 406740, "step": 145, "train_runtime": 430.6513, "train_tokens_per_second": 944.476 }, { "epoch": 0.01173185479820808, "grad_norm": 3.6439552307128906, "learning_rate": 1.8596491228070176e-05, "loss": 0.3286985754966736, "num_input_tokens_seen": 408922, "step": 146, "train_runtime": 433.1279, "train_tokens_per_second": 944.114 }, { "epoch": 0.01181220996805882, "grad_norm": 3.3875863552093506, "learning_rate": 1.8586466165413533e-05, "loss": 0.2598862051963806, "num_input_tokens_seen": 411180, "step": 147, "train_runtime": 435.6264, "train_tokens_per_second": 943.882 }, { "epoch": 0.01189256513790956, "grad_norm": 3.0596063137054443, "learning_rate": 1.8576441102756893e-05, "loss": 0.23106268048286438, "num_input_tokens_seen": 413016, "step": 148, "train_runtime": 437.9316, "train_tokens_per_second": 943.106 }, { "epoch": 0.011972920307760301, "grad_norm": 3.0926902294158936, "learning_rate": 1.8566416040100254e-05, "loss": 0.33295685052871704, "num_input_tokens_seen": 416140, "step": 149, "train_runtime": 440.6189, "train_tokens_per_second": 944.444 }, { "epoch": 0.012053275477611041, "grad_norm": 3.4687018394470215, "learning_rate": 1.855639097744361e-05, "loss": 0.2633710205554962, "num_input_tokens_seen": 418112, "step": 150, "train_runtime": 443.0333, "train_tokens_per_second": 943.749 }, { "epoch": 0.012133630647461782, "grad_norm": 3.5554466247558594, "learning_rate": 1.8546365914786967e-05, "loss": 0.24585884809494019, "num_input_tokens_seen": 422106, "step": 151, "train_runtime": 457.2604, "train_tokens_per_second": 923.12 }, { "epoch": 0.012213985817312522, "grad_norm": 3.000566005706787, "learning_rate": 1.8536340852130327e-05, "loss": 0.3520648181438446, "num_input_tokens_seen": 425326, "step": 152, "train_runtime": 459.9262, "train_tokens_per_second": 924.77 }, { "epoch": 0.012294340987163262, "grad_norm": 2.8457818031311035, "learning_rate": 1.8526315789473684e-05, "loss": 0.28346604108810425, "num_input_tokens_seen": 430664, "step": 153, "train_runtime": 463.1137, "train_tokens_per_second": 929.931 }, { "epoch": 0.012374696157014002, "grad_norm": 3.134417772293091, "learning_rate": 1.8516290726817044e-05, "loss": 0.31573018431663513, "num_input_tokens_seen": 432986, "step": 154, "train_runtime": 465.5339, "train_tokens_per_second": 930.085 }, { "epoch": 0.012455051326864742, "grad_norm": 3.6260793209075928, "learning_rate": 1.8506265664160404e-05, "loss": 0.3819829821586609, "num_input_tokens_seen": 435526, "step": 155, "train_runtime": 468.0106, "train_tokens_per_second": 930.59 }, { "epoch": 0.012535406496715482, "grad_norm": 3.070864200592041, "learning_rate": 1.849624060150376e-05, "loss": 0.33370253443717957, "num_input_tokens_seen": 438908, "step": 156, "train_runtime": 470.702, "train_tokens_per_second": 932.454 }, { "epoch": 0.012615761666566222, "grad_norm": 3.503136157989502, "learning_rate": 1.8486215538847118e-05, "loss": 0.32686448097229004, "num_input_tokens_seen": 441614, "step": 157, "train_runtime": 473.2452, "train_tokens_per_second": 933.161 }, { "epoch": 0.012696116836416964, "grad_norm": 3.129359006881714, "learning_rate": 1.8476190476190478e-05, "loss": 0.3641253709793091, "num_input_tokens_seen": 444594, "step": 158, "train_runtime": 476.0281, "train_tokens_per_second": 933.966 }, { "epoch": 0.012776472006267704, "grad_norm": 2.8638508319854736, "learning_rate": 1.8466165413533835e-05, "loss": 0.2783857583999634, "num_input_tokens_seen": 447042, "step": 159, "train_runtime": 478.5777, "train_tokens_per_second": 934.105 }, { "epoch": 0.012856827176118444, "grad_norm": 3.4965758323669434, "learning_rate": 1.8456140350877195e-05, "loss": 0.33323633670806885, "num_input_tokens_seen": 449378, "step": 160, "train_runtime": 481.0416, "train_tokens_per_second": 934.177 }, { "epoch": 0.012937182345969184, "grad_norm": 3.2660486698150635, "learning_rate": 1.8446115288220552e-05, "loss": 0.2735540270805359, "num_input_tokens_seen": 451002, "step": 161, "train_runtime": 483.3877, "train_tokens_per_second": 933.003 }, { "epoch": 0.013017537515819924, "grad_norm": 3.4547367095947266, "learning_rate": 1.8436090225563912e-05, "loss": 0.28112494945526123, "num_input_tokens_seen": 453018, "step": 162, "train_runtime": 485.8067, "train_tokens_per_second": 932.507 }, { "epoch": 0.013097892685670664, "grad_norm": 3.90035080909729, "learning_rate": 1.842606516290727e-05, "loss": 0.3287423253059387, "num_input_tokens_seen": 455666, "step": 163, "train_runtime": 488.4576, "train_tokens_per_second": 932.867 }, { "epoch": 0.013178247855521404, "grad_norm": 3.276613712310791, "learning_rate": 1.8416040100250626e-05, "loss": 0.2647658884525299, "num_input_tokens_seen": 458704, "step": 164, "train_runtime": 491.0581, "train_tokens_per_second": 934.114 }, { "epoch": 0.013258603025372144, "grad_norm": 3.1398704051971436, "learning_rate": 1.8406015037593986e-05, "loss": 0.3142634630203247, "num_input_tokens_seen": 462122, "step": 165, "train_runtime": 493.8596, "train_tokens_per_second": 935.736 }, { "epoch": 0.013338958195222884, "grad_norm": 5.588934898376465, "learning_rate": 1.8395989974937346e-05, "loss": 0.3987157344818115, "num_input_tokens_seen": 465348, "step": 166, "train_runtime": 496.5233, "train_tokens_per_second": 937.213 }, { "epoch": 0.013419313365073626, "grad_norm": 3.0949580669403076, "learning_rate": 1.8385964912280703e-05, "loss": 0.36319243907928467, "num_input_tokens_seen": 467798, "step": 167, "train_runtime": 499.0428, "train_tokens_per_second": 937.391 }, { "epoch": 0.013499668534924366, "grad_norm": 3.9428093433380127, "learning_rate": 1.8375939849624063e-05, "loss": 0.2748376727104187, "num_input_tokens_seen": 469458, "step": 168, "train_runtime": 501.4612, "train_tokens_per_second": 936.18 }, { "epoch": 0.013580023704775106, "grad_norm": 3.0122852325439453, "learning_rate": 1.836591478696742e-05, "loss": 0.2739015221595764, "num_input_tokens_seen": 471766, "step": 169, "train_runtime": 503.9968, "train_tokens_per_second": 936.05 }, { "epoch": 0.013660378874625847, "grad_norm": 3.3499648571014404, "learning_rate": 1.8355889724310777e-05, "loss": 0.3079832196235657, "num_input_tokens_seen": 473424, "step": 170, "train_runtime": 506.2511, "train_tokens_per_second": 935.156 }, { "epoch": 0.013740734044476587, "grad_norm": 2.7142419815063477, "learning_rate": 1.8345864661654137e-05, "loss": 0.22613102197647095, "num_input_tokens_seen": 476048, "step": 171, "train_runtime": 508.7747, "train_tokens_per_second": 935.675 }, { "epoch": 0.013821089214327327, "grad_norm": 3.5622923374176025, "learning_rate": 1.8335839598997497e-05, "loss": 0.3136799931526184, "num_input_tokens_seen": 478842, "step": 172, "train_runtime": 511.3115, "train_tokens_per_second": 936.498 }, { "epoch": 0.013901444384178067, "grad_norm": 3.6706583499908447, "learning_rate": 1.8325814536340854e-05, "loss": 0.20312413573265076, "num_input_tokens_seen": 480410, "step": 173, "train_runtime": 513.5962, "train_tokens_per_second": 935.385 }, { "epoch": 0.013981799554028807, "grad_norm": 2.960205078125, "learning_rate": 1.831578947368421e-05, "loss": 0.2867105305194855, "num_input_tokens_seen": 484278, "step": 174, "train_runtime": 516.4383, "train_tokens_per_second": 937.727 }, { "epoch": 0.014062154723879547, "grad_norm": 2.9632568359375, "learning_rate": 1.830576441102757e-05, "loss": 0.29494723677635193, "num_input_tokens_seen": 487098, "step": 175, "train_runtime": 519.0237, "train_tokens_per_second": 938.489 }, { "epoch": 0.014142509893730289, "grad_norm": 3.523531198501587, "learning_rate": 1.8295739348370928e-05, "loss": 0.33469343185424805, "num_input_tokens_seen": 489580, "step": 176, "train_runtime": 521.5063, "train_tokens_per_second": 938.781 }, { "epoch": 0.014222865063581029, "grad_norm": 4.074111461639404, "learning_rate": 1.8285714285714288e-05, "loss": 0.3341754674911499, "num_input_tokens_seen": 491342, "step": 177, "train_runtime": 523.7808, "train_tokens_per_second": 938.068 }, { "epoch": 0.014303220233431769, "grad_norm": 3.0969536304473877, "learning_rate": 1.8275689223057648e-05, "loss": 0.3931128978729248, "num_input_tokens_seen": 493974, "step": 178, "train_runtime": 526.3122, "train_tokens_per_second": 938.557 }, { "epoch": 0.014383575403282509, "grad_norm": 4.918959617614746, "learning_rate": 1.8265664160401005e-05, "loss": 0.34122270345687866, "num_input_tokens_seen": 495404, "step": 179, "train_runtime": 528.5165, "train_tokens_per_second": 937.348 }, { "epoch": 0.014463930573133249, "grad_norm": 2.8688724040985107, "learning_rate": 1.825563909774436e-05, "loss": 0.3398945927619934, "num_input_tokens_seen": 498320, "step": 180, "train_runtime": 531.0995, "train_tokens_per_second": 938.28 }, { "epoch": 0.014544285742983989, "grad_norm": 2.803664445877075, "learning_rate": 1.824561403508772e-05, "loss": 0.18443168699741364, "num_input_tokens_seen": 500406, "step": 181, "train_runtime": 545.1552, "train_tokens_per_second": 917.915 }, { "epoch": 0.01462464091283473, "grad_norm": 3.303579568862915, "learning_rate": 1.823558897243108e-05, "loss": 0.2465621381998062, "num_input_tokens_seen": 503466, "step": 182, "train_runtime": 547.8081, "train_tokens_per_second": 919.055 }, { "epoch": 0.01470499608268547, "grad_norm": 3.534769296646118, "learning_rate": 1.822556390977444e-05, "loss": 0.2632983922958374, "num_input_tokens_seen": 505476, "step": 183, "train_runtime": 550.1795, "train_tokens_per_second": 918.747 }, { "epoch": 0.01478535125253621, "grad_norm": 3.037271499633789, "learning_rate": 1.8215538847117796e-05, "loss": 0.2662435472011566, "num_input_tokens_seen": 508060, "step": 184, "train_runtime": 552.7072, "train_tokens_per_second": 919.221 }, { "epoch": 0.014865706422386951, "grad_norm": 2.9402449131011963, "learning_rate": 1.8205513784461156e-05, "loss": 0.29634687304496765, "num_input_tokens_seen": 510638, "step": 185, "train_runtime": 555.2283, "train_tokens_per_second": 919.69 }, { "epoch": 0.014946061592237691, "grad_norm": 4.127347946166992, "learning_rate": 1.8195488721804512e-05, "loss": 0.28563880920410156, "num_input_tokens_seen": 512494, "step": 186, "train_runtime": 557.5551, "train_tokens_per_second": 919.181 }, { "epoch": 0.015026416762088431, "grad_norm": 3.0700926780700684, "learning_rate": 1.818546365914787e-05, "loss": 0.293647438287735, "num_input_tokens_seen": 514994, "step": 187, "train_runtime": 560.0325, "train_tokens_per_second": 919.579 }, { "epoch": 0.015106771931939171, "grad_norm": 2.573157548904419, "learning_rate": 1.817543859649123e-05, "loss": 0.27615776658058167, "num_input_tokens_seen": 517600, "step": 188, "train_runtime": 562.5263, "train_tokens_per_second": 920.135 }, { "epoch": 0.015187127101789911, "grad_norm": 3.1237025260925293, "learning_rate": 1.816541353383459e-05, "loss": 0.31244227290153503, "num_input_tokens_seen": 524006, "step": 189, "train_runtime": 566.0702, "train_tokens_per_second": 925.691 }, { "epoch": 0.015267482271640652, "grad_norm": 4.022103786468506, "learning_rate": 1.8155388471177946e-05, "loss": 0.3071961998939514, "num_input_tokens_seen": 526242, "step": 190, "train_runtime": 568.5783, "train_tokens_per_second": 925.54 }, { "epoch": 0.015347837441491392, "grad_norm": 2.7209904193878174, "learning_rate": 1.8145363408521303e-05, "loss": 0.3356509506702423, "num_input_tokens_seen": 528778, "step": 191, "train_runtime": 571.1978, "train_tokens_per_second": 925.735 }, { "epoch": 0.015428192611342132, "grad_norm": 2.8003807067871094, "learning_rate": 1.8135338345864663e-05, "loss": 0.306357741355896, "num_input_tokens_seen": 531528, "step": 192, "train_runtime": 573.7706, "train_tokens_per_second": 926.377 }, { "epoch": 0.015508547781192872, "grad_norm": 3.5681028366088867, "learning_rate": 1.812531328320802e-05, "loss": 0.3600728511810303, "num_input_tokens_seen": 533018, "step": 193, "train_runtime": 575.9898, "train_tokens_per_second": 925.395 }, { "epoch": 0.015588902951043614, "grad_norm": 3.398167133331299, "learning_rate": 1.811528822055138e-05, "loss": 0.28078368306159973, "num_input_tokens_seen": 534850, "step": 194, "train_runtime": 578.2977, "train_tokens_per_second": 924.87 }, { "epoch": 0.015669258120894352, "grad_norm": 3.371738910675049, "learning_rate": 1.810526315789474e-05, "loss": 0.3097071945667267, "num_input_tokens_seen": 537720, "step": 195, "train_runtime": 580.9408, "train_tokens_per_second": 925.602 }, { "epoch": 0.015749613290745094, "grad_norm": 2.9839794635772705, "learning_rate": 1.8095238095238097e-05, "loss": 0.35705384612083435, "num_input_tokens_seen": 539378, "step": 196, "train_runtime": 583.2662, "train_tokens_per_second": 924.754 }, { "epoch": 0.015829968460595832, "grad_norm": 3.936265468597412, "learning_rate": 1.8085213032581454e-05, "loss": 0.4094323515892029, "num_input_tokens_seen": 540996, "step": 197, "train_runtime": 585.5502, "train_tokens_per_second": 923.911 }, { "epoch": 0.015910323630446574, "grad_norm": 2.447577953338623, "learning_rate": 1.8075187969924814e-05, "loss": 0.4294370412826538, "num_input_tokens_seen": 546750, "step": 198, "train_runtime": 588.8837, "train_tokens_per_second": 928.452 }, { "epoch": 0.015990678800297316, "grad_norm": 3.1359894275665283, "learning_rate": 1.806516290726817e-05, "loss": 0.30995383858680725, "num_input_tokens_seen": 549122, "step": 199, "train_runtime": 591.4333, "train_tokens_per_second": 928.46 }, { "epoch": 0.016071033970148054, "grad_norm": 3.35567569732666, "learning_rate": 1.805513784461153e-05, "loss": 0.2711547613143921, "num_input_tokens_seen": 551614, "step": 200, "train_runtime": 593.9139, "train_tokens_per_second": 928.778 }, { "epoch": 0.016071033970148054, "eval_loss": 1.2120285034179688, "eval_runtime": 69.0037, "eval_samples_per_second": 14.405, "eval_steps_per_second": 7.203, "num_input_tokens_seen": 551614, "step": 200 }, { "epoch": 0.016151389139998796, "grad_norm": 3.460120439529419, "learning_rate": 1.8045112781954888e-05, "loss": 0.3592950999736786, "num_input_tokens_seen": 554984, "step": 201, "train_runtime": 665.6885, "train_tokens_per_second": 833.699 }, { "epoch": 0.016231744309849534, "grad_norm": 3.522367238998413, "learning_rate": 1.8035087719298248e-05, "loss": 0.33141809701919556, "num_input_tokens_seen": 559668, "step": 202, "train_runtime": 668.7796, "train_tokens_per_second": 836.85 }, { "epoch": 0.016312099479700276, "grad_norm": 3.695065498352051, "learning_rate": 1.8025062656641605e-05, "loss": 0.3351621627807617, "num_input_tokens_seen": 562086, "step": 203, "train_runtime": 671.2511, "train_tokens_per_second": 837.371 }, { "epoch": 0.016392454649551014, "grad_norm": 2.8964600563049316, "learning_rate": 1.8015037593984962e-05, "loss": 0.3249494433403015, "num_input_tokens_seen": 564026, "step": 204, "train_runtime": 673.6057, "train_tokens_per_second": 837.324 }, { "epoch": 0.016472809819401756, "grad_norm": 3.574324607849121, "learning_rate": 1.8005012531328322e-05, "loss": 0.32394301891326904, "num_input_tokens_seen": 565572, "step": 205, "train_runtime": 675.8173, "train_tokens_per_second": 836.871 }, { "epoch": 0.016553164989252495, "grad_norm": 3.7345569133758545, "learning_rate": 1.7994987468671682e-05, "loss": 0.32426610589027405, "num_input_tokens_seen": 568346, "step": 206, "train_runtime": 678.4065, "train_tokens_per_second": 837.766 }, { "epoch": 0.016633520159103236, "grad_norm": 2.77427077293396, "learning_rate": 1.798496240601504e-05, "loss": 0.1888466775417328, "num_input_tokens_seen": 571736, "step": 207, "train_runtime": 681.1998, "train_tokens_per_second": 839.307 }, { "epoch": 0.016713875328953978, "grad_norm": 2.8614933490753174, "learning_rate": 1.79749373433584e-05, "loss": 0.2676595151424408, "num_input_tokens_seen": 573780, "step": 208, "train_runtime": 683.6456, "train_tokens_per_second": 839.295 }, { "epoch": 0.016794230498804717, "grad_norm": 2.7828643321990967, "learning_rate": 1.7964912280701756e-05, "loss": 0.26680800318717957, "num_input_tokens_seen": 577094, "step": 209, "train_runtime": 686.3545, "train_tokens_per_second": 840.81 }, { "epoch": 0.01687458566865546, "grad_norm": 2.821056365966797, "learning_rate": 1.7954887218045113e-05, "loss": 0.3167721927165985, "num_input_tokens_seen": 579430, "step": 210, "train_runtime": 688.8355, "train_tokens_per_second": 841.173 }, { "epoch": 0.016954940838506197, "grad_norm": 3.8467915058135986, "learning_rate": 1.7944862155388473e-05, "loss": 0.343126505613327, "num_input_tokens_seen": 581858, "step": 211, "train_runtime": 703.4193, "train_tokens_per_second": 827.185 }, { "epoch": 0.01703529600835694, "grad_norm": 3.1451239585876465, "learning_rate": 1.7934837092731833e-05, "loss": 0.33314090967178345, "num_input_tokens_seen": 585862, "step": 212, "train_runtime": 706.3866, "train_tokens_per_second": 829.379 }, { "epoch": 0.017115651178207677, "grad_norm": 3.04956316947937, "learning_rate": 1.792481203007519e-05, "loss": 0.30027827620506287, "num_input_tokens_seen": 589314, "step": 213, "train_runtime": 709.2711, "train_tokens_per_second": 830.873 }, { "epoch": 0.01719600634805842, "grad_norm": 2.990105628967285, "learning_rate": 1.7914786967418547e-05, "loss": 0.29545435309410095, "num_input_tokens_seen": 591532, "step": 214, "train_runtime": 711.7593, "train_tokens_per_second": 831.084 }, { "epoch": 0.017276361517909157, "grad_norm": 3.047029972076416, "learning_rate": 1.7904761904761907e-05, "loss": 0.34850576519966125, "num_input_tokens_seen": 594994, "step": 215, "train_runtime": 714.5858, "train_tokens_per_second": 832.642 }, { "epoch": 0.0173567166877599, "grad_norm": 3.086738348007202, "learning_rate": 1.7894736842105264e-05, "loss": 0.3032008111476898, "num_input_tokens_seen": 597280, "step": 216, "train_runtime": 717.1046, "train_tokens_per_second": 832.905 }, { "epoch": 0.01743707185761064, "grad_norm": 3.0148956775665283, "learning_rate": 1.7884711779448624e-05, "loss": 0.2793583273887634, "num_input_tokens_seen": 600080, "step": 217, "train_runtime": 719.6989, "train_tokens_per_second": 833.793 }, { "epoch": 0.01751742702746138, "grad_norm": 2.565553665161133, "learning_rate": 1.7874686716791984e-05, "loss": 0.2778613567352295, "num_input_tokens_seen": 601900, "step": 218, "train_runtime": 722.0386, "train_tokens_per_second": 833.612 }, { "epoch": 0.01759778219731212, "grad_norm": 2.57916522026062, "learning_rate": 1.786466165413534e-05, "loss": 0.23286838829517365, "num_input_tokens_seen": 604262, "step": 219, "train_runtime": 724.5579, "train_tokens_per_second": 833.973 }, { "epoch": 0.01767813736716286, "grad_norm": 3.229583740234375, "learning_rate": 1.7854636591478698e-05, "loss": 0.30804774165153503, "num_input_tokens_seen": 608006, "step": 220, "train_runtime": 727.4124, "train_tokens_per_second": 835.848 }, { "epoch": 0.0177584925370136, "grad_norm": 3.204101085662842, "learning_rate": 1.7844611528822054e-05, "loss": 0.20986182987689972, "num_input_tokens_seen": 609914, "step": 221, "train_runtime": 729.8596, "train_tokens_per_second": 835.659 }, { "epoch": 0.01783884770686434, "grad_norm": 3.0256690979003906, "learning_rate": 1.7834586466165415e-05, "loss": 0.28531599044799805, "num_input_tokens_seen": 611676, "step": 222, "train_runtime": 732.2344, "train_tokens_per_second": 835.355 }, { "epoch": 0.01791920287671508, "grad_norm": 3.115753173828125, "learning_rate": 1.7824561403508775e-05, "loss": 0.3498820960521698, "num_input_tokens_seen": 613872, "step": 223, "train_runtime": 734.6495, "train_tokens_per_second": 835.598 }, { "epoch": 0.01799955804656582, "grad_norm": 2.816206693649292, "learning_rate": 1.781453634085213e-05, "loss": 0.27496087551116943, "num_input_tokens_seen": 618118, "step": 224, "train_runtime": 737.6069, "train_tokens_per_second": 838.005 }, { "epoch": 0.01807991321641656, "grad_norm": 2.4712586402893066, "learning_rate": 1.7804511278195492e-05, "loss": 0.26049908995628357, "num_input_tokens_seen": 620236, "step": 225, "train_runtime": 740.0602, "train_tokens_per_second": 838.089 }, { "epoch": 0.018160268386267303, "grad_norm": 3.0243327617645264, "learning_rate": 1.779448621553885e-05, "loss": 0.3580141067504883, "num_input_tokens_seen": 623482, "step": 226, "train_runtime": 742.8121, "train_tokens_per_second": 839.354 }, { "epoch": 0.01824062355611804, "grad_norm": 3.20993709564209, "learning_rate": 1.7784461152882205e-05, "loss": 0.27118682861328125, "num_input_tokens_seen": 626424, "step": 227, "train_runtime": 745.437, "train_tokens_per_second": 840.345 }, { "epoch": 0.018320978725968783, "grad_norm": 3.7725906372070312, "learning_rate": 1.7774436090225566e-05, "loss": 0.2128567099571228, "num_input_tokens_seen": 628968, "step": 228, "train_runtime": 747.9795, "train_tokens_per_second": 840.889 }, { "epoch": 0.01840133389581952, "grad_norm": 3.113862991333008, "learning_rate": 1.7764411027568926e-05, "loss": 0.3590608835220337, "num_input_tokens_seen": 632106, "step": 229, "train_runtime": 750.6466, "train_tokens_per_second": 842.082 }, { "epoch": 0.018481689065670263, "grad_norm": 3.5587961673736572, "learning_rate": 1.7754385964912283e-05, "loss": 0.4478112757205963, "num_input_tokens_seen": 635142, "step": 230, "train_runtime": 753.261, "train_tokens_per_second": 843.19 }, { "epoch": 0.018562044235521002, "grad_norm": 2.8759124279022217, "learning_rate": 1.774436090225564e-05, "loss": 0.2309218943119049, "num_input_tokens_seen": 638020, "step": 231, "train_runtime": 755.8684, "train_tokens_per_second": 844.089 }, { "epoch": 0.018642399405371744, "grad_norm": 2.8777658939361572, "learning_rate": 1.7734335839599e-05, "loss": 0.281022846698761, "num_input_tokens_seen": 640314, "step": 232, "train_runtime": 758.3154, "train_tokens_per_second": 844.39 }, { "epoch": 0.018722754575222482, "grad_norm": 2.938976287841797, "learning_rate": 1.7724310776942356e-05, "loss": 0.20252320170402527, "num_input_tokens_seen": 642332, "step": 233, "train_runtime": 760.7152, "train_tokens_per_second": 844.379 }, { "epoch": 0.018803109745073224, "grad_norm": 3.3122262954711914, "learning_rate": 1.7714285714285717e-05, "loss": 0.4204859137535095, "num_input_tokens_seen": 645296, "step": 234, "train_runtime": 763.4239, "train_tokens_per_second": 845.266 }, { "epoch": 0.018883464914923966, "grad_norm": 3.373497724533081, "learning_rate": 1.7704260651629073e-05, "loss": 0.3417308032512665, "num_input_tokens_seen": 648600, "step": 235, "train_runtime": 766.1121, "train_tokens_per_second": 846.612 }, { "epoch": 0.018963820084774704, "grad_norm": 3.3100321292877197, "learning_rate": 1.7694235588972433e-05, "loss": 0.33300453424453735, "num_input_tokens_seen": 650114, "step": 236, "train_runtime": 768.3412, "train_tokens_per_second": 846.127 }, { "epoch": 0.019044175254625446, "grad_norm": 3.2101967334747314, "learning_rate": 1.768421052631579e-05, "loss": 0.36575010418891907, "num_input_tokens_seen": 653974, "step": 237, "train_runtime": 771.1438, "train_tokens_per_second": 848.057 }, { "epoch": 0.019124530424476184, "grad_norm": 3.441819906234741, "learning_rate": 1.767418546365915e-05, "loss": 0.3431284725666046, "num_input_tokens_seen": 656296, "step": 238, "train_runtime": 773.6367, "train_tokens_per_second": 848.326 }, { "epoch": 0.019204885594326926, "grad_norm": 3.2599246501922607, "learning_rate": 1.7664160401002507e-05, "loss": 0.3860718309879303, "num_input_tokens_seen": 659866, "step": 239, "train_runtime": 776.4147, "train_tokens_per_second": 849.889 }, { "epoch": 0.019285240764177664, "grad_norm": 3.200822591781616, "learning_rate": 1.7654135338345867e-05, "loss": 0.2970229983329773, "num_input_tokens_seen": 663134, "step": 240, "train_runtime": 779.1345, "train_tokens_per_second": 851.116 }, { "epoch": 0.019365595934028406, "grad_norm": 3.2202234268188477, "learning_rate": 1.7644110275689224e-05, "loss": 0.23008516430854797, "num_input_tokens_seen": 665690, "step": 241, "train_runtime": 793.7488, "train_tokens_per_second": 838.666 }, { "epoch": 0.019445951103879144, "grad_norm": 2.4978902339935303, "learning_rate": 1.7634085213032584e-05, "loss": 0.24421465396881104, "num_input_tokens_seen": 669346, "step": 242, "train_runtime": 796.598, "train_tokens_per_second": 840.256 }, { "epoch": 0.019526306273729886, "grad_norm": 3.2488372325897217, "learning_rate": 1.762406015037594e-05, "loss": 0.3928283452987671, "num_input_tokens_seen": 672434, "step": 243, "train_runtime": 799.279, "train_tokens_per_second": 841.301 }, { "epoch": 0.019606661443580628, "grad_norm": 3.360460042953491, "learning_rate": 1.7614035087719298e-05, "loss": 0.36841315031051636, "num_input_tokens_seen": 678340, "step": 244, "train_runtime": 802.7218, "train_tokens_per_second": 845.05 }, { "epoch": 0.019687016613431366, "grad_norm": 3.1337029933929443, "learning_rate": 1.7604010025062658e-05, "loss": 0.31601768732070923, "num_input_tokens_seen": 679850, "step": 245, "train_runtime": 805.0063, "train_tokens_per_second": 844.528 }, { "epoch": 0.019767371783282108, "grad_norm": 3.5034542083740234, "learning_rate": 1.7593984962406015e-05, "loss": 0.32139885425567627, "num_input_tokens_seen": 684852, "step": 246, "train_runtime": 808.1192, "train_tokens_per_second": 847.464 }, { "epoch": 0.019847726953132847, "grad_norm": 2.8113763332366943, "learning_rate": 1.7583959899749375e-05, "loss": 0.3611815571784973, "num_input_tokens_seen": 688162, "step": 247, "train_runtime": 810.8299, "train_tokens_per_second": 848.713 }, { "epoch": 0.01992808212298359, "grad_norm": 3.0140609741210938, "learning_rate": 1.7573934837092735e-05, "loss": 0.26797425746917725, "num_input_tokens_seen": 690376, "step": 248, "train_runtime": 813.2817, "train_tokens_per_second": 848.877 }, { "epoch": 0.020008437292834327, "grad_norm": 3.4975757598876953, "learning_rate": 1.7563909774436092e-05, "loss": 0.4344942271709442, "num_input_tokens_seen": 694308, "step": 249, "train_runtime": 816.1549, "train_tokens_per_second": 850.706 }, { "epoch": 0.02008879246268507, "grad_norm": 2.786160945892334, "learning_rate": 1.755388471177945e-05, "loss": 0.28215643763542175, "num_input_tokens_seen": 696508, "step": 250, "train_runtime": 818.5837, "train_tokens_per_second": 850.87 }, { "epoch": 0.020169147632535807, "grad_norm": 2.7811715602874756, "learning_rate": 1.754385964912281e-05, "loss": 0.30155014991760254, "num_input_tokens_seen": 700740, "step": 251, "train_runtime": 821.6581, "train_tokens_per_second": 852.836 }, { "epoch": 0.02024950280238655, "grad_norm": 2.469858169555664, "learning_rate": 1.7533834586466166e-05, "loss": 0.22723723948001862, "num_input_tokens_seen": 702702, "step": 252, "train_runtime": 824.0307, "train_tokens_per_second": 852.762 }, { "epoch": 0.02032985797223729, "grad_norm": 3.3452255725860596, "learning_rate": 1.7523809523809526e-05, "loss": 0.30738964676856995, "num_input_tokens_seen": 704222, "step": 253, "train_runtime": 826.285, "train_tokens_per_second": 852.275 }, { "epoch": 0.02041021314208803, "grad_norm": 2.8986828327178955, "learning_rate": 1.7513784461152883e-05, "loss": 0.28926005959510803, "num_input_tokens_seen": 706280, "step": 254, "train_runtime": 828.6741, "train_tokens_per_second": 852.301 }, { "epoch": 0.02049056831193877, "grad_norm": 4.258272171020508, "learning_rate": 1.7503759398496243e-05, "loss": 0.3986973762512207, "num_input_tokens_seen": 712178, "step": 255, "train_runtime": 832.0657, "train_tokens_per_second": 855.916 }, { "epoch": 0.02057092348178951, "grad_norm": 3.135568380355835, "learning_rate": 1.74937343358396e-05, "loss": 0.2749190330505371, "num_input_tokens_seen": 714994, "step": 256, "train_runtime": 834.6531, "train_tokens_per_second": 856.636 }, { "epoch": 0.02065127865164025, "grad_norm": 2.858424186706543, "learning_rate": 1.7483709273182957e-05, "loss": 0.3634519577026367, "num_input_tokens_seen": 718366, "step": 257, "train_runtime": 837.3786, "train_tokens_per_second": 857.875 }, { "epoch": 0.02073163382149099, "grad_norm": 2.4050028324127197, "learning_rate": 1.7473684210526317e-05, "loss": 0.24798567593097687, "num_input_tokens_seen": 721334, "step": 258, "train_runtime": 839.9885, "train_tokens_per_second": 858.743 }, { "epoch": 0.02081198899134173, "grad_norm": 3.5194098949432373, "learning_rate": 1.7463659147869677e-05, "loss": 0.2981254756450653, "num_input_tokens_seen": 724154, "step": 259, "train_runtime": 842.5565, "train_tokens_per_second": 859.472 }, { "epoch": 0.02089234416119247, "grad_norm": 3.8050601482391357, "learning_rate": 1.7453634085213034e-05, "loss": 0.3709879219532013, "num_input_tokens_seen": 727218, "step": 260, "train_runtime": 845.2015, "train_tokens_per_second": 860.408 }, { "epoch": 0.02097269933104321, "grad_norm": 2.898082733154297, "learning_rate": 1.744360902255639e-05, "loss": 0.3443959355354309, "num_input_tokens_seen": 729916, "step": 261, "train_runtime": 847.8359, "train_tokens_per_second": 860.917 }, { "epoch": 0.021053054500893953, "grad_norm": 3.0824203491210938, "learning_rate": 1.743358395989975e-05, "loss": 0.34597283601760864, "num_input_tokens_seen": 732508, "step": 262, "train_runtime": 850.4438, "train_tokens_per_second": 861.324 }, { "epoch": 0.02113340967074469, "grad_norm": 2.7052230834960938, "learning_rate": 1.7423558897243108e-05, "loss": 0.27982208132743835, "num_input_tokens_seen": 738102, "step": 263, "train_runtime": 853.735, "train_tokens_per_second": 864.556 }, { "epoch": 0.021213764840595433, "grad_norm": 3.7319936752319336, "learning_rate": 1.7413533834586468e-05, "loss": 0.3050349950790405, "num_input_tokens_seen": 739610, "step": 264, "train_runtime": 856.0117, "train_tokens_per_second": 864.019 }, { "epoch": 0.02129412001044617, "grad_norm": 3.164203405380249, "learning_rate": 1.7403508771929828e-05, "loss": 0.2995794415473938, "num_input_tokens_seen": 741790, "step": 265, "train_runtime": 858.4511, "train_tokens_per_second": 864.103 }, { "epoch": 0.021374475180296913, "grad_norm": 2.902195930480957, "learning_rate": 1.7393483709273185e-05, "loss": 0.254988431930542, "num_input_tokens_seen": 744398, "step": 266, "train_runtime": 860.9472, "train_tokens_per_second": 864.627 }, { "epoch": 0.02145483035014765, "grad_norm": 3.0125560760498047, "learning_rate": 1.738345864661654e-05, "loss": 0.25865864753723145, "num_input_tokens_seen": 747032, "step": 267, "train_runtime": 863.4554, "train_tokens_per_second": 865.166 }, { "epoch": 0.021535185519998393, "grad_norm": 3.451200246810913, "learning_rate": 1.7373433583959902e-05, "loss": 0.39031851291656494, "num_input_tokens_seen": 750976, "step": 268, "train_runtime": 866.3019, "train_tokens_per_second": 866.876 }, { "epoch": 0.021615540689849132, "grad_norm": 2.7546122074127197, "learning_rate": 1.736340852130326e-05, "loss": 0.2288772016763687, "num_input_tokens_seen": 753784, "step": 269, "train_runtime": 868.8656, "train_tokens_per_second": 867.55 }, { "epoch": 0.021695895859699874, "grad_norm": 3.000035047531128, "learning_rate": 1.735338345864662e-05, "loss": 0.3425222337245941, "num_input_tokens_seen": 756954, "step": 270, "train_runtime": 871.635, "train_tokens_per_second": 868.43 }, { "epoch": 0.021776251029550615, "grad_norm": 3.3948066234588623, "learning_rate": 1.7343358395989975e-05, "loss": 0.3390633165836334, "num_input_tokens_seen": 761204, "step": 271, "train_runtime": 886.4254, "train_tokens_per_second": 858.734 }, { "epoch": 0.021856606199401354, "grad_norm": 3.184051990509033, "learning_rate": 1.7333333333333336e-05, "loss": 0.321198970079422, "num_input_tokens_seen": 767484, "step": 272, "train_runtime": 889.9592, "train_tokens_per_second": 862.381 }, { "epoch": 0.021936961369252096, "grad_norm": 3.067659854888916, "learning_rate": 1.7323308270676692e-05, "loss": 0.25460851192474365, "num_input_tokens_seen": 769322, "step": 273, "train_runtime": 892.2321, "train_tokens_per_second": 862.244 }, { "epoch": 0.022017316539102834, "grad_norm": 2.8137221336364746, "learning_rate": 1.731328320802005e-05, "loss": 0.2153228223323822, "num_input_tokens_seen": 772100, "step": 274, "train_runtime": 894.78, "train_tokens_per_second": 862.894 }, { "epoch": 0.022097671708953576, "grad_norm": 3.047585964202881, "learning_rate": 1.730325814536341e-05, "loss": 0.29086732864379883, "num_input_tokens_seen": 775254, "step": 275, "train_runtime": 897.4936, "train_tokens_per_second": 863.799 }, { "epoch": 0.022178026878804314, "grad_norm": 3.2107648849487305, "learning_rate": 1.729323308270677e-05, "loss": 0.2905319929122925, "num_input_tokens_seen": 777766, "step": 276, "train_runtime": 900.0303, "train_tokens_per_second": 864.155 }, { "epoch": 0.022258382048655056, "grad_norm": 2.7808125019073486, "learning_rate": 1.7283208020050126e-05, "loss": 0.2845841646194458, "num_input_tokens_seen": 780202, "step": 277, "train_runtime": 902.5051, "train_tokens_per_second": 864.485 }, { "epoch": 0.022338737218505794, "grad_norm": 3.3890421390533447, "learning_rate": 1.7273182957393487e-05, "loss": 0.23502470552921295, "num_input_tokens_seen": 784098, "step": 278, "train_runtime": 905.3171, "train_tokens_per_second": 866.103 }, { "epoch": 0.022419092388356536, "grad_norm": 3.118842601776123, "learning_rate": 1.7263157894736843e-05, "loss": 0.26093435287475586, "num_input_tokens_seen": 790412, "step": 279, "train_runtime": 908.7924, "train_tokens_per_second": 869.739 }, { "epoch": 0.022499447558207278, "grad_norm": 2.634300708770752, "learning_rate": 1.72531328320802e-05, "loss": 0.3072972893714905, "num_input_tokens_seen": 793478, "step": 280, "train_runtime": 911.4051, "train_tokens_per_second": 870.61 }, { "epoch": 0.022579802728058016, "grad_norm": 2.758129596710205, "learning_rate": 1.724310776942356e-05, "loss": 0.248707115650177, "num_input_tokens_seen": 796114, "step": 281, "train_runtime": 914.0077, "train_tokens_per_second": 871.015 }, { "epoch": 0.022660157897908758, "grad_norm": 2.539311170578003, "learning_rate": 1.723308270676692e-05, "loss": 0.20980846881866455, "num_input_tokens_seen": 799108, "step": 282, "train_runtime": 916.6778, "train_tokens_per_second": 871.744 }, { "epoch": 0.022740513067759496, "grad_norm": 3.1046602725982666, "learning_rate": 1.7223057644110277e-05, "loss": 0.2095586210489273, "num_input_tokens_seen": 801470, "step": 283, "train_runtime": 919.1514, "train_tokens_per_second": 871.967 }, { "epoch": 0.022820868237610238, "grad_norm": 3.1335947513580322, "learning_rate": 1.7213032581453634e-05, "loss": 0.2636254131793976, "num_input_tokens_seen": 804454, "step": 284, "train_runtime": 921.7545, "train_tokens_per_second": 872.742 }, { "epoch": 0.022901223407460976, "grad_norm": 3.6494688987731934, "learning_rate": 1.7203007518796994e-05, "loss": 0.2901442050933838, "num_input_tokens_seen": 806156, "step": 285, "train_runtime": 924.0395, "train_tokens_per_second": 872.426 }, { "epoch": 0.02298157857731172, "grad_norm": 3.281369924545288, "learning_rate": 1.719298245614035e-05, "loss": 0.3514842092990875, "num_input_tokens_seen": 808634, "step": 286, "train_runtime": 926.5161, "train_tokens_per_second": 872.768 }, { "epoch": 0.023061933747162457, "grad_norm": 3.012890338897705, "learning_rate": 1.718295739348371e-05, "loss": 0.2031923234462738, "num_input_tokens_seen": 811476, "step": 287, "train_runtime": 929.1011, "train_tokens_per_second": 873.399 }, { "epoch": 0.0231422889170132, "grad_norm": 3.85345458984375, "learning_rate": 1.717293233082707e-05, "loss": 0.38497740030288696, "num_input_tokens_seen": 813274, "step": 288, "train_runtime": 931.4354, "train_tokens_per_second": 873.14 }, { "epoch": 0.02322264408686394, "grad_norm": 3.0317018032073975, "learning_rate": 1.7162907268170428e-05, "loss": 0.33740293979644775, "num_input_tokens_seen": 817156, "step": 289, "train_runtime": 934.2607, "train_tokens_per_second": 874.655 }, { "epoch": 0.02330299925671468, "grad_norm": 3.4287424087524414, "learning_rate": 1.7152882205513785e-05, "loss": 0.33195459842681885, "num_input_tokens_seen": 820368, "step": 290, "train_runtime": 937.0076, "train_tokens_per_second": 875.519 }, { "epoch": 0.02338335442656542, "grad_norm": 2.3367555141448975, "learning_rate": 1.7142857142857142e-05, "loss": 0.23529896140098572, "num_input_tokens_seen": 823350, "step": 291, "train_runtime": 939.6072, "train_tokens_per_second": 876.27 }, { "epoch": 0.02346370959641616, "grad_norm": 3.0639004707336426, "learning_rate": 1.7132832080200502e-05, "loss": 0.27256661653518677, "num_input_tokens_seen": 825918, "step": 292, "train_runtime": 942.152, "train_tokens_per_second": 876.629 }, { "epoch": 0.0235440647662669, "grad_norm": 2.9229602813720703, "learning_rate": 1.7122807017543862e-05, "loss": 0.3626100718975067, "num_input_tokens_seen": 829014, "step": 293, "train_runtime": 944.7975, "train_tokens_per_second": 877.452 }, { "epoch": 0.02362441993611764, "grad_norm": 2.813892126083374, "learning_rate": 1.711278195488722e-05, "loss": 0.2940147817134857, "num_input_tokens_seen": 831504, "step": 294, "train_runtime": 947.2985, "train_tokens_per_second": 877.763 }, { "epoch": 0.02370477510596838, "grad_norm": 2.8105359077453613, "learning_rate": 1.710275689223058e-05, "loss": 0.28730159997940063, "num_input_tokens_seen": 836602, "step": 295, "train_runtime": 950.4975, "train_tokens_per_second": 880.173 }, { "epoch": 0.02378513027581912, "grad_norm": 2.8816328048706055, "learning_rate": 1.7092731829573936e-05, "loss": 0.24477729201316833, "num_input_tokens_seen": 838756, "step": 296, "train_runtime": 952.9149, "train_tokens_per_second": 880.2 }, { "epoch": 0.02386548544566986, "grad_norm": 3.1478993892669678, "learning_rate": 1.7082706766917293e-05, "loss": 0.3519008457660675, "num_input_tokens_seen": 843740, "step": 297, "train_runtime": 956.0578, "train_tokens_per_second": 882.52 }, { "epoch": 0.023945840615520603, "grad_norm": 4.049642086029053, "learning_rate": 1.7072681704260653e-05, "loss": 0.31622883677482605, "num_input_tokens_seen": 846110, "step": 298, "train_runtime": 958.5436, "train_tokens_per_second": 882.704 }, { "epoch": 0.02402619578537134, "grad_norm": 3.3289341926574707, "learning_rate": 1.7062656641604013e-05, "loss": 0.22939565777778625, "num_input_tokens_seen": 847716, "step": 299, "train_runtime": 960.8665, "train_tokens_per_second": 882.241 }, { "epoch": 0.024106550955222083, "grad_norm": 2.923302173614502, "learning_rate": 1.705263157894737e-05, "loss": 0.37166160345077515, "num_input_tokens_seen": 851378, "step": 300, "train_runtime": 963.6532, "train_tokens_per_second": 883.49 }, { "epoch": 0.02418690612507282, "grad_norm": 2.582261323928833, "learning_rate": 1.7042606516290727e-05, "loss": 0.21705582737922668, "num_input_tokens_seen": 853854, "step": 301, "train_runtime": 976.9171, "train_tokens_per_second": 874.029 }, { "epoch": 0.024267261294923563, "grad_norm": 2.98903751373291, "learning_rate": 1.7032581453634087e-05, "loss": 0.39324602484703064, "num_input_tokens_seen": 856270, "step": 302, "train_runtime": 979.4343, "train_tokens_per_second": 874.25 }, { "epoch": 0.0243476164647743, "grad_norm": 2.781104803085327, "learning_rate": 1.7022556390977444e-05, "loss": 0.267890065908432, "num_input_tokens_seen": 860152, "step": 303, "train_runtime": 982.2686, "train_tokens_per_second": 875.679 }, { "epoch": 0.024427971634625043, "grad_norm": 2.645282506942749, "learning_rate": 1.7012531328320804e-05, "loss": 0.3215838074684143, "num_input_tokens_seen": 864000, "step": 304, "train_runtime": 985.1893, "train_tokens_per_second": 876.989 }, { "epoch": 0.02450832680447578, "grad_norm": 3.2293035984039307, "learning_rate": 1.7002506265664164e-05, "loss": 0.33724308013916016, "num_input_tokens_seen": 867372, "step": 305, "train_runtime": 987.9444, "train_tokens_per_second": 877.956 }, { "epoch": 0.024588681974326523, "grad_norm": 3.1866393089294434, "learning_rate": 1.699248120300752e-05, "loss": 0.32881247997283936, "num_input_tokens_seen": 869930, "step": 306, "train_runtime": 990.5061, "train_tokens_per_second": 878.268 }, { "epoch": 0.024669037144177265, "grad_norm": 3.15861177444458, "learning_rate": 1.6982456140350878e-05, "loss": 0.27569305896759033, "num_input_tokens_seen": 872358, "step": 307, "train_runtime": 993.0025, "train_tokens_per_second": 878.505 }, { "epoch": 0.024749392314028004, "grad_norm": 3.1236774921417236, "learning_rate": 1.6972431077694238e-05, "loss": 0.2648042142391205, "num_input_tokens_seen": 874168, "step": 308, "train_runtime": 995.516, "train_tokens_per_second": 878.105 }, { "epoch": 0.024829747483878745, "grad_norm": 2.745558738708496, "learning_rate": 1.6962406015037595e-05, "loss": 0.2354494035243988, "num_input_tokens_seen": 876790, "step": 309, "train_runtime": 998.0567, "train_tokens_per_second": 878.497 }, { "epoch": 0.024910102653729484, "grad_norm": 3.4442496299743652, "learning_rate": 1.6952380952380955e-05, "loss": 0.3092072010040283, "num_input_tokens_seen": 879470, "step": 310, "train_runtime": 1000.7103, "train_tokens_per_second": 878.846 }, { "epoch": 0.024990457823580225, "grad_norm": 3.5796656608581543, "learning_rate": 1.694235588972431e-05, "loss": 0.2775275707244873, "num_input_tokens_seen": 882244, "step": 311, "train_runtime": 1003.2862, "train_tokens_per_second": 879.354 }, { "epoch": 0.025070812993430964, "grad_norm": 2.843031406402588, "learning_rate": 1.6932330827067672e-05, "loss": 0.3159998953342438, "num_input_tokens_seen": 885262, "step": 312, "train_runtime": 1005.9611, "train_tokens_per_second": 880.016 }, { "epoch": 0.025151168163281706, "grad_norm": 3.534191608428955, "learning_rate": 1.692230576441103e-05, "loss": 0.3665875196456909, "num_input_tokens_seen": 888904, "step": 313, "train_runtime": 1008.8226, "train_tokens_per_second": 881.13 }, { "epoch": 0.025231523333132444, "grad_norm": 2.691452980041504, "learning_rate": 1.6912280701754385e-05, "loss": 0.3234063386917114, "num_input_tokens_seen": 892142, "step": 314, "train_runtime": 1011.5749, "train_tokens_per_second": 881.934 }, { "epoch": 0.025311878502983186, "grad_norm": 2.5050652027130127, "learning_rate": 1.6902255639097746e-05, "loss": 0.33579960465431213, "num_input_tokens_seen": 895016, "step": 315, "train_runtime": 1014.2233, "train_tokens_per_second": 882.464 }, { "epoch": 0.025392233672833928, "grad_norm": 4.014756202697754, "learning_rate": 1.6892230576441106e-05, "loss": 0.34510987997055054, "num_input_tokens_seen": 896346, "step": 316, "train_runtime": 1016.4671, "train_tokens_per_second": 881.825 }, { "epoch": 0.025472588842684666, "grad_norm": 2.5267333984375, "learning_rate": 1.6882205513784463e-05, "loss": 0.2872266173362732, "num_input_tokens_seen": 899296, "step": 317, "train_runtime": 1019.1135, "train_tokens_per_second": 882.43 }, { "epoch": 0.025552944012535408, "grad_norm": 3.070948362350464, "learning_rate": 1.6872180451127823e-05, "loss": 0.3514803647994995, "num_input_tokens_seen": 903046, "step": 318, "train_runtime": 1021.9519, "train_tokens_per_second": 883.648 }, { "epoch": 0.025633299182386146, "grad_norm": 2.7682411670684814, "learning_rate": 1.686215538847118e-05, "loss": 0.24477803707122803, "num_input_tokens_seen": 905406, "step": 319, "train_runtime": 1024.4311, "train_tokens_per_second": 883.813 }, { "epoch": 0.025713654352236888, "grad_norm": 2.9864275455474854, "learning_rate": 1.6852130325814536e-05, "loss": 0.24607419967651367, "num_input_tokens_seen": 908384, "step": 320, "train_runtime": 1027.0829, "train_tokens_per_second": 884.431 }, { "epoch": 0.025794009522087626, "grad_norm": 2.711859941482544, "learning_rate": 1.6842105263157896e-05, "loss": 0.3553354740142822, "num_input_tokens_seen": 911408, "step": 321, "train_runtime": 1029.7358, "train_tokens_per_second": 885.089 }, { "epoch": 0.025874364691938368, "grad_norm": 3.8021342754364014, "learning_rate": 1.6832080200501257e-05, "loss": 0.29201799631118774, "num_input_tokens_seen": 913842, "step": 322, "train_runtime": 1032.2361, "train_tokens_per_second": 885.303 }, { "epoch": 0.025954719861789106, "grad_norm": 2.8324108123779297, "learning_rate": 1.6822055137844613e-05, "loss": 0.3195113241672516, "num_input_tokens_seen": 917942, "step": 323, "train_runtime": 1035.118, "train_tokens_per_second": 886.799 }, { "epoch": 0.026035075031639848, "grad_norm": 2.861069440841675, "learning_rate": 1.681203007518797e-05, "loss": 0.24255013465881348, "num_input_tokens_seen": 921238, "step": 324, "train_runtime": 1037.824, "train_tokens_per_second": 887.663 }, { "epoch": 0.02611543020149059, "grad_norm": 2.9005484580993652, "learning_rate": 1.680200501253133e-05, "loss": 0.3186454772949219, "num_input_tokens_seen": 923518, "step": 325, "train_runtime": 1040.2576, "train_tokens_per_second": 887.778 }, { "epoch": 0.02619578537134133, "grad_norm": 2.71109938621521, "learning_rate": 1.6791979949874687e-05, "loss": 0.2985735535621643, "num_input_tokens_seen": 927426, "step": 326, "train_runtime": 1043.113, "train_tokens_per_second": 889.094 }, { "epoch": 0.02627614054119207, "grad_norm": 3.0954089164733887, "learning_rate": 1.6781954887218047e-05, "loss": 0.28040140867233276, "num_input_tokens_seen": 930736, "step": 327, "train_runtime": 1045.779, "train_tokens_per_second": 889.993 }, { "epoch": 0.02635649571104281, "grad_norm": 3.4521098136901855, "learning_rate": 1.6771929824561408e-05, "loss": 0.30392998456954956, "num_input_tokens_seen": 932676, "step": 328, "train_runtime": 1048.1246, "train_tokens_per_second": 889.852 }, { "epoch": 0.02643685088089355, "grad_norm": 3.140446662902832, "learning_rate": 1.6761904761904764e-05, "loss": 0.24223175644874573, "num_input_tokens_seen": 935376, "step": 329, "train_runtime": 1050.6567, "train_tokens_per_second": 890.277 }, { "epoch": 0.02651720605074429, "grad_norm": 2.8953723907470703, "learning_rate": 1.675187969924812e-05, "loss": 0.3038066625595093, "num_input_tokens_seen": 938690, "step": 330, "train_runtime": 1053.3702, "train_tokens_per_second": 891.13 }, { "epoch": 0.02659756122059503, "grad_norm": 3.29001784324646, "learning_rate": 1.6741854636591478e-05, "loss": 0.31074488162994385, "num_input_tokens_seen": 941440, "step": 331, "train_runtime": 1067.6271, "train_tokens_per_second": 881.806 }, { "epoch": 0.02667791639044577, "grad_norm": 2.6703059673309326, "learning_rate": 1.6731829573934838e-05, "loss": 0.2324913889169693, "num_input_tokens_seen": 944542, "step": 332, "train_runtime": 1070.2959, "train_tokens_per_second": 882.505 }, { "epoch": 0.02675827156029651, "grad_norm": 3.0003550052642822, "learning_rate": 1.67218045112782e-05, "loss": 0.3308841586112976, "num_input_tokens_seen": 947156, "step": 333, "train_runtime": 1072.8362, "train_tokens_per_second": 882.852 }, { "epoch": 0.026838626730147253, "grad_norm": 6.185088157653809, "learning_rate": 1.6711779448621555e-05, "loss": 0.3639359474182129, "num_input_tokens_seen": 949046, "step": 334, "train_runtime": 1075.2751, "train_tokens_per_second": 882.608 }, { "epoch": 0.02691898189999799, "grad_norm": 3.0791001319885254, "learning_rate": 1.6701754385964915e-05, "loss": 0.273248553276062, "num_input_tokens_seen": 951832, "step": 335, "train_runtime": 1077.8641, "train_tokens_per_second": 883.072 }, { "epoch": 0.026999337069848733, "grad_norm": 3.233905076980591, "learning_rate": 1.6691729323308272e-05, "loss": 0.3049735426902771, "num_input_tokens_seen": 954018, "step": 336, "train_runtime": 1080.27, "train_tokens_per_second": 883.129 }, { "epoch": 0.02707969223969947, "grad_norm": 4.3563127517700195, "learning_rate": 1.668170426065163e-05, "loss": 0.28589046001434326, "num_input_tokens_seen": 955632, "step": 337, "train_runtime": 1082.5592, "train_tokens_per_second": 882.753 }, { "epoch": 0.027160047409550213, "grad_norm": 2.993605852127075, "learning_rate": 1.667167919799499e-05, "loss": 0.3632199764251709, "num_input_tokens_seen": 958300, "step": 338, "train_runtime": 1085.1099, "train_tokens_per_second": 883.136 }, { "epoch": 0.02724040257940095, "grad_norm": 2.8447954654693604, "learning_rate": 1.666165413533835e-05, "loss": 0.20722845196723938, "num_input_tokens_seen": 961438, "step": 339, "train_runtime": 1087.7723, "train_tokens_per_second": 883.86 }, { "epoch": 0.027320757749251693, "grad_norm": 2.540973424911499, "learning_rate": 1.6651629072681706e-05, "loss": 0.28336578607559204, "num_input_tokens_seen": 964414, "step": 340, "train_runtime": 1090.411, "train_tokens_per_second": 884.45 }, { "epoch": 0.02740111291910243, "grad_norm": 3.250901699066162, "learning_rate": 1.6641604010025063e-05, "loss": 0.2816706895828247, "num_input_tokens_seen": 966870, "step": 341, "train_runtime": 1092.9573, "train_tokens_per_second": 884.637 }, { "epoch": 0.027481468088953173, "grad_norm": 3.1629106998443604, "learning_rate": 1.6631578947368423e-05, "loss": 0.3279896080493927, "num_input_tokens_seen": 969186, "step": 342, "train_runtime": 1095.4555, "train_tokens_per_second": 884.733 }, { "epoch": 0.027561823258803915, "grad_norm": 2.678255081176758, "learning_rate": 1.662155388471178e-05, "loss": 0.23265647888183594, "num_input_tokens_seen": 971200, "step": 343, "train_runtime": 1097.8587, "train_tokens_per_second": 884.631 }, { "epoch": 0.027642178428654653, "grad_norm": 2.325950860977173, "learning_rate": 1.661152882205514e-05, "loss": 0.3469184637069702, "num_input_tokens_seen": 975378, "step": 344, "train_runtime": 1100.7671, "train_tokens_per_second": 886.089 }, { "epoch": 0.027722533598505395, "grad_norm": 2.173321008682251, "learning_rate": 1.6601503759398497e-05, "loss": 0.23510408401489258, "num_input_tokens_seen": 980582, "step": 345, "train_runtime": 1103.9606, "train_tokens_per_second": 888.24 }, { "epoch": 0.027802888768356133, "grad_norm": 3.184866428375244, "learning_rate": 1.6591478696741857e-05, "loss": 0.2502746284008026, "num_input_tokens_seen": 983146, "step": 346, "train_runtime": 1106.4713, "train_tokens_per_second": 888.542 }, { "epoch": 0.027883243938206875, "grad_norm": 2.9400253295898438, "learning_rate": 1.6581453634085214e-05, "loss": 0.26074936985969543, "num_input_tokens_seen": 985198, "step": 347, "train_runtime": 1108.8553, "train_tokens_per_second": 888.482 }, { "epoch": 0.027963599108057614, "grad_norm": 2.9507036209106445, "learning_rate": 1.6571428571428574e-05, "loss": 0.2959950268268585, "num_input_tokens_seen": 987652, "step": 348, "train_runtime": 1111.324, "train_tokens_per_second": 888.717 }, { "epoch": 0.028043954277908355, "grad_norm": 3.6237893104553223, "learning_rate": 1.656140350877193e-05, "loss": 0.33771026134490967, "num_input_tokens_seen": 989016, "step": 349, "train_runtime": 1113.5401, "train_tokens_per_second": 888.173 }, { "epoch": 0.028124309447759094, "grad_norm": 2.9807822704315186, "learning_rate": 1.655137844611529e-05, "loss": 0.34814852476119995, "num_input_tokens_seen": 991946, "step": 350, "train_runtime": 1116.1418, "train_tokens_per_second": 888.728 }, { "epoch": 0.028204664617609836, "grad_norm": 2.927029609680176, "learning_rate": 1.6541353383458648e-05, "loss": 0.25223565101623535, "num_input_tokens_seen": 994280, "step": 351, "train_runtime": 1118.5665, "train_tokens_per_second": 888.888 }, { "epoch": 0.028285019787460577, "grad_norm": 3.623434543609619, "learning_rate": 1.6531328320802008e-05, "loss": 0.28349393606185913, "num_input_tokens_seen": 996028, "step": 352, "train_runtime": 1120.87, "train_tokens_per_second": 888.62 }, { "epoch": 0.028365374957311316, "grad_norm": 2.5638298988342285, "learning_rate": 1.6521303258145365e-05, "loss": 0.19310274720191956, "num_input_tokens_seen": 997816, "step": 353, "train_runtime": 1123.2326, "train_tokens_per_second": 888.343 }, { "epoch": 0.028445730127162058, "grad_norm": 3.2650277614593506, "learning_rate": 1.651127819548872e-05, "loss": 0.20673713088035583, "num_input_tokens_seen": 1000598, "step": 354, "train_runtime": 1125.8571, "train_tokens_per_second": 888.743 }, { "epoch": 0.028526085297012796, "grad_norm": 3.129873752593994, "learning_rate": 1.650125313283208e-05, "loss": 0.30608096718788147, "num_input_tokens_seen": 1002282, "step": 355, "train_runtime": 1128.165, "train_tokens_per_second": 888.418 }, { "epoch": 0.028606440466863538, "grad_norm": 2.393281936645508, "learning_rate": 1.649122807017544e-05, "loss": 0.22457966208457947, "num_input_tokens_seen": 1004498, "step": 356, "train_runtime": 1130.6008, "train_tokens_per_second": 888.464 }, { "epoch": 0.028686795636714276, "grad_norm": 2.8952462673187256, "learning_rate": 1.64812030075188e-05, "loss": 0.3065701723098755, "num_input_tokens_seen": 1009500, "step": 357, "train_runtime": 1133.715, "train_tokens_per_second": 890.435 }, { "epoch": 0.028767150806565018, "grad_norm": 3.095759153366089, "learning_rate": 1.647117794486216e-05, "loss": 0.2602217197418213, "num_input_tokens_seen": 1012560, "step": 358, "train_runtime": 1136.3622, "train_tokens_per_second": 891.054 }, { "epoch": 0.028847505976415756, "grad_norm": 3.529064416885376, "learning_rate": 1.6461152882205516e-05, "loss": 0.28875595331192017, "num_input_tokens_seen": 1015842, "step": 359, "train_runtime": 1139.0618, "train_tokens_per_second": 891.823 }, { "epoch": 0.028927861146266498, "grad_norm": 3.2159552574157715, "learning_rate": 1.6451127819548872e-05, "loss": 0.2264186143875122, "num_input_tokens_seen": 1018030, "step": 360, "train_runtime": 1141.5233, "train_tokens_per_second": 891.817 }, { "epoch": 0.02900821631611724, "grad_norm": 2.9752864837646484, "learning_rate": 1.6441102756892233e-05, "loss": 0.30945977568626404, "num_input_tokens_seen": 1020850, "step": 361, "train_runtime": 1156.8762, "train_tokens_per_second": 882.419 }, { "epoch": 0.029088571485967978, "grad_norm": 3.8603475093841553, "learning_rate": 1.643107769423559e-05, "loss": 0.30086299777030945, "num_input_tokens_seen": 1023098, "step": 362, "train_runtime": 1159.545, "train_tokens_per_second": 882.327 }, { "epoch": 0.02916892665581872, "grad_norm": 3.185044765472412, "learning_rate": 1.642105263157895e-05, "loss": 0.3729586601257324, "num_input_tokens_seen": 1025558, "step": 363, "train_runtime": 1162.1314, "train_tokens_per_second": 882.48 }, { "epoch": 0.02924928182566946, "grad_norm": 2.8934638500213623, "learning_rate": 1.6411027568922306e-05, "loss": 0.2639715373516083, "num_input_tokens_seen": 1028730, "step": 364, "train_runtime": 1164.9276, "train_tokens_per_second": 883.085 }, { "epoch": 0.0293296369955202, "grad_norm": 2.467963218688965, "learning_rate": 1.6401002506265667e-05, "loss": 0.2426169514656067, "num_input_tokens_seen": 1031678, "step": 365, "train_runtime": 1167.5847, "train_tokens_per_second": 883.6 }, { "epoch": 0.02940999216537094, "grad_norm": 3.7559967041015625, "learning_rate": 1.6390977443609023e-05, "loss": 0.34257516264915466, "num_input_tokens_seen": 1033998, "step": 366, "train_runtime": 1170.1374, "train_tokens_per_second": 883.655 }, { "epoch": 0.02949034733522168, "grad_norm": 3.033872365951538, "learning_rate": 1.6380952380952384e-05, "loss": 0.30197417736053467, "num_input_tokens_seen": 1036224, "step": 367, "train_runtime": 1172.6127, "train_tokens_per_second": 883.688 }, { "epoch": 0.02957070250507242, "grad_norm": 2.9688868522644043, "learning_rate": 1.637092731829574e-05, "loss": 0.26695260405540466, "num_input_tokens_seen": 1039486, "step": 368, "train_runtime": 1175.3231, "train_tokens_per_second": 884.426 }, { "epoch": 0.02965105767492316, "grad_norm": 3.4373786449432373, "learning_rate": 1.63609022556391e-05, "loss": 0.35735583305358887, "num_input_tokens_seen": 1042834, "step": 369, "train_runtime": 1178.1171, "train_tokens_per_second": 885.17 }, { "epoch": 0.029731412844773902, "grad_norm": 3.433124303817749, "learning_rate": 1.6350877192982457e-05, "loss": 0.3222700357437134, "num_input_tokens_seen": 1045598, "step": 370, "train_runtime": 1180.7689, "train_tokens_per_second": 885.523 }, { "epoch": 0.02981176801462464, "grad_norm": 3.435239315032959, "learning_rate": 1.6340852130325814e-05, "loss": 0.3286718428134918, "num_input_tokens_seen": 1049668, "step": 371, "train_runtime": 1183.6973, "train_tokens_per_second": 886.771 }, { "epoch": 0.029892123184475382, "grad_norm": 3.0817859172821045, "learning_rate": 1.6330827067669174e-05, "loss": 0.35218101739883423, "num_input_tokens_seen": 1053414, "step": 372, "train_runtime": 1186.5059, "train_tokens_per_second": 887.829 }, { "epoch": 0.02997247835432612, "grad_norm": 2.913379669189453, "learning_rate": 1.632080200501253e-05, "loss": 0.40495985746383667, "num_input_tokens_seen": 1056336, "step": 373, "train_runtime": 1189.1127, "train_tokens_per_second": 888.34 }, { "epoch": 0.030052833524176863, "grad_norm": 2.900643825531006, "learning_rate": 1.631077694235589e-05, "loss": 0.288735568523407, "num_input_tokens_seen": 1058250, "step": 374, "train_runtime": 1191.4232, "train_tokens_per_second": 888.223 }, { "epoch": 0.0301331886940276, "grad_norm": 2.9756662845611572, "learning_rate": 1.630075187969925e-05, "loss": 0.3011665642261505, "num_input_tokens_seen": 1060628, "step": 375, "train_runtime": 1193.9105, "train_tokens_per_second": 888.365 }, { "epoch": 0.030213543863878343, "grad_norm": 2.9925317764282227, "learning_rate": 1.6290726817042608e-05, "loss": 0.2644414007663727, "num_input_tokens_seen": 1063338, "step": 376, "train_runtime": 1196.5112, "train_tokens_per_second": 888.699 }, { "epoch": 0.03029389903372908, "grad_norm": 2.605008363723755, "learning_rate": 1.6280701754385965e-05, "loss": 0.24983012676239014, "num_input_tokens_seen": 1065384, "step": 377, "train_runtime": 1198.9066, "train_tokens_per_second": 888.63 }, { "epoch": 0.030374254203579823, "grad_norm": 3.3709828853607178, "learning_rate": 1.6270676691729325e-05, "loss": 0.339907705783844, "num_input_tokens_seen": 1067650, "step": 378, "train_runtime": 1201.3796, "train_tokens_per_second": 888.687 }, { "epoch": 0.030454609373430565, "grad_norm": 2.9063334465026855, "learning_rate": 1.6260651629072682e-05, "loss": 0.32940688729286194, "num_input_tokens_seen": 1070284, "step": 379, "train_runtime": 1203.9259, "train_tokens_per_second": 888.995 }, { "epoch": 0.030534964543281303, "grad_norm": 4.010578632354736, "learning_rate": 1.6250626566416042e-05, "loss": 0.3015247583389282, "num_input_tokens_seen": 1072566, "step": 380, "train_runtime": 1206.4041, "train_tokens_per_second": 889.06 }, { "epoch": 0.030615319713132045, "grad_norm": 2.9312222003936768, "learning_rate": 1.62406015037594e-05, "loss": 0.2930034399032593, "num_input_tokens_seen": 1074288, "step": 381, "train_runtime": 1208.7595, "train_tokens_per_second": 888.752 }, { "epoch": 0.030695674882982783, "grad_norm": 3.3023061752319336, "learning_rate": 1.623057644110276e-05, "loss": 0.35004299879074097, "num_input_tokens_seen": 1077372, "step": 382, "train_runtime": 1211.4174, "train_tokens_per_second": 889.348 }, { "epoch": 0.030776030052833525, "grad_norm": 3.05501651763916, "learning_rate": 1.6220551378446116e-05, "loss": 0.21290528774261475, "num_input_tokens_seen": 1079672, "step": 383, "train_runtime": 1213.8244, "train_tokens_per_second": 889.48 }, { "epoch": 0.030856385222684263, "grad_norm": 3.3656420707702637, "learning_rate": 1.6210526315789473e-05, "loss": 0.3058585524559021, "num_input_tokens_seen": 1081562, "step": 384, "train_runtime": 1216.1728, "train_tokens_per_second": 889.316 }, { "epoch": 0.030936740392535005, "grad_norm": 3.4209678173065186, "learning_rate": 1.6200501253132833e-05, "loss": 0.30159473419189453, "num_input_tokens_seen": 1084548, "step": 385, "train_runtime": 1218.7646, "train_tokens_per_second": 889.875 }, { "epoch": 0.031017095562385744, "grad_norm": 2.5945634841918945, "learning_rate": 1.6190476190476193e-05, "loss": 0.3786466419696808, "num_input_tokens_seen": 1088040, "step": 386, "train_runtime": 1221.4839, "train_tokens_per_second": 890.753 }, { "epoch": 0.031097450732236485, "grad_norm": 2.4741642475128174, "learning_rate": 1.618045112781955e-05, "loss": 0.20992669463157654, "num_input_tokens_seen": 1091676, "step": 387, "train_runtime": 1224.2507, "train_tokens_per_second": 891.71 }, { "epoch": 0.031177805902087227, "grad_norm": 2.642354726791382, "learning_rate": 1.617042606516291e-05, "loss": 0.21321657299995422, "num_input_tokens_seen": 1094354, "step": 388, "train_runtime": 1226.7867, "train_tokens_per_second": 892.049 }, { "epoch": 0.03125816107193797, "grad_norm": 2.4200429916381836, "learning_rate": 1.6160401002506267e-05, "loss": 0.15720219910144806, "num_input_tokens_seen": 1096008, "step": 389, "train_runtime": 1229.065, "train_tokens_per_second": 891.741 }, { "epoch": 0.031338516241788704, "grad_norm": 2.474097967147827, "learning_rate": 1.6150375939849624e-05, "loss": 0.29118612408638, "num_input_tokens_seen": 1099258, "step": 390, "train_runtime": 1231.8336, "train_tokens_per_second": 892.375 }, { "epoch": 0.031418871411639446, "grad_norm": 2.931605100631714, "learning_rate": 1.6140350877192984e-05, "loss": 0.26404663920402527, "num_input_tokens_seen": 1103194, "step": 391, "train_runtime": 1246.539, "train_tokens_per_second": 885.006 }, { "epoch": 0.03149922658149019, "grad_norm": 3.0591230392456055, "learning_rate": 1.6130325814536344e-05, "loss": 0.26877331733703613, "num_input_tokens_seen": 1107754, "step": 392, "train_runtime": 1249.6283, "train_tokens_per_second": 886.467 }, { "epoch": 0.03157958175134093, "grad_norm": 3.190993547439575, "learning_rate": 1.61203007518797e-05, "loss": 0.29649144411087036, "num_input_tokens_seen": 1109528, "step": 393, "train_runtime": 1251.934, "train_tokens_per_second": 886.251 }, { "epoch": 0.031659936921191664, "grad_norm": 3.8178114891052246, "learning_rate": 1.6110275689223058e-05, "loss": 0.3399081826210022, "num_input_tokens_seen": 1112572, "step": 394, "train_runtime": 1254.5895, "train_tokens_per_second": 886.802 }, { "epoch": 0.031740292091042406, "grad_norm": 3.2926313877105713, "learning_rate": 1.6100250626566418e-05, "loss": 0.3436011075973511, "num_input_tokens_seen": 1114518, "step": 395, "train_runtime": 1256.975, "train_tokens_per_second": 886.667 }, { "epoch": 0.03182064726089315, "grad_norm": 2.9628255367279053, "learning_rate": 1.6090225563909775e-05, "loss": 0.29956409335136414, "num_input_tokens_seen": 1120438, "step": 396, "train_runtime": 1260.4109, "train_tokens_per_second": 888.947 }, { "epoch": 0.03190100243074389, "grad_norm": 3.5722475051879883, "learning_rate": 1.6080200501253135e-05, "loss": 0.2980824410915375, "num_input_tokens_seen": 1123360, "step": 397, "train_runtime": 1262.9879, "train_tokens_per_second": 889.446 }, { "epoch": 0.03198135760059463, "grad_norm": 3.1230993270874023, "learning_rate": 1.6070175438596495e-05, "loss": 0.19605785608291626, "num_input_tokens_seen": 1126374, "step": 398, "train_runtime": 1265.6737, "train_tokens_per_second": 889.94 }, { "epoch": 0.032061712770445366, "grad_norm": 3.4061503410339355, "learning_rate": 1.6060150375939852e-05, "loss": 0.2907423973083496, "num_input_tokens_seen": 1128926, "step": 399, "train_runtime": 1268.2161, "train_tokens_per_second": 890.169 }, { "epoch": 0.03214206794029611, "grad_norm": 3.7919397354125977, "learning_rate": 1.605012531328321e-05, "loss": 0.3025698661804199, "num_input_tokens_seen": 1130724, "step": 400, "train_runtime": 1270.5494, "train_tokens_per_second": 889.949 }, { "epoch": 0.03214206794029611, "eval_loss": 1.1756271123886108, "eval_runtime": 69.0182, "eval_samples_per_second": 14.402, "eval_steps_per_second": 7.201, "num_input_tokens_seen": 1130724, "step": 400 }, { "epoch": 0.03222242311014685, "grad_norm": 3.4964044094085693, "learning_rate": 1.6040100250626565e-05, "loss": 0.34112071990966797, "num_input_tokens_seen": 1139560, "step": 401, "train_runtime": 1343.6787, "train_tokens_per_second": 848.09 }, { "epoch": 0.03230277827999759, "grad_norm": 2.734004259109497, "learning_rate": 1.6030075187969926e-05, "loss": 0.28023210167884827, "num_input_tokens_seen": 1142480, "step": 402, "train_runtime": 1346.3137, "train_tokens_per_second": 848.599 }, { "epoch": 0.03238313344984833, "grad_norm": 3.584583044052124, "learning_rate": 1.6020050125313286e-05, "loss": 0.33350223302841187, "num_input_tokens_seen": 1145048, "step": 403, "train_runtime": 1348.8506, "train_tokens_per_second": 848.906 }, { "epoch": 0.03246348861969907, "grad_norm": 2.4744067192077637, "learning_rate": 1.6010025062656642e-05, "loss": 0.19812653958797455, "num_input_tokens_seen": 1147356, "step": 404, "train_runtime": 1351.3068, "train_tokens_per_second": 849.071 }, { "epoch": 0.03254384378954981, "grad_norm": 2.9946377277374268, "learning_rate": 1.6000000000000003e-05, "loss": 0.3582034707069397, "num_input_tokens_seen": 1149730, "step": 405, "train_runtime": 1353.8127, "train_tokens_per_second": 849.253 }, { "epoch": 0.03262419895940055, "grad_norm": 3.0742878913879395, "learning_rate": 1.598997493734336e-05, "loss": 0.25629955530166626, "num_input_tokens_seen": 1151264, "step": 406, "train_runtime": 1356.1159, "train_tokens_per_second": 848.942 }, { "epoch": 0.032704554129251294, "grad_norm": 3.0237274169921875, "learning_rate": 1.5979949874686716e-05, "loss": 0.2870306670665741, "num_input_tokens_seen": 1153498, "step": 407, "train_runtime": 1358.5659, "train_tokens_per_second": 849.056 }, { "epoch": 0.03278490929910203, "grad_norm": 3.2956907749176025, "learning_rate": 1.5969924812030076e-05, "loss": 0.27075907588005066, "num_input_tokens_seen": 1155738, "step": 408, "train_runtime": 1361.0574, "train_tokens_per_second": 849.147 }, { "epoch": 0.03286526446895277, "grad_norm": 3.0270862579345703, "learning_rate": 1.5959899749373437e-05, "loss": 0.2017335742712021, "num_input_tokens_seen": 1158142, "step": 409, "train_runtime": 1363.6162, "train_tokens_per_second": 849.317 }, { "epoch": 0.03294561963880351, "grad_norm": 2.7161452770233154, "learning_rate": 1.5949874686716793e-05, "loss": 0.24648171663284302, "num_input_tokens_seen": 1162016, "step": 410, "train_runtime": 1366.4535, "train_tokens_per_second": 850.388 }, { "epoch": 0.033025974808654254, "grad_norm": 3.4280121326446533, "learning_rate": 1.593984962406015e-05, "loss": 0.39886415004730225, "num_input_tokens_seen": 1163854, "step": 411, "train_runtime": 1368.7893, "train_tokens_per_second": 850.28 }, { "epoch": 0.03310632997850499, "grad_norm": 2.863529682159424, "learning_rate": 1.592982456140351e-05, "loss": 0.27319756150245667, "num_input_tokens_seen": 1166486, "step": 412, "train_runtime": 1371.3366, "train_tokens_per_second": 850.62 }, { "epoch": 0.03318668514835573, "grad_norm": 3.4037463665008545, "learning_rate": 1.5919799498746867e-05, "loss": 0.30301928520202637, "num_input_tokens_seen": 1169350, "step": 413, "train_runtime": 1373.9179, "train_tokens_per_second": 851.106 }, { "epoch": 0.03326704031820647, "grad_norm": 3.3865318298339844, "learning_rate": 1.5909774436090227e-05, "loss": 0.2560475766658783, "num_input_tokens_seen": 1171968, "step": 414, "train_runtime": 1376.4251, "train_tokens_per_second": 851.458 }, { "epoch": 0.033347395488057215, "grad_norm": 3.1275112628936768, "learning_rate": 1.5899749373433588e-05, "loss": 0.25025707483291626, "num_input_tokens_seen": 1175492, "step": 415, "train_runtime": 1379.1942, "train_tokens_per_second": 852.303 }, { "epoch": 0.033427750657907956, "grad_norm": 3.040975570678711, "learning_rate": 1.5889724310776944e-05, "loss": 0.3704090118408203, "num_input_tokens_seen": 1179274, "step": 416, "train_runtime": 1382.0127, "train_tokens_per_second": 853.302 }, { "epoch": 0.03350810582775869, "grad_norm": 3.1968061923980713, "learning_rate": 1.58796992481203e-05, "loss": 0.32380276918411255, "num_input_tokens_seen": 1182732, "step": 417, "train_runtime": 1384.7479, "train_tokens_per_second": 854.114 }, { "epoch": 0.03358846099760943, "grad_norm": 2.7388741970062256, "learning_rate": 1.5869674185463658e-05, "loss": 0.2912801504135132, "num_input_tokens_seen": 1184842, "step": 418, "train_runtime": 1387.1616, "train_tokens_per_second": 854.149 }, { "epoch": 0.033668816167460175, "grad_norm": 3.220170497894287, "learning_rate": 1.5859649122807018e-05, "loss": 0.27821680903434753, "num_input_tokens_seen": 1188874, "step": 419, "train_runtime": 1390.0272, "train_tokens_per_second": 855.288 }, { "epoch": 0.03374917133731092, "grad_norm": 2.976245164871216, "learning_rate": 1.5849624060150378e-05, "loss": 0.2863772213459015, "num_input_tokens_seen": 1194400, "step": 420, "train_runtime": 1393.2606, "train_tokens_per_second": 857.27 }, { "epoch": 0.03382952650716165, "grad_norm": 3.8372294902801514, "learning_rate": 1.5839598997493735e-05, "loss": 0.3358113169670105, "num_input_tokens_seen": 1197974, "step": 421, "train_runtime": 1407.8635, "train_tokens_per_second": 850.916 }, { "epoch": 0.03390988167701239, "grad_norm": 2.8535056114196777, "learning_rate": 1.5829573934837095e-05, "loss": 0.23794673383235931, "num_input_tokens_seen": 1203570, "step": 422, "train_runtime": 1411.2327, "train_tokens_per_second": 852.85 }, { "epoch": 0.033990236846863135, "grad_norm": 3.9697251319885254, "learning_rate": 1.5819548872180452e-05, "loss": 0.2616580128669739, "num_input_tokens_seen": 1206490, "step": 423, "train_runtime": 1413.9145, "train_tokens_per_second": 853.298 }, { "epoch": 0.03407059201671388, "grad_norm": 2.55678391456604, "learning_rate": 1.580952380952381e-05, "loss": 0.2917073965072632, "num_input_tokens_seen": 1211792, "step": 424, "train_runtime": 1417.2321, "train_tokens_per_second": 855.041 }, { "epoch": 0.03415094718656462, "grad_norm": 2.956308364868164, "learning_rate": 1.579949874686717e-05, "loss": 0.30434995889663696, "num_input_tokens_seen": 1213306, "step": 425, "train_runtime": 1419.4925, "train_tokens_per_second": 854.746 }, { "epoch": 0.034231302356415354, "grad_norm": 2.2293848991394043, "learning_rate": 1.578947368421053e-05, "loss": 0.17356255650520325, "num_input_tokens_seen": 1216020, "step": 426, "train_runtime": 1422.0677, "train_tokens_per_second": 855.107 }, { "epoch": 0.034311657526266096, "grad_norm": 2.773193359375, "learning_rate": 1.5779448621553886e-05, "loss": 0.3809736669063568, "num_input_tokens_seen": 1220280, "step": 427, "train_runtime": 1425.0658, "train_tokens_per_second": 856.297 }, { "epoch": 0.03439201269611684, "grad_norm": 2.6261844635009766, "learning_rate": 1.5769423558897243e-05, "loss": 0.3115207552909851, "num_input_tokens_seen": 1225658, "step": 428, "train_runtime": 1428.3328, "train_tokens_per_second": 858.104 }, { "epoch": 0.03447236786596758, "grad_norm": 2.807551145553589, "learning_rate": 1.5759398496240603e-05, "loss": 0.2823522090911865, "num_input_tokens_seen": 1230038, "step": 429, "train_runtime": 1431.3013, "train_tokens_per_second": 859.384 }, { "epoch": 0.034552723035818314, "grad_norm": 3.6646246910095215, "learning_rate": 1.574937343358396e-05, "loss": 0.27005884051322937, "num_input_tokens_seen": 1233886, "step": 430, "train_runtime": 1434.2505, "train_tokens_per_second": 860.3 }, { "epoch": 0.034633078205669056, "grad_norm": 3.1736390590667725, "learning_rate": 1.573934837092732e-05, "loss": 0.30805572867393494, "num_input_tokens_seen": 1236222, "step": 431, "train_runtime": 1436.8653, "train_tokens_per_second": 860.36 }, { "epoch": 0.0347134333755198, "grad_norm": 2.969747304916382, "learning_rate": 1.572932330827068e-05, "loss": 0.2623918950557709, "num_input_tokens_seen": 1237768, "step": 432, "train_runtime": 1439.1174, "train_tokens_per_second": 860.088 }, { "epoch": 0.03479378854537054, "grad_norm": 3.017998218536377, "learning_rate": 1.5719298245614037e-05, "loss": 0.26893916726112366, "num_input_tokens_seen": 1240322, "step": 433, "train_runtime": 1441.61, "train_tokens_per_second": 860.373 }, { "epoch": 0.03487414371522128, "grad_norm": 2.760850667953491, "learning_rate": 1.5709273182957394e-05, "loss": 0.29734647274017334, "num_input_tokens_seen": 1243428, "step": 434, "train_runtime": 1444.2677, "train_tokens_per_second": 860.94 }, { "epoch": 0.034954498885072016, "grad_norm": 3.389772415161133, "learning_rate": 1.5699248120300754e-05, "loss": 0.3615173101425171, "num_input_tokens_seen": 1245808, "step": 435, "train_runtime": 1446.7813, "train_tokens_per_second": 861.089 }, { "epoch": 0.03503485405492276, "grad_norm": 3.183323621749878, "learning_rate": 1.568922305764411e-05, "loss": 0.23475828766822815, "num_input_tokens_seen": 1247390, "step": 436, "train_runtime": 1449.0462, "train_tokens_per_second": 860.835 }, { "epoch": 0.0351152092247735, "grad_norm": 2.748947858810425, "learning_rate": 1.567919799498747e-05, "loss": 0.3026242256164551, "num_input_tokens_seen": 1249876, "step": 437, "train_runtime": 1451.592, "train_tokens_per_second": 861.038 }, { "epoch": 0.03519556439462424, "grad_norm": 2.938328266143799, "learning_rate": 1.5669172932330828e-05, "loss": 0.2658514976501465, "num_input_tokens_seen": 1251766, "step": 438, "train_runtime": 1454.0322, "train_tokens_per_second": 860.893 }, { "epoch": 0.035275919564474977, "grad_norm": 3.5145623683929443, "learning_rate": 1.5659147869674188e-05, "loss": 0.29393747448921204, "num_input_tokens_seen": 1254856, "step": 439, "train_runtime": 1456.7915, "train_tokens_per_second": 861.383 }, { "epoch": 0.03535627473432572, "grad_norm": 2.912855625152588, "learning_rate": 1.5649122807017545e-05, "loss": 0.3583008050918579, "num_input_tokens_seen": 1257128, "step": 440, "train_runtime": 1459.2702, "train_tokens_per_second": 861.477 }, { "epoch": 0.03543662990417646, "grad_norm": 2.911714792251587, "learning_rate": 1.56390977443609e-05, "loss": 0.2746956944465637, "num_input_tokens_seen": 1259556, "step": 441, "train_runtime": 1461.8157, "train_tokens_per_second": 861.638 }, { "epoch": 0.0355169850740272, "grad_norm": 3.5647177696228027, "learning_rate": 1.562907268170426e-05, "loss": 0.3998049795627594, "num_input_tokens_seen": 1261792, "step": 442, "train_runtime": 1464.2535, "train_tokens_per_second": 861.731 }, { "epoch": 0.035597340243877944, "grad_norm": 3.0395331382751465, "learning_rate": 1.5619047619047622e-05, "loss": 0.30140477418899536, "num_input_tokens_seen": 1264812, "step": 443, "train_runtime": 1466.9131, "train_tokens_per_second": 862.227 }, { "epoch": 0.03567769541372868, "grad_norm": 2.811570644378662, "learning_rate": 1.560902255639098e-05, "loss": 0.2283719778060913, "num_input_tokens_seen": 1266946, "step": 444, "train_runtime": 1469.3483, "train_tokens_per_second": 862.25 }, { "epoch": 0.03575805058357942, "grad_norm": 2.806750774383545, "learning_rate": 1.559899749373434e-05, "loss": 0.24863821268081665, "num_input_tokens_seen": 1269520, "step": 445, "train_runtime": 1471.9917, "train_tokens_per_second": 862.451 }, { "epoch": 0.03583840575343016, "grad_norm": 2.771272897720337, "learning_rate": 1.5588972431077696e-05, "loss": 0.28894513845443726, "num_input_tokens_seen": 1273334, "step": 446, "train_runtime": 1474.9239, "train_tokens_per_second": 863.322 }, { "epoch": 0.035918760923280904, "grad_norm": 3.634634256362915, "learning_rate": 1.5578947368421052e-05, "loss": 0.3700304627418518, "num_input_tokens_seen": 1276172, "step": 447, "train_runtime": 1477.5447, "train_tokens_per_second": 863.711 }, { "epoch": 0.03599911609313164, "grad_norm": 2.821510076522827, "learning_rate": 1.5568922305764413e-05, "loss": 0.3684689402580261, "num_input_tokens_seen": 1279344, "step": 448, "train_runtime": 1480.2713, "train_tokens_per_second": 864.263 }, { "epoch": 0.03607947126298238, "grad_norm": 2.6971709728240967, "learning_rate": 1.5558897243107773e-05, "loss": 0.2639304995536804, "num_input_tokens_seen": 1281576, "step": 449, "train_runtime": 1482.788, "train_tokens_per_second": 864.302 }, { "epoch": 0.03615982643283312, "grad_norm": 3.385385513305664, "learning_rate": 1.554887218045113e-05, "loss": 0.3571033477783203, "num_input_tokens_seen": 1284058, "step": 450, "train_runtime": 1485.3285, "train_tokens_per_second": 864.494 }, { "epoch": 0.036240181602683864, "grad_norm": 2.6196072101593018, "learning_rate": 1.5538847117794486e-05, "loss": 0.2818796634674072, "num_input_tokens_seen": 1286946, "step": 451, "train_runtime": 1499.7291, "train_tokens_per_second": 858.119 }, { "epoch": 0.036320536772534606, "grad_norm": 3.5355074405670166, "learning_rate": 1.5528822055137847e-05, "loss": 0.29990047216415405, "num_input_tokens_seen": 1289254, "step": 452, "train_runtime": 1502.2247, "train_tokens_per_second": 858.23 }, { "epoch": 0.03640089194238534, "grad_norm": 2.6526682376861572, "learning_rate": 1.5518796992481203e-05, "loss": 0.26082050800323486, "num_input_tokens_seen": 1292260, "step": 453, "train_runtime": 1504.9556, "train_tokens_per_second": 858.67 }, { "epoch": 0.03648124711223608, "grad_norm": 3.1149792671203613, "learning_rate": 1.5508771929824563e-05, "loss": 0.26639389991760254, "num_input_tokens_seen": 1293784, "step": 454, "train_runtime": 1507.2211, "train_tokens_per_second": 858.39 }, { "epoch": 0.036561602282086825, "grad_norm": 3.541278839111328, "learning_rate": 1.5498746867167924e-05, "loss": 0.2926393151283264, "num_input_tokens_seen": 1295748, "step": 455, "train_runtime": 1509.5996, "train_tokens_per_second": 858.339 }, { "epoch": 0.03664195745193757, "grad_norm": 2.61978816986084, "learning_rate": 1.548872180451128e-05, "loss": 0.3007138967514038, "num_input_tokens_seen": 1299298, "step": 456, "train_runtime": 1512.3721, "train_tokens_per_second": 859.113 }, { "epoch": 0.0367223126217883, "grad_norm": 2.66868257522583, "learning_rate": 1.5478696741854637e-05, "loss": 0.2031121701002121, "num_input_tokens_seen": 1301388, "step": 457, "train_runtime": 1514.7621, "train_tokens_per_second": 859.137 }, { "epoch": 0.03680266779163904, "grad_norm": 3.153425931930542, "learning_rate": 1.5468671679197994e-05, "loss": 0.34795188903808594, "num_input_tokens_seen": 1303070, "step": 458, "train_runtime": 1517.1264, "train_tokens_per_second": 858.907 }, { "epoch": 0.036883022961489785, "grad_norm": 3.2693755626678467, "learning_rate": 1.5458646616541354e-05, "loss": 0.20093488693237305, "num_input_tokens_seen": 1306748, "step": 459, "train_runtime": 1519.9649, "train_tokens_per_second": 859.722 }, { "epoch": 0.03696337813134053, "grad_norm": 3.1116180419921875, "learning_rate": 1.5448621553884714e-05, "loss": 0.2953888475894928, "num_input_tokens_seen": 1308518, "step": 460, "train_runtime": 1522.2599, "train_tokens_per_second": 859.589 }, { "epoch": 0.03704373330119127, "grad_norm": 2.8764989376068115, "learning_rate": 1.543859649122807e-05, "loss": 0.33561351895332336, "num_input_tokens_seen": 1311470, "step": 461, "train_runtime": 1524.8597, "train_tokens_per_second": 860.059 }, { "epoch": 0.037124088471042004, "grad_norm": 3.5070552825927734, "learning_rate": 1.542857142857143e-05, "loss": 0.2915307581424713, "num_input_tokens_seen": 1314370, "step": 462, "train_runtime": 1527.6472, "train_tokens_per_second": 860.388 }, { "epoch": 0.037204443640892745, "grad_norm": 2.316959857940674, "learning_rate": 1.5418546365914788e-05, "loss": 0.1979905366897583, "num_input_tokens_seen": 1316582, "step": 463, "train_runtime": 1530.0863, "train_tokens_per_second": 860.463 }, { "epoch": 0.03728479881074349, "grad_norm": 2.5102152824401855, "learning_rate": 1.5408521303258145e-05, "loss": 0.27561643719673157, "num_input_tokens_seen": 1319664, "step": 464, "train_runtime": 1532.7694, "train_tokens_per_second": 860.967 }, { "epoch": 0.03736515398059423, "grad_norm": 2.724153757095337, "learning_rate": 1.5398496240601505e-05, "loss": 0.22378146648406982, "num_input_tokens_seen": 1322124, "step": 465, "train_runtime": 1535.2706, "train_tokens_per_second": 861.167 }, { "epoch": 0.037445509150444964, "grad_norm": 3.267641544342041, "learning_rate": 1.5388471177944865e-05, "loss": 0.2911773920059204, "num_input_tokens_seen": 1325782, "step": 466, "train_runtime": 1538.05, "train_tokens_per_second": 861.989 }, { "epoch": 0.037525864320295706, "grad_norm": 2.7459402084350586, "learning_rate": 1.5378446115288222e-05, "loss": 0.3501089811325073, "num_input_tokens_seen": 1328474, "step": 467, "train_runtime": 1540.5782, "train_tokens_per_second": 862.322 }, { "epoch": 0.03760621949014645, "grad_norm": 2.6960482597351074, "learning_rate": 1.536842105263158e-05, "loss": 0.23897308111190796, "num_input_tokens_seen": 1330716, "step": 468, "train_runtime": 1543.0989, "train_tokens_per_second": 862.366 }, { "epoch": 0.03768657465999719, "grad_norm": 2.7215046882629395, "learning_rate": 1.535839598997494e-05, "loss": 0.23288050293922424, "num_input_tokens_seen": 1333014, "step": 469, "train_runtime": 1545.5126, "train_tokens_per_second": 862.506 }, { "epoch": 0.03776692982984793, "grad_norm": 2.6348910331726074, "learning_rate": 1.5348370927318296e-05, "loss": 0.29265719652175903, "num_input_tokens_seen": 1335778, "step": 470, "train_runtime": 1548.0611, "train_tokens_per_second": 862.872 }, { "epoch": 0.037847284999698666, "grad_norm": 3.967195749282837, "learning_rate": 1.5338345864661656e-05, "loss": 0.32230350375175476, "num_input_tokens_seen": 1337650, "step": 471, "train_runtime": 1550.4159, "train_tokens_per_second": 862.768 }, { "epoch": 0.03792764016954941, "grad_norm": 3.1403439044952393, "learning_rate": 1.5328320802005013e-05, "loss": 0.3154650628566742, "num_input_tokens_seen": 1340718, "step": 472, "train_runtime": 1553.079, "train_tokens_per_second": 863.264 }, { "epoch": 0.03800799533940015, "grad_norm": 3.00761342048645, "learning_rate": 1.5318295739348373e-05, "loss": 0.3360331356525421, "num_input_tokens_seen": 1342930, "step": 473, "train_runtime": 1555.4896, "train_tokens_per_second": 863.349 }, { "epoch": 0.03808835050925089, "grad_norm": 2.6518497467041016, "learning_rate": 1.530827067669173e-05, "loss": 0.26505953073501587, "num_input_tokens_seen": 1346268, "step": 474, "train_runtime": 1558.1871, "train_tokens_per_second": 863.996 }, { "epoch": 0.038168705679101626, "grad_norm": 3.2083017826080322, "learning_rate": 1.529824561403509e-05, "loss": 0.2443038821220398, "num_input_tokens_seen": 1348176, "step": 475, "train_runtime": 1560.5705, "train_tokens_per_second": 863.899 }, { "epoch": 0.03824906084895237, "grad_norm": 3.296663284301758, "learning_rate": 1.5288220551378447e-05, "loss": 0.352725625038147, "num_input_tokens_seen": 1350662, "step": 476, "train_runtime": 1563.0618, "train_tokens_per_second": 864.113 }, { "epoch": 0.03832941601880311, "grad_norm": 3.1446797847747803, "learning_rate": 1.5278195488721807e-05, "loss": 0.3125939965248108, "num_input_tokens_seen": 1352746, "step": 477, "train_runtime": 1565.4633, "train_tokens_per_second": 864.119 }, { "epoch": 0.03840977118865385, "grad_norm": 3.544168710708618, "learning_rate": 1.5268170426065164e-05, "loss": 0.35394471883773804, "num_input_tokens_seen": 1354352, "step": 478, "train_runtime": 1567.8334, "train_tokens_per_second": 863.837 }, { "epoch": 0.038490126358504594, "grad_norm": 3.535618305206299, "learning_rate": 1.5258145363408522e-05, "loss": 0.39981409907341003, "num_input_tokens_seen": 1357096, "step": 479, "train_runtime": 1570.4825, "train_tokens_per_second": 864.127 }, { "epoch": 0.03857048152835533, "grad_norm": 2.7207283973693848, "learning_rate": 1.524812030075188e-05, "loss": 0.29447442293167114, "num_input_tokens_seen": 1359772, "step": 480, "train_runtime": 1573.0652, "train_tokens_per_second": 864.409 }, { "epoch": 0.03865083669820607, "grad_norm": 3.4513680934906006, "learning_rate": 1.523809523809524e-05, "loss": 0.34852829575538635, "num_input_tokens_seen": 1362576, "step": 481, "train_runtime": 1587.1956, "train_tokens_per_second": 858.48 }, { "epoch": 0.03873119186805681, "grad_norm": 3.0656425952911377, "learning_rate": 1.5228070175438598e-05, "loss": 0.2447851300239563, "num_input_tokens_seen": 1365624, "step": 482, "train_runtime": 1589.7962, "train_tokens_per_second": 858.993 }, { "epoch": 0.038811547037907554, "grad_norm": 3.050888776779175, "learning_rate": 1.5218045112781956e-05, "loss": 0.33830946683883667, "num_input_tokens_seen": 1367952, "step": 483, "train_runtime": 1592.2706, "train_tokens_per_second": 859.12 }, { "epoch": 0.03889190220775829, "grad_norm": 3.2872021198272705, "learning_rate": 1.5208020050125315e-05, "loss": 0.3517390191555023, "num_input_tokens_seen": 1370314, "step": 484, "train_runtime": 1594.7226, "train_tokens_per_second": 859.28 }, { "epoch": 0.03897225737760903, "grad_norm": 2.7993619441986084, "learning_rate": 1.5197994987468673e-05, "loss": 0.2503485381603241, "num_input_tokens_seen": 1374180, "step": 485, "train_runtime": 1597.5603, "train_tokens_per_second": 860.174 }, { "epoch": 0.03905261254745977, "grad_norm": 3.098541259765625, "learning_rate": 1.5187969924812032e-05, "loss": 0.34080803394317627, "num_input_tokens_seen": 1377594, "step": 486, "train_runtime": 1600.2556, "train_tokens_per_second": 860.859 }, { "epoch": 0.039132967717310514, "grad_norm": 2.828519821166992, "learning_rate": 1.5177944862155388e-05, "loss": 0.3073095381259918, "num_input_tokens_seen": 1380600, "step": 487, "train_runtime": 1602.853, "train_tokens_per_second": 861.339 }, { "epoch": 0.039213322887161256, "grad_norm": 3.2828986644744873, "learning_rate": 1.5167919799498747e-05, "loss": 0.35353776812553406, "num_input_tokens_seen": 1382842, "step": 488, "train_runtime": 1605.2335, "train_tokens_per_second": 861.458 }, { "epoch": 0.03929367805701199, "grad_norm": 3.4282615184783936, "learning_rate": 1.5157894736842107e-05, "loss": 0.3232978880405426, "num_input_tokens_seen": 1385756, "step": 489, "train_runtime": 1607.8496, "train_tokens_per_second": 861.869 }, { "epoch": 0.03937403322686273, "grad_norm": 4.491104602813721, "learning_rate": 1.5147869674185464e-05, "loss": 0.22646228969097137, "num_input_tokens_seen": 1388056, "step": 490, "train_runtime": 1610.3302, "train_tokens_per_second": 861.97 }, { "epoch": 0.039454388396713475, "grad_norm": 2.283034086227417, "learning_rate": 1.5137844611528822e-05, "loss": 0.2124098837375641, "num_input_tokens_seen": 1392648, "step": 491, "train_runtime": 1613.4556, "train_tokens_per_second": 863.146 }, { "epoch": 0.039534743566564216, "grad_norm": 3.596614122390747, "learning_rate": 1.5127819548872183e-05, "loss": 0.3218157887458801, "num_input_tokens_seen": 1394394, "step": 492, "train_runtime": 1615.8935, "train_tokens_per_second": 862.924 }, { "epoch": 0.03961509873641495, "grad_norm": 2.380276679992676, "learning_rate": 1.511779448621554e-05, "loss": 0.2117171287536621, "num_input_tokens_seen": 1396202, "step": 493, "train_runtime": 1618.3682, "train_tokens_per_second": 862.722 }, { "epoch": 0.03969545390626569, "grad_norm": 3.0455877780914307, "learning_rate": 1.5107769423558898e-05, "loss": 0.30723366141319275, "num_input_tokens_seen": 1398320, "step": 494, "train_runtime": 1620.8583, "train_tokens_per_second": 862.703 }, { "epoch": 0.039775809076116435, "grad_norm": 3.120380401611328, "learning_rate": 1.5097744360902258e-05, "loss": 0.2298741489648819, "num_input_tokens_seen": 1400732, "step": 495, "train_runtime": 1623.3164, "train_tokens_per_second": 862.883 }, { "epoch": 0.03985616424596718, "grad_norm": 3.3141303062438965, "learning_rate": 1.5087719298245615e-05, "loss": 0.2528960406780243, "num_input_tokens_seen": 1402944, "step": 496, "train_runtime": 1625.7671, "train_tokens_per_second": 862.943 }, { "epoch": 0.03993651941581792, "grad_norm": 3.06976056098938, "learning_rate": 1.5077694235588973e-05, "loss": 0.3393453061580658, "num_input_tokens_seen": 1404702, "step": 497, "train_runtime": 1628.0871, "train_tokens_per_second": 862.793 }, { "epoch": 0.04001687458566865, "grad_norm": 3.0676136016845703, "learning_rate": 1.506766917293233e-05, "loss": 0.28514528274536133, "num_input_tokens_seen": 1407538, "step": 498, "train_runtime": 1630.7287, "train_tokens_per_second": 863.134 }, { "epoch": 0.040097229755519395, "grad_norm": 2.79145884513855, "learning_rate": 1.505764411027569e-05, "loss": 0.2932468056678772, "num_input_tokens_seen": 1410390, "step": 499, "train_runtime": 1633.3503, "train_tokens_per_second": 863.495 }, { "epoch": 0.04017758492537014, "grad_norm": 3.1214354038238525, "learning_rate": 1.5047619047619049e-05, "loss": 0.25155845284461975, "num_input_tokens_seen": 1413128, "step": 500, "train_runtime": 1635.9316, "train_tokens_per_second": 863.806 }, { "epoch": 0.04025794009522088, "grad_norm": 2.3560523986816406, "learning_rate": 1.5037593984962406e-05, "loss": 0.23559823632240295, "num_input_tokens_seen": 1415440, "step": 501, "train_runtime": 1638.5097, "train_tokens_per_second": 863.858 }, { "epoch": 0.040338295265071614, "grad_norm": 2.8014540672302246, "learning_rate": 1.5027568922305766e-05, "loss": 0.2883181571960449, "num_input_tokens_seen": 1417162, "step": 502, "train_runtime": 1640.8239, "train_tokens_per_second": 863.689 }, { "epoch": 0.040418650434922355, "grad_norm": 3.7968125343322754, "learning_rate": 1.5017543859649124e-05, "loss": 0.6628416776657104, "num_input_tokens_seen": 1420222, "step": 503, "train_runtime": 1643.4579, "train_tokens_per_second": 864.167 }, { "epoch": 0.0404990056047731, "grad_norm": 2.82828688621521, "learning_rate": 1.5007518796992481e-05, "loss": 0.28036296367645264, "num_input_tokens_seen": 1422696, "step": 504, "train_runtime": 1645.9839, "train_tokens_per_second": 864.344 }, { "epoch": 0.04057936077462384, "grad_norm": 2.7034668922424316, "learning_rate": 1.4997493734335841e-05, "loss": 0.24519217014312744, "num_input_tokens_seen": 1425890, "step": 505, "train_runtime": 1648.6592, "train_tokens_per_second": 864.879 }, { "epoch": 0.04065971594447458, "grad_norm": 3.060004711151123, "learning_rate": 1.49874686716792e-05, "loss": 0.2792198061943054, "num_input_tokens_seen": 1427956, "step": 506, "train_runtime": 1651.0838, "train_tokens_per_second": 864.86 }, { "epoch": 0.040740071114325316, "grad_norm": 3.2986347675323486, "learning_rate": 1.4977443609022557e-05, "loss": 0.31121474504470825, "num_input_tokens_seen": 1432658, "step": 507, "train_runtime": 1654.1753, "train_tokens_per_second": 866.086 }, { "epoch": 0.04082042628417606, "grad_norm": 2.542386531829834, "learning_rate": 1.4967418546365915e-05, "loss": 0.3004966676235199, "num_input_tokens_seen": 1436322, "step": 508, "train_runtime": 1656.9701, "train_tokens_per_second": 866.836 }, { "epoch": 0.0409007814540268, "grad_norm": 3.392289161682129, "learning_rate": 1.4957393483709275e-05, "loss": 0.37108853459358215, "num_input_tokens_seen": 1438828, "step": 509, "train_runtime": 1659.4776, "train_tokens_per_second": 867.037 }, { "epoch": 0.04098113662387754, "grad_norm": 3.3409488201141357, "learning_rate": 1.4947368421052632e-05, "loss": 0.32543739676475525, "num_input_tokens_seen": 1440300, "step": 510, "train_runtime": 1661.7249, "train_tokens_per_second": 866.75 }, { "epoch": 0.041061491793728276, "grad_norm": 2.394435167312622, "learning_rate": 1.493734335839599e-05, "loss": 0.27898499369621277, "num_input_tokens_seen": 1443786, "step": 511, "train_runtime": 1676.131, "train_tokens_per_second": 861.38 }, { "epoch": 0.04114184696357902, "grad_norm": 1.9380141496658325, "learning_rate": 1.492731829573935e-05, "loss": 0.19266772270202637, "num_input_tokens_seen": 1447070, "step": 512, "train_runtime": 1678.8481, "train_tokens_per_second": 861.942 }, { "epoch": 0.04122220213342976, "grad_norm": 3.0689549446105957, "learning_rate": 1.4917293233082707e-05, "loss": 0.21198229491710663, "num_input_tokens_seen": 1449774, "step": 513, "train_runtime": 1681.412, "train_tokens_per_second": 862.236 }, { "epoch": 0.0413025573032805, "grad_norm": 3.324253797531128, "learning_rate": 1.4907268170426066e-05, "loss": 0.2911336421966553, "num_input_tokens_seen": 1452068, "step": 514, "train_runtime": 1683.901, "train_tokens_per_second": 862.324 }, { "epoch": 0.04138291247313124, "grad_norm": 2.9122893810272217, "learning_rate": 1.4897243107769426e-05, "loss": 0.24665817618370056, "num_input_tokens_seen": 1455458, "step": 515, "train_runtime": 1686.6044, "train_tokens_per_second": 862.952 }, { "epoch": 0.04146326764298198, "grad_norm": 3.327023506164551, "learning_rate": 1.4887218045112783e-05, "loss": 0.3381025791168213, "num_input_tokens_seen": 1458584, "step": 516, "train_runtime": 1689.2405, "train_tokens_per_second": 863.455 }, { "epoch": 0.04154362281283272, "grad_norm": 2.6475985050201416, "learning_rate": 1.4877192982456141e-05, "loss": 0.31457406282424927, "num_input_tokens_seen": 1461486, "step": 517, "train_runtime": 1691.8982, "train_tokens_per_second": 863.814 }, { "epoch": 0.04162397798268346, "grad_norm": 2.8047564029693604, "learning_rate": 1.4867167919799498e-05, "loss": 0.3272978663444519, "num_input_tokens_seen": 1466890, "step": 518, "train_runtime": 1695.1497, "train_tokens_per_second": 865.345 }, { "epoch": 0.041704333152534204, "grad_norm": 3.1899144649505615, "learning_rate": 1.4857142857142858e-05, "loss": 0.3464055359363556, "num_input_tokens_seen": 1471594, "step": 519, "train_runtime": 1698.2342, "train_tokens_per_second": 866.544 }, { "epoch": 0.04178468832238494, "grad_norm": 3.2701563835144043, "learning_rate": 1.4847117794486217e-05, "loss": 0.3693785071372986, "num_input_tokens_seen": 1473812, "step": 520, "train_runtime": 1700.7244, "train_tokens_per_second": 866.579 }, { "epoch": 0.04186504349223568, "grad_norm": 3.076849937438965, "learning_rate": 1.4837092731829574e-05, "loss": 0.2547920048236847, "num_input_tokens_seen": 1475738, "step": 521, "train_runtime": 1703.1128, "train_tokens_per_second": 866.495 }, { "epoch": 0.04194539866208642, "grad_norm": 3.232638359069824, "learning_rate": 1.4827067669172934e-05, "loss": 0.30758124589920044, "num_input_tokens_seen": 1479010, "step": 522, "train_runtime": 1705.8472, "train_tokens_per_second": 867.024 }, { "epoch": 0.042025753831937164, "grad_norm": 2.9567739963531494, "learning_rate": 1.4817042606516292e-05, "loss": 0.34875011444091797, "num_input_tokens_seen": 1481060, "step": 523, "train_runtime": 1708.3897, "train_tokens_per_second": 866.933 }, { "epoch": 0.042106109001787906, "grad_norm": 2.7298872470855713, "learning_rate": 1.4807017543859649e-05, "loss": 0.2419530749320984, "num_input_tokens_seen": 1484018, "step": 524, "train_runtime": 1711.0363, "train_tokens_per_second": 867.321 }, { "epoch": 0.04218646417163864, "grad_norm": 3.3870508670806885, "learning_rate": 1.479699248120301e-05, "loss": 0.3683491349220276, "num_input_tokens_seen": 1486288, "step": 525, "train_runtime": 1713.4606, "train_tokens_per_second": 867.419 }, { "epoch": 0.04226681934148938, "grad_norm": 2.653449773788452, "learning_rate": 1.4786967418546368e-05, "loss": 0.26108017563819885, "num_input_tokens_seen": 1490818, "step": 526, "train_runtime": 1716.4775, "train_tokens_per_second": 868.533 }, { "epoch": 0.042347174511340124, "grad_norm": 2.9990553855895996, "learning_rate": 1.4776942355889725e-05, "loss": 0.3187485933303833, "num_input_tokens_seen": 1492714, "step": 527, "train_runtime": 1718.8319, "train_tokens_per_second": 868.447 }, { "epoch": 0.042427529681190866, "grad_norm": 2.8503904342651367, "learning_rate": 1.4766917293233083e-05, "loss": 0.2494989037513733, "num_input_tokens_seen": 1494814, "step": 528, "train_runtime": 1721.242, "train_tokens_per_second": 868.451 }, { "epoch": 0.0425078848510416, "grad_norm": 2.783130407333374, "learning_rate": 1.4756892230576443e-05, "loss": 0.2761821150779724, "num_input_tokens_seen": 1496788, "step": 529, "train_runtime": 1723.6008, "train_tokens_per_second": 868.408 }, { "epoch": 0.04258824002089234, "grad_norm": 3.1766133308410645, "learning_rate": 1.47468671679198e-05, "loss": 0.29582470655441284, "num_input_tokens_seen": 1499546, "step": 530, "train_runtime": 1726.1639, "train_tokens_per_second": 868.716 }, { "epoch": 0.042668595190743085, "grad_norm": 3.0334694385528564, "learning_rate": 1.4736842105263159e-05, "loss": 0.2379629909992218, "num_input_tokens_seen": 1501096, "step": 531, "train_runtime": 1728.4528, "train_tokens_per_second": 868.462 }, { "epoch": 0.042748950360593826, "grad_norm": 3.1252830028533936, "learning_rate": 1.4726817042606519e-05, "loss": 0.305890291929245, "num_input_tokens_seen": 1504344, "step": 532, "train_runtime": 1731.1303, "train_tokens_per_second": 868.995 }, { "epoch": 0.04282930553044457, "grad_norm": 3.7489023208618164, "learning_rate": 1.4716791979949876e-05, "loss": 0.3382200002670288, "num_input_tokens_seen": 1507846, "step": 533, "train_runtime": 1733.9005, "train_tokens_per_second": 869.627 }, { "epoch": 0.0429096607002953, "grad_norm": 3.03436279296875, "learning_rate": 1.4706766917293234e-05, "loss": 0.26980990171432495, "num_input_tokens_seen": 1510124, "step": 534, "train_runtime": 1736.3382, "train_tokens_per_second": 869.718 }, { "epoch": 0.042990015870146045, "grad_norm": 3.014248847961426, "learning_rate": 1.4696741854636594e-05, "loss": 0.24341891705989838, "num_input_tokens_seen": 1512808, "step": 535, "train_runtime": 1738.8458, "train_tokens_per_second": 870.007 }, { "epoch": 0.04307037103999679, "grad_norm": 3.535560131072998, "learning_rate": 1.4686716791979951e-05, "loss": 0.4216445982456207, "num_input_tokens_seen": 1514670, "step": 536, "train_runtime": 1741.1482, "train_tokens_per_second": 869.926 }, { "epoch": 0.04315072620984753, "grad_norm": 3.0121147632598877, "learning_rate": 1.467669172932331e-05, "loss": 0.2686004638671875, "num_input_tokens_seen": 1516724, "step": 537, "train_runtime": 1743.5305, "train_tokens_per_second": 869.915 }, { "epoch": 0.043231081379698263, "grad_norm": 3.1252758502960205, "learning_rate": 1.4666666666666666e-05, "loss": 0.3464912176132202, "num_input_tokens_seen": 1520840, "step": 538, "train_runtime": 1746.412, "train_tokens_per_second": 870.837 }, { "epoch": 0.043311436549549005, "grad_norm": 2.089831590652466, "learning_rate": 1.4656641604010026e-05, "loss": 0.21078312397003174, "num_input_tokens_seen": 1523524, "step": 539, "train_runtime": 1748.9078, "train_tokens_per_second": 871.129 }, { "epoch": 0.04339179171939975, "grad_norm": 2.758943557739258, "learning_rate": 1.4646616541353385e-05, "loss": 0.31783950328826904, "num_input_tokens_seen": 1525706, "step": 540, "train_runtime": 1751.3245, "train_tokens_per_second": 871.173 }, { "epoch": 0.04347214688925049, "grad_norm": 2.2309701442718506, "learning_rate": 1.4636591478696742e-05, "loss": 0.2050892412662506, "num_input_tokens_seen": 1527712, "step": 541, "train_runtime": 1765.2932, "train_tokens_per_second": 865.415 }, { "epoch": 0.04355250205910123, "grad_norm": 3.1627402305603027, "learning_rate": 1.4626566416040102e-05, "loss": 0.23846906423568726, "num_input_tokens_seen": 1529966, "step": 542, "train_runtime": 1767.7751, "train_tokens_per_second": 865.475 }, { "epoch": 0.043632857228951966, "grad_norm": 2.7366840839385986, "learning_rate": 1.461654135338346e-05, "loss": 0.23605456948280334, "num_input_tokens_seen": 1533540, "step": 543, "train_runtime": 1770.5047, "train_tokens_per_second": 866.16 }, { "epoch": 0.04371321239880271, "grad_norm": 3.4041922092437744, "learning_rate": 1.4606516290726817e-05, "loss": 0.4136143922805786, "num_input_tokens_seen": 1535940, "step": 544, "train_runtime": 1773.0, "train_tokens_per_second": 866.294 }, { "epoch": 0.04379356756865345, "grad_norm": 3.4260804653167725, "learning_rate": 1.4596491228070177e-05, "loss": 0.28705301880836487, "num_input_tokens_seen": 1537822, "step": 545, "train_runtime": 1775.329, "train_tokens_per_second": 866.218 }, { "epoch": 0.04387392273850419, "grad_norm": 3.232818603515625, "learning_rate": 1.4586466165413536e-05, "loss": 0.24531015753746033, "num_input_tokens_seen": 1539660, "step": 546, "train_runtime": 1777.6408, "train_tokens_per_second": 866.126 }, { "epoch": 0.043954277908354926, "grad_norm": 3.0654499530792236, "learning_rate": 1.4576441102756893e-05, "loss": 0.25500035285949707, "num_input_tokens_seen": 1542448, "step": 547, "train_runtime": 1780.1779, "train_tokens_per_second": 866.457 }, { "epoch": 0.04403463307820567, "grad_norm": 2.7167179584503174, "learning_rate": 1.4566416040100251e-05, "loss": 0.21828681230545044, "num_input_tokens_seen": 1544412, "step": 548, "train_runtime": 1782.5401, "train_tokens_per_second": 866.411 }, { "epoch": 0.04411498824805641, "grad_norm": 3.158804416656494, "learning_rate": 1.4556390977443611e-05, "loss": 0.2506077289581299, "num_input_tokens_seen": 1547582, "step": 549, "train_runtime": 1785.2146, "train_tokens_per_second": 866.889 }, { "epoch": 0.04419534341790715, "grad_norm": 2.4271559715270996, "learning_rate": 1.4546365914786968e-05, "loss": 0.22938311100006104, "num_input_tokens_seen": 1550674, "step": 550, "train_runtime": 1787.8532, "train_tokens_per_second": 867.339 }, { "epoch": 0.04427569858775789, "grad_norm": 3.9599802494049072, "learning_rate": 1.4536340852130327e-05, "loss": 0.2798722982406616, "num_input_tokens_seen": 1552400, "step": 551, "train_runtime": 1790.1768, "train_tokens_per_second": 867.177 }, { "epoch": 0.04435605375760863, "grad_norm": 2.8766610622406006, "learning_rate": 1.4526315789473687e-05, "loss": 0.2502022087574005, "num_input_tokens_seen": 1556150, "step": 552, "train_runtime": 1793.0286, "train_tokens_per_second": 867.889 }, { "epoch": 0.04443640892745937, "grad_norm": 2.0394608974456787, "learning_rate": 1.4516290726817044e-05, "loss": 0.2013159692287445, "num_input_tokens_seen": 1558574, "step": 553, "train_runtime": 1795.5473, "train_tokens_per_second": 868.022 }, { "epoch": 0.04451676409731011, "grad_norm": 2.5627055168151855, "learning_rate": 1.4506265664160402e-05, "loss": 0.21650420129299164, "num_input_tokens_seen": 1560378, "step": 554, "train_runtime": 1797.9176, "train_tokens_per_second": 867.881 }, { "epoch": 0.044597119267160854, "grad_norm": 2.2948267459869385, "learning_rate": 1.4496240601503762e-05, "loss": 0.23037126660346985, "num_input_tokens_seen": 1562900, "step": 555, "train_runtime": 1800.4369, "train_tokens_per_second": 868.067 }, { "epoch": 0.04467747443701159, "grad_norm": 2.8712704181671143, "learning_rate": 1.4486215538847119e-05, "loss": 0.25969430804252625, "num_input_tokens_seen": 1565406, "step": 556, "train_runtime": 1802.9261, "train_tokens_per_second": 868.259 }, { "epoch": 0.04475782960686233, "grad_norm": 3.0390191078186035, "learning_rate": 1.4476190476190478e-05, "loss": 0.22287069261074066, "num_input_tokens_seen": 1568046, "step": 557, "train_runtime": 1805.4956, "train_tokens_per_second": 868.485 }, { "epoch": 0.04483818477671307, "grad_norm": 2.714883804321289, "learning_rate": 1.4466165413533834e-05, "loss": 0.279304176568985, "num_input_tokens_seen": 1571150, "step": 558, "train_runtime": 1808.2129, "train_tokens_per_second": 868.897 }, { "epoch": 0.044918539946563814, "grad_norm": 3.1715149879455566, "learning_rate": 1.4456140350877195e-05, "loss": 0.34002357721328735, "num_input_tokens_seen": 1574144, "step": 559, "train_runtime": 1810.8781, "train_tokens_per_second": 869.271 }, { "epoch": 0.044998895116414556, "grad_norm": 2.2238173484802246, "learning_rate": 1.4446115288220553e-05, "loss": 0.1641891896724701, "num_input_tokens_seen": 1576632, "step": 560, "train_runtime": 1813.3875, "train_tokens_per_second": 869.44 }, { "epoch": 0.04507925028626529, "grad_norm": 2.898747444152832, "learning_rate": 1.443609022556391e-05, "loss": 0.27240127325057983, "num_input_tokens_seen": 1579388, "step": 561, "train_runtime": 1816.0099, "train_tokens_per_second": 869.702 }, { "epoch": 0.04515960545611603, "grad_norm": 2.9486331939697266, "learning_rate": 1.442606516290727e-05, "loss": 0.2826988101005554, "num_input_tokens_seen": 1582576, "step": 562, "train_runtime": 1818.6775, "train_tokens_per_second": 870.18 }, { "epoch": 0.045239960625966774, "grad_norm": 3.0795254707336426, "learning_rate": 1.4416040100250628e-05, "loss": 0.29547619819641113, "num_input_tokens_seen": 1587280, "step": 563, "train_runtime": 1821.8295, "train_tokens_per_second": 871.256 }, { "epoch": 0.045320315795817516, "grad_norm": 3.9188342094421387, "learning_rate": 1.4406015037593985e-05, "loss": 0.3751507103443146, "num_input_tokens_seen": 1589554, "step": 564, "train_runtime": 1824.3463, "train_tokens_per_second": 871.301 }, { "epoch": 0.04540067096566825, "grad_norm": 2.7381668090820312, "learning_rate": 1.4395989974937345e-05, "loss": 0.3298999071121216, "num_input_tokens_seen": 1592076, "step": 565, "train_runtime": 1826.8601, "train_tokens_per_second": 871.482 }, { "epoch": 0.04548102613551899, "grad_norm": 2.9839556217193604, "learning_rate": 1.4385964912280704e-05, "loss": 0.3729553520679474, "num_input_tokens_seen": 1593870, "step": 566, "train_runtime": 1829.2286, "train_tokens_per_second": 871.335 }, { "epoch": 0.045561381305369734, "grad_norm": 3.271955966949463, "learning_rate": 1.437593984962406e-05, "loss": 0.32570868730545044, "num_input_tokens_seen": 1596290, "step": 567, "train_runtime": 1831.7186, "train_tokens_per_second": 871.471 }, { "epoch": 0.045641736475220476, "grad_norm": 2.670545816421509, "learning_rate": 1.436591478696742e-05, "loss": 0.17393577098846436, "num_input_tokens_seen": 1599362, "step": 568, "train_runtime": 1834.3632, "train_tokens_per_second": 871.889 }, { "epoch": 0.04572209164507122, "grad_norm": 2.821702241897583, "learning_rate": 1.435588972431078e-05, "loss": 0.310310423374176, "num_input_tokens_seen": 1604282, "step": 569, "train_runtime": 1837.4299, "train_tokens_per_second": 873.112 }, { "epoch": 0.04580244681492195, "grad_norm": 3.5008559226989746, "learning_rate": 1.4345864661654136e-05, "loss": 0.2781810760498047, "num_input_tokens_seen": 1606524, "step": 570, "train_runtime": 1839.9123, "train_tokens_per_second": 873.152 }, { "epoch": 0.045882801984772695, "grad_norm": 2.7812283039093018, "learning_rate": 1.4335839598997495e-05, "loss": 0.2434626966714859, "num_input_tokens_seen": 1608744, "step": 571, "train_runtime": 1853.9801, "train_tokens_per_second": 867.725 }, { "epoch": 0.04596315715462344, "grad_norm": 3.587242841720581, "learning_rate": 1.4325814536340855e-05, "loss": 0.3550419807434082, "num_input_tokens_seen": 1612198, "step": 572, "train_runtime": 1856.7229, "train_tokens_per_second": 868.303 }, { "epoch": 0.04604351232447418, "grad_norm": 3.0360004901885986, "learning_rate": 1.4315789473684212e-05, "loss": 0.3085740804672241, "num_input_tokens_seen": 1614602, "step": 573, "train_runtime": 1859.1896, "train_tokens_per_second": 868.444 }, { "epoch": 0.04612386749432491, "grad_norm": 3.2381155490875244, "learning_rate": 1.430576441102757e-05, "loss": 0.3477135896682739, "num_input_tokens_seen": 1618068, "step": 574, "train_runtime": 1861.9362, "train_tokens_per_second": 869.024 }, { "epoch": 0.046204222664175655, "grad_norm": 3.5969059467315674, "learning_rate": 1.4295739348370929e-05, "loss": 0.2693406045436859, "num_input_tokens_seen": 1620680, "step": 575, "train_runtime": 1864.4321, "train_tokens_per_second": 869.262 }, { "epoch": 0.0462845778340264, "grad_norm": 2.951134443283081, "learning_rate": 1.4285714285714287e-05, "loss": 0.3496609926223755, "num_input_tokens_seen": 1624514, "step": 576, "train_runtime": 1867.2271, "train_tokens_per_second": 870.014 }, { "epoch": 0.04636493300387714, "grad_norm": 2.7670092582702637, "learning_rate": 1.4275689223057646e-05, "loss": 0.23010921478271484, "num_input_tokens_seen": 1628540, "step": 577, "train_runtime": 1870.0945, "train_tokens_per_second": 870.833 }, { "epoch": 0.04644528817372788, "grad_norm": 2.8113555908203125, "learning_rate": 1.4265664160401002e-05, "loss": 0.30008959770202637, "num_input_tokens_seen": 1632160, "step": 578, "train_runtime": 1872.9021, "train_tokens_per_second": 871.46 }, { "epoch": 0.046525643343578615, "grad_norm": 2.7752041816711426, "learning_rate": 1.4255639097744363e-05, "loss": 0.26254767179489136, "num_input_tokens_seen": 1635032, "step": 579, "train_runtime": 1875.5129, "train_tokens_per_second": 871.779 }, { "epoch": 0.04660599851342936, "grad_norm": 2.9898886680603027, "learning_rate": 1.4245614035087721e-05, "loss": 0.3409383296966553, "num_input_tokens_seen": 1638442, "step": 580, "train_runtime": 1878.2736, "train_tokens_per_second": 872.313 }, { "epoch": 0.0466863536832801, "grad_norm": 2.7706613540649414, "learning_rate": 1.4235588972431078e-05, "loss": 0.25528547167778015, "num_input_tokens_seen": 1642506, "step": 581, "train_runtime": 1881.1121, "train_tokens_per_second": 873.157 }, { "epoch": 0.04676670885313084, "grad_norm": 3.0313048362731934, "learning_rate": 1.4225563909774438e-05, "loss": 0.39156675338745117, "num_input_tokens_seen": 1644554, "step": 582, "train_runtime": 1883.4718, "train_tokens_per_second": 873.15 }, { "epoch": 0.046847064022981576, "grad_norm": 2.8468167781829834, "learning_rate": 1.4215538847117797e-05, "loss": 0.24175627529621124, "num_input_tokens_seen": 1646628, "step": 583, "train_runtime": 1885.8818, "train_tokens_per_second": 873.134 }, { "epoch": 0.04692741919283232, "grad_norm": 3.092499017715454, "learning_rate": 1.4205513784461153e-05, "loss": 0.23448869585990906, "num_input_tokens_seen": 1650096, "step": 584, "train_runtime": 1888.6359, "train_tokens_per_second": 873.697 }, { "epoch": 0.04700777436268306, "grad_norm": 3.5609030723571777, "learning_rate": 1.4195488721804514e-05, "loss": 0.31450241804122925, "num_input_tokens_seen": 1652456, "step": 585, "train_runtime": 1891.1314, "train_tokens_per_second": 873.792 }, { "epoch": 0.0470881295325338, "grad_norm": 3.4701812267303467, "learning_rate": 1.4185463659147872e-05, "loss": 0.2631363570690155, "num_input_tokens_seen": 1653906, "step": 586, "train_runtime": 1893.4013, "train_tokens_per_second": 873.511 }, { "epoch": 0.04716848470238454, "grad_norm": 2.7920117378234863, "learning_rate": 1.4175438596491229e-05, "loss": 0.3138284683227539, "num_input_tokens_seen": 1656056, "step": 587, "train_runtime": 1895.7955, "train_tokens_per_second": 873.541 }, { "epoch": 0.04724883987223528, "grad_norm": 2.7733755111694336, "learning_rate": 1.4165413533834587e-05, "loss": 0.2623884379863739, "num_input_tokens_seen": 1658828, "step": 588, "train_runtime": 1898.3514, "train_tokens_per_second": 873.826 }, { "epoch": 0.04732919504208602, "grad_norm": 2.7091736793518066, "learning_rate": 1.4155388471177946e-05, "loss": 0.2920924127101898, "num_input_tokens_seen": 1662298, "step": 589, "train_runtime": 1901.1781, "train_tokens_per_second": 874.352 }, { "epoch": 0.04740955021193676, "grad_norm": 2.6660897731781006, "learning_rate": 1.4145363408521304e-05, "loss": 0.28765198588371277, "num_input_tokens_seen": 1664048, "step": 590, "train_runtime": 1903.599, "train_tokens_per_second": 874.159 }, { "epoch": 0.0474899053817875, "grad_norm": 3.03718638420105, "learning_rate": 1.4135338345864663e-05, "loss": 0.2665225863456726, "num_input_tokens_seen": 1665786, "step": 591, "train_runtime": 1905.984, "train_tokens_per_second": 873.977 }, { "epoch": 0.04757026055163824, "grad_norm": 3.444945812225342, "learning_rate": 1.4125313283208021e-05, "loss": 0.3407847285270691, "num_input_tokens_seen": 1667348, "step": 592, "train_runtime": 1908.2799, "train_tokens_per_second": 873.744 }, { "epoch": 0.04765061572148898, "grad_norm": 2.9096648693084717, "learning_rate": 1.411528822055138e-05, "loss": 0.26239296793937683, "num_input_tokens_seen": 1669890, "step": 593, "train_runtime": 1910.8028, "train_tokens_per_second": 873.921 }, { "epoch": 0.04773097089133972, "grad_norm": 3.274122953414917, "learning_rate": 1.4105263157894738e-05, "loss": 0.29693636298179626, "num_input_tokens_seen": 1671744, "step": 594, "train_runtime": 1913.1217, "train_tokens_per_second": 873.83 }, { "epoch": 0.047811326061190464, "grad_norm": 3.882530689239502, "learning_rate": 1.4095238095238097e-05, "loss": 0.4560999274253845, "num_input_tokens_seen": 1673946, "step": 595, "train_runtime": 1915.567, "train_tokens_per_second": 873.865 }, { "epoch": 0.047891681231041205, "grad_norm": 3.5009660720825195, "learning_rate": 1.4085213032581455e-05, "loss": 0.2948603928089142, "num_input_tokens_seen": 1675936, "step": 596, "train_runtime": 1917.9417, "train_tokens_per_second": 873.82 }, { "epoch": 0.04797203640089194, "grad_norm": 4.356101989746094, "learning_rate": 1.4075187969924814e-05, "loss": 0.4128536581993103, "num_input_tokens_seen": 1677602, "step": 597, "train_runtime": 1920.2151, "train_tokens_per_second": 873.653 }, { "epoch": 0.04805239157074268, "grad_norm": 3.148362159729004, "learning_rate": 1.406516290726817e-05, "loss": 0.2090621292591095, "num_input_tokens_seen": 1679824, "step": 598, "train_runtime": 1922.6494, "train_tokens_per_second": 873.703 }, { "epoch": 0.048132746740593424, "grad_norm": 2.914428949356079, "learning_rate": 1.405513784461153e-05, "loss": 0.23675161600112915, "num_input_tokens_seen": 1682860, "step": 599, "train_runtime": 1925.3064, "train_tokens_per_second": 874.074 }, { "epoch": 0.048213101910444166, "grad_norm": 2.7452480792999268, "learning_rate": 1.4045112781954887e-05, "loss": 0.2620311677455902, "num_input_tokens_seen": 1685492, "step": 600, "train_runtime": 1927.8401, "train_tokens_per_second": 874.29 }, { "epoch": 0.048213101910444166, "eval_loss": 1.1551827192306519, "eval_runtime": 68.4628, "eval_samples_per_second": 14.519, "eval_steps_per_second": 7.259, "num_input_tokens_seen": 1685492, "step": 600 }, { "epoch": 0.0482934570802949, "grad_norm": 2.5968258380889893, "learning_rate": 1.4035087719298246e-05, "loss": 0.21236193180084229, "num_input_tokens_seen": 1687672, "step": 601, "train_runtime": 2009.9991, "train_tokens_per_second": 839.638 }, { "epoch": 0.04837381225014564, "grad_norm": 3.3968238830566406, "learning_rate": 1.4025062656641606e-05, "loss": 0.33055582642555237, "num_input_tokens_seen": 1689866, "step": 602, "train_runtime": 2012.4516, "train_tokens_per_second": 839.705 }, { "epoch": 0.048454167419996384, "grad_norm": 3.2553207874298096, "learning_rate": 1.4015037593984963e-05, "loss": 0.31621479988098145, "num_input_tokens_seen": 1692804, "step": 603, "train_runtime": 2015.0572, "train_tokens_per_second": 840.077 }, { "epoch": 0.048534522589847126, "grad_norm": 3.6723170280456543, "learning_rate": 1.4005012531328321e-05, "loss": 0.3579470217227936, "num_input_tokens_seen": 1694786, "step": 604, "train_runtime": 2017.4139, "train_tokens_per_second": 840.078 }, { "epoch": 0.04861487775969787, "grad_norm": 3.066075086593628, "learning_rate": 1.3994987468671682e-05, "loss": 0.26271671056747437, "num_input_tokens_seen": 1698726, "step": 605, "train_runtime": 2020.2594, "train_tokens_per_second": 840.845 }, { "epoch": 0.0486952329295486, "grad_norm": 3.0611395835876465, "learning_rate": 1.3984962406015038e-05, "loss": 0.3656150698661804, "num_input_tokens_seen": 1701924, "step": 606, "train_runtime": 2022.9474, "train_tokens_per_second": 841.309 }, { "epoch": 0.048775588099399345, "grad_norm": 2.2558176517486572, "learning_rate": 1.3974937343358397e-05, "loss": 0.1754308044910431, "num_input_tokens_seen": 1704664, "step": 607, "train_runtime": 2025.5139, "train_tokens_per_second": 841.596 }, { "epoch": 0.048855943269250086, "grad_norm": 2.7675423622131348, "learning_rate": 1.3964912280701755e-05, "loss": 0.30169665813446045, "num_input_tokens_seen": 1707604, "step": 608, "train_runtime": 2028.1252, "train_tokens_per_second": 841.962 }, { "epoch": 0.04893629843910083, "grad_norm": 3.326284885406494, "learning_rate": 1.3954887218045114e-05, "loss": 0.31554681062698364, "num_input_tokens_seen": 1710062, "step": 609, "train_runtime": 2030.6372, "train_tokens_per_second": 842.131 }, { "epoch": 0.04901665360895156, "grad_norm": 3.253532648086548, "learning_rate": 1.3944862155388472e-05, "loss": 0.19874432682991028, "num_input_tokens_seen": 1713178, "step": 610, "train_runtime": 2033.3032, "train_tokens_per_second": 842.559 }, { "epoch": 0.049097008778802305, "grad_norm": 2.384195566177368, "learning_rate": 1.3934837092731829e-05, "loss": 0.2646555006504059, "num_input_tokens_seen": 1720132, "step": 611, "train_runtime": 2037.1263, "train_tokens_per_second": 844.391 }, { "epoch": 0.04917736394865305, "grad_norm": 2.53721284866333, "learning_rate": 1.392481203007519e-05, "loss": 0.22302576899528503, "num_input_tokens_seen": 1723848, "step": 612, "train_runtime": 2039.913, "train_tokens_per_second": 845.06 }, { "epoch": 0.04925771911850379, "grad_norm": 3.2206151485443115, "learning_rate": 1.3914786967418548e-05, "loss": 0.25008195638656616, "num_input_tokens_seen": 1730680, "step": 613, "train_runtime": 2043.5009, "train_tokens_per_second": 846.919 }, { "epoch": 0.04933807428835453, "grad_norm": 2.865894079208374, "learning_rate": 1.3904761904761905e-05, "loss": 0.34075915813446045, "num_input_tokens_seen": 1732318, "step": 614, "train_runtime": 2045.8154, "train_tokens_per_second": 846.762 }, { "epoch": 0.049418429458205265, "grad_norm": 2.801156520843506, "learning_rate": 1.3894736842105265e-05, "loss": 0.19296248257160187, "num_input_tokens_seen": 1733578, "step": 615, "train_runtime": 2047.9659, "train_tokens_per_second": 846.488 }, { "epoch": 0.04949878462805601, "grad_norm": 2.6612067222595215, "learning_rate": 1.3884711779448623e-05, "loss": 0.21209879219532013, "num_input_tokens_seen": 1735718, "step": 616, "train_runtime": 2050.3958, "train_tokens_per_second": 846.528 }, { "epoch": 0.04957913979790675, "grad_norm": 3.3813745975494385, "learning_rate": 1.387468671679198e-05, "loss": 0.2648746073246002, "num_input_tokens_seen": 1737268, "step": 617, "train_runtime": 2052.7077, "train_tokens_per_second": 846.33 }, { "epoch": 0.04965949496775749, "grad_norm": 3.608549118041992, "learning_rate": 1.3864661654135339e-05, "loss": 0.2968538701534271, "num_input_tokens_seen": 1739064, "step": 618, "train_runtime": 2055.0062, "train_tokens_per_second": 846.257 }, { "epoch": 0.049739850137608226, "grad_norm": 3.120656967163086, "learning_rate": 1.3854636591478699e-05, "loss": 0.3218134939670563, "num_input_tokens_seen": 1741882, "step": 619, "train_runtime": 2057.6368, "train_tokens_per_second": 846.545 }, { "epoch": 0.04982020530745897, "grad_norm": 2.4669291973114014, "learning_rate": 1.3844611528822056e-05, "loss": 0.2805180549621582, "num_input_tokens_seen": 1745514, "step": 620, "train_runtime": 2060.3914, "train_tokens_per_second": 847.176 }, { "epoch": 0.04990056047730971, "grad_norm": 4.341863632202148, "learning_rate": 1.3834586466165414e-05, "loss": 0.2129952609539032, "num_input_tokens_seen": 1747038, "step": 621, "train_runtime": 2062.6366, "train_tokens_per_second": 846.993 }, { "epoch": 0.04998091564716045, "grad_norm": 3.1166579723358154, "learning_rate": 1.3824561403508774e-05, "loss": 0.2970042824745178, "num_input_tokens_seen": 1749444, "step": 622, "train_runtime": 2065.106, "train_tokens_per_second": 847.145 }, { "epoch": 0.05006127081701119, "grad_norm": 2.9399631023406982, "learning_rate": 1.3814536340852131e-05, "loss": 0.3203200101852417, "num_input_tokens_seen": 1752924, "step": 623, "train_runtime": 2067.8302, "train_tokens_per_second": 847.712 }, { "epoch": 0.05014162598686193, "grad_norm": 2.739675760269165, "learning_rate": 1.380451127819549e-05, "loss": 0.2688198983669281, "num_input_tokens_seen": 1756136, "step": 624, "train_runtime": 2070.4856, "train_tokens_per_second": 848.176 }, { "epoch": 0.05022198115671267, "grad_norm": 3.0830225944519043, "learning_rate": 1.379448621553885e-05, "loss": 0.3489181101322174, "num_input_tokens_seen": 1759438, "step": 625, "train_runtime": 2073.1706, "train_tokens_per_second": 848.67 }, { "epoch": 0.05030233632656341, "grad_norm": 2.940826416015625, "learning_rate": 1.3784461152882206e-05, "loss": 0.2296193540096283, "num_input_tokens_seen": 1761142, "step": 626, "train_runtime": 2075.4463, "train_tokens_per_second": 848.561 }, { "epoch": 0.05038269149641415, "grad_norm": 4.391513824462891, "learning_rate": 1.3774436090225565e-05, "loss": 0.2284136712551117, "num_input_tokens_seen": 1762970, "step": 627, "train_runtime": 2077.7545, "train_tokens_per_second": 848.498 }, { "epoch": 0.05046304666626489, "grad_norm": 2.627239465713501, "learning_rate": 1.3764411027568922e-05, "loss": 0.24553167819976807, "num_input_tokens_seen": 1764678, "step": 628, "train_runtime": 2080.0625, "train_tokens_per_second": 848.377 }, { "epoch": 0.05054340183611563, "grad_norm": 3.0661940574645996, "learning_rate": 1.3754385964912282e-05, "loss": 0.28851979970932007, "num_input_tokens_seen": 1766990, "step": 629, "train_runtime": 2082.5293, "train_tokens_per_second": 848.483 }, { "epoch": 0.05062375700596637, "grad_norm": 2.551525831222534, "learning_rate": 1.374436090225564e-05, "loss": 0.31731733679771423, "num_input_tokens_seen": 1769930, "step": 630, "train_runtime": 2085.1296, "train_tokens_per_second": 848.835 }, { "epoch": 0.05070411217581711, "grad_norm": 3.30733060836792, "learning_rate": 1.3734335839598997e-05, "loss": 0.23765036463737488, "num_input_tokens_seen": 1773194, "step": 631, "train_runtime": 2099.9084, "train_tokens_per_second": 844.415 }, { "epoch": 0.050784467345667855, "grad_norm": 2.6271886825561523, "learning_rate": 1.3724310776942357e-05, "loss": 0.27127909660339355, "num_input_tokens_seen": 1774720, "step": 632, "train_runtime": 2102.1441, "train_tokens_per_second": 844.243 }, { "epoch": 0.05086482251551859, "grad_norm": 2.812711715698242, "learning_rate": 1.3714285714285716e-05, "loss": 0.33196860551834106, "num_input_tokens_seen": 1776740, "step": 633, "train_runtime": 2104.5133, "train_tokens_per_second": 844.252 }, { "epoch": 0.05094517768536933, "grad_norm": 3.027647018432617, "learning_rate": 1.3704260651629073e-05, "loss": 0.36906951665878296, "num_input_tokens_seen": 1781722, "step": 634, "train_runtime": 2107.661, "train_tokens_per_second": 845.355 }, { "epoch": 0.051025532855220074, "grad_norm": 2.9225051403045654, "learning_rate": 1.3694235588972433e-05, "loss": 0.32247769832611084, "num_input_tokens_seen": 1785088, "step": 635, "train_runtime": 2110.3822, "train_tokens_per_second": 845.86 }, { "epoch": 0.051105888025070816, "grad_norm": 2.510363817214966, "learning_rate": 1.3684210526315791e-05, "loss": 0.22477424144744873, "num_input_tokens_seen": 1789554, "step": 636, "train_runtime": 2113.3746, "train_tokens_per_second": 846.776 }, { "epoch": 0.05118624319492155, "grad_norm": 2.1739397048950195, "learning_rate": 1.3674185463659148e-05, "loss": 0.2022925317287445, "num_input_tokens_seen": 1792596, "step": 637, "train_runtime": 2116.0193, "train_tokens_per_second": 847.155 }, { "epoch": 0.05126659836477229, "grad_norm": 2.644911766052246, "learning_rate": 1.3664160401002507e-05, "loss": 0.2350763976573944, "num_input_tokens_seen": 1795494, "step": 638, "train_runtime": 2118.722, "train_tokens_per_second": 847.442 }, { "epoch": 0.051346953534623034, "grad_norm": 3.1598801612854004, "learning_rate": 1.3654135338345867e-05, "loss": 0.3328752815723419, "num_input_tokens_seen": 1798278, "step": 639, "train_runtime": 2121.282, "train_tokens_per_second": 847.732 }, { "epoch": 0.051427308704473776, "grad_norm": 3.557187080383301, "learning_rate": 1.3644110275689224e-05, "loss": 0.26554226875305176, "num_input_tokens_seen": 1800912, "step": 640, "train_runtime": 2123.8576, "train_tokens_per_second": 847.944 }, { "epoch": 0.05150766387432452, "grad_norm": 2.1795055866241455, "learning_rate": 1.3634085213032582e-05, "loss": 0.15878361463546753, "num_input_tokens_seen": 1804786, "step": 641, "train_runtime": 2126.6773, "train_tokens_per_second": 848.641 }, { "epoch": 0.05158801904417525, "grad_norm": 3.2361507415771484, "learning_rate": 1.3624060150375942e-05, "loss": 0.3802279531955719, "num_input_tokens_seen": 1808034, "step": 642, "train_runtime": 2129.5516, "train_tokens_per_second": 849.021 }, { "epoch": 0.051668374214025994, "grad_norm": 2.938929796218872, "learning_rate": 1.3614035087719299e-05, "loss": 0.2998857796192169, "num_input_tokens_seen": 1810516, "step": 643, "train_runtime": 2132.0698, "train_tokens_per_second": 849.182 }, { "epoch": 0.051748729383876736, "grad_norm": 2.978208303451538, "learning_rate": 1.3604010025062658e-05, "loss": 0.4231499433517456, "num_input_tokens_seen": 1814888, "step": 644, "train_runtime": 2135.0084, "train_tokens_per_second": 850.061 }, { "epoch": 0.05182908455372748, "grad_norm": 4.2538299560546875, "learning_rate": 1.3593984962406018e-05, "loss": 0.2943754196166992, "num_input_tokens_seen": 1820326, "step": 645, "train_runtime": 2138.235, "train_tokens_per_second": 851.322 }, { "epoch": 0.05190943972357821, "grad_norm": 3.5172996520996094, "learning_rate": 1.3583959899749374e-05, "loss": 0.31165650486946106, "num_input_tokens_seen": 1822132, "step": 646, "train_runtime": 2140.5577, "train_tokens_per_second": 851.242 }, { "epoch": 0.051989794893428955, "grad_norm": 2.9702513217926025, "learning_rate": 1.3573934837092733e-05, "loss": 0.2500317394733429, "num_input_tokens_seen": 1823578, "step": 647, "train_runtime": 2142.7561, "train_tokens_per_second": 851.043 }, { "epoch": 0.052070150063279697, "grad_norm": 3.51141357421875, "learning_rate": 1.356390977443609e-05, "loss": 0.32684528827667236, "num_input_tokens_seen": 1825182, "step": 648, "train_runtime": 2145.1161, "train_tokens_per_second": 850.855 }, { "epoch": 0.05215050523313044, "grad_norm": 3.6421401500701904, "learning_rate": 1.355388471177945e-05, "loss": 0.28116369247436523, "num_input_tokens_seen": 1826670, "step": 649, "train_runtime": 2147.3674, "train_tokens_per_second": 850.656 }, { "epoch": 0.05223086040298118, "grad_norm": 2.8052337169647217, "learning_rate": 1.3543859649122808e-05, "loss": 0.3427644371986389, "num_input_tokens_seen": 1828504, "step": 650, "train_runtime": 2149.7165, "train_tokens_per_second": 850.579 }, { "epoch": 0.052311215572831915, "grad_norm": 2.4777185916900635, "learning_rate": 1.3533834586466165e-05, "loss": 0.319729745388031, "num_input_tokens_seen": 1830492, "step": 651, "train_runtime": 2152.106, "train_tokens_per_second": 850.558 }, { "epoch": 0.05239157074268266, "grad_norm": 2.2239251136779785, "learning_rate": 1.3523809523809525e-05, "loss": 0.21376140415668488, "num_input_tokens_seen": 1833708, "step": 652, "train_runtime": 2154.8059, "train_tokens_per_second": 850.985 }, { "epoch": 0.0524719259125334, "grad_norm": 3.161543607711792, "learning_rate": 1.3513784461152884e-05, "loss": 0.3152637779712677, "num_input_tokens_seen": 1836482, "step": 653, "train_runtime": 2157.3778, "train_tokens_per_second": 851.257 }, { "epoch": 0.05255228108238414, "grad_norm": 3.664531946182251, "learning_rate": 1.350375939849624e-05, "loss": 0.2820356786251068, "num_input_tokens_seen": 1838800, "step": 654, "train_runtime": 2159.8401, "train_tokens_per_second": 851.359 }, { "epoch": 0.052632636252234875, "grad_norm": 2.8876495361328125, "learning_rate": 1.3493734335839601e-05, "loss": 0.2726529538631439, "num_input_tokens_seen": 1843808, "step": 655, "train_runtime": 2162.9771, "train_tokens_per_second": 852.44 }, { "epoch": 0.05271299142208562, "grad_norm": 2.637519359588623, "learning_rate": 1.348370927318296e-05, "loss": 0.2864427864551544, "num_input_tokens_seen": 1847714, "step": 656, "train_runtime": 2165.8269, "train_tokens_per_second": 853.122 }, { "epoch": 0.05279334659193636, "grad_norm": 3.108640432357788, "learning_rate": 1.3473684210526316e-05, "loss": 0.27912938594818115, "num_input_tokens_seen": 1849836, "step": 657, "train_runtime": 2168.269, "train_tokens_per_second": 853.14 }, { "epoch": 0.0528737017617871, "grad_norm": 2.71067214012146, "learning_rate": 1.3463659147869675e-05, "loss": 0.18870022892951965, "num_input_tokens_seen": 1852480, "step": 658, "train_runtime": 2170.776, "train_tokens_per_second": 853.372 }, { "epoch": 0.05295405693163784, "grad_norm": 3.19543194770813, "learning_rate": 1.3453634085213035e-05, "loss": 0.27085942029953003, "num_input_tokens_seen": 1855074, "step": 659, "train_runtime": 2173.2912, "train_tokens_per_second": 853.578 }, { "epoch": 0.05303441210148858, "grad_norm": 1.9311610460281372, "learning_rate": 1.3443609022556392e-05, "loss": 0.1847735345363617, "num_input_tokens_seen": 1858828, "step": 660, "train_runtime": 2176.1143, "train_tokens_per_second": 854.196 }, { "epoch": 0.05311476727133932, "grad_norm": 3.0223422050476074, "learning_rate": 1.343358395989975e-05, "loss": 0.35366198420524597, "num_input_tokens_seen": 1861174, "step": 661, "train_runtime": 2190.0994, "train_tokens_per_second": 849.813 }, { "epoch": 0.05319512244119006, "grad_norm": 3.538344383239746, "learning_rate": 1.342355889724311e-05, "loss": 0.2725224196910858, "num_input_tokens_seen": 1864990, "step": 662, "train_runtime": 2192.9586, "train_tokens_per_second": 850.445 }, { "epoch": 0.0532754776110408, "grad_norm": 2.75211238861084, "learning_rate": 1.3413533834586467e-05, "loss": 0.21504676342010498, "num_input_tokens_seen": 1868278, "step": 663, "train_runtime": 2195.6818, "train_tokens_per_second": 850.887 }, { "epoch": 0.05335583278089154, "grad_norm": 2.9910168647766113, "learning_rate": 1.3403508771929826e-05, "loss": 0.2507106363773346, "num_input_tokens_seen": 1872778, "step": 664, "train_runtime": 2198.6645, "train_tokens_per_second": 851.78 }, { "epoch": 0.05343618795074228, "grad_norm": 3.351558208465576, "learning_rate": 1.3393483709273186e-05, "loss": 0.2646216154098511, "num_input_tokens_seen": 1875038, "step": 665, "train_runtime": 2201.1236, "train_tokens_per_second": 851.855 }, { "epoch": 0.05351654312059302, "grad_norm": 3.5287859439849854, "learning_rate": 1.3383458646616543e-05, "loss": 0.3360460698604584, "num_input_tokens_seen": 1879922, "step": 666, "train_runtime": 2204.2924, "train_tokens_per_second": 852.846 }, { "epoch": 0.05359689829044376, "grad_norm": 2.708070755004883, "learning_rate": 1.3373433583959901e-05, "loss": 0.22727471590042114, "num_input_tokens_seen": 1882562, "step": 667, "train_runtime": 2206.8687, "train_tokens_per_second": 853.047 }, { "epoch": 0.053677253460294505, "grad_norm": 2.6637134552001953, "learning_rate": 1.3363408521303258e-05, "loss": 0.2622554302215576, "num_input_tokens_seen": 1885098, "step": 668, "train_runtime": 2209.4432, "train_tokens_per_second": 853.2 }, { "epoch": 0.05375760863014524, "grad_norm": 3.351876735687256, "learning_rate": 1.3353383458646618e-05, "loss": 0.30854332447052, "num_input_tokens_seen": 1887120, "step": 669, "train_runtime": 2211.8337, "train_tokens_per_second": 853.193 }, { "epoch": 0.05383796379999598, "grad_norm": 2.7834854125976562, "learning_rate": 1.3343358395989977e-05, "loss": 0.3723748028278351, "num_input_tokens_seen": 1890722, "step": 670, "train_runtime": 2214.6446, "train_tokens_per_second": 853.736 }, { "epoch": 0.053918318969846724, "grad_norm": 2.919356346130371, "learning_rate": 1.3333333333333333e-05, "loss": 0.19046790897846222, "num_input_tokens_seen": 1892162, "step": 671, "train_runtime": 2216.9084, "train_tokens_per_second": 853.514 }, { "epoch": 0.053998674139697465, "grad_norm": 3.359084367752075, "learning_rate": 1.3323308270676693e-05, "loss": 0.25587698817253113, "num_input_tokens_seen": 1894718, "step": 672, "train_runtime": 2219.3895, "train_tokens_per_second": 853.711 }, { "epoch": 0.0540790293095482, "grad_norm": 3.0106875896453857, "learning_rate": 1.3313283208020052e-05, "loss": 0.227071613073349, "num_input_tokens_seen": 1896454, "step": 673, "train_runtime": 2221.6625, "train_tokens_per_second": 853.619 }, { "epoch": 0.05415938447939894, "grad_norm": 3.170067548751831, "learning_rate": 1.3303258145363409e-05, "loss": 0.2881281077861786, "num_input_tokens_seen": 1898252, "step": 674, "train_runtime": 2223.9777, "train_tokens_per_second": 853.539 }, { "epoch": 0.054239739649249684, "grad_norm": 3.260258913040161, "learning_rate": 1.3293233082706769e-05, "loss": 0.2771885395050049, "num_input_tokens_seen": 1900172, "step": 675, "train_runtime": 2226.3989, "train_tokens_per_second": 853.473 }, { "epoch": 0.054320094819100426, "grad_norm": 2.3612301349639893, "learning_rate": 1.3283208020050127e-05, "loss": 0.25770312547683716, "num_input_tokens_seen": 1903126, "step": 676, "train_runtime": 2229.0402, "train_tokens_per_second": 853.787 }, { "epoch": 0.05440044998895117, "grad_norm": 3.1166040897369385, "learning_rate": 1.3273182957393484e-05, "loss": 0.3346172869205475, "num_input_tokens_seen": 1905518, "step": 677, "train_runtime": 2231.5935, "train_tokens_per_second": 853.882 }, { "epoch": 0.0544808051588019, "grad_norm": 3.45450496673584, "learning_rate": 1.3263157894736843e-05, "loss": 0.29421189427375793, "num_input_tokens_seen": 1908022, "step": 678, "train_runtime": 2234.122, "train_tokens_per_second": 854.037 }, { "epoch": 0.054561160328652644, "grad_norm": 2.934697151184082, "learning_rate": 1.3253132832080203e-05, "loss": 0.2912865877151489, "num_input_tokens_seen": 1911234, "step": 679, "train_runtime": 2236.8479, "train_tokens_per_second": 854.432 }, { "epoch": 0.054641515498503386, "grad_norm": 2.8125381469726562, "learning_rate": 1.324310776942356e-05, "loss": 0.23185798525810242, "num_input_tokens_seen": 1913438, "step": 680, "train_runtime": 2239.2664, "train_tokens_per_second": 854.493 }, { "epoch": 0.05472187066835413, "grad_norm": 2.69460129737854, "learning_rate": 1.3233082706766918e-05, "loss": 0.4150112569332123, "num_input_tokens_seen": 1916426, "step": 681, "train_runtime": 2241.8676, "train_tokens_per_second": 854.835 }, { "epoch": 0.05480222583820486, "grad_norm": 2.402989387512207, "learning_rate": 1.3223057644110278e-05, "loss": 0.28885847330093384, "num_input_tokens_seen": 1920296, "step": 682, "train_runtime": 2244.6557, "train_tokens_per_second": 855.497 }, { "epoch": 0.054882581008055605, "grad_norm": 2.586798667907715, "learning_rate": 1.3213032581453635e-05, "loss": 0.2565203011035919, "num_input_tokens_seen": 1922240, "step": 683, "train_runtime": 2247.0345, "train_tokens_per_second": 855.456 }, { "epoch": 0.054962936177906346, "grad_norm": 2.9696907997131348, "learning_rate": 1.3203007518796994e-05, "loss": 0.34296274185180664, "num_input_tokens_seen": 1924500, "step": 684, "train_runtime": 2249.5095, "train_tokens_per_second": 855.52 }, { "epoch": 0.05504329134775709, "grad_norm": 2.6322405338287354, "learning_rate": 1.3192982456140354e-05, "loss": 0.26834651827812195, "num_input_tokens_seen": 1927784, "step": 685, "train_runtime": 2252.2409, "train_tokens_per_second": 855.94 }, { "epoch": 0.05512364651760783, "grad_norm": 3.479985475540161, "learning_rate": 1.318295739348371e-05, "loss": 0.3691520690917969, "num_input_tokens_seen": 1934020, "step": 686, "train_runtime": 2255.824, "train_tokens_per_second": 857.345 }, { "epoch": 0.055204001687458565, "grad_norm": 2.857056140899658, "learning_rate": 1.3172932330827069e-05, "loss": 0.3064947724342346, "num_input_tokens_seen": 1936712, "step": 687, "train_runtime": 2258.4185, "train_tokens_per_second": 857.552 }, { "epoch": 0.05528435685730931, "grad_norm": 3.084240436553955, "learning_rate": 1.3162907268170426e-05, "loss": 0.34522297978401184, "num_input_tokens_seen": 1939710, "step": 688, "train_runtime": 2261.0134, "train_tokens_per_second": 857.894 }, { "epoch": 0.05536471202716005, "grad_norm": 3.5253946781158447, "learning_rate": 1.3152882205513786e-05, "loss": 0.26836851239204407, "num_input_tokens_seen": 1942828, "step": 689, "train_runtime": 2263.6682, "train_tokens_per_second": 858.265 }, { "epoch": 0.05544506719701079, "grad_norm": 2.921954870223999, "learning_rate": 1.3142857142857145e-05, "loss": 0.25162479281425476, "num_input_tokens_seen": 1945126, "step": 690, "train_runtime": 2266.1356, "train_tokens_per_second": 858.345 }, { "epoch": 0.055525422366861525, "grad_norm": 2.917914867401123, "learning_rate": 1.3132832080200501e-05, "loss": 0.31794431805610657, "num_input_tokens_seen": 1946844, "step": 691, "train_runtime": 2280.0542, "train_tokens_per_second": 853.859 }, { "epoch": 0.05560577753671227, "grad_norm": 2.4738786220550537, "learning_rate": 1.3122807017543862e-05, "loss": 0.26946914196014404, "num_input_tokens_seen": 1949554, "step": 692, "train_runtime": 2282.6544, "train_tokens_per_second": 854.073 }, { "epoch": 0.05568613270656301, "grad_norm": 3.40885329246521, "learning_rate": 1.311278195488722e-05, "loss": 0.3421843647956848, "num_input_tokens_seen": 1952036, "step": 693, "train_runtime": 2285.2532, "train_tokens_per_second": 854.188 }, { "epoch": 0.05576648787641375, "grad_norm": 3.214125871658325, "learning_rate": 1.3102756892230577e-05, "loss": 0.21185483038425446, "num_input_tokens_seen": 1955012, "step": 694, "train_runtime": 2287.8711, "train_tokens_per_second": 854.511 }, { "epoch": 0.05584684304626449, "grad_norm": 2.9773783683776855, "learning_rate": 1.3092731829573937e-05, "loss": 0.27962541580200195, "num_input_tokens_seen": 1957382, "step": 695, "train_runtime": 2290.354, "train_tokens_per_second": 854.62 }, { "epoch": 0.05592719821611523, "grad_norm": 3.8800296783447266, "learning_rate": 1.3082706766917295e-05, "loss": 0.3565332591533661, "num_input_tokens_seen": 1961586, "step": 696, "train_runtime": 2293.2921, "train_tokens_per_second": 855.358 }, { "epoch": 0.05600755338596597, "grad_norm": 2.1255645751953125, "learning_rate": 1.3072681704260652e-05, "loss": 0.20530018210411072, "num_input_tokens_seen": 1963672, "step": 697, "train_runtime": 2295.694, "train_tokens_per_second": 855.372 }, { "epoch": 0.05608790855581671, "grad_norm": 3.502772808074951, "learning_rate": 1.306265664160401e-05, "loss": 0.3294326961040497, "num_input_tokens_seen": 1965692, "step": 698, "train_runtime": 2298.1512, "train_tokens_per_second": 855.336 }, { "epoch": 0.05616826372566745, "grad_norm": 2.7191543579101562, "learning_rate": 1.305263157894737e-05, "loss": 0.21986062824726105, "num_input_tokens_seen": 1969126, "step": 699, "train_runtime": 2300.9922, "train_tokens_per_second": 855.773 }, { "epoch": 0.05624861889551819, "grad_norm": 2.9529592990875244, "learning_rate": 1.3042606516290728e-05, "loss": 0.2995322644710541, "num_input_tokens_seen": 1971102, "step": 700, "train_runtime": 2303.5126, "train_tokens_per_second": 855.694 }, { "epoch": 0.05632897406536893, "grad_norm": 2.3385748863220215, "learning_rate": 1.3032581453634086e-05, "loss": 0.20204994082450867, "num_input_tokens_seen": 1973002, "step": 701, "train_runtime": 2305.8584, "train_tokens_per_second": 855.647 }, { "epoch": 0.05640932923521967, "grad_norm": 3.070664167404175, "learning_rate": 1.3022556390977445e-05, "loss": 0.33333444595336914, "num_input_tokens_seen": 1975182, "step": 702, "train_runtime": 2308.3224, "train_tokens_per_second": 855.679 }, { "epoch": 0.05648968440507041, "grad_norm": 3.0550460815429688, "learning_rate": 1.3012531328320803e-05, "loss": 0.2802373766899109, "num_input_tokens_seen": 1978232, "step": 703, "train_runtime": 2310.9662, "train_tokens_per_second": 856.019 }, { "epoch": 0.056570039574921155, "grad_norm": 3.1352860927581787, "learning_rate": 1.3002506265664162e-05, "loss": 0.20022815465927124, "num_input_tokens_seen": 1982480, "step": 704, "train_runtime": 2313.9088, "train_tokens_per_second": 856.767 }, { "epoch": 0.05665039474477189, "grad_norm": 2.783033847808838, "learning_rate": 1.299248120300752e-05, "loss": 0.30796366930007935, "num_input_tokens_seen": 1985814, "step": 705, "train_runtime": 2316.5968, "train_tokens_per_second": 857.212 }, { "epoch": 0.05673074991462263, "grad_norm": 2.5930073261260986, "learning_rate": 1.2982456140350879e-05, "loss": 0.2709652781486511, "num_input_tokens_seen": 1988434, "step": 706, "train_runtime": 2319.1096, "train_tokens_per_second": 857.413 }, { "epoch": 0.05681110508447337, "grad_norm": 2.735032796859741, "learning_rate": 1.2972431077694237e-05, "loss": 0.30349603295326233, "num_input_tokens_seen": 1990426, "step": 707, "train_runtime": 2321.5142, "train_tokens_per_second": 857.383 }, { "epoch": 0.056891460254324115, "grad_norm": 2.7197277545928955, "learning_rate": 1.2962406015037594e-05, "loss": 0.29959139227867126, "num_input_tokens_seen": 1992628, "step": 708, "train_runtime": 2323.9176, "train_tokens_per_second": 857.443 }, { "epoch": 0.05697181542417485, "grad_norm": 3.2913830280303955, "learning_rate": 1.2952380952380954e-05, "loss": 0.26303839683532715, "num_input_tokens_seen": 1994768, "step": 709, "train_runtime": 2326.2908, "train_tokens_per_second": 857.489 }, { "epoch": 0.05705217059402559, "grad_norm": 2.6862263679504395, "learning_rate": 1.2942355889724313e-05, "loss": 0.26484471559524536, "num_input_tokens_seen": 1998140, "step": 710, "train_runtime": 2329.017, "train_tokens_per_second": 857.933 }, { "epoch": 0.057132525763876334, "grad_norm": 2.7618422508239746, "learning_rate": 1.293233082706767e-05, "loss": 0.36666426062583923, "num_input_tokens_seen": 2000496, "step": 711, "train_runtime": 2331.456, "train_tokens_per_second": 858.046 }, { "epoch": 0.057212880933727076, "grad_norm": 2.6680479049682617, "learning_rate": 1.292230576441103e-05, "loss": 0.2736525535583496, "num_input_tokens_seen": 2002786, "step": 712, "train_runtime": 2333.8661, "train_tokens_per_second": 858.141 }, { "epoch": 0.05729323610357782, "grad_norm": 3.9863438606262207, "learning_rate": 1.2912280701754386e-05, "loss": 0.34666991233825684, "num_input_tokens_seen": 2004966, "step": 713, "train_runtime": 2336.2705, "train_tokens_per_second": 858.191 }, { "epoch": 0.05737359127342855, "grad_norm": 3.1319849491119385, "learning_rate": 1.2902255639097745e-05, "loss": 0.219292551279068, "num_input_tokens_seen": 2006704, "step": 714, "train_runtime": 2338.5893, "train_tokens_per_second": 858.083 }, { "epoch": 0.057453946443279294, "grad_norm": 2.8223040103912354, "learning_rate": 1.2892230576441105e-05, "loss": 0.30268675088882446, "num_input_tokens_seen": 2011030, "step": 715, "train_runtime": 2341.5104, "train_tokens_per_second": 858.86 }, { "epoch": 0.057534301613130036, "grad_norm": 2.8956298828125, "learning_rate": 1.2882205513784462e-05, "loss": 0.2694479823112488, "num_input_tokens_seen": 2013824, "step": 716, "train_runtime": 2344.0726, "train_tokens_per_second": 859.113 }, { "epoch": 0.05761465678298078, "grad_norm": 2.994716167449951, "learning_rate": 1.287218045112782e-05, "loss": 0.30355149507522583, "num_input_tokens_seen": 2016036, "step": 717, "train_runtime": 2346.5115, "train_tokens_per_second": 859.163 }, { "epoch": 0.05769501195283151, "grad_norm": 3.2853922843933105, "learning_rate": 1.2862155388471179e-05, "loss": 0.332498162984848, "num_input_tokens_seen": 2018750, "step": 718, "train_runtime": 2349.0677, "train_tokens_per_second": 859.383 }, { "epoch": 0.057775367122682254, "grad_norm": 2.9221301078796387, "learning_rate": 1.2852130325814537e-05, "loss": 0.25970330834388733, "num_input_tokens_seen": 2022500, "step": 719, "train_runtime": 2351.8767, "train_tokens_per_second": 859.952 }, { "epoch": 0.057855722292532996, "grad_norm": 3.1856865882873535, "learning_rate": 1.2842105263157896e-05, "loss": 0.29233789443969727, "num_input_tokens_seen": 2025130, "step": 720, "train_runtime": 2354.3924, "train_tokens_per_second": 860.15 }, { "epoch": 0.05793607746238374, "grad_norm": 3.168943405151367, "learning_rate": 1.2832080200501254e-05, "loss": 0.2875809967517853, "num_input_tokens_seen": 2027530, "step": 721, "train_runtime": 2368.3638, "train_tokens_per_second": 856.089 }, { "epoch": 0.05801643263223448, "grad_norm": 2.473194122314453, "learning_rate": 1.2822055137844613e-05, "loss": 0.23746222257614136, "num_input_tokens_seen": 2029678, "step": 722, "train_runtime": 2370.8, "train_tokens_per_second": 856.115 }, { "epoch": 0.058096787802085215, "grad_norm": 3.199448823928833, "learning_rate": 1.2812030075187971e-05, "loss": 0.24388766288757324, "num_input_tokens_seen": 2031940, "step": 723, "train_runtime": 2373.2932, "train_tokens_per_second": 856.169 }, { "epoch": 0.058177142971935956, "grad_norm": 2.9941396713256836, "learning_rate": 1.2802005012531328e-05, "loss": 0.3648560345172882, "num_input_tokens_seen": 2034580, "step": 724, "train_runtime": 2375.8407, "train_tokens_per_second": 856.362 }, { "epoch": 0.0582574981417867, "grad_norm": 2.9326486587524414, "learning_rate": 1.2791979949874688e-05, "loss": 0.32754868268966675, "num_input_tokens_seen": 2038314, "step": 725, "train_runtime": 2378.7036, "train_tokens_per_second": 856.901 }, { "epoch": 0.05833785331163744, "grad_norm": 2.7603845596313477, "learning_rate": 1.2781954887218047e-05, "loss": 0.33794474601745605, "num_input_tokens_seen": 2041380, "step": 726, "train_runtime": 2381.3581, "train_tokens_per_second": 857.234 }, { "epoch": 0.058418208481488175, "grad_norm": 2.7896475791931152, "learning_rate": 1.2771929824561404e-05, "loss": 0.28347307443618774, "num_input_tokens_seen": 2043568, "step": 727, "train_runtime": 2383.7867, "train_tokens_per_second": 857.278 }, { "epoch": 0.05849856365133892, "grad_norm": 3.4421300888061523, "learning_rate": 1.2761904761904762e-05, "loss": 0.2643086016178131, "num_input_tokens_seen": 2046902, "step": 728, "train_runtime": 2386.5587, "train_tokens_per_second": 857.679 }, { "epoch": 0.05857891882118966, "grad_norm": 3.087042808532715, "learning_rate": 1.2751879699248122e-05, "loss": 0.2249894142150879, "num_input_tokens_seen": 2050022, "step": 729, "train_runtime": 2389.251, "train_tokens_per_second": 858.019 }, { "epoch": 0.0586592739910404, "grad_norm": 3.35132098197937, "learning_rate": 1.2741854636591479e-05, "loss": 0.3189818859100342, "num_input_tokens_seen": 2054130, "step": 730, "train_runtime": 2392.191, "train_tokens_per_second": 858.681 }, { "epoch": 0.05873962916089114, "grad_norm": 2.5226151943206787, "learning_rate": 1.2731829573934837e-05, "loss": 0.2420227974653244, "num_input_tokens_seen": 2056002, "step": 731, "train_runtime": 2394.5995, "train_tokens_per_second": 858.6 }, { "epoch": 0.05881998433074188, "grad_norm": 3.000455379486084, "learning_rate": 1.2721804511278198e-05, "loss": 0.313071608543396, "num_input_tokens_seen": 2058612, "step": 732, "train_runtime": 2397.1684, "train_tokens_per_second": 858.768 }, { "epoch": 0.05890033950059262, "grad_norm": 3.360764265060425, "learning_rate": 1.2711779448621554e-05, "loss": 0.32356250286102295, "num_input_tokens_seen": 2062730, "step": 733, "train_runtime": 2400.0825, "train_tokens_per_second": 859.441 }, { "epoch": 0.05898069467044336, "grad_norm": 3.4393975734710693, "learning_rate": 1.2701754385964913e-05, "loss": 0.3170442283153534, "num_input_tokens_seen": 2065054, "step": 734, "train_runtime": 2402.5488, "train_tokens_per_second": 859.526 }, { "epoch": 0.0590610498402941, "grad_norm": 3.1318342685699463, "learning_rate": 1.2691729323308273e-05, "loss": 0.352206289768219, "num_input_tokens_seen": 2067152, "step": 735, "train_runtime": 2404.9524, "train_tokens_per_second": 859.54 }, { "epoch": 0.05914140501014484, "grad_norm": 2.8122692108154297, "learning_rate": 1.268170426065163e-05, "loss": 0.25028109550476074, "num_input_tokens_seen": 2068852, "step": 736, "train_runtime": 2407.2423, "train_tokens_per_second": 859.428 }, { "epoch": 0.05922176017999558, "grad_norm": 2.837824583053589, "learning_rate": 1.2671679197994988e-05, "loss": 0.26481619477272034, "num_input_tokens_seen": 2071602, "step": 737, "train_runtime": 2409.7843, "train_tokens_per_second": 859.663 }, { "epoch": 0.05930211534984632, "grad_norm": 3.4425995349884033, "learning_rate": 1.2661654135338345e-05, "loss": 0.3492182791233063, "num_input_tokens_seen": 2074314, "step": 738, "train_runtime": 2412.3045, "train_tokens_per_second": 859.889 }, { "epoch": 0.05938247051969706, "grad_norm": 2.9108664989471436, "learning_rate": 1.2651629072681705e-05, "loss": 0.2601884603500366, "num_input_tokens_seen": 2078108, "step": 739, "train_runtime": 2415.1446, "train_tokens_per_second": 860.449 }, { "epoch": 0.059462825689547805, "grad_norm": 2.6832640171051025, "learning_rate": 1.2641604010025064e-05, "loss": 0.26210784912109375, "num_input_tokens_seen": 2080674, "step": 740, "train_runtime": 2417.6684, "train_tokens_per_second": 860.612 }, { "epoch": 0.05954318085939854, "grad_norm": 3.2683944702148438, "learning_rate": 1.263157894736842e-05, "loss": 0.330169677734375, "num_input_tokens_seen": 2083142, "step": 741, "train_runtime": 2420.1576, "train_tokens_per_second": 860.746 }, { "epoch": 0.05962353602924928, "grad_norm": 2.8443398475646973, "learning_rate": 1.262155388471178e-05, "loss": 0.275868684053421, "num_input_tokens_seen": 2087506, "step": 742, "train_runtime": 2423.1063, "train_tokens_per_second": 861.5 }, { "epoch": 0.05970389119910002, "grad_norm": 2.9205875396728516, "learning_rate": 1.261152882205514e-05, "loss": 0.24210676550865173, "num_input_tokens_seen": 2090052, "step": 743, "train_runtime": 2425.668, "train_tokens_per_second": 861.64 }, { "epoch": 0.059784246368950765, "grad_norm": 2.823946237564087, "learning_rate": 1.2601503759398496e-05, "loss": 0.2695673704147339, "num_input_tokens_seen": 2092922, "step": 744, "train_runtime": 2428.3038, "train_tokens_per_second": 861.886 }, { "epoch": 0.0598646015388015, "grad_norm": 3.1429476737976074, "learning_rate": 1.2591478696741856e-05, "loss": 0.3264291286468506, "num_input_tokens_seen": 2094718, "step": 745, "train_runtime": 2430.7115, "train_tokens_per_second": 861.772 }, { "epoch": 0.05994495670865224, "grad_norm": 3.067852020263672, "learning_rate": 1.2581453634085215e-05, "loss": 0.22391746938228607, "num_input_tokens_seen": 2096236, "step": 746, "train_runtime": 2433.048, "train_tokens_per_second": 861.568 }, { "epoch": 0.060025311878502983, "grad_norm": 2.8143184185028076, "learning_rate": 1.2571428571428572e-05, "loss": 0.24153998494148254, "num_input_tokens_seen": 2098838, "step": 747, "train_runtime": 2435.6764, "train_tokens_per_second": 861.706 }, { "epoch": 0.060105667048353725, "grad_norm": 2.3908045291900635, "learning_rate": 1.256140350877193e-05, "loss": 0.2455410659313202, "num_input_tokens_seen": 2101132, "step": 748, "train_runtime": 2438.1221, "train_tokens_per_second": 861.783 }, { "epoch": 0.06018602221820447, "grad_norm": 2.4056248664855957, "learning_rate": 1.255137844611529e-05, "loss": 0.20279547572135925, "num_input_tokens_seen": 2107006, "step": 749, "train_runtime": 2441.498, "train_tokens_per_second": 862.997 }, { "epoch": 0.0602663773880552, "grad_norm": 3.636943817138672, "learning_rate": 1.2541353383458647e-05, "loss": 0.3395722210407257, "num_input_tokens_seen": 2110764, "step": 750, "train_runtime": 2444.3084, "train_tokens_per_second": 863.542 }, { "epoch": 0.060346732557905944, "grad_norm": 2.799332857131958, "learning_rate": 1.2531328320802006e-05, "loss": 0.29942968487739563, "num_input_tokens_seen": 2113552, "step": 751, "train_runtime": 2458.5634, "train_tokens_per_second": 859.67 }, { "epoch": 0.060427087727756686, "grad_norm": 2.3057260513305664, "learning_rate": 1.2521303258145366e-05, "loss": 0.21573303639888763, "num_input_tokens_seen": 2116872, "step": 752, "train_runtime": 2461.3105, "train_tokens_per_second": 860.059 }, { "epoch": 0.06050744289760743, "grad_norm": 2.782414674758911, "learning_rate": 1.2511278195488723e-05, "loss": 0.25436198711395264, "num_input_tokens_seen": 2119440, "step": 753, "train_runtime": 2463.8296, "train_tokens_per_second": 860.222 }, { "epoch": 0.06058779806745816, "grad_norm": 3.568317413330078, "learning_rate": 1.2501253132832081e-05, "loss": 0.3253176212310791, "num_input_tokens_seen": 2122124, "step": 754, "train_runtime": 2466.3807, "train_tokens_per_second": 860.42 }, { "epoch": 0.060668153237308904, "grad_norm": 2.734286069869995, "learning_rate": 1.2491228070175441e-05, "loss": 0.2492453008890152, "num_input_tokens_seen": 2126464, "step": 755, "train_runtime": 2469.3623, "train_tokens_per_second": 861.139 }, { "epoch": 0.060748508407159646, "grad_norm": 3.554206371307373, "learning_rate": 1.2481203007518798e-05, "loss": 0.2971659004688263, "num_input_tokens_seen": 2132378, "step": 756, "train_runtime": 2472.8221, "train_tokens_per_second": 862.326 }, { "epoch": 0.06082886357701039, "grad_norm": 2.579406261444092, "learning_rate": 1.2471177944862156e-05, "loss": 0.1930014044046402, "num_input_tokens_seen": 2134070, "step": 757, "train_runtime": 2475.1451, "train_tokens_per_second": 862.2 }, { "epoch": 0.06090921874686113, "grad_norm": 2.7121148109436035, "learning_rate": 1.2461152882205513e-05, "loss": 0.23829159140586853, "num_input_tokens_seen": 2136364, "step": 758, "train_runtime": 2477.5946, "train_tokens_per_second": 862.273 }, { "epoch": 0.060989573916711864, "grad_norm": 2.152038335800171, "learning_rate": 1.2451127819548873e-05, "loss": 0.21540594100952148, "num_input_tokens_seen": 2138950, "step": 759, "train_runtime": 2480.1271, "train_tokens_per_second": 862.436 }, { "epoch": 0.061069929086562606, "grad_norm": 2.7821602821350098, "learning_rate": 1.2441102756892232e-05, "loss": 0.2620702087879181, "num_input_tokens_seen": 2141456, "step": 760, "train_runtime": 2482.6073, "train_tokens_per_second": 862.583 }, { "epoch": 0.06115028425641335, "grad_norm": 3.0144550800323486, "learning_rate": 1.2431077694235589e-05, "loss": 0.3187815546989441, "num_input_tokens_seen": 2143180, "step": 761, "train_runtime": 2485.018, "train_tokens_per_second": 862.44 }, { "epoch": 0.06123063942626409, "grad_norm": 2.9687743186950684, "learning_rate": 1.2421052631578949e-05, "loss": 0.2873280644416809, "num_input_tokens_seen": 2146560, "step": 762, "train_runtime": 2487.8305, "train_tokens_per_second": 862.824 }, { "epoch": 0.061310994596114825, "grad_norm": 2.583993911743164, "learning_rate": 1.2411027568922307e-05, "loss": 0.29108160734176636, "num_input_tokens_seen": 2150406, "step": 763, "train_runtime": 2490.67, "train_tokens_per_second": 863.385 }, { "epoch": 0.06139134976596557, "grad_norm": 3.725745677947998, "learning_rate": 1.2401002506265664e-05, "loss": 0.36730512976646423, "num_input_tokens_seen": 2152674, "step": 764, "train_runtime": 2493.0849, "train_tokens_per_second": 863.458 }, { "epoch": 0.06147170493581631, "grad_norm": 2.7844672203063965, "learning_rate": 1.2390977443609024e-05, "loss": 0.30903497338294983, "num_input_tokens_seen": 2155570, "step": 765, "train_runtime": 2495.6751, "train_tokens_per_second": 863.722 }, { "epoch": 0.06155206010566705, "grad_norm": 2.7151501178741455, "learning_rate": 1.2380952380952383e-05, "loss": 0.23024988174438477, "num_input_tokens_seen": 2158284, "step": 766, "train_runtime": 2498.2331, "train_tokens_per_second": 863.924 }, { "epoch": 0.06163241527551779, "grad_norm": 2.610064744949341, "learning_rate": 1.237092731829574e-05, "loss": 0.19674576818943024, "num_input_tokens_seen": 2160528, "step": 767, "train_runtime": 2500.6846, "train_tokens_per_second": 863.975 }, { "epoch": 0.06171277044536853, "grad_norm": 3.0285184383392334, "learning_rate": 1.2360902255639098e-05, "loss": 0.26799285411834717, "num_input_tokens_seen": 2163062, "step": 768, "train_runtime": 2503.1735, "train_tokens_per_second": 864.128 }, { "epoch": 0.06179312561521927, "grad_norm": 3.4095847606658936, "learning_rate": 1.2350877192982458e-05, "loss": 0.2050393521785736, "num_input_tokens_seen": 2166096, "step": 769, "train_runtime": 2505.7871, "train_tokens_per_second": 864.437 }, { "epoch": 0.06187348078507001, "grad_norm": 2.8052728176116943, "learning_rate": 1.2340852130325815e-05, "loss": 0.35179954767227173, "num_input_tokens_seen": 2167602, "step": 770, "train_runtime": 2508.0183, "train_tokens_per_second": 864.269 }, { "epoch": 0.06195383595492075, "grad_norm": 3.4134974479675293, "learning_rate": 1.2330827067669174e-05, "loss": 0.21670857071876526, "num_input_tokens_seen": 2169358, "step": 771, "train_runtime": 2510.3281, "train_tokens_per_second": 864.173 }, { "epoch": 0.06203419112477149, "grad_norm": 2.407679319381714, "learning_rate": 1.2320802005012534e-05, "loss": 0.23209697008132935, "num_input_tokens_seen": 2173620, "step": 772, "train_runtime": 2513.2555, "train_tokens_per_second": 864.862 }, { "epoch": 0.06211454629462223, "grad_norm": 3.3431007862091064, "learning_rate": 1.231077694235589e-05, "loss": 0.30953449010849, "num_input_tokens_seen": 2175542, "step": 773, "train_runtime": 2515.6133, "train_tokens_per_second": 864.816 }, { "epoch": 0.06219490146447297, "grad_norm": 2.2891623973846436, "learning_rate": 1.2300751879699249e-05, "loss": 0.18018221855163574, "num_input_tokens_seen": 2181638, "step": 774, "train_runtime": 2519.0839, "train_tokens_per_second": 866.044 }, { "epoch": 0.06227525663432371, "grad_norm": 2.7475595474243164, "learning_rate": 1.229072681704261e-05, "loss": 0.2909543812274933, "num_input_tokens_seen": 2184276, "step": 775, "train_runtime": 2521.6467, "train_tokens_per_second": 866.21 }, { "epoch": 0.062355611804174454, "grad_norm": 2.765322208404541, "learning_rate": 1.2280701754385966e-05, "loss": 0.2712576985359192, "num_input_tokens_seen": 2186672, "step": 776, "train_runtime": 2524.1173, "train_tokens_per_second": 866.312 }, { "epoch": 0.06243596697402519, "grad_norm": 3.0189335346221924, "learning_rate": 1.2270676691729325e-05, "loss": 0.3496021330356598, "num_input_tokens_seen": 2188374, "step": 777, "train_runtime": 2526.4128, "train_tokens_per_second": 866.198 }, { "epoch": 0.06251632214387594, "grad_norm": 3.2362682819366455, "learning_rate": 1.2260651629072681e-05, "loss": 0.25949907302856445, "num_input_tokens_seen": 2192396, "step": 778, "train_runtime": 2529.3147, "train_tokens_per_second": 866.794 }, { "epoch": 0.06259667731372667, "grad_norm": 2.8941972255706787, "learning_rate": 1.2250626566416041e-05, "loss": 0.33396419882774353, "num_input_tokens_seen": 2194086, "step": 779, "train_runtime": 2531.6388, "train_tokens_per_second": 866.666 }, { "epoch": 0.06267703248357741, "grad_norm": 3.1877472400665283, "learning_rate": 1.22406015037594e-05, "loss": 0.3487219214439392, "num_input_tokens_seen": 2196372, "step": 780, "train_runtime": 2534.0947, "train_tokens_per_second": 866.728 }, { "epoch": 0.06275738765342816, "grad_norm": 2.8576133251190186, "learning_rate": 1.2230576441102757e-05, "loss": 0.26556044816970825, "num_input_tokens_seen": 2199122, "step": 781, "train_runtime": 2548.1838, "train_tokens_per_second": 863.015 }, { "epoch": 0.06283774282327889, "grad_norm": 3.0474417209625244, "learning_rate": 1.2220551378446117e-05, "loss": 0.32222989201545715, "num_input_tokens_seen": 2202668, "step": 782, "train_runtime": 2550.9341, "train_tokens_per_second": 863.475 }, { "epoch": 0.06291809799312963, "grad_norm": 2.0292720794677734, "learning_rate": 1.2210526315789475e-05, "loss": 0.25332799553871155, "num_input_tokens_seen": 2207240, "step": 783, "train_runtime": 2553.9956, "train_tokens_per_second": 864.23 }, { "epoch": 0.06299845316298038, "grad_norm": 2.600461483001709, "learning_rate": 1.2200501253132832e-05, "loss": 0.2454194724559784, "num_input_tokens_seen": 2210724, "step": 784, "train_runtime": 2556.7765, "train_tokens_per_second": 864.653 }, { "epoch": 0.06307880833283111, "grad_norm": 3.5251662731170654, "learning_rate": 1.2190476190476192e-05, "loss": 0.3073396384716034, "num_input_tokens_seen": 2212986, "step": 785, "train_runtime": 2559.2261, "train_tokens_per_second": 864.709 }, { "epoch": 0.06315916350268186, "grad_norm": 2.7284512519836426, "learning_rate": 1.2180451127819551e-05, "loss": 0.2464675009250641, "num_input_tokens_seen": 2215062, "step": 786, "train_runtime": 2561.6119, "train_tokens_per_second": 864.714 }, { "epoch": 0.0632395186725326, "grad_norm": 3.402259111404419, "learning_rate": 1.2170426065162908e-05, "loss": 0.25732147693634033, "num_input_tokens_seen": 2218322, "step": 787, "train_runtime": 2564.2899, "train_tokens_per_second": 865.082 }, { "epoch": 0.06331987384238333, "grad_norm": 2.6554465293884277, "learning_rate": 1.2160401002506266e-05, "loss": 0.32181304693222046, "num_input_tokens_seen": 2222348, "step": 788, "train_runtime": 2567.1306, "train_tokens_per_second": 865.693 }, { "epoch": 0.06340022901223408, "grad_norm": 2.8244717121124268, "learning_rate": 1.2150375939849626e-05, "loss": 0.27541273832321167, "num_input_tokens_seen": 2224960, "step": 789, "train_runtime": 2569.6559, "train_tokens_per_second": 865.859 }, { "epoch": 0.06348058418208481, "grad_norm": 3.246001720428467, "learning_rate": 1.2140350877192983e-05, "loss": 0.2977929711341858, "num_input_tokens_seen": 2227156, "step": 790, "train_runtime": 2572.0612, "train_tokens_per_second": 865.903 }, { "epoch": 0.06356093935193556, "grad_norm": 3.336674213409424, "learning_rate": 1.2130325814536342e-05, "loss": 0.3430129587650299, "num_input_tokens_seen": 2229860, "step": 791, "train_runtime": 2574.6404, "train_tokens_per_second": 866.086 }, { "epoch": 0.0636412945217863, "grad_norm": 3.276735305786133, "learning_rate": 1.2120300751879702e-05, "loss": 0.35523948073387146, "num_input_tokens_seen": 2232812, "step": 792, "train_runtime": 2577.2376, "train_tokens_per_second": 866.359 }, { "epoch": 0.06372164969163703, "grad_norm": 2.5701100826263428, "learning_rate": 1.2110275689223059e-05, "loss": 0.34134089946746826, "num_input_tokens_seen": 2234462, "step": 793, "train_runtime": 2579.5223, "train_tokens_per_second": 866.231 }, { "epoch": 0.06380200486148778, "grad_norm": 2.437366247177124, "learning_rate": 1.2100250626566417e-05, "loss": 0.2768482565879822, "num_input_tokens_seen": 2236186, "step": 794, "train_runtime": 2581.8174, "train_tokens_per_second": 866.129 }, { "epoch": 0.06388236003133851, "grad_norm": 2.999232053756714, "learning_rate": 1.2090225563909777e-05, "loss": 0.31885021924972534, "num_input_tokens_seen": 2239030, "step": 795, "train_runtime": 2584.4176, "train_tokens_per_second": 866.358 }, { "epoch": 0.06396271520118926, "grad_norm": 2.989140510559082, "learning_rate": 1.2080200501253134e-05, "loss": 0.2512443959712982, "num_input_tokens_seen": 2241616, "step": 796, "train_runtime": 2586.9429, "train_tokens_per_second": 866.512 }, { "epoch": 0.06404307037104, "grad_norm": 2.7428035736083984, "learning_rate": 1.2070175438596493e-05, "loss": 0.26089048385620117, "num_input_tokens_seen": 2243362, "step": 797, "train_runtime": 2589.2923, "train_tokens_per_second": 866.4 }, { "epoch": 0.06412342554089073, "grad_norm": 3.046943187713623, "learning_rate": 1.206015037593985e-05, "loss": 0.23675644397735596, "num_input_tokens_seen": 2244900, "step": 798, "train_runtime": 2591.6095, "train_tokens_per_second": 866.218 }, { "epoch": 0.06420378071074148, "grad_norm": 2.5427329540252686, "learning_rate": 1.205012531328321e-05, "loss": 0.30876022577285767, "num_input_tokens_seen": 2251908, "step": 799, "train_runtime": 2595.3081, "train_tokens_per_second": 867.684 }, { "epoch": 0.06428413588059222, "grad_norm": 3.0693864822387695, "learning_rate": 1.2040100250626568e-05, "loss": 0.25820907950401306, "num_input_tokens_seen": 2256274, "step": 800, "train_runtime": 2598.3651, "train_tokens_per_second": 868.344 }, { "epoch": 0.06428413588059222, "eval_loss": 1.1370564699172974, "eval_runtime": 68.3523, "eval_samples_per_second": 14.542, "eval_steps_per_second": 7.271, "num_input_tokens_seen": 2256274, "step": 800 }, { "epoch": 0.06436449105044295, "grad_norm": 3.2291626930236816, "learning_rate": 1.2030075187969925e-05, "loss": 0.2551375925540924, "num_input_tokens_seen": 2259448, "step": 801, "train_runtime": 2669.3607, "train_tokens_per_second": 846.438 }, { "epoch": 0.0644448462202937, "grad_norm": 2.9300308227539062, "learning_rate": 1.2020050125313285e-05, "loss": 0.27280884981155396, "num_input_tokens_seen": 2261828, "step": 802, "train_runtime": 2671.7898, "train_tokens_per_second": 846.559 }, { "epoch": 0.06452520139014443, "grad_norm": 3.3189568519592285, "learning_rate": 1.2010025062656644e-05, "loss": 0.34705424308776855, "num_input_tokens_seen": 2263514, "step": 803, "train_runtime": 2674.0877, "train_tokens_per_second": 846.462 }, { "epoch": 0.06460555655999518, "grad_norm": 3.697053909301758, "learning_rate": 1.2e-05, "loss": 0.3187881112098694, "num_input_tokens_seen": 2266586, "step": 804, "train_runtime": 2676.6946, "train_tokens_per_second": 846.785 }, { "epoch": 0.06468591172984592, "grad_norm": 2.109485149383545, "learning_rate": 1.1989974937343359e-05, "loss": 0.2484457790851593, "num_input_tokens_seen": 2269446, "step": 805, "train_runtime": 2679.2586, "train_tokens_per_second": 847.043 }, { "epoch": 0.06476626689969665, "grad_norm": 2.90828800201416, "learning_rate": 1.1979949874686719e-05, "loss": 0.3012159764766693, "num_input_tokens_seen": 2272596, "step": 806, "train_runtime": 2681.9151, "train_tokens_per_second": 847.378 }, { "epoch": 0.0648466220695474, "grad_norm": 3.402026414871216, "learning_rate": 1.1969924812030076e-05, "loss": 0.3304890990257263, "num_input_tokens_seen": 2276194, "step": 807, "train_runtime": 2684.6616, "train_tokens_per_second": 847.851 }, { "epoch": 0.06492697723939814, "grad_norm": 1.9149166345596313, "learning_rate": 1.1959899749373434e-05, "loss": 0.15809530019760132, "num_input_tokens_seen": 2278282, "step": 808, "train_runtime": 2687.0608, "train_tokens_per_second": 847.871 }, { "epoch": 0.06500733240924889, "grad_norm": 2.186581611633301, "learning_rate": 1.1949874686716794e-05, "loss": 0.22400712966918945, "num_input_tokens_seen": 2280792, "step": 809, "train_runtime": 2689.5611, "train_tokens_per_second": 848.016 }, { "epoch": 0.06508768757909962, "grad_norm": 3.1930742263793945, "learning_rate": 1.1939849624060151e-05, "loss": 0.3012041449546814, "num_input_tokens_seen": 2282948, "step": 810, "train_runtime": 2691.9705, "train_tokens_per_second": 848.058 }, { "epoch": 0.06516804274895036, "grad_norm": 2.901463031768799, "learning_rate": 1.192982456140351e-05, "loss": 0.2528325319290161, "num_input_tokens_seen": 2285570, "step": 811, "train_runtime": 2706.0401, "train_tokens_per_second": 844.618 }, { "epoch": 0.0652483979188011, "grad_norm": 2.5245742797851562, "learning_rate": 1.1919799498746868e-05, "loss": 0.19094139337539673, "num_input_tokens_seen": 2287570, "step": 812, "train_runtime": 2708.4193, "train_tokens_per_second": 844.614 }, { "epoch": 0.06532875308865184, "grad_norm": 3.0036656856536865, "learning_rate": 1.1909774436090227e-05, "loss": 0.2308330088853836, "num_input_tokens_seen": 2289566, "step": 813, "train_runtime": 2710.7571, "train_tokens_per_second": 844.622 }, { "epoch": 0.06540910825850259, "grad_norm": 3.3483850955963135, "learning_rate": 1.1899749373433585e-05, "loss": 0.3905947506427765, "num_input_tokens_seen": 2291628, "step": 814, "train_runtime": 2713.135, "train_tokens_per_second": 844.642 }, { "epoch": 0.06548946342835332, "grad_norm": 3.3075528144836426, "learning_rate": 1.1889724310776942e-05, "loss": 0.28439873456954956, "num_input_tokens_seen": 2293440, "step": 815, "train_runtime": 2715.447, "train_tokens_per_second": 844.59 }, { "epoch": 0.06556981859820406, "grad_norm": 3.069700002670288, "learning_rate": 1.1879699248120302e-05, "loss": 0.29112881422042847, "num_input_tokens_seen": 2296128, "step": 816, "train_runtime": 2718.0366, "train_tokens_per_second": 844.774 }, { "epoch": 0.0656501737680548, "grad_norm": 2.810053825378418, "learning_rate": 1.186967418546366e-05, "loss": 0.296118825674057, "num_input_tokens_seen": 2298476, "step": 817, "train_runtime": 2720.5291, "train_tokens_per_second": 844.864 }, { "epoch": 0.06573052893790554, "grad_norm": 3.226889133453369, "learning_rate": 1.1859649122807017e-05, "loss": 0.3280177414417267, "num_input_tokens_seen": 2301454, "step": 818, "train_runtime": 2723.2226, "train_tokens_per_second": 845.122 }, { "epoch": 0.06581088410775628, "grad_norm": 3.180469274520874, "learning_rate": 1.1849624060150378e-05, "loss": 0.3562939763069153, "num_input_tokens_seen": 2304936, "step": 819, "train_runtime": 2726.0189, "train_tokens_per_second": 845.532 }, { "epoch": 0.06589123927760702, "grad_norm": 3.3186299800872803, "learning_rate": 1.1839598997493736e-05, "loss": 0.25684499740600586, "num_input_tokens_seen": 2308426, "step": 820, "train_runtime": 2728.7066, "train_tokens_per_second": 845.978 }, { "epoch": 0.06597159444745776, "grad_norm": 3.414411783218384, "learning_rate": 1.1829573934837093e-05, "loss": 0.3466445803642273, "num_input_tokens_seen": 2310388, "step": 821, "train_runtime": 2731.0802, "train_tokens_per_second": 845.961 }, { "epoch": 0.06605194961730851, "grad_norm": 2.922644853591919, "learning_rate": 1.1819548872180453e-05, "loss": 0.22471588850021362, "num_input_tokens_seen": 2312722, "step": 822, "train_runtime": 2733.5514, "train_tokens_per_second": 846.05 }, { "epoch": 0.06613230478715924, "grad_norm": 3.0060935020446777, "learning_rate": 1.180952380952381e-05, "loss": 0.3773375153541565, "num_input_tokens_seen": 2318406, "step": 823, "train_runtime": 2736.8599, "train_tokens_per_second": 847.104 }, { "epoch": 0.06621265995700998, "grad_norm": 3.2040343284606934, "learning_rate": 1.1799498746867168e-05, "loss": 0.24648547172546387, "num_input_tokens_seen": 2322552, "step": 824, "train_runtime": 2739.7442, "train_tokens_per_second": 847.726 }, { "epoch": 0.06629301512686073, "grad_norm": 3.1924192905426025, "learning_rate": 1.1789473684210527e-05, "loss": 0.38184764981269836, "num_input_tokens_seen": 2324106, "step": 825, "train_runtime": 2741.9931, "train_tokens_per_second": 847.597 }, { "epoch": 0.06637337029671146, "grad_norm": 2.8940579891204834, "learning_rate": 1.1779448621553885e-05, "loss": 0.24251845479011536, "num_input_tokens_seen": 2326672, "step": 826, "train_runtime": 2744.5234, "train_tokens_per_second": 847.751 }, { "epoch": 0.06645372546656221, "grad_norm": 3.8691294193267822, "learning_rate": 1.1769423558897244e-05, "loss": 0.33357876539230347, "num_input_tokens_seen": 2333836, "step": 827, "train_runtime": 2748.2206, "train_tokens_per_second": 849.217 }, { "epoch": 0.06653408063641295, "grad_norm": 2.9479005336761475, "learning_rate": 1.1759398496240602e-05, "loss": 0.365050733089447, "num_input_tokens_seen": 2336590, "step": 828, "train_runtime": 2750.7864, "train_tokens_per_second": 849.426 }, { "epoch": 0.06661443580626368, "grad_norm": 2.0450966358184814, "learning_rate": 1.174937343358396e-05, "loss": 0.1982790231704712, "num_input_tokens_seen": 2339490, "step": 829, "train_runtime": 2753.381, "train_tokens_per_second": 849.679 }, { "epoch": 0.06669479097611443, "grad_norm": 2.5706307888031006, "learning_rate": 1.173934837092732e-05, "loss": 0.31383752822875977, "num_input_tokens_seen": 2342592, "step": 830, "train_runtime": 2756.1298, "train_tokens_per_second": 849.957 }, { "epoch": 0.06677514614596516, "grad_norm": 2.5175678730010986, "learning_rate": 1.1729323308270678e-05, "loss": 0.23741817474365234, "num_input_tokens_seen": 2344956, "step": 831, "train_runtime": 2758.6228, "train_tokens_per_second": 850.046 }, { "epoch": 0.06685550131581591, "grad_norm": 2.140591621398926, "learning_rate": 1.1719298245614036e-05, "loss": 0.19060128927230835, "num_input_tokens_seen": 2347736, "step": 832, "train_runtime": 2761.2243, "train_tokens_per_second": 850.252 }, { "epoch": 0.06693585648566665, "grad_norm": 3.0951597690582275, "learning_rate": 1.1709273182957395e-05, "loss": 0.33035385608673096, "num_input_tokens_seen": 2350688, "step": 833, "train_runtime": 2763.8146, "train_tokens_per_second": 850.523 }, { "epoch": 0.06701621165551738, "grad_norm": 3.2192423343658447, "learning_rate": 1.1699248120300753e-05, "loss": 0.3007463216781616, "num_input_tokens_seen": 2354138, "step": 834, "train_runtime": 2766.5426, "train_tokens_per_second": 850.931 }, { "epoch": 0.06709656682536813, "grad_norm": 2.9447860717773438, "learning_rate": 1.168922305764411e-05, "loss": 0.3031856417655945, "num_input_tokens_seen": 2357974, "step": 835, "train_runtime": 2769.3849, "train_tokens_per_second": 851.443 }, { "epoch": 0.06717692199521887, "grad_norm": 3.6809303760528564, "learning_rate": 1.167919799498747e-05, "loss": 0.2824529707431793, "num_input_tokens_seen": 2360716, "step": 836, "train_runtime": 2771.9363, "train_tokens_per_second": 851.649 }, { "epoch": 0.0672572771650696, "grad_norm": 3.2521047592163086, "learning_rate": 1.1669172932330827e-05, "loss": 0.3381871283054352, "num_input_tokens_seen": 2362678, "step": 837, "train_runtime": 2774.2933, "train_tokens_per_second": 851.632 }, { "epoch": 0.06733763233492035, "grad_norm": 2.5861077308654785, "learning_rate": 1.1659147869674185e-05, "loss": 0.2722276747226715, "num_input_tokens_seen": 2364514, "step": 838, "train_runtime": 2776.6006, "train_tokens_per_second": 851.586 }, { "epoch": 0.06741798750477108, "grad_norm": 3.3274059295654297, "learning_rate": 1.1649122807017546e-05, "loss": 0.30140984058380127, "num_input_tokens_seen": 2369170, "step": 839, "train_runtime": 2779.6036, "train_tokens_per_second": 852.341 }, { "epoch": 0.06749834267462183, "grad_norm": 2.834169626235962, "learning_rate": 1.1639097744360902e-05, "loss": 0.2673649191856384, "num_input_tokens_seen": 2373256, "step": 840, "train_runtime": 2782.4625, "train_tokens_per_second": 852.934 }, { "epoch": 0.06757869784447257, "grad_norm": 3.0885422229766846, "learning_rate": 1.1629072681704261e-05, "loss": 0.28609558939933777, "num_input_tokens_seen": 2376258, "step": 841, "train_runtime": 2796.1932, "train_tokens_per_second": 849.819 }, { "epoch": 0.0676590530143233, "grad_norm": 2.7441048622131348, "learning_rate": 1.1619047619047621e-05, "loss": 0.2724419832229614, "num_input_tokens_seen": 2377962, "step": 842, "train_runtime": 2798.4984, "train_tokens_per_second": 849.728 }, { "epoch": 0.06773940818417405, "grad_norm": 3.0750732421875, "learning_rate": 1.1609022556390978e-05, "loss": 0.21906113624572754, "num_input_tokens_seen": 2379658, "step": 843, "train_runtime": 2800.7626, "train_tokens_per_second": 849.646 }, { "epoch": 0.06781976335402479, "grad_norm": 2.9323623180389404, "learning_rate": 1.1598997493734336e-05, "loss": 0.32173338532447815, "num_input_tokens_seen": 2383732, "step": 844, "train_runtime": 2803.6724, "train_tokens_per_second": 850.218 }, { "epoch": 0.06790011852387554, "grad_norm": 2.686721086502075, "learning_rate": 1.1588972431077695e-05, "loss": 0.24609172344207764, "num_input_tokens_seen": 2387400, "step": 845, "train_runtime": 2806.5124, "train_tokens_per_second": 850.664 }, { "epoch": 0.06798047369372627, "grad_norm": 3.380343437194824, "learning_rate": 1.1578947368421053e-05, "loss": 0.2745164632797241, "num_input_tokens_seen": 2390728, "step": 846, "train_runtime": 2809.2375, "train_tokens_per_second": 851.024 }, { "epoch": 0.068060828863577, "grad_norm": 3.3192689418792725, "learning_rate": 1.1568922305764412e-05, "loss": 0.37721115350723267, "num_input_tokens_seen": 2394720, "step": 847, "train_runtime": 2812.12, "train_tokens_per_second": 851.571 }, { "epoch": 0.06814118403342775, "grad_norm": 3.3431966304779053, "learning_rate": 1.1558897243107769e-05, "loss": 0.32320156693458557, "num_input_tokens_seen": 2397740, "step": 848, "train_runtime": 2814.7859, "train_tokens_per_second": 851.837 }, { "epoch": 0.06822153920327849, "grad_norm": 3.164013147354126, "learning_rate": 1.1548872180451129e-05, "loss": 0.32989904284477234, "num_input_tokens_seen": 2400966, "step": 849, "train_runtime": 2817.4769, "train_tokens_per_second": 852.169 }, { "epoch": 0.06830189437312924, "grad_norm": 2.7197554111480713, "learning_rate": 1.1538847117794487e-05, "loss": 0.23005303740501404, "num_input_tokens_seen": 2403542, "step": 850, "train_runtime": 2820.0432, "train_tokens_per_second": 852.307 }, { "epoch": 0.06838224954297997, "grad_norm": 3.0320212841033936, "learning_rate": 1.1528822055137844e-05, "loss": 0.33020418882369995, "num_input_tokens_seen": 2405274, "step": 851, "train_runtime": 2822.4314, "train_tokens_per_second": 852.199 }, { "epoch": 0.06846260471283071, "grad_norm": 2.902709722518921, "learning_rate": 1.1518796992481204e-05, "loss": 0.3048401474952698, "num_input_tokens_seen": 2407668, "step": 852, "train_runtime": 2824.9282, "train_tokens_per_second": 852.294 }, { "epoch": 0.06854295988268146, "grad_norm": 3.3414578437805176, "learning_rate": 1.1508771929824563e-05, "loss": 0.34276214241981506, "num_input_tokens_seen": 2410880, "step": 853, "train_runtime": 2827.5857, "train_tokens_per_second": 852.628 }, { "epoch": 0.06862331505253219, "grad_norm": 2.3540456295013428, "learning_rate": 1.149874686716792e-05, "loss": 0.21741557121276855, "num_input_tokens_seen": 2414056, "step": 854, "train_runtime": 2830.2874, "train_tokens_per_second": 852.937 }, { "epoch": 0.06870367022238293, "grad_norm": 3.273486375808716, "learning_rate": 1.1488721804511278e-05, "loss": 0.30173686146736145, "num_input_tokens_seen": 2416810, "step": 855, "train_runtime": 2832.8249, "train_tokens_per_second": 853.145 }, { "epoch": 0.06878402539223367, "grad_norm": 2.766336679458618, "learning_rate": 1.1478696741854638e-05, "loss": 0.26270854473114014, "num_input_tokens_seen": 2419084, "step": 856, "train_runtime": 2835.2551, "train_tokens_per_second": 853.216 }, { "epoch": 0.06886438056208441, "grad_norm": 2.394879102706909, "learning_rate": 1.1468671679197995e-05, "loss": 0.2380744367837906, "num_input_tokens_seen": 2422558, "step": 857, "train_runtime": 2838.0168, "train_tokens_per_second": 853.609 }, { "epoch": 0.06894473573193516, "grad_norm": 3.3358161449432373, "learning_rate": 1.1458646616541354e-05, "loss": 0.252188503742218, "num_input_tokens_seen": 2424698, "step": 858, "train_runtime": 2840.4029, "train_tokens_per_second": 853.646 }, { "epoch": 0.0690250909017859, "grad_norm": 2.712773323059082, "learning_rate": 1.1448621553884714e-05, "loss": 0.23814445734024048, "num_input_tokens_seen": 2426968, "step": 859, "train_runtime": 2842.8427, "train_tokens_per_second": 853.712 }, { "epoch": 0.06910544607163663, "grad_norm": 2.6414430141448975, "learning_rate": 1.143859649122807e-05, "loss": 0.2549075782299042, "num_input_tokens_seen": 2429084, "step": 860, "train_runtime": 2845.2482, "train_tokens_per_second": 853.734 }, { "epoch": 0.06918580124148738, "grad_norm": 3.0256922245025635, "learning_rate": 1.1428571428571429e-05, "loss": 0.2339504212141037, "num_input_tokens_seen": 2431882, "step": 861, "train_runtime": 2847.8126, "train_tokens_per_second": 853.947 }, { "epoch": 0.06926615641133811, "grad_norm": 2.381401777267456, "learning_rate": 1.141854636591479e-05, "loss": 0.20784461498260498, "num_input_tokens_seen": 2435424, "step": 862, "train_runtime": 2850.5694, "train_tokens_per_second": 854.364 }, { "epoch": 0.06934651158118886, "grad_norm": 3.066648244857788, "learning_rate": 1.1408521303258146e-05, "loss": 0.30620497465133667, "num_input_tokens_seen": 2437234, "step": 863, "train_runtime": 2852.8853, "train_tokens_per_second": 854.305 }, { "epoch": 0.0694268667510396, "grad_norm": 2.7175960540771484, "learning_rate": 1.1398496240601504e-05, "loss": 0.25860556960105896, "num_input_tokens_seen": 2441074, "step": 864, "train_runtime": 2855.6943, "train_tokens_per_second": 854.809 }, { "epoch": 0.06950722192089033, "grad_norm": 3.146047353744507, "learning_rate": 1.1388471177944861e-05, "loss": 0.3009291887283325, "num_input_tokens_seen": 2443040, "step": 865, "train_runtime": 2858.053, "train_tokens_per_second": 854.792 }, { "epoch": 0.06958757709074108, "grad_norm": 3.2384097576141357, "learning_rate": 1.1378446115288221e-05, "loss": 0.18501576781272888, "num_input_tokens_seen": 2446488, "step": 866, "train_runtime": 2860.7788, "train_tokens_per_second": 855.183 }, { "epoch": 0.06966793226059181, "grad_norm": 3.0936598777770996, "learning_rate": 1.136842105263158e-05, "loss": 0.33881914615631104, "num_input_tokens_seen": 2449056, "step": 867, "train_runtime": 2863.2686, "train_tokens_per_second": 855.336 }, { "epoch": 0.06974828743044256, "grad_norm": 2.7397212982177734, "learning_rate": 1.1358395989974937e-05, "loss": 0.2787984609603882, "num_input_tokens_seen": 2451800, "step": 868, "train_runtime": 2865.8353, "train_tokens_per_second": 855.527 }, { "epoch": 0.0698286426002933, "grad_norm": 2.962069034576416, "learning_rate": 1.1348370927318297e-05, "loss": 0.2494591623544693, "num_input_tokens_seen": 2455114, "step": 869, "train_runtime": 2868.5381, "train_tokens_per_second": 855.876 }, { "epoch": 0.06990899777014403, "grad_norm": 2.6867611408233643, "learning_rate": 1.1338345864661655e-05, "loss": 0.2679477035999298, "num_input_tokens_seen": 2458982, "step": 870, "train_runtime": 2871.3465, "train_tokens_per_second": 856.386 }, { "epoch": 0.06998935293999478, "grad_norm": 2.541769027709961, "learning_rate": 1.1328320802005012e-05, "loss": 0.3020399212837219, "num_input_tokens_seen": 2462436, "step": 871, "train_runtime": 2885.6952, "train_tokens_per_second": 853.325 }, { "epoch": 0.07006970810984552, "grad_norm": 2.5876219272613525, "learning_rate": 1.1318295739348372e-05, "loss": 0.25577884912490845, "num_input_tokens_seen": 2464638, "step": 872, "train_runtime": 2888.1592, "train_tokens_per_second": 853.359 }, { "epoch": 0.07015006327969625, "grad_norm": 2.554353952407837, "learning_rate": 1.1308270676691731e-05, "loss": 0.28523603081703186, "num_input_tokens_seen": 2466722, "step": 873, "train_runtime": 2890.548, "train_tokens_per_second": 853.375 }, { "epoch": 0.070230418449547, "grad_norm": 2.9560370445251465, "learning_rate": 1.1298245614035088e-05, "loss": 0.25039082765579224, "num_input_tokens_seen": 2469044, "step": 874, "train_runtime": 2892.9936, "train_tokens_per_second": 853.456 }, { "epoch": 0.07031077361939773, "grad_norm": 3.0631520748138428, "learning_rate": 1.1288220551378446e-05, "loss": 0.3140687346458435, "num_input_tokens_seen": 2472666, "step": 875, "train_runtime": 2895.7746, "train_tokens_per_second": 853.888 }, { "epoch": 0.07039112878924848, "grad_norm": 3.1806013584136963, "learning_rate": 1.1278195488721806e-05, "loss": 0.31442970037460327, "num_input_tokens_seen": 2475484, "step": 876, "train_runtime": 2898.3375, "train_tokens_per_second": 854.105 }, { "epoch": 0.07047148395909922, "grad_norm": 2.9442179203033447, "learning_rate": 1.1268170426065163e-05, "loss": 0.4229345917701721, "num_input_tokens_seen": 2478716, "step": 877, "train_runtime": 2900.9924, "train_tokens_per_second": 854.437 }, { "epoch": 0.07055183912894995, "grad_norm": 2.8847992420196533, "learning_rate": 1.1258145363408522e-05, "loss": 0.2810676395893097, "num_input_tokens_seen": 2481338, "step": 878, "train_runtime": 2903.5132, "train_tokens_per_second": 854.599 }, { "epoch": 0.0706321942988007, "grad_norm": 3.0552022457122803, "learning_rate": 1.1248120300751882e-05, "loss": 0.27646005153656006, "num_input_tokens_seen": 2483482, "step": 879, "train_runtime": 2905.8879, "train_tokens_per_second": 854.638 }, { "epoch": 0.07071254946865144, "grad_norm": 2.3963708877563477, "learning_rate": 1.1238095238095239e-05, "loss": 0.2934857904911041, "num_input_tokens_seen": 2486044, "step": 880, "train_runtime": 2908.4349, "train_tokens_per_second": 854.77 }, { "epoch": 0.07079290463850219, "grad_norm": 3.189134359359741, "learning_rate": 1.1228070175438597e-05, "loss": 0.2951675057411194, "num_input_tokens_seen": 2488032, "step": 881, "train_runtime": 2910.785, "train_tokens_per_second": 854.763 }, { "epoch": 0.07087325980835292, "grad_norm": 3.145353317260742, "learning_rate": 1.1218045112781957e-05, "loss": 0.3777672052383423, "num_input_tokens_seen": 2490256, "step": 882, "train_runtime": 2913.1907, "train_tokens_per_second": 854.821 }, { "epoch": 0.07095361497820366, "grad_norm": 2.863912582397461, "learning_rate": 1.1208020050125314e-05, "loss": 0.3316439688205719, "num_input_tokens_seen": 2492516, "step": 883, "train_runtime": 2915.6586, "train_tokens_per_second": 854.872 }, { "epoch": 0.0710339701480544, "grad_norm": 2.7762043476104736, "learning_rate": 1.1197994987468673e-05, "loss": 0.3250517249107361, "num_input_tokens_seen": 2494358, "step": 884, "train_runtime": 2917.9679, "train_tokens_per_second": 854.827 }, { "epoch": 0.07111432531790514, "grad_norm": 3.6697604656219482, "learning_rate": 1.118796992481203e-05, "loss": 0.3572869300842285, "num_input_tokens_seen": 2495880, "step": 885, "train_runtime": 2920.2156, "train_tokens_per_second": 854.69 }, { "epoch": 0.07119468048775589, "grad_norm": 3.345015287399292, "learning_rate": 1.117794486215539e-05, "loss": 0.2328757494688034, "num_input_tokens_seen": 2498010, "step": 886, "train_runtime": 2922.586, "train_tokens_per_second": 854.726 }, { "epoch": 0.07127503565760662, "grad_norm": 2.9286606311798096, "learning_rate": 1.1167919799498748e-05, "loss": 0.2707301378250122, "num_input_tokens_seen": 2499924, "step": 887, "train_runtime": 2924.9304, "train_tokens_per_second": 854.695 }, { "epoch": 0.07135539082745736, "grad_norm": 2.4948160648345947, "learning_rate": 1.1157894736842105e-05, "loss": 0.28970158100128174, "num_input_tokens_seen": 2501794, "step": 888, "train_runtime": 2927.2691, "train_tokens_per_second": 854.651 }, { "epoch": 0.0714357459973081, "grad_norm": 2.6450035572052, "learning_rate": 1.1147869674185465e-05, "loss": 0.27675482630729675, "num_input_tokens_seen": 2505034, "step": 889, "train_runtime": 2929.9237, "train_tokens_per_second": 854.983 }, { "epoch": 0.07151610116715884, "grad_norm": 2.7640318870544434, "learning_rate": 1.1137844611528823e-05, "loss": 0.20892232656478882, "num_input_tokens_seen": 2507526, "step": 890, "train_runtime": 2932.4134, "train_tokens_per_second": 855.107 }, { "epoch": 0.07159645633700958, "grad_norm": 2.5910139083862305, "learning_rate": 1.112781954887218e-05, "loss": 0.2586798369884491, "num_input_tokens_seen": 2511154, "step": 891, "train_runtime": 2935.1674, "train_tokens_per_second": 855.54 }, { "epoch": 0.07167681150686032, "grad_norm": 2.934616804122925, "learning_rate": 1.111779448621554e-05, "loss": 0.3174422085285187, "num_input_tokens_seen": 2513732, "step": 892, "train_runtime": 2937.6957, "train_tokens_per_second": 855.682 }, { "epoch": 0.07175716667671106, "grad_norm": 3.296107292175293, "learning_rate": 1.1107769423558899e-05, "loss": 0.36007875204086304, "num_input_tokens_seen": 2516034, "step": 893, "train_runtime": 2940.1516, "train_tokens_per_second": 855.75 }, { "epoch": 0.07183752184656181, "grad_norm": 3.3363406658172607, "learning_rate": 1.1097744360902256e-05, "loss": 0.2184358537197113, "num_input_tokens_seen": 2517878, "step": 894, "train_runtime": 2942.5006, "train_tokens_per_second": 855.693 }, { "epoch": 0.07191787701641254, "grad_norm": 3.6419906616210938, "learning_rate": 1.1087719298245614e-05, "loss": 0.31250089406967163, "num_input_tokens_seen": 2519918, "step": 895, "train_runtime": 2944.8727, "train_tokens_per_second": 855.697 }, { "epoch": 0.07199823218626328, "grad_norm": 2.5691442489624023, "learning_rate": 1.1077694235588974e-05, "loss": 0.2788035273551941, "num_input_tokens_seen": 2523422, "step": 896, "train_runtime": 2947.6241, "train_tokens_per_second": 856.087 }, { "epoch": 0.07207858735611403, "grad_norm": 3.0096826553344727, "learning_rate": 1.1067669172932331e-05, "loss": 0.31573477387428284, "num_input_tokens_seen": 2526436, "step": 897, "train_runtime": 2950.2564, "train_tokens_per_second": 856.345 }, { "epoch": 0.07215894252596476, "grad_norm": 2.6826555728912354, "learning_rate": 1.105764411027569e-05, "loss": 0.20941345393657684, "num_input_tokens_seen": 2529046, "step": 898, "train_runtime": 2952.7523, "train_tokens_per_second": 856.505 }, { "epoch": 0.07223929769581551, "grad_norm": 2.7025513648986816, "learning_rate": 1.104761904761905e-05, "loss": 0.2592368423938751, "num_input_tokens_seen": 2532954, "step": 899, "train_runtime": 2955.528, "train_tokens_per_second": 857.022 }, { "epoch": 0.07231965286566625, "grad_norm": 2.2408924102783203, "learning_rate": 1.1037593984962407e-05, "loss": 0.21881291270256042, "num_input_tokens_seen": 2536206, "step": 900, "train_runtime": 2958.1914, "train_tokens_per_second": 857.35 }, { "epoch": 0.07240000803551698, "grad_norm": 2.42545485496521, "learning_rate": 1.1027568922305765e-05, "loss": 0.2336631417274475, "num_input_tokens_seen": 2540696, "step": 901, "train_runtime": 2972.1145, "train_tokens_per_second": 854.845 }, { "epoch": 0.07248036320536773, "grad_norm": 3.4606752395629883, "learning_rate": 1.1017543859649125e-05, "loss": 0.34886184334754944, "num_input_tokens_seen": 2542916, "step": 902, "train_runtime": 2974.5204, "train_tokens_per_second": 854.9 }, { "epoch": 0.07256071837521846, "grad_norm": 3.267016887664795, "learning_rate": 1.1007518796992482e-05, "loss": 0.32064512372016907, "num_input_tokens_seen": 2545442, "step": 903, "train_runtime": 2977.029, "train_tokens_per_second": 855.028 }, { "epoch": 0.07264107354506921, "grad_norm": 3.088024377822876, "learning_rate": 1.099749373433584e-05, "loss": 0.2829703688621521, "num_input_tokens_seen": 2548038, "step": 904, "train_runtime": 2979.5773, "train_tokens_per_second": 855.168 }, { "epoch": 0.07272142871491995, "grad_norm": 2.9237823486328125, "learning_rate": 1.0987468671679197e-05, "loss": 0.38408032059669495, "num_input_tokens_seen": 2551642, "step": 905, "train_runtime": 2982.3122, "train_tokens_per_second": 855.592 }, { "epoch": 0.07280178388477068, "grad_norm": 2.35085129737854, "learning_rate": 1.0977443609022558e-05, "loss": 0.19211620092391968, "num_input_tokens_seen": 2555158, "step": 906, "train_runtime": 2985.0505, "train_tokens_per_second": 855.985 }, { "epoch": 0.07288213905462143, "grad_norm": 3.5112156867980957, "learning_rate": 1.0967418546365916e-05, "loss": 0.2657938003540039, "num_input_tokens_seen": 2557426, "step": 907, "train_runtime": 2987.4565, "train_tokens_per_second": 856.055 }, { "epoch": 0.07296249422447217, "grad_norm": 3.499185085296631, "learning_rate": 1.0957393483709273e-05, "loss": 0.33689403533935547, "num_input_tokens_seen": 2559798, "step": 908, "train_runtime": 2989.8968, "train_tokens_per_second": 856.149 }, { "epoch": 0.0730428493943229, "grad_norm": 3.5011346340179443, "learning_rate": 1.0947368421052633e-05, "loss": 0.29301249980926514, "num_input_tokens_seen": 2563598, "step": 909, "train_runtime": 2992.7143, "train_tokens_per_second": 856.613 }, { "epoch": 0.07312320456417365, "grad_norm": 3.0111422538757324, "learning_rate": 1.0937343358395992e-05, "loss": 0.2713438868522644, "num_input_tokens_seen": 2565922, "step": 910, "train_runtime": 2995.1897, "train_tokens_per_second": 856.681 }, { "epoch": 0.07320355973402438, "grad_norm": 3.3043766021728516, "learning_rate": 1.0927318295739348e-05, "loss": 0.34866422414779663, "num_input_tokens_seen": 2570004, "step": 911, "train_runtime": 2998.1309, "train_tokens_per_second": 857.202 }, { "epoch": 0.07328391490387513, "grad_norm": 3.054727554321289, "learning_rate": 1.0917293233082709e-05, "loss": 0.18902209401130676, "num_input_tokens_seen": 2571808, "step": 912, "train_runtime": 3000.4496, "train_tokens_per_second": 857.141 }, { "epoch": 0.07336427007372587, "grad_norm": 2.884338617324829, "learning_rate": 1.0907268170426067e-05, "loss": 0.322395920753479, "num_input_tokens_seen": 2575322, "step": 913, "train_runtime": 3003.3275, "train_tokens_per_second": 857.49 }, { "epoch": 0.0734446252435766, "grad_norm": 3.417935848236084, "learning_rate": 1.0897243107769424e-05, "loss": 0.3774549961090088, "num_input_tokens_seen": 2577886, "step": 914, "train_runtime": 3005.8377, "train_tokens_per_second": 857.626 }, { "epoch": 0.07352498041342735, "grad_norm": 3.3805410861968994, "learning_rate": 1.0887218045112782e-05, "loss": 0.35853803157806396, "num_input_tokens_seen": 2583050, "step": 915, "train_runtime": 3009.0282, "train_tokens_per_second": 858.433 }, { "epoch": 0.07360533558327809, "grad_norm": 2.988804340362549, "learning_rate": 1.0877192982456142e-05, "loss": 0.30890414118766785, "num_input_tokens_seen": 2587608, "step": 916, "train_runtime": 3012.0379, "train_tokens_per_second": 859.089 }, { "epoch": 0.07368569075312884, "grad_norm": 2.9539706707000732, "learning_rate": 1.08671679197995e-05, "loss": 0.2871193289756775, "num_input_tokens_seen": 2589274, "step": 917, "train_runtime": 3014.2882, "train_tokens_per_second": 859.0 }, { "epoch": 0.07376604592297957, "grad_norm": 1.9494750499725342, "learning_rate": 1.0857142857142858e-05, "loss": 0.14205417037010193, "num_input_tokens_seen": 2591092, "step": 918, "train_runtime": 3016.6056, "train_tokens_per_second": 858.943 }, { "epoch": 0.0738464010928303, "grad_norm": 2.887657880783081, "learning_rate": 1.0847117794486218e-05, "loss": 0.2674184739589691, "num_input_tokens_seen": 2594500, "step": 919, "train_runtime": 3019.287, "train_tokens_per_second": 859.309 }, { "epoch": 0.07392675626268105, "grad_norm": 3.1741626262664795, "learning_rate": 1.0837092731829575e-05, "loss": 0.3731027841567993, "num_input_tokens_seen": 2596780, "step": 920, "train_runtime": 3021.7274, "train_tokens_per_second": 859.369 }, { "epoch": 0.07400711143253179, "grad_norm": 2.750936508178711, "learning_rate": 1.0827067669172933e-05, "loss": 0.2692050635814667, "num_input_tokens_seen": 2601548, "step": 921, "train_runtime": 3024.8055, "train_tokens_per_second": 860.071 }, { "epoch": 0.07408746660238254, "grad_norm": 2.9356539249420166, "learning_rate": 1.0817042606516293e-05, "loss": 0.24668651819229126, "num_input_tokens_seen": 2603926, "step": 922, "train_runtime": 3027.2819, "train_tokens_per_second": 860.153 }, { "epoch": 0.07416782177223327, "grad_norm": 3.0405921936035156, "learning_rate": 1.080701754385965e-05, "loss": 0.2326321303844452, "num_input_tokens_seen": 2607208, "step": 923, "train_runtime": 3029.9706, "train_tokens_per_second": 860.473 }, { "epoch": 0.07424817694208401, "grad_norm": 2.9016056060791016, "learning_rate": 1.0796992481203009e-05, "loss": 0.3299196660518646, "num_input_tokens_seen": 2610570, "step": 924, "train_runtime": 3032.732, "train_tokens_per_second": 860.798 }, { "epoch": 0.07432853211193476, "grad_norm": 2.5427145957946777, "learning_rate": 1.0786967418546365e-05, "loss": 0.25243812799453735, "num_input_tokens_seen": 2612472, "step": 925, "train_runtime": 3035.0538, "train_tokens_per_second": 860.766 }, { "epoch": 0.07440888728178549, "grad_norm": 3.4277877807617188, "learning_rate": 1.0776942355889726e-05, "loss": 0.349045068025589, "num_input_tokens_seen": 2617510, "step": 926, "train_runtime": 3038.2016, "train_tokens_per_second": 861.533 }, { "epoch": 0.07448924245163623, "grad_norm": 3.1395628452301025, "learning_rate": 1.0766917293233084e-05, "loss": 0.30842944979667664, "num_input_tokens_seen": 2620594, "step": 927, "train_runtime": 3040.84, "train_tokens_per_second": 861.799 }, { "epoch": 0.07456959762148697, "grad_norm": 2.8847174644470215, "learning_rate": 1.0756892230576441e-05, "loss": 0.30633026361465454, "num_input_tokens_seen": 2622998, "step": 928, "train_runtime": 3043.3687, "train_tokens_per_second": 861.873 }, { "epoch": 0.07464995279133771, "grad_norm": 3.185054302215576, "learning_rate": 1.0746867167919801e-05, "loss": 0.22780334949493408, "num_input_tokens_seen": 2624440, "step": 929, "train_runtime": 3045.5952, "train_tokens_per_second": 861.717 }, { "epoch": 0.07473030796118846, "grad_norm": 2.9074995517730713, "learning_rate": 1.073684210526316e-05, "loss": 0.35684502124786377, "num_input_tokens_seen": 2628624, "step": 930, "train_runtime": 3048.555, "train_tokens_per_second": 862.252 }, { "epoch": 0.07481066313103919, "grad_norm": 3.058561086654663, "learning_rate": 1.0726817042606516e-05, "loss": 0.24309319257736206, "num_input_tokens_seen": 2631244, "step": 931, "train_runtime": 3061.5738, "train_tokens_per_second": 859.442 }, { "epoch": 0.07489101830088993, "grad_norm": 3.51434063911438, "learning_rate": 1.0716791979949877e-05, "loss": 0.27488744258880615, "num_input_tokens_seen": 2632922, "step": 932, "train_runtime": 3064.0394, "train_tokens_per_second": 859.298 }, { "epoch": 0.07497137347074068, "grad_norm": 3.7143125534057617, "learning_rate": 1.0706766917293235e-05, "loss": 0.2860804796218872, "num_input_tokens_seen": 2634736, "step": 933, "train_runtime": 3066.3421, "train_tokens_per_second": 859.244 }, { "epoch": 0.07505172864059141, "grad_norm": 3.395794630050659, "learning_rate": 1.0696741854636592e-05, "loss": 0.30575066804885864, "num_input_tokens_seen": 2636844, "step": 934, "train_runtime": 3068.7354, "train_tokens_per_second": 859.261 }, { "epoch": 0.07513208381044216, "grad_norm": 3.1410999298095703, "learning_rate": 1.068671679197995e-05, "loss": 0.29593583941459656, "num_input_tokens_seen": 2638380, "step": 935, "train_runtime": 3071.0141, "train_tokens_per_second": 859.123 }, { "epoch": 0.0752124389802929, "grad_norm": 3.0170605182647705, "learning_rate": 1.0676691729323309e-05, "loss": 0.2192925065755844, "num_input_tokens_seen": 2639908, "step": 936, "train_runtime": 3073.2643, "train_tokens_per_second": 858.992 }, { "epoch": 0.07529279415014363, "grad_norm": 2.8403615951538086, "learning_rate": 1.0666666666666667e-05, "loss": 0.27189669013023376, "num_input_tokens_seen": 2642592, "step": 937, "train_runtime": 3075.7881, "train_tokens_per_second": 859.159 }, { "epoch": 0.07537314931999438, "grad_norm": 3.5006563663482666, "learning_rate": 1.0656641604010026e-05, "loss": 0.25356900691986084, "num_input_tokens_seen": 2645672, "step": 938, "train_runtime": 3078.404, "train_tokens_per_second": 859.43 }, { "epoch": 0.07545350448984511, "grad_norm": 2.8472111225128174, "learning_rate": 1.0646616541353384e-05, "loss": 0.22225934267044067, "num_input_tokens_seen": 2648698, "step": 939, "train_runtime": 3081.0059, "train_tokens_per_second": 859.686 }, { "epoch": 0.07553385965969586, "grad_norm": 1.8143668174743652, "learning_rate": 1.0636591478696743e-05, "loss": 0.12409442663192749, "num_input_tokens_seen": 2652294, "step": 940, "train_runtime": 3083.7636, "train_tokens_per_second": 860.083 }, { "epoch": 0.0756142148295466, "grad_norm": 3.1159932613372803, "learning_rate": 1.0626566416040101e-05, "loss": 0.2769079804420471, "num_input_tokens_seen": 2654368, "step": 941, "train_runtime": 3086.1396, "train_tokens_per_second": 860.093 }, { "epoch": 0.07569456999939733, "grad_norm": 3.694201946258545, "learning_rate": 1.061654135338346e-05, "loss": 0.2216227948665619, "num_input_tokens_seen": 2657062, "step": 942, "train_runtime": 3088.694, "train_tokens_per_second": 860.254 }, { "epoch": 0.07577492516924808, "grad_norm": 3.5044589042663574, "learning_rate": 1.0606516290726818e-05, "loss": 0.2904451787471771, "num_input_tokens_seen": 2659348, "step": 943, "train_runtime": 3091.1411, "train_tokens_per_second": 860.313 }, { "epoch": 0.07585528033909882, "grad_norm": 3.1539793014526367, "learning_rate": 1.0596491228070177e-05, "loss": 0.29085615277290344, "num_input_tokens_seen": 2662136, "step": 944, "train_runtime": 3093.7384, "train_tokens_per_second": 860.492 }, { "epoch": 0.07593563550894955, "grad_norm": 2.5806024074554443, "learning_rate": 1.0586466165413534e-05, "loss": 0.2519429326057434, "num_input_tokens_seen": 2664614, "step": 945, "train_runtime": 3096.1837, "train_tokens_per_second": 860.612 }, { "epoch": 0.0760159906788003, "grad_norm": 2.9692866802215576, "learning_rate": 1.0576441102756894e-05, "loss": 0.3095383048057556, "num_input_tokens_seen": 2666958, "step": 946, "train_runtime": 3098.6359, "train_tokens_per_second": 860.688 }, { "epoch": 0.07609634584865103, "grad_norm": 2.8070459365844727, "learning_rate": 1.0566416040100252e-05, "loss": 0.24489399790763855, "num_input_tokens_seen": 2670506, "step": 947, "train_runtime": 3101.3917, "train_tokens_per_second": 861.067 }, { "epoch": 0.07617670101850178, "grad_norm": 2.559191942214966, "learning_rate": 1.0556390977443609e-05, "loss": 0.24660678207874298, "num_input_tokens_seen": 2673816, "step": 948, "train_runtime": 3104.0521, "train_tokens_per_second": 861.395 }, { "epoch": 0.07625705618835252, "grad_norm": 2.632732391357422, "learning_rate": 1.054636591478697e-05, "loss": 0.26406505703926086, "num_input_tokens_seen": 2675642, "step": 949, "train_runtime": 3106.4124, "train_tokens_per_second": 861.329 }, { "epoch": 0.07633741135820325, "grad_norm": 4.315592288970947, "learning_rate": 1.0536340852130326e-05, "loss": 0.2956114113330841, "num_input_tokens_seen": 2677732, "step": 950, "train_runtime": 3108.8232, "train_tokens_per_second": 861.333 }, { "epoch": 0.076417766528054, "grad_norm": 2.6697747707366943, "learning_rate": 1.0526315789473684e-05, "loss": 0.27512234449386597, "num_input_tokens_seen": 2679740, "step": 951, "train_runtime": 3111.2419, "train_tokens_per_second": 861.309 }, { "epoch": 0.07649812169790474, "grad_norm": 3.5777106285095215, "learning_rate": 1.0516290726817045e-05, "loss": 0.39105916023254395, "num_input_tokens_seen": 2683260, "step": 952, "train_runtime": 3113.9801, "train_tokens_per_second": 861.682 }, { "epoch": 0.07657847686775549, "grad_norm": 2.491039514541626, "learning_rate": 1.0506265664160401e-05, "loss": 0.23402318358421326, "num_input_tokens_seen": 2688696, "step": 953, "train_runtime": 3117.1429, "train_tokens_per_second": 862.551 }, { "epoch": 0.07665883203760622, "grad_norm": 3.17685866355896, "learning_rate": 1.049624060150376e-05, "loss": 0.23363497853279114, "num_input_tokens_seen": 2690590, "step": 954, "train_runtime": 3119.4488, "train_tokens_per_second": 862.521 }, { "epoch": 0.07673918720745695, "grad_norm": 2.934950590133667, "learning_rate": 1.0486215538847118e-05, "loss": 0.32324397563934326, "num_input_tokens_seen": 2693776, "step": 955, "train_runtime": 3122.1218, "train_tokens_per_second": 862.803 }, { "epoch": 0.0768195423773077, "grad_norm": 3.0018794536590576, "learning_rate": 1.0476190476190477e-05, "loss": 0.28532862663269043, "num_input_tokens_seen": 2695678, "step": 956, "train_runtime": 3124.4711, "train_tokens_per_second": 862.763 }, { "epoch": 0.07689989754715844, "grad_norm": 2.84800386428833, "learning_rate": 1.0466165413533835e-05, "loss": 0.1978975385427475, "num_input_tokens_seen": 2698078, "step": 957, "train_runtime": 3126.9471, "train_tokens_per_second": 862.847 }, { "epoch": 0.07698025271700919, "grad_norm": 2.46484375, "learning_rate": 1.0456140350877194e-05, "loss": 0.27531129121780396, "num_input_tokens_seen": 2700892, "step": 958, "train_runtime": 3129.5195, "train_tokens_per_second": 863.037 }, { "epoch": 0.07706060788685992, "grad_norm": 2.919818878173828, "learning_rate": 1.0446115288220552e-05, "loss": 0.21053661406040192, "num_input_tokens_seen": 2703672, "step": 959, "train_runtime": 3132.0642, "train_tokens_per_second": 863.224 }, { "epoch": 0.07714096305671066, "grad_norm": 2.8706934452056885, "learning_rate": 1.043609022556391e-05, "loss": 0.2930707037448883, "num_input_tokens_seen": 2706322, "step": 960, "train_runtime": 3134.5988, "train_tokens_per_second": 863.371 }, { "epoch": 0.0772213182265614, "grad_norm": 2.9488773345947266, "learning_rate": 1.0426065162907268e-05, "loss": 0.2253037393093109, "num_input_tokens_seen": 2708152, "step": 961, "train_runtime": 3148.898, "train_tokens_per_second": 860.032 }, { "epoch": 0.07730167339641214, "grad_norm": 3.013538360595703, "learning_rate": 1.0416040100250628e-05, "loss": 0.24188920855522156, "num_input_tokens_seen": 2710120, "step": 962, "train_runtime": 3151.2441, "train_tokens_per_second": 860.016 }, { "epoch": 0.07738202856626288, "grad_norm": 3.2728631496429443, "learning_rate": 1.0406015037593986e-05, "loss": 0.3330007791519165, "num_input_tokens_seen": 2713192, "step": 963, "train_runtime": 3153.8789, "train_tokens_per_second": 860.271 }, { "epoch": 0.07746238373611362, "grad_norm": 3.1263556480407715, "learning_rate": 1.0395989974937343e-05, "loss": 0.3454062342643738, "num_input_tokens_seen": 2715268, "step": 964, "train_runtime": 3156.2976, "train_tokens_per_second": 860.27 }, { "epoch": 0.07754273890596436, "grad_norm": 3.428396224975586, "learning_rate": 1.0385964912280702e-05, "loss": 0.36283326148986816, "num_input_tokens_seen": 2717654, "step": 965, "train_runtime": 3158.71, "train_tokens_per_second": 860.368 }, { "epoch": 0.07762309407581511, "grad_norm": 3.0883429050445557, "learning_rate": 1.0375939849624062e-05, "loss": 0.3031189739704132, "num_input_tokens_seen": 2721414, "step": 966, "train_runtime": 3161.5075, "train_tokens_per_second": 860.796 }, { "epoch": 0.07770344924566584, "grad_norm": 3.2181687355041504, "learning_rate": 1.0365914786967419e-05, "loss": 0.19726848602294922, "num_input_tokens_seen": 2723236, "step": 967, "train_runtime": 3163.8256, "train_tokens_per_second": 860.742 }, { "epoch": 0.07778380441551658, "grad_norm": 2.425121545791626, "learning_rate": 1.0355889724310777e-05, "loss": 0.2667858898639679, "num_input_tokens_seen": 2726264, "step": 968, "train_runtime": 3166.4209, "train_tokens_per_second": 860.992 }, { "epoch": 0.07786415958536733, "grad_norm": 3.255161762237549, "learning_rate": 1.0345864661654137e-05, "loss": 0.3482424020767212, "num_input_tokens_seen": 2729624, "step": 969, "train_runtime": 3169.1074, "train_tokens_per_second": 861.323 }, { "epoch": 0.07794451475521806, "grad_norm": 2.7471694946289062, "learning_rate": 1.0335839598997494e-05, "loss": 0.34294867515563965, "num_input_tokens_seen": 2732154, "step": 970, "train_runtime": 3171.652, "train_tokens_per_second": 861.429 }, { "epoch": 0.07802486992506881, "grad_norm": 3.1254217624664307, "learning_rate": 1.0325814536340853e-05, "loss": 0.33439940214157104, "num_input_tokens_seen": 2735410, "step": 971, "train_runtime": 3174.3474, "train_tokens_per_second": 861.724 }, { "epoch": 0.07810522509491954, "grad_norm": 2.7895805835723877, "learning_rate": 1.0315789473684213e-05, "loss": 0.2601195275783539, "num_input_tokens_seen": 2739300, "step": 972, "train_runtime": 3177.1623, "train_tokens_per_second": 862.184 }, { "epoch": 0.07818558026477028, "grad_norm": 3.3237452507019043, "learning_rate": 1.030576441102757e-05, "loss": 0.30988389253616333, "num_input_tokens_seen": 2741528, "step": 973, "train_runtime": 3179.5633, "train_tokens_per_second": 862.234 }, { "epoch": 0.07826593543462103, "grad_norm": 3.1637649536132812, "learning_rate": 1.0295739348370928e-05, "loss": 0.3148415982723236, "num_input_tokens_seen": 2743338, "step": 974, "train_runtime": 3181.8745, "train_tokens_per_second": 862.177 }, { "epoch": 0.07834629060447176, "grad_norm": 3.603101968765259, "learning_rate": 1.0285714285714285e-05, "loss": 0.3045535683631897, "num_input_tokens_seen": 2744998, "step": 975, "train_runtime": 3184.1841, "train_tokens_per_second": 862.073 }, { "epoch": 0.07842664577432251, "grad_norm": 3.35391902923584, "learning_rate": 1.0275689223057645e-05, "loss": 0.35841816663742065, "num_input_tokens_seen": 2747840, "step": 976, "train_runtime": 3186.7945, "train_tokens_per_second": 862.258 }, { "epoch": 0.07850700094417325, "grad_norm": 3.3913121223449707, "learning_rate": 1.0265664160401003e-05, "loss": 0.3307485580444336, "num_input_tokens_seen": 2754754, "step": 977, "train_runtime": 3190.488, "train_tokens_per_second": 863.427 }, { "epoch": 0.07858735611402398, "grad_norm": 3.251598596572876, "learning_rate": 1.025563909774436e-05, "loss": 0.2884286642074585, "num_input_tokens_seen": 2757312, "step": 978, "train_runtime": 3193.0461, "train_tokens_per_second": 863.537 }, { "epoch": 0.07866771128387473, "grad_norm": 3.008368492126465, "learning_rate": 1.024561403508772e-05, "loss": 0.2905981242656708, "num_input_tokens_seen": 2761544, "step": 979, "train_runtime": 3196.0189, "train_tokens_per_second": 864.057 }, { "epoch": 0.07874806645372547, "grad_norm": 3.4248290061950684, "learning_rate": 1.0235588972431079e-05, "loss": 0.3442487120628357, "num_input_tokens_seen": 2764154, "step": 980, "train_runtime": 3198.5496, "train_tokens_per_second": 864.19 }, { "epoch": 0.0788284216235762, "grad_norm": 2.64013671875, "learning_rate": 1.0225563909774436e-05, "loss": 0.2109047919511795, "num_input_tokens_seen": 2765938, "step": 981, "train_runtime": 3200.8614, "train_tokens_per_second": 864.123 }, { "epoch": 0.07890877679342695, "grad_norm": 2.815533399581909, "learning_rate": 1.0215538847117796e-05, "loss": 0.22510233521461487, "num_input_tokens_seen": 2770294, "step": 982, "train_runtime": 3203.8113, "train_tokens_per_second": 864.687 }, { "epoch": 0.07898913196327768, "grad_norm": 3.0152463912963867, "learning_rate": 1.0205513784461154e-05, "loss": 0.22924399375915527, "num_input_tokens_seen": 2772500, "step": 983, "train_runtime": 3206.2199, "train_tokens_per_second": 864.725 }, { "epoch": 0.07906948713312843, "grad_norm": 3.2430624961853027, "learning_rate": 1.0195488721804511e-05, "loss": 0.3210355043411255, "num_input_tokens_seen": 2775494, "step": 984, "train_runtime": 3208.8106, "train_tokens_per_second": 864.96 }, { "epoch": 0.07914984230297917, "grad_norm": 2.4131264686584473, "learning_rate": 1.018546365914787e-05, "loss": 0.3185635805130005, "num_input_tokens_seen": 2779578, "step": 985, "train_runtime": 3211.6956, "train_tokens_per_second": 865.455 }, { "epoch": 0.0792301974728299, "grad_norm": 2.7430477142333984, "learning_rate": 1.017543859649123e-05, "loss": 0.3092443645000458, "num_input_tokens_seen": 2781786, "step": 986, "train_runtime": 3214.0821, "train_tokens_per_second": 865.499 }, { "epoch": 0.07931055264268065, "grad_norm": 3.2494089603424072, "learning_rate": 1.0165413533834587e-05, "loss": 0.4144340753555298, "num_input_tokens_seen": 2784496, "step": 987, "train_runtime": 3216.652, "train_tokens_per_second": 865.65 }, { "epoch": 0.07939090781253139, "grad_norm": 2.719303607940674, "learning_rate": 1.0155388471177945e-05, "loss": 0.25890401005744934, "num_input_tokens_seen": 2789168, "step": 988, "train_runtime": 3219.8329, "train_tokens_per_second": 866.246 }, { "epoch": 0.07947126298238213, "grad_norm": 3.0966217517852783, "learning_rate": 1.0145363408521305e-05, "loss": 0.21390646696090698, "num_input_tokens_seen": 2791346, "step": 989, "train_runtime": 3222.2533, "train_tokens_per_second": 866.271 }, { "epoch": 0.07955161815223287, "grad_norm": 2.971519708633423, "learning_rate": 1.0135338345864662e-05, "loss": 0.20476631820201874, "num_input_tokens_seen": 2794442, "step": 990, "train_runtime": 3224.9371, "train_tokens_per_second": 866.511 }, { "epoch": 0.0796319733220836, "grad_norm": 2.7565529346466064, "learning_rate": 1.012531328320802e-05, "loss": 0.3130325376987457, "num_input_tokens_seen": 2796730, "step": 991, "train_runtime": 3238.9655, "train_tokens_per_second": 863.464 }, { "epoch": 0.07971232849193435, "grad_norm": 3.078580141067505, "learning_rate": 1.011528822055138e-05, "loss": 0.19416363537311554, "num_input_tokens_seen": 2798354, "step": 992, "train_runtime": 3241.2519, "train_tokens_per_second": 863.356 }, { "epoch": 0.07979268366178509, "grad_norm": 3.077559232711792, "learning_rate": 1.0105263157894738e-05, "loss": 0.2500746250152588, "num_input_tokens_seen": 2801432, "step": 993, "train_runtime": 3243.8894, "train_tokens_per_second": 863.603 }, { "epoch": 0.07987303883163584, "grad_norm": 2.39316987991333, "learning_rate": 1.0095238095238096e-05, "loss": 0.21953001618385315, "num_input_tokens_seen": 2803966, "step": 994, "train_runtime": 3246.3801, "train_tokens_per_second": 863.721 }, { "epoch": 0.07995339400148657, "grad_norm": 3.5589749813079834, "learning_rate": 1.0085213032581453e-05, "loss": 0.3521823585033417, "num_input_tokens_seen": 2805674, "step": 995, "train_runtime": 3248.6939, "train_tokens_per_second": 863.631 }, { "epoch": 0.0800337491713373, "grad_norm": 3.0713658332824707, "learning_rate": 1.0075187969924813e-05, "loss": 0.34391069412231445, "num_input_tokens_seen": 2807158, "step": 996, "train_runtime": 3250.9863, "train_tokens_per_second": 863.479 }, { "epoch": 0.08011410434118806, "grad_norm": 2.4753036499023438, "learning_rate": 1.0065162907268171e-05, "loss": 0.20421439409255981, "num_input_tokens_seen": 2809064, "step": 997, "train_runtime": 3253.3559, "train_tokens_per_second": 863.436 }, { "epoch": 0.08019445951103879, "grad_norm": 2.644885301589966, "learning_rate": 1.0055137844611528e-05, "loss": 0.2542359530925751, "num_input_tokens_seen": 2811416, "step": 998, "train_runtime": 3255.8136, "train_tokens_per_second": 863.506 }, { "epoch": 0.08027481468088953, "grad_norm": 3.1423861980438232, "learning_rate": 1.0045112781954888e-05, "loss": 0.24088791012763977, "num_input_tokens_seen": 2814322, "step": 999, "train_runtime": 3258.4258, "train_tokens_per_second": 863.706 }, { "epoch": 0.08035516985074027, "grad_norm": 2.8667550086975098, "learning_rate": 1.0035087719298247e-05, "loss": 0.28831619024276733, "num_input_tokens_seen": 2819396, "step": 1000, "train_runtime": 3261.5568, "train_tokens_per_second": 864.433 }, { "epoch": 0.08035516985074027, "eval_loss": 1.1204386949539185, "eval_runtime": 68.5296, "eval_samples_per_second": 14.505, "eval_steps_per_second": 7.252, "num_input_tokens_seen": 2819396, "step": 1000 }, { "epoch": 0.08043552502059101, "grad_norm": 2.7726781368255615, "learning_rate": 1.0025062656641604e-05, "loss": 0.32432037591934204, "num_input_tokens_seen": 2823204, "step": 1001, "train_runtime": 3333.0119, "train_tokens_per_second": 847.043 }, { "epoch": 0.08051588019044176, "grad_norm": 2.7974565029144287, "learning_rate": 1.0015037593984964e-05, "loss": 0.249294251203537, "num_input_tokens_seen": 2824732, "step": 1002, "train_runtime": 3335.2818, "train_tokens_per_second": 846.925 }, { "epoch": 0.08059623536029249, "grad_norm": 2.480847120285034, "learning_rate": 1.0005012531328322e-05, "loss": 0.2426639050245285, "num_input_tokens_seen": 2828156, "step": 1003, "train_runtime": 3338.0331, "train_tokens_per_second": 847.252 }, { "epoch": 0.08067659053014323, "grad_norm": 2.542202949523926, "learning_rate": 9.99498746867168e-06, "loss": 0.2917858362197876, "num_input_tokens_seen": 2830618, "step": 1004, "train_runtime": 3340.5168, "train_tokens_per_second": 847.359 }, { "epoch": 0.08075694569999398, "grad_norm": 2.7174367904663086, "learning_rate": 9.984962406015038e-06, "loss": 0.26602327823638916, "num_input_tokens_seen": 2832394, "step": 1005, "train_runtime": 3342.8761, "train_tokens_per_second": 847.293 }, { "epoch": 0.08083730086984471, "grad_norm": 3.6784749031066895, "learning_rate": 9.974937343358396e-06, "loss": 0.28863030672073364, "num_input_tokens_seen": 2835260, "step": 1006, "train_runtime": 3345.4793, "train_tokens_per_second": 847.49 }, { "epoch": 0.08091765603969546, "grad_norm": 3.05899977684021, "learning_rate": 9.964912280701755e-06, "loss": 0.24351707100868225, "num_input_tokens_seen": 2838126, "step": 1007, "train_runtime": 3348.118, "train_tokens_per_second": 847.678 }, { "epoch": 0.0809980112095462, "grad_norm": 2.5002386569976807, "learning_rate": 9.954887218045113e-06, "loss": 0.222204327583313, "num_input_tokens_seen": 2841434, "step": 1008, "train_runtime": 3350.8623, "train_tokens_per_second": 847.971 }, { "epoch": 0.08107836637939693, "grad_norm": 3.223135232925415, "learning_rate": 9.944862155388472e-06, "loss": 0.3133132755756378, "num_input_tokens_seen": 2843808, "step": 1009, "train_runtime": 3353.4085, "train_tokens_per_second": 848.035 }, { "epoch": 0.08115872154924768, "grad_norm": 2.6144323348999023, "learning_rate": 9.93483709273183e-06, "loss": 0.26953643560409546, "num_input_tokens_seen": 2846198, "step": 1010, "train_runtime": 3355.9443, "train_tokens_per_second": 848.106 }, { "epoch": 0.08123907671909841, "grad_norm": 2.6100590229034424, "learning_rate": 9.924812030075189e-06, "loss": 0.24415908753871918, "num_input_tokens_seen": 2848558, "step": 1011, "train_runtime": 3358.4686, "train_tokens_per_second": 848.172 }, { "epoch": 0.08131943188894916, "grad_norm": 2.4113831520080566, "learning_rate": 9.914786967418547e-06, "loss": 0.20661228895187378, "num_input_tokens_seen": 2850636, "step": 1012, "train_runtime": 3360.8631, "train_tokens_per_second": 848.186 }, { "epoch": 0.0813997870587999, "grad_norm": 2.3410634994506836, "learning_rate": 9.904761904761906e-06, "loss": 0.19426091015338898, "num_input_tokens_seen": 2856798, "step": 1013, "train_runtime": 3364.4141, "train_tokens_per_second": 849.122 }, { "epoch": 0.08148014222865063, "grad_norm": 3.571234703063965, "learning_rate": 9.894736842105264e-06, "loss": 0.34215855598449707, "num_input_tokens_seen": 2858596, "step": 1014, "train_runtime": 3366.7362, "train_tokens_per_second": 849.07 }, { "epoch": 0.08156049739850138, "grad_norm": 2.4953017234802246, "learning_rate": 9.884711779448623e-06, "loss": 0.29377248883247375, "num_input_tokens_seen": 2862112, "step": 1015, "train_runtime": 3369.4845, "train_tokens_per_second": 849.421 }, { "epoch": 0.08164085256835212, "grad_norm": 2.2020606994628906, "learning_rate": 9.87468671679198e-06, "loss": 0.1879524290561676, "num_input_tokens_seen": 2863874, "step": 1016, "train_runtime": 3371.8024, "train_tokens_per_second": 849.36 }, { "epoch": 0.08172120773820285, "grad_norm": 4.262032985687256, "learning_rate": 9.86466165413534e-06, "loss": 0.2627931237220764, "num_input_tokens_seen": 2866248, "step": 1017, "train_runtime": 3374.2817, "train_tokens_per_second": 849.439 }, { "epoch": 0.0818015629080536, "grad_norm": 2.9921622276306152, "learning_rate": 9.854636591478698e-06, "loss": 0.3278368413448334, "num_input_tokens_seen": 2870002, "step": 1018, "train_runtime": 3377.059, "train_tokens_per_second": 849.852 }, { "epoch": 0.08188191807790433, "grad_norm": 3.2916033267974854, "learning_rate": 9.844611528822055e-06, "loss": 0.26203659176826477, "num_input_tokens_seen": 2871812, "step": 1019, "train_runtime": 3379.4007, "train_tokens_per_second": 849.799 }, { "epoch": 0.08196227324775508, "grad_norm": 2.786294460296631, "learning_rate": 9.834586466165415e-06, "loss": 0.2923906147480011, "num_input_tokens_seen": 2874062, "step": 1020, "train_runtime": 3381.8625, "train_tokens_per_second": 849.846 }, { "epoch": 0.08204262841760582, "grad_norm": 3.1308836936950684, "learning_rate": 9.824561403508772e-06, "loss": 0.28525033593177795, "num_input_tokens_seen": 2875618, "step": 1021, "train_runtime": 3395.7315, "train_tokens_per_second": 846.833 }, { "epoch": 0.08212298358745655, "grad_norm": 3.0777547359466553, "learning_rate": 9.81453634085213e-06, "loss": 0.2667570114135742, "num_input_tokens_seen": 2878032, "step": 1022, "train_runtime": 3398.1994, "train_tokens_per_second": 846.929 }, { "epoch": 0.0822033387573073, "grad_norm": 2.993479013442993, "learning_rate": 9.80451127819549e-06, "loss": 0.31288808584213257, "num_input_tokens_seen": 2881054, "step": 1023, "train_runtime": 3400.8618, "train_tokens_per_second": 847.154 }, { "epoch": 0.08228369392715804, "grad_norm": 3.3210372924804688, "learning_rate": 9.794486215538847e-06, "loss": 0.2599511444568634, "num_input_tokens_seen": 2884632, "step": 1024, "train_runtime": 3403.6363, "train_tokens_per_second": 847.515 }, { "epoch": 0.08236404909700878, "grad_norm": 3.1501388549804688, "learning_rate": 9.784461152882206e-06, "loss": 0.2744535505771637, "num_input_tokens_seen": 2886538, "step": 1025, "train_runtime": 3405.9786, "train_tokens_per_second": 847.492 }, { "epoch": 0.08244440426685952, "grad_norm": 3.850752830505371, "learning_rate": 9.774436090225564e-06, "loss": 0.2422037124633789, "num_input_tokens_seen": 2888370, "step": 1026, "train_runtime": 3408.3167, "train_tokens_per_second": 847.448 }, { "epoch": 0.08252475943671025, "grad_norm": 2.8923146724700928, "learning_rate": 9.764411027568923e-06, "loss": 0.26897644996643066, "num_input_tokens_seen": 2891068, "step": 1027, "train_runtime": 3410.8834, "train_tokens_per_second": 847.601 }, { "epoch": 0.082605114606561, "grad_norm": 3.4990756511688232, "learning_rate": 9.754385964912281e-06, "loss": 0.28393685817718506, "num_input_tokens_seen": 2893400, "step": 1028, "train_runtime": 3413.3251, "train_tokens_per_second": 847.678 }, { "epoch": 0.08268546977641174, "grad_norm": 2.6899726390838623, "learning_rate": 9.74436090225564e-06, "loss": 0.2518342137336731, "num_input_tokens_seen": 2895070, "step": 1029, "train_runtime": 3415.6126, "train_tokens_per_second": 847.599 }, { "epoch": 0.08276582494626249, "grad_norm": 3.351921319961548, "learning_rate": 9.734335839598998e-06, "loss": 0.2554769515991211, "num_input_tokens_seen": 2896682, "step": 1030, "train_runtime": 3417.8895, "train_tokens_per_second": 847.506 }, { "epoch": 0.08284618011611322, "grad_norm": 3.2560954093933105, "learning_rate": 9.724310776942357e-06, "loss": 0.2378174364566803, "num_input_tokens_seen": 2898368, "step": 1031, "train_runtime": 3420.2509, "train_tokens_per_second": 847.414 }, { "epoch": 0.08292653528596396, "grad_norm": 2.5922584533691406, "learning_rate": 9.714285714285715e-06, "loss": 0.29033413529396057, "num_input_tokens_seen": 2902294, "step": 1032, "train_runtime": 3423.0606, "train_tokens_per_second": 847.865 }, { "epoch": 0.0830068904558147, "grad_norm": 3.2347123622894287, "learning_rate": 9.704260651629074e-06, "loss": 0.2644709348678589, "num_input_tokens_seen": 2905722, "step": 1033, "train_runtime": 3425.7526, "train_tokens_per_second": 848.2 }, { "epoch": 0.08308724562566544, "grad_norm": 3.0091824531555176, "learning_rate": 9.694235588972432e-06, "loss": 0.38773226737976074, "num_input_tokens_seen": 2909180, "step": 1034, "train_runtime": 3428.4653, "train_tokens_per_second": 848.537 }, { "epoch": 0.08316760079551618, "grad_norm": 2.2834033966064453, "learning_rate": 9.68421052631579e-06, "loss": 0.13752481341362, "num_input_tokens_seen": 2916058, "step": 1035, "train_runtime": 3432.07, "train_tokens_per_second": 849.65 }, { "epoch": 0.08324795596536692, "grad_norm": 2.6717710494995117, "learning_rate": 9.674185463659147e-06, "loss": 0.21808898448944092, "num_input_tokens_seen": 2918726, "step": 1036, "train_runtime": 3434.5517, "train_tokens_per_second": 849.813 }, { "epoch": 0.08332831113521766, "grad_norm": 2.577711582183838, "learning_rate": 9.664160401002508e-06, "loss": 0.24005763232707977, "num_input_tokens_seen": 2921418, "step": 1037, "train_runtime": 3437.0707, "train_tokens_per_second": 849.973 }, { "epoch": 0.08340866630506841, "grad_norm": 3.181042194366455, "learning_rate": 9.654135338345866e-06, "loss": 0.29091519117355347, "num_input_tokens_seen": 2923508, "step": 1038, "train_runtime": 3439.4277, "train_tokens_per_second": 849.998 }, { "epoch": 0.08348902147491914, "grad_norm": 2.7881429195404053, "learning_rate": 9.644110275689223e-06, "loss": 0.22305895388126373, "num_input_tokens_seen": 2924958, "step": 1039, "train_runtime": 3441.6281, "train_tokens_per_second": 849.876 }, { "epoch": 0.08356937664476988, "grad_norm": 2.954681873321533, "learning_rate": 9.634085213032583e-06, "loss": 0.21325847506523132, "num_input_tokens_seen": 2926804, "step": 1040, "train_runtime": 3443.9684, "train_tokens_per_second": 849.835 }, { "epoch": 0.08364973181462063, "grad_norm": 3.353553533554077, "learning_rate": 9.62406015037594e-06, "loss": 0.27024149894714355, "num_input_tokens_seen": 2931078, "step": 1041, "train_runtime": 3446.8722, "train_tokens_per_second": 850.359 }, { "epoch": 0.08373008698447136, "grad_norm": 2.2735166549682617, "learning_rate": 9.614035087719298e-06, "loss": 0.208747997879982, "num_input_tokens_seen": 2934016, "step": 1042, "train_runtime": 3449.4652, "train_tokens_per_second": 850.571 }, { "epoch": 0.08381044215432211, "grad_norm": 2.9170501232147217, "learning_rate": 9.604010025062659e-06, "loss": 0.3145195543766022, "num_input_tokens_seen": 2938282, "step": 1043, "train_runtime": 3452.364, "train_tokens_per_second": 851.093 }, { "epoch": 0.08389079732417284, "grad_norm": 3.2619576454162598, "learning_rate": 9.593984962406015e-06, "loss": 0.27522629499435425, "num_input_tokens_seen": 2940258, "step": 1044, "train_runtime": 3454.7411, "train_tokens_per_second": 851.079 }, { "epoch": 0.08397115249402358, "grad_norm": 2.8148863315582275, "learning_rate": 9.583959899749374e-06, "loss": 0.39351987838745117, "num_input_tokens_seen": 2943046, "step": 1045, "train_runtime": 3457.2992, "train_tokens_per_second": 851.256 }, { "epoch": 0.08405150766387433, "grad_norm": 3.0083775520324707, "learning_rate": 9.573934837092732e-06, "loss": 0.27277255058288574, "num_input_tokens_seen": 2945298, "step": 1046, "train_runtime": 3459.7037, "train_tokens_per_second": 851.315 }, { "epoch": 0.08413186283372506, "grad_norm": 3.4814062118530273, "learning_rate": 9.56390977443609e-06, "loss": 0.3063601851463318, "num_input_tokens_seen": 2948158, "step": 1047, "train_runtime": 3462.2912, "train_tokens_per_second": 851.505 }, { "epoch": 0.08421221800357581, "grad_norm": 2.7683229446411133, "learning_rate": 9.55388471177945e-06, "loss": 0.3377416431903839, "num_input_tokens_seen": 2951362, "step": 1048, "train_runtime": 3464.9527, "train_tokens_per_second": 851.776 }, { "epoch": 0.08429257317342655, "grad_norm": 2.934075117111206, "learning_rate": 9.543859649122808e-06, "loss": 0.24475271999835968, "num_input_tokens_seen": 2956378, "step": 1049, "train_runtime": 3468.054, "train_tokens_per_second": 852.46 }, { "epoch": 0.08437292834327728, "grad_norm": 2.979227066040039, "learning_rate": 9.533834586466166e-06, "loss": 0.3251151442527771, "num_input_tokens_seen": 2960138, "step": 1050, "train_runtime": 3470.8335, "train_tokens_per_second": 852.861 }, { "epoch": 0.08445328351312803, "grad_norm": 2.301772117614746, "learning_rate": 9.523809523809525e-06, "loss": 0.2732616066932678, "num_input_tokens_seen": 2962108, "step": 1051, "train_runtime": 3484.5911, "train_tokens_per_second": 850.059 }, { "epoch": 0.08453363868297877, "grad_norm": 3.0667169094085693, "learning_rate": 9.513784461152883e-06, "loss": 0.2762908935546875, "num_input_tokens_seen": 2964582, "step": 1052, "train_runtime": 3487.2211, "train_tokens_per_second": 850.127 }, { "epoch": 0.0846139938528295, "grad_norm": 2.809988021850586, "learning_rate": 9.503759398496242e-06, "loss": 0.25662127137184143, "num_input_tokens_seen": 2966588, "step": 1053, "train_runtime": 3489.5777, "train_tokens_per_second": 850.128 }, { "epoch": 0.08469434902268025, "grad_norm": 3.0787670612335205, "learning_rate": 9.4937343358396e-06, "loss": 0.18731394410133362, "num_input_tokens_seen": 2968546, "step": 1054, "train_runtime": 3492.0246, "train_tokens_per_second": 850.093 }, { "epoch": 0.08477470419253098, "grad_norm": 3.279913902282715, "learning_rate": 9.483709273182959e-06, "loss": 0.21910035610198975, "num_input_tokens_seen": 2971032, "step": 1055, "train_runtime": 3494.5687, "train_tokens_per_second": 850.186 }, { "epoch": 0.08485505936238173, "grad_norm": 3.3003053665161133, "learning_rate": 9.473684210526315e-06, "loss": 0.3318665027618408, "num_input_tokens_seen": 2972554, "step": 1056, "train_runtime": 3496.8565, "train_tokens_per_second": 850.065 }, { "epoch": 0.08493541453223247, "grad_norm": 3.7690956592559814, "learning_rate": 9.463659147869676e-06, "loss": 0.25538554787635803, "num_input_tokens_seen": 2973722, "step": 1057, "train_runtime": 3499.0468, "train_tokens_per_second": 849.866 }, { "epoch": 0.0850157697020832, "grad_norm": 3.3318698406219482, "learning_rate": 9.453634085213034e-06, "loss": 0.3181800842285156, "num_input_tokens_seen": 2977500, "step": 1058, "train_runtime": 3501.8686, "train_tokens_per_second": 850.26 }, { "epoch": 0.08509612487193395, "grad_norm": 2.618157148361206, "learning_rate": 9.443609022556391e-06, "loss": 0.2461847960948944, "num_input_tokens_seen": 2979654, "step": 1059, "train_runtime": 3504.2864, "train_tokens_per_second": 850.288 }, { "epoch": 0.08517648004178469, "grad_norm": 2.746332883834839, "learning_rate": 9.43358395989975e-06, "loss": 0.2694500982761383, "num_input_tokens_seen": 2982084, "step": 1060, "train_runtime": 3506.8156, "train_tokens_per_second": 850.368 }, { "epoch": 0.08525683521163543, "grad_norm": 2.402766704559326, "learning_rate": 9.423558897243108e-06, "loss": 0.17988912761211395, "num_input_tokens_seen": 2983730, "step": 1061, "train_runtime": 3509.0712, "train_tokens_per_second": 850.291 }, { "epoch": 0.08533719038148617, "grad_norm": 3.228522300720215, "learning_rate": 9.413533834586466e-06, "loss": 0.300514817237854, "num_input_tokens_seen": 2987404, "step": 1062, "train_runtime": 3511.8355, "train_tokens_per_second": 850.667 }, { "epoch": 0.0854175455513369, "grad_norm": 2.8541412353515625, "learning_rate": 9.403508771929825e-06, "loss": 0.3240264058113098, "num_input_tokens_seen": 2989606, "step": 1063, "train_runtime": 3514.23, "train_tokens_per_second": 850.714 }, { "epoch": 0.08549790072118765, "grad_norm": 2.663964033126831, "learning_rate": 9.393483709273183e-06, "loss": 0.23833400011062622, "num_input_tokens_seen": 2993028, "step": 1064, "train_runtime": 3516.9622, "train_tokens_per_second": 851.026 }, { "epoch": 0.08557825589103839, "grad_norm": 3.504387855529785, "learning_rate": 9.383458646616542e-06, "loss": 0.3533208966255188, "num_input_tokens_seen": 2998372, "step": 1065, "train_runtime": 3520.2963, "train_tokens_per_second": 851.739 }, { "epoch": 0.08565861106088914, "grad_norm": 3.0442399978637695, "learning_rate": 9.3734335839599e-06, "loss": 0.38174551725387573, "num_input_tokens_seen": 3000038, "step": 1066, "train_runtime": 3522.5687, "train_tokens_per_second": 851.662 }, { "epoch": 0.08573896623073987, "grad_norm": 3.1089744567871094, "learning_rate": 9.363408521303259e-06, "loss": 0.3180496394634247, "num_input_tokens_seen": 3002902, "step": 1067, "train_runtime": 3525.1506, "train_tokens_per_second": 851.851 }, { "epoch": 0.0858193214005906, "grad_norm": 3.20108699798584, "learning_rate": 9.353383458646617e-06, "loss": 0.29105740785598755, "num_input_tokens_seen": 3004518, "step": 1068, "train_runtime": 3527.4505, "train_tokens_per_second": 851.753 }, { "epoch": 0.08589967657044136, "grad_norm": 2.801586627960205, "learning_rate": 9.343358395989976e-06, "loss": 0.28173720836639404, "num_input_tokens_seen": 3006432, "step": 1069, "train_runtime": 3529.8545, "train_tokens_per_second": 851.716 }, { "epoch": 0.08598003174029209, "grad_norm": 2.9772539138793945, "learning_rate": 9.333333333333334e-06, "loss": 0.24833260476589203, "num_input_tokens_seen": 3009028, "step": 1070, "train_runtime": 3532.3964, "train_tokens_per_second": 851.838 }, { "epoch": 0.08606038691014282, "grad_norm": 2.633957862854004, "learning_rate": 9.323308270676693e-06, "loss": 0.28082433342933655, "num_input_tokens_seen": 3011662, "step": 1071, "train_runtime": 3534.8985, "train_tokens_per_second": 851.98 }, { "epoch": 0.08614074207999357, "grad_norm": 2.510115623474121, "learning_rate": 9.313283208020051e-06, "loss": 0.2830372750759125, "num_input_tokens_seen": 3014518, "step": 1072, "train_runtime": 3537.4674, "train_tokens_per_second": 852.168 }, { "epoch": 0.08622109724984431, "grad_norm": 2.6606836318969727, "learning_rate": 9.30325814536341e-06, "loss": 0.26171478629112244, "num_input_tokens_seen": 3018554, "step": 1073, "train_runtime": 3540.3229, "train_tokens_per_second": 852.621 }, { "epoch": 0.08630145241969506, "grad_norm": 2.6702425479888916, "learning_rate": 9.293233082706767e-06, "loss": 0.2230178713798523, "num_input_tokens_seen": 3020706, "step": 1074, "train_runtime": 3542.6901, "train_tokens_per_second": 852.659 }, { "epoch": 0.08638180758954579, "grad_norm": 3.058614492416382, "learning_rate": 9.283208020050127e-06, "loss": 0.27942901849746704, "num_input_tokens_seen": 3024460, "step": 1075, "train_runtime": 3545.462, "train_tokens_per_second": 853.051 }, { "epoch": 0.08646216275939653, "grad_norm": 2.9077789783477783, "learning_rate": 9.273182957393484e-06, "loss": 0.22575780749320984, "num_input_tokens_seen": 3026108, "step": 1076, "train_runtime": 3547.7098, "train_tokens_per_second": 852.975 }, { "epoch": 0.08654251792924728, "grad_norm": 2.7554054260253906, "learning_rate": 9.263157894736842e-06, "loss": 0.27153584361076355, "num_input_tokens_seen": 3028578, "step": 1077, "train_runtime": 3550.2501, "train_tokens_per_second": 853.06 }, { "epoch": 0.08662287309909801, "grad_norm": 3.188560724258423, "learning_rate": 9.253132832080202e-06, "loss": 0.22757884860038757, "num_input_tokens_seen": 3031566, "step": 1078, "train_runtime": 3552.877, "train_tokens_per_second": 853.271 }, { "epoch": 0.08670322826894876, "grad_norm": 3.3173720836639404, "learning_rate": 9.243107769423559e-06, "loss": 0.33387550711631775, "num_input_tokens_seen": 3034484, "step": 1079, "train_runtime": 3555.5567, "train_tokens_per_second": 853.448 }, { "epoch": 0.0867835834387995, "grad_norm": 3.155432939529419, "learning_rate": 9.233082706766918e-06, "loss": 0.23940780758857727, "num_input_tokens_seen": 3036764, "step": 1080, "train_runtime": 3558.0344, "train_tokens_per_second": 853.495 }, { "epoch": 0.08686393860865023, "grad_norm": 3.4120397567749023, "learning_rate": 9.223057644110276e-06, "loss": 0.2906268537044525, "num_input_tokens_seen": 3039028, "step": 1081, "train_runtime": 3572.1693, "train_tokens_per_second": 850.751 }, { "epoch": 0.08694429377850098, "grad_norm": 2.5552213191986084, "learning_rate": 9.213032581453634e-06, "loss": 0.2579834759235382, "num_input_tokens_seen": 3042544, "step": 1082, "train_runtime": 3574.962, "train_tokens_per_second": 851.07 }, { "epoch": 0.08702464894835171, "grad_norm": 2.9964680671691895, "learning_rate": 9.203007518796993e-06, "loss": 0.31168627738952637, "num_input_tokens_seen": 3045792, "step": 1083, "train_runtime": 3577.6574, "train_tokens_per_second": 851.337 }, { "epoch": 0.08710500411820246, "grad_norm": 2.9274113178253174, "learning_rate": 9.192982456140351e-06, "loss": 0.32569068670272827, "num_input_tokens_seen": 3052498, "step": 1084, "train_runtime": 3581.2956, "train_tokens_per_second": 852.345 }, { "epoch": 0.0871853592880532, "grad_norm": 2.2886106967926025, "learning_rate": 9.18295739348371e-06, "loss": 0.1805725246667862, "num_input_tokens_seen": 3054488, "step": 1085, "train_runtime": 3583.7674, "train_tokens_per_second": 852.312 }, { "epoch": 0.08726571445790393, "grad_norm": 3.319037437438965, "learning_rate": 9.172932330827068e-06, "loss": 0.2565021216869354, "num_input_tokens_seen": 3056842, "step": 1086, "train_runtime": 3586.2705, "train_tokens_per_second": 852.374 }, { "epoch": 0.08734606962775468, "grad_norm": 2.804027795791626, "learning_rate": 9.162907268170427e-06, "loss": 0.31864815950393677, "num_input_tokens_seen": 3058518, "step": 1087, "train_runtime": 3588.6604, "train_tokens_per_second": 852.273 }, { "epoch": 0.08742642479760541, "grad_norm": 2.6863903999328613, "learning_rate": 9.152882205513785e-06, "loss": 0.2194186896085739, "num_input_tokens_seen": 3062620, "step": 1088, "train_runtime": 3591.5222, "train_tokens_per_second": 852.736 }, { "epoch": 0.08750677996745615, "grad_norm": 2.8325083255767822, "learning_rate": 9.142857142857144e-06, "loss": 0.23367822170257568, "num_input_tokens_seen": 3065360, "step": 1089, "train_runtime": 3594.1897, "train_tokens_per_second": 852.865 }, { "epoch": 0.0875871351373069, "grad_norm": 2.2934718132019043, "learning_rate": 9.132832080200502e-06, "loss": 0.2181030660867691, "num_input_tokens_seen": 3069040, "step": 1090, "train_runtime": 3597.0127, "train_tokens_per_second": 853.219 }, { "epoch": 0.08766749030715763, "grad_norm": 1.8729697465896606, "learning_rate": 9.12280701754386e-06, "loss": 0.18227432668209076, "num_input_tokens_seen": 3072316, "step": 1091, "train_runtime": 3599.7197, "train_tokens_per_second": 853.488 }, { "epoch": 0.08774784547700838, "grad_norm": 2.8503057956695557, "learning_rate": 9.11278195488722e-06, "loss": 0.2774672508239746, "num_input_tokens_seen": 3075542, "step": 1092, "train_runtime": 3602.3991, "train_tokens_per_second": 853.748 }, { "epoch": 0.08782820064685912, "grad_norm": 3.327531099319458, "learning_rate": 9.102756892230578e-06, "loss": 0.3142390847206116, "num_input_tokens_seen": 3077334, "step": 1093, "train_runtime": 3604.7504, "train_tokens_per_second": 853.689 }, { "epoch": 0.08790855581670985, "grad_norm": 2.7247509956359863, "learning_rate": 9.092731829573935e-06, "loss": 0.31622743606567383, "num_input_tokens_seen": 3080638, "step": 1094, "train_runtime": 3607.4146, "train_tokens_per_second": 853.974 }, { "epoch": 0.0879889109865606, "grad_norm": 2.4906461238861084, "learning_rate": 9.082706766917295e-06, "loss": 0.17167414724826813, "num_input_tokens_seen": 3083748, "step": 1095, "train_runtime": 3610.0722, "train_tokens_per_second": 854.207 }, { "epoch": 0.08806926615641134, "grad_norm": 2.8842532634735107, "learning_rate": 9.072681704260652e-06, "loss": 0.25497299432754517, "num_input_tokens_seen": 3087130, "step": 1096, "train_runtime": 3612.7723, "train_tokens_per_second": 854.504 }, { "epoch": 0.08814962132626208, "grad_norm": 3.0470404624938965, "learning_rate": 9.06265664160401e-06, "loss": 0.29288047552108765, "num_input_tokens_seen": 3088470, "step": 1097, "train_runtime": 3614.9941, "train_tokens_per_second": 854.35 }, { "epoch": 0.08822997649611282, "grad_norm": 2.4805948734283447, "learning_rate": 9.05263157894737e-06, "loss": 0.2315695881843567, "num_input_tokens_seen": 3091986, "step": 1098, "train_runtime": 3617.7101, "train_tokens_per_second": 854.68 }, { "epoch": 0.08831033166596355, "grad_norm": 2.4088752269744873, "learning_rate": 9.042606516290727e-06, "loss": 0.28510117530822754, "num_input_tokens_seen": 3094952, "step": 1099, "train_runtime": 3620.3129, "train_tokens_per_second": 854.885 }, { "epoch": 0.0883906868358143, "grad_norm": 2.9681971073150635, "learning_rate": 9.032581453634086e-06, "loss": 0.26909980177879333, "num_input_tokens_seen": 3097312, "step": 1100, "train_runtime": 3622.7856, "train_tokens_per_second": 854.953 }, { "epoch": 0.08847104200566504, "grad_norm": 2.2669382095336914, "learning_rate": 9.022556390977444e-06, "loss": 0.24719354510307312, "num_input_tokens_seen": 3098980, "step": 1101, "train_runtime": 3625.0325, "train_tokens_per_second": 854.883 }, { "epoch": 0.08855139717551579, "grad_norm": 4.001906394958496, "learning_rate": 9.012531328320803e-06, "loss": 0.35922515392303467, "num_input_tokens_seen": 3102620, "step": 1102, "train_runtime": 3627.7766, "train_tokens_per_second": 855.24 }, { "epoch": 0.08863175234536652, "grad_norm": 3.7869842052459717, "learning_rate": 9.002506265664161e-06, "loss": 0.25262606143951416, "num_input_tokens_seen": 3104730, "step": 1103, "train_runtime": 3630.1464, "train_tokens_per_second": 855.263 }, { "epoch": 0.08871210751521726, "grad_norm": 2.6303040981292725, "learning_rate": 8.99248120300752e-06, "loss": 0.26815345883369446, "num_input_tokens_seen": 3106440, "step": 1104, "train_runtime": 3632.4604, "train_tokens_per_second": 855.189 }, { "epoch": 0.088792462685068, "grad_norm": 2.8536343574523926, "learning_rate": 8.982456140350878e-06, "loss": 0.27207204699516296, "num_input_tokens_seen": 3108826, "step": 1105, "train_runtime": 3634.893, "train_tokens_per_second": 855.273 }, { "epoch": 0.08887281785491874, "grad_norm": 2.915250301361084, "learning_rate": 8.972431077694236e-06, "loss": 0.2552780508995056, "num_input_tokens_seen": 3112392, "step": 1106, "train_runtime": 3637.643, "train_tokens_per_second": 855.607 }, { "epoch": 0.08895317302476947, "grad_norm": 2.6338889598846436, "learning_rate": 8.962406015037595e-06, "loss": 0.3244155943393707, "num_input_tokens_seen": 3115802, "step": 1107, "train_runtime": 3640.3851, "train_tokens_per_second": 855.899 }, { "epoch": 0.08903352819462022, "grad_norm": 3.3019986152648926, "learning_rate": 8.952380952380953e-06, "loss": 0.31214478611946106, "num_input_tokens_seen": 3118826, "step": 1108, "train_runtime": 3642.9999, "train_tokens_per_second": 856.115 }, { "epoch": 0.08911388336447096, "grad_norm": 2.15661883354187, "learning_rate": 8.942355889724312e-06, "loss": 0.1894627809524536, "num_input_tokens_seen": 3122342, "step": 1109, "train_runtime": 3645.7285, "train_tokens_per_second": 856.438 }, { "epoch": 0.08919423853432171, "grad_norm": 3.273404359817505, "learning_rate": 8.93233082706767e-06, "loss": 0.30990320444107056, "num_input_tokens_seen": 3125338, "step": 1110, "train_runtime": 3648.3788, "train_tokens_per_second": 856.637 }, { "epoch": 0.08927459370417244, "grad_norm": 3.169673442840576, "learning_rate": 8.922305764411027e-06, "loss": 0.25750330090522766, "num_input_tokens_seen": 3128180, "step": 1111, "train_runtime": 3662.8964, "train_tokens_per_second": 854.018 }, { "epoch": 0.08935494887402318, "grad_norm": 2.9008629322052, "learning_rate": 8.912280701754387e-06, "loss": 0.2972443103790283, "num_input_tokens_seen": 3131596, "step": 1112, "train_runtime": 3665.6207, "train_tokens_per_second": 854.315 }, { "epoch": 0.08943530404387393, "grad_norm": 2.5140528678894043, "learning_rate": 8.902255639097746e-06, "loss": 0.22682245075702667, "num_input_tokens_seen": 3134412, "step": 1113, "train_runtime": 3668.1595, "train_tokens_per_second": 854.492 }, { "epoch": 0.08951565921372466, "grad_norm": 3.0421128273010254, "learning_rate": 8.892230576441103e-06, "loss": 0.25906410813331604, "num_input_tokens_seen": 3137504, "step": 1114, "train_runtime": 3670.7777, "train_tokens_per_second": 854.725 }, { "epoch": 0.08959601438357541, "grad_norm": 3.2030932903289795, "learning_rate": 8.882205513784463e-06, "loss": 0.30559325218200684, "num_input_tokens_seen": 3139896, "step": 1115, "train_runtime": 3673.2219, "train_tokens_per_second": 854.807 }, { "epoch": 0.08967636955342614, "grad_norm": 3.059091567993164, "learning_rate": 8.87218045112782e-06, "loss": 0.30283451080322266, "num_input_tokens_seen": 3143684, "step": 1116, "train_runtime": 3676.0851, "train_tokens_per_second": 855.172 }, { "epoch": 0.08975672472327688, "grad_norm": 2.613288402557373, "learning_rate": 8.862155388471178e-06, "loss": 0.2585943341255188, "num_input_tokens_seen": 3146556, "step": 1117, "train_runtime": 3678.6523, "train_tokens_per_second": 855.356 }, { "epoch": 0.08983707989312763, "grad_norm": 3.0170950889587402, "learning_rate": 8.852130325814537e-06, "loss": 0.27338603138923645, "num_input_tokens_seen": 3150956, "step": 1118, "train_runtime": 3681.7208, "train_tokens_per_second": 855.838 }, { "epoch": 0.08991743506297836, "grad_norm": 2.3550713062286377, "learning_rate": 8.842105263157895e-06, "loss": 0.2898193597793579, "num_input_tokens_seen": 3153398, "step": 1119, "train_runtime": 3684.2711, "train_tokens_per_second": 855.908 }, { "epoch": 0.08999779023282911, "grad_norm": 2.919173240661621, "learning_rate": 8.832080200501254e-06, "loss": 0.27399080991744995, "num_input_tokens_seen": 3155970, "step": 1120, "train_runtime": 3686.8677, "train_tokens_per_second": 856.003 }, { "epoch": 0.09007814540267985, "grad_norm": 2.9430360794067383, "learning_rate": 8.822055137844612e-06, "loss": 0.3093157708644867, "num_input_tokens_seen": 3159768, "step": 1121, "train_runtime": 3689.766, "train_tokens_per_second": 856.36 }, { "epoch": 0.09015850057253058, "grad_norm": 2.7869479656219482, "learning_rate": 8.81203007518797e-06, "loss": 0.3200400769710541, "num_input_tokens_seen": 3161724, "step": 1122, "train_runtime": 3692.1381, "train_tokens_per_second": 856.34 }, { "epoch": 0.09023885574238133, "grad_norm": 2.9397168159484863, "learning_rate": 8.802005012531329e-06, "loss": 0.29338183999061584, "num_input_tokens_seen": 3166170, "step": 1123, "train_runtime": 3695.113, "train_tokens_per_second": 856.853 }, { "epoch": 0.09031921091223206, "grad_norm": 4.112015724182129, "learning_rate": 8.791979949874688e-06, "loss": 0.30180227756500244, "num_input_tokens_seen": 3167388, "step": 1124, "train_runtime": 3697.2695, "train_tokens_per_second": 856.683 }, { "epoch": 0.0903995660820828, "grad_norm": 2.7407000064849854, "learning_rate": 8.781954887218046e-06, "loss": 0.3127600848674774, "num_input_tokens_seen": 3170298, "step": 1125, "train_runtime": 3699.872, "train_tokens_per_second": 856.867 }, { "epoch": 0.09047992125193355, "grad_norm": 3.0575273036956787, "learning_rate": 8.771929824561405e-06, "loss": 0.28493809700012207, "num_input_tokens_seen": 3172614, "step": 1126, "train_runtime": 3702.3407, "train_tokens_per_second": 856.921 }, { "epoch": 0.09056027642178428, "grad_norm": 3.0020124912261963, "learning_rate": 8.761904761904763e-06, "loss": 0.2901063859462738, "num_input_tokens_seen": 3174992, "step": 1127, "train_runtime": 3704.7776, "train_tokens_per_second": 856.999 }, { "epoch": 0.09064063159163503, "grad_norm": 2.9716427326202393, "learning_rate": 8.751879699248122e-06, "loss": 0.29861336946487427, "num_input_tokens_seen": 3178492, "step": 1128, "train_runtime": 3707.5408, "train_tokens_per_second": 857.305 }, { "epoch": 0.09072098676148577, "grad_norm": 2.472208023071289, "learning_rate": 8.741854636591478e-06, "loss": 0.23268243670463562, "num_input_tokens_seen": 3181154, "step": 1129, "train_runtime": 3710.0707, "train_tokens_per_second": 857.438 }, { "epoch": 0.0908013419313365, "grad_norm": 3.1195871829986572, "learning_rate": 8.731829573934839e-06, "loss": 0.31297188997268677, "num_input_tokens_seen": 3184244, "step": 1130, "train_runtime": 3712.7338, "train_tokens_per_second": 857.655 }, { "epoch": 0.09088169710118725, "grad_norm": 2.9426820278167725, "learning_rate": 8.721804511278195e-06, "loss": 0.2540207505226135, "num_input_tokens_seen": 3186506, "step": 1131, "train_runtime": 3715.235, "train_tokens_per_second": 857.686 }, { "epoch": 0.09096205227103799, "grad_norm": 3.344089984893799, "learning_rate": 8.711779448621554e-06, "loss": 0.30014848709106445, "num_input_tokens_seen": 3189356, "step": 1132, "train_runtime": 3717.9208, "train_tokens_per_second": 857.833 }, { "epoch": 0.09104240744088873, "grad_norm": 2.444239377975464, "learning_rate": 8.701754385964914e-06, "loss": 0.23984336853027344, "num_input_tokens_seen": 3191756, "step": 1133, "train_runtime": 3720.4532, "train_tokens_per_second": 857.894 }, { "epoch": 0.09112276261073947, "grad_norm": 3.2191689014434814, "learning_rate": 8.69172932330827e-06, "loss": 0.33059006929397583, "num_input_tokens_seen": 3193550, "step": 1134, "train_runtime": 3722.8601, "train_tokens_per_second": 857.822 }, { "epoch": 0.0912031177805902, "grad_norm": 2.0156381130218506, "learning_rate": 8.68170426065163e-06, "loss": 0.18222689628601074, "num_input_tokens_seen": 3196928, "step": 1135, "train_runtime": 3725.5991, "train_tokens_per_second": 858.098 }, { "epoch": 0.09128347295044095, "grad_norm": 2.3665623664855957, "learning_rate": 8.671679197994988e-06, "loss": 0.2675105333328247, "num_input_tokens_seen": 3199560, "step": 1136, "train_runtime": 3728.1, "train_tokens_per_second": 858.228 }, { "epoch": 0.09136382812029169, "grad_norm": 2.4238121509552, "learning_rate": 8.661654135338346e-06, "loss": 0.26898425817489624, "num_input_tokens_seen": 3203656, "step": 1137, "train_runtime": 3730.9707, "train_tokens_per_second": 858.666 }, { "epoch": 0.09144418329014244, "grad_norm": 2.7386136054992676, "learning_rate": 8.651629072681705e-06, "loss": 0.14335662126541138, "num_input_tokens_seen": 3205826, "step": 1138, "train_runtime": 3733.4185, "train_tokens_per_second": 858.684 }, { "epoch": 0.09152453845999317, "grad_norm": 3.0974209308624268, "learning_rate": 8.641604010025063e-06, "loss": 0.30975887179374695, "num_input_tokens_seen": 3208286, "step": 1139, "train_runtime": 3735.8916, "train_tokens_per_second": 858.774 }, { "epoch": 0.0916048936298439, "grad_norm": 3.4228551387786865, "learning_rate": 8.631578947368422e-06, "loss": 0.32829299569129944, "num_input_tokens_seen": 3210886, "step": 1140, "train_runtime": 3738.4119, "train_tokens_per_second": 858.89 }, { "epoch": 0.09168524879969465, "grad_norm": 2.7802693843841553, "learning_rate": 8.62155388471178e-06, "loss": 0.33544057607650757, "num_input_tokens_seen": 3213498, "step": 1141, "train_runtime": 3752.9271, "train_tokens_per_second": 856.264 }, { "epoch": 0.09176560396954539, "grad_norm": 3.2307186126708984, "learning_rate": 8.611528822055139e-06, "loss": 0.292299747467041, "num_input_tokens_seen": 3214980, "step": 1142, "train_runtime": 3755.1962, "train_tokens_per_second": 856.142 }, { "epoch": 0.09184595913939612, "grad_norm": 3.1838762760162354, "learning_rate": 8.601503759398497e-06, "loss": 0.29825225472450256, "num_input_tokens_seen": 3218510, "step": 1143, "train_runtime": 3757.9413, "train_tokens_per_second": 856.456 }, { "epoch": 0.09192631430924687, "grad_norm": 2.90578556060791, "learning_rate": 8.591478696741856e-06, "loss": 0.25112801790237427, "num_input_tokens_seen": 3221156, "step": 1144, "train_runtime": 3760.4579, "train_tokens_per_second": 856.586 }, { "epoch": 0.09200666947909761, "grad_norm": 2.946054458618164, "learning_rate": 8.581453634085214e-06, "loss": 0.31097450852394104, "num_input_tokens_seen": 3224030, "step": 1145, "train_runtime": 3763.0695, "train_tokens_per_second": 856.755 }, { "epoch": 0.09208702464894836, "grad_norm": 3.8333687782287598, "learning_rate": 8.571428571428571e-06, "loss": 0.36663222312927246, "num_input_tokens_seen": 3226330, "step": 1146, "train_runtime": 3765.5221, "train_tokens_per_second": 856.808 }, { "epoch": 0.09216737981879909, "grad_norm": 3.2993428707122803, "learning_rate": 8.561403508771931e-06, "loss": 0.32325536012649536, "num_input_tokens_seen": 3228622, "step": 1147, "train_runtime": 3768.0154, "train_tokens_per_second": 856.849 }, { "epoch": 0.09224773498864983, "grad_norm": 3.1793479919433594, "learning_rate": 8.55137844611529e-06, "loss": 0.2385995239019394, "num_input_tokens_seen": 3231396, "step": 1148, "train_runtime": 3770.6035, "train_tokens_per_second": 856.997 }, { "epoch": 0.09232809015850058, "grad_norm": 2.9457011222839355, "learning_rate": 8.541353383458646e-06, "loss": 0.28244975209236145, "num_input_tokens_seen": 3234578, "step": 1149, "train_runtime": 3773.3179, "train_tokens_per_second": 857.224 }, { "epoch": 0.09240844532835131, "grad_norm": 2.9778213500976562, "learning_rate": 8.531328320802007e-06, "loss": 0.23317573964595795, "num_input_tokens_seen": 3237168, "step": 1150, "train_runtime": 3775.8023, "train_tokens_per_second": 857.346 }, { "epoch": 0.09248880049820206, "grad_norm": 2.9976768493652344, "learning_rate": 8.521303258145363e-06, "loss": 0.327251672744751, "num_input_tokens_seen": 3239038, "step": 1151, "train_runtime": 3778.13, "train_tokens_per_second": 857.312 }, { "epoch": 0.0925691556680528, "grad_norm": 3.3894107341766357, "learning_rate": 8.511278195488722e-06, "loss": 0.2839146852493286, "num_input_tokens_seen": 3242032, "step": 1152, "train_runtime": 3780.7731, "train_tokens_per_second": 857.505 }, { "epoch": 0.09264951083790353, "grad_norm": 2.642024040222168, "learning_rate": 8.501253132832082e-06, "loss": 0.232076033949852, "num_input_tokens_seen": 3245250, "step": 1153, "train_runtime": 3783.4712, "train_tokens_per_second": 857.744 }, { "epoch": 0.09272986600775428, "grad_norm": 3.0953457355499268, "learning_rate": 8.491228070175439e-06, "loss": 0.2820534110069275, "num_input_tokens_seen": 3247936, "step": 1154, "train_runtime": 3786.0155, "train_tokens_per_second": 857.877 }, { "epoch": 0.09281022117760501, "grad_norm": 3.054499864578247, "learning_rate": 8.481203007518797e-06, "loss": 0.2444484531879425, "num_input_tokens_seen": 3250066, "step": 1155, "train_runtime": 3788.3878, "train_tokens_per_second": 857.902 }, { "epoch": 0.09289057634745576, "grad_norm": 3.5810904502868652, "learning_rate": 8.471177944862156e-06, "loss": 0.27406686544418335, "num_input_tokens_seen": 3251356, "step": 1156, "train_runtime": 3790.6537, "train_tokens_per_second": 857.73 }, { "epoch": 0.0929709315173065, "grad_norm": 2.8003427982330322, "learning_rate": 8.461152882205514e-06, "loss": 0.29646754264831543, "num_input_tokens_seen": 3253502, "step": 1157, "train_runtime": 3793.0769, "train_tokens_per_second": 857.747 }, { "epoch": 0.09305128668715723, "grad_norm": 2.714930772781372, "learning_rate": 8.451127819548873e-06, "loss": 0.3359104096889496, "num_input_tokens_seen": 3256928, "step": 1158, "train_runtime": 3795.8049, "train_tokens_per_second": 858.034 }, { "epoch": 0.09313164185700798, "grad_norm": 2.7674224376678467, "learning_rate": 8.441102756892231e-06, "loss": 0.19646523892879486, "num_input_tokens_seen": 3259598, "step": 1159, "train_runtime": 3798.419, "train_tokens_per_second": 858.146 }, { "epoch": 0.09321199702685871, "grad_norm": 3.0799527168273926, "learning_rate": 8.43107769423559e-06, "loss": 0.31579288840293884, "num_input_tokens_seen": 3261912, "step": 1160, "train_runtime": 3800.8599, "train_tokens_per_second": 858.204 }, { "epoch": 0.09329235219670945, "grad_norm": 3.373354196548462, "learning_rate": 8.421052631578948e-06, "loss": 0.23240306973457336, "num_input_tokens_seen": 3264176, "step": 1161, "train_runtime": 3803.3007, "train_tokens_per_second": 858.248 }, { "epoch": 0.0933727073665602, "grad_norm": 1.7892944812774658, "learning_rate": 8.411027568922307e-06, "loss": 0.1150800883769989, "num_input_tokens_seen": 3266536, "step": 1162, "train_runtime": 3805.7403, "train_tokens_per_second": 858.318 }, { "epoch": 0.09345306253641093, "grad_norm": 2.6740939617156982, "learning_rate": 8.401002506265665e-06, "loss": 0.22116731107234955, "num_input_tokens_seen": 3268722, "step": 1163, "train_runtime": 3808.1706, "train_tokens_per_second": 858.344 }, { "epoch": 0.09353341770626168, "grad_norm": 3.2153854370117188, "learning_rate": 8.390977443609024e-06, "loss": 0.3411394953727722, "num_input_tokens_seen": 3272210, "step": 1164, "train_runtime": 3810.9048, "train_tokens_per_second": 858.644 }, { "epoch": 0.09361377287611242, "grad_norm": 3.2527852058410645, "learning_rate": 8.380952380952382e-06, "loss": 0.2883263826370239, "num_input_tokens_seen": 3274874, "step": 1165, "train_runtime": 3813.4285, "train_tokens_per_second": 858.774 }, { "epoch": 0.09369412804596315, "grad_norm": 2.4813098907470703, "learning_rate": 8.370927318295739e-06, "loss": 0.20611388981342316, "num_input_tokens_seen": 3278162, "step": 1166, "train_runtime": 3816.1033, "train_tokens_per_second": 859.034 }, { "epoch": 0.0937744832158139, "grad_norm": 3.446521759033203, "learning_rate": 8.3609022556391e-06, "loss": 0.2597815990447998, "num_input_tokens_seen": 3280158, "step": 1167, "train_runtime": 3818.4741, "train_tokens_per_second": 859.023 }, { "epoch": 0.09385483838566464, "grad_norm": 2.654390811920166, "learning_rate": 8.350877192982458e-06, "loss": 0.24222707748413086, "num_input_tokens_seen": 3283160, "step": 1168, "train_runtime": 3821.1105, "train_tokens_per_second": 859.216 }, { "epoch": 0.09393519355551538, "grad_norm": 2.7661592960357666, "learning_rate": 8.340852130325814e-06, "loss": 0.32253751158714294, "num_input_tokens_seen": 3285000, "step": 1169, "train_runtime": 3823.452, "train_tokens_per_second": 859.171 }, { "epoch": 0.09401554872536612, "grad_norm": 2.8930723667144775, "learning_rate": 8.330827067669175e-06, "loss": 0.2635021209716797, "num_input_tokens_seen": 3287900, "step": 1170, "train_runtime": 3826.1352, "train_tokens_per_second": 859.327 }, { "epoch": 0.09409590389521685, "grad_norm": 3.0205187797546387, "learning_rate": 8.320802005012531e-06, "loss": 0.25306233763694763, "num_input_tokens_seen": 3290116, "step": 1171, "train_runtime": 3841.0161, "train_tokens_per_second": 856.574 }, { "epoch": 0.0941762590650676, "grad_norm": 2.744584321975708, "learning_rate": 8.31077694235589e-06, "loss": 0.22004106640815735, "num_input_tokens_seen": 3292660, "step": 1172, "train_runtime": 3843.6269, "train_tokens_per_second": 856.654 }, { "epoch": 0.09425661423491834, "grad_norm": 2.230405569076538, "learning_rate": 8.300751879699248e-06, "loss": 0.13979382812976837, "num_input_tokens_seen": 3294330, "step": 1173, "train_runtime": 3846.0933, "train_tokens_per_second": 856.539 }, { "epoch": 0.09433696940476909, "grad_norm": 2.8869242668151855, "learning_rate": 8.290726817042607e-06, "loss": 0.23333315551280975, "num_input_tokens_seen": 3298498, "step": 1174, "train_runtime": 3849.1124, "train_tokens_per_second": 856.95 }, { "epoch": 0.09441732457461982, "grad_norm": 3.0218491554260254, "learning_rate": 8.280701754385965e-06, "loss": 0.334041953086853, "num_input_tokens_seen": 3301164, "step": 1175, "train_runtime": 3851.6658, "train_tokens_per_second": 857.074 }, { "epoch": 0.09449767974447056, "grad_norm": 3.9566650390625, "learning_rate": 8.270676691729324e-06, "loss": 0.40120652318000793, "num_input_tokens_seen": 3305362, "step": 1176, "train_runtime": 3854.5647, "train_tokens_per_second": 857.519 }, { "epoch": 0.0945780349143213, "grad_norm": 3.423213243484497, "learning_rate": 8.260651629072682e-06, "loss": 0.25451239943504333, "num_input_tokens_seen": 3308328, "step": 1177, "train_runtime": 3857.2122, "train_tokens_per_second": 857.699 }, { "epoch": 0.09465839008417204, "grad_norm": 2.8770368099212646, "learning_rate": 8.25062656641604e-06, "loss": 0.30318745970726013, "num_input_tokens_seen": 3310002, "step": 1178, "train_runtime": 3859.4792, "train_tokens_per_second": 857.629 }, { "epoch": 0.09473874525402277, "grad_norm": 2.44460129737854, "learning_rate": 8.2406015037594e-06, "loss": 0.2280236780643463, "num_input_tokens_seen": 3312374, "step": 1179, "train_runtime": 3861.9394, "train_tokens_per_second": 857.697 }, { "epoch": 0.09481910042387352, "grad_norm": 3.0745186805725098, "learning_rate": 8.230576441102758e-06, "loss": 0.3264968991279602, "num_input_tokens_seen": 3314878, "step": 1180, "train_runtime": 3864.4301, "train_tokens_per_second": 857.792 }, { "epoch": 0.09489945559372426, "grad_norm": 3.383007764816284, "learning_rate": 8.220551378446116e-06, "loss": 0.3342382311820984, "num_input_tokens_seen": 3317542, "step": 1181, "train_runtime": 3867.0627, "train_tokens_per_second": 857.897 }, { "epoch": 0.094979810763575, "grad_norm": 2.721928596496582, "learning_rate": 8.210526315789475e-06, "loss": 0.2191690057516098, "num_input_tokens_seen": 3319704, "step": 1182, "train_runtime": 3869.5481, "train_tokens_per_second": 857.905 }, { "epoch": 0.09506016593342574, "grad_norm": 2.779350757598877, "learning_rate": 8.200501253132833e-06, "loss": 0.24846196174621582, "num_input_tokens_seen": 3323544, "step": 1183, "train_runtime": 3872.3964, "train_tokens_per_second": 858.265 }, { "epoch": 0.09514052110327648, "grad_norm": 3.1472787857055664, "learning_rate": 8.190476190476192e-06, "loss": 0.25563085079193115, "num_input_tokens_seen": 3325492, "step": 1184, "train_runtime": 3874.7701, "train_tokens_per_second": 858.242 }, { "epoch": 0.09522087627312723, "grad_norm": 2.750178098678589, "learning_rate": 8.18045112781955e-06, "loss": 0.28222528100013733, "num_input_tokens_seen": 3328700, "step": 1185, "train_runtime": 3877.4898, "train_tokens_per_second": 858.468 }, { "epoch": 0.09530123144297796, "grad_norm": 2.834480047225952, "learning_rate": 8.170426065162907e-06, "loss": 0.2701455354690552, "num_input_tokens_seen": 3331914, "step": 1186, "train_runtime": 3880.1776, "train_tokens_per_second": 858.701 }, { "epoch": 0.09538158661282871, "grad_norm": 2.6727957725524902, "learning_rate": 8.160401002506266e-06, "loss": 0.26388850808143616, "num_input_tokens_seen": 3334784, "step": 1187, "train_runtime": 3882.8961, "train_tokens_per_second": 858.839 }, { "epoch": 0.09546194178267944, "grad_norm": 3.5899851322174072, "learning_rate": 8.150375939849626e-06, "loss": 0.4315168857574463, "num_input_tokens_seen": 3338970, "step": 1188, "train_runtime": 3885.867, "train_tokens_per_second": 859.26 }, { "epoch": 0.09554229695253018, "grad_norm": 2.2768161296844482, "learning_rate": 8.140350877192983e-06, "loss": 0.27162060141563416, "num_input_tokens_seen": 3341986, "step": 1189, "train_runtime": 3888.5074, "train_tokens_per_second": 859.452 }, { "epoch": 0.09562265212238093, "grad_norm": 3.9469542503356934, "learning_rate": 8.130325814536341e-06, "loss": 0.31582677364349365, "num_input_tokens_seen": 3344786, "step": 1190, "train_runtime": 3891.0758, "train_tokens_per_second": 859.604 }, { "epoch": 0.09570300729223166, "grad_norm": 3.0052027702331543, "learning_rate": 8.1203007518797e-06, "loss": 0.27772459387779236, "num_input_tokens_seen": 3347440, "step": 1191, "train_runtime": 3893.5977, "train_tokens_per_second": 859.729 }, { "epoch": 0.09578336246208241, "grad_norm": 2.402719020843506, "learning_rate": 8.110275689223058e-06, "loss": 0.18139885365962982, "num_input_tokens_seen": 3349690, "step": 1192, "train_runtime": 3896.0964, "train_tokens_per_second": 859.755 }, { "epoch": 0.09586371763193315, "grad_norm": 3.154900550842285, "learning_rate": 8.100250626566416e-06, "loss": 0.26014313101768494, "num_input_tokens_seen": 3351398, "step": 1193, "train_runtime": 3898.4154, "train_tokens_per_second": 859.682 }, { "epoch": 0.09594407280178388, "grad_norm": 3.1407954692840576, "learning_rate": 8.090225563909775e-06, "loss": 0.3488783538341522, "num_input_tokens_seen": 3354666, "step": 1194, "train_runtime": 3901.1375, "train_tokens_per_second": 859.92 }, { "epoch": 0.09602442797163463, "grad_norm": 2.5230674743652344, "learning_rate": 8.080200501253133e-06, "loss": 0.25091272592544556, "num_input_tokens_seen": 3358072, "step": 1195, "train_runtime": 3903.866, "train_tokens_per_second": 860.191 }, { "epoch": 0.09610478314148536, "grad_norm": 3.295290470123291, "learning_rate": 8.070175438596492e-06, "loss": 0.28005850315093994, "num_input_tokens_seen": 3360466, "step": 1196, "train_runtime": 3906.3436, "train_tokens_per_second": 860.259 }, { "epoch": 0.0961851383113361, "grad_norm": 2.2329554557800293, "learning_rate": 8.06015037593985e-06, "loss": 0.24669934809207916, "num_input_tokens_seen": 3362618, "step": 1197, "train_runtime": 3908.7541, "train_tokens_per_second": 860.279 }, { "epoch": 0.09626549348118685, "grad_norm": 2.875924825668335, "learning_rate": 8.050125313283209e-06, "loss": 0.35174018144607544, "num_input_tokens_seen": 3366584, "step": 1198, "train_runtime": 3911.602, "train_tokens_per_second": 860.666 }, { "epoch": 0.09634584865103758, "grad_norm": 2.012368679046631, "learning_rate": 8.040100250626567e-06, "loss": 0.15056964755058289, "num_input_tokens_seen": 3368778, "step": 1199, "train_runtime": 3914.0224, "train_tokens_per_second": 860.695 }, { "epoch": 0.09642620382088833, "grad_norm": 2.8406872749328613, "learning_rate": 8.030075187969926e-06, "loss": 0.3512207269668579, "num_input_tokens_seen": 3371510, "step": 1200, "train_runtime": 3916.6015, "train_tokens_per_second": 860.825 }, { "epoch": 0.09642620382088833, "eval_loss": 1.109608769416809, "eval_runtime": 68.7896, "eval_samples_per_second": 14.45, "eval_steps_per_second": 7.225, "num_input_tokens_seen": 3371510, "step": 1200 }, { "epoch": 0.09650655899073907, "grad_norm": 2.5624406337738037, "learning_rate": 8.020050125313283e-06, "loss": 0.2971264719963074, "num_input_tokens_seen": 3374196, "step": 1201, "train_runtime": 3999.299, "train_tokens_per_second": 843.697 }, { "epoch": 0.0965869141605898, "grad_norm": 2.989069938659668, "learning_rate": 8.010025062656643e-06, "loss": 0.23883947730064392, "num_input_tokens_seen": 3377918, "step": 1202, "train_runtime": 4002.1412, "train_tokens_per_second": 844.028 }, { "epoch": 0.09666726933044055, "grad_norm": 3.273341417312622, "learning_rate": 8.000000000000001e-06, "loss": 0.3458550274372101, "num_input_tokens_seen": 3379654, "step": 1203, "train_runtime": 4004.4647, "train_tokens_per_second": 843.971 }, { "epoch": 0.09674762450029128, "grad_norm": 2.483314037322998, "learning_rate": 7.989974937343358e-06, "loss": 0.2504233717918396, "num_input_tokens_seen": 3383040, "step": 1204, "train_runtime": 4007.1983, "train_tokens_per_second": 844.241 }, { "epoch": 0.09682797967014203, "grad_norm": 3.1416079998016357, "learning_rate": 7.979949874686718e-06, "loss": 0.28146371245384216, "num_input_tokens_seen": 3385382, "step": 1205, "train_runtime": 4009.672, "train_tokens_per_second": 844.304 }, { "epoch": 0.09690833483999277, "grad_norm": 3.211169958114624, "learning_rate": 7.969924812030075e-06, "loss": 0.367956280708313, "num_input_tokens_seen": 3391238, "step": 1206, "train_runtime": 4012.9884, "train_tokens_per_second": 845.065 }, { "epoch": 0.0969886900098435, "grad_norm": 3.2680113315582275, "learning_rate": 7.959899749373434e-06, "loss": 0.3160693347454071, "num_input_tokens_seen": 3393532, "step": 1207, "train_runtime": 4015.4128, "train_tokens_per_second": 845.127 }, { "epoch": 0.09706904517969425, "grad_norm": 2.1852521896362305, "learning_rate": 7.949874686716794e-06, "loss": 0.19716721773147583, "num_input_tokens_seen": 3397874, "step": 1208, "train_runtime": 4018.3701, "train_tokens_per_second": 845.585 }, { "epoch": 0.09714940034954499, "grad_norm": 3.7123451232910156, "learning_rate": 7.93984962406015e-06, "loss": 0.2891196608543396, "num_input_tokens_seen": 3400938, "step": 1209, "train_runtime": 4021.0169, "train_tokens_per_second": 845.791 }, { "epoch": 0.09722975551939574, "grad_norm": 2.604372501373291, "learning_rate": 7.929824561403509e-06, "loss": 0.23744642734527588, "num_input_tokens_seen": 3403552, "step": 1210, "train_runtime": 4023.5348, "train_tokens_per_second": 845.911 }, { "epoch": 0.09731011068924647, "grad_norm": 3.140507698059082, "learning_rate": 7.919799498746868e-06, "loss": 0.2738659381866455, "num_input_tokens_seen": 3405574, "step": 1211, "train_runtime": 4025.8953, "train_tokens_per_second": 845.917 }, { "epoch": 0.0973904658590972, "grad_norm": 2.250427007675171, "learning_rate": 7.909774436090226e-06, "loss": 0.18239513039588928, "num_input_tokens_seen": 3409398, "step": 1212, "train_runtime": 4028.6887, "train_tokens_per_second": 846.28 }, { "epoch": 0.09747082102894795, "grad_norm": 3.2674877643585205, "learning_rate": 7.899749373433585e-06, "loss": 0.2691424787044525, "num_input_tokens_seen": 3411476, "step": 1213, "train_runtime": 4031.035, "train_tokens_per_second": 846.303 }, { "epoch": 0.09755117619879869, "grad_norm": 3.0355498790740967, "learning_rate": 7.889724310776943e-06, "loss": 0.24186620116233826, "num_input_tokens_seen": 3414032, "step": 1214, "train_runtime": 4033.5487, "train_tokens_per_second": 846.409 }, { "epoch": 0.09763153136864942, "grad_norm": 3.833584785461426, "learning_rate": 7.879699248120301e-06, "loss": 0.31686660647392273, "num_input_tokens_seen": 3416512, "step": 1215, "train_runtime": 4035.9935, "train_tokens_per_second": 846.511 }, { "epoch": 0.09771188653850017, "grad_norm": 2.64996337890625, "learning_rate": 7.86967418546366e-06, "loss": 0.19599565863609314, "num_input_tokens_seen": 3419360, "step": 1216, "train_runtime": 4038.5612, "train_tokens_per_second": 846.678 }, { "epoch": 0.09779224170835091, "grad_norm": 2.9499833583831787, "learning_rate": 7.859649122807018e-06, "loss": 0.29638680815696716, "num_input_tokens_seen": 3421626, "step": 1217, "train_runtime": 4040.966, "train_tokens_per_second": 846.735 }, { "epoch": 0.09787259687820166, "grad_norm": 2.8141889572143555, "learning_rate": 7.849624060150377e-06, "loss": 0.25919899344444275, "num_input_tokens_seen": 3424834, "step": 1218, "train_runtime": 4043.6111, "train_tokens_per_second": 846.974 }, { "epoch": 0.09795295204805239, "grad_norm": 2.9556543827056885, "learning_rate": 7.839598997493735e-06, "loss": 0.25455695390701294, "num_input_tokens_seen": 3427164, "step": 1219, "train_runtime": 4046.0636, "train_tokens_per_second": 847.037 }, { "epoch": 0.09803330721790313, "grad_norm": 2.998471736907959, "learning_rate": 7.829573934837094e-06, "loss": 0.2511589527130127, "num_input_tokens_seen": 3433066, "step": 1220, "train_runtime": 4049.502, "train_tokens_per_second": 847.775 }, { "epoch": 0.09811366238775387, "grad_norm": 2.585073471069336, "learning_rate": 7.81954887218045e-06, "loss": 0.21295492351055145, "num_input_tokens_seen": 3436088, "step": 1221, "train_runtime": 4052.0557, "train_tokens_per_second": 847.986 }, { "epoch": 0.09819401755760461, "grad_norm": 2.7970006465911865, "learning_rate": 7.809523809523811e-06, "loss": 0.36667370796203613, "num_input_tokens_seen": 3438240, "step": 1222, "train_runtime": 4054.4624, "train_tokens_per_second": 848.014 }, { "epoch": 0.09827437272745536, "grad_norm": 2.7556533813476562, "learning_rate": 7.79949874686717e-06, "loss": 0.26070499420166016, "num_input_tokens_seen": 3440584, "step": 1223, "train_runtime": 4056.9103, "train_tokens_per_second": 848.08 }, { "epoch": 0.0983547278973061, "grad_norm": 2.688927173614502, "learning_rate": 7.789473684210526e-06, "loss": 0.27351608872413635, "num_input_tokens_seen": 3443382, "step": 1224, "train_runtime": 4059.5029, "train_tokens_per_second": 848.227 }, { "epoch": 0.09843508306715683, "grad_norm": 3.170353889465332, "learning_rate": 7.779448621553886e-06, "loss": 0.32282915711402893, "num_input_tokens_seen": 3447320, "step": 1225, "train_runtime": 4062.3992, "train_tokens_per_second": 848.592 }, { "epoch": 0.09851543823700758, "grad_norm": 3.119405746459961, "learning_rate": 7.769423558897243e-06, "loss": 0.21286459267139435, "num_input_tokens_seen": 3450538, "step": 1226, "train_runtime": 4065.0523, "train_tokens_per_second": 848.83 }, { "epoch": 0.09859579340685831, "grad_norm": 2.715303659439087, "learning_rate": 7.759398496240602e-06, "loss": 0.22451232373714447, "num_input_tokens_seen": 3453710, "step": 1227, "train_runtime": 4067.7289, "train_tokens_per_second": 849.051 }, { "epoch": 0.09867614857670906, "grad_norm": 2.035274028778076, "learning_rate": 7.749373433583962e-06, "loss": 0.22338321805000305, "num_input_tokens_seen": 3458012, "step": 1228, "train_runtime": 4070.6916, "train_tokens_per_second": 849.49 }, { "epoch": 0.0987565037465598, "grad_norm": 2.7421329021453857, "learning_rate": 7.739348370927319e-06, "loss": 0.2550141215324402, "num_input_tokens_seen": 3460988, "step": 1229, "train_runtime": 4073.3367, "train_tokens_per_second": 849.669 }, { "epoch": 0.09883685891641053, "grad_norm": 3.1457486152648926, "learning_rate": 7.729323308270677e-06, "loss": 0.2319430559873581, "num_input_tokens_seen": 3463460, "step": 1230, "train_runtime": 4075.9317, "train_tokens_per_second": 849.735 }, { "epoch": 0.09891721408626128, "grad_norm": 2.7177469730377197, "learning_rate": 7.719298245614036e-06, "loss": 0.23517662286758423, "num_input_tokens_seen": 3465954, "step": 1231, "train_runtime": 4089.7404, "train_tokens_per_second": 847.475 }, { "epoch": 0.09899756925611201, "grad_norm": 3.9900853633880615, "learning_rate": 7.709273182957394e-06, "loss": 0.3090720772743225, "num_input_tokens_seen": 3469212, "step": 1232, "train_runtime": 4092.526, "train_tokens_per_second": 847.695 }, { "epoch": 0.09907792442596275, "grad_norm": 3.1974992752075195, "learning_rate": 7.699248120300753e-06, "loss": 0.24765890836715698, "num_input_tokens_seen": 3471130, "step": 1233, "train_runtime": 4094.9516, "train_tokens_per_second": 847.661 }, { "epoch": 0.0991582795958135, "grad_norm": 3.2709391117095947, "learning_rate": 7.689223057644111e-06, "loss": 0.3343440890312195, "num_input_tokens_seen": 3473514, "step": 1234, "train_runtime": 4097.4497, "train_tokens_per_second": 847.726 }, { "epoch": 0.09923863476566423, "grad_norm": 2.6102488040924072, "learning_rate": 7.67919799498747e-06, "loss": 0.243694469332695, "num_input_tokens_seen": 3476372, "step": 1235, "train_runtime": 4100.0709, "train_tokens_per_second": 847.881 }, { "epoch": 0.09931898993551498, "grad_norm": 2.851473569869995, "learning_rate": 7.669172932330828e-06, "loss": 0.29606375098228455, "num_input_tokens_seen": 3479468, "step": 1236, "train_runtime": 4102.7278, "train_tokens_per_second": 848.086 }, { "epoch": 0.09939934510536572, "grad_norm": 3.4026975631713867, "learning_rate": 7.659147869674187e-06, "loss": 0.2673945426940918, "num_input_tokens_seen": 3483672, "step": 1237, "train_runtime": 4105.675, "train_tokens_per_second": 848.502 }, { "epoch": 0.09947970027521645, "grad_norm": 2.699486255645752, "learning_rate": 7.649122807017545e-06, "loss": 0.33283883333206177, "num_input_tokens_seen": 3487418, "step": 1238, "train_runtime": 4108.5781, "train_tokens_per_second": 848.814 }, { "epoch": 0.0995600554450672, "grad_norm": 3.9876625537872314, "learning_rate": 7.639097744360904e-06, "loss": 0.3779451549053192, "num_input_tokens_seen": 3491594, "step": 1239, "train_runtime": 4111.4957, "train_tokens_per_second": 849.227 }, { "epoch": 0.09964041061491793, "grad_norm": 2.7607078552246094, "learning_rate": 7.629072681704261e-06, "loss": 0.3556692600250244, "num_input_tokens_seen": 3494174, "step": 1240, "train_runtime": 4114.1167, "train_tokens_per_second": 849.313 }, { "epoch": 0.09972076578476868, "grad_norm": 3.105757474899292, "learning_rate": 7.61904761904762e-06, "loss": 0.3475748300552368, "num_input_tokens_seen": 3496178, "step": 1241, "train_runtime": 4116.5142, "train_tokens_per_second": 849.305 }, { "epoch": 0.09980112095461942, "grad_norm": 2.8426575660705566, "learning_rate": 7.609022556390978e-06, "loss": 0.20897287130355835, "num_input_tokens_seen": 3498144, "step": 1242, "train_runtime": 4118.9685, "train_tokens_per_second": 849.277 }, { "epoch": 0.09988147612447015, "grad_norm": 2.9536170959472656, "learning_rate": 7.598997493734337e-06, "loss": 0.2257828265428543, "num_input_tokens_seen": 3501536, "step": 1243, "train_runtime": 4121.7381, "train_tokens_per_second": 849.529 }, { "epoch": 0.0999618312943209, "grad_norm": 2.7218315601348877, "learning_rate": 7.588972431077694e-06, "loss": 0.26391175389289856, "num_input_tokens_seen": 3504582, "step": 1244, "train_runtime": 4124.3866, "train_tokens_per_second": 849.722 }, { "epoch": 0.10004218646417164, "grad_norm": 2.7987897396087646, "learning_rate": 7.578947368421054e-06, "loss": 0.3179968595504761, "num_input_tokens_seen": 3510218, "step": 1245, "train_runtime": 4127.9716, "train_tokens_per_second": 850.349 }, { "epoch": 0.10012254163402239, "grad_norm": 2.9945554733276367, "learning_rate": 7.568922305764411e-06, "loss": 0.18606728315353394, "num_input_tokens_seen": 3512330, "step": 1246, "train_runtime": 4130.3801, "train_tokens_per_second": 850.365 }, { "epoch": 0.10020289680387312, "grad_norm": 3.0329389572143555, "learning_rate": 7.55889724310777e-06, "loss": 0.4226483106613159, "num_input_tokens_seen": 3518946, "step": 1247, "train_runtime": 4133.9253, "train_tokens_per_second": 851.236 }, { "epoch": 0.10028325197372386, "grad_norm": 3.2576375007629395, "learning_rate": 7.548872180451129e-06, "loss": 0.27298611402511597, "num_input_tokens_seen": 3521432, "step": 1248, "train_runtime": 4136.46, "train_tokens_per_second": 851.315 }, { "epoch": 0.1003636071435746, "grad_norm": 2.6053500175476074, "learning_rate": 7.538847117794487e-06, "loss": 0.25645941495895386, "num_input_tokens_seen": 3523594, "step": 1249, "train_runtime": 4138.992, "train_tokens_per_second": 851.317 }, { "epoch": 0.10044396231342534, "grad_norm": 2.37628173828125, "learning_rate": 7.528822055137845e-06, "loss": 0.21910059452056885, "num_input_tokens_seen": 3526028, "step": 1250, "train_runtime": 4141.5085, "train_tokens_per_second": 851.387 }, { "epoch": 0.10052431748327607, "grad_norm": 2.436614990234375, "learning_rate": 7.518796992481203e-06, "loss": 0.2092609405517578, "num_input_tokens_seen": 3528308, "step": 1251, "train_runtime": 4144.0188, "train_tokens_per_second": 851.422 }, { "epoch": 0.10060467265312682, "grad_norm": 3.291579246520996, "learning_rate": 7.508771929824562e-06, "loss": 0.33822739124298096, "num_input_tokens_seen": 3530712, "step": 1252, "train_runtime": 4146.526, "train_tokens_per_second": 851.487 }, { "epoch": 0.10068502782297756, "grad_norm": 2.736966133117676, "learning_rate": 7.498746867167921e-06, "loss": 0.2830503582954407, "num_input_tokens_seen": 3533942, "step": 1253, "train_runtime": 4149.1922, "train_tokens_per_second": 851.718 }, { "epoch": 0.1007653829928283, "grad_norm": 3.055858850479126, "learning_rate": 7.488721804511278e-06, "loss": 0.3185797333717346, "num_input_tokens_seen": 3536868, "step": 1254, "train_runtime": 4151.7992, "train_tokens_per_second": 851.888 }, { "epoch": 0.10084573816267904, "grad_norm": 3.149435043334961, "learning_rate": 7.478696741854638e-06, "loss": 0.302920937538147, "num_input_tokens_seen": 3538614, "step": 1255, "train_runtime": 4154.0979, "train_tokens_per_second": 851.837 }, { "epoch": 0.10092609333252978, "grad_norm": 2.3568084239959717, "learning_rate": 7.468671679197995e-06, "loss": 0.228249654173851, "num_input_tokens_seen": 3542866, "step": 1256, "train_runtime": 4157.0805, "train_tokens_per_second": 852.249 }, { "epoch": 0.10100644850238052, "grad_norm": 2.682399034500122, "learning_rate": 7.458646616541354e-06, "loss": 0.2700612246990204, "num_input_tokens_seen": 3544556, "step": 1257, "train_runtime": 4159.3435, "train_tokens_per_second": 852.191 }, { "epoch": 0.10108680367223126, "grad_norm": 3.101486921310425, "learning_rate": 7.448621553884713e-06, "loss": 0.24518723785877228, "num_input_tokens_seen": 3551050, "step": 1258, "train_runtime": 4162.8676, "train_tokens_per_second": 853.03 }, { "epoch": 0.10116715884208201, "grad_norm": 3.364250898361206, "learning_rate": 7.438596491228071e-06, "loss": 0.37361615896224976, "num_input_tokens_seen": 3553098, "step": 1259, "train_runtime": 4165.269, "train_tokens_per_second": 853.03 }, { "epoch": 0.10124751401193274, "grad_norm": 3.4735288619995117, "learning_rate": 7.428571428571429e-06, "loss": 0.316500723361969, "num_input_tokens_seen": 3556686, "step": 1260, "train_runtime": 4168.0487, "train_tokens_per_second": 853.322 }, { "epoch": 0.10132786918178348, "grad_norm": 2.5838074684143066, "learning_rate": 7.418546365914787e-06, "loss": 0.2603965401649475, "num_input_tokens_seen": 3559964, "step": 1261, "train_runtime": 4182.2577, "train_tokens_per_second": 851.206 }, { "epoch": 0.10140822435163423, "grad_norm": 2.809246063232422, "learning_rate": 7.408521303258146e-06, "loss": 0.24488122761249542, "num_input_tokens_seen": 3562418, "step": 1262, "train_runtime": 4184.7438, "train_tokens_per_second": 851.287 }, { "epoch": 0.10148857952148496, "grad_norm": 3.5139620304107666, "learning_rate": 7.398496240601505e-06, "loss": 0.2722415626049042, "num_input_tokens_seen": 3564632, "step": 1263, "train_runtime": 4187.1846, "train_tokens_per_second": 851.32 }, { "epoch": 0.10156893469133571, "grad_norm": 3.238068103790283, "learning_rate": 7.388471177944862e-06, "loss": 0.3223555088043213, "num_input_tokens_seen": 3568038, "step": 1264, "train_runtime": 4189.8977, "train_tokens_per_second": 851.581 }, { "epoch": 0.10164928986118645, "grad_norm": 2.9235353469848633, "learning_rate": 7.378446115288222e-06, "loss": 0.23329806327819824, "num_input_tokens_seen": 3569636, "step": 1265, "train_runtime": 4192.1541, "train_tokens_per_second": 851.504 }, { "epoch": 0.10172964503103718, "grad_norm": 2.906095266342163, "learning_rate": 7.368421052631579e-06, "loss": 0.3109259009361267, "num_input_tokens_seen": 3571950, "step": 1266, "train_runtime": 4194.5816, "train_tokens_per_second": 851.563 }, { "epoch": 0.10181000020088793, "grad_norm": 2.6556291580200195, "learning_rate": 7.358395989974938e-06, "loss": 0.281730979681015, "num_input_tokens_seen": 3574924, "step": 1267, "train_runtime": 4197.2098, "train_tokens_per_second": 851.738 }, { "epoch": 0.10189035537073866, "grad_norm": 3.209303617477417, "learning_rate": 7.348370927318297e-06, "loss": 0.27695825695991516, "num_input_tokens_seen": 3581474, "step": 1268, "train_runtime": 4200.7951, "train_tokens_per_second": 852.571 }, { "epoch": 0.1019707105405894, "grad_norm": 2.317187547683716, "learning_rate": 7.338345864661655e-06, "loss": 0.23622731864452362, "num_input_tokens_seen": 3584580, "step": 1269, "train_runtime": 4203.454, "train_tokens_per_second": 852.77 }, { "epoch": 0.10205106571044015, "grad_norm": 2.713881015777588, "learning_rate": 7.328320802005013e-06, "loss": 0.23750698566436768, "num_input_tokens_seen": 3587060, "step": 1270, "train_runtime": 4205.9912, "train_tokens_per_second": 852.845 }, { "epoch": 0.10213142088029088, "grad_norm": 3.137974977493286, "learning_rate": 7.318295739348371e-06, "loss": 0.3108150362968445, "num_input_tokens_seen": 3590276, "step": 1271, "train_runtime": 4208.7146, "train_tokens_per_second": 853.058 }, { "epoch": 0.10221177605014163, "grad_norm": 2.92746639251709, "learning_rate": 7.30827067669173e-06, "loss": 0.26072806119918823, "num_input_tokens_seen": 3592644, "step": 1272, "train_runtime": 4211.203, "train_tokens_per_second": 853.116 }, { "epoch": 0.10229213121999237, "grad_norm": 3.100482940673828, "learning_rate": 7.298245614035089e-06, "loss": 0.3293037712574005, "num_input_tokens_seen": 3596030, "step": 1273, "train_runtime": 4213.9768, "train_tokens_per_second": 853.358 }, { "epoch": 0.1023724863898431, "grad_norm": 2.5122861862182617, "learning_rate": 7.288220551378446e-06, "loss": 0.26580604910850525, "num_input_tokens_seen": 3598824, "step": 1274, "train_runtime": 4216.584, "train_tokens_per_second": 853.493 }, { "epoch": 0.10245284155969385, "grad_norm": 2.775573492050171, "learning_rate": 7.278195488721806e-06, "loss": 0.26910942792892456, "num_input_tokens_seen": 3601502, "step": 1275, "train_runtime": 4219.1201, "train_tokens_per_second": 853.614 }, { "epoch": 0.10253319672954458, "grad_norm": 2.876359224319458, "learning_rate": 7.268170426065163e-06, "loss": 0.21550068259239197, "num_input_tokens_seen": 3603390, "step": 1276, "train_runtime": 4221.4879, "train_tokens_per_second": 853.583 }, { "epoch": 0.10261355189939533, "grad_norm": 2.95550274848938, "learning_rate": 7.258145363408522e-06, "loss": 0.1885908991098404, "num_input_tokens_seen": 3605652, "step": 1277, "train_runtime": 4223.923, "train_tokens_per_second": 853.626 }, { "epoch": 0.10269390706924607, "grad_norm": 1.8608531951904297, "learning_rate": 7.248120300751881e-06, "loss": 0.1685124933719635, "num_input_tokens_seen": 3608088, "step": 1278, "train_runtime": 4226.3682, "train_tokens_per_second": 853.709 }, { "epoch": 0.1027742622390968, "grad_norm": 3.2943778038024902, "learning_rate": 7.238095238095239e-06, "loss": 0.347849577665329, "num_input_tokens_seen": 3610700, "step": 1279, "train_runtime": 4228.9111, "train_tokens_per_second": 853.813 }, { "epoch": 0.10285461740894755, "grad_norm": 2.7399518489837646, "learning_rate": 7.228070175438597e-06, "loss": 0.31600484251976013, "num_input_tokens_seen": 3613626, "step": 1280, "train_runtime": 4231.5199, "train_tokens_per_second": 853.978 }, { "epoch": 0.10293497257879829, "grad_norm": 3.1097514629364014, "learning_rate": 7.218045112781955e-06, "loss": 0.3257511854171753, "num_input_tokens_seen": 3616224, "step": 1281, "train_runtime": 4234.0656, "train_tokens_per_second": 854.078 }, { "epoch": 0.10301532774864904, "grad_norm": 2.9192843437194824, "learning_rate": 7.208020050125314e-06, "loss": 0.24569165706634521, "num_input_tokens_seen": 3618942, "step": 1282, "train_runtime": 4236.6042, "train_tokens_per_second": 854.208 }, { "epoch": 0.10309568291849977, "grad_norm": 3.0147223472595215, "learning_rate": 7.197994987468673e-06, "loss": 0.2591594159603119, "num_input_tokens_seen": 3621398, "step": 1283, "train_runtime": 4239.1385, "train_tokens_per_second": 854.277 }, { "epoch": 0.1031760380883505, "grad_norm": 2.953045129776001, "learning_rate": 7.18796992481203e-06, "loss": 0.3617849349975586, "num_input_tokens_seen": 3624130, "step": 1284, "train_runtime": 4241.6833, "train_tokens_per_second": 854.408 }, { "epoch": 0.10325639325820125, "grad_norm": 2.5360052585601807, "learning_rate": 7.17794486215539e-06, "loss": 0.23290663957595825, "num_input_tokens_seen": 3627700, "step": 1285, "train_runtime": 4244.4767, "train_tokens_per_second": 854.687 }, { "epoch": 0.10333674842805199, "grad_norm": 2.7292211055755615, "learning_rate": 7.167919799498747e-06, "loss": 0.2580949664115906, "num_input_tokens_seen": 3629382, "step": 1286, "train_runtime": 4246.8532, "train_tokens_per_second": 854.605 }, { "epoch": 0.10341710359790272, "grad_norm": 2.583487033843994, "learning_rate": 7.157894736842106e-06, "loss": 0.2227247953414917, "num_input_tokens_seen": 3631996, "step": 1287, "train_runtime": 4249.3629, "train_tokens_per_second": 854.715 }, { "epoch": 0.10349745876775347, "grad_norm": 3.1658170223236084, "learning_rate": 7.147869674185464e-06, "loss": 0.260337769985199, "num_input_tokens_seen": 3633916, "step": 1288, "train_runtime": 4251.7443, "train_tokens_per_second": 854.688 }, { "epoch": 0.10357781393760421, "grad_norm": 2.8405299186706543, "learning_rate": 7.137844611528823e-06, "loss": 0.2515084147453308, "num_input_tokens_seen": 3636840, "step": 1289, "train_runtime": 4254.3682, "train_tokens_per_second": 854.848 }, { "epoch": 0.10365816910745496, "grad_norm": 3.2480762004852295, "learning_rate": 7.127819548872181e-06, "loss": 0.21861658990383148, "num_input_tokens_seen": 3638722, "step": 1290, "train_runtime": 4256.7023, "train_tokens_per_second": 854.822 }, { "epoch": 0.10373852427730569, "grad_norm": 2.115570545196533, "learning_rate": 7.117794486215539e-06, "loss": 0.1968001425266266, "num_input_tokens_seen": 3641746, "step": 1291, "train_runtime": 4270.8456, "train_tokens_per_second": 852.699 }, { "epoch": 0.10381887944715643, "grad_norm": 2.6528449058532715, "learning_rate": 7.107769423558898e-06, "loss": 0.30368080735206604, "num_input_tokens_seen": 3648440, "step": 1292, "train_runtime": 4274.4509, "train_tokens_per_second": 853.546 }, { "epoch": 0.10389923461700717, "grad_norm": 3.5016634464263916, "learning_rate": 7.097744360902257e-06, "loss": 0.3148006796836853, "num_input_tokens_seen": 3650962, "step": 1293, "train_runtime": 4276.9824, "train_tokens_per_second": 853.63 }, { "epoch": 0.10397958978685791, "grad_norm": 2.7445638179779053, "learning_rate": 7.087719298245614e-06, "loss": 0.24015653133392334, "num_input_tokens_seen": 3652882, "step": 1294, "train_runtime": 4279.3657, "train_tokens_per_second": 853.604 }, { "epoch": 0.10405994495670866, "grad_norm": 2.9767911434173584, "learning_rate": 7.077694235588973e-06, "loss": 0.26700857281684875, "num_input_tokens_seen": 3654714, "step": 1295, "train_runtime": 4281.6971, "train_tokens_per_second": 853.567 }, { "epoch": 0.10414030012655939, "grad_norm": 3.252373218536377, "learning_rate": 7.067669172932331e-06, "loss": 0.309529572725296, "num_input_tokens_seen": 3659822, "step": 1296, "train_runtime": 4284.8012, "train_tokens_per_second": 854.14 }, { "epoch": 0.10422065529641013, "grad_norm": 2.8370306491851807, "learning_rate": 7.05764411027569e-06, "loss": 0.27076029777526855, "num_input_tokens_seen": 3662826, "step": 1297, "train_runtime": 4287.4297, "train_tokens_per_second": 854.317 }, { "epoch": 0.10430101046626088, "grad_norm": 3.398254156112671, "learning_rate": 7.047619047619048e-06, "loss": 0.3058130145072937, "num_input_tokens_seen": 3665884, "step": 1298, "train_runtime": 4290.1609, "train_tokens_per_second": 854.486 }, { "epoch": 0.10438136563611161, "grad_norm": 2.7050766944885254, "learning_rate": 7.037593984962407e-06, "loss": 0.27482491731643677, "num_input_tokens_seen": 3668864, "step": 1299, "train_runtime": 4292.7936, "train_tokens_per_second": 854.657 }, { "epoch": 0.10446172080596236, "grad_norm": 3.201122283935547, "learning_rate": 7.027568922305765e-06, "loss": 0.3411877751350403, "num_input_tokens_seen": 3671992, "step": 1300, "train_runtime": 4295.4259, "train_tokens_per_second": 854.861 }, { "epoch": 0.1045420759758131, "grad_norm": 2.843256950378418, "learning_rate": 7.017543859649123e-06, "loss": 0.2543523609638214, "num_input_tokens_seen": 3675840, "step": 1301, "train_runtime": 4298.2739, "train_tokens_per_second": 855.19 }, { "epoch": 0.10462243114566383, "grad_norm": 2.247267007827759, "learning_rate": 7.0075187969924815e-06, "loss": 0.2120831310749054, "num_input_tokens_seen": 3679250, "step": 1302, "train_runtime": 4300.9715, "train_tokens_per_second": 855.446 }, { "epoch": 0.10470278631551458, "grad_norm": 2.678691864013672, "learning_rate": 6.997493734335841e-06, "loss": 0.319938600063324, "num_input_tokens_seen": 3682314, "step": 1303, "train_runtime": 4303.6234, "train_tokens_per_second": 855.631 }, { "epoch": 0.10478314148536531, "grad_norm": 2.6315672397613525, "learning_rate": 6.9874686716791984e-06, "loss": 0.23649345338344574, "num_input_tokens_seen": 3684310, "step": 1304, "train_runtime": 4305.9837, "train_tokens_per_second": 855.626 }, { "epoch": 0.10486349665521605, "grad_norm": 3.104888439178467, "learning_rate": 6.977443609022557e-06, "loss": 0.20615467429161072, "num_input_tokens_seen": 3686790, "step": 1305, "train_runtime": 4308.5203, "train_tokens_per_second": 855.697 }, { "epoch": 0.1049438518250668, "grad_norm": 4.1220903396606445, "learning_rate": 6.9674185463659146e-06, "loss": 0.3183768093585968, "num_input_tokens_seen": 3690192, "step": 1306, "train_runtime": 4311.2393, "train_tokens_per_second": 855.947 }, { "epoch": 0.10502420699491753, "grad_norm": 2.7987046241760254, "learning_rate": 6.957393483709274e-06, "loss": 0.283966064453125, "num_input_tokens_seen": 3692570, "step": 1307, "train_runtime": 4313.701, "train_tokens_per_second": 856.01 }, { "epoch": 0.10510456216476828, "grad_norm": 3.1144962310791016, "learning_rate": 6.947368421052632e-06, "loss": 0.2901330888271332, "num_input_tokens_seen": 3695726, "step": 1308, "train_runtime": 4316.3834, "train_tokens_per_second": 856.209 }, { "epoch": 0.10518491733461902, "grad_norm": 3.18627667427063, "learning_rate": 6.93734335839599e-06, "loss": 0.4019319713115692, "num_input_tokens_seen": 3697730, "step": 1309, "train_runtime": 4318.8175, "train_tokens_per_second": 856.19 }, { "epoch": 0.10526527250446975, "grad_norm": 3.0186328887939453, "learning_rate": 6.927318295739349e-06, "loss": 0.27146396040916443, "num_input_tokens_seen": 3700818, "step": 1310, "train_runtime": 4321.4469, "train_tokens_per_second": 856.384 }, { "epoch": 0.1053456276743205, "grad_norm": 2.6918671131134033, "learning_rate": 6.917293233082707e-06, "loss": 0.22707262635231018, "num_input_tokens_seen": 3702486, "step": 1311, "train_runtime": 4323.7392, "train_tokens_per_second": 856.316 }, { "epoch": 0.10542598284417123, "grad_norm": 2.8633460998535156, "learning_rate": 6.9072681704260655e-06, "loss": 0.3115274906158447, "num_input_tokens_seen": 3704388, "step": 1312, "train_runtime": 4326.1019, "train_tokens_per_second": 856.288 }, { "epoch": 0.10550633801402198, "grad_norm": 2.5421924591064453, "learning_rate": 6.897243107769425e-06, "loss": 0.3022477626800537, "num_input_tokens_seen": 3706656, "step": 1313, "train_runtime": 4328.5723, "train_tokens_per_second": 856.323 }, { "epoch": 0.10558669318387272, "grad_norm": 2.2379984855651855, "learning_rate": 6.8872180451127825e-06, "loss": 0.24730360507965088, "num_input_tokens_seen": 3708640, "step": 1314, "train_runtime": 4330.8905, "train_tokens_per_second": 856.323 }, { "epoch": 0.10566704835372345, "grad_norm": 2.756762981414795, "learning_rate": 6.877192982456141e-06, "loss": 0.2432839423418045, "num_input_tokens_seen": 3712166, "step": 1315, "train_runtime": 4333.6306, "train_tokens_per_second": 856.595 }, { "epoch": 0.1057474035235742, "grad_norm": 3.4246673583984375, "learning_rate": 6.867167919799499e-06, "loss": 0.28765740990638733, "num_input_tokens_seen": 3716088, "step": 1316, "train_runtime": 4336.4485, "train_tokens_per_second": 856.943 }, { "epoch": 0.10582775869342494, "grad_norm": 3.375300884246826, "learning_rate": 6.857142857142858e-06, "loss": 0.2924841344356537, "num_input_tokens_seen": 3717968, "step": 1317, "train_runtime": 4338.8415, "train_tokens_per_second": 856.903 }, { "epoch": 0.10590811386327569, "grad_norm": 2.547553539276123, "learning_rate": 6.847117794486216e-06, "loss": 0.2873426377773285, "num_input_tokens_seen": 3720002, "step": 1318, "train_runtime": 4341.2207, "train_tokens_per_second": 856.902 }, { "epoch": 0.10598846903312642, "grad_norm": 2.692870616912842, "learning_rate": 6.837092731829574e-06, "loss": 0.3553043603897095, "num_input_tokens_seen": 3723374, "step": 1319, "train_runtime": 4343.9732, "train_tokens_per_second": 857.136 }, { "epoch": 0.10606882420297715, "grad_norm": 3.2462286949157715, "learning_rate": 6.827067669172933e-06, "loss": 0.3482048511505127, "num_input_tokens_seen": 3726284, "step": 1320, "train_runtime": 4346.6034, "train_tokens_per_second": 857.286 }, { "epoch": 0.1061491793728279, "grad_norm": 3.100785732269287, "learning_rate": 6.817042606516291e-06, "loss": 0.3801589608192444, "num_input_tokens_seen": 3728236, "step": 1321, "train_runtime": 4360.5239, "train_tokens_per_second": 854.997 }, { "epoch": 0.10622953454267864, "grad_norm": 2.8300514221191406, "learning_rate": 6.8070175438596495e-06, "loss": 0.2705637514591217, "num_input_tokens_seen": 3730838, "step": 1322, "train_runtime": 4363.2011, "train_tokens_per_second": 855.069 }, { "epoch": 0.10630988971252937, "grad_norm": 3.3140461444854736, "learning_rate": 6.796992481203009e-06, "loss": 0.21537010371685028, "num_input_tokens_seen": 3732784, "step": 1323, "train_runtime": 4365.5437, "train_tokens_per_second": 855.056 }, { "epoch": 0.10639024488238012, "grad_norm": 2.9539787769317627, "learning_rate": 6.7869674185463665e-06, "loss": 0.3178909122943878, "num_input_tokens_seen": 3736290, "step": 1324, "train_runtime": 4368.3046, "train_tokens_per_second": 855.318 }, { "epoch": 0.10647060005223086, "grad_norm": 2.20052170753479, "learning_rate": 6.776942355889725e-06, "loss": 0.2049206644296646, "num_input_tokens_seen": 3740506, "step": 1325, "train_runtime": 4371.2307, "train_tokens_per_second": 855.71 }, { "epoch": 0.1065509552220816, "grad_norm": 3.1912245750427246, "learning_rate": 6.766917293233083e-06, "loss": 0.25804394483566284, "num_input_tokens_seen": 3742546, "step": 1326, "train_runtime": 4373.6207, "train_tokens_per_second": 855.709 }, { "epoch": 0.10663131039193234, "grad_norm": 3.229161262512207, "learning_rate": 6.756892230576442e-06, "loss": 0.20261827111244202, "num_input_tokens_seen": 3747008, "step": 1327, "train_runtime": 4376.6424, "train_tokens_per_second": 856.138 }, { "epoch": 0.10671166556178308, "grad_norm": 3.8558263778686523, "learning_rate": 6.7468671679198004e-06, "loss": 0.3340679705142975, "num_input_tokens_seen": 3748554, "step": 1328, "train_runtime": 4378.8908, "train_tokens_per_second": 856.051 }, { "epoch": 0.10679202073163382, "grad_norm": 2.991812229156494, "learning_rate": 6.736842105263158e-06, "loss": 0.2880048453807831, "num_input_tokens_seen": 3751250, "step": 1329, "train_runtime": 4381.4561, "train_tokens_per_second": 856.165 }, { "epoch": 0.10687237590148456, "grad_norm": 3.273391008377075, "learning_rate": 6.726817042606517e-06, "loss": 0.2723652720451355, "num_input_tokens_seen": 3753522, "step": 1330, "train_runtime": 4383.907, "train_tokens_per_second": 856.205 }, { "epoch": 0.10695273107133531, "grad_norm": 2.4918224811553955, "learning_rate": 6.716791979949875e-06, "loss": 0.22401857376098633, "num_input_tokens_seen": 3756504, "step": 1331, "train_runtime": 4386.5892, "train_tokens_per_second": 856.361 }, { "epoch": 0.10703308624118604, "grad_norm": 2.621143341064453, "learning_rate": 6.7067669172932335e-06, "loss": 0.22797007858753204, "num_input_tokens_seen": 3758752, "step": 1332, "train_runtime": 4389.1096, "train_tokens_per_second": 856.381 }, { "epoch": 0.10711344141103678, "grad_norm": 3.085921049118042, "learning_rate": 6.696741854636593e-06, "loss": 0.31925728917121887, "num_input_tokens_seen": 3762710, "step": 1333, "train_runtime": 4391.9752, "train_tokens_per_second": 856.724 }, { "epoch": 0.10719379658088753, "grad_norm": 2.342050313949585, "learning_rate": 6.6867167919799505e-06, "loss": 0.22503532469272614, "num_input_tokens_seen": 3764736, "step": 1334, "train_runtime": 4394.3375, "train_tokens_per_second": 856.724 }, { "epoch": 0.10727415175073826, "grad_norm": 2.829094409942627, "learning_rate": 6.676691729323309e-06, "loss": 0.31050387024879456, "num_input_tokens_seen": 3767322, "step": 1335, "train_runtime": 4396.8296, "train_tokens_per_second": 856.827 }, { "epoch": 0.10735450692058901, "grad_norm": 2.099871873855591, "learning_rate": 6.666666666666667e-06, "loss": 0.18152445554733276, "num_input_tokens_seen": 3769890, "step": 1336, "train_runtime": 4399.3544, "train_tokens_per_second": 856.919 }, { "epoch": 0.10743486209043974, "grad_norm": 2.7306768894195557, "learning_rate": 6.656641604010026e-06, "loss": 0.2584064304828644, "num_input_tokens_seen": 3772298, "step": 1337, "train_runtime": 4401.8591, "train_tokens_per_second": 856.978 }, { "epoch": 0.10751521726029048, "grad_norm": 2.3899106979370117, "learning_rate": 6.6466165413533845e-06, "loss": 0.1989639401435852, "num_input_tokens_seen": 3773790, "step": 1338, "train_runtime": 4404.1391, "train_tokens_per_second": 856.873 }, { "epoch": 0.10759557243014123, "grad_norm": 2.755117893218994, "learning_rate": 6.636591478696742e-06, "loss": 0.22161421179771423, "num_input_tokens_seen": 3776620, "step": 1339, "train_runtime": 4406.7509, "train_tokens_per_second": 857.008 }, { "epoch": 0.10767592759999196, "grad_norm": 2.4736201763153076, "learning_rate": 6.6265664160401014e-06, "loss": 0.33031490445137024, "num_input_tokens_seen": 3779372, "step": 1340, "train_runtime": 4409.3456, "train_tokens_per_second": 857.128 }, { "epoch": 0.1077562827698427, "grad_norm": 3.675243854522705, "learning_rate": 6.616541353383459e-06, "loss": 0.34945181012153625, "num_input_tokens_seen": 3782802, "step": 1341, "train_runtime": 4412.1645, "train_tokens_per_second": 857.357 }, { "epoch": 0.10783663793969345, "grad_norm": 2.3972744941711426, "learning_rate": 6.6065162907268176e-06, "loss": 0.1905646026134491, "num_input_tokens_seen": 3784380, "step": 1342, "train_runtime": 4414.5266, "train_tokens_per_second": 857.256 }, { "epoch": 0.10791699310954418, "grad_norm": 3.2512881755828857, "learning_rate": 6.596491228070177e-06, "loss": 0.2532309889793396, "num_input_tokens_seen": 3786536, "step": 1343, "train_runtime": 4416.9389, "train_tokens_per_second": 857.276 }, { "epoch": 0.10799734827939493, "grad_norm": 2.9417724609375, "learning_rate": 6.5864661654135345e-06, "loss": 0.27173182368278503, "num_input_tokens_seen": 3789156, "step": 1344, "train_runtime": 4419.4781, "train_tokens_per_second": 857.376 }, { "epoch": 0.10807770344924567, "grad_norm": 3.538637638092041, "learning_rate": 6.576441102756893e-06, "loss": 0.32068124413490295, "num_input_tokens_seen": 3790846, "step": 1345, "train_runtime": 4421.8597, "train_tokens_per_second": 857.297 }, { "epoch": 0.1081580586190964, "grad_norm": 2.9422028064727783, "learning_rate": 6.566416040100251e-06, "loss": 0.2736411988735199, "num_input_tokens_seen": 3794414, "step": 1346, "train_runtime": 4424.5802, "train_tokens_per_second": 857.576 }, { "epoch": 0.10823841378894715, "grad_norm": 3.2272634506225586, "learning_rate": 6.55639097744361e-06, "loss": 0.2898770570755005, "num_input_tokens_seen": 3798204, "step": 1347, "train_runtime": 4427.3847, "train_tokens_per_second": 857.889 }, { "epoch": 0.10831876895879788, "grad_norm": 2.4257845878601074, "learning_rate": 6.5463659147869685e-06, "loss": 0.23071610927581787, "num_input_tokens_seen": 3800226, "step": 1348, "train_runtime": 4429.8344, "train_tokens_per_second": 857.871 }, { "epoch": 0.10839912412864863, "grad_norm": 3.085305690765381, "learning_rate": 6.536340852130326e-06, "loss": 0.2288166582584381, "num_input_tokens_seen": 3802320, "step": 1349, "train_runtime": 4432.1683, "train_tokens_per_second": 857.892 }, { "epoch": 0.10847947929849937, "grad_norm": 2.586228609085083, "learning_rate": 6.526315789473685e-06, "loss": 0.2783668637275696, "num_input_tokens_seen": 3806882, "step": 1350, "train_runtime": 4435.1415, "train_tokens_per_second": 858.345 }, { "epoch": 0.1085598344683501, "grad_norm": 3.1222634315490723, "learning_rate": 6.516290726817043e-06, "loss": 0.301727831363678, "num_input_tokens_seen": 3808798, "step": 1351, "train_runtime": 4448.968, "train_tokens_per_second": 856.108 }, { "epoch": 0.10864018963820085, "grad_norm": 3.1318600177764893, "learning_rate": 6.506265664160402e-06, "loss": 0.24911057949066162, "num_input_tokens_seen": 3810700, "step": 1352, "train_runtime": 4451.3856, "train_tokens_per_second": 856.071 }, { "epoch": 0.10872054480805159, "grad_norm": 2.8164403438568115, "learning_rate": 6.49624060150376e-06, "loss": 0.2827795445919037, "num_input_tokens_seen": 3813242, "step": 1353, "train_runtime": 4453.8971, "train_tokens_per_second": 856.159 }, { "epoch": 0.10880089997790234, "grad_norm": 2.815059185028076, "learning_rate": 6.486215538847119e-06, "loss": 0.2893601059913635, "num_input_tokens_seen": 3815692, "step": 1354, "train_runtime": 4456.335, "train_tokens_per_second": 856.24 }, { "epoch": 0.10888125514775307, "grad_norm": 2.7385289669036865, "learning_rate": 6.476190476190477e-06, "loss": 0.28072428703308105, "num_input_tokens_seen": 3818398, "step": 1355, "train_runtime": 4458.8631, "train_tokens_per_second": 856.361 }, { "epoch": 0.1089616103176038, "grad_norm": 3.790722131729126, "learning_rate": 6.466165413533835e-06, "loss": 0.30124592781066895, "num_input_tokens_seen": 3822112, "step": 1356, "train_runtime": 4461.6356, "train_tokens_per_second": 856.662 }, { "epoch": 0.10904196548745455, "grad_norm": 2.898178815841675, "learning_rate": 6.456140350877193e-06, "loss": 0.29472798109054565, "num_input_tokens_seen": 3825130, "step": 1357, "train_runtime": 4464.2597, "train_tokens_per_second": 856.834 }, { "epoch": 0.10912232065730529, "grad_norm": 3.9905450344085693, "learning_rate": 6.4461152882205525e-06, "loss": 0.3842206299304962, "num_input_tokens_seen": 3827578, "step": 1358, "train_runtime": 4466.7228, "train_tokens_per_second": 856.91 }, { "epoch": 0.10920267582715602, "grad_norm": 2.66606068611145, "learning_rate": 6.43609022556391e-06, "loss": 0.20043233036994934, "num_input_tokens_seen": 3829440, "step": 1359, "train_runtime": 4469.1134, "train_tokens_per_second": 856.868 }, { "epoch": 0.10928303099700677, "grad_norm": 2.8895816802978516, "learning_rate": 6.426065162907269e-06, "loss": 0.32120281457901, "num_input_tokens_seen": 3832000, "step": 1360, "train_runtime": 4471.6281, "train_tokens_per_second": 856.959 }, { "epoch": 0.1093633861668575, "grad_norm": 3.2627344131469727, "learning_rate": 6.416040100250627e-06, "loss": 0.3577735126018524, "num_input_tokens_seen": 3836148, "step": 1361, "train_runtime": 4474.53, "train_tokens_per_second": 857.33 }, { "epoch": 0.10944374133670826, "grad_norm": 2.6659417152404785, "learning_rate": 6.406015037593986e-06, "loss": 0.25677013397216797, "num_input_tokens_seen": 3838086, "step": 1362, "train_runtime": 4477.0481, "train_tokens_per_second": 857.28 }, { "epoch": 0.10952409650655899, "grad_norm": 2.7092063426971436, "learning_rate": 6.395989974937344e-06, "loss": 0.29768452048301697, "num_input_tokens_seen": 3840800, "step": 1363, "train_runtime": 4479.6269, "train_tokens_per_second": 857.393 }, { "epoch": 0.10960445167640973, "grad_norm": 2.942011833190918, "learning_rate": 6.385964912280702e-06, "loss": 0.3464399576187134, "num_input_tokens_seen": 3843350, "step": 1364, "train_runtime": 4482.1382, "train_tokens_per_second": 857.481 }, { "epoch": 0.10968480684626047, "grad_norm": 2.979142904281616, "learning_rate": 6.375939849624061e-06, "loss": 0.31967002153396606, "num_input_tokens_seen": 3846952, "step": 1365, "train_runtime": 4484.8957, "train_tokens_per_second": 857.757 }, { "epoch": 0.10976516201611121, "grad_norm": 2.9331562519073486, "learning_rate": 6.365914786967419e-06, "loss": 0.27625903487205505, "num_input_tokens_seen": 3849676, "step": 1366, "train_runtime": 4487.4468, "train_tokens_per_second": 857.877 }, { "epoch": 0.10984551718596196, "grad_norm": 2.299562692642212, "learning_rate": 6.355889724310777e-06, "loss": 0.20642195641994476, "num_input_tokens_seen": 3852356, "step": 1367, "train_runtime": 4489.9964, "train_tokens_per_second": 857.986 }, { "epoch": 0.10992587235581269, "grad_norm": 2.7156784534454346, "learning_rate": 6.3458646616541366e-06, "loss": 0.18016330897808075, "num_input_tokens_seen": 3854014, "step": 1368, "train_runtime": 4492.2789, "train_tokens_per_second": 857.92 }, { "epoch": 0.11000622752566343, "grad_norm": 2.8953890800476074, "learning_rate": 6.335839598997494e-06, "loss": 0.2447168231010437, "num_input_tokens_seen": 3857912, "step": 1369, "train_runtime": 4495.0981, "train_tokens_per_second": 858.249 }, { "epoch": 0.11008658269551418, "grad_norm": 3.2375199794769287, "learning_rate": 6.325814536340853e-06, "loss": 0.20718032121658325, "num_input_tokens_seen": 3860610, "step": 1370, "train_runtime": 4497.6299, "train_tokens_per_second": 858.365 }, { "epoch": 0.11016693786536491, "grad_norm": 3.0049266815185547, "learning_rate": 6.31578947368421e-06, "loss": 0.3404928743839264, "num_input_tokens_seen": 3862446, "step": 1371, "train_runtime": 4499.9686, "train_tokens_per_second": 858.327 }, { "epoch": 0.11024729303521566, "grad_norm": 3.322697162628174, "learning_rate": 6.30576441102757e-06, "loss": 0.20526009798049927, "num_input_tokens_seen": 3863642, "step": 1372, "train_runtime": 4502.1172, "train_tokens_per_second": 858.183 }, { "epoch": 0.1103276482050664, "grad_norm": 2.308042287826538, "learning_rate": 6.295739348370928e-06, "loss": 0.23909851908683777, "num_input_tokens_seen": 3867162, "step": 1373, "train_runtime": 4504.8359, "train_tokens_per_second": 858.447 }, { "epoch": 0.11040800337491713, "grad_norm": 3.050081729888916, "learning_rate": 6.285714285714286e-06, "loss": 0.26798489689826965, "num_input_tokens_seen": 3869030, "step": 1374, "train_runtime": 4507.1587, "train_tokens_per_second": 858.419 }, { "epoch": 0.11048835854476788, "grad_norm": 3.36409330368042, "learning_rate": 6.275689223057645e-06, "loss": 0.3549877405166626, "num_input_tokens_seen": 3872066, "step": 1375, "train_runtime": 4509.8413, "train_tokens_per_second": 858.581 }, { "epoch": 0.11056871371461861, "grad_norm": 3.251760244369507, "learning_rate": 6.265664160401003e-06, "loss": 0.2705858051776886, "num_input_tokens_seen": 3875920, "step": 1376, "train_runtime": 4512.7217, "train_tokens_per_second": 858.887 }, { "epoch": 0.11064906888446935, "grad_norm": 2.5173656940460205, "learning_rate": 6.255639097744361e-06, "loss": 0.27834269404411316, "num_input_tokens_seen": 3879418, "step": 1377, "train_runtime": 4515.4799, "train_tokens_per_second": 859.137 }, { "epoch": 0.1107294240543201, "grad_norm": 2.2818567752838135, "learning_rate": 6.245614035087721e-06, "loss": 0.19807076454162598, "num_input_tokens_seen": 3881096, "step": 1378, "train_runtime": 4517.8084, "train_tokens_per_second": 859.066 }, { "epoch": 0.11080977922417083, "grad_norm": 3.382110118865967, "learning_rate": 6.235588972431078e-06, "loss": 0.2918010950088501, "num_input_tokens_seen": 3882768, "step": 1379, "train_runtime": 4520.0766, "train_tokens_per_second": 859.005 }, { "epoch": 0.11089013439402158, "grad_norm": 2.324683666229248, "learning_rate": 6.225563909774437e-06, "loss": 0.23027238249778748, "num_input_tokens_seen": 3886536, "step": 1380, "train_runtime": 4522.8625, "train_tokens_per_second": 859.309 }, { "epoch": 0.11097048956387232, "grad_norm": 3.263043165206909, "learning_rate": 6.215538847117794e-06, "loss": 0.21772350370883942, "num_input_tokens_seen": 3889062, "step": 1381, "train_runtime": 4536.9103, "train_tokens_per_second": 857.205 }, { "epoch": 0.11105084473372305, "grad_norm": 2.6889545917510986, "learning_rate": 6.205513784461154e-06, "loss": 0.25899720191955566, "num_input_tokens_seen": 3892476, "step": 1382, "train_runtime": 4539.6083, "train_tokens_per_second": 857.448 }, { "epoch": 0.1111311999035738, "grad_norm": 3.3619232177734375, "learning_rate": 6.195488721804512e-06, "loss": 0.35660725831985474, "num_input_tokens_seen": 3898034, "step": 1383, "train_runtime": 4542.9193, "train_tokens_per_second": 858.046 }, { "epoch": 0.11121155507342453, "grad_norm": 3.7298805713653564, "learning_rate": 6.18546365914787e-06, "loss": 0.3183562457561493, "num_input_tokens_seen": 3899536, "step": 1384, "train_runtime": 4545.1719, "train_tokens_per_second": 857.951 }, { "epoch": 0.11129191024327528, "grad_norm": 3.123117446899414, "learning_rate": 6.175438596491229e-06, "loss": 0.2808802127838135, "num_input_tokens_seen": 3902482, "step": 1385, "train_runtime": 4547.8047, "train_tokens_per_second": 858.102 }, { "epoch": 0.11137226541312602, "grad_norm": 3.1751973628997803, "learning_rate": 6.165413533834587e-06, "loss": 0.3033704161643982, "num_input_tokens_seen": 3904344, "step": 1386, "train_runtime": 4550.1934, "train_tokens_per_second": 858.061 }, { "epoch": 0.11145262058297675, "grad_norm": 2.535676956176758, "learning_rate": 6.155388471177945e-06, "loss": 0.26164472103118896, "num_input_tokens_seen": 3906898, "step": 1387, "train_runtime": 4552.6565, "train_tokens_per_second": 858.158 }, { "epoch": 0.1115329757528275, "grad_norm": 2.351003885269165, "learning_rate": 6.145363408521305e-06, "loss": 0.19835935533046722, "num_input_tokens_seen": 3910690, "step": 1388, "train_runtime": 4555.4777, "train_tokens_per_second": 858.459 }, { "epoch": 0.11161333092267824, "grad_norm": 3.258298635482788, "learning_rate": 6.135338345864662e-06, "loss": 0.2820468544960022, "num_input_tokens_seen": 3913044, "step": 1389, "train_runtime": 4557.9674, "train_tokens_per_second": 858.506 }, { "epoch": 0.11169368609252898, "grad_norm": 2.7637014389038086, "learning_rate": 6.125313283208021e-06, "loss": 0.32364901900291443, "num_input_tokens_seen": 3917154, "step": 1390, "train_runtime": 4560.859, "train_tokens_per_second": 858.863 }, { "epoch": 0.11177404126237972, "grad_norm": 3.1862659454345703, "learning_rate": 6.115288220551378e-06, "loss": 0.32775211334228516, "num_input_tokens_seen": 3920028, "step": 1391, "train_runtime": 4563.4717, "train_tokens_per_second": 859.001 }, { "epoch": 0.11185439643223045, "grad_norm": 2.235316276550293, "learning_rate": 6.105263157894738e-06, "loss": 0.20273257791996002, "num_input_tokens_seen": 3923440, "step": 1392, "train_runtime": 4566.2251, "train_tokens_per_second": 859.231 }, { "epoch": 0.1119347516020812, "grad_norm": 3.2075371742248535, "learning_rate": 6.095238095238096e-06, "loss": 0.2053486704826355, "num_input_tokens_seen": 3925454, "step": 1393, "train_runtime": 4568.6246, "train_tokens_per_second": 859.22 }, { "epoch": 0.11201510677193194, "grad_norm": 3.0833842754364014, "learning_rate": 6.085213032581454e-06, "loss": 0.26964426040649414, "num_input_tokens_seen": 3927870, "step": 1394, "train_runtime": 4571.1324, "train_tokens_per_second": 859.277 }, { "epoch": 0.11209546194178267, "grad_norm": 3.0298945903778076, "learning_rate": 6.075187969924813e-06, "loss": 0.28605589270591736, "num_input_tokens_seen": 3931646, "step": 1395, "train_runtime": 4573.9514, "train_tokens_per_second": 859.573 }, { "epoch": 0.11217581711163342, "grad_norm": 3.138096570968628, "learning_rate": 6.065162907268171e-06, "loss": 0.23132188618183136, "num_input_tokens_seen": 3933974, "step": 1396, "train_runtime": 4576.4112, "train_tokens_per_second": 859.62 }, { "epoch": 0.11225617228148416, "grad_norm": 2.880998134613037, "learning_rate": 6.055137844611529e-06, "loss": 0.2844209671020508, "num_input_tokens_seen": 3936200, "step": 1397, "train_runtime": 4578.8511, "train_tokens_per_second": 859.648 }, { "epoch": 0.1123365274513349, "grad_norm": 3.0479013919830322, "learning_rate": 6.045112781954889e-06, "loss": 0.2697030305862427, "num_input_tokens_seen": 3937856, "step": 1398, "train_runtime": 4581.1679, "train_tokens_per_second": 859.575 }, { "epoch": 0.11241688262118564, "grad_norm": 3.169602155685425, "learning_rate": 6.035087719298246e-06, "loss": 0.31326133012771606, "num_input_tokens_seen": 3941210, "step": 1399, "train_runtime": 4583.8531, "train_tokens_per_second": 859.803 }, { "epoch": 0.11249723779103638, "grad_norm": 3.241184949874878, "learning_rate": 6.025062656641605e-06, "loss": 0.28021031618118286, "num_input_tokens_seen": 3943362, "step": 1400, "train_runtime": 4586.2276, "train_tokens_per_second": 859.827 }, { "epoch": 0.11249723779103638, "eval_loss": 1.1007806062698364, "eval_runtime": 69.212, "eval_samples_per_second": 14.362, "eval_steps_per_second": 7.181, "num_input_tokens_seen": 3943362, "step": 1400 }, { "epoch": 0.11257759296088712, "grad_norm": 2.3216493129730225, "learning_rate": 6.015037593984962e-06, "loss": 0.25376856327056885, "num_input_tokens_seen": 3946604, "step": 1401, "train_runtime": 4658.1102, "train_tokens_per_second": 847.254 }, { "epoch": 0.11265794813073786, "grad_norm": 2.891674280166626, "learning_rate": 6.005012531328322e-06, "loss": 0.23303399980068207, "num_input_tokens_seen": 3948074, "step": 1402, "train_runtime": 4660.3901, "train_tokens_per_second": 847.155 }, { "epoch": 0.11273830330058861, "grad_norm": 3.0930569171905518, "learning_rate": 5.994987468671679e-06, "loss": 0.41975679993629456, "num_input_tokens_seen": 3950634, "step": 1403, "train_runtime": 4662.9449, "train_tokens_per_second": 847.24 }, { "epoch": 0.11281865847043934, "grad_norm": 3.1307642459869385, "learning_rate": 5.984962406015038e-06, "loss": 0.31344935297966003, "num_input_tokens_seen": 3952892, "step": 1404, "train_runtime": 4665.4244, "train_tokens_per_second": 847.274 }, { "epoch": 0.11289901364029008, "grad_norm": 3.224762201309204, "learning_rate": 5.974937343358397e-06, "loss": 0.28411367535591125, "num_input_tokens_seen": 3956460, "step": 1405, "train_runtime": 4668.209, "train_tokens_per_second": 847.533 }, { "epoch": 0.11297936881014083, "grad_norm": 2.4652411937713623, "learning_rate": 5.964912280701755e-06, "loss": 0.2480407953262329, "num_input_tokens_seen": 3959524, "step": 1406, "train_runtime": 4670.863, "train_tokens_per_second": 847.707 }, { "epoch": 0.11305972397999156, "grad_norm": 3.117076873779297, "learning_rate": 5.954887218045113e-06, "loss": 0.24491529166698456, "num_input_tokens_seen": 3962392, "step": 1407, "train_runtime": 4673.4887, "train_tokens_per_second": 847.845 }, { "epoch": 0.11314007914984231, "grad_norm": 1.871842622756958, "learning_rate": 5.944862155388471e-06, "loss": 0.17226003110408783, "num_input_tokens_seen": 3968610, "step": 1408, "train_runtime": 4676.9529, "train_tokens_per_second": 848.546 }, { "epoch": 0.11322043431969304, "grad_norm": 2.0373785495758057, "learning_rate": 5.93483709273183e-06, "loss": 0.19375091791152954, "num_input_tokens_seen": 3971812, "step": 1409, "train_runtime": 4679.7346, "train_tokens_per_second": 848.726 }, { "epoch": 0.11330078948954378, "grad_norm": 3.4503040313720703, "learning_rate": 5.924812030075189e-06, "loss": 0.36595362424850464, "num_input_tokens_seen": 3974418, "step": 1410, "train_runtime": 4682.288, "train_tokens_per_second": 848.82 }, { "epoch": 0.11338114465939453, "grad_norm": 2.9706382751464844, "learning_rate": 5.9147869674185465e-06, "loss": 0.2891871929168701, "num_input_tokens_seen": 3977208, "step": 1411, "train_runtime": 4696.248, "train_tokens_per_second": 846.891 }, { "epoch": 0.11346149982924526, "grad_norm": 2.825423240661621, "learning_rate": 5.904761904761905e-06, "loss": 0.26031816005706787, "num_input_tokens_seen": 3980262, "step": 1412, "train_runtime": 4699.0389, "train_tokens_per_second": 847.037 }, { "epoch": 0.113541854999096, "grad_norm": 3.342594861984253, "learning_rate": 5.8947368421052634e-06, "loss": 0.3189372718334198, "num_input_tokens_seen": 3984388, "step": 1413, "train_runtime": 4701.936, "train_tokens_per_second": 847.393 }, { "epoch": 0.11362221016894675, "grad_norm": 2.2773900032043457, "learning_rate": 5.884711779448622e-06, "loss": 0.21895751357078552, "num_input_tokens_seen": 3987734, "step": 1414, "train_runtime": 4704.6282, "train_tokens_per_second": 847.619 }, { "epoch": 0.11370256533879748, "grad_norm": 2.818767547607422, "learning_rate": 5.87468671679198e-06, "loss": 0.2894251048564911, "num_input_tokens_seen": 3990654, "step": 1415, "train_runtime": 4707.2953, "train_tokens_per_second": 847.759 }, { "epoch": 0.11378292050864823, "grad_norm": 2.6037795543670654, "learning_rate": 5.864661654135339e-06, "loss": 0.2447981834411621, "num_input_tokens_seen": 3993390, "step": 1416, "train_runtime": 4709.9721, "train_tokens_per_second": 847.859 }, { "epoch": 0.11386327567849897, "grad_norm": 3.049138307571411, "learning_rate": 5.854636591478697e-06, "loss": 0.3157178461551666, "num_input_tokens_seen": 3998610, "step": 1417, "train_runtime": 4713.1395, "train_tokens_per_second": 848.396 }, { "epoch": 0.1139436308483497, "grad_norm": 3.4487335681915283, "learning_rate": 5.844611528822055e-06, "loss": 0.35990622639656067, "num_input_tokens_seen": 4002292, "step": 1418, "train_runtime": 4715.9466, "train_tokens_per_second": 848.672 }, { "epoch": 0.11402398601820045, "grad_norm": 2.486720085144043, "learning_rate": 5.8345864661654135e-06, "loss": 0.19976626336574554, "num_input_tokens_seen": 4004118, "step": 1419, "train_runtime": 4718.3006, "train_tokens_per_second": 848.636 }, { "epoch": 0.11410434118805118, "grad_norm": 2.3615756034851074, "learning_rate": 5.824561403508773e-06, "loss": 0.23037317395210266, "num_input_tokens_seen": 4005838, "step": 1420, "train_runtime": 4720.668, "train_tokens_per_second": 848.574 }, { "epoch": 0.11418469635790193, "grad_norm": 2.7046210765838623, "learning_rate": 5.8145363408521305e-06, "loss": 0.23145315051078796, "num_input_tokens_seen": 4010674, "step": 1421, "train_runtime": 4723.7521, "train_tokens_per_second": 849.044 }, { "epoch": 0.11426505152775267, "grad_norm": 3.0042569637298584, "learning_rate": 5.804511278195489e-06, "loss": 0.3014902174472809, "num_input_tokens_seen": 4013586, "step": 1422, "train_runtime": 4726.3108, "train_tokens_per_second": 849.201 }, { "epoch": 0.1143454066976034, "grad_norm": 3.2757630348205566, "learning_rate": 5.7944862155388475e-06, "loss": 0.3366795778274536, "num_input_tokens_seen": 4018878, "step": 1423, "train_runtime": 4729.5312, "train_tokens_per_second": 849.741 }, { "epoch": 0.11442576186745415, "grad_norm": 2.9781508445739746, "learning_rate": 5.784461152882206e-06, "loss": 0.3883008360862732, "num_input_tokens_seen": 4021792, "step": 1424, "train_runtime": 4732.1608, "train_tokens_per_second": 849.885 }, { "epoch": 0.11450611703730489, "grad_norm": 2.9627015590667725, "learning_rate": 5.7744360902255644e-06, "loss": 0.2552022635936737, "num_input_tokens_seen": 4023978, "step": 1425, "train_runtime": 4734.617, "train_tokens_per_second": 849.906 }, { "epoch": 0.11458647220715563, "grad_norm": 3.167212724685669, "learning_rate": 5.764411027568922e-06, "loss": 0.23747900128364563, "num_input_tokens_seen": 4027076, "step": 1426, "train_runtime": 4737.3368, "train_tokens_per_second": 850.072 }, { "epoch": 0.11466682737700637, "grad_norm": 2.6414806842803955, "learning_rate": 5.754385964912281e-06, "loss": 0.24899795651435852, "num_input_tokens_seen": 4031002, "step": 1427, "train_runtime": 4740.2125, "train_tokens_per_second": 850.384 }, { "epoch": 0.1147471825468571, "grad_norm": 3.0701868534088135, "learning_rate": 5.744360902255639e-06, "loss": 0.24891497194766998, "num_input_tokens_seen": 4034136, "step": 1428, "train_runtime": 4742.9376, "train_tokens_per_second": 850.556 }, { "epoch": 0.11482753771670785, "grad_norm": 2.501899480819702, "learning_rate": 5.7343358395989975e-06, "loss": 0.2020815908908844, "num_input_tokens_seen": 4036614, "step": 1429, "train_runtime": 4745.5424, "train_tokens_per_second": 850.612 }, { "epoch": 0.11490789288655859, "grad_norm": 2.769407033920288, "learning_rate": 5.724310776942357e-06, "loss": 0.337311714887619, "num_input_tokens_seen": 4039282, "step": 1430, "train_runtime": 4748.1243, "train_tokens_per_second": 850.711 }, { "epoch": 0.11498824805640932, "grad_norm": 3.089515447616577, "learning_rate": 5.7142857142857145e-06, "loss": 0.3752546012401581, "num_input_tokens_seen": 4042290, "step": 1431, "train_runtime": 4750.8178, "train_tokens_per_second": 850.862 }, { "epoch": 0.11506860322626007, "grad_norm": 3.7693467140197754, "learning_rate": 5.704260651629073e-06, "loss": 0.3394043445587158, "num_input_tokens_seen": 4044860, "step": 1432, "train_runtime": 4753.3901, "train_tokens_per_second": 850.942 }, { "epoch": 0.1151489583961108, "grad_norm": 2.9814698696136475, "learning_rate": 5.694235588972431e-06, "loss": 0.22202201187610626, "num_input_tokens_seen": 4048752, "step": 1433, "train_runtime": 4756.2812, "train_tokens_per_second": 851.243 }, { "epoch": 0.11522931356596156, "grad_norm": 2.665430784225464, "learning_rate": 5.68421052631579e-06, "loss": 0.2997194826602936, "num_input_tokens_seen": 4050396, "step": 1434, "train_runtime": 4758.5815, "train_tokens_per_second": 851.177 }, { "epoch": 0.11530966873581229, "grad_norm": 2.938098907470703, "learning_rate": 5.6741854636591485e-06, "loss": 0.26819586753845215, "num_input_tokens_seen": 4053442, "step": 1435, "train_runtime": 4761.2123, "train_tokens_per_second": 851.347 }, { "epoch": 0.11539002390566303, "grad_norm": 2.904787302017212, "learning_rate": 5.664160401002506e-06, "loss": 0.25698572397232056, "num_input_tokens_seen": 4056416, "step": 1436, "train_runtime": 4763.8869, "train_tokens_per_second": 851.493 }, { "epoch": 0.11547037907551377, "grad_norm": 2.5543854236602783, "learning_rate": 5.6541353383458654e-06, "loss": 0.22885577380657196, "num_input_tokens_seen": 4060170, "step": 1437, "train_runtime": 4766.6821, "train_tokens_per_second": 851.781 }, { "epoch": 0.11555073424536451, "grad_norm": 3.2168142795562744, "learning_rate": 5.644110275689223e-06, "loss": 0.3062664270401001, "num_input_tokens_seen": 4061702, "step": 1438, "train_runtime": 4768.9308, "train_tokens_per_second": 851.701 }, { "epoch": 0.11563108941521526, "grad_norm": 2.782468795776367, "learning_rate": 5.6340852130325816e-06, "loss": 0.33539703488349915, "num_input_tokens_seen": 4065558, "step": 1439, "train_runtime": 4771.76, "train_tokens_per_second": 852.004 }, { "epoch": 0.11571144458506599, "grad_norm": 2.6941494941711426, "learning_rate": 5.624060150375941e-06, "loss": 0.25932425260543823, "num_input_tokens_seen": 4068598, "step": 1440, "train_runtime": 4774.4118, "train_tokens_per_second": 852.167 }, { "epoch": 0.11579179975491673, "grad_norm": 2.5505142211914062, "learning_rate": 5.6140350877192985e-06, "loss": 0.2479671686887741, "num_input_tokens_seen": 4071176, "step": 1441, "train_runtime": 4788.518, "train_tokens_per_second": 850.195 }, { "epoch": 0.11587215492476748, "grad_norm": 2.4900782108306885, "learning_rate": 5.604010025062657e-06, "loss": 0.21096110343933105, "num_input_tokens_seen": 4074916, "step": 1442, "train_runtime": 4791.3106, "train_tokens_per_second": 850.48 }, { "epoch": 0.11595251009461821, "grad_norm": 2.9089365005493164, "learning_rate": 5.593984962406015e-06, "loss": 0.21680693328380585, "num_input_tokens_seen": 4076536, "step": 1443, "train_runtime": 4793.5941, "train_tokens_per_second": 850.413 }, { "epoch": 0.11603286526446896, "grad_norm": 2.8015928268432617, "learning_rate": 5.583959899749374e-06, "loss": 0.2591073513031006, "num_input_tokens_seen": 4080466, "step": 1444, "train_runtime": 4796.437, "train_tokens_per_second": 850.729 }, { "epoch": 0.1161132204343197, "grad_norm": 2.996544361114502, "learning_rate": 5.5739348370927325e-06, "loss": 0.3013087213039398, "num_input_tokens_seen": 4085082, "step": 1445, "train_runtime": 4799.4552, "train_tokens_per_second": 851.155 }, { "epoch": 0.11619357560417043, "grad_norm": 2.682084321975708, "learning_rate": 5.56390977443609e-06, "loss": 0.20050327479839325, "num_input_tokens_seen": 4087502, "step": 1446, "train_runtime": 4801.9806, "train_tokens_per_second": 851.212 }, { "epoch": 0.11627393077402118, "grad_norm": 3.974398374557495, "learning_rate": 5.5538847117794495e-06, "loss": 0.27522775530815125, "num_input_tokens_seen": 4089386, "step": 1447, "train_runtime": 4804.3084, "train_tokens_per_second": 851.191 }, { "epoch": 0.11635428594387191, "grad_norm": 2.9742932319641113, "learning_rate": 5.543859649122807e-06, "loss": 0.27682211995124817, "num_input_tokens_seen": 4092472, "step": 1448, "train_runtime": 4806.9989, "train_tokens_per_second": 851.357 }, { "epoch": 0.11643464111372265, "grad_norm": 2.6279585361480713, "learning_rate": 5.533834586466166e-06, "loss": 0.15858615934848785, "num_input_tokens_seen": 4095132, "step": 1449, "train_runtime": 4809.5557, "train_tokens_per_second": 851.457 }, { "epoch": 0.1165149962835734, "grad_norm": 2.832261800765991, "learning_rate": 5.523809523809525e-06, "loss": 0.2748209834098816, "num_input_tokens_seen": 4097818, "step": 1450, "train_runtime": 4812.0692, "train_tokens_per_second": 851.571 }, { "epoch": 0.11659535145342413, "grad_norm": 3.4616706371307373, "learning_rate": 5.5137844611528826e-06, "loss": 0.3417867422103882, "num_input_tokens_seen": 4100468, "step": 1451, "train_runtime": 4814.6059, "train_tokens_per_second": 851.673 }, { "epoch": 0.11667570662327488, "grad_norm": 3.1236157417297363, "learning_rate": 5.503759398496241e-06, "loss": 0.25979772210121155, "num_input_tokens_seen": 4103948, "step": 1452, "train_runtime": 4817.3404, "train_tokens_per_second": 851.912 }, { "epoch": 0.11675606179312562, "grad_norm": 3.460836887359619, "learning_rate": 5.493734335839599e-06, "loss": 0.2971200942993164, "num_input_tokens_seen": 4105406, "step": 1453, "train_runtime": 4819.5691, "train_tokens_per_second": 851.82 }, { "epoch": 0.11683641696297635, "grad_norm": 3.683802843093872, "learning_rate": 5.483709273182958e-06, "loss": 0.34153273701667786, "num_input_tokens_seen": 4107814, "step": 1454, "train_runtime": 4822.0441, "train_tokens_per_second": 851.882 }, { "epoch": 0.1169167721328271, "grad_norm": 3.4445955753326416, "learning_rate": 5.4736842105263165e-06, "loss": 0.28405457735061646, "num_input_tokens_seen": 4110320, "step": 1455, "train_runtime": 4824.5116, "train_tokens_per_second": 851.966 }, { "epoch": 0.11699712730267783, "grad_norm": 3.1234376430511475, "learning_rate": 5.463659147869674e-06, "loss": 0.3268517851829529, "num_input_tokens_seen": 4112882, "step": 1456, "train_runtime": 4827.015, "train_tokens_per_second": 852.055 }, { "epoch": 0.11707748247252858, "grad_norm": 3.3386764526367188, "learning_rate": 5.4536340852130335e-06, "loss": 0.29514551162719727, "num_input_tokens_seen": 4115136, "step": 1457, "train_runtime": 4829.4628, "train_tokens_per_second": 852.09 }, { "epoch": 0.11715783764237932, "grad_norm": 3.2079219818115234, "learning_rate": 5.443609022556391e-06, "loss": 0.2872256934642792, "num_input_tokens_seen": 4117018, "step": 1458, "train_runtime": 4831.7754, "train_tokens_per_second": 852.071 }, { "epoch": 0.11723819281223005, "grad_norm": 2.7889466285705566, "learning_rate": 5.43358395989975e-06, "loss": 0.376557856798172, "num_input_tokens_seen": 4119836, "step": 1459, "train_runtime": 4834.3505, "train_tokens_per_second": 852.201 }, { "epoch": 0.1173185479820808, "grad_norm": 2.4744255542755127, "learning_rate": 5.423558897243109e-06, "loss": 0.23823916912078857, "num_input_tokens_seen": 4122918, "step": 1460, "train_runtime": 4836.9611, "train_tokens_per_second": 852.378 }, { "epoch": 0.11739890315193154, "grad_norm": 3.5117545127868652, "learning_rate": 5.413533834586467e-06, "loss": 0.27018940448760986, "num_input_tokens_seen": 4124874, "step": 1461, "train_runtime": 4839.3015, "train_tokens_per_second": 852.37 }, { "epoch": 0.11747925832178228, "grad_norm": 2.6579322814941406, "learning_rate": 5.403508771929825e-06, "loss": 0.27787190675735474, "num_input_tokens_seen": 4129068, "step": 1462, "train_runtime": 4842.2339, "train_tokens_per_second": 852.72 }, { "epoch": 0.11755961349163302, "grad_norm": 2.5772063732147217, "learning_rate": 5.393483709273183e-06, "loss": 0.2821792960166931, "num_input_tokens_seen": 4131834, "step": 1463, "train_runtime": 4844.7999, "train_tokens_per_second": 852.839 }, { "epoch": 0.11763996866148375, "grad_norm": 2.435159683227539, "learning_rate": 5.383458646616542e-06, "loss": 0.2968067526817322, "num_input_tokens_seen": 4134496, "step": 1464, "train_runtime": 4847.341, "train_tokens_per_second": 852.941 }, { "epoch": 0.1177203238313345, "grad_norm": 2.4741392135620117, "learning_rate": 5.3734335839599006e-06, "loss": 0.28385621309280396, "num_input_tokens_seen": 4136898, "step": 1465, "train_runtime": 4849.8257, "train_tokens_per_second": 852.999 }, { "epoch": 0.11780067900118524, "grad_norm": 2.592153549194336, "learning_rate": 5.363408521303258e-06, "loss": 0.23821529746055603, "num_input_tokens_seen": 4140278, "step": 1466, "train_runtime": 4852.5902, "train_tokens_per_second": 853.21 }, { "epoch": 0.11788103417103597, "grad_norm": 3.089360475540161, "learning_rate": 5.3533834586466175e-06, "loss": 0.30427590012550354, "num_input_tokens_seen": 4143442, "step": 1467, "train_runtime": 4855.2481, "train_tokens_per_second": 853.394 }, { "epoch": 0.11796138934088672, "grad_norm": 3.0564987659454346, "learning_rate": 5.343358395989975e-06, "loss": 0.23271848261356354, "num_input_tokens_seen": 4146128, "step": 1468, "train_runtime": 4857.791, "train_tokens_per_second": 853.501 }, { "epoch": 0.11804174451073746, "grad_norm": 2.777866840362549, "learning_rate": 5.333333333333334e-06, "loss": 0.3335420489311218, "num_input_tokens_seen": 4148144, "step": 1469, "train_runtime": 4860.2423, "train_tokens_per_second": 853.485 }, { "epoch": 0.1181220996805882, "grad_norm": 3.293527841567993, "learning_rate": 5.323308270676692e-06, "loss": 0.2835467755794525, "num_input_tokens_seen": 4150420, "step": 1470, "train_runtime": 4862.7393, "train_tokens_per_second": 853.515 }, { "epoch": 0.11820245485043894, "grad_norm": 2.3825302124023438, "learning_rate": 5.313283208020051e-06, "loss": 0.18579310178756714, "num_input_tokens_seen": 4152292, "step": 1471, "train_runtime": 4876.9001, "train_tokens_per_second": 851.42 }, { "epoch": 0.11828281002028967, "grad_norm": 3.277602434158325, "learning_rate": 5.303258145363409e-06, "loss": 0.25148677825927734, "num_input_tokens_seen": 4155118, "step": 1472, "train_runtime": 4879.5727, "train_tokens_per_second": 851.533 }, { "epoch": 0.11836316519014042, "grad_norm": 3.3161399364471436, "learning_rate": 5.293233082706767e-06, "loss": 0.3530995845794678, "num_input_tokens_seen": 4157428, "step": 1473, "train_runtime": 4882.0776, "train_tokens_per_second": 851.569 }, { "epoch": 0.11844352035999116, "grad_norm": 3.0307016372680664, "learning_rate": 5.283208020050126e-06, "loss": 0.28325870633125305, "num_input_tokens_seen": 4159094, "step": 1474, "train_runtime": 4884.3571, "train_tokens_per_second": 851.513 }, { "epoch": 0.11852387552984191, "grad_norm": 2.9295613765716553, "learning_rate": 5.273182957393485e-06, "loss": 0.2938195466995239, "num_input_tokens_seen": 4162194, "step": 1475, "train_runtime": 4887.0306, "train_tokens_per_second": 851.682 }, { "epoch": 0.11860423069969264, "grad_norm": 2.558558702468872, "learning_rate": 5.263157894736842e-06, "loss": 0.22293132543563843, "num_input_tokens_seen": 4164838, "step": 1476, "train_runtime": 4889.6419, "train_tokens_per_second": 851.767 }, { "epoch": 0.11868458586954338, "grad_norm": 2.743661642074585, "learning_rate": 5.253132832080201e-06, "loss": 0.32872116565704346, "num_input_tokens_seen": 4167888, "step": 1477, "train_runtime": 4892.2855, "train_tokens_per_second": 851.931 }, { "epoch": 0.11876494103939413, "grad_norm": 2.9900410175323486, "learning_rate": 5.243107769423559e-06, "loss": 0.39512765407562256, "num_input_tokens_seen": 4170216, "step": 1478, "train_runtime": 4894.764, "train_tokens_per_second": 851.975 }, { "epoch": 0.11884529620924486, "grad_norm": 2.8836870193481445, "learning_rate": 5.233082706766918e-06, "loss": 0.27355799078941345, "num_input_tokens_seen": 4173502, "step": 1479, "train_runtime": 4897.4485, "train_tokens_per_second": 852.179 }, { "epoch": 0.11892565137909561, "grad_norm": 3.90472412109375, "learning_rate": 5.223057644110276e-06, "loss": 0.30894801020622253, "num_input_tokens_seen": 4179600, "step": 1480, "train_runtime": 4900.8417, "train_tokens_per_second": 852.833 }, { "epoch": 0.11900600654894634, "grad_norm": 3.3749773502349854, "learning_rate": 5.213032581453634e-06, "loss": 0.24909842014312744, "num_input_tokens_seen": 4182504, "step": 1481, "train_runtime": 4903.5299, "train_tokens_per_second": 852.958 }, { "epoch": 0.11908636171879708, "grad_norm": 3.291069984436035, "learning_rate": 5.203007518796993e-06, "loss": 0.26263687014579773, "num_input_tokens_seen": 4185082, "step": 1482, "train_runtime": 4906.0745, "train_tokens_per_second": 853.041 }, { "epoch": 0.11916671688864783, "grad_norm": 2.6687004566192627, "learning_rate": 5.192982456140351e-06, "loss": 0.2763051390647888, "num_input_tokens_seen": 4187184, "step": 1483, "train_runtime": 4908.5094, "train_tokens_per_second": 853.046 }, { "epoch": 0.11924707205849856, "grad_norm": 3.367006540298462, "learning_rate": 5.182957393483709e-06, "loss": 0.3233863115310669, "num_input_tokens_seen": 4189872, "step": 1484, "train_runtime": 4911.0564, "train_tokens_per_second": 853.151 }, { "epoch": 0.1193274272283493, "grad_norm": 2.947126865386963, "learning_rate": 5.172932330827069e-06, "loss": 0.25550344586372375, "num_input_tokens_seen": 4191594, "step": 1485, "train_runtime": 4913.3781, "train_tokens_per_second": 853.098 }, { "epoch": 0.11940778239820005, "grad_norm": 3.731600284576416, "learning_rate": 5.162907268170426e-06, "loss": 0.32960042357444763, "num_input_tokens_seen": 4194254, "step": 1486, "train_runtime": 4915.9213, "train_tokens_per_second": 853.198 }, { "epoch": 0.11948813756805078, "grad_norm": 2.716449022293091, "learning_rate": 5.152882205513785e-06, "loss": 0.27400752902030945, "num_input_tokens_seen": 4198598, "step": 1487, "train_runtime": 4918.9242, "train_tokens_per_second": 853.56 }, { "epoch": 0.11956849273790153, "grad_norm": 3.319211959838867, "learning_rate": 5.142857142857142e-06, "loss": 0.2725558876991272, "num_input_tokens_seen": 4200520, "step": 1488, "train_runtime": 4921.3299, "train_tokens_per_second": 853.534 }, { "epoch": 0.11964884790775226, "grad_norm": 3.1523752212524414, "learning_rate": 5.132832080200502e-06, "loss": 0.30173444747924805, "num_input_tokens_seen": 4203038, "step": 1489, "train_runtime": 4923.8084, "train_tokens_per_second": 853.615 }, { "epoch": 0.119729203077603, "grad_norm": 3.45601487159729, "learning_rate": 5.12280701754386e-06, "loss": 0.3390136957168579, "num_input_tokens_seen": 4205196, "step": 1490, "train_runtime": 4926.2214, "train_tokens_per_second": 853.635 }, { "epoch": 0.11980955824745375, "grad_norm": 3.055856227874756, "learning_rate": 5.112781954887218e-06, "loss": 0.22803093492984772, "num_input_tokens_seen": 4207762, "step": 1491, "train_runtime": 4928.7254, "train_tokens_per_second": 853.722 }, { "epoch": 0.11988991341730448, "grad_norm": 3.1930654048919678, "learning_rate": 5.102756892230577e-06, "loss": 0.2162226438522339, "num_input_tokens_seen": 4209532, "step": 1492, "train_runtime": 4931.0424, "train_tokens_per_second": 853.68 }, { "epoch": 0.11997026858715523, "grad_norm": 2.6579906940460205, "learning_rate": 5.092731829573935e-06, "loss": 0.3245127499103546, "num_input_tokens_seen": 4212336, "step": 1493, "train_runtime": 4933.6537, "train_tokens_per_second": 853.796 }, { "epoch": 0.12005062375700597, "grad_norm": 2.995706081390381, "learning_rate": 5.082706766917293e-06, "loss": 0.3383377194404602, "num_input_tokens_seen": 4214588, "step": 1494, "train_runtime": 4936.0943, "train_tokens_per_second": 853.831 }, { "epoch": 0.1201309789268567, "grad_norm": 2.2819552421569824, "learning_rate": 5.072681704260653e-06, "loss": 0.16781920194625854, "num_input_tokens_seen": 4216386, "step": 1495, "train_runtime": 4938.4656, "train_tokens_per_second": 853.785 }, { "epoch": 0.12021133409670745, "grad_norm": 3.0651237964630127, "learning_rate": 5.06265664160401e-06, "loss": 0.2646363079547882, "num_input_tokens_seen": 4219542, "step": 1496, "train_runtime": 4941.1374, "train_tokens_per_second": 853.962 }, { "epoch": 0.12029168926655819, "grad_norm": 2.806114673614502, "learning_rate": 5.052631578947369e-06, "loss": 0.2691468596458435, "num_input_tokens_seen": 4223734, "step": 1497, "train_runtime": 4944.0414, "train_tokens_per_second": 854.308 }, { "epoch": 0.12037204443640893, "grad_norm": 2.7190909385681152, "learning_rate": 5.042606516290726e-06, "loss": 0.3344663083553314, "num_input_tokens_seen": 4227088, "step": 1498, "train_runtime": 4946.8889, "train_tokens_per_second": 854.494 }, { "epoch": 0.12045239960625967, "grad_norm": 2.5185353755950928, "learning_rate": 5.032581453634086e-06, "loss": 0.2771476209163666, "num_input_tokens_seen": 4230194, "step": 1499, "train_runtime": 4949.6166, "train_tokens_per_second": 854.651 }, { "epoch": 0.1205327547761104, "grad_norm": 3.1708884239196777, "learning_rate": 5.022556390977444e-06, "loss": 0.3453780710697174, "num_input_tokens_seen": 4232796, "step": 1500, "train_runtime": 4952.1604, "train_tokens_per_second": 854.737 }, { "epoch": 0.12061310994596115, "grad_norm": 2.731300115585327, "learning_rate": 5.012531328320802e-06, "loss": 0.28870445489883423, "num_input_tokens_seen": 4234906, "step": 1501, "train_runtime": 4966.1278, "train_tokens_per_second": 852.758 }, { "epoch": 0.12069346511581189, "grad_norm": 3.055309295654297, "learning_rate": 5.002506265664161e-06, "loss": 0.339389443397522, "num_input_tokens_seen": 4238744, "step": 1502, "train_runtime": 4968.9552, "train_tokens_per_second": 853.045 }, { "epoch": 0.12077382028566262, "grad_norm": 3.203106164932251, "learning_rate": 4.992481203007519e-06, "loss": 0.2770470082759857, "num_input_tokens_seen": 4241516, "step": 1503, "train_runtime": 4971.5367, "train_tokens_per_second": 853.16 }, { "epoch": 0.12085417545551337, "grad_norm": 3.0062098503112793, "learning_rate": 4.982456140350877e-06, "loss": 0.307538777589798, "num_input_tokens_seen": 4245350, "step": 1504, "train_runtime": 4974.3324, "train_tokens_per_second": 853.451 }, { "epoch": 0.1209345306253641, "grad_norm": 2.947699785232544, "learning_rate": 4.972431077694236e-06, "loss": 0.3436700701713562, "num_input_tokens_seen": 4251428, "step": 1505, "train_runtime": 4977.6859, "train_tokens_per_second": 854.097 }, { "epoch": 0.12101488579521485, "grad_norm": 3.601120948791504, "learning_rate": 4.962406015037594e-06, "loss": 0.23005445301532745, "num_input_tokens_seen": 4256740, "step": 1506, "train_runtime": 4980.91, "train_tokens_per_second": 854.611 }, { "epoch": 0.12109524096506559, "grad_norm": 2.8040966987609863, "learning_rate": 4.952380952380953e-06, "loss": 0.2858981192111969, "num_input_tokens_seen": 4260056, "step": 1507, "train_runtime": 4983.6051, "train_tokens_per_second": 854.814 }, { "epoch": 0.12117559613491632, "grad_norm": 3.309481143951416, "learning_rate": 4.942355889724311e-06, "loss": 0.23356567323207855, "num_input_tokens_seen": 4262662, "step": 1508, "train_runtime": 4986.1422, "train_tokens_per_second": 854.902 }, { "epoch": 0.12125595130476707, "grad_norm": 3.4939756393432617, "learning_rate": 4.93233082706767e-06, "loss": 0.24671828746795654, "num_input_tokens_seen": 4269290, "step": 1509, "train_runtime": 4989.751, "train_tokens_per_second": 855.612 }, { "epoch": 0.12133630647461781, "grad_norm": 3.2025961875915527, "learning_rate": 4.922305764411027e-06, "loss": 0.35596320033073425, "num_input_tokens_seen": 4271732, "step": 1510, "train_runtime": 4992.2496, "train_tokens_per_second": 855.673 }, { "epoch": 0.12141666164446856, "grad_norm": 2.937068223953247, "learning_rate": 4.912280701754386e-06, "loss": 0.27411335706710815, "num_input_tokens_seen": 4274282, "step": 1511, "train_runtime": 4994.7396, "train_tokens_per_second": 855.757 }, { "epoch": 0.12149701681431929, "grad_norm": 2.546400308609009, "learning_rate": 4.902255639097745e-06, "loss": 0.24476122856140137, "num_input_tokens_seen": 4276958, "step": 1512, "train_runtime": 4997.2774, "train_tokens_per_second": 855.858 }, { "epoch": 0.12157737198417003, "grad_norm": 2.64227294921875, "learning_rate": 4.892230576441103e-06, "loss": 0.2522956132888794, "num_input_tokens_seen": 4281258, "step": 1513, "train_runtime": 5000.2246, "train_tokens_per_second": 856.213 }, { "epoch": 0.12165772715402078, "grad_norm": 2.707197904586792, "learning_rate": 4.882205513784461e-06, "loss": 0.2728644013404846, "num_input_tokens_seen": 4284598, "step": 1514, "train_runtime": 5002.9321, "train_tokens_per_second": 856.417 }, { "epoch": 0.12173808232387151, "grad_norm": 2.7783658504486084, "learning_rate": 4.87218045112782e-06, "loss": 0.24135765433311462, "num_input_tokens_seen": 4287166, "step": 1515, "train_runtime": 5005.4379, "train_tokens_per_second": 856.502 }, { "epoch": 0.12181843749372226, "grad_norm": 3.088732957839966, "learning_rate": 4.862155388471178e-06, "loss": 0.2880747616291046, "num_input_tokens_seen": 4289400, "step": 1516, "train_runtime": 5007.8662, "train_tokens_per_second": 856.532 }, { "epoch": 0.121898792663573, "grad_norm": 2.1554574966430664, "learning_rate": 4.852130325814537e-06, "loss": 0.17148688435554504, "num_input_tokens_seen": 4291706, "step": 1517, "train_runtime": 5010.3306, "train_tokens_per_second": 856.571 }, { "epoch": 0.12197914783342373, "grad_norm": 2.8585317134857178, "learning_rate": 4.842105263157895e-06, "loss": 0.26538702845573425, "num_input_tokens_seen": 4294024, "step": 1518, "train_runtime": 5012.7898, "train_tokens_per_second": 856.614 }, { "epoch": 0.12205950300327448, "grad_norm": 2.864565849304199, "learning_rate": 4.832080200501254e-06, "loss": 0.2918477952480316, "num_input_tokens_seen": 4297434, "step": 1519, "train_runtime": 5015.5146, "train_tokens_per_second": 856.828 }, { "epoch": 0.12213985817312521, "grad_norm": 2.9734835624694824, "learning_rate": 4.8220551378446114e-06, "loss": 0.2435590922832489, "num_input_tokens_seen": 4300646, "step": 1520, "train_runtime": 5018.2207, "train_tokens_per_second": 857.006 }, { "epoch": 0.12222021334297595, "grad_norm": 3.4042928218841553, "learning_rate": 4.81203007518797e-06, "loss": 0.2870469391345978, "num_input_tokens_seen": 4303588, "step": 1521, "train_runtime": 5020.8259, "train_tokens_per_second": 857.147 }, { "epoch": 0.1223005685128267, "grad_norm": 2.7857882976531982, "learning_rate": 4.802005012531329e-06, "loss": 0.24200549721717834, "num_input_tokens_seen": 4307516, "step": 1522, "train_runtime": 5023.7512, "train_tokens_per_second": 857.43 }, { "epoch": 0.12238092368267743, "grad_norm": 3.156872272491455, "learning_rate": 4.791979949874687e-06, "loss": 0.32763203978538513, "num_input_tokens_seen": 4310940, "step": 1523, "train_runtime": 5026.604, "train_tokens_per_second": 857.625 }, { "epoch": 0.12246127885252818, "grad_norm": 2.694005250930786, "learning_rate": 4.781954887218045e-06, "loss": 0.26268887519836426, "num_input_tokens_seen": 4315316, "step": 1524, "train_runtime": 5029.6645, "train_tokens_per_second": 857.973 }, { "epoch": 0.12254163402237891, "grad_norm": 2.697019577026367, "learning_rate": 4.771929824561404e-06, "loss": 0.28071093559265137, "num_input_tokens_seen": 4318212, "step": 1525, "train_runtime": 5032.3154, "train_tokens_per_second": 858.096 }, { "epoch": 0.12262198919222965, "grad_norm": 2.4412550926208496, "learning_rate": 4.761904761904762e-06, "loss": 0.18292629718780518, "num_input_tokens_seen": 4320550, "step": 1526, "train_runtime": 5034.8215, "train_tokens_per_second": 858.134 }, { "epoch": 0.1227023443620804, "grad_norm": 2.777437448501587, "learning_rate": 4.751879699248121e-06, "loss": 0.33663976192474365, "num_input_tokens_seen": 4323602, "step": 1527, "train_runtime": 5037.4788, "train_tokens_per_second": 858.287 }, { "epoch": 0.12278269953193113, "grad_norm": 3.4766111373901367, "learning_rate": 4.741854636591479e-06, "loss": 0.32571977376937866, "num_input_tokens_seen": 4326382, "step": 1528, "train_runtime": 5040.0627, "train_tokens_per_second": 858.398 }, { "epoch": 0.12286305470178188, "grad_norm": 3.6585536003112793, "learning_rate": 4.731829573934838e-06, "loss": 0.3420848250389099, "num_input_tokens_seen": 4328718, "step": 1529, "train_runtime": 5042.542, "train_tokens_per_second": 858.44 }, { "epoch": 0.12294340987163262, "grad_norm": 2.642838954925537, "learning_rate": 4.7218045112781955e-06, "loss": 0.2982402741909027, "num_input_tokens_seen": 4330576, "step": 1530, "train_runtime": 5044.8905, "train_tokens_per_second": 858.408 }, { "epoch": 0.12302376504148335, "grad_norm": 2.6689939498901367, "learning_rate": 4.711779448621554e-06, "loss": 0.23559099435806274, "num_input_tokens_seen": 4332412, "step": 1531, "train_runtime": 5059.6916, "train_tokens_per_second": 856.26 }, { "epoch": 0.1231041202113341, "grad_norm": 2.2403523921966553, "learning_rate": 4.7017543859649125e-06, "loss": 0.17892217636108398, "num_input_tokens_seen": 4334566, "step": 1532, "train_runtime": 5062.2366, "train_tokens_per_second": 856.255 }, { "epoch": 0.12318447538118484, "grad_norm": 3.4011120796203613, "learning_rate": 4.691729323308271e-06, "loss": 0.30782049894332886, "num_input_tokens_seen": 4335908, "step": 1533, "train_runtime": 5064.42, "train_tokens_per_second": 856.151 }, { "epoch": 0.12326483055103558, "grad_norm": 3.101360559463501, "learning_rate": 4.6817042606516294e-06, "loss": 0.3074231445789337, "num_input_tokens_seen": 4337674, "step": 1534, "train_runtime": 5066.7929, "train_tokens_per_second": 856.099 }, { "epoch": 0.12334518572088632, "grad_norm": 3.431312084197998, "learning_rate": 4.671679197994988e-06, "loss": 0.3185311257839203, "num_input_tokens_seen": 4342852, "step": 1535, "train_runtime": 5069.9968, "train_tokens_per_second": 856.579 }, { "epoch": 0.12342554089073705, "grad_norm": 2.893465995788574, "learning_rate": 4.661654135338346e-06, "loss": 0.29798492789268494, "num_input_tokens_seen": 4345052, "step": 1536, "train_runtime": 5072.405, "train_tokens_per_second": 856.606 }, { "epoch": 0.1235058960605878, "grad_norm": 3.156512498855591, "learning_rate": 4.651629072681705e-06, "loss": 0.37329667806625366, "num_input_tokens_seen": 4348746, "step": 1537, "train_runtime": 5075.2527, "train_tokens_per_second": 856.853 }, { "epoch": 0.12358625123043854, "grad_norm": 3.0637717247009277, "learning_rate": 4.641604010025063e-06, "loss": 0.26409435272216797, "num_input_tokens_seen": 4350448, "step": 1538, "train_runtime": 5077.5591, "train_tokens_per_second": 856.799 }, { "epoch": 0.12366660640028927, "grad_norm": 2.770779848098755, "learning_rate": 4.631578947368421e-06, "loss": 0.18116110563278198, "num_input_tokens_seen": 4354074, "step": 1539, "train_runtime": 5080.4589, "train_tokens_per_second": 857.024 }, { "epoch": 0.12374696157014002, "grad_norm": 2.8981900215148926, "learning_rate": 4.6215538847117795e-06, "loss": 0.25696051120758057, "num_input_tokens_seen": 4356000, "step": 1540, "train_runtime": 5082.8077, "train_tokens_per_second": 857.007 }, { "epoch": 0.12382731673999076, "grad_norm": 2.8401684761047363, "learning_rate": 4.611528822055138e-06, "loss": 0.24523131549358368, "num_input_tokens_seen": 4357428, "step": 1541, "train_runtime": 5085.0395, "train_tokens_per_second": 856.911 }, { "epoch": 0.1239076719098415, "grad_norm": 3.0031232833862305, "learning_rate": 4.6015037593984965e-06, "loss": 0.3003750145435333, "num_input_tokens_seen": 4361422, "step": 1542, "train_runtime": 5087.8999, "train_tokens_per_second": 857.215 }, { "epoch": 0.12398802707969224, "grad_norm": 2.579862356185913, "learning_rate": 4.591478696741855e-06, "loss": 0.1940459907054901, "num_input_tokens_seen": 4364618, "step": 1543, "train_runtime": 5090.5851, "train_tokens_per_second": 857.39 }, { "epoch": 0.12406838224954297, "grad_norm": 2.6171000003814697, "learning_rate": 4.5814536340852135e-06, "loss": 0.27032357454299927, "num_input_tokens_seen": 4367230, "step": 1544, "train_runtime": 5093.1882, "train_tokens_per_second": 857.465 }, { "epoch": 0.12414873741939372, "grad_norm": 3.1257200241088867, "learning_rate": 4.571428571428572e-06, "loss": 0.2770206034183502, "num_input_tokens_seen": 4370048, "step": 1545, "train_runtime": 5095.7655, "train_tokens_per_second": 857.584 }, { "epoch": 0.12422909258924446, "grad_norm": 2.8265321254730225, "learning_rate": 4.56140350877193e-06, "loss": 0.3434055745601654, "num_input_tokens_seen": 4375584, "step": 1546, "train_runtime": 5099.1357, "train_tokens_per_second": 858.103 }, { "epoch": 0.1243094477590952, "grad_norm": 3.3013854026794434, "learning_rate": 4.551378446115289e-06, "loss": 0.32925301790237427, "num_input_tokens_seen": 4377734, "step": 1547, "train_runtime": 5101.6529, "train_tokens_per_second": 858.101 }, { "epoch": 0.12438980292894594, "grad_norm": 2.6121668815612793, "learning_rate": 4.541353383458647e-06, "loss": 0.29640352725982666, "num_input_tokens_seen": 4381692, "step": 1548, "train_runtime": 5104.5262, "train_tokens_per_second": 858.393 }, { "epoch": 0.12447015809879668, "grad_norm": 3.764406204223633, "learning_rate": 4.531328320802005e-06, "loss": 0.26686936616897583, "num_input_tokens_seen": 4384800, "step": 1549, "train_runtime": 5107.2766, "train_tokens_per_second": 858.54 }, { "epoch": 0.12455051326864743, "grad_norm": 3.0191783905029297, "learning_rate": 4.5213032581453635e-06, "loss": 0.30864834785461426, "num_input_tokens_seen": 4387842, "step": 1550, "train_runtime": 5109.8898, "train_tokens_per_second": 858.696 }, { "epoch": 0.12463086843849816, "grad_norm": 4.339770317077637, "learning_rate": 4.511278195488722e-06, "loss": 0.32658612728118896, "num_input_tokens_seen": 4391202, "step": 1551, "train_runtime": 5112.6863, "train_tokens_per_second": 858.884 }, { "epoch": 0.12471122360834891, "grad_norm": 3.5194594860076904, "learning_rate": 4.5012531328320805e-06, "loss": 0.2709728479385376, "num_input_tokens_seen": 4392702, "step": 1552, "train_runtime": 5114.9812, "train_tokens_per_second": 858.791 }, { "epoch": 0.12479157877819964, "grad_norm": 2.785592794418335, "learning_rate": 4.491228070175439e-06, "loss": 0.29841822385787964, "num_input_tokens_seen": 4397348, "step": 1553, "train_runtime": 5118.0413, "train_tokens_per_second": 859.186 }, { "epoch": 0.12487193394805038, "grad_norm": 2.76931095123291, "learning_rate": 4.4812030075187975e-06, "loss": 0.22457507252693176, "num_input_tokens_seen": 4400502, "step": 1554, "train_runtime": 5120.7257, "train_tokens_per_second": 859.351 }, { "epoch": 0.12495228911790113, "grad_norm": 3.8501720428466797, "learning_rate": 4.471177944862156e-06, "loss": 0.21528595685958862, "num_input_tokens_seen": 4403072, "step": 1555, "train_runtime": 5123.3571, "train_tokens_per_second": 859.411 }, { "epoch": 0.12503264428775188, "grad_norm": 2.8727564811706543, "learning_rate": 4.461152882205514e-06, "loss": 0.2829807996749878, "num_input_tokens_seen": 4404606, "step": 1556, "train_runtime": 5125.6166, "train_tokens_per_second": 859.332 }, { "epoch": 0.1251129994576026, "grad_norm": 3.1478970050811768, "learning_rate": 4.451127819548873e-06, "loss": 0.3154464066028595, "num_input_tokens_seen": 4406812, "step": 1557, "train_runtime": 5128.0316, "train_tokens_per_second": 859.357 }, { "epoch": 0.12519335462745335, "grad_norm": 2.7064621448516846, "learning_rate": 4.4411027568922314e-06, "loss": 0.2691633701324463, "num_input_tokens_seen": 4410586, "step": 1558, "train_runtime": 5130.9059, "train_tokens_per_second": 859.612 }, { "epoch": 0.1252737097973041, "grad_norm": 3.097320079803467, "learning_rate": 4.431077694235589e-06, "loss": 0.378887802362442, "num_input_tokens_seen": 4413906, "step": 1559, "train_runtime": 5133.6102, "train_tokens_per_second": 859.805 }, { "epoch": 0.12535406496715482, "grad_norm": 2.5535833835601807, "learning_rate": 4.4210526315789476e-06, "loss": 0.23710960149765015, "num_input_tokens_seen": 4416256, "step": 1560, "train_runtime": 5136.098, "train_tokens_per_second": 859.847 }, { "epoch": 0.12543442013700556, "grad_norm": 2.7486019134521484, "learning_rate": 4.411027568922306e-06, "loss": 0.17418622970581055, "num_input_tokens_seen": 4418430, "step": 1561, "train_runtime": 5148.8192, "train_tokens_per_second": 858.144 }, { "epoch": 0.1255147753068563, "grad_norm": 3.28747296333313, "learning_rate": 4.4010025062656645e-06, "loss": 0.3157857656478882, "num_input_tokens_seen": 4421136, "step": 1562, "train_runtime": 5151.4976, "train_tokens_per_second": 858.223 }, { "epoch": 0.12559513047670703, "grad_norm": 3.486398220062256, "learning_rate": 4.390977443609023e-06, "loss": 0.29543566703796387, "num_input_tokens_seen": 4423166, "step": 1563, "train_runtime": 5153.9763, "train_tokens_per_second": 858.205 }, { "epoch": 0.12567548564655778, "grad_norm": 2.46333909034729, "learning_rate": 4.3809523809523815e-06, "loss": 0.24228858947753906, "num_input_tokens_seen": 4425078, "step": 1564, "train_runtime": 5156.3954, "train_tokens_per_second": 858.173 }, { "epoch": 0.12575584081640853, "grad_norm": 2.4858338832855225, "learning_rate": 4.370927318295739e-06, "loss": 0.17682811617851257, "num_input_tokens_seen": 4427772, "step": 1565, "train_runtime": 5158.9592, "train_tokens_per_second": 858.268 }, { "epoch": 0.12583619598625925, "grad_norm": 3.7024409770965576, "learning_rate": 4.360902255639098e-06, "loss": 0.3545113801956177, "num_input_tokens_seen": 4429158, "step": 1566, "train_runtime": 5161.1816, "train_tokens_per_second": 858.167 }, { "epoch": 0.12591655115611, "grad_norm": 3.2303340435028076, "learning_rate": 4.350877192982457e-06, "loss": 0.3681087791919708, "num_input_tokens_seen": 4432702, "step": 1567, "train_runtime": 5163.9438, "train_tokens_per_second": 858.395 }, { "epoch": 0.12599690632596075, "grad_norm": 3.467773199081421, "learning_rate": 4.340852130325815e-06, "loss": 0.3672095835208893, "num_input_tokens_seen": 4434552, "step": 1568, "train_runtime": 5166.3036, "train_tokens_per_second": 858.361 }, { "epoch": 0.1260772614958115, "grad_norm": 3.213836669921875, "learning_rate": 4.330827067669173e-06, "loss": 0.27541661262512207, "num_input_tokens_seen": 4436262, "step": 1569, "train_runtime": 5168.6735, "train_tokens_per_second": 858.298 }, { "epoch": 0.12615761666566222, "grad_norm": 3.145123243331909, "learning_rate": 4.320802005012532e-06, "loss": 0.31866660714149475, "num_input_tokens_seen": 4439336, "step": 1570, "train_runtime": 5171.3378, "train_tokens_per_second": 858.45 }, { "epoch": 0.12623797183551297, "grad_norm": 2.2678487300872803, "learning_rate": 4.31077694235589e-06, "loss": 0.2358078807592392, "num_input_tokens_seen": 4442212, "step": 1571, "train_runtime": 5173.8882, "train_tokens_per_second": 858.583 }, { "epoch": 0.12631832700536372, "grad_norm": 3.1890170574188232, "learning_rate": 4.3007518796992486e-06, "loss": 0.27289140224456787, "num_input_tokens_seen": 4443570, "step": 1572, "train_runtime": 5176.1062, "train_tokens_per_second": 858.477 }, { "epoch": 0.12639868217521444, "grad_norm": 2.9951303005218506, "learning_rate": 4.290726817042607e-06, "loss": 0.32761648297309875, "num_input_tokens_seen": 4445578, "step": 1573, "train_runtime": 5178.4598, "train_tokens_per_second": 858.475 }, { "epoch": 0.1264790373450652, "grad_norm": 3.1068384647369385, "learning_rate": 4.2807017543859656e-06, "loss": 0.20398426055908203, "num_input_tokens_seen": 4450036, "step": 1574, "train_runtime": 5181.4818, "train_tokens_per_second": 858.835 }, { "epoch": 0.12655939251491594, "grad_norm": 3.521606922149658, "learning_rate": 4.270676691729323e-06, "loss": 0.2504396438598633, "num_input_tokens_seen": 4452236, "step": 1575, "train_runtime": 5183.9039, "train_tokens_per_second": 858.858 }, { "epoch": 0.12663974768476666, "grad_norm": 3.0894813537597656, "learning_rate": 4.260651629072682e-06, "loss": 0.25354331731796265, "num_input_tokens_seen": 4455016, "step": 1576, "train_runtime": 5186.4485, "train_tokens_per_second": 858.972 }, { "epoch": 0.1267201028546174, "grad_norm": 3.273036003112793, "learning_rate": 4.250626566416041e-06, "loss": 0.2948225140571594, "num_input_tokens_seen": 4456720, "step": 1577, "train_runtime": 5188.7269, "train_tokens_per_second": 858.924 }, { "epoch": 0.12680045802446815, "grad_norm": 2.1898884773254395, "learning_rate": 4.240601503759399e-06, "loss": 0.24644124507904053, "num_input_tokens_seen": 4459716, "step": 1578, "train_runtime": 5191.3339, "train_tokens_per_second": 859.069 }, { "epoch": 0.1268808131943189, "grad_norm": 2.5287981033325195, "learning_rate": 4.230576441102757e-06, "loss": 0.2850573658943176, "num_input_tokens_seen": 4465662, "step": 1579, "train_runtime": 5194.6501, "train_tokens_per_second": 859.666 }, { "epoch": 0.12696116836416962, "grad_norm": 3.5528640747070312, "learning_rate": 4.220551378446116e-06, "loss": 0.24940133094787598, "num_input_tokens_seen": 4468002, "step": 1580, "train_runtime": 5197.0972, "train_tokens_per_second": 859.711 }, { "epoch": 0.12704152353402037, "grad_norm": 2.957428455352783, "learning_rate": 4.210526315789474e-06, "loss": 0.28707653284072876, "num_input_tokens_seen": 4470394, "step": 1581, "train_runtime": 5199.5302, "train_tokens_per_second": 859.769 }, { "epoch": 0.12712187870387112, "grad_norm": 2.5958659648895264, "learning_rate": 4.200501253132833e-06, "loss": 0.19343242049217224, "num_input_tokens_seen": 4472830, "step": 1582, "train_runtime": 5202.0483, "train_tokens_per_second": 859.821 }, { "epoch": 0.12720223387372184, "grad_norm": 3.040623188018799, "learning_rate": 4.190476190476191e-06, "loss": 0.23508931696414948, "num_input_tokens_seen": 4475820, "step": 1583, "train_runtime": 5204.6431, "train_tokens_per_second": 859.967 }, { "epoch": 0.1272825890435726, "grad_norm": 3.039968729019165, "learning_rate": 4.18045112781955e-06, "loss": 0.2774313688278198, "num_input_tokens_seen": 4478900, "step": 1584, "train_runtime": 5207.2741, "train_tokens_per_second": 860.124 }, { "epoch": 0.12736294421342334, "grad_norm": 2.825068235397339, "learning_rate": 4.170426065162907e-06, "loss": 0.36368417739868164, "num_input_tokens_seen": 4481784, "step": 1585, "train_runtime": 5209.8696, "train_tokens_per_second": 860.249 }, { "epoch": 0.12744329938327406, "grad_norm": 2.442612886428833, "learning_rate": 4.160401002506266e-06, "loss": 0.27431151270866394, "num_input_tokens_seen": 4484728, "step": 1586, "train_runtime": 5212.4843, "train_tokens_per_second": 860.382 }, { "epoch": 0.1275236545531248, "grad_norm": 3.1194427013397217, "learning_rate": 4.150375939849624e-06, "loss": 0.31946033239364624, "num_input_tokens_seen": 4486630, "step": 1587, "train_runtime": 5214.8405, "train_tokens_per_second": 860.358 }, { "epoch": 0.12760400972297556, "grad_norm": 3.2785778045654297, "learning_rate": 4.140350877192983e-06, "loss": 0.362075537443161, "num_input_tokens_seen": 4490214, "step": 1588, "train_runtime": 5217.5909, "train_tokens_per_second": 860.591 }, { "epoch": 0.12768436489282628, "grad_norm": 2.912715435028076, "learning_rate": 4.130325814536341e-06, "loss": 0.28062406182289124, "num_input_tokens_seen": 4494472, "step": 1589, "train_runtime": 5220.4973, "train_tokens_per_second": 860.928 }, { "epoch": 0.12776472006267703, "grad_norm": 2.3348228931427, "learning_rate": 4.1203007518797e-06, "loss": 0.21747004985809326, "num_input_tokens_seen": 4496950, "step": 1590, "train_runtime": 5222.9672, "train_tokens_per_second": 860.995 }, { "epoch": 0.12784507523252778, "grad_norm": 3.111332654953003, "learning_rate": 4.110275689223058e-06, "loss": 0.32543712854385376, "num_input_tokens_seen": 4500420, "step": 1591, "train_runtime": 5237.9647, "train_tokens_per_second": 859.193 }, { "epoch": 0.12792543040237853, "grad_norm": 3.049992799758911, "learning_rate": 4.100250626566417e-06, "loss": 0.27476322650909424, "num_input_tokens_seen": 4501874, "step": 1592, "train_runtime": 5240.2324, "train_tokens_per_second": 859.098 }, { "epoch": 0.12800578557222925, "grad_norm": 3.612494945526123, "learning_rate": 4.090225563909775e-06, "loss": 0.22073635458946228, "num_input_tokens_seen": 4503296, "step": 1593, "train_runtime": 5242.4951, "train_tokens_per_second": 858.999 }, { "epoch": 0.12808614074208, "grad_norm": 2.910612106323242, "learning_rate": 4.080200501253133e-06, "loss": 0.28355786204338074, "num_input_tokens_seen": 4505626, "step": 1594, "train_runtime": 5244.9609, "train_tokens_per_second": 859.039 }, { "epoch": 0.12816649591193074, "grad_norm": 3.1616811752319336, "learning_rate": 4.070175438596491e-06, "loss": 0.3286800980567932, "num_input_tokens_seen": 4508960, "step": 1595, "train_runtime": 5247.6886, "train_tokens_per_second": 859.228 }, { "epoch": 0.12824685108178147, "grad_norm": 3.134964942932129, "learning_rate": 4.06015037593985e-06, "loss": 0.3819255232810974, "num_input_tokens_seen": 4512476, "step": 1596, "train_runtime": 5250.4906, "train_tokens_per_second": 859.439 }, { "epoch": 0.12832720625163221, "grad_norm": 3.2321438789367676, "learning_rate": 4.050125313283208e-06, "loss": 0.2924082279205322, "num_input_tokens_seen": 4515072, "step": 1597, "train_runtime": 5253.0507, "train_tokens_per_second": 859.514 }, { "epoch": 0.12840756142148296, "grad_norm": 3.253018379211426, "learning_rate": 4.040100250626567e-06, "loss": 0.2898171544075012, "num_input_tokens_seen": 4516792, "step": 1598, "train_runtime": 5255.3396, "train_tokens_per_second": 859.467 }, { "epoch": 0.12848791659133368, "grad_norm": 3.2549591064453125, "learning_rate": 4.030075187969925e-06, "loss": 0.25828102231025696, "num_input_tokens_seen": 4521138, "step": 1599, "train_runtime": 5258.3352, "train_tokens_per_second": 859.804 }, { "epoch": 0.12856827176118443, "grad_norm": 2.734081983566284, "learning_rate": 4.020050125313284e-06, "loss": 0.24324357509613037, "num_input_tokens_seen": 4524996, "step": 1600, "train_runtime": 5261.1944, "train_tokens_per_second": 860.07 }, { "epoch": 0.12856827176118443, "eval_loss": 1.096724033355713, "eval_runtime": 69.0307, "eval_samples_per_second": 14.399, "eval_steps_per_second": 7.2, "num_input_tokens_seen": 4524996, "step": 1600 }, { "epoch": 0.12864862693103518, "grad_norm": 2.6073803901672363, "learning_rate": 4.010025062656641e-06, "loss": 0.3659200072288513, "num_input_tokens_seen": 4528174, "step": 1601, "train_runtime": 5332.9625, "train_tokens_per_second": 849.092 }, { "epoch": 0.1287289821008859, "grad_norm": 3.789525032043457, "learning_rate": 4.000000000000001e-06, "loss": 0.35582900047302246, "num_input_tokens_seen": 4532116, "step": 1602, "train_runtime": 5336.0749, "train_tokens_per_second": 849.335 }, { "epoch": 0.12880933727073665, "grad_norm": 1.9736008644104004, "learning_rate": 3.989974937343359e-06, "loss": 0.21662628650665283, "num_input_tokens_seen": 4536096, "step": 1603, "train_runtime": 5338.9604, "train_tokens_per_second": 849.622 }, { "epoch": 0.1288896924405874, "grad_norm": 2.658519983291626, "learning_rate": 3.979949874686717e-06, "loss": 0.2852608263492584, "num_input_tokens_seen": 4540288, "step": 1604, "train_runtime": 5342.1181, "train_tokens_per_second": 849.904 }, { "epoch": 0.12897004761043815, "grad_norm": 2.6562435626983643, "learning_rate": 3.969924812030075e-06, "loss": 0.22401344776153564, "num_input_tokens_seen": 4544080, "step": 1605, "train_runtime": 5345.1276, "train_tokens_per_second": 850.135 }, { "epoch": 0.12905040278028887, "grad_norm": 2.404392719268799, "learning_rate": 3.959899749373434e-06, "loss": 0.25996530055999756, "num_input_tokens_seen": 4547124, "step": 1606, "train_runtime": 5347.817, "train_tokens_per_second": 850.277 }, { "epoch": 0.12913075795013962, "grad_norm": 4.269808292388916, "learning_rate": 3.949874686716792e-06, "loss": 0.28057506680488586, "num_input_tokens_seen": 4549892, "step": 1607, "train_runtime": 5350.4004, "train_tokens_per_second": 850.383 }, { "epoch": 0.12921111311999037, "grad_norm": 3.1833913326263428, "learning_rate": 3.939849624060151e-06, "loss": 0.2964240610599518, "num_input_tokens_seen": 4554236, "step": 1608, "train_runtime": 5353.4266, "train_tokens_per_second": 850.714 }, { "epoch": 0.1292914682898411, "grad_norm": 2.6291544437408447, "learning_rate": 3.929824561403509e-06, "loss": 0.24665242433547974, "num_input_tokens_seen": 4557620, "step": 1609, "train_runtime": 5356.1205, "train_tokens_per_second": 850.918 }, { "epoch": 0.12937182345969184, "grad_norm": 3.098881483078003, "learning_rate": 3.919799498746868e-06, "loss": 0.24723756313323975, "num_input_tokens_seen": 4560608, "step": 1610, "train_runtime": 5358.7426, "train_tokens_per_second": 851.059 }, { "epoch": 0.12945217862954259, "grad_norm": 2.770930767059326, "learning_rate": 3.909774436090225e-06, "loss": 0.2384951114654541, "num_input_tokens_seen": 4562086, "step": 1611, "train_runtime": 5361.0632, "train_tokens_per_second": 850.967 }, { "epoch": 0.1295325337993933, "grad_norm": 3.298766613006592, "learning_rate": 3.899749373433585e-06, "loss": 0.25935304164886475, "num_input_tokens_seen": 4566794, "step": 1612, "train_runtime": 5364.1709, "train_tokens_per_second": 851.351 }, { "epoch": 0.12961288896924406, "grad_norm": 3.339351177215576, "learning_rate": 3.889724310776943e-06, "loss": 0.32792750000953674, "num_input_tokens_seen": 4569148, "step": 1613, "train_runtime": 5366.6934, "train_tokens_per_second": 851.39 }, { "epoch": 0.1296932441390948, "grad_norm": 2.990955114364624, "learning_rate": 3.879699248120301e-06, "loss": 0.2783536911010742, "num_input_tokens_seen": 4571608, "step": 1614, "train_runtime": 5369.2419, "train_tokens_per_second": 851.444 }, { "epoch": 0.12977359930894555, "grad_norm": 3.0776004791259766, "learning_rate": 3.869674185463659e-06, "loss": 0.30076685547828674, "num_input_tokens_seen": 4573378, "step": 1615, "train_runtime": 5371.5695, "train_tokens_per_second": 851.404 }, { "epoch": 0.12985395447879627, "grad_norm": 2.89322829246521, "learning_rate": 3.859649122807018e-06, "loss": 0.24292536079883575, "num_input_tokens_seen": 4576894, "step": 1616, "train_runtime": 5374.3871, "train_tokens_per_second": 851.612 }, { "epoch": 0.12993430964864702, "grad_norm": 3.08730411529541, "learning_rate": 3.849624060150376e-06, "loss": 0.2734002470970154, "num_input_tokens_seen": 4579864, "step": 1617, "train_runtime": 5377.008, "train_tokens_per_second": 851.75 }, { "epoch": 0.13001466481849777, "grad_norm": 2.974649667739868, "learning_rate": 3.839598997493735e-06, "loss": 0.27682948112487793, "num_input_tokens_seen": 4582040, "step": 1618, "train_runtime": 5379.3763, "train_tokens_per_second": 851.779 }, { "epoch": 0.1300950199883485, "grad_norm": 2.4524142742156982, "learning_rate": 3.829573934837093e-06, "loss": 0.2066311091184616, "num_input_tokens_seen": 4583614, "step": 1619, "train_runtime": 5381.618, "train_tokens_per_second": 851.717 }, { "epoch": 0.13017537515819924, "grad_norm": 3.290361166000366, "learning_rate": 3.819548872180452e-06, "loss": 0.2680341899394989, "num_input_tokens_seen": 4585582, "step": 1620, "train_runtime": 5383.9497, "train_tokens_per_second": 851.713 }, { "epoch": 0.13025573032805, "grad_norm": 2.974470615386963, "learning_rate": 3.80952380952381e-06, "loss": 0.30632761120796204, "num_input_tokens_seen": 4587204, "step": 1621, "train_runtime": 5397.8204, "train_tokens_per_second": 849.825 }, { "epoch": 0.1303360854979007, "grad_norm": 2.5080368518829346, "learning_rate": 3.7994987468671683e-06, "loss": 0.2757986783981323, "num_input_tokens_seen": 4590164, "step": 1622, "train_runtime": 5400.4095, "train_tokens_per_second": 849.966 }, { "epoch": 0.13041644066775146, "grad_norm": 2.9822890758514404, "learning_rate": 3.789473684210527e-06, "loss": 0.2486990988254547, "num_input_tokens_seen": 4594244, "step": 1623, "train_runtime": 5403.2814, "train_tokens_per_second": 850.269 }, { "epoch": 0.1304967958376022, "grad_norm": 3.143510580062866, "learning_rate": 3.779448621553885e-06, "loss": 0.28181135654449463, "num_input_tokens_seen": 4597712, "step": 1624, "train_runtime": 5405.9948, "train_tokens_per_second": 850.484 }, { "epoch": 0.13057715100745293, "grad_norm": 2.578763246536255, "learning_rate": 3.7694235588972433e-06, "loss": 0.19986355304718018, "num_input_tokens_seen": 4600786, "step": 1625, "train_runtime": 5408.6383, "train_tokens_per_second": 850.637 }, { "epoch": 0.13065750617730368, "grad_norm": 2.900402784347534, "learning_rate": 3.7593984962406014e-06, "loss": 0.2729087471961975, "num_input_tokens_seen": 4604596, "step": 1626, "train_runtime": 5411.4841, "train_tokens_per_second": 850.893 }, { "epoch": 0.13073786134715443, "grad_norm": 2.9734342098236084, "learning_rate": 3.7493734335839603e-06, "loss": 0.23576216399669647, "num_input_tokens_seen": 4606224, "step": 1627, "train_runtime": 5413.7598, "train_tokens_per_second": 850.836 }, { "epoch": 0.13081821651700518, "grad_norm": 2.236919403076172, "learning_rate": 3.739348370927319e-06, "loss": 0.1833740770816803, "num_input_tokens_seen": 4608290, "step": 1628, "train_runtime": 5416.1531, "train_tokens_per_second": 850.842 }, { "epoch": 0.1308985716868559, "grad_norm": 2.6004817485809326, "learning_rate": 3.729323308270677e-06, "loss": 0.21652820706367493, "num_input_tokens_seen": 4609878, "step": 1629, "train_runtime": 5418.4284, "train_tokens_per_second": 850.778 }, { "epoch": 0.13097892685670665, "grad_norm": 3.064807653427124, "learning_rate": 3.7192982456140354e-06, "loss": 0.3599761128425598, "num_input_tokens_seen": 4613090, "step": 1630, "train_runtime": 5421.112, "train_tokens_per_second": 850.949 }, { "epoch": 0.1310592820265574, "grad_norm": 3.5223734378814697, "learning_rate": 3.7092731829573934e-06, "loss": 0.3077417314052582, "num_input_tokens_seen": 4615258, "step": 1631, "train_runtime": 5423.5705, "train_tokens_per_second": 850.963 }, { "epoch": 0.13113963719640812, "grad_norm": 3.220855951309204, "learning_rate": 3.6992481203007523e-06, "loss": 0.2985258102416992, "num_input_tokens_seen": 4617094, "step": 1632, "train_runtime": 5425.892, "train_tokens_per_second": 850.937 }, { "epoch": 0.13121999236625886, "grad_norm": 3.4946353435516357, "learning_rate": 3.689223057644111e-06, "loss": 0.2741073966026306, "num_input_tokens_seen": 4620036, "step": 1633, "train_runtime": 5428.4893, "train_tokens_per_second": 851.072 }, { "epoch": 0.1313003475361096, "grad_norm": 2.7613909244537354, "learning_rate": 3.679197994987469e-06, "loss": 0.17059437930583954, "num_input_tokens_seen": 4624012, "step": 1634, "train_runtime": 5431.3526, "train_tokens_per_second": 851.356 }, { "epoch": 0.13138070270596033, "grad_norm": 2.818030834197998, "learning_rate": 3.6691729323308274e-06, "loss": 0.22798065841197968, "num_input_tokens_seen": 4627478, "step": 1635, "train_runtime": 5434.1088, "train_tokens_per_second": 851.562 }, { "epoch": 0.13146105787581108, "grad_norm": 2.6767661571502686, "learning_rate": 3.6591478696741854e-06, "loss": 0.26027464866638184, "num_input_tokens_seen": 4629720, "step": 1636, "train_runtime": 5436.5222, "train_tokens_per_second": 851.596 }, { "epoch": 0.13154141304566183, "grad_norm": 2.838437080383301, "learning_rate": 3.6491228070175443e-06, "loss": 0.3061496317386627, "num_input_tokens_seen": 4632306, "step": 1637, "train_runtime": 5439.0357, "train_tokens_per_second": 851.678 }, { "epoch": 0.13162176821551255, "grad_norm": 3.1988513469696045, "learning_rate": 3.639097744360903e-06, "loss": 0.3043241500854492, "num_input_tokens_seen": 4634490, "step": 1638, "train_runtime": 5441.5929, "train_tokens_per_second": 851.679 }, { "epoch": 0.1317021233853633, "grad_norm": 2.745096206665039, "learning_rate": 3.629072681704261e-06, "loss": 0.24540069699287415, "num_input_tokens_seen": 4636418, "step": 1639, "train_runtime": 5443.9671, "train_tokens_per_second": 851.662 }, { "epoch": 0.13178247855521405, "grad_norm": 3.031689405441284, "learning_rate": 3.6190476190476194e-06, "loss": 0.29343026876449585, "num_input_tokens_seen": 4638926, "step": 1640, "train_runtime": 5446.5221, "train_tokens_per_second": 851.723 }, { "epoch": 0.1318628337250648, "grad_norm": 2.6522254943847656, "learning_rate": 3.6090225563909775e-06, "loss": 0.1807267665863037, "num_input_tokens_seen": 4642282, "step": 1641, "train_runtime": 5449.244, "train_tokens_per_second": 851.913 }, { "epoch": 0.13194318889491552, "grad_norm": 2.6758854389190674, "learning_rate": 3.5989974937343364e-06, "loss": 0.22280552983283997, "num_input_tokens_seen": 4645342, "step": 1642, "train_runtime": 5451.9141, "train_tokens_per_second": 852.057 }, { "epoch": 0.13202354406476627, "grad_norm": 3.334634304046631, "learning_rate": 3.588972431077695e-06, "loss": 0.27707386016845703, "num_input_tokens_seen": 4646950, "step": 1643, "train_runtime": 5454.16, "train_tokens_per_second": 852.001 }, { "epoch": 0.13210389923461702, "grad_norm": 3.1396286487579346, "learning_rate": 3.578947368421053e-06, "loss": 0.24969074130058289, "num_input_tokens_seen": 4649516, "step": 1644, "train_runtime": 5456.6901, "train_tokens_per_second": 852.076 }, { "epoch": 0.13218425440446774, "grad_norm": 3.771538257598877, "learning_rate": 3.5689223057644114e-06, "loss": 0.2481880486011505, "num_input_tokens_seen": 4651044, "step": 1645, "train_runtime": 5458.9646, "train_tokens_per_second": 852.001 }, { "epoch": 0.1322646095743185, "grad_norm": 2.2247567176818848, "learning_rate": 3.5588972431077695e-06, "loss": 0.22516387701034546, "num_input_tokens_seen": 4654060, "step": 1646, "train_runtime": 5461.5947, "train_tokens_per_second": 852.143 }, { "epoch": 0.13234496474416924, "grad_norm": 2.957327365875244, "learning_rate": 3.5488721804511284e-06, "loss": 0.27436932921409607, "num_input_tokens_seen": 4655470, "step": 1647, "train_runtime": 5463.809, "train_tokens_per_second": 852.056 }, { "epoch": 0.13242531991401996, "grad_norm": 2.762390613555908, "learning_rate": 3.5388471177944864e-06, "loss": 0.2583554983139038, "num_input_tokens_seen": 4659028, "step": 1648, "train_runtime": 5466.5976, "train_tokens_per_second": 852.272 }, { "epoch": 0.1325056750838707, "grad_norm": 2.484198570251465, "learning_rate": 3.528822055137845e-06, "loss": 0.261231392621994, "num_input_tokens_seen": 4662422, "step": 1649, "train_runtime": 5469.3661, "train_tokens_per_second": 852.461 }, { "epoch": 0.13258603025372145, "grad_norm": 3.2716386318206787, "learning_rate": 3.5187969924812034e-06, "loss": 0.26729220151901245, "num_input_tokens_seen": 4664048, "step": 1650, "train_runtime": 5471.6494, "train_tokens_per_second": 852.403 }, { "epoch": 0.1326663854235722, "grad_norm": 2.701476812362671, "learning_rate": 3.5087719298245615e-06, "loss": 0.29118356108665466, "num_input_tokens_seen": 4667058, "step": 1651, "train_runtime": 5486.1161, "train_tokens_per_second": 850.703 }, { "epoch": 0.13274674059342292, "grad_norm": 3.2550511360168457, "learning_rate": 3.4987468671679204e-06, "loss": 0.30382823944091797, "num_input_tokens_seen": 4668912, "step": 1652, "train_runtime": 5488.5548, "train_tokens_per_second": 850.663 }, { "epoch": 0.13282709576327367, "grad_norm": 2.937861204147339, "learning_rate": 3.4887218045112785e-06, "loss": 0.3328143060207367, "num_input_tokens_seen": 4671694, "step": 1653, "train_runtime": 5491.1275, "train_tokens_per_second": 850.771 }, { "epoch": 0.13290745093312442, "grad_norm": 2.9086825847625732, "learning_rate": 3.478696741854637e-06, "loss": 0.283705472946167, "num_input_tokens_seen": 4674172, "step": 1654, "train_runtime": 5493.625, "train_tokens_per_second": 850.836 }, { "epoch": 0.13298780610297514, "grad_norm": 3.0282235145568848, "learning_rate": 3.468671679197995e-06, "loss": 0.2438664436340332, "num_input_tokens_seen": 4676406, "step": 1655, "train_runtime": 5496.0645, "train_tokens_per_second": 850.864 }, { "epoch": 0.1330681612728259, "grad_norm": 2.2459821701049805, "learning_rate": 3.4586466165413535e-06, "loss": 0.1973123848438263, "num_input_tokens_seen": 4679000, "step": 1656, "train_runtime": 5498.5613, "train_tokens_per_second": 850.95 }, { "epoch": 0.13314851644267664, "grad_norm": 3.679588556289673, "learning_rate": 3.4486215538847124e-06, "loss": 0.21945686638355255, "num_input_tokens_seen": 4684334, "step": 1657, "train_runtime": 5501.8569, "train_tokens_per_second": 851.41 }, { "epoch": 0.13322887161252736, "grad_norm": 2.7196996212005615, "learning_rate": 3.4385964912280705e-06, "loss": 0.2902495563030243, "num_input_tokens_seen": 4689096, "step": 1658, "train_runtime": 5504.8961, "train_tokens_per_second": 851.805 }, { "epoch": 0.1333092267823781, "grad_norm": 2.8299779891967773, "learning_rate": 3.428571428571429e-06, "loss": 0.2582407295703888, "num_input_tokens_seen": 4692796, "step": 1659, "train_runtime": 5507.6482, "train_tokens_per_second": 852.051 }, { "epoch": 0.13338958195222886, "grad_norm": 3.1810829639434814, "learning_rate": 3.418546365914787e-06, "loss": 0.2985352873802185, "num_input_tokens_seen": 4695054, "step": 1660, "train_runtime": 5510.2229, "train_tokens_per_second": 852.062 }, { "epoch": 0.13346993712207958, "grad_norm": 2.752344846725464, "learning_rate": 3.4085213032581455e-06, "loss": 0.2649127244949341, "num_input_tokens_seen": 4699016, "step": 1661, "train_runtime": 5513.1889, "train_tokens_per_second": 852.323 }, { "epoch": 0.13355029229193033, "grad_norm": 3.066635847091675, "learning_rate": 3.3984962406015044e-06, "loss": 0.26856717467308044, "num_input_tokens_seen": 4703458, "step": 1662, "train_runtime": 5516.1392, "train_tokens_per_second": 852.672 }, { "epoch": 0.13363064746178108, "grad_norm": 3.4716458320617676, "learning_rate": 3.3884711779448625e-06, "loss": 0.35104241967201233, "num_input_tokens_seen": 4705366, "step": 1663, "train_runtime": 5518.5018, "train_tokens_per_second": 852.653 }, { "epoch": 0.13371100263163183, "grad_norm": 2.786759376525879, "learning_rate": 3.378446115288221e-06, "loss": 0.27460747957229614, "num_input_tokens_seen": 4707308, "step": 1664, "train_runtime": 5520.9617, "train_tokens_per_second": 852.625 }, { "epoch": 0.13379135780148255, "grad_norm": 3.4815237522125244, "learning_rate": 3.368421052631579e-06, "loss": 0.22290366888046265, "num_input_tokens_seen": 4711218, "step": 1665, "train_runtime": 5523.8099, "train_tokens_per_second": 852.893 }, { "epoch": 0.1338717129713333, "grad_norm": 2.517517328262329, "learning_rate": 3.3583959899749375e-06, "loss": 0.27269166707992554, "num_input_tokens_seen": 4715610, "step": 1666, "train_runtime": 5526.7525, "train_tokens_per_second": 853.233 }, { "epoch": 0.13395206814118404, "grad_norm": 2.6733028888702393, "learning_rate": 3.3483709273182964e-06, "loss": 0.20389196276664734, "num_input_tokens_seen": 4718486, "step": 1667, "train_runtime": 5529.383, "train_tokens_per_second": 853.348 }, { "epoch": 0.13403242331103477, "grad_norm": 2.9722750186920166, "learning_rate": 3.3383458646616545e-06, "loss": 0.28279006481170654, "num_input_tokens_seen": 4721206, "step": 1668, "train_runtime": 5531.9204, "train_tokens_per_second": 853.448 }, { "epoch": 0.13411277848088551, "grad_norm": 3.168781042098999, "learning_rate": 3.328320802005013e-06, "loss": 0.21676430106163025, "num_input_tokens_seen": 4725592, "step": 1669, "train_runtime": 5534.9054, "train_tokens_per_second": 853.78 }, { "epoch": 0.13419313365073626, "grad_norm": 3.6228220462799072, "learning_rate": 3.318295739348371e-06, "loss": 0.33118686079978943, "num_input_tokens_seen": 4727774, "step": 1670, "train_runtime": 5537.2872, "train_tokens_per_second": 853.807 }, { "epoch": 0.13427348882058698, "grad_norm": 2.861109495162964, "learning_rate": 3.3082706766917295e-06, "loss": 0.28560683131217957, "num_input_tokens_seen": 4732266, "step": 1671, "train_runtime": 5540.2899, "train_tokens_per_second": 854.155 }, { "epoch": 0.13435384399043773, "grad_norm": 2.9555437564849854, "learning_rate": 3.2982456140350885e-06, "loss": 0.18611425161361694, "num_input_tokens_seen": 4734460, "step": 1672, "train_runtime": 5542.6913, "train_tokens_per_second": 854.181 }, { "epoch": 0.13443419916028848, "grad_norm": 2.838218927383423, "learning_rate": 3.2882205513784465e-06, "loss": 0.20387740433216095, "num_input_tokens_seen": 4736278, "step": 1673, "train_runtime": 5545.0263, "train_tokens_per_second": 854.149 }, { "epoch": 0.1345145543301392, "grad_norm": 3.360383987426758, "learning_rate": 3.278195488721805e-06, "loss": 0.2792057693004608, "num_input_tokens_seen": 4738546, "step": 1674, "train_runtime": 5547.4388, "train_tokens_per_second": 854.186 }, { "epoch": 0.13459490949998995, "grad_norm": 3.0020039081573486, "learning_rate": 3.268170426065163e-06, "loss": 0.3126697540283203, "num_input_tokens_seen": 4740414, "step": 1675, "train_runtime": 5549.7478, "train_tokens_per_second": 854.167 }, { "epoch": 0.1346752646698407, "grad_norm": 3.2556471824645996, "learning_rate": 3.2581453634085216e-06, "loss": 0.28695642948150635, "num_input_tokens_seen": 4742572, "step": 1676, "train_runtime": 5552.1785, "train_tokens_per_second": 854.182 }, { "epoch": 0.13475561983969145, "grad_norm": 3.175236701965332, "learning_rate": 3.24812030075188e-06, "loss": 0.29978281259536743, "num_input_tokens_seen": 4747054, "step": 1677, "train_runtime": 5555.1406, "train_tokens_per_second": 854.534 }, { "epoch": 0.13483597500954217, "grad_norm": 3.0875840187072754, "learning_rate": 3.2380952380952385e-06, "loss": 0.3312967121601105, "num_input_tokens_seen": 4748990, "step": 1678, "train_runtime": 5557.5399, "train_tokens_per_second": 854.513 }, { "epoch": 0.13491633017939292, "grad_norm": 2.8591248989105225, "learning_rate": 3.2280701754385966e-06, "loss": 0.24268457293510437, "num_input_tokens_seen": 4751712, "step": 1679, "train_runtime": 5560.0895, "train_tokens_per_second": 854.611 }, { "epoch": 0.13499668534924367, "grad_norm": 2.7852845191955566, "learning_rate": 3.218045112781955e-06, "loss": 0.3064137399196625, "num_input_tokens_seen": 4753928, "step": 1680, "train_runtime": 5562.5156, "train_tokens_per_second": 854.636 }, { "epoch": 0.1350770405190944, "grad_norm": 3.3103504180908203, "learning_rate": 3.2080200501253136e-06, "loss": 0.32759934663772583, "num_input_tokens_seen": 4756108, "step": 1681, "train_runtime": 5576.3851, "train_tokens_per_second": 852.902 }, { "epoch": 0.13515739568894514, "grad_norm": 2.993309259414673, "learning_rate": 3.197994987468672e-06, "loss": 0.2757419943809509, "num_input_tokens_seen": 4759590, "step": 1682, "train_runtime": 5579.0999, "train_tokens_per_second": 853.111 }, { "epoch": 0.13523775085879589, "grad_norm": 3.4947266578674316, "learning_rate": 3.1879699248120305e-06, "loss": 0.3096398711204529, "num_input_tokens_seen": 4761258, "step": 1683, "train_runtime": 5581.4184, "train_tokens_per_second": 853.055 }, { "epoch": 0.1353181060286466, "grad_norm": 3.5111193656921387, "learning_rate": 3.1779448621553886e-06, "loss": 0.30844801664352417, "num_input_tokens_seen": 4764860, "step": 1684, "train_runtime": 5584.2112, "train_tokens_per_second": 853.274 }, { "epoch": 0.13539846119849736, "grad_norm": 2.764714241027832, "learning_rate": 3.167919799498747e-06, "loss": 0.2784709632396698, "num_input_tokens_seen": 4767366, "step": 1685, "train_runtime": 5586.7406, "train_tokens_per_second": 853.336 }, { "epoch": 0.1354788163683481, "grad_norm": 2.7272424697875977, "learning_rate": 3.157894736842105e-06, "loss": 0.26437634229660034, "num_input_tokens_seen": 4769486, "step": 1686, "train_runtime": 5589.1822, "train_tokens_per_second": 853.342 }, { "epoch": 0.13555917153819885, "grad_norm": 3.261348247528076, "learning_rate": 3.147869674185464e-06, "loss": 0.31559816002845764, "num_input_tokens_seen": 4772472, "step": 1687, "train_runtime": 5591.7755, "train_tokens_per_second": 853.481 }, { "epoch": 0.13563952670804957, "grad_norm": 2.8197181224823, "learning_rate": 3.1378446115288226e-06, "loss": 0.2338036596775055, "num_input_tokens_seen": 4774460, "step": 1688, "train_runtime": 5594.1821, "train_tokens_per_second": 853.469 }, { "epoch": 0.13571988187790032, "grad_norm": 3.3962323665618896, "learning_rate": 3.1278195488721806e-06, "loss": 0.30098965764045715, "num_input_tokens_seen": 4777100, "step": 1689, "train_runtime": 5596.7339, "train_tokens_per_second": 853.551 }, { "epoch": 0.13580023704775107, "grad_norm": 3.3569490909576416, "learning_rate": 3.117794486215539e-06, "loss": 0.3160550892353058, "num_input_tokens_seen": 4778940, "step": 1690, "train_runtime": 5599.1007, "train_tokens_per_second": 853.519 }, { "epoch": 0.1358805922176018, "grad_norm": 3.248988389968872, "learning_rate": 3.107769423558897e-06, "loss": 0.3484092950820923, "num_input_tokens_seen": 4781116, "step": 1691, "train_runtime": 5601.6019, "train_tokens_per_second": 853.527 }, { "epoch": 0.13596094738745254, "grad_norm": 2.6077263355255127, "learning_rate": 3.097744360902256e-06, "loss": 0.24307739734649658, "num_input_tokens_seen": 4785740, "step": 1692, "train_runtime": 5604.6857, "train_tokens_per_second": 853.882 }, { "epoch": 0.1360413025573033, "grad_norm": 3.6286673545837402, "learning_rate": 3.0877192982456146e-06, "loss": 0.3503590524196625, "num_input_tokens_seen": 4787890, "step": 1693, "train_runtime": 5607.1016, "train_tokens_per_second": 853.897 }, { "epoch": 0.136121657727154, "grad_norm": 3.190042018890381, "learning_rate": 3.0776942355889726e-06, "loss": 0.243970587849617, "num_input_tokens_seen": 4789880, "step": 1694, "train_runtime": 5609.4952, "train_tokens_per_second": 853.888 }, { "epoch": 0.13620201289700476, "grad_norm": 3.1295032501220703, "learning_rate": 3.067669172932331e-06, "loss": 0.22541047632694244, "num_input_tokens_seen": 4795884, "step": 1695, "train_runtime": 5612.8518, "train_tokens_per_second": 854.447 }, { "epoch": 0.1362823680668555, "grad_norm": 2.7716140747070312, "learning_rate": 3.057644110275689e-06, "loss": 0.2466445416212082, "num_input_tokens_seen": 4797810, "step": 1696, "train_runtime": 5615.2444, "train_tokens_per_second": 854.426 }, { "epoch": 0.13636272323670623, "grad_norm": 2.531359910964966, "learning_rate": 3.047619047619048e-06, "loss": 0.26966169476509094, "num_input_tokens_seen": 4800126, "step": 1697, "train_runtime": 5617.7142, "train_tokens_per_second": 854.462 }, { "epoch": 0.13644307840655698, "grad_norm": 3.0783119201660156, "learning_rate": 3.0375939849624066e-06, "loss": 0.24386481940746307, "num_input_tokens_seen": 4802108, "step": 1698, "train_runtime": 5620.1255, "train_tokens_per_second": 854.449 }, { "epoch": 0.13652343357640773, "grad_norm": 2.9699504375457764, "learning_rate": 3.0275689223057647e-06, "loss": 0.30408671498298645, "num_input_tokens_seen": 4804608, "step": 1699, "train_runtime": 5622.6256, "train_tokens_per_second": 854.513 }, { "epoch": 0.13660378874625848, "grad_norm": 3.2536258697509766, "learning_rate": 3.017543859649123e-06, "loss": 0.3324778079986572, "num_input_tokens_seen": 4806768, "step": 1700, "train_runtime": 5625.0762, "train_tokens_per_second": 854.525 }, { "epoch": 0.1366841439161092, "grad_norm": 2.673616886138916, "learning_rate": 3.007518796992481e-06, "loss": 0.2577211260795593, "num_input_tokens_seen": 4810836, "step": 1701, "train_runtime": 5627.962, "train_tokens_per_second": 854.81 }, { "epoch": 0.13676449908595995, "grad_norm": 3.0187020301818848, "learning_rate": 2.9974937343358397e-06, "loss": 0.25227558612823486, "num_input_tokens_seen": 4812810, "step": 1702, "train_runtime": 5630.3315, "train_tokens_per_second": 854.8 }, { "epoch": 0.1368448542558107, "grad_norm": 2.976358413696289, "learning_rate": 2.9874686716791986e-06, "loss": 0.31490010023117065, "num_input_tokens_seen": 4815168, "step": 1703, "train_runtime": 5632.8116, "train_tokens_per_second": 854.843 }, { "epoch": 0.13692520942566141, "grad_norm": 2.686866521835327, "learning_rate": 2.9774436090225567e-06, "loss": 0.24525614082813263, "num_input_tokens_seen": 4817380, "step": 1704, "train_runtime": 5635.2116, "train_tokens_per_second": 854.871 }, { "epoch": 0.13700556459551216, "grad_norm": 4.0214619636535645, "learning_rate": 2.967418546365915e-06, "loss": 0.351442813873291, "num_input_tokens_seen": 4820524, "step": 1705, "train_runtime": 5637.8653, "train_tokens_per_second": 855.026 }, { "epoch": 0.1370859197653629, "grad_norm": 3.166612386703491, "learning_rate": 2.9573934837092732e-06, "loss": 0.4158138036727905, "num_input_tokens_seen": 4823384, "step": 1706, "train_runtime": 5640.4999, "train_tokens_per_second": 855.134 }, { "epoch": 0.13716627493521363, "grad_norm": 2.8576693534851074, "learning_rate": 2.9473684210526317e-06, "loss": 0.3365335166454315, "num_input_tokens_seen": 4827694, "step": 1707, "train_runtime": 5643.4813, "train_tokens_per_second": 855.446 }, { "epoch": 0.13724663010506438, "grad_norm": 2.8427538871765137, "learning_rate": 2.93734335839599e-06, "loss": 0.24599894881248474, "num_input_tokens_seen": 4829820, "step": 1708, "train_runtime": 5645.9549, "train_tokens_per_second": 855.448 }, { "epoch": 0.13732698527491513, "grad_norm": 4.192246913909912, "learning_rate": 2.9273182957393487e-06, "loss": 0.2771722674369812, "num_input_tokens_seen": 4835648, "step": 1709, "train_runtime": 5649.5291, "train_tokens_per_second": 855.938 }, { "epoch": 0.13740734044476585, "grad_norm": 3.05594539642334, "learning_rate": 2.9172932330827068e-06, "loss": 0.2552582323551178, "num_input_tokens_seen": 4838116, "step": 1710, "train_runtime": 5652.0725, "train_tokens_per_second": 855.99 }, { "epoch": 0.1374876956146166, "grad_norm": 2.9709534645080566, "learning_rate": 2.9072681704260652e-06, "loss": 0.23290160298347473, "num_input_tokens_seen": 4840374, "step": 1711, "train_runtime": 5666.1695, "train_tokens_per_second": 854.259 }, { "epoch": 0.13756805078446735, "grad_norm": 3.022853374481201, "learning_rate": 2.8972431077694237e-06, "loss": 0.20388667285442352, "num_input_tokens_seen": 4843582, "step": 1712, "train_runtime": 5669.0016, "train_tokens_per_second": 854.398 }, { "epoch": 0.1376484059543181, "grad_norm": 2.9590768814086914, "learning_rate": 2.8872180451127822e-06, "loss": 0.43798813223838806, "num_input_tokens_seen": 4847098, "step": 1713, "train_runtime": 5671.7453, "train_tokens_per_second": 854.604 }, { "epoch": 0.13772876112416882, "grad_norm": 2.6649250984191895, "learning_rate": 2.8771929824561407e-06, "loss": 0.24580538272857666, "num_input_tokens_seen": 4849606, "step": 1714, "train_runtime": 5674.2445, "train_tokens_per_second": 854.67 }, { "epoch": 0.13780911629401957, "grad_norm": 2.708113670349121, "learning_rate": 2.8671679197994988e-06, "loss": 0.2604023218154907, "num_input_tokens_seen": 4851394, "step": 1715, "train_runtime": 5676.6282, "train_tokens_per_second": 854.626 }, { "epoch": 0.13788947146387032, "grad_norm": 2.6693947315216064, "learning_rate": 2.8571428571428573e-06, "loss": 0.20540615916252136, "num_input_tokens_seen": 4854832, "step": 1716, "train_runtime": 5679.3906, "train_tokens_per_second": 854.816 }, { "epoch": 0.13796982663372104, "grad_norm": 3.218764305114746, "learning_rate": 2.8471177944862153e-06, "loss": 0.334343284368515, "num_input_tokens_seen": 4857964, "step": 1717, "train_runtime": 5682.0251, "train_tokens_per_second": 854.971 }, { "epoch": 0.1380501818035718, "grad_norm": 2.944223403930664, "learning_rate": 2.8370927318295742e-06, "loss": 0.2597845494747162, "num_input_tokens_seen": 4859788, "step": 1718, "train_runtime": 5684.3737, "train_tokens_per_second": 854.938 }, { "epoch": 0.13813053697342254, "grad_norm": 2.8793723583221436, "learning_rate": 2.8270676691729327e-06, "loss": 0.26263901591300964, "num_input_tokens_seen": 4863038, "step": 1719, "train_runtime": 5687.1218, "train_tokens_per_second": 855.097 }, { "epoch": 0.13821089214327326, "grad_norm": 2.8462417125701904, "learning_rate": 2.8170426065162908e-06, "loss": 0.2839650809764862, "num_input_tokens_seen": 4868122, "step": 1720, "train_runtime": 5690.2834, "train_tokens_per_second": 855.515 }, { "epoch": 0.138291247313124, "grad_norm": 3.2740869522094727, "learning_rate": 2.8070175438596493e-06, "loss": 0.261339396238327, "num_input_tokens_seen": 4870898, "step": 1721, "train_runtime": 5692.854, "train_tokens_per_second": 855.616 }, { "epoch": 0.13837160248297475, "grad_norm": 2.389802932739258, "learning_rate": 2.7969924812030073e-06, "loss": 0.24968945980072021, "num_input_tokens_seen": 4873324, "step": 1722, "train_runtime": 5695.3614, "train_tokens_per_second": 855.665 }, { "epoch": 0.1384519576528255, "grad_norm": 2.8070433139801025, "learning_rate": 2.7869674185463662e-06, "loss": 0.2357427179813385, "num_input_tokens_seen": 4876296, "step": 1723, "train_runtime": 5697.9583, "train_tokens_per_second": 855.797 }, { "epoch": 0.13853231282267622, "grad_norm": 2.787797689437866, "learning_rate": 2.7769423558897247e-06, "loss": 0.2982591390609741, "num_input_tokens_seen": 4879174, "step": 1724, "train_runtime": 5700.4845, "train_tokens_per_second": 855.923 }, { "epoch": 0.13861266799252697, "grad_norm": 2.3699374198913574, "learning_rate": 2.766917293233083e-06, "loss": 0.2170788198709488, "num_input_tokens_seen": 4881500, "step": 1725, "train_runtime": 5702.9477, "train_tokens_per_second": 855.961 }, { "epoch": 0.13869302316237772, "grad_norm": 2.2282285690307617, "learning_rate": 2.7568922305764413e-06, "loss": 0.1959061622619629, "num_input_tokens_seen": 4883856, "step": 1726, "train_runtime": 5705.3946, "train_tokens_per_second": 856.007 }, { "epoch": 0.13877337833222844, "grad_norm": 3.293076515197754, "learning_rate": 2.7468671679197994e-06, "loss": 0.2725188434123993, "num_input_tokens_seen": 4886358, "step": 1727, "train_runtime": 5707.8635, "train_tokens_per_second": 856.075 }, { "epoch": 0.1388537335020792, "grad_norm": 2.8111913204193115, "learning_rate": 2.7368421052631583e-06, "loss": 0.25002121925354004, "num_input_tokens_seen": 4888002, "step": 1728, "train_runtime": 5710.1133, "train_tokens_per_second": 856.025 }, { "epoch": 0.13893408867192994, "grad_norm": 2.880429267883301, "learning_rate": 2.7268170426065167e-06, "loss": 0.26038026809692383, "num_input_tokens_seen": 4889516, "step": 1729, "train_runtime": 5712.3206, "train_tokens_per_second": 855.96 }, { "epoch": 0.13901444384178066, "grad_norm": 3.2697722911834717, "learning_rate": 2.716791979949875e-06, "loss": 0.2712022364139557, "num_input_tokens_seen": 4892952, "step": 1730, "train_runtime": 5715.0361, "train_tokens_per_second": 856.154 }, { "epoch": 0.1390947990116314, "grad_norm": 3.6746299266815186, "learning_rate": 2.7067669172932333e-06, "loss": 0.3591930568218231, "num_input_tokens_seen": 4895872, "step": 1731, "train_runtime": 5717.6025, "train_tokens_per_second": 856.281 }, { "epoch": 0.13917515418148216, "grad_norm": 2.7356319427490234, "learning_rate": 2.6967418546365914e-06, "loss": 0.25933602452278137, "num_input_tokens_seen": 4898122, "step": 1732, "train_runtime": 5720.0138, "train_tokens_per_second": 856.313 }, { "epoch": 0.13925550935133288, "grad_norm": 3.2641971111297607, "learning_rate": 2.6867167919799503e-06, "loss": 0.32112735509872437, "num_input_tokens_seen": 4901306, "step": 1733, "train_runtime": 5722.7471, "train_tokens_per_second": 856.46 }, { "epoch": 0.13933586452118363, "grad_norm": 3.3800313472747803, "learning_rate": 2.6766917293233088e-06, "loss": 0.24829038977622986, "num_input_tokens_seen": 4903998, "step": 1734, "train_runtime": 5725.3557, "train_tokens_per_second": 856.54 }, { "epoch": 0.13941621969103438, "grad_norm": 3.217944383621216, "learning_rate": 2.666666666666667e-06, "loss": 0.32297271490097046, "num_input_tokens_seen": 4906694, "step": 1735, "train_runtime": 5727.8987, "train_tokens_per_second": 856.631 }, { "epoch": 0.13949657486088513, "grad_norm": 3.110604763031006, "learning_rate": 2.6566416040100253e-06, "loss": 0.2506431043148041, "num_input_tokens_seen": 4912368, "step": 1736, "train_runtime": 5731.2862, "train_tokens_per_second": 857.114 }, { "epoch": 0.13957693003073585, "grad_norm": 2.7314085960388184, "learning_rate": 2.6466165413533834e-06, "loss": 0.22235555946826935, "num_input_tokens_seen": 4915770, "step": 1737, "train_runtime": 5734.0772, "train_tokens_per_second": 857.291 }, { "epoch": 0.1396572852005866, "grad_norm": 3.047297477722168, "learning_rate": 2.6365914786967423e-06, "loss": 0.27471205592155457, "num_input_tokens_seen": 4918184, "step": 1738, "train_runtime": 5736.5756, "train_tokens_per_second": 857.338 }, { "epoch": 0.13973764037043734, "grad_norm": 3.369499921798706, "learning_rate": 2.6265664160401004e-06, "loss": 0.2966940701007843, "num_input_tokens_seen": 4920818, "step": 1739, "train_runtime": 5739.1045, "train_tokens_per_second": 857.419 }, { "epoch": 0.13981799554028806, "grad_norm": 3.0819008350372314, "learning_rate": 2.616541353383459e-06, "loss": 0.3102680444717407, "num_input_tokens_seen": 4923154, "step": 1740, "train_runtime": 5741.6111, "train_tokens_per_second": 857.452 }, { "epoch": 0.1398983507101388, "grad_norm": 3.159250259399414, "learning_rate": 2.606516290726817e-06, "loss": 0.41482383012771606, "num_input_tokens_seen": 4926194, "step": 1741, "train_runtime": 5755.9003, "train_tokens_per_second": 855.851 }, { "epoch": 0.13997870587998956, "grad_norm": 2.979243516921997, "learning_rate": 2.5964912280701754e-06, "loss": 0.3266957402229309, "num_input_tokens_seen": 4928786, "step": 1742, "train_runtime": 5758.4089, "train_tokens_per_second": 855.928 }, { "epoch": 0.14005906104984028, "grad_norm": 3.109147548675537, "learning_rate": 2.5864661654135343e-06, "loss": 0.23772843182086945, "num_input_tokens_seen": 4931322, "step": 1743, "train_runtime": 5760.938, "train_tokens_per_second": 855.993 }, { "epoch": 0.14013941621969103, "grad_norm": 2.6563000679016113, "learning_rate": 2.5764411027568924e-06, "loss": 0.25501877069473267, "num_input_tokens_seen": 4933076, "step": 1744, "train_runtime": 5763.3275, "train_tokens_per_second": 855.942 }, { "epoch": 0.14021977138954178, "grad_norm": 2.468778133392334, "learning_rate": 2.566416040100251e-06, "loss": 0.23249754309654236, "num_input_tokens_seen": 4936058, "step": 1745, "train_runtime": 5765.9502, "train_tokens_per_second": 856.07 }, { "epoch": 0.1403001265593925, "grad_norm": 3.2199158668518066, "learning_rate": 2.556390977443609e-06, "loss": 0.3400820195674896, "num_input_tokens_seen": 4939552, "step": 1746, "train_runtime": 5768.7139, "train_tokens_per_second": 856.266 }, { "epoch": 0.14038048172924325, "grad_norm": 3.546126365661621, "learning_rate": 2.5463659147869674e-06, "loss": 0.33273935317993164, "num_input_tokens_seen": 4941606, "step": 1747, "train_runtime": 5771.087, "train_tokens_per_second": 856.27 }, { "epoch": 0.140460836899094, "grad_norm": 3.551957607269287, "learning_rate": 2.5363408521303263e-06, "loss": 0.24110406637191772, "num_input_tokens_seen": 4944022, "step": 1748, "train_runtime": 5773.5698, "train_tokens_per_second": 856.32 }, { "epoch": 0.14054119206894475, "grad_norm": 2.933347225189209, "learning_rate": 2.5263157894736844e-06, "loss": 0.3032664656639099, "num_input_tokens_seen": 4945836, "step": 1749, "train_runtime": 5775.8844, "train_tokens_per_second": 856.291 }, { "epoch": 0.14062154723879547, "grad_norm": 3.9250261783599854, "learning_rate": 2.516290726817043e-06, "loss": 0.31265705823898315, "num_input_tokens_seen": 4947814, "step": 1750, "train_runtime": 5778.42, "train_tokens_per_second": 856.257 }, { "epoch": 0.14070190240864622, "grad_norm": 2.5885422229766846, "learning_rate": 2.506265664160401e-06, "loss": 0.1828916072845459, "num_input_tokens_seen": 4950214, "step": 1751, "train_runtime": 5781.0411, "train_tokens_per_second": 856.284 }, { "epoch": 0.14078225757849697, "grad_norm": 2.916311502456665, "learning_rate": 2.4962406015037594e-06, "loss": 0.23075290024280548, "num_input_tokens_seen": 4952226, "step": 1752, "train_runtime": 5783.4074, "train_tokens_per_second": 856.282 }, { "epoch": 0.1408626127483477, "grad_norm": 2.677844762802124, "learning_rate": 2.486215538847118e-06, "loss": 0.21280820667743683, "num_input_tokens_seen": 4954334, "step": 1753, "train_runtime": 5785.7679, "train_tokens_per_second": 856.297 }, { "epoch": 0.14094296791819844, "grad_norm": 2.3096137046813965, "learning_rate": 2.4761904761904764e-06, "loss": 0.21944457292556763, "num_input_tokens_seen": 4956184, "step": 1754, "train_runtime": 5788.1154, "train_tokens_per_second": 856.269 }, { "epoch": 0.14102332308804919, "grad_norm": 3.290259838104248, "learning_rate": 2.466165413533835e-06, "loss": 0.27066901326179504, "num_input_tokens_seen": 4957990, "step": 1755, "train_runtime": 5790.4136, "train_tokens_per_second": 856.241 }, { "epoch": 0.1411036782578999, "grad_norm": 3.2820167541503906, "learning_rate": 2.456140350877193e-06, "loss": 0.24739360809326172, "num_input_tokens_seen": 4960542, "step": 1756, "train_runtime": 5792.9132, "train_tokens_per_second": 856.312 }, { "epoch": 0.14118403342775065, "grad_norm": 2.9319188594818115, "learning_rate": 2.4461152882205514e-06, "loss": 0.22173424065113068, "num_input_tokens_seen": 4963364, "step": 1757, "train_runtime": 5795.5777, "train_tokens_per_second": 856.405 }, { "epoch": 0.1412643885976014, "grad_norm": 2.9257237911224365, "learning_rate": 2.43609022556391e-06, "loss": 0.28933677077293396, "num_input_tokens_seen": 4966100, "step": 1758, "train_runtime": 5798.1506, "train_tokens_per_second": 856.497 }, { "epoch": 0.14134474376745215, "grad_norm": 2.5945324897766113, "learning_rate": 2.4260651629072684e-06, "loss": 0.18981416523456573, "num_input_tokens_seen": 4967824, "step": 1759, "train_runtime": 5800.4326, "train_tokens_per_second": 856.457 }, { "epoch": 0.14142509893730287, "grad_norm": 3.077648162841797, "learning_rate": 2.416040100250627e-06, "loss": 0.2695406377315521, "num_input_tokens_seen": 4970768, "step": 1760, "train_runtime": 5803.0487, "train_tokens_per_second": 856.579 }, { "epoch": 0.14150545410715362, "grad_norm": 2.457252025604248, "learning_rate": 2.406015037593985e-06, "loss": 0.35060200095176697, "num_input_tokens_seen": 4975176, "step": 1761, "train_runtime": 5805.9951, "train_tokens_per_second": 856.903 }, { "epoch": 0.14158580927700437, "grad_norm": 2.959620952606201, "learning_rate": 2.3959899749373435e-06, "loss": 0.21713030338287354, "num_input_tokens_seen": 4976632, "step": 1762, "train_runtime": 5808.2113, "train_tokens_per_second": 856.827 }, { "epoch": 0.1416661644468551, "grad_norm": 2.6666371822357178, "learning_rate": 2.385964912280702e-06, "loss": 0.2876514494419098, "num_input_tokens_seen": 4980000, "step": 1763, "train_runtime": 5810.9356, "train_tokens_per_second": 857.005 }, { "epoch": 0.14174651961670584, "grad_norm": 2.9948835372924805, "learning_rate": 2.3759398496240604e-06, "loss": 0.2741466462612152, "num_input_tokens_seen": 4983180, "step": 1764, "train_runtime": 5813.5676, "train_tokens_per_second": 857.164 }, { "epoch": 0.1418268747865566, "grad_norm": 2.6717231273651123, "learning_rate": 2.365914786967419e-06, "loss": 0.26970410346984863, "num_input_tokens_seen": 4986272, "step": 1765, "train_runtime": 5816.2097, "train_tokens_per_second": 857.306 }, { "epoch": 0.1419072299564073, "grad_norm": 3.3086459636688232, "learning_rate": 2.355889724310777e-06, "loss": 0.285335510969162, "num_input_tokens_seen": 4988050, "step": 1766, "train_runtime": 5818.5285, "train_tokens_per_second": 857.27 }, { "epoch": 0.14198758512625806, "grad_norm": 2.8983347415924072, "learning_rate": 2.3458646616541355e-06, "loss": 0.2980395555496216, "num_input_tokens_seen": 4990548, "step": 1767, "train_runtime": 5821.0324, "train_tokens_per_second": 857.33 }, { "epoch": 0.1420679402961088, "grad_norm": 3.099055290222168, "learning_rate": 2.335839598997494e-06, "loss": 0.28661343455314636, "num_input_tokens_seen": 4993588, "step": 1768, "train_runtime": 5823.6383, "train_tokens_per_second": 857.469 }, { "epoch": 0.14214829546595953, "grad_norm": 2.8428077697753906, "learning_rate": 2.3258145363408524e-06, "loss": 0.2536507248878479, "num_input_tokens_seen": 4995354, "step": 1769, "train_runtime": 5825.9952, "train_tokens_per_second": 857.425 }, { "epoch": 0.14222865063581028, "grad_norm": 3.2767207622528076, "learning_rate": 2.3157894736842105e-06, "loss": 0.30807361006736755, "num_input_tokens_seen": 4998470, "step": 1770, "train_runtime": 5828.6569, "train_tokens_per_second": 857.568 }, { "epoch": 0.14230900580566103, "grad_norm": 3.228936195373535, "learning_rate": 2.305764411027569e-06, "loss": 0.26145169138908386, "num_input_tokens_seen": 5001068, "step": 1771, "train_runtime": 5842.8111, "train_tokens_per_second": 855.935 }, { "epoch": 0.14238936097551178, "grad_norm": 3.473923444747925, "learning_rate": 2.2957393483709275e-06, "loss": 0.3374575972557068, "num_input_tokens_seen": 5003002, "step": 1772, "train_runtime": 5845.174, "train_tokens_per_second": 855.92 }, { "epoch": 0.1424697161453625, "grad_norm": 3.1981804370880127, "learning_rate": 2.285714285714286e-06, "loss": 0.2892587184906006, "num_input_tokens_seen": 5004776, "step": 1773, "train_runtime": 5847.4921, "train_tokens_per_second": 855.884 }, { "epoch": 0.14255007131521324, "grad_norm": 2.845667600631714, "learning_rate": 2.2756892230576445e-06, "loss": 0.2820015549659729, "num_input_tokens_seen": 5006770, "step": 1774, "train_runtime": 5849.8379, "train_tokens_per_second": 855.882 }, { "epoch": 0.142630426485064, "grad_norm": 2.3059911727905273, "learning_rate": 2.2656641604010025e-06, "loss": 0.22226737439632416, "num_input_tokens_seen": 5008736, "step": 1775, "train_runtime": 5852.2085, "train_tokens_per_second": 855.871 }, { "epoch": 0.14271078165491471, "grad_norm": 2.2758615016937256, "learning_rate": 2.255639097744361e-06, "loss": 0.2226129025220871, "num_input_tokens_seen": 5010812, "step": 1776, "train_runtime": 5854.6218, "train_tokens_per_second": 855.873 }, { "epoch": 0.14279113682476546, "grad_norm": 3.078495740890503, "learning_rate": 2.2456140350877195e-06, "loss": 0.2911476194858551, "num_input_tokens_seen": 5013546, "step": 1777, "train_runtime": 5857.1272, "train_tokens_per_second": 855.974 }, { "epoch": 0.1428714919946162, "grad_norm": 3.199249505996704, "learning_rate": 2.235588972431078e-06, "loss": 0.279721736907959, "num_input_tokens_seen": 5015932, "step": 1778, "train_runtime": 5859.5941, "train_tokens_per_second": 856.02 }, { "epoch": 0.14295184716446693, "grad_norm": 2.5613837242126465, "learning_rate": 2.2255639097744365e-06, "loss": 0.2372240424156189, "num_input_tokens_seen": 5018864, "step": 1779, "train_runtime": 5862.1992, "train_tokens_per_second": 856.14 }, { "epoch": 0.14303220233431768, "grad_norm": 3.3952341079711914, "learning_rate": 2.2155388471177945e-06, "loss": 0.3520214855670929, "num_input_tokens_seen": 5023806, "step": 1780, "train_runtime": 5865.4866, "train_tokens_per_second": 856.503 }, { "epoch": 0.14311255750416843, "grad_norm": 2.527592658996582, "learning_rate": 2.205513784461153e-06, "loss": 0.2609686255455017, "num_input_tokens_seen": 5028642, "step": 1781, "train_runtime": 5868.6106, "train_tokens_per_second": 856.871 }, { "epoch": 0.14319291267401915, "grad_norm": 2.6321048736572266, "learning_rate": 2.1954887218045115e-06, "loss": 0.21098420023918152, "num_input_tokens_seen": 5031104, "step": 1782, "train_runtime": 5871.0882, "train_tokens_per_second": 856.929 }, { "epoch": 0.1432732678438699, "grad_norm": 2.7917685508728027, "learning_rate": 2.1854636591478696e-06, "loss": 0.34251588582992554, "num_input_tokens_seen": 5034410, "step": 1783, "train_runtime": 5873.811, "train_tokens_per_second": 857.094 }, { "epoch": 0.14335362301372065, "grad_norm": 2.8712425231933594, "learning_rate": 2.1754385964912285e-06, "loss": 0.2683536410331726, "num_input_tokens_seen": 5036982, "step": 1784, "train_runtime": 5876.382, "train_tokens_per_second": 857.157 }, { "epoch": 0.1434339781835714, "grad_norm": 2.3242106437683105, "learning_rate": 2.1654135338345866e-06, "loss": 0.22497516870498657, "num_input_tokens_seen": 5041274, "step": 1785, "train_runtime": 5879.3009, "train_tokens_per_second": 857.461 }, { "epoch": 0.14351433335342212, "grad_norm": 2.5213968753814697, "learning_rate": 2.155388471177945e-06, "loss": 0.20407596230506897, "num_input_tokens_seen": 5045470, "step": 1786, "train_runtime": 5882.2159, "train_tokens_per_second": 857.75 }, { "epoch": 0.14359468852327287, "grad_norm": 2.6716811656951904, "learning_rate": 2.1453634085213035e-06, "loss": 0.20244643092155457, "num_input_tokens_seen": 5047518, "step": 1787, "train_runtime": 5884.6267, "train_tokens_per_second": 857.747 }, { "epoch": 0.14367504369312362, "grad_norm": 2.7488417625427246, "learning_rate": 2.1353383458646616e-06, "loss": 0.2526586055755615, "num_input_tokens_seen": 5050020, "step": 1788, "train_runtime": 5887.1855, "train_tokens_per_second": 857.799 }, { "epoch": 0.14375539886297434, "grad_norm": 3.080665111541748, "learning_rate": 2.1253132832080205e-06, "loss": 0.2783365845680237, "num_input_tokens_seen": 5054504, "step": 1789, "train_runtime": 5890.1902, "train_tokens_per_second": 858.122 }, { "epoch": 0.14383575403282509, "grad_norm": 3.3435699939727783, "learning_rate": 2.1152882205513786e-06, "loss": 0.29115796089172363, "num_input_tokens_seen": 5057294, "step": 1790, "train_runtime": 5892.8153, "train_tokens_per_second": 858.214 }, { "epoch": 0.14391610920267583, "grad_norm": 3.0493433475494385, "learning_rate": 2.105263157894737e-06, "loss": 0.3140782117843628, "num_input_tokens_seen": 5059974, "step": 1791, "train_runtime": 5895.3682, "train_tokens_per_second": 858.297 }, { "epoch": 0.14399646437252656, "grad_norm": 2.8110744953155518, "learning_rate": 2.0952380952380955e-06, "loss": 0.25942495465278625, "num_input_tokens_seen": 5062792, "step": 1792, "train_runtime": 5897.9399, "train_tokens_per_second": 858.4 }, { "epoch": 0.1440768195423773, "grad_norm": 3.364863157272339, "learning_rate": 2.0852130325814536e-06, "loss": 0.2991800904273987, "num_input_tokens_seen": 5064544, "step": 1793, "train_runtime": 5900.2563, "train_tokens_per_second": 858.36 }, { "epoch": 0.14415717471222805, "grad_norm": 3.3775980472564697, "learning_rate": 2.075187969924812e-06, "loss": 0.31868231296539307, "num_input_tokens_seen": 5066614, "step": 1794, "train_runtime": 5902.7312, "train_tokens_per_second": 858.351 }, { "epoch": 0.1442375298820788, "grad_norm": 3.5605974197387695, "learning_rate": 2.0651629072681706e-06, "loss": 0.27477923035621643, "num_input_tokens_seen": 5069384, "step": 1795, "train_runtime": 5905.2997, "train_tokens_per_second": 858.447 }, { "epoch": 0.14431788505192952, "grad_norm": 3.5581214427948, "learning_rate": 2.055137844611529e-06, "loss": 0.38893455266952515, "num_input_tokens_seen": 5071496, "step": 1796, "train_runtime": 5907.7153, "train_tokens_per_second": 858.453 }, { "epoch": 0.14439824022178027, "grad_norm": 2.8780107498168945, "learning_rate": 2.0451127819548876e-06, "loss": 0.24192000925540924, "num_input_tokens_seen": 5074104, "step": 1797, "train_runtime": 5910.2522, "train_tokens_per_second": 858.526 }, { "epoch": 0.14447859539163102, "grad_norm": 3.0316121578216553, "learning_rate": 2.0350877192982456e-06, "loss": 0.38636767864227295, "num_input_tokens_seen": 5076962, "step": 1798, "train_runtime": 5912.8925, "train_tokens_per_second": 858.626 }, { "epoch": 0.14455895056148174, "grad_norm": 2.8951122760772705, "learning_rate": 2.025062656641604e-06, "loss": 0.26132163405418396, "num_input_tokens_seen": 5079086, "step": 1799, "train_runtime": 5915.367, "train_tokens_per_second": 858.626 }, { "epoch": 0.1446393057313325, "grad_norm": 3.0799503326416016, "learning_rate": 2.0150375939849626e-06, "loss": 0.3290315270423889, "num_input_tokens_seen": 5080764, "step": 1800, "train_runtime": 5917.6431, "train_tokens_per_second": 858.579 }, { "epoch": 0.1446393057313325, "eval_loss": 1.094569206237793, "eval_runtime": 69.3529, "eval_samples_per_second": 14.332, "eval_steps_per_second": 7.166, "num_input_tokens_seen": 5080764, "step": 1800 }, { "epoch": 0.14471966090118324, "grad_norm": 3.234051465988159, "learning_rate": 2.0050125313283207e-06, "loss": 0.3210298418998718, "num_input_tokens_seen": 5083492, "step": 1801, "train_runtime": 6001.3515, "train_tokens_per_second": 847.058 }, { "epoch": 0.14480001607103396, "grad_norm": 2.805809259414673, "learning_rate": 1.9949874686716796e-06, "loss": 0.19775530695915222, "num_input_tokens_seen": 5087218, "step": 1802, "train_runtime": 6004.1827, "train_tokens_per_second": 847.279 }, { "epoch": 0.1448803712408847, "grad_norm": 2.187941789627075, "learning_rate": 1.9849624060150376e-06, "loss": 0.25018227100372314, "num_input_tokens_seen": 5092828, "step": 1803, "train_runtime": 6007.5914, "train_tokens_per_second": 847.732 }, { "epoch": 0.14496072641073546, "grad_norm": 3.373016119003296, "learning_rate": 1.974937343358396e-06, "loss": 0.3361077606678009, "num_input_tokens_seen": 5095000, "step": 1804, "train_runtime": 6010.1337, "train_tokens_per_second": 847.735 }, { "epoch": 0.14504108158058618, "grad_norm": 2.6214230060577393, "learning_rate": 1.9649122807017546e-06, "loss": 0.2352142632007599, "num_input_tokens_seen": 5096756, "step": 1805, "train_runtime": 6012.6087, "train_tokens_per_second": 847.678 }, { "epoch": 0.14512143675043693, "grad_norm": 3.2118759155273438, "learning_rate": 1.9548872180451127e-06, "loss": 0.31477999687194824, "num_input_tokens_seen": 5102154, "step": 1806, "train_runtime": 6016.02, "train_tokens_per_second": 848.095 }, { "epoch": 0.14520179192028768, "grad_norm": 2.5432589054107666, "learning_rate": 1.9448621553884716e-06, "loss": 0.2258012890815735, "num_input_tokens_seen": 5104608, "step": 1807, "train_runtime": 6018.5802, "train_tokens_per_second": 848.142 }, { "epoch": 0.14528214709013842, "grad_norm": 2.234830856323242, "learning_rate": 1.9348370927318297e-06, "loss": 0.19798319041728973, "num_input_tokens_seen": 5107754, "step": 1808, "train_runtime": 6021.216, "train_tokens_per_second": 848.293 }, { "epoch": 0.14536250225998915, "grad_norm": 3.272050380706787, "learning_rate": 1.924812030075188e-06, "loss": 0.32475775480270386, "num_input_tokens_seen": 5109126, "step": 1809, "train_runtime": 6023.4686, "train_tokens_per_second": 848.203 }, { "epoch": 0.1454428574298399, "grad_norm": 3.0805652141571045, "learning_rate": 1.9147869674185466e-06, "loss": 0.32856374979019165, "num_input_tokens_seen": 5112152, "step": 1810, "train_runtime": 6026.2374, "train_tokens_per_second": 848.316 }, { "epoch": 0.14552321259969064, "grad_norm": 3.1885104179382324, "learning_rate": 1.904761904761905e-06, "loss": 0.2722591757774353, "num_input_tokens_seen": 5115274, "step": 1811, "train_runtime": 6028.9503, "train_tokens_per_second": 848.452 }, { "epoch": 0.14560356776954136, "grad_norm": 2.7088003158569336, "learning_rate": 1.8947368421052634e-06, "loss": 0.20962053537368774, "num_input_tokens_seen": 5117010, "step": 1812, "train_runtime": 6031.2571, "train_tokens_per_second": 848.415 }, { "epoch": 0.1456839229393921, "grad_norm": 2.511005401611328, "learning_rate": 1.8847117794486217e-06, "loss": 0.28807705640792847, "num_input_tokens_seen": 5119524, "step": 1813, "train_runtime": 6033.7931, "train_tokens_per_second": 848.475 }, { "epoch": 0.14576427810924286, "grad_norm": 2.99678635597229, "learning_rate": 1.8746867167919802e-06, "loss": 0.3517390787601471, "num_input_tokens_seen": 5121770, "step": 1814, "train_runtime": 6036.2452, "train_tokens_per_second": 848.503 }, { "epoch": 0.14584463327909358, "grad_norm": 3.514350414276123, "learning_rate": 1.8646616541353384e-06, "loss": 0.3392871022224426, "num_input_tokens_seen": 5125308, "step": 1815, "train_runtime": 6038.9944, "train_tokens_per_second": 848.702 }, { "epoch": 0.14592498844894433, "grad_norm": 2.4640581607818604, "learning_rate": 1.8546365914786967e-06, "loss": 0.2564847767353058, "num_input_tokens_seen": 5128704, "step": 1816, "train_runtime": 6041.7263, "train_tokens_per_second": 848.881 }, { "epoch": 0.14600534361879508, "grad_norm": 2.95751953125, "learning_rate": 1.8446115288220554e-06, "loss": 0.2142380326986313, "num_input_tokens_seen": 5130768, "step": 1817, "train_runtime": 6044.1371, "train_tokens_per_second": 848.883 }, { "epoch": 0.1460856987886458, "grad_norm": 2.522202968597412, "learning_rate": 1.8345864661654137e-06, "loss": 0.27815118432044983, "num_input_tokens_seen": 5133206, "step": 1818, "train_runtime": 6046.6136, "train_tokens_per_second": 848.939 }, { "epoch": 0.14616605395849655, "grad_norm": 3.7731821537017822, "learning_rate": 1.8245614035087722e-06, "loss": 0.29861539602279663, "num_input_tokens_seen": 5135314, "step": 1819, "train_runtime": 6049.0676, "train_tokens_per_second": 848.943 }, { "epoch": 0.1462464091283473, "grad_norm": 2.3285205364227295, "learning_rate": 1.8145363408521305e-06, "loss": 0.20536713302135468, "num_input_tokens_seen": 5138202, "step": 1820, "train_runtime": 6051.7196, "train_tokens_per_second": 849.048 }, { "epoch": 0.14632676429819805, "grad_norm": 2.7543375492095947, "learning_rate": 1.8045112781954887e-06, "loss": 0.22493687272071838, "num_input_tokens_seen": 5139756, "step": 1821, "train_runtime": 6054.0425, "train_tokens_per_second": 848.979 }, { "epoch": 0.14640711946804877, "grad_norm": 2.397141218185425, "learning_rate": 1.7944862155388474e-06, "loss": 0.1969124972820282, "num_input_tokens_seen": 5141150, "step": 1822, "train_runtime": 6056.3349, "train_tokens_per_second": 848.888 }, { "epoch": 0.14648747463789952, "grad_norm": 2.5878233909606934, "learning_rate": 1.7844611528822057e-06, "loss": 0.28525903820991516, "num_input_tokens_seen": 5143180, "step": 1823, "train_runtime": 6058.8164, "train_tokens_per_second": 848.875 }, { "epoch": 0.14656782980775027, "grad_norm": 3.562067747116089, "learning_rate": 1.7744360902255642e-06, "loss": 0.24516353011131287, "num_input_tokens_seen": 5146412, "step": 1824, "train_runtime": 6061.495, "train_tokens_per_second": 849.033 }, { "epoch": 0.146648184977601, "grad_norm": 2.9113643169403076, "learning_rate": 1.7644110275689225e-06, "loss": 0.3307884931564331, "num_input_tokens_seen": 5149242, "step": 1825, "train_runtime": 6064.2177, "train_tokens_per_second": 849.119 }, { "epoch": 0.14672854014745174, "grad_norm": 2.887085199356079, "learning_rate": 1.7543859649122807e-06, "loss": 0.2549254596233368, "num_input_tokens_seen": 5152842, "step": 1826, "train_runtime": 6066.9576, "train_tokens_per_second": 849.329 }, { "epoch": 0.14680889531730248, "grad_norm": 2.711107015609741, "learning_rate": 1.7443609022556392e-06, "loss": 0.1816461980342865, "num_input_tokens_seen": 5154858, "step": 1827, "train_runtime": 6069.348, "train_tokens_per_second": 849.326 }, { "epoch": 0.1468892504871532, "grad_norm": 2.771200656890869, "learning_rate": 1.7343358395989975e-06, "loss": 0.2940555810928345, "num_input_tokens_seen": 5156284, "step": 1828, "train_runtime": 6071.5779, "train_tokens_per_second": 849.249 }, { "epoch": 0.14696960565700395, "grad_norm": 2.4687721729278564, "learning_rate": 1.7243107769423562e-06, "loss": 0.2302398383617401, "num_input_tokens_seen": 5159856, "step": 1829, "train_runtime": 6074.4339, "train_tokens_per_second": 849.438 }, { "epoch": 0.1470499608268547, "grad_norm": 3.4937784671783447, "learning_rate": 1.7142857142857145e-06, "loss": 0.31223994493484497, "num_input_tokens_seen": 5161446, "step": 1830, "train_runtime": 6076.725, "train_tokens_per_second": 849.38 }, { "epoch": 0.14713031599670545, "grad_norm": 3.526752233505249, "learning_rate": 1.7042606516290728e-06, "loss": 0.36472854018211365, "num_input_tokens_seen": 5164636, "step": 1831, "train_runtime": 6091.0458, "train_tokens_per_second": 847.906 }, { "epoch": 0.14721067116655617, "grad_norm": 3.710327386856079, "learning_rate": 1.6942355889724312e-06, "loss": 0.38389042019844055, "num_input_tokens_seen": 5167450, "step": 1832, "train_runtime": 6093.6852, "train_tokens_per_second": 848.001 }, { "epoch": 0.14729102633640692, "grad_norm": 3.132692337036133, "learning_rate": 1.6842105263157895e-06, "loss": 0.28569597005844116, "num_input_tokens_seen": 5171362, "step": 1833, "train_runtime": 6096.5389, "train_tokens_per_second": 848.246 }, { "epoch": 0.14737138150625767, "grad_norm": 3.359224319458008, "learning_rate": 1.6741854636591482e-06, "loss": 0.41540682315826416, "num_input_tokens_seen": 5174048, "step": 1834, "train_runtime": 6099.0903, "train_tokens_per_second": 848.331 }, { "epoch": 0.1474517366761084, "grad_norm": 2.9024569988250732, "learning_rate": 1.6641604010025065e-06, "loss": 0.33381617069244385, "num_input_tokens_seen": 5178648, "step": 1835, "train_runtime": 6102.0976, "train_tokens_per_second": 848.667 }, { "epoch": 0.14753209184595914, "grad_norm": 2.884857177734375, "learning_rate": 1.6541353383458648e-06, "loss": 0.23676171898841858, "num_input_tokens_seen": 5181396, "step": 1836, "train_runtime": 6104.69, "train_tokens_per_second": 848.757 }, { "epoch": 0.1476124470158099, "grad_norm": 3.1737420558929443, "learning_rate": 1.6441102756892233e-06, "loss": 0.33823448419570923, "num_input_tokens_seen": 5184230, "step": 1837, "train_runtime": 6107.2942, "train_tokens_per_second": 848.859 }, { "epoch": 0.1476928021856606, "grad_norm": 3.3363163471221924, "learning_rate": 1.6340852130325815e-06, "loss": 0.2511402368545532, "num_input_tokens_seen": 5186670, "step": 1838, "train_runtime": 6109.7986, "train_tokens_per_second": 848.91 }, { "epoch": 0.14777315735551136, "grad_norm": 2.703573226928711, "learning_rate": 1.62406015037594e-06, "loss": 0.28338298201560974, "num_input_tokens_seen": 5189754, "step": 1839, "train_runtime": 6112.5042, "train_tokens_per_second": 849.039 }, { "epoch": 0.1478535125253621, "grad_norm": 2.719547986984253, "learning_rate": 1.6140350877192983e-06, "loss": 0.32121896743774414, "num_input_tokens_seen": 5192990, "step": 1840, "train_runtime": 6115.3899, "train_tokens_per_second": 849.167 }, { "epoch": 0.14793386769521283, "grad_norm": 3.1726865768432617, "learning_rate": 1.6040100250626568e-06, "loss": 0.32551372051239014, "num_input_tokens_seen": 5195162, "step": 1841, "train_runtime": 6117.855, "train_tokens_per_second": 849.18 }, { "epoch": 0.14801422286506358, "grad_norm": 3.2489185333251953, "learning_rate": 1.5939849624060153e-06, "loss": 0.3134799599647522, "num_input_tokens_seen": 5196734, "step": 1842, "train_runtime": 6120.1455, "train_tokens_per_second": 849.119 }, { "epoch": 0.14809457803491433, "grad_norm": 2.9012372493743896, "learning_rate": 1.5839598997493736e-06, "loss": 0.1781068742275238, "num_input_tokens_seen": 5199838, "step": 1843, "train_runtime": 6122.8074, "train_tokens_per_second": 849.257 }, { "epoch": 0.14817493320476507, "grad_norm": 2.9657764434814453, "learning_rate": 1.573934837092732e-06, "loss": 0.2534942328929901, "num_input_tokens_seen": 5201452, "step": 1844, "train_runtime": 6125.0979, "train_tokens_per_second": 849.203 }, { "epoch": 0.1482552883746158, "grad_norm": 3.2403035163879395, "learning_rate": 1.5639097744360903e-06, "loss": 0.2998630106449127, "num_input_tokens_seen": 5204442, "step": 1845, "train_runtime": 6127.722, "train_tokens_per_second": 849.327 }, { "epoch": 0.14833564354446654, "grad_norm": 2.4931602478027344, "learning_rate": 1.5538847117794486e-06, "loss": 0.21516990661621094, "num_input_tokens_seen": 5207238, "step": 1846, "train_runtime": 6130.3477, "train_tokens_per_second": 849.42 }, { "epoch": 0.1484159987143173, "grad_norm": 2.398267984390259, "learning_rate": 1.5438596491228073e-06, "loss": 0.2641172409057617, "num_input_tokens_seen": 5212128, "step": 1847, "train_runtime": 6133.5269, "train_tokens_per_second": 849.777 }, { "epoch": 0.14849635388416801, "grad_norm": 3.874631404876709, "learning_rate": 1.5338345864661656e-06, "loss": 0.3232131898403168, "num_input_tokens_seen": 5214726, "step": 1848, "train_runtime": 6136.1031, "train_tokens_per_second": 849.843 }, { "epoch": 0.14857670905401876, "grad_norm": 2.4705722332000732, "learning_rate": 1.523809523809524e-06, "loss": 0.189859077334404, "num_input_tokens_seen": 5217510, "step": 1849, "train_runtime": 6138.7375, "train_tokens_per_second": 849.932 }, { "epoch": 0.1486570642238695, "grad_norm": 3.447075128555298, "learning_rate": 1.5137844611528823e-06, "loss": 0.3698558807373047, "num_input_tokens_seen": 5220786, "step": 1850, "train_runtime": 6141.446, "train_tokens_per_second": 850.091 }, { "epoch": 0.14873741939372023, "grad_norm": 3.309690237045288, "learning_rate": 1.5037593984962406e-06, "loss": 0.253084659576416, "num_input_tokens_seen": 5223268, "step": 1851, "train_runtime": 6143.986, "train_tokens_per_second": 850.143 }, { "epoch": 0.14881777456357098, "grad_norm": 3.5335569381713867, "learning_rate": 1.4937343358395993e-06, "loss": 0.39097344875335693, "num_input_tokens_seen": 5226400, "step": 1852, "train_runtime": 6146.8367, "train_tokens_per_second": 850.258 }, { "epoch": 0.14889812973342173, "grad_norm": 3.3545398712158203, "learning_rate": 1.4837092731829576e-06, "loss": 0.3471872806549072, "num_input_tokens_seen": 5228710, "step": 1853, "train_runtime": 6149.3049, "train_tokens_per_second": 850.293 }, { "epoch": 0.14897848490327245, "grad_norm": 2.426208257675171, "learning_rate": 1.4736842105263159e-06, "loss": 0.24170702695846558, "num_input_tokens_seen": 5231192, "step": 1854, "train_runtime": 6151.8031, "train_tokens_per_second": 850.351 }, { "epoch": 0.1490588400731232, "grad_norm": 2.6966443061828613, "learning_rate": 1.4636591478696743e-06, "loss": 0.2835777699947357, "num_input_tokens_seen": 5233734, "step": 1855, "train_runtime": 6154.2809, "train_tokens_per_second": 850.422 }, { "epoch": 0.14913919524297395, "grad_norm": 3.060317039489746, "learning_rate": 1.4536340852130326e-06, "loss": 0.2831120789051056, "num_input_tokens_seen": 5237364, "step": 1856, "train_runtime": 6157.0239, "train_tokens_per_second": 850.632 }, { "epoch": 0.1492195504128247, "grad_norm": 3.5655839443206787, "learning_rate": 1.4436090225563911e-06, "loss": 0.3785494863986969, "num_input_tokens_seen": 5239586, "step": 1857, "train_runtime": 6159.5219, "train_tokens_per_second": 850.648 }, { "epoch": 0.14929990558267542, "grad_norm": 3.9480326175689697, "learning_rate": 1.4335839598997494e-06, "loss": 0.30042004585266113, "num_input_tokens_seen": 5242376, "step": 1858, "train_runtime": 6162.1, "train_tokens_per_second": 850.745 }, { "epoch": 0.14938026075252617, "grad_norm": 3.295776844024658, "learning_rate": 1.4235588972431077e-06, "loss": 0.3053576946258545, "num_input_tokens_seen": 5246954, "step": 1859, "train_runtime": 6165.1164, "train_tokens_per_second": 851.071 }, { "epoch": 0.14946061592237692, "grad_norm": 3.5838963985443115, "learning_rate": 1.4135338345864664e-06, "loss": 0.3052842915058136, "num_input_tokens_seen": 5248336, "step": 1860, "train_runtime": 6167.3313, "train_tokens_per_second": 850.99 }, { "epoch": 0.14954097109222764, "grad_norm": 3.5218191146850586, "learning_rate": 1.4035087719298246e-06, "loss": 0.2905220687389374, "num_input_tokens_seen": 5250982, "step": 1861, "train_runtime": 6181.1554, "train_tokens_per_second": 849.515 }, { "epoch": 0.14962132626207839, "grad_norm": 3.380802631378174, "learning_rate": 1.3934837092731831e-06, "loss": 0.3273152709007263, "num_input_tokens_seen": 5253054, "step": 1862, "train_runtime": 6183.5442, "train_tokens_per_second": 849.522 }, { "epoch": 0.14970168143192913, "grad_norm": 2.9948363304138184, "learning_rate": 1.3834586466165414e-06, "loss": 0.2760758697986603, "num_input_tokens_seen": 5258022, "step": 1863, "train_runtime": 6186.6671, "train_tokens_per_second": 849.896 }, { "epoch": 0.14978203660177986, "grad_norm": 2.917080879211426, "learning_rate": 1.3734335839598997e-06, "loss": 0.23967498540878296, "num_input_tokens_seen": 5259710, "step": 1864, "train_runtime": 6188.9269, "train_tokens_per_second": 849.858 }, { "epoch": 0.1498623917716306, "grad_norm": 3.005389451980591, "learning_rate": 1.3634085213032584e-06, "loss": 0.30800870060920715, "num_input_tokens_seen": 5262426, "step": 1865, "train_runtime": 6191.489, "train_tokens_per_second": 849.945 }, { "epoch": 0.14994274694148135, "grad_norm": 2.991507053375244, "learning_rate": 1.3533834586466167e-06, "loss": 0.22139325737953186, "num_input_tokens_seen": 5264508, "step": 1866, "train_runtime": 6193.8756, "train_tokens_per_second": 849.954 }, { "epoch": 0.15002310211133207, "grad_norm": 3.5413694381713867, "learning_rate": 1.3433583959899751e-06, "loss": 0.2844051122665405, "num_input_tokens_seen": 5267542, "step": 1867, "train_runtime": 6196.5881, "train_tokens_per_second": 850.071 }, { "epoch": 0.15010345728118282, "grad_norm": 3.2032620906829834, "learning_rate": 1.3333333333333334e-06, "loss": 0.2877874970436096, "num_input_tokens_seen": 5269638, "step": 1868, "train_runtime": 6198.9949, "train_tokens_per_second": 850.079 }, { "epoch": 0.15018381245103357, "grad_norm": 3.3595235347747803, "learning_rate": 1.3233082706766917e-06, "loss": 0.2722119390964508, "num_input_tokens_seen": 5273694, "step": 1869, "train_runtime": 6201.9552, "train_tokens_per_second": 850.328 }, { "epoch": 0.15026416762088432, "grad_norm": 3.2541401386260986, "learning_rate": 1.3132832080200502e-06, "loss": 0.29996857047080994, "num_input_tokens_seen": 5276588, "step": 1870, "train_runtime": 6204.585, "train_tokens_per_second": 850.434 }, { "epoch": 0.15034452279073504, "grad_norm": 3.0926527976989746, "learning_rate": 1.3032581453634085e-06, "loss": 0.24269652366638184, "num_input_tokens_seen": 5283728, "step": 1871, "train_runtime": 6208.3334, "train_tokens_per_second": 851.07 }, { "epoch": 0.1504248779605858, "grad_norm": 3.0363216400146484, "learning_rate": 1.2932330827067672e-06, "loss": 0.3188760578632355, "num_input_tokens_seen": 5286564, "step": 1872, "train_runtime": 6210.9199, "train_tokens_per_second": 851.172 }, { "epoch": 0.15050523313043654, "grad_norm": 3.2666068077087402, "learning_rate": 1.2832080200501254e-06, "loss": 0.36560237407684326, "num_input_tokens_seen": 5289004, "step": 1873, "train_runtime": 6213.3997, "train_tokens_per_second": 851.225 }, { "epoch": 0.15058558830028726, "grad_norm": 2.6077520847320557, "learning_rate": 1.2731829573934837e-06, "loss": 0.2903512716293335, "num_input_tokens_seen": 5291894, "step": 1874, "train_runtime": 6215.9785, "train_tokens_per_second": 851.337 }, { "epoch": 0.150665943470138, "grad_norm": 3.5012166500091553, "learning_rate": 1.2631578947368422e-06, "loss": 0.330009788274765, "num_input_tokens_seen": 5296040, "step": 1875, "train_runtime": 6218.8704, "train_tokens_per_second": 851.608 }, { "epoch": 0.15074629863998876, "grad_norm": 3.2377772331237793, "learning_rate": 1.2531328320802005e-06, "loss": 0.3203224241733551, "num_input_tokens_seen": 5298980, "step": 1876, "train_runtime": 6221.4747, "train_tokens_per_second": 851.724 }, { "epoch": 0.15082665380983948, "grad_norm": 2.445685625076294, "learning_rate": 1.243107769423559e-06, "loss": 0.23817768692970276, "num_input_tokens_seen": 5301168, "step": 1877, "train_runtime": 6223.8966, "train_tokens_per_second": 851.744 }, { "epoch": 0.15090700897969023, "grad_norm": 4.016617298126221, "learning_rate": 1.2330827067669174e-06, "loss": 0.22632461786270142, "num_input_tokens_seen": 5303824, "step": 1878, "train_runtime": 6226.4271, "train_tokens_per_second": 851.825 }, { "epoch": 0.15098736414954098, "grad_norm": 2.841778516769409, "learning_rate": 1.2230576441102757e-06, "loss": 0.2677749991416931, "num_input_tokens_seen": 5305686, "step": 1879, "train_runtime": 6228.7755, "train_tokens_per_second": 851.802 }, { "epoch": 0.15106771931939172, "grad_norm": 3.427945375442505, "learning_rate": 1.2130325814536342e-06, "loss": 0.32309070229530334, "num_input_tokens_seen": 5307106, "step": 1880, "train_runtime": 6230.9867, "train_tokens_per_second": 851.728 }, { "epoch": 0.15114807448924245, "grad_norm": 3.5696756839752197, "learning_rate": 1.2030075187969925e-06, "loss": 0.38240352272987366, "num_input_tokens_seen": 5310178, "step": 1881, "train_runtime": 6233.5804, "train_tokens_per_second": 851.866 }, { "epoch": 0.1512284296590932, "grad_norm": 3.3981826305389404, "learning_rate": 1.192982456140351e-06, "loss": 0.24049322307109833, "num_input_tokens_seen": 5312424, "step": 1882, "train_runtime": 6236.0291, "train_tokens_per_second": 851.892 }, { "epoch": 0.15130878482894394, "grad_norm": 2.9497156143188477, "learning_rate": 1.1829573934837095e-06, "loss": 0.2853478789329529, "num_input_tokens_seen": 5314320, "step": 1883, "train_runtime": 6238.3774, "train_tokens_per_second": 851.875 }, { "epoch": 0.15138913999879466, "grad_norm": 2.7669665813446045, "learning_rate": 1.1729323308270677e-06, "loss": 0.24880701303482056, "num_input_tokens_seen": 5317336, "step": 1884, "train_runtime": 6241.0482, "train_tokens_per_second": 851.994 }, { "epoch": 0.1514694951686454, "grad_norm": 2.9516966342926025, "learning_rate": 1.1629072681704262e-06, "loss": 0.2909119427204132, "num_input_tokens_seen": 5319842, "step": 1885, "train_runtime": 6243.5937, "train_tokens_per_second": 852.048 }, { "epoch": 0.15154985033849616, "grad_norm": 2.674546957015991, "learning_rate": 1.1528822055137845e-06, "loss": 0.2621927857398987, "num_input_tokens_seen": 5322424, "step": 1886, "train_runtime": 6246.2026, "train_tokens_per_second": 852.106 }, { "epoch": 0.15163020550834688, "grad_norm": 3.0213022232055664, "learning_rate": 1.142857142857143e-06, "loss": 0.26921045780181885, "num_input_tokens_seen": 5324876, "step": 1887, "train_runtime": 6248.7869, "train_tokens_per_second": 852.146 }, { "epoch": 0.15171056067819763, "grad_norm": 3.2578752040863037, "learning_rate": 1.1328320802005013e-06, "loss": 0.28392139077186584, "num_input_tokens_seen": 5326232, "step": 1888, "train_runtime": 6251.0214, "train_tokens_per_second": 852.058 }, { "epoch": 0.15179091584804838, "grad_norm": 2.9834201335906982, "learning_rate": 1.1228070175438598e-06, "loss": 0.2287100851535797, "num_input_tokens_seen": 5327652, "step": 1889, "train_runtime": 6253.251, "train_tokens_per_second": 851.981 }, { "epoch": 0.1518712710178991, "grad_norm": 2.5924243927001953, "learning_rate": 1.1127819548872182e-06, "loss": 0.205129012465477, "num_input_tokens_seen": 5329680, "step": 1890, "train_runtime": 6255.659, "train_tokens_per_second": 851.977 }, { "epoch": 0.15195162618774985, "grad_norm": 2.6682045459747314, "learning_rate": 1.1027568922305765e-06, "loss": 0.25801795721054077, "num_input_tokens_seen": 5331016, "step": 1891, "train_runtime": 6269.8183, "train_tokens_per_second": 850.266 }, { "epoch": 0.1520319813576006, "grad_norm": 3.045374631881714, "learning_rate": 1.0927318295739348e-06, "loss": 0.32460835576057434, "num_input_tokens_seen": 5333208, "step": 1892, "train_runtime": 6272.2683, "train_tokens_per_second": 850.284 }, { "epoch": 0.15211233652745135, "grad_norm": 2.8790879249572754, "learning_rate": 1.0827067669172933e-06, "loss": 0.26253741979599, "num_input_tokens_seen": 5336020, "step": 1893, "train_runtime": 6274.8367, "train_tokens_per_second": 850.384 }, { "epoch": 0.15219269169730207, "grad_norm": 3.370601177215576, "learning_rate": 1.0726817042606518e-06, "loss": 0.24914756417274475, "num_input_tokens_seen": 5338552, "step": 1894, "train_runtime": 6277.4212, "train_tokens_per_second": 850.437 }, { "epoch": 0.15227304686715282, "grad_norm": 2.6314282417297363, "learning_rate": 1.0626566416040103e-06, "loss": 0.2269984632730484, "num_input_tokens_seen": 5341684, "step": 1895, "train_runtime": 6280.0716, "train_tokens_per_second": 850.577 }, { "epoch": 0.15235340203700357, "grad_norm": 3.9569621086120605, "learning_rate": 1.0526315789473685e-06, "loss": 0.23344630002975464, "num_input_tokens_seen": 5342822, "step": 1896, "train_runtime": 6282.313, "train_tokens_per_second": 850.455 }, { "epoch": 0.1524337572068543, "grad_norm": 2.572211742401123, "learning_rate": 1.0426065162907268e-06, "loss": 0.22743040323257446, "num_input_tokens_seen": 5344938, "step": 1897, "train_runtime": 6284.727, "train_tokens_per_second": 850.465 }, { "epoch": 0.15251411237670504, "grad_norm": 3.2575736045837402, "learning_rate": 1.0325814536340853e-06, "loss": 0.2278350293636322, "num_input_tokens_seen": 5346656, "step": 1898, "train_runtime": 6287.0493, "train_tokens_per_second": 850.424 }, { "epoch": 0.15259446754655578, "grad_norm": 2.725104331970215, "learning_rate": 1.0225563909774438e-06, "loss": 0.2828885018825531, "num_input_tokens_seen": 5349808, "step": 1899, "train_runtime": 6289.702, "train_tokens_per_second": 850.566 }, { "epoch": 0.1526748227164065, "grad_norm": 2.6800684928894043, "learning_rate": 1.012531328320802e-06, "loss": 0.3034268617630005, "num_input_tokens_seen": 5351746, "step": 1900, "train_runtime": 6292.1182, "train_tokens_per_second": 850.548 }, { "epoch": 0.15275517788625725, "grad_norm": 3.3724982738494873, "learning_rate": 1.0025062656641603e-06, "loss": 0.282732218503952, "num_input_tokens_seen": 5355248, "step": 1901, "train_runtime": 6295.0011, "train_tokens_per_second": 850.714 }, { "epoch": 0.152835533056108, "grad_norm": 3.5052149295806885, "learning_rate": 9.924812030075188e-07, "loss": 0.2908499240875244, "num_input_tokens_seen": 5357348, "step": 1902, "train_runtime": 6297.5131, "train_tokens_per_second": 850.709 }, { "epoch": 0.15291588822595872, "grad_norm": 2.709437370300293, "learning_rate": 9.824561403508773e-07, "loss": 0.2121705710887909, "num_input_tokens_seen": 5359760, "step": 1903, "train_runtime": 6300.0755, "train_tokens_per_second": 850.745 }, { "epoch": 0.15299624339580947, "grad_norm": 2.9009225368499756, "learning_rate": 9.724310776942358e-07, "loss": 0.2923988699913025, "num_input_tokens_seen": 5364586, "step": 1904, "train_runtime": 6303.1542, "train_tokens_per_second": 851.095 }, { "epoch": 0.15307659856566022, "grad_norm": 2.719784736633301, "learning_rate": 9.62406015037594e-07, "loss": 0.2966820001602173, "num_input_tokens_seen": 5367740, "step": 1905, "train_runtime": 6305.8315, "train_tokens_per_second": 851.234 }, { "epoch": 0.15315695373551097, "grad_norm": 2.6837711334228516, "learning_rate": 9.523809523809525e-07, "loss": 0.2750667929649353, "num_input_tokens_seen": 5369812, "step": 1906, "train_runtime": 6308.1966, "train_tokens_per_second": 851.244 }, { "epoch": 0.1532373089053617, "grad_norm": 2.7049546241760254, "learning_rate": 9.423558897243108e-07, "loss": 0.2653680443763733, "num_input_tokens_seen": 5371706, "step": 1907, "train_runtime": 6310.6277, "train_tokens_per_second": 851.216 }, { "epoch": 0.15331766407521244, "grad_norm": 2.9241247177124023, "learning_rate": 9.323308270676692e-07, "loss": 0.28262242674827576, "num_input_tokens_seen": 5374204, "step": 1908, "train_runtime": 6313.1322, "train_tokens_per_second": 851.274 }, { "epoch": 0.1533980192450632, "grad_norm": 2.9811325073242188, "learning_rate": 9.223057644110277e-07, "loss": 0.2916770279407501, "num_input_tokens_seen": 5376410, "step": 1909, "train_runtime": 6315.5545, "train_tokens_per_second": 851.297 }, { "epoch": 0.1534783744149139, "grad_norm": 2.9105608463287354, "learning_rate": 9.122807017543861e-07, "loss": 0.27223366498947144, "num_input_tokens_seen": 5378758, "step": 1910, "train_runtime": 6318.0254, "train_tokens_per_second": 851.335 }, { "epoch": 0.15355872958476466, "grad_norm": 3.421970844268799, "learning_rate": 9.022556390977444e-07, "loss": 0.4022532105445862, "num_input_tokens_seen": 5381050, "step": 1911, "train_runtime": 6320.5663, "train_tokens_per_second": 851.356 }, { "epoch": 0.1536390847546154, "grad_norm": 2.547138214111328, "learning_rate": 8.922305764411029e-07, "loss": 0.2474459707736969, "num_input_tokens_seen": 5383546, "step": 1912, "train_runtime": 6323.159, "train_tokens_per_second": 851.401 }, { "epoch": 0.15371943992446613, "grad_norm": 3.228506565093994, "learning_rate": 8.822055137844612e-07, "loss": 0.5346630215644836, "num_input_tokens_seen": 5388768, "step": 1913, "train_runtime": 6326.4343, "train_tokens_per_second": 851.786 }, { "epoch": 0.15379979509431688, "grad_norm": 3.1771228313446045, "learning_rate": 8.721804511278196e-07, "loss": 0.28206270933151245, "num_input_tokens_seen": 5391934, "step": 1914, "train_runtime": 6329.1712, "train_tokens_per_second": 851.918 }, { "epoch": 0.15388015026416763, "grad_norm": 2.406018018722534, "learning_rate": 8.621553884711781e-07, "loss": 0.2116585671901703, "num_input_tokens_seen": 5393534, "step": 1915, "train_runtime": 6331.5523, "train_tokens_per_second": 851.85 }, { "epoch": 0.15396050543401837, "grad_norm": 2.7135403156280518, "learning_rate": 8.521303258145364e-07, "loss": 0.27939337491989136, "num_input_tokens_seen": 5396880, "step": 1916, "train_runtime": 6334.2995, "train_tokens_per_second": 852.009 }, { "epoch": 0.1540408606038691, "grad_norm": 3.530348062515259, "learning_rate": 8.421052631578948e-07, "loss": 0.26044297218322754, "num_input_tokens_seen": 5398468, "step": 1917, "train_runtime": 6336.6209, "train_tokens_per_second": 851.947 }, { "epoch": 0.15412121577371984, "grad_norm": 3.2478277683258057, "learning_rate": 8.320802005012532e-07, "loss": 0.2745926082134247, "num_input_tokens_seen": 5399860, "step": 1918, "train_runtime": 6338.7976, "train_tokens_per_second": 851.874 }, { "epoch": 0.1542015709435706, "grad_norm": 2.3195953369140625, "learning_rate": 8.220551378446116e-07, "loss": 0.19344393908977509, "num_input_tokens_seen": 5402214, "step": 1919, "train_runtime": 6341.2307, "train_tokens_per_second": 851.919 }, { "epoch": 0.1542819261134213, "grad_norm": 2.9964892864227295, "learning_rate": 8.1203007518797e-07, "loss": 0.2635641098022461, "num_input_tokens_seen": 5403832, "step": 1920, "train_runtime": 6343.4747, "train_tokens_per_second": 851.873 }, { "epoch": 0.15436228128327206, "grad_norm": 2.8591902256011963, "learning_rate": 8.020050125313284e-07, "loss": 0.23747482895851135, "num_input_tokens_seen": 5405868, "step": 1921, "train_runtime": 6357.1485, "train_tokens_per_second": 850.361 }, { "epoch": 0.1544426364531228, "grad_norm": 3.1445531845092773, "learning_rate": 7.919799498746868e-07, "loss": 0.34054768085479736, "num_input_tokens_seen": 5408978, "step": 1922, "train_runtime": 6359.774, "train_tokens_per_second": 850.498 }, { "epoch": 0.15452299162297353, "grad_norm": 3.0237293243408203, "learning_rate": 7.819548872180452e-07, "loss": 0.31414127349853516, "num_input_tokens_seen": 5410674, "step": 1923, "train_runtime": 6362.1737, "train_tokens_per_second": 850.444 }, { "epoch": 0.15460334679282428, "grad_norm": 2.6085968017578125, "learning_rate": 7.719298245614036e-07, "loss": 0.2482825070619583, "num_input_tokens_seen": 5412796, "step": 1924, "train_runtime": 6364.5866, "train_tokens_per_second": 850.455 }, { "epoch": 0.15468370196267503, "grad_norm": 2.7537918090820312, "learning_rate": 7.61904761904762e-07, "loss": 0.2909414768218994, "num_input_tokens_seen": 5416514, "step": 1925, "train_runtime": 6367.4377, "train_tokens_per_second": 850.658 }, { "epoch": 0.15476405713252575, "grad_norm": 3.300318956375122, "learning_rate": 7.518796992481203e-07, "loss": 0.404562771320343, "num_input_tokens_seen": 5418704, "step": 1926, "train_runtime": 6369.8899, "train_tokens_per_second": 850.675 }, { "epoch": 0.1548444123023765, "grad_norm": 3.194606065750122, "learning_rate": 7.418546365914788e-07, "loss": 0.3142998516559601, "num_input_tokens_seen": 5421914, "step": 1927, "train_runtime": 6372.5683, "train_tokens_per_second": 850.821 }, { "epoch": 0.15492476747222725, "grad_norm": 2.9768946170806885, "learning_rate": 7.318295739348372e-07, "loss": 0.3042275309562683, "num_input_tokens_seen": 5423886, "step": 1928, "train_runtime": 6374.9581, "train_tokens_per_second": 850.811 }, { "epoch": 0.155005122642078, "grad_norm": 2.6181235313415527, "learning_rate": 7.218045112781956e-07, "loss": 0.30179303884506226, "num_input_tokens_seen": 5427414, "step": 1929, "train_runtime": 6377.728, "train_tokens_per_second": 850.995 }, { "epoch": 0.15508547781192872, "grad_norm": 3.6934986114501953, "learning_rate": 7.117794486215538e-07, "loss": 0.3255188465118408, "num_input_tokens_seen": 5431516, "step": 1930, "train_runtime": 6380.7129, "train_tokens_per_second": 851.24 }, { "epoch": 0.15516583298177947, "grad_norm": 3.028489828109741, "learning_rate": 7.017543859649123e-07, "loss": 0.2759728729724884, "num_input_tokens_seen": 5433732, "step": 1931, "train_runtime": 6383.1996, "train_tokens_per_second": 851.255 }, { "epoch": 0.15524618815163022, "grad_norm": 2.511678695678711, "learning_rate": 6.917293233082707e-07, "loss": 0.18361154198646545, "num_input_tokens_seen": 5437744, "step": 1932, "train_runtime": 6386.1412, "train_tokens_per_second": 851.491 }, { "epoch": 0.15532654332148094, "grad_norm": 2.9492247104644775, "learning_rate": 6.817042606516292e-07, "loss": 0.26676538586616516, "num_input_tokens_seen": 5440494, "step": 1933, "train_runtime": 6388.8064, "train_tokens_per_second": 851.567 }, { "epoch": 0.15540689849133169, "grad_norm": 2.5319151878356934, "learning_rate": 6.716791979949876e-07, "loss": 0.2374674379825592, "num_input_tokens_seen": 5442726, "step": 1934, "train_runtime": 6391.2587, "train_tokens_per_second": 851.589 }, { "epoch": 0.15548725366118243, "grad_norm": 2.658557653427124, "learning_rate": 6.616541353383458e-07, "loss": 0.18091371655464172, "num_input_tokens_seen": 5444802, "step": 1935, "train_runtime": 6393.6462, "train_tokens_per_second": 851.596 }, { "epoch": 0.15556760883103316, "grad_norm": 3.1839990615844727, "learning_rate": 6.516290726817042e-07, "loss": 0.28924867510795593, "num_input_tokens_seen": 5447394, "step": 1936, "train_runtime": 6396.2843, "train_tokens_per_second": 851.65 }, { "epoch": 0.1556479640008839, "grad_norm": 2.8640079498291016, "learning_rate": 6.416040100250627e-07, "loss": 0.2441353052854538, "num_input_tokens_seen": 5449472, "step": 1937, "train_runtime": 6398.6739, "train_tokens_per_second": 851.656 }, { "epoch": 0.15572831917073465, "grad_norm": 2.515584945678711, "learning_rate": 6.315789473684211e-07, "loss": 0.334287166595459, "num_input_tokens_seen": 5452976, "step": 1938, "train_runtime": 6401.5098, "train_tokens_per_second": 851.827 }, { "epoch": 0.15580867434058537, "grad_norm": 2.9840104579925537, "learning_rate": 6.215538847117795e-07, "loss": 0.2988210916519165, "num_input_tokens_seen": 5456446, "step": 1939, "train_runtime": 6404.2362, "train_tokens_per_second": 852.006 }, { "epoch": 0.15588902951043612, "grad_norm": 2.701909303665161, "learning_rate": 6.115288220551379e-07, "loss": 0.27468201518058777, "num_input_tokens_seen": 5459582, "step": 1940, "train_runtime": 6406.9381, "train_tokens_per_second": 852.136 }, { "epoch": 0.15596938468028687, "grad_norm": 2.5192322731018066, "learning_rate": 6.015037593984962e-07, "loss": 0.2132025957107544, "num_input_tokens_seen": 5461910, "step": 1941, "train_runtime": 6409.398, "train_tokens_per_second": 852.172 }, { "epoch": 0.15604973985013762, "grad_norm": 2.43003511428833, "learning_rate": 5.914786967418547e-07, "loss": 0.27549299597740173, "num_input_tokens_seen": 5463756, "step": 1942, "train_runtime": 6411.7552, "train_tokens_per_second": 852.147 }, { "epoch": 0.15613009501998834, "grad_norm": 3.027409315109253, "learning_rate": 5.814536340852131e-07, "loss": 0.2662180960178375, "num_input_tokens_seen": 5465906, "step": 1943, "train_runtime": 6414.1632, "train_tokens_per_second": 852.162 }, { "epoch": 0.1562104501898391, "grad_norm": 3.3979382514953613, "learning_rate": 5.714285714285715e-07, "loss": 0.28874868154525757, "num_input_tokens_seen": 5468452, "step": 1944, "train_runtime": 6416.6676, "train_tokens_per_second": 852.226 }, { "epoch": 0.15629080535968984, "grad_norm": 3.051954507827759, "learning_rate": 5.614035087719299e-07, "loss": 0.2529032230377197, "num_input_tokens_seen": 5471230, "step": 1945, "train_runtime": 6419.2041, "train_tokens_per_second": 852.322 }, { "epoch": 0.15637116052954056, "grad_norm": 3.236360788345337, "learning_rate": 5.513784461152883e-07, "loss": 0.2616214156150818, "num_input_tokens_seen": 5474106, "step": 1946, "train_runtime": 6421.8144, "train_tokens_per_second": 852.424 }, { "epoch": 0.1564515156993913, "grad_norm": 3.454623222351074, "learning_rate": 5.413533834586466e-07, "loss": 0.36226797103881836, "num_input_tokens_seen": 5478008, "step": 1947, "train_runtime": 6424.6445, "train_tokens_per_second": 852.655 }, { "epoch": 0.15653187086924206, "grad_norm": 3.6148548126220703, "learning_rate": 5.313283208020051e-07, "loss": 0.2593331038951874, "num_input_tokens_seen": 5479722, "step": 1948, "train_runtime": 6426.9341, "train_tokens_per_second": 852.618 }, { "epoch": 0.15661222603909278, "grad_norm": 2.946627140045166, "learning_rate": 5.213032581453634e-07, "loss": 0.3075236678123474, "num_input_tokens_seen": 5482650, "step": 1949, "train_runtime": 6429.5599, "train_tokens_per_second": 852.726 }, { "epoch": 0.15669258120894353, "grad_norm": 3.2776122093200684, "learning_rate": 5.112781954887219e-07, "loss": 0.20214340090751648, "num_input_tokens_seen": 5484450, "step": 1950, "train_runtime": 6431.8638, "train_tokens_per_second": 852.7 }, { "epoch": 0.15677293637879428, "grad_norm": 2.9741241931915283, "learning_rate": 5.012531328320802e-07, "loss": 0.2980518043041229, "num_input_tokens_seen": 5486316, "step": 1951, "train_runtime": 6445.5017, "train_tokens_per_second": 851.185 }, { "epoch": 0.15685329154864502, "grad_norm": 2.8899412155151367, "learning_rate": 4.912280701754387e-07, "loss": 0.3211197853088379, "num_input_tokens_seen": 5489074, "step": 1952, "train_runtime": 6448.0571, "train_tokens_per_second": 851.276 }, { "epoch": 0.15693364671849575, "grad_norm": 3.412649393081665, "learning_rate": 4.81203007518797e-07, "loss": 0.34056374430656433, "num_input_tokens_seen": 5491714, "step": 1953, "train_runtime": 6450.5893, "train_tokens_per_second": 851.351 }, { "epoch": 0.1570140018883465, "grad_norm": 2.2660436630249023, "learning_rate": 4.711779448621554e-07, "loss": 0.18520742654800415, "num_input_tokens_seen": 5494608, "step": 1954, "train_runtime": 6453.1883, "train_tokens_per_second": 851.456 }, { "epoch": 0.15709435705819724, "grad_norm": 2.5797133445739746, "learning_rate": 4.6115288220551385e-07, "loss": 0.21464145183563232, "num_input_tokens_seen": 5496704, "step": 1955, "train_runtime": 6455.5962, "train_tokens_per_second": 851.463 }, { "epoch": 0.15717471222804796, "grad_norm": 3.2148096561431885, "learning_rate": 4.511278195488722e-07, "loss": 0.30257993936538696, "num_input_tokens_seen": 5499730, "step": 1956, "train_runtime": 6458.1946, "train_tokens_per_second": 851.589 }, { "epoch": 0.1572550673978987, "grad_norm": 1.921965479850769, "learning_rate": 4.411027568922306e-07, "loss": 0.14744429290294647, "num_input_tokens_seen": 5502526, "step": 1957, "train_runtime": 6460.7576, "train_tokens_per_second": 851.684 }, { "epoch": 0.15733542256774946, "grad_norm": 2.946406364440918, "learning_rate": 4.3107769423558905e-07, "loss": 0.2980283498764038, "num_input_tokens_seen": 5505536, "step": 1958, "train_runtime": 6463.3422, "train_tokens_per_second": 851.809 }, { "epoch": 0.15741577773760018, "grad_norm": 2.671752452850342, "learning_rate": 4.210526315789474e-07, "loss": 0.33904901146888733, "num_input_tokens_seen": 5507778, "step": 1959, "train_runtime": 6465.8043, "train_tokens_per_second": 851.832 }, { "epoch": 0.15749613290745093, "grad_norm": 2.290651559829712, "learning_rate": 4.110275689223058e-07, "loss": 0.31553128361701965, "num_input_tokens_seen": 5511684, "step": 1960, "train_runtime": 6468.8669, "train_tokens_per_second": 852.032 }, { "epoch": 0.15757648807730168, "grad_norm": 3.115382194519043, "learning_rate": 4.010025062656642e-07, "loss": 0.34248441457748413, "num_input_tokens_seen": 5514098, "step": 1961, "train_runtime": 6471.3348, "train_tokens_per_second": 852.08 }, { "epoch": 0.1576568432471524, "grad_norm": 3.3842735290527344, "learning_rate": 3.909774436090226e-07, "loss": 0.31714296340942383, "num_input_tokens_seen": 5516842, "step": 1962, "train_runtime": 6473.8637, "train_tokens_per_second": 852.171 }, { "epoch": 0.15773719841700315, "grad_norm": 3.1095826625823975, "learning_rate": 3.80952380952381e-07, "loss": 0.2989776134490967, "num_input_tokens_seen": 5519764, "step": 1963, "train_runtime": 6476.454, "train_tokens_per_second": 852.282 }, { "epoch": 0.1578175535868539, "grad_norm": 3.3863987922668457, "learning_rate": 3.709273182957394e-07, "loss": 0.45265576243400574, "num_input_tokens_seen": 5523028, "step": 1964, "train_runtime": 6479.2156, "train_tokens_per_second": 852.422 }, { "epoch": 0.15789790875670465, "grad_norm": 2.737675428390503, "learning_rate": 3.609022556390978e-07, "loss": 0.3501412570476532, "num_input_tokens_seen": 5526378, "step": 1965, "train_runtime": 6481.9068, "train_tokens_per_second": 852.585 }, { "epoch": 0.15797826392655537, "grad_norm": 3.511582851409912, "learning_rate": 3.5087719298245616e-07, "loss": 0.3853951096534729, "num_input_tokens_seen": 5528822, "step": 1966, "train_runtime": 6484.3813, "train_tokens_per_second": 852.637 }, { "epoch": 0.15805861909640612, "grad_norm": 2.5564558506011963, "learning_rate": 3.408521303258146e-07, "loss": 0.2719891667366028, "num_input_tokens_seen": 5531908, "step": 1967, "train_runtime": 6487.1286, "train_tokens_per_second": 852.751 }, { "epoch": 0.15813897426625687, "grad_norm": 2.9515230655670166, "learning_rate": 3.308270676691729e-07, "loss": 0.24117296934127808, "num_input_tokens_seen": 5534324, "step": 1968, "train_runtime": 6489.6997, "train_tokens_per_second": 852.786 }, { "epoch": 0.1582193294361076, "grad_norm": 2.9954075813293457, "learning_rate": 3.2080200501253136e-07, "loss": 0.3434407413005829, "num_input_tokens_seen": 5538560, "step": 1969, "train_runtime": 6492.6992, "train_tokens_per_second": 853.044 }, { "epoch": 0.15829968460595834, "grad_norm": 3.227487325668335, "learning_rate": 3.1077694235588974e-07, "loss": 0.2884277403354645, "num_input_tokens_seen": 5541418, "step": 1970, "train_runtime": 6495.4129, "train_tokens_per_second": 853.128 }, { "epoch": 0.15838003977580908, "grad_norm": 2.4196465015411377, "learning_rate": 3.007518796992481e-07, "loss": 0.2532389760017395, "num_input_tokens_seen": 5545778, "step": 1971, "train_runtime": 6498.3665, "train_tokens_per_second": 853.411 }, { "epoch": 0.1584603949456598, "grad_norm": 2.4480080604553223, "learning_rate": 2.9072681704260656e-07, "loss": 0.2545238733291626, "num_input_tokens_seen": 5551620, "step": 1972, "train_runtime": 6501.7389, "train_tokens_per_second": 853.867 }, { "epoch": 0.15854075011551055, "grad_norm": 3.46783185005188, "learning_rate": 2.8070175438596494e-07, "loss": 0.2769492268562317, "num_input_tokens_seen": 5553568, "step": 1973, "train_runtime": 6504.0559, "train_tokens_per_second": 853.862 }, { "epoch": 0.1586211052853613, "grad_norm": 3.1666955947875977, "learning_rate": 2.706766917293233e-07, "loss": 0.33319470286369324, "num_input_tokens_seen": 5555244, "step": 1974, "train_runtime": 6506.3672, "train_tokens_per_second": 853.817 }, { "epoch": 0.15870146045521202, "grad_norm": 2.547426223754883, "learning_rate": 2.606516290726817e-07, "loss": 0.15900322794914246, "num_input_tokens_seen": 5557538, "step": 1975, "train_runtime": 6508.8481, "train_tokens_per_second": 853.844 }, { "epoch": 0.15878181562506277, "grad_norm": 2.592071056365967, "learning_rate": 2.506265664160401e-07, "loss": 0.19326552748680115, "num_input_tokens_seen": 5561404, "step": 1976, "train_runtime": 6511.6617, "train_tokens_per_second": 854.068 }, { "epoch": 0.15886217079491352, "grad_norm": 2.4860169887542725, "learning_rate": 2.406015037593985e-07, "loss": 0.2290671467781067, "num_input_tokens_seen": 5564424, "step": 1977, "train_runtime": 6514.2857, "train_tokens_per_second": 854.188 }, { "epoch": 0.15894252596476427, "grad_norm": 3.48280930519104, "learning_rate": 2.3057644110275693e-07, "loss": 0.39272210001945496, "num_input_tokens_seen": 5566282, "step": 1978, "train_runtime": 6516.6414, "train_tokens_per_second": 854.164 }, { "epoch": 0.159022881134615, "grad_norm": 2.5583174228668213, "learning_rate": 2.205513784461153e-07, "loss": 0.23420238494873047, "num_input_tokens_seen": 5567746, "step": 1979, "train_runtime": 6518.8609, "train_tokens_per_second": 854.098 }, { "epoch": 0.15910323630446574, "grad_norm": 2.5037450790405273, "learning_rate": 2.105263157894737e-07, "loss": 0.25029388070106506, "num_input_tokens_seen": 5570132, "step": 1980, "train_runtime": 6521.3054, "train_tokens_per_second": 854.144 }, { "epoch": 0.1591835914743165, "grad_norm": 2.5811026096343994, "learning_rate": 2.005012531328321e-07, "loss": 0.24229183793067932, "num_input_tokens_seen": 5572656, "step": 1981, "train_runtime": 6534.6671, "train_tokens_per_second": 852.783 }, { "epoch": 0.1592639466441672, "grad_norm": 2.69059419631958, "learning_rate": 1.904761904761905e-07, "loss": 0.26899397373199463, "num_input_tokens_seen": 5575726, "step": 1982, "train_runtime": 6537.3706, "train_tokens_per_second": 852.9 }, { "epoch": 0.15934430181401796, "grad_norm": 3.0376384258270264, "learning_rate": 1.804511278195489e-07, "loss": 0.24955542385578156, "num_input_tokens_seen": 5578290, "step": 1983, "train_runtime": 6539.9954, "train_tokens_per_second": 852.95 }, { "epoch": 0.1594246569838687, "grad_norm": 2.928349494934082, "learning_rate": 1.704260651629073e-07, "loss": 0.33316171169281006, "num_input_tokens_seen": 5581222, "step": 1984, "train_runtime": 6542.6555, "train_tokens_per_second": 853.051 }, { "epoch": 0.15950501215371943, "grad_norm": 3.242380142211914, "learning_rate": 1.6040100250626568e-07, "loss": 0.2893308103084564, "num_input_tokens_seen": 5584084, "step": 1985, "train_runtime": 6545.29, "train_tokens_per_second": 853.145 }, { "epoch": 0.15958536732357018, "grad_norm": 3.3536312580108643, "learning_rate": 1.5037593984962406e-07, "loss": 0.3599575459957123, "num_input_tokens_seen": 5585696, "step": 1986, "train_runtime": 6547.6335, "train_tokens_per_second": 853.086 }, { "epoch": 0.15966572249342093, "grad_norm": 3.1292190551757812, "learning_rate": 1.4035087719298247e-07, "loss": 0.302339106798172, "num_input_tokens_seen": 5588066, "step": 1987, "train_runtime": 6550.1202, "train_tokens_per_second": 853.124 }, { "epoch": 0.15974607766327167, "grad_norm": 2.870380401611328, "learning_rate": 1.3032581453634085e-07, "loss": 0.2822530269622803, "num_input_tokens_seen": 5591122, "step": 1988, "train_runtime": 6552.8238, "train_tokens_per_second": 853.239 }, { "epoch": 0.1598264328331224, "grad_norm": 2.5619113445281982, "learning_rate": 1.2030075187969926e-07, "loss": 0.3257730007171631, "num_input_tokens_seen": 5594242, "step": 1989, "train_runtime": 6555.459, "train_tokens_per_second": 853.372 }, { "epoch": 0.15990678800297314, "grad_norm": 3.3447911739349365, "learning_rate": 1.1027568922305765e-07, "loss": 0.28598421812057495, "num_input_tokens_seen": 5596586, "step": 1990, "train_runtime": 6557.9115, "train_tokens_per_second": 853.41 }, { "epoch": 0.1599871431728239, "grad_norm": 3.302879810333252, "learning_rate": 1.0025062656641605e-07, "loss": 0.2787710130214691, "num_input_tokens_seen": 5602684, "step": 1991, "train_runtime": 6561.7987, "train_tokens_per_second": 853.834 }, { "epoch": 0.1600674983426746, "grad_norm": 2.8427138328552246, "learning_rate": 9.022556390977444e-08, "loss": 0.32147353887557983, "num_input_tokens_seen": 5605566, "step": 1992, "train_runtime": 6564.6308, "train_tokens_per_second": 853.904 }, { "epoch": 0.16014785351252536, "grad_norm": 2.6374619007110596, "learning_rate": 8.020050125313284e-08, "loss": 0.22111660242080688, "num_input_tokens_seen": 5607598, "step": 1993, "train_runtime": 6567.0762, "train_tokens_per_second": 853.896 }, { "epoch": 0.1602282086823761, "grad_norm": 2.8294408321380615, "learning_rate": 7.017543859649123e-08, "loss": 0.17189015448093414, "num_input_tokens_seen": 5613232, "step": 1994, "train_runtime": 6570.466, "train_tokens_per_second": 854.313 }, { "epoch": 0.16030856385222683, "grad_norm": 3.3328566551208496, "learning_rate": 6.015037593984963e-08, "loss": 0.3289675712585449, "num_input_tokens_seen": 5615784, "step": 1995, "train_runtime": 6573.0989, "train_tokens_per_second": 854.359 }, { "epoch": 0.16038891902207758, "grad_norm": 3.014392614364624, "learning_rate": 5.0125313283208025e-08, "loss": 0.25987833738327026, "num_input_tokens_seen": 5618638, "step": 1996, "train_runtime": 6575.6556, "train_tokens_per_second": 854.461 }, { "epoch": 0.16046927419192833, "grad_norm": 3.1398026943206787, "learning_rate": 4.010025062656642e-08, "loss": 0.3443450927734375, "num_input_tokens_seen": 5621372, "step": 1997, "train_runtime": 6578.1939, "train_tokens_per_second": 854.546 }, { "epoch": 0.16054962936177905, "grad_norm": 3.0891308784484863, "learning_rate": 3.0075187969924815e-08, "loss": 0.2890098989009857, "num_input_tokens_seen": 5626454, "step": 1998, "train_runtime": 6581.3013, "train_tokens_per_second": 854.915 }, { "epoch": 0.1606299845316298, "grad_norm": 2.7004165649414062, "learning_rate": 2.005012531328321e-08, "loss": 0.2717441916465759, "num_input_tokens_seen": 5628146, "step": 1999, "train_runtime": 6583.5535, "train_tokens_per_second": 854.88 }, { "epoch": 0.16071033970148055, "grad_norm": 3.3573083877563477, "learning_rate": 1.0025062656641605e-08, "loss": 0.3191164433956146, "num_input_tokens_seen": 5631090, "step": 2000, "train_runtime": 6586.1537, "train_tokens_per_second": 854.989 }, { "epoch": 0.16071033970148055, "eval_loss": 1.0941919088363647, "eval_runtime": 68.8867, "eval_samples_per_second": 14.43, "eval_steps_per_second": 7.215, "num_input_tokens_seen": 5631090, "step": 2000 } ], "logging_steps": 1, "max_steps": 2000, "num_input_tokens_seen": 5631090, "num_train_epochs": 1, "save_steps": 30, "stateful_callbacks": { "TrainerControl": { "args": { "should_epoch_stop": false, "should_evaluate": false, "should_log": false, "should_save": true, "should_training_stop": true }, "attributes": {} } }, "total_flos": 5.751560440271232e+16, "train_batch_size": 2, "trial_name": null, "trial_params": null }