fast-model / training.jsonl
hanswalt's picture
Upload folder using huggingface_hub
57e31be verified
Raw History Blame Contribute Delete
156 kB
{"step": 0, "epoch": 0, "timestamp": "2026-10-11T20:16:38.679Z", "loss": 2.068072557449341, "grad_norm": 2.8421684352263825, "lr": 3.54e-08, "mem": 72.42932653427124, "tps": 7295.450877976608, "tps_per_gpu": 911.931359747076, "mfu": 0.0, "num_tokens_per_step": 88692, "num_label_tokens": 88692}
{"step": 1, "epoch": 0, "timestamp": "2026-10-11T20:16:46.446Z", "loss": 1.335546612739563, "grad_norm": 2.3335007881297183, "lr": 4.0799999999999995e-08, "mem": 87.21395492553711, "tps": 11718.781931142563, "tps_per_gpu": 1464.8477413928204, "mfu": 0.0, "num_tokens_per_step": 91025, "num_label_tokens": 91025}
{"step": 2, "epoch": 0, "timestamp": "2026-10-11T20:16:54.057Z", "loss": 1.3774056434631348, "grad_norm": 2.3126628787874517, "lr": 4.6200000000000003e-08, "mem": 86.13407897949219, "tps": 11934.535979103448, "tps_per_gpu": 1491.816997387931, "mfu": 0.0, "num_tokens_per_step": 90818, "num_label_tokens": 90818}
{"step": 3, "epoch": 0, "timestamp": "2026-10-11T20:17:01.543Z", "loss": 1.1758997440338135, "grad_norm": 2.252929464361691, "lr": 5.16e-08, "mem": 78.60670375823975, "tps": 12186.501337616815, "tps_per_gpu": 1523.312667202102, "mfu": 0.0, "num_tokens_per_step": 91187, "num_label_tokens": 91187}
{"step": 4, "epoch": 0, "timestamp": "2026-10-11T20:17:09.052Z", "loss": 1.1946529150009155, "grad_norm": 2.1597923558108585, "lr": 5.6999999999999994e-08, "mem": 81.84387254714966, "tps": 12233.725911935693, "tps_per_gpu": 1529.2157389919616, "mfu": 0.0, "num_tokens_per_step": 91884, "num_label_tokens": 91884}
{"step": 5, "epoch": 0, "timestamp": "2026-10-11T20:17:20.766Z", "loss": 1.243971347808838, "grad_norm": 2.2294260984792156, "lr": 6.24e-08, "mem": 79.694167137146, "tps": 7901.271827609335, "tps_per_gpu": 987.6589784511668, "mfu": 0.0, "num_tokens_per_step": 92558, "num_label_tokens": 92558}
{"step": 6, "epoch": 0, "timestamp": "2026-10-11T20:17:28.313Z", "loss": 1.2443562746047974, "grad_norm": 2.292689676102313, "lr": 6.78e-08, "mem": 86.13408088684082, "tps": 12211.422717597638, "tps_per_gpu": 1526.4278396997047, "mfu": 0.0, "num_tokens_per_step": 92159, "num_label_tokens": 92159}
{"step": 7, "epoch": 0, "timestamp": "2026-10-11T20:17:35.943Z", "loss": 1.3890854120254517, "grad_norm": 2.258811494514446, "lr": 7.32e-08, "mem": 87.21395778656006, "tps": 12438.362302275165, "tps_per_gpu": 1554.7952877843957, "mfu": 0.0, "num_tokens_per_step": 94919, "num_label_tokens": 94919}
{"step": 8, "epoch": 0, "timestamp": "2026-10-11T20:17:43.575Z", "loss": 1.3164693117141724, "grad_norm": 2.2962064850661204, "lr": 7.86e-08, "mem": 86.13408184051514, "tps": 12156.870844802464, "tps_per_gpu": 1519.608855600308, "mfu": 0.0, "num_tokens_per_step": 92787, "num_label_tokens": 92787}
{"step": 9, "epoch": 0, "timestamp": "2026-10-11T20:17:51.090Z", "loss": 1.4041621685028076, "grad_norm": 2.378738438811575, "lr": 8.4e-08, "mem": 82.9194540977478, "tps": 12482.54982802747, "tps_per_gpu": 1560.3187285034337, "mfu": 0.0, "num_tokens_per_step": 93787, "num_label_tokens": 93787}
{"step": 10, "epoch": 0, "timestamp": "2026-10-11T20:17:58.735Z", "loss": 1.1956603527069092, "grad_norm": 2.215547579689906, "lr": 8.939999999999999e-08, "mem": 85.0511736869812, "tps": 12256.1684450611, "tps_per_gpu": 1532.0210556326374, "mfu": 0.0, "num_tokens_per_step": 93705, "num_label_tokens": 93705}
{"step": 11, "epoch": 0, "timestamp": "2026-10-11T20:18:06.279Z", "loss": 1.2307325601577759, "grad_norm": 2.3767003487352656, "lr": 9.48e-08, "mem": 81.84387588500977, "tps": 12190.89786784091, "tps_per_gpu": 1523.8622334801137, "mfu": 0.0, "num_tokens_per_step": 91939, "num_label_tokens": 91939}
{"step": 12, "epoch": 0, "timestamp": "2026-10-11T20:18:13.791Z", "loss": 1.3338472843170166, "grad_norm": 2.2465245216882836, "lr": 1.0020000000000001e-07, "mem": 80.76884269714355, "tps": 12566.122353525987, "tps_per_gpu": 1570.7652941907484, "mfu": 0.0, "num_tokens_per_step": 94424, "num_label_tokens": 94424}
{"step": 13, "epoch": 0, "timestamp": "2026-10-11T20:18:21.289Z", "loss": 1.7403279542922974, "grad_norm": 2.6888544829721748, "lr": 1.0559999999999999e-07, "mem": 83.98414659500122, "tps": 12363.862689361391, "tps_per_gpu": 1545.4828361701739, "mfu": 0.0, "num_tokens_per_step": 92695, "num_label_tokens": 92695}
{"step": 14, "epoch": 0, "timestamp": "2026-10-11T20:18:28.840Z", "loss": 1.5906370878219604, "grad_norm": 2.7640628546839787, "lr": 1.11e-07, "mem": 83.98414707183838, "tps": 11941.018458018323, "tps_per_gpu": 1492.6273072522904, "mfu": 0.0, "num_tokens_per_step": 90149, "num_label_tokens": 90149}
{"step": 15, "epoch": 0, "timestamp": "2026-10-11T20:18:36.400Z", "loss": 1.4349256753921509, "grad_norm": 3.1229665213158415, "lr": 1.1640000000000001e-07, "mem": 86.13408517837524, "tps": 11734.529912055768, "tps_per_gpu": 1466.816239006971, "mfu": 0.0, "num_tokens_per_step": 88714, "num_label_tokens": 88714}
{"step": 16, "epoch": 0, "timestamp": "2026-10-11T20:18:43.917Z", "loss": 1.4047420024871826, "grad_norm": 2.887860167521192, "lr": 1.218e-07, "mem": 85.05117654800415, "tps": 12225.194957943206, "tps_per_gpu": 1528.1493697429007, "mfu": 0.0, "num_tokens_per_step": 91922, "num_label_tokens": 91922}
{"step": 17, "epoch": 0, "timestamp": "2026-10-11T20:18:51.437Z", "loss": 1.5150606632232666, "grad_norm": 2.9233280110710793, "lr": 1.272e-07, "mem": 85.05117702484131, "tps": 11412.769715401157, "tps_per_gpu": 1426.5962144251446, "mfu": 0.0, "num_tokens_per_step": 85825, "num_label_tokens": 85825}
{"step": 18, "epoch": 0, "timestamp": "2026-10-11T20:18:58.517Z", "loss": 1.977741003036499, "grad_norm": 2.7112383276873624, "lr": 1.326e-07, "mem": 86.13553190231323, "tps": 12684.00727219714, "tps_per_gpu": 1585.5009090246424, "mfu": 0.0, "num_tokens_per_step": 89792, "num_label_tokens": 89792}
{"step": 19, "epoch": 0, "timestamp": "2026-10-11T20:19:06.057Z", "loss": 1.9578453302383423, "grad_norm": 2.9513828532793065, "lr": 1.38e-07, "mem": 82.91945886611938, "tps": 12237.093691061531, "tps_per_gpu": 1529.6367113826914, "mfu": 0.0, "num_tokens_per_step": 92278, "num_label_tokens": 92278}
{"step": 20, "epoch": 0, "timestamp": "2026-10-11T20:19:13.062Z", "loss": 1.9747538566589355, "grad_norm": 2.8507688594047833, "lr": 1.434e-07, "mem": 86.13553285598755, "tps": 12091.851380279457, "tps_per_gpu": 1511.481422534932, "mfu": 0.0, "num_tokens_per_step": 84699, "num_label_tokens": 84699}
{"step": 21, "epoch": 0, "timestamp": "2026-10-11T20:19:20.659Z", "loss": 1.377044439315796, "grad_norm": 2.7379170415118588, "lr": 1.4879999999999998e-07, "mem": 80.76884698867798, "tps": 11604.623366242575, "tps_per_gpu": 1450.5779207803218, "mfu": 0.0, "num_tokens_per_step": 88164, "num_label_tokens": 88164}
{"step": 22, "epoch": 0, "timestamp": "2026-10-11T20:19:28.232Z", "loss": 1.411926507949829, "grad_norm": 2.6433199755924743, "lr": 1.542e-07, "mem": 86.13408851623535, "tps": 11779.78050150911, "tps_per_gpu": 1472.4725626886388, "mfu": 0.0, "num_tokens_per_step": 89197, "num_label_tokens": 89197}
{"step": 23, "epoch": 0, "timestamp": "2026-10-11T20:19:35.222Z", "loss": 1.4065980911254883, "grad_norm": 2.6760591510710485, "lr": 1.5960000000000003e-07, "mem": 81.84388160705566, "tps": 12415.8872551017, "tps_per_gpu": 1551.9859068877124, "mfu": 0.0, "num_tokens_per_step": 86781, "num_label_tokens": 86781}
{"step": 24, "epoch": 0, "timestamp": "2026-10-11T20:19:42.667Z", "loss": 2.0188755989074707, "grad_norm": 2.65446318364701, "lr": 1.65e-07, "mem": 85.05118036270142, "tps": 12211.194868594146, "tps_per_gpu": 1526.3993585742683, "mfu": 0.0, "num_tokens_per_step": 90912, "num_label_tokens": 90912}
{"step": 25, "epoch": 0, "timestamp": "2026-10-11T20:19:54.201Z", "loss": 1.4495446681976318, "grad_norm": 2.3410368116422338, "lr": 1.7040000000000002e-07, "mem": 87.2139663696289, "tps": 8210.896072817588, "tps_per_gpu": 1026.3620091021985, "mfu": 0.0, "num_tokens_per_step": 94722, "num_label_tokens": 94722}
{"step": 26, "epoch": 0, "timestamp": "2026-10-11T20:20:01.796Z", "loss": 1.2621926069259644, "grad_norm": 2.2128333833408984, "lr": 1.7580000000000003e-07, "mem": 82.91946220397949, "tps": 12043.042782460881, "tps_per_gpu": 1505.3803478076102, "mfu": 0.0, "num_tokens_per_step": 91433, "num_label_tokens": 91433}
{"step": 27, "epoch": 0, "timestamp": "2026-10-11T20:20:09.312Z", "loss": 1.275265097618103, "grad_norm": 2.377626624328147, "lr": 1.812e-07, "mem": 87.21396732330322, "tps": 11771.906733901089, "tps_per_gpu": 1471.4883417376361, "mfu": 0.0, "num_tokens_per_step": 88508, "num_label_tokens": 88508}
{"step": 28, "epoch": 0, "timestamp": "2026-10-11T20:20:16.865Z", "loss": 1.3768572807312012, "grad_norm": 2.3033692362475833, "lr": 1.866e-07, "mem": 83.9841537475586, "tps": 12320.914337425156, "tps_per_gpu": 1540.1142921781445, "mfu": 0.0, "num_tokens_per_step": 93063, "num_label_tokens": 93063}
{"step": 29, "epoch": 0, "timestamp": "2026-10-11T20:20:24.409Z", "loss": 1.2727131843566895, "grad_norm": 2.2725402565892514, "lr": 1.92e-07, "mem": 83.98415422439575, "tps": 12306.359377546496, "tps_per_gpu": 1538.294922193312, "mfu": 0.0, "num_tokens_per_step": 92834, "num_label_tokens": 92834}
{"step": 30, "epoch": 0, "timestamp": "2026-10-11T20:20:31.947Z", "loss": 1.2157642841339111, "grad_norm": 2.336761037717293, "lr": 1.9739999999999998e-07, "mem": 86.13409233093262, "tps": 12490.018003954892, "tps_per_gpu": 1561.2522504943615, "mfu": 0.0, "num_tokens_per_step": 94158, "num_label_tokens": 94158}
{"step": 31, "epoch": 0, "timestamp": "2026-10-11T20:20:39.036Z", "loss": 1.3215315341949463, "grad_norm": 2.319588954776086, "lr": 2.0280000000000002e-07, "mem": 86.13553810119629, "tps": 12482.412444644257, "tps_per_gpu": 1560.301555580532, "mfu": 0.0, "num_tokens_per_step": 88479, "num_label_tokens": 88479}
{"step": 32, "epoch": 0, "timestamp": "2026-10-11T20:20:46.626Z", "loss": 1.1922903060913086, "grad_norm": 2.206827089852109, "lr": 2.0820000000000002e-07, "mem": 87.21396970748901, "tps": 12090.3370115211, "tps_per_gpu": 1511.2921264401375, "mfu": 0.0, "num_tokens_per_step": 91775, "num_label_tokens": 91775}
{"step": 33, "epoch": 0, "timestamp": "2026-10-11T20:20:54.169Z", "loss": 1.334112524986267, "grad_norm": 2.2894995825896767, "lr": 2.136e-07, "mem": 80.76885271072388, "tps": 12186.3688617456, "tps_per_gpu": 1523.2961077182, "mfu": 0.0, "num_tokens_per_step": 91906, "num_label_tokens": 91906}
{"step": 34, "epoch": 0, "timestamp": "2026-10-11T20:21:01.713Z", "loss": 1.3083481788635254, "grad_norm": 2.326645288762346, "lr": 2.1900000000000002e-07, "mem": 83.98415660858154, "tps": 12386.128778488255, "tps_per_gpu": 1548.266097311032, "mfu": 0.0, "num_tokens_per_step": 93434, "num_label_tokens": 93434}
{"step": 35, "epoch": 0, "timestamp": "2026-10-11T20:21:08.698Z", "loss": 1.3682105541229248, "grad_norm": 2.795546118811681, "lr": 2.244e-07, "mem": 82.91946649551392, "tps": 12289.412991015186, "tps_per_gpu": 1536.1766238768982, "mfu": 0.0, "num_tokens_per_step": 85841, "num_label_tokens": 85841}
{"step": 36, "epoch": 0, "timestamp": "2026-10-11T20:21:16.228Z", "loss": 1.4581153392791748, "grad_norm": 2.985317597304623, "lr": 2.2980000000000003e-07, "mem": 86.13409519195557, "tps": 11955.849433803738, "tps_per_gpu": 1494.4811792254673, "mfu": 0.0, "num_tokens_per_step": 90054, "num_label_tokens": 90054}
{"step": 37, "epoch": 0, "timestamp": "2026-10-11T20:21:23.793Z", "loss": 1.4990702867507935, "grad_norm": 2.629046223363658, "lr": 2.3520000000000001e-07, "mem": 86.13409566879272, "tps": 12185.600220620841, "tps_per_gpu": 1523.2000275776052, "mfu": 0.0, "num_tokens_per_step": 92158, "num_label_tokens": 92158}
{"step": 38, "epoch": 0, "timestamp": "2026-10-11T20:21:31.360Z", "loss": 1.2494808435440063, "grad_norm": 2.257167970840762, "lr": 2.406e-07, "mem": 87.21397256851196, "tps": 12424.380600503542, "tps_per_gpu": 1553.0475750629428, "mfu": 0.0, "num_tokens_per_step": 94031, "num_label_tokens": 94031}
{"step": 39, "epoch": 0, "timestamp": "2026-10-11T20:21:38.964Z", "loss": 1.3288094997406006, "grad_norm": 2.3239857651324844, "lr": 2.46e-07, "mem": 83.98415899276733, "tps": 12255.717000370383, "tps_per_gpu": 1531.964625046298, "mfu": 0.0, "num_tokens_per_step": 93184, "num_label_tokens": 93184}
{"step": 40, "epoch": 0, "timestamp": "2026-10-11T20:21:50.510Z", "loss": 1.1980632543563843, "grad_norm": 2.210060823701554, "lr": 2.514e-07, "mem": 86.1340970993042, "tps": 8117.027952479485, "tps_per_gpu": 1014.6284940599356, "mfu": 0.0, "num_tokens_per_step": 93737, "num_label_tokens": 93737}
{"step": 41, "epoch": 0, "timestamp": "2026-10-11T20:21:57.938Z", "loss": 1.3710070848464966, "grad_norm": 2.498566833636716, "lr": 2.5679999999999997e-07, "mem": 80.76885652542114, "tps": 11857.337597822614, "tps_per_gpu": 1482.1671997278268, "mfu": 0.0, "num_tokens_per_step": 88052, "num_label_tokens": 88052}
{"step": 42, "epoch": 0, "timestamp": "2026-10-11T20:22:05.426Z", "loss": 1.494611382484436, "grad_norm": 2.851686152277163, "lr": 2.622e-07, "mem": 80.7688570022583, "tps": 12082.102087989013, "tps_per_gpu": 1510.2627609986266, "mfu": 0.0, "num_tokens_per_step": 90494, "num_label_tokens": 90494}
{"step": 43, "epoch": 0, "timestamp": "2026-10-11T20:22:12.940Z", "loss": 1.4463214874267578, "grad_norm": 2.7334601706897486, "lr": 2.676e-07, "mem": 85.05118942260742, "tps": 12052.749318569646, "tps_per_gpu": 1506.5936648212057, "mfu": 0.0, "num_tokens_per_step": 90558, "num_label_tokens": 90558}
{"step": 44, "epoch": 0, "timestamp": "2026-10-11T20:22:20.061Z", "loss": 1.7825859785079956, "grad_norm": 2.6376777418714026, "lr": 2.7299999999999997e-07, "mem": 86.13554430007935, "tps": 12544.02866302575, "tps_per_gpu": 1568.0035828782188, "mfu": 0.0, "num_tokens_per_step": 89301, "num_label_tokens": 89301}
{"step": 45, "epoch": 0, "timestamp": "2026-10-11T20:22:27.543Z", "loss": 1.919927716255188, "grad_norm": 2.870684565431075, "lr": 2.7839999999999995e-07, "mem": 83.98416185379028, "tps": 11557.428480581963, "tps_per_gpu": 1444.6785600727453, "mfu": 0.0, "num_tokens_per_step": 86484, "num_label_tokens": 86484}
{"step": 46, "epoch": 0, "timestamp": "2026-10-11T20:22:34.994Z", "loss": 1.7424097061157227, "grad_norm": 2.599924163499551, "lr": 2.838e-07, "mem": 87.21397638320923, "tps": 11843.380408975348, "tps_per_gpu": 1480.4225511219186, "mfu": 0.0, "num_tokens_per_step": 88236, "num_label_tokens": 88236}
{"step": 47, "epoch": 0, "timestamp": "2026-10-11T20:22:42.537Z", "loss": 1.1943023204803467, "grad_norm": 2.334971519621497, "lr": 2.892e-07, "mem": 86.1341004371643, "tps": 12568.413814516329, "tps_per_gpu": 1571.051726814541, "mfu": 0.0, "num_tokens_per_step": 94795, "num_label_tokens": 94795}
{"step": 48, "epoch": 0, "timestamp": "2026-10-11T20:22:50.110Z", "loss": 1.3159751892089844, "grad_norm": 2.256069596097778, "lr": 2.946e-07, "mem": 86.13410091400146, "tps": 12352.299328536354, "tps_per_gpu": 1544.0374160670442, "mfu": 0.0, "num_tokens_per_step": 93557, "num_label_tokens": 93557}
{"step": 49, "epoch": 0, "timestamp": "2026-10-11T20:22:57.616Z", "loss": 1.3156242370605469, "grad_norm": 2.383777199470824, "lr": 3e-07, "mem": 82.91947317123413, "tps": 12140.142406833003, "tps_per_gpu": 1517.5178008541254, "mfu": 0.0, "num_tokens_per_step": 91114, "num_label_tokens": 91114}
{"step": 50, "epoch": 0, "timestamp": "2026-10-11T20:23:04.666Z", "loss": 1.4647108316421509, "grad_norm": 2.7838043852702032, "lr": 2.9999997986000596e-07, "mem": 83.98416423797607, "tps": 12513.286531780688, "tps_per_gpu": 1564.160816472586, "mfu": 0.0, "num_tokens_per_step": 88236, "num_label_tokens": 88236}
{"step": 51, "epoch": 0, "timestamp": "2026-10-11T20:23:11.762Z", "loss": 1.641648292541504, "grad_norm": 2.8369045903703163, "lr": 2.9999991944003204e-07, "mem": 82.91947412490845, "tps": 12451.39625988162, "tps_per_gpu": 1556.4245324852025, "mfu": 0.0, "num_tokens_per_step": 88328, "num_label_tokens": 88328}
{"step": 52, "epoch": 0, "timestamp": "2026-10-11T20:23:19.267Z", "loss": 1.8492529392242432, "grad_norm": 2.6940266988870176, "lr": 2.9999981874010246e-07, "mem": 87.21397924423218, "tps": 11933.089755134857, "tps_per_gpu": 1491.6362193918571, "mfu": 0.0, "num_tokens_per_step": 89586, "num_label_tokens": 89586}
{"step": 53, "epoch": 0, "timestamp": "2026-10-11T20:23:26.860Z", "loss": 1.6814225912094116, "grad_norm": 2.5462276861733417, "lr": 2.999996777602579e-07, "mem": 83.98416566848755, "tps": 12152.714645667176, "tps_per_gpu": 1519.089330708397, "mfu": 0.0, "num_tokens_per_step": 92260, "num_label_tokens": 92260}
{"step": 54, "epoch": 0, "timestamp": "2026-10-11T20:23:34.405Z", "loss": 1.2602039575576782, "grad_norm": 2.216008867795225, "lr": 2.999994965005551e-07, "mem": 87.2139801979065, "tps": 12200.653591047034, "tps_per_gpu": 1525.0816988808792, "mfu": 0.0, "num_tokens_per_step": 92072, "num_label_tokens": 92072}
{"step": 55, "epoch": 0, "timestamp": "2026-10-11T20:23:41.973Z", "loss": 1.330971598625183, "grad_norm": 2.285791610476865, "lr": 2.999992749610671e-07, "mem": 86.13410425186157, "tps": 12401.02338726507, "tps_per_gpu": 1550.1279234081337, "mfu": 0.0, "num_tokens_per_step": 93846, "num_label_tokens": 93846}
{"step": 56, "epoch": 0, "timestamp": "2026-10-11T20:23:56.690Z", "loss": 1.2246201038360596, "grad_norm": 2.233455399056461, "lr": 2.99999013141883e-07, "mem": 87.21398115158081, "tps": 6398.03652978144, "tps_per_gpu": 799.75456622268, "mfu": 0.0, "num_tokens_per_step": 94155, "num_label_tokens": 94155}
{"step": 57, "epoch": 0, "timestamp": "2026-10-11T20:24:04.168Z", "loss": 1.2920657396316528, "grad_norm": 2.3045725347857777, "lr": 2.9999871104310844e-07, "mem": 87.21398162841797, "tps": 11866.671717436313, "tps_per_gpu": 1483.3339646795391, "mfu": 0.0, "num_tokens_per_step": 88734, "num_label_tokens": 88734}
{"step": 58, "epoch": 0, "timestamp": "2026-10-11T20:24:11.326Z", "loss": 1.271222710609436, "grad_norm": 2.273995943678822, "lr": 2.99998368664865e-07, "mem": 85.0511965751648, "tps": 12991.517150898313, "tps_per_gpu": 1623.9396438622891, "mfu": 0.0, "num_tokens_per_step": 92998, "num_label_tokens": 92998}
{"step": 59, "epoch": 0, "timestamp": "2026-10-11T20:24:18.823Z", "loss": 1.28052818775177, "grad_norm": 2.283177296655609, "lr": 2.999979860072906e-07, "mem": 85.05119705200195, "tps": 11976.81289503017, "tps_per_gpu": 1497.1016118787713, "mfu": 0.0, "num_tokens_per_step": 89793, "num_label_tokens": 89793}
{"step": 60, "epoch": 0, "timestamp": "2026-10-11T20:24:26.412Z", "loss": 1.2395806312561035, "grad_norm": 2.2836458947522593, "lr": 2.9999756307053946e-07, "mem": 87.21398305892944, "tps": 12236.535125141561, "tps_per_gpu": 1529.5668906426952, "mfu": 0.0, "num_tokens_per_step": 92876, "num_label_tokens": 92876}
{"step": 61, "epoch": 0, "timestamp": "2026-10-11T20:24:33.891Z", "loss": 1.2408634424209595, "grad_norm": 2.2667061797043004, "lr": 2.9999709985478187e-07, "mem": 87.2139835357666, "tps": 12054.231246980398, "tps_per_gpu": 1506.7789058725498, "mfu": 0.0, "num_tokens_per_step": 90154, "num_label_tokens": 90154}
{"step": 62, "epoch": 0, "timestamp": "2026-10-11T20:24:41.446Z", "loss": 1.2469303607940674, "grad_norm": 2.258545853699972, "lr": 2.999965963602044e-07, "mem": 86.13410758972168, "tps": 12483.415629987587, "tps_per_gpu": 1560.4269537484483, "mfu": 0.0, "num_tokens_per_step": 94282, "num_label_tokens": 94282}
{"step": 63, "epoch": 0, "timestamp": "2026-10-11T20:24:49.046Z", "loss": 1.9804461002349854, "grad_norm": 2.8675560354336382, "lr": 2.999960525870098e-07, "mem": 87.21398448944092, "tps": 12179.665866794703, "tps_per_gpu": 1522.458233349338, "mfu": 0.0, "num_tokens_per_step": 92601, "num_label_tokens": 92601}
{"step": 64, "epoch": 0, "timestamp": "2026-10-11T20:24:56.447Z", "loss": 1.350180745124817, "grad_norm": 2.330386636060657, "lr": 2.9999546853541727e-07, "mem": 87.21398496627808, "tps": 11778.93695231105, "tps_per_gpu": 1472.3671190388814, "mfu": 0.0, "num_tokens_per_step": 87142, "num_label_tokens": 87142}
{"step": 65, "epoch": 0, "timestamp": "2026-10-11T20:25:03.523Z", "loss": 1.18169105052948, "grad_norm": 2.2265458055453315, "lr": 2.9999484420566195e-07, "mem": 82.91948080062866, "tps": 12849.651739070776, "tps_per_gpu": 1606.206467383847, "mfu": 0.0, "num_tokens_per_step": 90974, "num_label_tokens": 90974}
{"step": 66, "epoch": 0, "timestamp": "2026-10-11T20:25:11.054Z", "loss": 1.384746789932251, "grad_norm": 2.359101991802221, "lr": 2.9999417959799535e-07, "mem": 80.7688684463501, "tps": 12023.829418878662, "tps_per_gpu": 1502.9786773598328, "mfu": 0.0, "num_tokens_per_step": 90532, "num_label_tokens": 90532}
{"step": 67, "epoch": 0, "timestamp": "2026-10-11T20:25:18.140Z", "loss": 1.1693288087844849, "grad_norm": 2.2418145066826662, "lr": 2.9999347471268513e-07, "mem": 86.13555526733398, "tps": 12700.237192537166, "tps_per_gpu": 1587.5296490671458, "mfu": 0.0, "num_tokens_per_step": 90000, "num_label_tokens": 90000}
{"step": 68, "epoch": 0, "timestamp": "2026-10-11T20:25:25.748Z", "loss": 1.2916594743728638, "grad_norm": 2.3260283215844026, "lr": 2.9999272955001526e-07, "mem": 86.13411045074463, "tps": 11724.224027450464, "tps_per_gpu": 1465.528003431308, "mfu": 0.0, "num_tokens_per_step": 89179, "num_label_tokens": 89179}
{"step": 69, "epoch": 0, "timestamp": "2026-10-11T20:25:33.351Z", "loss": 1.2323424816131592, "grad_norm": 2.1626666755635497, "lr": 2.999919441102859e-07, "mem": 85.05120182037354, "tps": 12657.293712859664, "tps_per_gpu": 1582.161714107458, "mfu": 0.0, "num_tokens_per_step": 96232, "num_label_tokens": 96232}
{"step": 70, "epoch": 0, "timestamp": "2026-10-11T20:25:40.871Z", "loss": 1.2737373113632202, "grad_norm": 2.245433258864425, "lr": 2.999911183938134e-07, "mem": 79.69419813156128, "tps": 12483.246543133573, "tps_per_gpu": 1560.4058178916966, "mfu": 0.0, "num_tokens_per_step": 93890, "num_label_tokens": 93890}
{"step": 71, "epoch": 0, "timestamp": "2026-10-11T20:25:48.436Z", "loss": 1.2664002180099487, "grad_norm": 2.2515392532120484, "lr": 2.9999025240093044e-07, "mem": 83.9841742515564, "tps": 12331.03515348265, "tps_per_gpu": 1541.3793941853312, "mfu": 0.0, "num_tokens_per_step": 93278, "num_label_tokens": 93278}
{"step": 72, "epoch": 0, "timestamp": "2026-10-11T20:25:56.063Z", "loss": 1.3757957220077515, "grad_norm": 2.3436608276494284, "lr": 2.9998934613198566e-07, "mem": 87.21398878097534, "tps": 12495.370414067906, "tps_per_gpu": 1561.9213017584882, "mfu": 0.0, "num_tokens_per_step": 95309, "num_label_tokens": 95309}
{"step": 73, "epoch": 0, "timestamp": "2026-10-11T20:26:07.412Z", "loss": 1.3267561197280884, "grad_norm": 2.328148983051962, "lr": 2.999883995873443e-07, "mem": 77.54820823669434, "tps": 8070.30000549251, "tps_per_gpu": 1008.7875006865637, "mfu": 0.0, "num_tokens_per_step": 91584, "num_label_tokens": 91584}
{"step": 74, "epoch": 0, "timestamp": "2026-10-11T20:26:15.002Z", "loss": 1.1722583770751953, "grad_norm": 2.246710514066396, "lr": 2.999874127673875e-07, "mem": 86.13411331176758, "tps": 12166.331146447359, "tps_per_gpu": 1520.7913933059199, "mfu": 0.0, "num_tokens_per_step": 92340, "num_label_tokens": 92340}
{"step": 75, "epoch": 0, "timestamp": "2026-10-11T20:26:22.582Z", "loss": 1.2920678853988647, "grad_norm": 2.206353637581207, "lr": 2.999863856725128e-07, "mem": 85.05120468139648, "tps": 12221.91277916638, "tps_per_gpu": 1527.7390973957974, "mfu": 0.0, "num_tokens_per_step": 92643, "num_label_tokens": 92643}
{"step": 76, "epoch": 0, "timestamp": "2026-10-11T20:26:30.037Z", "loss": 1.3447068929672241, "grad_norm": 2.6232513970074858, "lr": 2.999853183031339e-07, "mem": 82.9194860458374, "tps": 12174.478214254801, "tps_per_gpu": 1521.8097767818501, "mfu": 0.0, "num_tokens_per_step": 90791, "num_label_tokens": 90791}
{"step": 77, "epoch": 0, "timestamp": "2026-10-11T20:26:37.552Z", "loss": 1.4442217350006104, "grad_norm": 2.9124562373014395, "lr": 2.999842106596808e-07, "mem": 77.54821014404297, "tps": 11868.602809429965, "tps_per_gpu": 1483.5753511787457, "mfu": 0.0, "num_tokens_per_step": 89165, "num_label_tokens": 89165}
{"step": 78, "epoch": 0, "timestamp": "2026-10-11T20:26:45.105Z", "loss": 1.5545052289962769, "grad_norm": 2.966372619381236, "lr": 2.9998306274259956e-07, "mem": 83.9841775894165, "tps": 11958.629416493943, "tps_per_gpu": 1494.8286770617428, "mfu": 0.0, "num_tokens_per_step": 90313, "num_label_tokens": 90313}
{"step": 79, "epoch": 0, "timestamp": "2026-10-11T20:26:52.602Z", "loss": 1.479148507118225, "grad_norm": 2.4858956341531027, "lr": 2.999818745523526e-07, "mem": 80.76887464523315, "tps": 11807.315431642635, "tps_per_gpu": 1475.9144289553294, "mfu": 0.0, "num_tokens_per_step": 88535, "num_label_tokens": 88535}
{"step": 80, "epoch": 0, "timestamp": "2026-10-11T20:27:00.240Z", "loss": 1.3046467304229736, "grad_norm": 2.284553349812241, "lr": 2.999806460894185e-07, "mem": 85.05120706558228, "tps": 12089.50681727443, "tps_per_gpu": 1511.1883521593038, "mfu": 0.0, "num_tokens_per_step": 92324, "num_label_tokens": 92324}
{"step": 81, "epoch": 0, "timestamp": "2026-10-11T20:27:07.722Z", "loss": 1.2953006029129028, "grad_norm": 2.3461692019293388, "lr": 2.9997937735429216e-07, "mem": 87.21399307250977, "tps": 11912.94511381035, "tps_per_gpu": 1489.1181392262938, "mfu": 0.0, "num_tokens_per_step": 89155, "num_label_tokens": 89155}
{"step": 82, "epoch": 0, "timestamp": "2026-10-11T20:27:15.237Z", "loss": 1.4293768405914307, "grad_norm": 2.671009408513967, "lr": 2.999780683474845e-07, "mem": 77.54821252822876, "tps": 12145.63623031754, "tps_per_gpu": 1518.2045287896924, "mfu": 0.0, "num_tokens_per_step": 91250, "num_label_tokens": 91250}
{"step": 83, "epoch": 0, "timestamp": "2026-10-11T20:27:22.774Z", "loss": 1.497978925704956, "grad_norm": 2.7044041908996777, "lr": 2.9997671906952296e-07, "mem": 87.21399402618408, "tps": 12033.003643178135, "tps_per_gpu": 1504.125455397267, "mfu": 0.0, "num_tokens_per_step": 90703, "num_label_tokens": 90703}
{"step": 84, "epoch": 0, "timestamp": "2026-10-11T20:27:30.416Z", "loss": 1.5420668125152588, "grad_norm": 2.772372301853594, "lr": 2.999753295209509e-07, "mem": 85.05120897293091, "tps": 12329.886792481295, "tps_per_gpu": 1541.2358490601619, "mfu": 0.0, "num_tokens_per_step": 94214, "num_label_tokens": 94214}
{"step": 85, "epoch": 0, "timestamp": "2026-10-11T20:27:37.979Z", "loss": 1.3109986782073975, "grad_norm": 2.369110286796, "lr": 2.999738997023281e-07, "mem": 78.60674285888672, "tps": 11633.421400159057, "tps_per_gpu": 1454.1776750198821, "mfu": 0.0, "num_tokens_per_step": 87988, "num_label_tokens": 87988}
{"step": 86, "epoch": 0, "timestamp": "2026-10-11T20:27:45.181Z", "loss": 1.3240423202514648, "grad_norm": 2.183040688388079, "lr": 2.999724296142304e-07, "mem": 85.05120992660522, "tps": 13294.642228328968, "tps_per_gpu": 1661.830278541121, "mfu": 0.0, "num_tokens_per_step": 95750, "num_label_tokens": 95750}
{"step": 87, "epoch": 0, "timestamp": "2026-10-11T20:27:52.254Z", "loss": 1.2444791793823242, "grad_norm": 2.271789659341812, "lr": 2.9997091925725e-07, "mem": 79.69420623779297, "tps": 13105.146105425712, "tps_per_gpu": 1638.143263178214, "mfu": 0.0, "num_tokens_per_step": 92699, "num_label_tokens": 92699}
{"step": 88, "epoch": 0, "timestamp": "2026-10-11T20:27:59.777Z", "loss": 1.354720950126648, "grad_norm": 2.5018327070740685, "lr": 2.9996936863199536e-07, "mem": 83.98418235778809, "tps": 12151.394983949529, "tps_per_gpu": 1518.9243729936911, "mfu": 0.0, "num_tokens_per_step": 91411, "num_label_tokens": 91411}
{"step": 89, "epoch": 0, "timestamp": "2026-10-11T20:28:07.286Z", "loss": 1.4245935678482056, "grad_norm": 2.7108353304001014, "lr": 2.9996777773909087e-07, "mem": 80.76887941360474, "tps": 11998.527148952562, "tps_per_gpu": 1499.8158936190703, "mfu": 0.0, "num_tokens_per_step": 90105, "num_label_tokens": 90105}
{"step": 90, "epoch": 0, "timestamp": "2026-10-11T20:28:18.123Z", "loss": 1.4921514987945557, "grad_norm": 2.8535907716122293, "lr": 2.9996614657917757e-07, "mem": 86.13412094116211, "tps": 8354.039144147635, "tps_per_gpu": 1044.2548930184544, "mfu": 0.0, "num_tokens_per_step": 90523, "num_label_tokens": 90523}
{"step": 91, "epoch": 0, "timestamp": "2026-10-11T20:28:25.675Z", "loss": 1.236900806427002, "grad_norm": 2.2650563604521388, "lr": 2.999644751529123e-07, "mem": 87.21399784088135, "tps": 12317.241442373534, "tps_per_gpu": 1539.6551802966917, "mfu": 0.0, "num_tokens_per_step": 93038, "num_label_tokens": 93038}
{"step": 92, "epoch": 0, "timestamp": "2026-10-11T20:28:33.238Z", "loss": 1.272038459777832, "grad_norm": 2.2695481663115658, "lr": 2.9996276346096847e-07, "mem": 86.13412189483643, "tps": 12062.918467222878, "tps_per_gpu": 1507.8648084028598, "mfu": 0.0, "num_tokens_per_step": 91215, "num_label_tokens": 91215}
{"step": 93, "epoch": 0, "timestamp": "2026-10-11T20:28:40.791Z", "loss": 1.205022931098938, "grad_norm": 2.3177872330579072, "lr": 2.9996101150403543e-07, "mem": 80.76888132095337, "tps": 11847.356822752457, "tps_per_gpu": 1480.9196028440572, "mfu": 0.0, "num_tokens_per_step": 89479, "num_label_tokens": 89479}
{"step": 94, "epoch": 0, "timestamp": "2026-10-11T20:28:47.792Z", "loss": 1.4446009397506714, "grad_norm": 2.5957640470911145, "lr": 2.999592192828189e-07, "mem": 80.76888179779053, "tps": 12417.215980356528, "tps_per_gpu": 1552.151997544566, "mfu": 0.0, "num_tokens_per_step": 86940, "num_label_tokens": 86940}
{"step": 95, "epoch": 0, "timestamp": "2026-10-11T20:28:55.217Z", "loss": 1.548079013824463, "grad_norm": 2.733785041515086, "lr": 2.999573867980408e-07, "mem": 79.69421005249023, "tps": 11708.629071946125, "tps_per_gpu": 1463.5786339932656, "mfu": 0.0, "num_tokens_per_step": 86930, "num_label_tokens": 86930}
{"step": 96, "epoch": 0, "timestamp": "2026-10-11T20:29:02.317Z", "loss": 1.502723217010498, "grad_norm": 2.6482994268805795, "lr": 2.999555140504393e-07, "mem": 82.91949558258057, "tps": 12939.721339621236, "tps_per_gpu": 1617.4651674526544, "mfu": 0.0, "num_tokens_per_step": 91897, "num_label_tokens": 91897}
{"step": 97, "epoch": 0, "timestamp": "2026-10-11T20:29:09.879Z", "loss": 1.4360308647155762, "grad_norm": 2.7403941745203055, "lr": 2.9995360104076867e-07, "mem": 85.05121517181396, "tps": 11560.217065941359, "tps_per_gpu": 1445.0271332426698, "mfu": 0.0, "num_tokens_per_step": 87402, "num_label_tokens": 87402}
{"step": 98, "epoch": 0, "timestamp": "2026-10-11T20:29:17.342Z", "loss": 1.4702179431915283, "grad_norm": 2.6207896549358596, "lr": 2.9995164776979945e-07, "mem": 87.21400117874146, "tps": 11568.110665751987, "tps_per_gpu": 1446.0138332189983, "mfu": 0.0, "num_tokens_per_step": 86327, "num_label_tokens": 86327}
{"step": 99, "epoch": 0, "timestamp": "2026-10-11T20:29:24.795Z", "loss": 1.5611705780029297, "grad_norm": 2.7850381305906518, "lr": 2.9994965423831854e-07, "mem": 87.21400165557861, "tps": 11805.618522105153, "tps_per_gpu": 1475.702315263144, "mfu": 0.0, "num_tokens_per_step": 88003, "num_label_tokens": 88003}
{"step": 100, "epoch": 0, "timestamp": "2026-10-11T20:30:12.302Z", "loss": 1.9966115951538086, "grad_norm": 2.8081359287534973, "lr": 2.999476204471288e-07, "mem": 76.45979595184326, "tps": 1883.174049428675, "tps_per_gpu": 235.39675617858438, "mfu": 0.0, "num_tokens_per_step": 89466, "num_label_tokens": 89466}
{"step": 101, "epoch": 0, "timestamp": "2026-10-11T20:30:19.851Z", "loss": 1.2436461448669434, "grad_norm": 2.2711952947646123, "lr": 2.9994554639704947e-07, "mem": 80.76883745193481, "tps": 12277.186321348589, "tps_per_gpu": 1534.6482901685736, "mfu": 0.0, "num_tokens_per_step": 92694, "num_label_tokens": 92694}
{"step": 102, "epoch": 0, "timestamp": "2026-10-11T20:30:27.407Z", "loss": 1.2511717081069946, "grad_norm": 2.2458346853865043, "lr": 2.9994343208891605e-07, "mem": 81.84387159347534, "tps": 12259.71397751772, "tps_per_gpu": 1532.464247189715, "mfu": 0.0, "num_tokens_per_step": 92618, "num_label_tokens": 92618}
{"step": 103, "epoch": 0, "timestamp": "2026-10-11T20:30:35.020Z", "loss": 1.2263036966323853, "grad_norm": 2.32327966295446, "lr": 2.999412775235801e-07, "mem": 87.21395587921143, "tps": 11835.52750103037, "tps_per_gpu": 1479.4409376287963, "mfu": 0.0, "num_tokens_per_step": 90086, "num_label_tokens": 90086}
{"step": 104, "epoch": 0, "timestamp": "2026-10-11T20:30:42.111Z", "loss": 1.3692610263824463, "grad_norm": 2.4427233805968704, "lr": 2.999390827019096e-07, "mem": 83.9841423034668, "tps": 12843.123760012868, "tps_per_gpu": 1605.3904700016085, "mfu": 0.0, "num_tokens_per_step": 91100, "num_label_tokens": 91100}
{"step": 105, "epoch": 0, "timestamp": "2026-10-11T20:30:49.142Z", "loss": 1.2300924062728882, "grad_norm": 2.6552719574801897, "lr": 2.9993684762478846e-07, "mem": 79.694167137146, "tps": 12460.796738771276, "tps_per_gpu": 1557.5995923464095, "mfu": 0.0, "num_tokens_per_step": 87606, "num_label_tokens": 87606}
{"step": 106, "epoch": 0, "timestamp": "2026-10-11T20:30:55.911Z", "loss": 1.4394065141677856, "grad_norm": 2.626088296548205, "lr": 2.9993457229311705e-07, "mem": 83.98414325714111, "tps": 12723.958586872444, "tps_per_gpu": 1590.4948233590555, "mfu": 0.0, "num_tokens_per_step": 86117, "num_label_tokens": 86117}
{"step": 107, "epoch": 0, "timestamp": "2026-10-11T20:31:13.341Z", "loss": 1.303235411643982, "grad_norm": 2.3325283871880615, "lr": 2.9993225670781185e-07, "mem": 82.91945314407349, "tps": 5213.9517592022785, "tps_per_gpu": 651.7439699002848, "mfu": 0.0, "num_tokens_per_step": 90862, "num_label_tokens": 90862}
{"step": 108, "epoch": 0, "timestamp": "2026-10-11T20:31:20.964Z", "loss": 1.155618667602539, "grad_norm": 2.1647062097062206, "lr": 2.9992990086980566e-07, "mem": 86.13408184051514, "tps": 12216.070181938023, "tps_per_gpu": 1527.0087727422529, "mfu": 0.0, "num_tokens_per_step": 93147, "num_label_tokens": 93147}
{"step": 109, "epoch": 0, "timestamp": "2026-10-11T20:31:28.498Z", "loss": 1.2563371658325195, "grad_norm": 2.323392394391878, "lr": 2.9992750478004737e-07, "mem": 75.40293025970459, "tps": 11881.732666555883, "tps_per_gpu": 1485.2165833194854, "mfu": 0.0, "num_tokens_per_step": 89537, "num_label_tokens": 89537}
{"step": 110, "epoch": 0, "timestamp": "2026-10-11T20:31:35.922Z", "loss": 1.7238860130310059, "grad_norm": 2.742366875411881, "lr": 2.9992506843950207e-07, "mem": 87.21395921707153, "tps": 12101.615332512738, "tps_per_gpu": 1512.7019165640922, "mfu": 0.0, "num_tokens_per_step": 89824, "num_label_tokens": 89824}
{"step": 111, "epoch": 0, "timestamp": "2026-10-11T20:31:43.498Z", "loss": 1.5272619724273682, "grad_norm": 2.5003483439500473, "lr": 2.9992259184915114e-07, "mem": 87.21395969390869, "tps": 12190.288496573052, "tps_per_gpu": 1523.7860620716315, "mfu": 0.0, "num_tokens_per_step": 92364, "num_label_tokens": 92364}
{"step": 112, "epoch": 0, "timestamp": "2026-10-11T20:31:51.008Z", "loss": 1.1937235593795776, "grad_norm": 2.379347293668464, "lr": 2.999200750099921e-07, "mem": 85.05117464065552, "tps": 12368.975752429056, "tps_per_gpu": 1546.121969053632, "mfu": 0.0, "num_tokens_per_step": 92895, "num_label_tokens": 92895}
{"step": 113, "epoch": 0, "timestamp": "2026-10-11T20:31:58.559Z", "loss": 1.2296867370605469, "grad_norm": 2.3186126116459542, "lr": 2.999175179230389e-07, "mem": 82.91945600509644, "tps": 12469.903192867907, "tps_per_gpu": 1558.7378991084884, "mfu": 0.0, "num_tokens_per_step": 94147, "num_label_tokens": 94147}
{"step": 114, "epoch": 0, "timestamp": "2026-10-11T20:32:06.162Z", "loss": 1.2081271409988403, "grad_norm": 2.2464391706377906, "lr": 2.9991492058932137e-07, "mem": 86.13408470153809, "tps": 12208.594777667155, "tps_per_gpu": 1526.0743472083943, "mfu": 0.0, "num_tokens_per_step": 92837, "num_label_tokens": 92837}
{"step": 115, "epoch": 0, "timestamp": "2026-10-11T20:32:13.774Z", "loss": 1.2171638011932373, "grad_norm": 2.2860454679074516, "lr": 2.999122830098858e-07, "mem": 85.05117607116699, "tps": 11990.428224147025, "tps_per_gpu": 1498.803528018378, "mfu": 0.0, "num_tokens_per_step": 91277, "num_label_tokens": 91277}
{"step": 116, "epoch": 0, "timestamp": "2026-10-11T20:32:21.416Z", "loss": 1.2662631273269653, "grad_norm": 2.282208327853355, "lr": 2.999096051857946e-07, "mem": 85.05117654800415, "tps": 11873.331730645348, "tps_per_gpu": 1484.1664663306685, "mfu": 0.0, "num_tokens_per_step": 90740, "num_label_tokens": 90740}
{"step": 117, "epoch": 0, "timestamp": "2026-10-11T20:32:28.560Z", "loss": 1.5895161628723145, "grad_norm": 2.588277236898345, "lr": 2.9990688711812634e-07, "mem": 81.84387874603271, "tps": 12492.522208156954, "tps_per_gpu": 1561.5652760196192, "mfu": 0.0, "num_tokens_per_step": 89245, "num_label_tokens": 89245}
{"step": 118, "epoch": 0, "timestamp": "2026-10-11T20:32:36.118Z", "loss": 1.9373877048492432, "grad_norm": 2.9243816416696715, "lr": 2.9990412880797596e-07, "mem": 87.2139630317688, "tps": 11715.019600267302, "tps_per_gpu": 1464.3774500334127, "mfu": 0.0, "num_tokens_per_step": 88523, "num_label_tokens": 88523}
{"step": 119, "epoch": 0, "timestamp": "2026-10-11T20:32:43.176Z", "loss": 1.8240116834640503, "grad_norm": 2.8588325219959203, "lr": 2.999013302564544e-07, "mem": 81.84387969970703, "tps": 12886.744751103535, "tps_per_gpu": 1610.843093887942, "mfu": 0.0, "num_tokens_per_step": 90954, "num_label_tokens": 90954}
{"step": 120, "epoch": 0, "timestamp": "2026-10-11T20:32:50.765Z", "loss": 1.242078185081482, "grad_norm": 2.2413814203423277, "lr": 2.9989849146468896e-07, "mem": 79.69417428970337, "tps": 12023.655973052772, "tps_per_gpu": 1502.9569966315964, "mfu": 0.0, "num_tokens_per_step": 91244, "num_label_tokens": 91244}
{"step": 121, "epoch": 0, "timestamp": "2026-10-11T20:32:58.410Z", "loss": 1.3702549934387207, "grad_norm": 2.318699246668153, "lr": 2.998956124338231e-07, "mem": 87.21396446228027, "tps": 11902.684793380737, "tps_per_gpu": 1487.835599172592, "mfu": 0.0, "num_tokens_per_step": 90999, "num_label_tokens": 90999}
{"step": 122, "epoch": 0, "timestamp": "2026-10-11T20:33:05.977Z", "loss": 1.322862148284912, "grad_norm": 2.264481448085031, "lr": 2.9989269316501647e-07, "mem": 87.21396493911743, "tps": 12131.47810249482, "tps_per_gpu": 1516.4347628118526, "mfu": 0.0, "num_tokens_per_step": 91721, "num_label_tokens": 91721}
{"step": 123, "epoch": 0, "timestamp": "2026-10-11T20:33:13.566Z", "loss": 1.5590583086013794, "grad_norm": 2.709422870327726, "lr": 2.99889733659445e-07, "mem": 81.84388160705566, "tps": 12435.692104582746, "tps_per_gpu": 1554.4615130728432, "mfu": 0.0, "num_tokens_per_step": 94464, "num_label_tokens": 94464}
{"step": 124, "epoch": 0, "timestamp": "2026-10-11T20:33:21.127Z", "loss": 1.491477131843567, "grad_norm": 2.5849838853059923, "lr": 2.998867339183008e-07, "mem": 83.98415184020996, "tps": 12527.80309989551, "tps_per_gpu": 1565.9753874869386, "mfu": 0.0, "num_tokens_per_step": 94714, "num_label_tokens": 94714}
{"step": 125, "epoch": 0, "timestamp": "2026-10-11T20:33:38.441Z", "loss": 1.5023157596588135, "grad_norm": 2.611385697126493, "lr": 2.9988369394279205e-07, "mem": 81.84388256072998, "tps": 5254.8568013950335, "tps_per_gpu": 656.8571001743792, "mfu": 0.0, "num_tokens_per_step": 90986, "num_label_tokens": 90986}
{"step": 126, "epoch": 0, "timestamp": "2026-10-11T20:33:45.997Z", "loss": 1.9359865188598633, "grad_norm": 2.8621353788798976, "lr": 2.998806137341434e-07, "mem": 80.76884937286377, "tps": 11951.318937529168, "tps_per_gpu": 1493.914867191146, "mfu": 0.0, "num_tokens_per_step": 90310, "num_label_tokens": 90310}
{"step": 127, "epoch": 0, "timestamp": "2026-10-11T20:33:53.580Z", "loss": 1.4685944318771362, "grad_norm": 2.774461752979249, "lr": 2.9987749329359546e-07, "mem": 79.69417762756348, "tps": 11665.698778038373, "tps_per_gpu": 1458.2123472547967, "mfu": 0.0, "num_tokens_per_step": 88452, "num_label_tokens": 88452}
{"step": 128, "epoch": 0, "timestamp": "2026-10-11T20:34:01.083Z", "loss": 1.441120982170105, "grad_norm": 2.6806128166422587, "lr": 2.9987433262240515e-07, "mem": 86.1340913772583, "tps": 11713.945304817255, "tps_per_gpu": 1464.2431631021568, "mfu": 0.0, "num_tokens_per_step": 87917, "num_label_tokens": 87917}
{"step": 129, "epoch": 0, "timestamp": "2026-10-11T20:34:08.623Z", "loss": 1.3091098070144653, "grad_norm": 2.6586318770504898, "lr": 2.998711317218456e-07, "mem": 82.91946363449097, "tps": 11646.373822488053, "tps_per_gpu": 1455.7967278110066, "mfu": 0.0, "num_tokens_per_step": 87803, "num_label_tokens": 87803}
{"step": 130, "epoch": 0, "timestamp": "2026-10-11T20:34:16.214Z", "loss": 1.2285040616989136, "grad_norm": 2.3790938780429487, "lr": 2.9986789059320615e-07, "mem": 83.98415470123291, "tps": 12154.164326717688, "tps_per_gpu": 1519.270540839711, "mfu": 0.0, "num_tokens_per_step": 92257, "num_label_tokens": 92257}
{"step": 131, "epoch": 0, "timestamp": "2026-10-11T20:34:23.801Z", "loss": 1.3621050119400024, "grad_norm": 2.3160386029248, "lr": 2.998646092377923e-07, "mem": 87.21396923065186, "tps": 12112.50376286414, "tps_per_gpu": 1514.0629703580175, "mfu": 0.0, "num_tokens_per_step": 91883, "num_label_tokens": 91883}
{"step": 132, "epoch": 0, "timestamp": "2026-10-11T20:34:31.374Z", "loss": 1.311484932899475, "grad_norm": 2.2631461741704295, "lr": 2.9986128765692577e-07, "mem": 86.13409328460693, "tps": 12317.80507004896, "tps_per_gpu": 1539.72563375612, "mfu": 0.0, "num_tokens_per_step": 93308, "num_label_tokens": 93308}
{"step": 133, "epoch": 0, "timestamp": "2026-10-11T20:34:38.874Z", "loss": 2.0138936042785645, "grad_norm": 2.853791058500735, "lr": 2.9985792585194455e-07, "mem": 79.69418048858643, "tps": 12091.323226672772, "tps_per_gpu": 1511.4154033340965, "mfu": 0.0, "num_tokens_per_step": 90669, "num_label_tokens": 90669}
{"step": 134, "epoch": 0, "timestamp": "2026-10-11T20:34:46.442Z", "loss": 1.2624361515045166, "grad_norm": 2.2872872403611626, "lr": 2.9985452382420273e-07, "mem": 86.13409423828125, "tps": 12575.979589272447, "tps_per_gpu": 1571.9974486590559, "mfu": 0.0, "num_tokens_per_step": 95166, "num_label_tokens": 95166}
{"step": 135, "epoch": 0, "timestamp": "2026-10-11T20:34:54.063Z", "loss": 1.3029577732086182, "grad_norm": 2.216839758776394, "lr": 2.9985108157507066e-07, "mem": 85.05118560791016, "tps": 12308.523658738886, "tps_per_gpu": 1538.5654573423608, "mfu": 0.0, "num_tokens_per_step": 93790, "num_label_tokens": 93790}
{"step": 136, "epoch": 0, "timestamp": "2026-10-11T20:35:01.633Z", "loss": 1.3055777549743652, "grad_norm": 2.37508757053865, "lr": 2.9984759910593486e-07, "mem": 87.21397161483765, "tps": 12272.952127029337, "tps_per_gpu": 1534.1190158786671, "mfu": 0.0, "num_tokens_per_step": 92943, "num_label_tokens": 92943}
{"step": 137, "epoch": 0, "timestamp": "2026-10-11T20:35:09.162Z", "loss": 1.3161736726760864, "grad_norm": 2.183916658391497, "lr": 2.998440764181981e-07, "mem": 87.2139720916748, "tps": 12675.648737052314, "tps_per_gpu": 1584.4560921315392, "mfu": 0.0, "num_tokens_per_step": 95403, "num_label_tokens": 95403}
{"step": 138, "epoch": 0, "timestamp": "2026-10-11T20:35:16.707Z", "loss": 1.2575311660766602, "grad_norm": 2.2490280297301024, "lr": 2.998405135132793e-07, "mem": 80.76885509490967, "tps": 12114.88027230139, "tps_per_gpu": 1514.3600340376738, "mfu": 0.0, "num_tokens_per_step": 91404, "num_label_tokens": 91404}
{"step": 139, "epoch": 0, "timestamp": "2026-10-11T20:35:24.298Z", "loss": 1.2991384267807007, "grad_norm": 2.2250641660565242, "lr": 2.9983691039261357e-07, "mem": 81.8438892364502, "tps": 11952.69881720146, "tps_per_gpu": 1494.0873521501826, "mfu": 0.0, "num_tokens_per_step": 90741, "num_label_tokens": 90741}
{"step": 140, "epoch": 0, "timestamp": "2026-10-11T20:35:31.823Z", "loss": 1.5357728004455566, "grad_norm": 2.5745716930044305, "lr": 2.998332670576523e-07, "mem": 81.84388971328735, "tps": 11892.057323917634, "tps_per_gpu": 1486.5071654897042, "mfu": 0.0, "num_tokens_per_step": 89479, "num_label_tokens": 89479}
{"step": 141, "epoch": 0, "timestamp": "2026-10-11T20:35:39.317Z", "loss": 1.449852466583252, "grad_norm": 2.7793742979070495, "lr": 2.9982958350986293e-07, "mem": 77.54819297790527, "tps": 11668.703877583917, "tps_per_gpu": 1458.5879846979897, "mfu": 0.0, "num_tokens_per_step": 87450, "num_label_tokens": 87450}
{"step": 142, "epoch": 0, "timestamp": "2026-10-11T20:35:46.924Z", "loss": 1.3876420259475708, "grad_norm": 2.4964324249017364, "lr": 2.9982585975072934e-07, "mem": 85.05118894577026, "tps": 11687.384032134632, "tps_per_gpu": 1460.923004016829, "mfu": 0.0, "num_tokens_per_step": 88912, "num_label_tokens": 88912}
{"step": 143, "epoch": 0, "timestamp": "2026-10-11T20:36:00.993Z", "loss": 1.9826335906982422, "grad_norm": 2.71919397256746, "lr": 2.9982209578175135e-07, "mem": 86.13409852981567, "tps": 6350.261476922265, "tps_per_gpu": 793.7826846152831, "mfu": 0.0, "num_tokens_per_step": 89332, "num_label_tokens": 89332}
{"step": 144, "epoch": 0, "timestamp": "2026-10-11T20:36:12.002Z", "loss": 1.446877360343933, "grad_norm": 2.5584373415637662, "lr": 2.998182916044451e-07, "mem": 80.76885795593262, "tps": 8297.11507866535, "tps_per_gpu": 1037.1393848331688, "mfu": 0.0, "num_tokens_per_step": 91345, "num_label_tokens": 91345}
{"step": 145, "epoch": 0, "timestamp": "2026-10-11T20:36:19.593Z", "loss": 1.34660005569458, "grad_norm": 2.2642467551977177, "lr": 2.99814447220343e-07, "mem": 79.69418621063232, "tps": 12547.671345987375, "tps_per_gpu": 1568.458918248422, "mfu": 0.0, "num_tokens_per_step": 95274, "num_label_tokens": 95274}
{"step": 146, "epoch": 0, "timestamp": "2026-10-11T20:36:27.129Z", "loss": 1.2516909837722778, "grad_norm": 2.3679305503385595, "lr": 2.9981056263099345e-07, "mem": 77.54819536209106, "tps": 11793.270754814315, "tps_per_gpu": 1474.1588443517894, "mfu": 0.0, "num_tokens_per_step": 88853, "num_label_tokens": 88853}
{"step": 147, "epoch": 0, "timestamp": "2026-10-11T20:36:34.680Z", "loss": 1.197153925895691, "grad_norm": 2.1225084928444815, "lr": 2.9980663783796113e-07, "mem": 85.05119132995605, "tps": 12719.183397411569, "tps_per_gpu": 1589.897924676446, "mfu": 0.0, "num_tokens_per_step": 96060, "num_label_tokens": 96060}
{"step": 148, "epoch": 0, "timestamp": "2026-10-11T20:36:42.253Z", "loss": 1.3231645822525024, "grad_norm": 2.358101518356727, "lr": 2.9980267284282714e-07, "mem": 86.13410091400146, "tps": 12241.353540781809, "tps_per_gpu": 1530.1691925977261, "mfu": 0.0, "num_tokens_per_step": 92704, "num_label_tokens": 92704}
{"step": 149, "epoch": 0, "timestamp": "2026-10-11T20:36:49.834Z", "loss": 1.287225604057312, "grad_norm": 2.293840071967452, "lr": 2.9979866764718845e-07, "mem": 87.2139778137207, "tps": 12384.705498249183, "tps_per_gpu": 1548.0881872811478, "mfu": 0.0, "num_tokens_per_step": 93886, "num_label_tokens": 93886}
{"step": 150, "epoch": 0, "timestamp": "2026-10-11T20:36:57.416Z", "loss": 1.2707841396331787, "grad_norm": 2.282963183917318, "lr": 2.9979462225265833e-07, "mem": 87.21397829055786, "tps": 12094.704648623025, "tps_per_gpu": 1511.838081077878, "mfu": 0.0, "num_tokens_per_step": 91705, "num_label_tokens": 91705}
{"step": 151, "epoch": 0, "timestamp": "2026-10-11T20:37:04.877Z", "loss": 1.2462835311889648, "grad_norm": 2.341027838197105, "lr": 2.997905366608663e-07, "mem": 87.21397876739502, "tps": 11712.301633957983, "tps_per_gpu": 1464.0377042447478, "mfu": 0.0, "num_tokens_per_step": 87373, "num_label_tokens": 87373}
{"step": 152, "epoch": 0, "timestamp": "2026-10-11T20:37:12.023Z", "loss": 1.332382321357727, "grad_norm": 2.2597427029992856, "lr": 2.9978641087345806e-07, "mem": 86.13554811477661, "tps": 13361.205690751445, "tps_per_gpu": 1670.1507113439307, "mfu": 0.0, "num_tokens_per_step": 95478, "num_label_tokens": 95478}
{"step": 153, "epoch": 0, "timestamp": "2026-10-11T20:37:19.207Z", "loss": 1.6503008604049683, "grad_norm": 2.5800290127504515, "lr": 2.9978224489209546e-07, "mem": 86.13554859161377, "tps": 12583.727892975585, "tps_per_gpu": 1572.965986621948, "mfu": 0.0, "num_tokens_per_step": 90381, "num_label_tokens": 90381}
{"step": 154, "epoch": 0, "timestamp": "2026-10-11T20:37:26.765Z", "loss": 1.552734613418579, "grad_norm": 2.4766554440445745, "lr": 2.9977803871845645e-07, "mem": 85.05119466781616, "tps": 11949.112445241313, "tps_per_gpu": 1493.6390556551642, "mfu": 0.0, "num_tokens_per_step": 90343, "num_label_tokens": 90343}
{"step": 155, "epoch": 0, "timestamp": "2026-10-11T20:37:34.370Z", "loss": 1.403090476989746, "grad_norm": 2.304127410653248, "lr": 2.997737923542355e-07, "mem": 86.13410425186157, "tps": 12418.944053456851, "tps_per_gpu": 1552.3680066821064, "mfu": 0.0, "num_tokens_per_step": 94442, "num_label_tokens": 94442}
{"step": 156, "epoch": 0, "timestamp": "2026-10-11T20:37:41.939Z", "loss": 1.2407500743865967, "grad_norm": 2.199993397079476, "lr": 2.9976950580114287e-07, "mem": 82.91947650909424, "tps": 12105.94752486545, "tps_per_gpu": 1513.2434406081813, "mfu": 0.0, "num_tokens_per_step": 91616, "num_label_tokens": 91616}
{"step": 157, "epoch": 0, "timestamp": "2026-10-11T20:37:49.000Z", "loss": 1.664760947227478, "grad_norm": 2.48293507787477, "lr": 2.9976517906090524e-07, "mem": 79.69419193267822, "tps": 12726.252989546188, "tps_per_gpu": 1590.7816236932736, "mfu": 0.0, "num_tokens_per_step": 89857, "num_label_tokens": 89857}
{"step": 158, "epoch": 0, "timestamp": "2026-10-11T20:37:56.067Z", "loss": 1.8458287715911865, "grad_norm": 2.757509045025219, "lr": 2.9976081213526546e-07, "mem": 78.60672998428345, "tps": 12564.390748574064, "tps_per_gpu": 1570.548843571758, "mfu": 0.0, "num_tokens_per_step": 88823, "num_label_tokens": 88823}
{"step": 159, "epoch": 0, "timestamp": "2026-10-11T20:38:03.543Z", "loss": 1.345874547958374, "grad_norm": 2.655224300313788, "lr": 2.9975640502598243e-07, "mem": 74.34162092208862, "tps": 11555.818468902142, "tps_per_gpu": 1444.4773086127677, "mfu": 0.0, "num_tokens_per_step": 86394, "num_label_tokens": 86394}
{"step": 160, "epoch": 0, "timestamp": "2026-10-11T20:38:11.117Z", "loss": 1.3675721883773804, "grad_norm": 2.789209996564737, "lr": 2.9975195773483134e-07, "mem": 87.21398305892944, "tps": 11981.38023707234, "tps_per_gpu": 1497.6725296340426, "mfu": 0.0, "num_tokens_per_step": 90729, "num_label_tokens": 90729}
{"step": 161, "epoch": 0, "timestamp": "2026-10-11T20:38:18.653Z", "loss": 1.3828680515289307, "grad_norm": 2.4684170546508284, "lr": 2.997474702636037e-07, "mem": 82.91947889328003, "tps": 12050.10837341991, "tps_per_gpu": 1506.2635466774886, "mfu": 0.0, "num_tokens_per_step": 90834, "num_label_tokens": 90834}
{"step": 162, "epoch": 0, "timestamp": "2026-10-11T20:38:28.936Z", "loss": 1.3064602613449097, "grad_norm": 2.3806178517134295, "lr": 2.997429426141069e-07, "mem": 83.98416996002197, "tps": 8996.893313073868, "tps_per_gpu": 1124.6116641342335, "mfu": 0.0, "num_tokens_per_step": 92505, "num_label_tokens": 92505}
{"step": 163, "epoch": 0, "timestamp": "2026-10-11T20:38:39.172Z", "loss": 1.3268847465515137, "grad_norm": 2.429778126338447, "lr": 2.9973837478816483e-07, "mem": 80.76886701583862, "tps": 8738.132695944398, "tps_per_gpu": 1092.2665869930497, "mfu": 0.0, "num_tokens_per_step": 89450, "num_label_tokens": 89450}
{"step": 164, "epoch": 0, "timestamp": "2026-10-11T20:38:48.606Z", "loss": 1.453507900238037, "grad_norm": 2.4670584402995317, "lr": 2.9973376678761724e-07, "mem": 86.134108543396, "tps": 9810.013363020953, "tps_per_gpu": 1226.2516703776191, "mfu": 0.0, "num_tokens_per_step": 92554, "num_label_tokens": 92554}
{"step": 165, "epoch": 0, "timestamp": "2026-10-11T20:38:56.182Z", "loss": 1.9016553163528442, "grad_norm": 2.7471363893290417, "lr": 2.997291186143203e-07, "mem": 87.21398544311523, "tps": 12203.277611837371, "tps_per_gpu": 1525.4097014796714, "mfu": 0.0, "num_tokens_per_step": 92435, "num_label_tokens": 92435}
{"step": 166, "epoch": 0, "timestamp": "2026-10-11T20:39:03.738Z", "loss": 1.2502366304397583, "grad_norm": 2.211872600193686, "lr": 2.9972443027014635e-07, "mem": 80.7688684463501, "tps": 12322.82209910953, "tps_per_gpu": 1540.3527623886912, "mfu": 0.0, "num_tokens_per_step": 93094, "num_label_tokens": 93094}
{"step": 167, "epoch": 0, "timestamp": "2026-10-11T20:39:11.356Z", "loss": 1.2823710441589355, "grad_norm": 2.2219660363847096, "lr": 2.9971970175698383e-07, "mem": 85.05120086669922, "tps": 12577.420196171523, "tps_per_gpu": 1572.1775245214403, "mfu": 0.0, "num_tokens_per_step": 95794, "num_label_tokens": 95794}
{"step": 168, "epoch": 0, "timestamp": "2026-10-11T20:39:18.449Z", "loss": 1.1278183460235596, "grad_norm": 2.2891362971724485, "lr": 2.997149330767373e-07, "mem": 81.84390306472778, "tps": 13038.293726790896, "tps_per_gpu": 1629.786715848862, "mfu": 0.0, "num_tokens_per_step": 92494, "num_label_tokens": 92494}
{"step": 169, "epoch": 0, "timestamp": "2026-10-11T20:39:25.398Z", "loss": 1.4732682704925537, "grad_norm": 2.9150550962480235, "lr": 2.997101242313277e-07, "mem": 86.1355562210083, "tps": 11965.48470418359, "tps_per_gpu": 1495.6855880229487, "mfu": 0.0, "num_tokens_per_step": 83175, "num_label_tokens": 83175}
{"step": 170, "epoch": 0, "timestamp": "2026-10-11T20:39:32.848Z", "loss": 1.436087727546692, "grad_norm": 2.7862768340092736, "lr": 2.99705275222692e-07, "mem": 76.47957468032837, "tps": 11788.439203395874, "tps_per_gpu": 1473.5549004244842, "mfu": 0.0, "num_tokens_per_step": 87805, "num_label_tokens": 87805}
{"step": 171, "epoch": 0, "timestamp": "2026-10-11T20:39:40.394Z", "loss": 1.4053562879562378, "grad_norm": 3.0093419120224687, "lr": 2.997003860527834e-07, "mem": 85.05120277404785, "tps": 12138.116054602677, "tps_per_gpu": 1517.2645068253346, "mfu": 0.0, "num_tokens_per_step": 91608, "num_label_tokens": 91608}
{"step": 172, "epoch": 0, "timestamp": "2026-10-11T20:39:47.528Z", "loss": 1.7059714794158936, "grad_norm": 2.714832774470152, "lr": 2.996954567235711e-07, "mem": 85.05120325088501, "tps": 12460.392759331573, "tps_per_gpu": 1557.5490949164466, "mfu": 0.0, "num_tokens_per_step": 88868, "num_label_tokens": 88868}
{"step": 173, "epoch": 0, "timestamp": "2026-10-11T20:39:55.066Z", "loss": 1.6264972686767578, "grad_norm": 2.645315828315187, "lr": 2.9969048723704087e-07, "mem": 80.7688717842102, "tps": 11953.43069160134, "tps_per_gpu": 1494.1788364501674, "mfu": 0.0, "num_tokens_per_step": 90109, "num_label_tokens": 90109}
{"step": 174, "epoch": 0, "timestamp": "2026-10-11T20:40:02.658Z", "loss": 1.5390313863754272, "grad_norm": 2.7957209121707827, "lr": 2.9968547759519423e-07, "mem": 82.91948509216309, "tps": 11742.412095353953, "tps_per_gpu": 1467.8015119192441, "mfu": 0.0, "num_tokens_per_step": 89179, "num_label_tokens": 89179}
{"step": 175, "epoch": 0, "timestamp": "2026-10-11T20:40:10.208Z", "loss": 1.4712468385696411, "grad_norm": 2.8297841645043023, "lr": 2.9968042780004915e-07, "mem": 87.21399021148682, "tps": 12579.76654543754, "tps_per_gpu": 1572.4708181796925, "mfu": 0.0, "num_tokens_per_step": 94936, "num_label_tokens": 94936}
{"step": 176, "epoch": 0, "timestamp": "2026-10-11T20:40:17.270Z", "loss": 1.3151708841323853, "grad_norm": 2.3867093414833196, "lr": 2.9967533785363967e-07, "mem": 85.05120515823364, "tps": 12755.546941765198, "tps_per_gpu": 1594.4433677206498, "mfu": 0.0, "num_tokens_per_step": 90103, "num_label_tokens": 90103}
{"step": 177, "epoch": 0, "timestamp": "2026-10-11T20:40:24.862Z", "loss": 1.345815658569336, "grad_norm": 2.3385402420659887, "lr": 2.99670207758016e-07, "mem": 81.8439073562622, "tps": 11928.384354386844, "tps_per_gpu": 1491.0480442983555, "mfu": 0.0, "num_tokens_per_step": 90541, "num_label_tokens": 90541}
{"step": 178, "epoch": 0, "timestamp": "2026-10-11T20:40:32.355Z", "loss": 1.2521153688430786, "grad_norm": 2.3544635602170647, "lr": 2.996650375152446e-07, "mem": 79.69420194625854, "tps": 11624.676281933276, "tps_per_gpu": 1453.0845352416595, "mfu": 0.0, "num_tokens_per_step": 87139, "num_label_tokens": 87139}
{"step": 179, "epoch": 0, "timestamp": "2026-10-11T20:40:39.866Z", "loss": 1.7510563135147095, "grad_norm": 2.6756185606780254, "lr": 2.996598271274081e-07, "mem": 80.76887464523315, "tps": 11996.397737485662, "tps_per_gpu": 1499.5497171857078, "mfu": 0.0, "num_tokens_per_step": 90081, "num_label_tokens": 90081}
{"step": 180, "epoch": 0, "timestamp": "2026-10-11T20:40:47.438Z", "loss": 2.0237345695495605, "grad_norm": 2.7446560347633278, "lr": 2.99654576596605e-07, "mem": 80.76887512207031, "tps": 12477.983173215007, "tps_per_gpu": 1559.7478966518759, "mfu": 0.0, "num_tokens_per_step": 94475, "num_label_tokens": 94475}
{"step": 181, "epoch": 0, "timestamp": "2026-10-11T20:40:54.880Z", "loss": 1.8724768161773682, "grad_norm": 2.8099437053152116, "lr": 2.9964928592495046e-07, "mem": 82.9194884300232, "tps": 12030.906730280924, "tps_per_gpu": 1503.8633412851154, "mfu": 0.0, "num_tokens_per_step": 89540, "num_label_tokens": 89540}
{"step": 182, "epoch": 0, "timestamp": "2026-10-11T20:41:05.073Z", "loss": 1.2313079833984375, "grad_norm": 3.1508103488936765, "lr": 2.996439551145754e-07, "mem": 85.05120801925659, "tps": 8857.939808912966, "tps_per_gpu": 1107.2424761141208, "mfu": 0.0, "num_tokens_per_step": 90281, "num_label_tokens": 90281}
{"step": 183, "epoch": 0, "timestamp": "2026-10-11T20:41:15.346Z", "loss": 1.2721848487854004, "grad_norm": 2.291236782243969, "lr": 2.9963858416762713e-07, "mem": 83.9841799736023, "tps": 8766.705788395588, "tps_per_gpu": 1095.8382235494485, "mfu": 0.0, "num_tokens_per_step": 90040, "num_label_tokens": 90040}
{"step": 184, "epoch": 0, "timestamp": "2026-10-11T20:41:25.340Z", "loss": 1.3823527097702026, "grad_norm": 2.280923582159368, "lr": 2.9963317308626915e-07, "mem": 82.91948986053467, "tps": 9349.585190501155, "tps_per_gpu": 1168.6981488126444, "mfu": 0.0, "num_tokens_per_step": 93464, "num_label_tokens": 93464}
{"step": 185, "epoch": 0, "timestamp": "2026-10-11T20:41:32.963Z", "loss": 1.2996104955673218, "grad_norm": 2.3638081747439985, "lr": 2.9962772187268093e-07, "mem": 77.54821395874023, "tps": 11566.237401723742, "tps_per_gpu": 1445.7796752154677, "mfu": 0.0, "num_tokens_per_step": 88168, "num_label_tokens": 88168}
{"step": 186, "epoch": 0, "timestamp": "2026-10-11T20:41:40.566Z", "loss": 1.2650892734527588, "grad_norm": 2.249759775150226, "lr": 2.996222305290582e-07, "mem": 83.98418140411377, "tps": 12203.349030574373, "tps_per_gpu": 1525.4186288217966, "mfu": 0.0, "num_tokens_per_step": 92774, "num_label_tokens": 92774}
{"step": 187, "epoch": 0, "timestamp": "2026-10-11T20:41:48.143Z", "loss": 1.3661444187164307, "grad_norm": 2.276551026149106, "lr": 2.9961669905761303e-07, "mem": 83.98418188095093, "tps": 12334.148986537793, "tps_per_gpu": 1541.7686233172242, "mfu": 0.0, "num_tokens_per_step": 93449, "num_label_tokens": 93449}
{"step": 188, "epoch": 0, "timestamp": "2026-10-11T20:41:55.807Z", "loss": 1.3235939741134644, "grad_norm": 2.2765826934035918, "lr": 2.996111274605733e-07, "mem": 86.1341199874878, "tps": 12164.4770360778, "tps_per_gpu": 1520.559629509725, "mfu": 0.0, "num_tokens_per_step": 93220, "num_label_tokens": 93220}
{"step": 189, "epoch": 0, "timestamp": "2026-10-11T20:42:03.375Z", "loss": 1.3021159172058105, "grad_norm": 2.2965477455961736, "lr": 2.996055157401834e-07, "mem": 85.0512113571167, "tps": 12895.710210769223, "tps_per_gpu": 1611.9637763461528, "mfu": 0.0, "num_tokens_per_step": 97642, "num_label_tokens": 97642}
{"step": 190, "epoch": 0, "timestamp": "2026-10-11T20:42:10.476Z", "loss": 1.2820404767990112, "grad_norm": 2.3561676083094727, "lr": 2.9959986389870364e-07, "mem": 81.84391355514526, "tps": 12643.741409914994, "tps_per_gpu": 1580.4676762393742, "mfu": 0.0, "num_tokens_per_step": 89760, "num_label_tokens": 89760}
{"step": 191, "epoch": 0, "timestamp": "2026-10-11T20:42:18.005Z", "loss": 1.7447971105575562, "grad_norm": 2.713788411469737, "lr": 2.995941719384106e-07, "mem": 83.98418378829956, "tps": 12012.060431698472, "tps_per_gpu": 1501.507553962309, "mfu": 0.0, "num_tokens_per_step": 90436, "num_label_tokens": 90436}
{"step": 192, "epoch": 0, "timestamp": "2026-10-11T20:42:25.515Z", "loss": 1.4367883205413818, "grad_norm": 2.494374067949921, "lr": 2.99588439861597e-07, "mem": 82.91949367523193, "tps": 12342.299667227178, "tps_per_gpu": 1542.7874584033973, "mfu": 0.0, "num_tokens_per_step": 92670, "num_label_tokens": 92670}
{"step": 193, "epoch": 0, "timestamp": "2026-10-11T20:42:33.100Z", "loss": 1.2631916999816895, "grad_norm": 2.2438534254087426, "lr": 2.995826676705718e-07, "mem": 87.21399879455566, "tps": 12394.538762986085, "tps_per_gpu": 1549.3173453732607, "mfu": 0.0, "num_tokens_per_step": 94029, "num_label_tokens": 94029}
{"step": 194, "epoch": 0, "timestamp": "2026-10-11T20:42:40.622Z", "loss": 1.2564185857772827, "grad_norm": 2.3809265903841186, "lr": 2.995768553676599e-07, "mem": 82.91949462890625, "tps": 12161.653857913288, "tps_per_gpu": 1520.206732239161, "mfu": 0.0, "num_tokens_per_step": 91494, "num_label_tokens": 91494}
{"step": 195, "epoch": 0, "timestamp": "2026-10-11T20:42:48.188Z", "loss": 1.46108877658844, "grad_norm": 2.5315931051011376, "lr": 2.9957100295520266e-07, "mem": 83.9841856956482, "tps": 12176.123839213633, "tps_per_gpu": 1522.015479901704, "mfu": 0.0, "num_tokens_per_step": 92105, "num_label_tokens": 92105}
{"step": 196, "epoch": 0, "timestamp": "2026-10-11T20:42:55.767Z", "loss": 1.4193114042282104, "grad_norm": 2.751171293730331, "lr": 2.995651104355573e-07, "mem": 87.21400022506714, "tps": 12227.732026241092, "tps_per_gpu": 1528.4665032801365, "mfu": 0.0, "num_tokens_per_step": 92691, "num_label_tokens": 92691}
{"step": 197, "epoch": 0, "timestamp": "2026-10-11T20:43:03.296Z", "loss": 1.4461320638656616, "grad_norm": 3.0187314330920834, "lr": 2.995591778110973e-07, "mem": 79.69421100616455, "tps": 11607.342273976265, "tps_per_gpu": 1450.917784247033, "mfu": 0.0, "num_tokens_per_step": 87369, "num_label_tokens": 87369}
{"step": 198, "epoch": 0, "timestamp": "2026-10-11T20:43:10.871Z", "loss": 1.468044638633728, "grad_norm": 2.9596428135188524, "lr": 2.995532050842124e-07, "mem": 87.21400117874146, "tps": 12268.992230384785, "tps_per_gpu": 1533.6240287980982, "mfu": 0.0, "num_tokens_per_step": 92936, "num_label_tokens": 92936}
{"step": 199, "epoch": 0, "timestamp": "2026-10-11T20:43:18.449Z", "loss": 1.5996018648147583, "grad_norm": 2.753729704591547, "lr": 2.995471922573084e-07, "mem": 87.21400165557861, "tps": 11758.464358486535, "tps_per_gpu": 1469.808044810817, "mfu": 0.0, "num_tokens_per_step": 89107, "num_label_tokens": 89107}
{"step": 200, "epoch": 0, "timestamp": "2026-10-11T20:44:05.889Z", "loss": 1.4678725004196167, "grad_norm": 2.6744346866796542, "lr": 2.9954113933280734e-07, "mem": 79.67750978469849, "tps": 1847.870195309183, "tps_per_gpu": 230.98377441364786, "mfu": 0.0, "num_tokens_per_step": 87667, "num_label_tokens": 87667}
{"step": 201, "epoch": 0, "timestamp": "2026-10-11T20:44:13.475Z", "loss": 1.2033779621124268, "grad_norm": 2.192570574729385, "lr": 2.9953504631314715e-07, "mem": 83.98414087295532, "tps": 12232.728110198199, "tps_per_gpu": 1529.0910137747749, "mfu": 0.0, "num_tokens_per_step": 92779, "num_label_tokens": 92779}
{"step": 202, "epoch": 0, "timestamp": "2026-10-11T20:44:23.809Z", "loss": 1.2716869115829468, "grad_norm": 2.3398172782556514, "lr": 2.9952891320078234e-07, "mem": 87.21395540237427, "tps": 8651.75903783125, "tps_per_gpu": 1081.4698797289063, "mfu": 0.0, "num_tokens_per_step": 89403, "num_label_tokens": 89403}
{"step": 203, "epoch": 0, "timestamp": "2026-10-11T20:44:34.110Z", "loss": 1.419685959815979, "grad_norm": 2.366744716405403, "lr": 2.9952273999818307e-07, "mem": 86.13407945632935, "tps": 9195.805510075934, "tps_per_gpu": 1149.4756887594917, "mfu": 0.0, "num_tokens_per_step": 94747, "num_label_tokens": 94747}
{"step": 204, "epoch": 0, "timestamp": "2026-10-11T20:44:43.860Z", "loss": 1.3938900232315063, "grad_norm": 2.339423573203925, "lr": 2.9951652670783613e-07, "mem": 81.84387254714966, "tps": 9371.45201395132, "tps_per_gpu": 1171.431501743915, "mfu": 0.0, "num_tokens_per_step": 91332, "num_label_tokens": 91332}
{"step": 205, "epoch": 0, "timestamp": "2026-10-11T20:44:51.461Z", "loss": 1.7999320030212402, "grad_norm": 2.7217266516731327, "lr": 2.9951027333224407e-07, "mem": 82.91945219039917, "tps": 12068.504702203722, "tps_per_gpu": 1508.5630877754652, "mfu": 0.0, "num_tokens_per_step": 91757, "num_label_tokens": 91757}
{"step": 206, "epoch": 0, "timestamp": "2026-10-11T20:44:58.972Z", "loss": 1.4004417657852173, "grad_norm": 2.3646343656066975, "lr": 2.995039798739258e-07, "mem": 83.98414325714111, "tps": 11827.556300542961, "tps_per_gpu": 1478.4445375678702, "mfu": 0.0, "num_tokens_per_step": 88869, "num_label_tokens": 88869}
{"step": 207, "epoch": 0, "timestamp": "2026-10-11T20:45:06.528Z", "loss": 1.210687279701233, "grad_norm": 2.205737243823109, "lr": 2.994976463354164e-07, "mem": 85.05117225646973, "tps": 12210.127082042749, "tps_per_gpu": 1526.2658852553436, "mfu": 0.0, "num_tokens_per_step": 92213, "num_label_tokens": 92213}
{"step": 208, "epoch": 0, "timestamp": "2026-10-11T20:45:14.043Z", "loss": 1.3241150379180908, "grad_norm": 2.2907594344090754, "lr": 2.994912727192669e-07, "mem": 83.98414421081543, "tps": 11961.682600701854, "tps_per_gpu": 1495.2103250877317, "mfu": 0.0, "num_tokens_per_step": 89905, "num_label_tokens": 89905}
{"step": 209, "epoch": 0, "timestamp": "2026-10-11T20:45:21.564Z", "loss": 2.025088310241699, "grad_norm": 2.8204682014906264, "lr": 2.994848590280447e-07, "mem": 83.98414468765259, "tps": 12012.08256760373, "tps_per_gpu": 1501.5103209504662, "mfu": 0.0, "num_tokens_per_step": 90348, "num_label_tokens": 90348}
{"step": 210, "epoch": 0, "timestamp": "2026-10-11T20:45:28.655Z", "loss": 1.9029042720794678, "grad_norm": 2.8821184939484166, "lr": 2.994784052643331e-07, "mem": 85.0511736869812, "tps": 12649.869950561966, "tps_per_gpu": 1581.2337438202458, "mfu": 0.0, "num_tokens_per_step": 89693, "num_label_tokens": 89693}
{"step": 211, "epoch": 0, "timestamp": "2026-10-11T20:45:36.129Z", "loss": 1.894195795059204, "grad_norm": 2.7953135145542056, "lr": 2.9947191143073184e-07, "mem": 87.21395969390869, "tps": 11800.974850876024, "tps_per_gpu": 1475.121856359503, "mfu": 0.0, "num_tokens_per_step": 88183, "num_label_tokens": 88183}
{"step": 212, "epoch": 0, "timestamp": "2026-10-11T20:45:43.646Z", "loss": 1.3717803955078125, "grad_norm": 2.301228870780263, "lr": 2.994653775298565e-07, "mem": 82.91945552825928, "tps": 12245.110456199149, "tps_per_gpu": 1530.6388070248936, "mfu": 0.0, "num_tokens_per_step": 92077, "num_label_tokens": 92077}
{"step": 213, "epoch": 0, "timestamp": "2026-10-11T20:45:50.763Z", "loss": 1.3230431079864502, "grad_norm": 2.3094755155646434, "lr": 2.99458803564339e-07, "mem": 86.13552951812744, "tps": 13306.793834211618, "tps_per_gpu": 1663.3492292764522, "mfu": 0.0, "num_tokens_per_step": 94679, "num_label_tokens": 94679}
{"step": 214, "epoch": 0, "timestamp": "2026-10-11T20:45:58.334Z", "loss": 1.2789440155029297, "grad_norm": 2.2416337339824155, "lr": 2.994521895368273e-07, "mem": 86.13408470153809, "tps": 12277.343935774734, "tps_per_gpu": 1534.6679919718417, "mfu": 0.0, "num_tokens_per_step": 92973, "num_label_tokens": 92973}
{"step": 215, "epoch": 0, "timestamp": "2026-10-11T20:46:05.859Z", "loss": 1.6497738361358643, "grad_norm": 2.523600982085557, "lr": 2.9944553544998557e-07, "mem": 83.98414754867554, "tps": 12348.314843481765, "tps_per_gpu": 1543.5393554352206, "mfu": 0.0, "num_tokens_per_step": 92939, "num_label_tokens": 92939}
{"step": 216, "epoch": 0, "timestamp": "2026-10-11T20:46:13.400Z", "loss": 1.4721932411193848, "grad_norm": 2.469901664116806, "lr": 2.9943884130649407e-07, "mem": 87.21396207809448, "tps": 12254.655744784048, "tps_per_gpu": 1531.831968098006, "mfu": 0.0, "num_tokens_per_step": 92388, "num_label_tokens": 92388}
{"step": 217, "epoch": 0, "timestamp": "2026-10-11T20:46:21.023Z", "loss": 1.2207624912261963, "grad_norm": 2.226302719435923, "lr": 2.994321071090492e-07, "mem": 87.21396255493164, "tps": 12036.955513943007, "tps_per_gpu": 1504.6194392428758, "mfu": 0.0, "num_tokens_per_step": 91762, "num_label_tokens": 91762}
{"step": 218, "epoch": 0, "timestamp": "2026-10-11T20:46:28.606Z", "loss": 1.3798023462295532, "grad_norm": 2.363348684202752, "lr": 2.994253328603634e-07, "mem": 86.13408660888672, "tps": 12119.236551013328, "tps_per_gpu": 1514.904568876666, "mfu": 0.0, "num_tokens_per_step": 91890, "num_label_tokens": 91890}
{"step": 219, "epoch": 0, "timestamp": "2026-10-11T20:46:36.138Z", "loss": 1.323103427886963, "grad_norm": 2.724175097613213, "lr": 2.994185185631654e-07, "mem": 81.84387969970703, "tps": 11849.48422995021, "tps_per_gpu": 1481.1855287437763, "mfu": 0.0, "num_tokens_per_step": 89250, "num_label_tokens": 89250}
{"step": 220, "epoch": 0, "timestamp": "2026-10-11T20:46:46.271Z", "loss": 1.340603232383728, "grad_norm": 3.2042971558788182, "lr": 2.994116642202001e-07, "mem": 83.98414993286133, "tps": 8989.316387348888, "tps_per_gpu": 1123.664548418611, "mfu": 0.0, "num_tokens_per_step": 91109, "num_label_tokens": 91109}
{"step": 221, "epoch": 0, "timestamp": "2026-10-11T20:46:56.538Z", "loss": 1.4280580282211304, "grad_norm": 2.6554606556305482, "lr": 2.9940476983422833e-07, "mem": 86.1340880393982, "tps": 9076.390747192623, "tps_per_gpu": 1134.548843399078, "mfu": 0.0, "num_tokens_per_step": 93174, "num_label_tokens": 93174}
{"step": 222, "epoch": 0, "timestamp": "2026-10-11T20:47:06.542Z", "loss": 1.7682170867919922, "grad_norm": 2.5787798542410276, "lr": 2.9939783540802713e-07, "mem": 85.0511794090271, "tps": 9237.741768761978, "tps_per_gpu": 1154.7177210952473, "mfu": 0.0, "num_tokens_per_step": 92390, "num_label_tokens": 92390}
{"step": 223, "epoch": 0, "timestamp": "2026-10-11T20:47:14.082Z", "loss": 1.9349268674850464, "grad_norm": 2.7297587270165518, "lr": 2.9939086094438973e-07, "mem": 83.9841513633728, "tps": 11977.98068564234, "tps_per_gpu": 1497.2475857052925, "mfu": 0.0, "num_tokens_per_step": 90341, "num_label_tokens": 90341}
{"step": 224, "epoch": 0, "timestamp": "2026-10-11T20:47:21.653Z", "loss": 1.5548059940338135, "grad_norm": 2.58096110293234, "lr": 2.9938384644612537e-07, "mem": 86.13408946990967, "tps": 11758.177280814949, "tps_per_gpu": 1469.7721601018686, "mfu": 0.0, "num_tokens_per_step": 89026, "num_label_tokens": 89026}
{"step": 225, "epoch": 0, "timestamp": "2026-10-11T20:47:29.170Z", "loss": 1.309499979019165, "grad_norm": 2.3037342605687163, "lr": 2.993767919160596e-07, "mem": 85.05118083953857, "tps": 12547.333756779364, "tps_per_gpu": 1568.4167195974205, "mfu": 0.0, "num_tokens_per_step": 94328, "num_label_tokens": 94328}
{"step": 226, "epoch": 0, "timestamp": "2026-10-11T20:47:36.768Z", "loss": 1.2663581371307373, "grad_norm": 2.538231480801296, "lr": 2.9936969735703394e-07, "mem": 87.21396684646606, "tps": 12167.540042135777, "tps_per_gpu": 1520.9425052669721, "mfu": 0.0, "num_tokens_per_step": 92424, "num_label_tokens": 92424}
{"step": 227, "epoch": 0, "timestamp": "2026-10-11T20:47:44.284Z", "loss": 1.592764973640442, "grad_norm": 2.675055478855385, "lr": 2.993625627719061e-07, "mem": 79.69417762756348, "tps": 12023.358584083657, "tps_per_gpu": 1502.919823010457, "mfu": 0.0, "num_tokens_per_step": 90354, "num_label_tokens": 90354}
{"step": 228, "epoch": 0, "timestamp": "2026-10-11T20:47:51.862Z", "loss": 1.9214023351669312, "grad_norm": 2.71778148084021, "lr": 2.993553881635498e-07, "mem": 82.91946315765381, "tps": 11973.11251455063, "tps_per_gpu": 1496.6390643188288, "mfu": 0.0, "num_tokens_per_step": 90798, "num_label_tokens": 90798}
{"step": 229, "epoch": 0, "timestamp": "2026-10-11T20:47:59.368Z", "loss": 1.8700809478759766, "grad_norm": 2.7577783398575724, "lr": 2.99348173534855e-07, "mem": 87.21396827697754, "tps": 12163.02156480839, "tps_per_gpu": 1520.3776956010488, "mfu": 0.0, "num_tokens_per_step": 91250, "num_label_tokens": 91250}
{"step": 230, "epoch": 0, "timestamp": "2026-10-11T20:48:06.976Z", "loss": 1.373528003692627, "grad_norm": 2.779900104452237, "lr": 2.9934091888872785e-07, "mem": 87.2139687538147, "tps": 12150.655167441928, "tps_per_gpu": 1518.831895930241, "mfu": 0.0, "num_tokens_per_step": 92414, "num_label_tokens": 92414}
{"step": 231, "epoch": 0, "timestamp": "2026-10-11T20:48:14.566Z", "loss": 1.3963624238967896, "grad_norm": 2.754801464092948, "lr": 2.993336242280904e-07, "mem": 85.05118370056152, "tps": 11831.599128867962, "tps_per_gpu": 1478.9498911084952, "mfu": 0.0, "num_tokens_per_step": 89807, "num_label_tokens": 89807}
{"step": 232, "epoch": 0, "timestamp": "2026-10-11T20:48:22.082Z", "loss": 1.5060787200927734, "grad_norm": 2.9546129501547105, "lr": 2.9932628955588105e-07, "mem": 81.84388589859009, "tps": 11775.471519132323, "tps_per_gpu": 1471.9339398915404, "mfu": 0.0, "num_tokens_per_step": 88522, "num_label_tokens": 88522}
{"step": 233, "epoch": 0, "timestamp": "2026-10-11T20:48:29.612Z", "loss": 1.803970456123352, "grad_norm": 2.8492901258145205, "lr": 2.993189148750541e-07, "mem": 85.05118465423584, "tps": 11659.88490599722, "tps_per_gpu": 1457.4856132496525, "mfu": 0.0, "num_tokens_per_step": 87800, "num_label_tokens": 87800}
{"step": 234, "epoch": 0, "timestamp": "2026-10-11T20:48:37.215Z", "loss": 1.4824316501617432, "grad_norm": 2.8271758033958103, "lr": 2.9931150018858007e-07, "mem": 80.76885318756104, "tps": 11789.157729055732, "tps_per_gpu": 1473.6447161319666, "mfu": 0.0, "num_tokens_per_step": 89609, "num_label_tokens": 89609}
{"step": 235, "epoch": 0, "timestamp": "2026-10-11T20:48:44.812Z", "loss": 1.4551832675933838, "grad_norm": 2.700718525215197, "lr": 2.993040454994457e-07, "mem": 80.7688536643982, "tps": 12172.854249232509, "tps_per_gpu": 1521.6067811540636, "mfu": 0.0, "num_tokens_per_step": 92494, "num_label_tokens": 92494}
{"step": 236, "epoch": 0, "timestamp": "2026-10-11T20:48:52.372Z", "loss": 1.4600640535354614, "grad_norm": 2.658384603670386, "lr": 2.9929655081065364e-07, "mem": 87.21397161483765, "tps": 12570.881323914722, "tps_per_gpu": 1571.3601654893403, "mfu": 0.0, "num_tokens_per_step": 95041, "num_label_tokens": 95041}
{"step": 237, "epoch": 0, "timestamp": "2026-10-11T20:48:59.926Z", "loss": 1.2574512958526611, "grad_norm": 2.458495928029486, "lr": 2.9928901612522287e-07, "mem": 81.84388828277588, "tps": 11873.670938349609, "tps_per_gpu": 1484.2088672937011, "mfu": 0.0, "num_tokens_per_step": 89698, "num_label_tokens": 89698}
{"step": 238, "epoch": 0, "timestamp": "2026-10-11T20:49:10.199Z", "loss": 1.1688880920410156, "grad_norm": 2.25065819209745, "lr": 2.992814414461882e-07, "mem": 87.21397256851196, "tps": 8890.670652964413, "tps_per_gpu": 1111.3338316205516, "mfu": 0.0, "num_tokens_per_step": 91330, "num_label_tokens": 91330}
{"step": 239, "epoch": 0, "timestamp": "2026-10-11T20:49:20.470Z", "loss": 1.1773067712783813, "grad_norm": 2.2981663779969863, "lr": 2.9927382677660085e-07, "mem": 85.05118751525879, "tps": 8737.789686931124, "tps_per_gpu": 1092.2237108663905, "mfu": 0.0, "num_tokens_per_step": 89771, "num_label_tokens": 89771}
{"step": 240, "epoch": 0, "timestamp": "2026-10-11T20:49:30.519Z", "loss": 1.57753324508667, "grad_norm": 2.605267180253999, "lr": 2.9926617211952793e-07, "mem": 87.21397352218628, "tps": 9295.645868482792, "tps_per_gpu": 1161.955733560349, "mfu": 0.0, "num_tokens_per_step": 93408, "num_label_tokens": 93408}
{"step": 241, "epoch": 0, "timestamp": "2026-10-11T20:49:37.575Z", "loss": 1.9702389240264893, "grad_norm": 2.851972619835036, "lr": 2.992584774780527e-07, "mem": 86.13554286956787, "tps": 12514.846619839991, "tps_per_gpu": 1564.3558274799989, "mfu": 0.0, "num_tokens_per_step": 88294, "num_label_tokens": 88294}
{"step": 242, "epoch": 0, "timestamp": "2026-10-11T20:49:45.195Z", "loss": 2.0230929851531982, "grad_norm": 2.757577345577828, "lr": 2.9925074285527466e-07, "mem": 86.13409805297852, "tps": 12097.055361259561, "tps_per_gpu": 1512.1319201574452, "mfu": 0.0, "num_tokens_per_step": 92142, "num_label_tokens": 92142}
{"step": 243, "epoch": 0, "timestamp": "2026-10-11T20:49:52.763Z", "loss": 1.8464717864990234, "grad_norm": 2.753465567654673, "lr": 2.9924296825430925e-07, "mem": 85.05118942260742, "tps": 12090.516376352003, "tps_per_gpu": 1511.3145470440004, "mfu": 0.0, "num_tokens_per_step": 91542, "num_label_tokens": 91542}
{"step": 244, "epoch": 0, "timestamp": "2026-10-11T20:50:00.325Z", "loss": 1.4317137002944946, "grad_norm": 2.348458847866324, "lr": 2.9923515367828807e-07, "mem": 82.91947078704834, "tps": 12175.861096155184, "tps_per_gpu": 1521.982637019398, "mfu": 0.0, "num_tokens_per_step": 92060, "num_label_tokens": 92060}
{"step": 245, "epoch": 0, "timestamp": "2026-10-11T20:50:07.915Z", "loss": 1.259578824043274, "grad_norm": 2.152748354142031, "lr": 2.992272991303589e-07, "mem": 83.98416185379028, "tps": 12765.752446102495, "tps_per_gpu": 1595.7190557628119, "mfu": 0.0, "num_tokens_per_step": 96871, "num_label_tokens": 96871}
{"step": 246, "epoch": 0, "timestamp": "2026-10-11T20:50:15.577Z", "loss": 1.3487331867218018, "grad_norm": 2.3219987448436465, "lr": 2.992194046136855e-07, "mem": 86.13409996032715, "tps": 12518.72327703074, "tps_per_gpu": 1564.8404096288425, "mfu": 0.0, "num_tokens_per_step": 95944, "num_label_tokens": 95944}
{"step": 247, "epoch": 0, "timestamp": "2026-10-11T20:50:23.145Z", "loss": 1.2158279418945312, "grad_norm": 2.201669027028036, "lr": 2.992114701314478e-07, "mem": 85.05119132995605, "tps": 11741.082997446592, "tps_per_gpu": 1467.635374680824, "mfu": 0.0, "num_tokens_per_step": 88849, "num_label_tokens": 88849}
{"step": 248, "epoch": 0, "timestamp": "2026-10-11T20:50:30.705Z", "loss": 1.2814481258392334, "grad_norm": 2.3646149255049322, "lr": 2.9920349568684176e-07, "mem": 86.13410091400146, "tps": 12279.988970355194, "tps_per_gpu": 1534.9986212943993, "mfu": 0.0, "num_tokens_per_step": 92834, "num_label_tokens": 92834}
{"step": 249, "epoch": 0, "timestamp": "2026-10-11T20:50:38.237Z", "loss": 1.3255624771118164, "grad_norm": 2.5180522448316776, "lr": 2.991954812830795e-07, "mem": 82.91947317123413, "tps": 12163.89373841751, "tps_per_gpu": 1520.4867173021887, "mfu": 0.0, "num_tokens_per_step": 91619, "num_label_tokens": 91619}
{"step": 250, "epoch": 0, "timestamp": "2026-10-11T20:50:45.817Z", "loss": 1.3210666179656982, "grad_norm": 2.346534796787263, "lr": 2.991874269233893e-07, "mem": 87.21397829055786, "tps": 12179.81659993402, "tps_per_gpu": 1522.4770749917525, "mfu": 0.0, "num_tokens_per_step": 92320, "num_label_tokens": 92320}
{"step": 251, "epoch": 0, "timestamp": "2026-10-11T20:50:53.457Z", "loss": 1.3412141799926758, "grad_norm": 2.360418674211901, "lr": 2.991793326110154e-07, "mem": 86.13410234451294, "tps": 11982.273568571873, "tps_per_gpu": 1497.784196071484, "mfu": 0.0, "num_tokens_per_step": 91532, "num_label_tokens": 91532}
{"step": 252, "epoch": 0, "timestamp": "2026-10-11T20:51:00.593Z", "loss": 1.3063260316848755, "grad_norm": 2.28608196998781, "lr": 2.9917119834921813e-07, "mem": 85.05119371414185, "tps": 12598.52873712008, "tps_per_gpu": 1574.81609214001, "mfu": 0.0, "num_tokens_per_step": 89941, "num_label_tokens": 89941}
{"step": 253, "epoch": 0, "timestamp": "2026-10-11T20:51:08.202Z", "loss": 1.337470531463623, "grad_norm": 2.2573690512245608, "lr": 2.9916302414127403e-07, "mem": 87.21397972106934, "tps": 12287.196278476242, "tps_per_gpu": 1535.8995348095302, "mfu": 0.0, "num_tokens_per_step": 93468, "num_label_tokens": 93468}
{"step": 254, "epoch": 0, "timestamp": "2026-10-11T20:51:15.789Z", "loss": 1.28476083278656, "grad_norm": 2.2764452760466303, "lr": 2.991548099904757e-07, "mem": 81.84389638900757, "tps": 12085.803172227774, "tps_per_gpu": 1510.7253965284717, "mfu": 0.0, "num_tokens_per_step": 91687, "num_label_tokens": 91687}
{"step": 255, "epoch": 0, "timestamp": "2026-10-11T20:51:23.427Z", "loss": 1.266070008277893, "grad_norm": 2.3104199508632766, "lr": 2.991465559001318e-07, "mem": 87.21398067474365, "tps": 12225.486844788158, "tps_per_gpu": 1528.1858555985198, "mfu": 0.0, "num_tokens_per_step": 93384, "num_label_tokens": 93384}
{"step": 256, "epoch": 0, "timestamp": "2026-10-11T20:51:30.511Z", "loss": 1.3413056135177612, "grad_norm": 2.2762186399022633, "lr": 2.9913826187356695e-07, "mem": 83.98416709899902, "tps": 12814.942774182553, "tps_per_gpu": 1601.8678467728191, "mfu": 0.0, "num_tokens_per_step": 90785, "num_label_tokens": 90785}
{"step": 257, "epoch": 0, "timestamp": "2026-10-11T20:51:38.096Z", "loss": 1.3568788766860962, "grad_norm": 2.30479043568583, "lr": 2.991299279141221e-07, "mem": 86.13410520553589, "tps": 12149.423741006864, "tps_per_gpu": 1518.677967625858, "mfu": 0.0, "num_tokens_per_step": 92137, "num_label_tokens": 92137}
{"step": 258, "epoch": 0, "timestamp": "2026-10-11T20:51:48.281Z", "loss": 1.3155485391616821, "grad_norm": 2.2468423137547995, "lr": 2.9912155402515415e-07, "mem": 86.13410568237305, "tps": 9031.316435905313, "tps_per_gpu": 1128.9145544881642, "mfu": 0.0, "num_tokens_per_step": 92005, "num_label_tokens": 92005}
{"step": 259, "epoch": 0, "timestamp": "2026-10-11T20:51:58.665Z", "loss": 1.4330874681472778, "grad_norm": 2.4987143042602606, "lr": 2.991131402100361e-07, "mem": 87.21398258209229, "tps": 8675.582366770748, "tps_per_gpu": 1084.4477958463435, "mfu": 0.0, "num_tokens_per_step": 90084, "num_label_tokens": 90084}
{"step": 260, "epoch": 0, "timestamp": "2026-10-11T20:52:08.632Z", "loss": 1.440597653388977, "grad_norm": 2.9119115843397423, "lr": 2.99104686472157e-07, "mem": 87.21398305892944, "tps": 8943.356720082653, "tps_per_gpu": 1117.9195900103316, "mfu": 0.0, "num_tokens_per_step": 89124, "num_label_tokens": 89124}
{"step": 261, "epoch": 0, "timestamp": "2026-10-11T20:52:15.784Z", "loss": 1.4490305185317993, "grad_norm": 2.939585356411383, "lr": 2.990961928149221e-07, "mem": 85.05119800567627, "tps": 13010.820246840613, "tps_per_gpu": 1626.3525308550766, "mfu": 0.0, "num_tokens_per_step": 93089, "num_label_tokens": 93089}
{"step": 262, "epoch": 0, "timestamp": "2026-10-11T20:52:22.861Z", "loss": 1.2992998361587524, "grad_norm": 2.500234306677895, "lr": 2.9908765924175257e-07, "mem": 81.84390020370483, "tps": 12841.340061920297, "tps_per_gpu": 1605.167507740037, "mfu": 0.0, "num_tokens_per_step": 90875, "num_label_tokens": 90875}
{"step": 263, "epoch": 0, "timestamp": "2026-10-11T20:52:30.402Z", "loss": 1.248408317565918, "grad_norm": 2.3332491864340126, "lr": 2.990790857560857e-07, "mem": 87.21398448944092, "tps": 11559.097810012552, "tps_per_gpu": 1444.887226251569, "mfu": 0.0, "num_tokens_per_step": 87193, "num_label_tokens": 87193}
{"step": 264, "epoch": 0, "timestamp": "2026-10-11T20:52:38.013Z", "loss": 1.2810614109039307, "grad_norm": 2.198218245902946, "lr": 2.9907047236137497e-07, "mem": 86.134108543396, "tps": 12215.991898827351, "tps_per_gpu": 1526.998987353419, "mfu": 0.0, "num_tokens_per_step": 92931, "num_label_tokens": 92931}
{"step": 265, "epoch": 0, "timestamp": "2026-10-11T20:52:45.503Z", "loss": 1.3447760343551636, "grad_norm": 2.5397370782127955, "lr": 2.9906181906108983e-07, "mem": 83.98417139053345, "tps": 12100.623460936522, "tps_per_gpu": 1512.5779326170652, "mfu": 0.0, "num_tokens_per_step": 90636, "num_label_tokens": 90636}
{"step": 266, "epoch": 0, "timestamp": "2026-10-11T20:52:53.105Z", "loss": 1.5669808387756348, "grad_norm": 2.723855856948629, "lr": 2.990531258587158e-07, "mem": 87.21398591995239, "tps": 12299.702199548232, "tps_per_gpu": 1537.462774943529, "mfu": 0.0, "num_tokens_per_step": 93504, "num_label_tokens": 93504}
{"step": 267, "epoch": 0, "timestamp": "2026-10-11T20:53:00.640Z", "loss": 1.3689364194869995, "grad_norm": 2.9407142275680456, "lr": 2.990443927577545e-07, "mem": 87.21398639678955, "tps": 11126.922774633358, "tps_per_gpu": 1390.8653468291698, "mfu": 0.0, "num_tokens_per_step": 83852, "num_label_tokens": 83852}
{"step": 268, "epoch": 0, "timestamp": "2026-10-11T20:53:08.221Z", "loss": 1.3449008464813232, "grad_norm": 2.4892769722584935, "lr": 2.9903561976172365e-07, "mem": 83.98417282104492, "tps": 11992.423488477742, "tps_per_gpu": 1499.0529360597177, "mfu": 0.0, "num_tokens_per_step": 90925, "num_label_tokens": 90925}
{"step": 269, "epoch": 0, "timestamp": "2026-10-11T20:53:15.875Z", "loss": 1.301695704460144, "grad_norm": 2.3276414605763103, "lr": 2.99026806874157e-07, "mem": 85.05120182037354, "tps": 11860.47717666031, "tps_per_gpu": 1482.5596470825387, "mfu": 0.0, "num_tokens_per_step": 90730, "num_label_tokens": 90730}
{"step": 270, "epoch": 0, "timestamp": "2026-10-11T20:53:22.989Z", "loss": 1.2825500965118408, "grad_norm": 2.2509137924204063, "lr": 2.990179540986044e-07, "mem": 80.76887035369873, "tps": 12884.213177424124, "tps_per_gpu": 1610.5266471780155, "mfu": 0.0, "num_tokens_per_step": 91668, "num_label_tokens": 91668}
{"step": 271, "epoch": 0, "timestamp": "2026-10-11T20:53:30.602Z", "loss": 1.3331950902938843, "grad_norm": 2.3615543278727977, "lr": 2.9900906143863173e-07, "mem": 83.9841742515564, "tps": 12114.192952377547, "tps_per_gpu": 1514.2741190471934, "mfu": 0.0, "num_tokens_per_step": 92256, "num_label_tokens": 92256}
{"step": 272, "epoch": 0, "timestamp": "2026-10-11T20:53:38.188Z", "loss": 1.3788740634918213, "grad_norm": 2.7223664080496595, "lr": 2.99000128897821e-07, "mem": 82.91948413848877, "tps": 11764.282123868728, "tps_per_gpu": 1470.535265483591, "mfu": 0.0, "num_tokens_per_step": 89241, "num_label_tokens": 89241}
{"step": 273, "epoch": 0, "timestamp": "2026-10-11T20:53:45.751Z", "loss": 1.4067496061325073, "grad_norm": 3.0889636390798203, "lr": 2.989911564797701e-07, "mem": 80.7688717842102, "tps": 11997.923640570603, "tps_per_gpu": 1499.7404550713254, "mfu": 0.0, "num_tokens_per_step": 90756, "num_label_tokens": 90756}
{"step": 274, "epoch": 0, "timestamp": "2026-10-11T20:53:53.299Z", "loss": 1.3490699529647827, "grad_norm": 2.651781634855357, "lr": 2.9898214418809324e-07, "mem": 82.91948509216309, "tps": 12101.873334684498, "tps_per_gpu": 1512.7341668355623, "mfu": 0.0, "num_tokens_per_step": 91337, "num_label_tokens": 91337}
{"step": 275, "epoch": 0, "timestamp": "2026-10-11T20:54:00.819Z", "loss": 1.4497038125991821, "grad_norm": 2.665061871428744, "lr": 2.989730920264206e-07, "mem": 86.13411378860474, "tps": 11957.72619310456, "tps_per_gpu": 1494.71577413807, "mfu": 0.0, "num_tokens_per_step": 89904, "num_label_tokens": 89904}
{"step": 276, "epoch": 0, "timestamp": "2026-10-11T20:54:08.278Z", "loss": 1.341806173324585, "grad_norm": 2.6979623348976958, "lr": 2.9896399999839824e-07, "mem": 81.84390687942505, "tps": 11797.72227001848, "tps_per_gpu": 1474.71528375231, "mfu": 0.0, "num_tokens_per_step": 88031, "num_label_tokens": 88031}
{"step": 277, "epoch": 0, "timestamp": "2026-10-11T20:54:15.925Z", "loss": 1.3624393939971924, "grad_norm": 2.5629464612413497, "lr": 2.989548681076886e-07, "mem": 86.13411474227905, "tps": 11951.07536178253, "tps_per_gpu": 1493.8844202228163, "mfu": 0.0, "num_tokens_per_step": 91356, "num_label_tokens": 91356}
{"step": 278, "epoch": 0, "timestamp": "2026-10-11T20:54:26.179Z", "loss": 1.283730149269104, "grad_norm": 2.446844710690297, "lr": 2.9894569635796985e-07, "mem": 85.05120611190796, "tps": 8928.803187055222, "tps_per_gpu": 1116.1003983819028, "mfu": 0.0, "num_tokens_per_step": 91553, "num_label_tokens": 91553}
{"step": 279, "epoch": 0, "timestamp": "2026-10-11T20:54:35.711Z", "loss": 1.2087467908859253, "grad_norm": 2.2449124798523568, "lr": 2.9893648475293643e-07, "mem": 80.76216077804565, "tps": 9431.571453853436, "tps_per_gpu": 1178.9464317316795, "mfu": 0.0, "num_tokens_per_step": 89922, "num_label_tokens": 89922}
{"step": 280, "epoch": 0, "timestamp": "2026-10-11T20:54:45.232Z", "loss": 1.2956993579864502, "grad_norm": 2.3515238570010535, "lr": 2.989272332962988e-07, "mem": 85.05120706558228, "tps": 9608.703314276027, "tps_per_gpu": 1201.0879142845033, "mfu": 0.0, "num_tokens_per_step": 91478, "num_label_tokens": 91478}
{"step": 281, "epoch": 0, "timestamp": "2026-10-11T20:54:52.866Z", "loss": 1.2923694849014282, "grad_norm": 2.2970042563955775, "lr": 2.9891794199178344e-07, "mem": 87.21399307250977, "tps": 11996.34808052118, "tps_per_gpu": 1499.5435100651475, "mfu": 0.0, "num_tokens_per_step": 91560, "num_label_tokens": 91560}
{"step": 282, "epoch": 0, "timestamp": "2026-10-11T20:55:00.453Z", "loss": 1.320482611656189, "grad_norm": 2.392497829047071, "lr": 2.989086108431329e-07, "mem": 86.13411712646484, "tps": 11562.249817412667, "tps_per_gpu": 1445.2812271765833, "mfu": 0.0, "num_tokens_per_step": 87730, "num_label_tokens": 87730}
{"step": 283, "epoch": 0, "timestamp": "2026-10-11T20:55:08.043Z", "loss": 1.2531399726867676, "grad_norm": 2.1966870537242746, "lr": 2.9889923985410573e-07, "mem": 87.21399402618408, "tps": 12505.101292723788, "tps_per_gpu": 1563.1376615904735, "mfu": 0.0, "num_tokens_per_step": 94928, "num_label_tokens": 94928}
{"step": 284, "epoch": 0, "timestamp": "2026-10-11T20:55:15.497Z", "loss": 1.797574520111084, "grad_norm": 3.233886011825128, "lr": 2.9888982902847657e-07, "mem": 83.98418045043945, "tps": 11843.45919752435, "tps_per_gpu": 1480.4323996905437, "mfu": 0.0, "num_tokens_per_step": 88271, "num_label_tokens": 88271}
{"step": 285, "epoch": 0, "timestamp": "2026-10-11T20:55:23.014Z", "loss": 1.5212286710739136, "grad_norm": 2.4147687174812, "lr": 2.9888037837003607e-07, "mem": 87.2139949798584, "tps": 11962.55486643182, "tps_per_gpu": 1495.3193583039774, "mfu": 0.0, "num_tokens_per_step": 89912, "num_label_tokens": 89912}
{"step": 286, "epoch": 0, "timestamp": "2026-10-11T20:55:30.605Z", "loss": 1.3057204484939575, "grad_norm": 2.221123730738431, "lr": 2.9887088788259104e-07, "mem": 87.21399545669556, "tps": 12459.551126077657, "tps_per_gpu": 1557.443890759707, "mfu": 0.0, "num_tokens_per_step": 94610, "num_label_tokens": 94610}
{"step": 287, "epoch": 0, "timestamp": "2026-10-11T20:55:38.264Z", "loss": 1.3553845882415771, "grad_norm": 2.3839082310397974, "lr": 2.9886135756996415e-07, "mem": 83.98418188095093, "tps": 12024.066580491446, "tps_per_gpu": 1503.0083225614308, "mfu": 0.0, "num_tokens_per_step": 92074, "num_label_tokens": 92074}
{"step": 288, "epoch": 0, "timestamp": "2026-10-11T20:55:45.849Z", "loss": 1.1729005575180054, "grad_norm": 2.1699120427544054, "lr": 2.9885178743599425e-07, "mem": 86.1341199874878, "tps": 12325.050782754608, "tps_per_gpu": 1540.631347844326, "mfu": 0.0, "num_tokens_per_step": 93475, "num_label_tokens": 93475}
{"step": 289, "epoch": 0, "timestamp": "2026-10-11T20:55:53.380Z", "loss": 1.3011616468429565, "grad_norm": 2.361586699114576, "lr": 2.988421774845362e-07, "mem": 75.40296840667725, "tps": 11863.022556029415, "tps_per_gpu": 1482.8778195036768, "mfu": 0.0, "num_tokens_per_step": 88892, "num_label_tokens": 88892}
{"step": 290, "epoch": 0, "timestamp": "2026-10-11T20:56:00.960Z", "loss": 1.3656338453292847, "grad_norm": 2.25092418776675, "lr": 2.9883252771946094e-07, "mem": 80.7688798904419, "tps": 12349.0676251874, "tps_per_gpu": 1543.633453148425, "mfu": 0.0, "num_tokens_per_step": 94087, "num_label_tokens": 94087}
{"step": 291, "epoch": 0, "timestamp": "2026-10-11T20:56:08.615Z", "loss": 1.4429417848587036, "grad_norm": 2.5252655039677085, "lr": 2.988228381446553e-07, "mem": 83.98418378829956, "tps": 11753.33032164242, "tps_per_gpu": 1469.1662902053024, "mfu": 0.0, "num_tokens_per_step": 89983, "num_label_tokens": 89983}
{"step": 292, "epoch": 0, "timestamp": "2026-10-11T20:56:16.195Z", "loss": 2.0102078914642334, "grad_norm": 2.8996543450328276, "lr": 2.9881310876402225e-07, "mem": 87.2139983177185, "tps": 11662.090568540594, "tps_per_gpu": 1457.7613210675743, "mfu": 0.0, "num_tokens_per_step": 88400, "num_label_tokens": 88400}
{"step": 293, "epoch": 0, "timestamp": "2026-10-11T20:56:23.776Z", "loss": 2.0525407791137695, "grad_norm": 2.796798271865149, "lr": 2.988033395814808e-07, "mem": 83.98418474197388, "tps": 12080.627944607622, "tps_per_gpu": 1510.0784930759528, "mfu": 0.0, "num_tokens_per_step": 91587, "num_label_tokens": 91587}
{"step": 294, "epoch": 0, "timestamp": "2026-10-11T20:56:31.329Z", "loss": 2.0097663402557373, "grad_norm": 2.981577807041368, "lr": 2.98793530600966e-07, "mem": 85.05121374130249, "tps": 11672.849217443516, "tps_per_gpu": 1459.1061521804395, "mfu": 0.0, "num_tokens_per_step": 88165, "num_label_tokens": 88165}
{"step": 295, "epoch": 0, "timestamp": "2026-10-11T20:56:38.915Z", "loss": 2.06673002243042, "grad_norm": 2.755423464944216, "lr": 2.9878368182642886e-07, "mem": 81.84391593933105, "tps": 11779.161588320028, "tps_per_gpu": 1472.3951985400035, "mfu": 0.0, "num_tokens_per_step": 89357, "num_label_tokens": 89357}
{"step": 296, "epoch": 0, "timestamp": "2026-10-11T20:56:46.020Z", "loss": 2.05088472366333, "grad_norm": 2.919646144919302, "lr": 2.9877379326183653e-07, "mem": 86.13556909561157, "tps": 12278.202888436197, "tps_per_gpu": 1534.7753610545246, "mfu": 0.0, "num_tokens_per_step": 87225, "num_label_tokens": 87225}
{"step": 297, "epoch": 0, "timestamp": "2026-10-11T20:56:53.567Z", "loss": 2.030245304107666, "grad_norm": 2.915743275865324, "lr": 2.987638649111721e-07, "mem": 81.84391689300537, "tps": 11654.742037001868, "tps_per_gpu": 1456.8427546252335, "mfu": 0.0, "num_tokens_per_step": 87943, "num_label_tokens": 87943}
{"step": 298, "epoch": 0, "timestamp": "2026-10-11T20:57:00.665Z", "loss": 1.2057801485061646, "grad_norm": 2.2406372019193133, "lr": 2.9875389677843466e-07, "mem": 82.91949653625488, "tps": 12876.058546394514, "tps_per_gpu": 1609.5073182993142, "mfu": 0.0, "num_tokens_per_step": 91404, "num_label_tokens": 91404}
{"step": 299, "epoch": 0, "timestamp": "2026-10-11T20:57:10.913Z", "loss": 1.4054332971572876, "grad_norm": 2.2830210820577217, "lr": 2.987438888676394e-07, "mem": 86.13412523269653, "tps": 9135.636992733043, "tps_per_gpu": 1141.9546240916304, "mfu": 0.0, "num_tokens_per_step": 93615, "num_label_tokens": 93615}
{"step": 300, "epoch": 0, "timestamp": "2026-10-11T20:58:05.007Z", "loss": 1.357941746711731, "grad_norm": 2.3532393837930488, "lr": 2.9873384118281754e-07, "mem": 85.0465612411499, "tps": 1676.0918596989222, "tps_per_gpu": 209.51148246236528, "mfu": 0.0, "num_tokens_per_step": 90673, "num_label_tokens": 90673}
{"step": 301, "epoch": 0, "timestamp": "2026-10-11T20:58:13.664Z", "loss": 1.8079462051391602, "grad_norm": 2.687957787991689, "lr": 2.9872375372801627e-07, "mem": 83.98414087295532, "tps": 10286.626083583453, "tps_per_gpu": 1285.8282604479316, "mfu": 0.0, "num_tokens_per_step": 89035, "num_label_tokens": 89035}
{"step": 302, "epoch": 0, "timestamp": "2026-10-11T20:58:21.246Z", "loss": 1.532355546951294, "grad_norm": 2.9543659182219675, "lr": 2.987136265072988e-07, "mem": 81.84387159347534, "tps": 11643.05566634412, "tps_per_gpu": 1455.381958293015, "mfu": 0.0, "num_tokens_per_step": 88266, "num_label_tokens": 88266}
{"step": 303, "epoch": 0, "timestamp": "2026-10-11T20:58:28.795Z", "loss": 1.4819005727767944, "grad_norm": 2.784206533893309, "lr": 2.9870345952474437e-07, "mem": 85.0511703491211, "tps": 12134.908960490238, "tps_per_gpu": 1516.8636200612798, "mfu": 0.0, "num_tokens_per_step": 91598, "num_label_tokens": 91598}
{"step": 304, "epoch": 0, "timestamp": "2026-10-11T20:58:35.760Z", "loss": 1.6046364307403564, "grad_norm": 2.8743846583906105, "lr": 2.986932527844482e-07, "mem": 80.76883888244629, "tps": 12368.462172238276, "tps_per_gpu": 1546.0577715297845, "mfu": 0.0, "num_tokens_per_step": 86142, "num_label_tokens": 86142}
{"step": 305, "epoch": 0, "timestamp": "2026-10-11T20:58:42.935Z", "loss": 2.136017322540283, "grad_norm": 2.7894830234770662, "lr": 2.986830062905217e-07, "mem": 85.05117130279541, "tps": 12642.192316023078, "tps_per_gpu": 1580.2740395028848, "mfu": 0.0, "num_tokens_per_step": 90720, "num_label_tokens": 90720}
{"step": 306, "epoch": 0, "timestamp": "2026-10-11T20:58:50.443Z", "loss": 1.510349988937378, "grad_norm": 2.6244913339399094, "lr": 2.98672720047092e-07, "mem": 80.7688398361206, "tps": 11912.636231834304, "tps_per_gpu": 1489.079528979288, "mfu": 0.0, "num_tokens_per_step": 89458, "num_label_tokens": 89458}
{"step": 307, "epoch": 0, "timestamp": "2026-10-11T20:58:58.031Z", "loss": 1.4116345643997192, "grad_norm": 2.520250772362132, "lr": 2.986623940583025e-07, "mem": 87.21395778656006, "tps": 12467.645261818327, "tps_per_gpu": 1558.4556577272908, "mfu": 0.0, "num_tokens_per_step": 94594, "num_label_tokens": 94594}
{"step": 308, "epoch": 0, "timestamp": "2026-10-11T20:59:05.602Z", "loss": 1.4669307470321655, "grad_norm": 2.6842627110123964, "lr": 2.9865202832831245e-07, "mem": 81.84387445449829, "tps": 11879.318257775434, "tps_per_gpu": 1484.9147822219293, "mfu": 0.0, "num_tokens_per_step": 89905, "num_label_tokens": 89905}
{"step": 309, "epoch": 0, "timestamp": "2026-10-11T20:59:13.270Z", "loss": 1.435701847076416, "grad_norm": 2.7244002340693307, "lr": 2.986416228612972e-07, "mem": 86.1340823173523, "tps": 12149.211706990569, "tps_per_gpu": 1518.6514633738211, "mfu": 0.0, "num_tokens_per_step": 93187, "num_label_tokens": 93187}
{"step": 310, "epoch": 0, "timestamp": "2026-10-11T20:59:20.785Z", "loss": 1.419723391532898, "grad_norm": 2.6074583640576434, "lr": 2.9863117766144803e-07, "mem": 87.21395921707153, "tps": 11414.823226654622, "tps_per_gpu": 1426.8529033318277, "mfu": 0.0, "num_tokens_per_step": 85784, "num_label_tokens": 85784}
{"step": 311, "epoch": 0, "timestamp": "2026-10-11T20:59:28.345Z", "loss": 1.5251145362854004, "grad_norm": 2.630508648770957, "lr": 2.986206927329723e-07, "mem": 85.05117416381836, "tps": 11986.524400690581, "tps_per_gpu": 1498.3155500863227, "mfu": 0.0, "num_tokens_per_step": 90623, "num_label_tokens": 90623}
{"step": 312, "epoch": 0, "timestamp": "2026-10-11T20:59:35.907Z", "loss": 1.8388571739196777, "grad_norm": 2.7828642381983246, "lr": 2.9861016808009334e-07, "mem": 86.13408374786377, "tps": 11918.510849667637, "tps_per_gpu": 1489.8138562084546, "mfu": 0.0, "num_tokens_per_step": 90130, "num_label_tokens": 90130}
{"step": 313, "epoch": 0, "timestamp": "2026-10-11T20:59:43.468Z", "loss": 1.9558665752410889, "grad_norm": 2.824544936748544, "lr": 2.985996037070505e-07, "mem": 87.21396064758301, "tps": 12345.130196867985, "tps_per_gpu": 1543.1412746084982, "mfu": 0.0, "num_tokens_per_step": 93349, "num_label_tokens": 93349}
{"step": 314, "epoch": 0, "timestamp": "2026-10-11T20:59:50.970Z", "loss": 1.9926031827926636, "grad_norm": 2.82126368930971, "lr": 2.98588999618099e-07, "mem": 83.98414707183838, "tps": 11903.509927475327, "tps_per_gpu": 1487.9387409344158, "mfu": 0.0, "num_tokens_per_step": 89281, "num_label_tokens": 89281}
{"step": 315, "epoch": 0, "timestamp": "2026-10-11T20:59:58.456Z", "loss": 1.9282050132751465, "grad_norm": 2.731640769077903, "lr": 2.9857835581751033e-07, "mem": 79.69417190551758, "tps": 11916.296837623913, "tps_per_gpu": 1489.5371047029892, "mfu": 0.0, "num_tokens_per_step": 89217, "num_label_tokens": 89217}
{"step": 316, "epoch": 0, "timestamp": "2026-10-11T21:00:06.117Z", "loss": 1.9571391344070435, "grad_norm": 2.7857665170496886, "lr": 2.9856767230957173e-07, "mem": 87.21396207809448, "tps": 11816.029540910857, "tps_per_gpu": 1477.003692613857, "mfu": 0.0, "num_tokens_per_step": 90516, "num_label_tokens": 90516}
{"step": 317, "epoch": 0, "timestamp": "2026-10-11T21:00:13.719Z", "loss": 2.067148447036743, "grad_norm": 2.788314503861375, "lr": 2.985569490985864e-07, "mem": 87.21396255493164, "tps": 12055.566665267263, "tps_per_gpu": 1506.9458331584078, "mfu": 0.0, "num_tokens_per_step": 91643, "num_label_tokens": 91643}
{"step": 318, "epoch": 0, "timestamp": "2026-10-11T21:00:24.462Z", "loss": 1.8731578588485718, "grad_norm": 3.0182652336019733, "lr": 2.9854618618887386e-07, "mem": 85.05117750167847, "tps": 8097.401832217504, "tps_per_gpu": 1012.175229027188, "mfu": 0.0, "num_tokens_per_step": 86998, "num_label_tokens": 86998}
{"step": 319, "epoch": 0, "timestamp": "2026-10-11T21:00:34.479Z", "loss": 2.0158326625823975, "grad_norm": 2.8078122965096557, "lr": 2.985353835847693e-07, "mem": 79.69417381286621, "tps": 8945.58698405904, "tps_per_gpu": 1118.19837300738, "mfu": 0.0, "num_tokens_per_step": 89621, "num_label_tokens": 89621}
{"step": 320, "epoch": 0, "timestamp": "2026-10-11T21:00:44.729Z", "loss": 1.429748773574829, "grad_norm": 3.421330886335405, "lr": 2.9852454129062395e-07, "mem": 81.84388017654419, "tps": 8743.668291209546, "tps_per_gpu": 1092.9585364011932, "mfu": 0.0, "num_tokens_per_step": 89608, "num_label_tokens": 89608}
{"step": 321, "epoch": 0, "timestamp": "2026-10-11T21:00:52.255Z", "loss": 1.4183582067489624, "grad_norm": 2.702462533624323, "lr": 2.9851365931080527e-07, "mem": 81.84388065338135, "tps": 11539.34442911863, "tps_per_gpu": 1442.4180536398287, "mfu": 0.0, "num_tokens_per_step": 86845, "num_label_tokens": 86845}
{"step": 322, "epoch": 0, "timestamp": "2026-10-11T21:00:59.841Z", "loss": 1.4407424926757812, "grad_norm": 2.5840375157305946, "lr": 2.985027376496963e-07, "mem": 85.0511794090271, "tps": 11954.792508628716, "tps_per_gpu": 1494.3490635785895, "mfu": 0.0, "num_tokens_per_step": 90706, "num_label_tokens": 90706}
{"step": 323, "epoch": 0, "timestamp": "2026-10-11T21:01:07.496Z", "loss": 1.25514817237854, "grad_norm": 2.2643091932010453, "lr": 2.984917763116964e-07, "mem": 86.13408899307251, "tps": 12042.221441816053, "tps_per_gpu": 1505.2776802270066, "mfu": 0.0, "num_tokens_per_step": 92150, "num_label_tokens": 92150}
{"step": 324, "epoch": 0, "timestamp": "2026-10-11T21:01:15.012Z", "loss": 1.2590582370758057, "grad_norm": 2.2080607085002093, "lr": 2.984807753012208e-07, "mem": 83.98415184020996, "tps": 12226.644038173696, "tps_per_gpu": 1528.330504771712, "mfu": 0.0, "num_tokens_per_step": 91905, "num_label_tokens": 91905}
{"step": 325, "epoch": 0, "timestamp": "2026-10-11T21:01:22.573Z", "loss": 1.2257201671600342, "grad_norm": 2.1913721298566715, "lr": 2.9846973462270065e-07, "mem": 87.2139663696289, "tps": 12487.656885830975, "tps_per_gpu": 1560.9571107288718, "mfu": 0.0, "num_tokens_per_step": 94444, "num_label_tokens": 94444}
{"step": 326, "epoch": 0, "timestamp": "2026-10-11T21:01:30.131Z", "loss": 1.2876980304718018, "grad_norm": 2.2760923406489693, "lr": 2.9845865428058324e-07, "mem": 83.98415279388428, "tps": 12237.824265514006, "tps_per_gpu": 1529.7280331892507, "mfu": 0.0, "num_tokens_per_step": 92498, "num_label_tokens": 92498}
{"step": 327, "epoch": 0, "timestamp": "2026-10-11T21:01:37.684Z", "loss": 1.2707737684249878, "grad_norm": 2.2528036041846793, "lr": 2.984475342793316e-07, "mem": 82.91946268081665, "tps": 12454.171259126095, "tps_per_gpu": 1556.7714073907619, "mfu": 0.0, "num_tokens_per_step": 94038, "num_label_tokens": 94038}
{"step": 328, "epoch": 0, "timestamp": "2026-10-11T21:01:44.733Z", "loss": 1.283106803894043, "grad_norm": 2.245724084131357, "lr": 2.984363746234249e-07, "mem": 83.9841537475586, "tps": 12899.595645734766, "tps_per_gpu": 1612.4494557168457, "mfu": 0.0, "num_tokens_per_step": 90941, "num_label_tokens": 90941}
{"step": 329, "epoch": 0, "timestamp": "2026-10-11T21:01:51.779Z", "loss": 1.3333773612976074, "grad_norm": 2.347824246408989, "lr": 2.984251753173584e-07, "mem": 76.47955513000488, "tps": 12702.638418266884, "tps_per_gpu": 1587.8298022833606, "mfu": 0.0, "num_tokens_per_step": 89517, "num_label_tokens": 89517}
{"step": 330, "epoch": 0, "timestamp": "2026-10-11T21:01:59.376Z", "loss": 1.36785089969635, "grad_norm": 2.4699248539742316, "lr": 2.984139363656429e-07, "mem": 85.05118322372437, "tps": 12501.960167879071, "tps_per_gpu": 1562.745020984884, "mfu": 0.0, "num_tokens_per_step": 94972, "num_label_tokens": 94972}
{"step": 331, "epoch": 0, "timestamp": "2026-10-11T21:02:06.980Z", "loss": 1.307024359703064, "grad_norm": 2.2921799427863374, "lr": 2.984026577728057e-07, "mem": 83.98415517807007, "tps": 12047.802160323581, "tps_per_gpu": 1505.9752700404476, "mfu": 0.0, "num_tokens_per_step": 91626, "num_label_tokens": 91626}
{"step": 332, "epoch": 0, "timestamp": "2026-10-11T21:02:14.589Z", "loss": 1.3693034648895264, "grad_norm": 2.3321410883284726, "lr": 2.983913395433897e-07, "mem": 87.21396970748901, "tps": 12233.64545395135, "tps_per_gpu": 1529.2056817439188, "mfu": 0.0, "num_tokens_per_step": 93064, "num_label_tokens": 93064}
{"step": 333, "epoch": 0, "timestamp": "2026-10-11T21:02:22.180Z", "loss": 1.3837127685546875, "grad_norm": 2.4208095617355174, "lr": 2.9837998168195384e-07, "mem": 83.98415613174438, "tps": 12471.586256728333, "tps_per_gpu": 1558.9482820910416, "mfu": 0.0, "num_tokens_per_step": 94659, "num_label_tokens": 94659}
{"step": 334, "epoch": 0, "timestamp": "2026-10-11T21:02:29.738Z", "loss": 1.48969566822052, "grad_norm": 2.278285213948061, "lr": 2.983685841930732e-07, "mem": 86.13409423828125, "tps": 12402.847873391567, "tps_per_gpu": 1550.355984173946, "mfu": 0.0, "num_tokens_per_step": 93740, "num_label_tokens": 93740}
{"step": 335, "epoch": 0, "timestamp": "2026-10-11T21:02:36.667Z", "loss": 1.9320894479751587, "grad_norm": 2.8191138640346183, "lr": 2.983571470813386e-07, "mem": 81.84476900100708, "tps": 12212.701025412698, "tps_per_gpu": 1526.5876281765873, "mfu": 0.0, "num_tokens_per_step": 84616, "num_label_tokens": 84616}
{"step": 336, "epoch": 0, "timestamp": "2026-10-11T21:02:44.253Z", "loss": 1.2504891157150269, "grad_norm": 2.604771528769889, "lr": 2.983456703513569e-07, "mem": 85.05118608474731, "tps": 12065.096541695477, "tps_per_gpu": 1508.1370677119346, "mfu": 0.0, "num_tokens_per_step": 91547, "num_label_tokens": 91547}
{"step": 337, "epoch": 0, "timestamp": "2026-10-11T21:02:57.938Z", "loss": 1.510871171951294, "grad_norm": 2.7647206865454304, "lr": 2.983341540077509e-07, "mem": 86.13409566879272, "tps": 6586.535361345485, "tps_per_gpu": 823.3169201681857, "mfu": 0.0, "num_tokens_per_step": 90128, "num_label_tokens": 90128}
{"step": 338, "epoch": 0, "timestamp": "2026-10-11T21:03:07.916Z", "loss": 1.3915740251541138, "grad_norm": 2.608208891683058, "lr": 2.983225980551595e-07, "mem": 81.84388875961304, "tps": 9101.23283981128, "tps_per_gpu": 1137.65410497641, "mfu": 0.0, "num_tokens_per_step": 90815, "num_label_tokens": 90815}
{"step": 339, "epoch": 0, "timestamp": "2026-10-11T21:03:15.525Z", "loss": 1.4458943605422974, "grad_norm": 2.493782558135643, "lr": 2.9831100249823734e-07, "mem": 80.76885557174683, "tps": 12056.348428509138, "tps_per_gpu": 1507.0435535636423, "mfu": 0.0, "num_tokens_per_step": 91732, "num_label_tokens": 91732}
{"step": 340, "epoch": 0, "timestamp": "2026-10-11T21:03:23.034Z", "loss": 1.519487977027893, "grad_norm": 2.7940963245213464, "lr": 2.982993673416551e-07, "mem": 85.05118799209595, "tps": 12123.84248092031, "tps_per_gpu": 1515.4803101150387, "mfu": 0.0, "num_tokens_per_step": 91041, "num_label_tokens": 91041}
{"step": 341, "epoch": 0, "timestamp": "2026-10-11T21:03:30.583Z", "loss": 1.4430570602416992, "grad_norm": 2.625824922074781, "lr": 2.982876925900995e-07, "mem": 81.84389019012451, "tps": 11822.69107383855, "tps_per_gpu": 1477.8363842298188, "mfu": 0.0, "num_tokens_per_step": 89272, "num_label_tokens": 89272}
{"step": 342, "epoch": 0, "timestamp": "2026-10-11T21:03:38.186Z", "loss": 1.1827828884124756, "grad_norm": 2.1776871547432846, "lr": 2.9827597824827305e-07, "mem": 86.13409805297852, "tps": 12425.864785785589, "tps_per_gpu": 1553.2330982231986, "mfu": 0.0, "num_tokens_per_step": 94473, "num_label_tokens": 94473}
{"step": 343, "epoch": 0, "timestamp": "2026-10-11T21:03:45.797Z", "loss": 1.337180495262146, "grad_norm": 2.2901739901923115, "lr": 2.982642243208943e-07, "mem": 86.13409852981567, "tps": 12342.788095352336, "tps_per_gpu": 1542.848511919042, "mfu": 0.0, "num_tokens_per_step": 93908, "num_label_tokens": 93908}
{"step": 344, "epoch": 0, "timestamp": "2026-10-11T21:03:53.470Z", "loss": 1.2419912815093994, "grad_norm": 2.257550379805011, "lr": 2.982524308126977e-07, "mem": 86.13409900665283, "tps": 12028.61302494044, "tps_per_gpu": 1503.576628117555, "mfu": 0.0, "num_tokens_per_step": 92328, "num_label_tokens": 92328}
{"step": 345, "epoch": 0, "timestamp": "2026-10-11T21:04:00.570Z", "loss": 1.3997145891189575, "grad_norm": 2.6660476770638026, "lr": 2.982405977284338e-07, "mem": 86.1355447769165, "tps": 12895.440746279506, "tps_per_gpu": 1611.9300932849383, "mfu": 0.0, "num_tokens_per_step": 91519, "num_label_tokens": 91519}
{"step": 346, "epoch": 0, "timestamp": "2026-10-11T21:04:07.335Z", "loss": 1.3781440258026123, "grad_norm": 2.628055276109605, "lr": 2.9822872507286884e-07, "mem": 80.76214504241943, "tps": 12753.835626971457, "tps_per_gpu": 1594.229453371432, "mfu": 0.0, "num_tokens_per_step": 86284, "num_label_tokens": 86284}
{"step": 347, "epoch": 0, "timestamp": "2026-10-11T21:04:14.843Z", "loss": 1.4060765504837036, "grad_norm": 2.7828134174189962, "lr": 2.982168128507852e-07, "mem": 86.1341004371643, "tps": 12097.826329054808, "tps_per_gpu": 1512.228291131851, "mfu": 0.0, "num_tokens_per_step": 90828, "num_label_tokens": 90828}
{"step": 348, "epoch": 0, "timestamp": "2026-10-11T21:04:22.326Z", "loss": 1.4112910032272339, "grad_norm": 2.6773365455073783, "lr": 2.9820486106698105e-07, "mem": 85.05119180679321, "tps": 12493.77710338575, "tps_per_gpu": 1561.7221379232187, "mfu": 0.0, "num_tokens_per_step": 93520, "num_label_tokens": 93520}
{"step": 349, "epoch": 0, "timestamp": "2026-10-11T21:04:29.940Z", "loss": 1.4820077419281006, "grad_norm": 2.673763257074947, "lr": 2.9819286972627064e-07, "mem": 82.91947317123413, "tps": 12164.598380360116, "tps_per_gpu": 1520.5747975450145, "mfu": 0.0, "num_tokens_per_step": 92611, "num_label_tokens": 92611}
{"step": 350, "epoch": 0, "timestamp": "2026-10-11T21:04:37.060Z", "loss": 1.4318361282348633, "grad_norm": 2.5266742759857967, "lr": 2.9818083883348403e-07, "mem": 76.4795651435852, "tps": 12565.761602302035, "tps_per_gpu": 1570.7202002877543, "mfu": 0.0, "num_tokens_per_step": 88772, "num_label_tokens": 88772}
{"step": 351, "epoch": 0, "timestamp": "2026-10-11T21:04:44.605Z", "loss": 1.2393560409545898, "grad_norm": 2.250882674411414, "lr": 2.981687683934673e-07, "mem": 85.05119323730469, "tps": 12413.239324864515, "tps_per_gpu": 1551.6549156080644, "mfu": 0.0, "num_tokens_per_step": 94337, "num_label_tokens": 94337}
{"step": 352, "epoch": 0, "timestamp": "2026-10-11T21:04:52.182Z", "loss": 1.3797253370285034, "grad_norm": 2.442285681130088, "lr": 2.981566584110824e-07, "mem": 83.98416519165039, "tps": 12555.908966063535, "tps_per_gpu": 1569.4886207579418, "mfu": 0.0, "num_tokens_per_step": 95139, "num_label_tokens": 95139}
{"step": 353, "epoch": 0, "timestamp": "2026-10-11T21:04:59.843Z", "loss": 1.292540192604065, "grad_norm": 2.3165841087375614, "lr": 2.9814450889120724e-07, "mem": 87.21397972106934, "tps": 12162.31197879838, "tps_per_gpu": 1520.2889973497975, "mfu": 0.0, "num_tokens_per_step": 93190, "num_label_tokens": 93190}
{"step": 354, "epoch": 0, "timestamp": "2026-10-11T21:05:10.016Z", "loss": 1.9809246063232422, "grad_norm": 2.840674874602906, "lr": 2.981323198387356e-07, "mem": 80.7688627243042, "tps": 8778.378505746003, "tps_per_gpu": 1097.2973132182503, "mfu": 0.0, "num_tokens_per_step": 89291, "num_label_tokens": 89291}
{"step": 355, "epoch": 0, "timestamp": "2026-10-11T21:05:17.590Z", "loss": 1.422617793083191, "grad_norm": 2.424132081316111, "lr": 2.981200912585773e-07, "mem": 87.21398067474365, "tps": 11771.65754797047, "tps_per_gpu": 1471.4571934963087, "mfu": 0.0, "num_tokens_per_step": 89158, "num_label_tokens": 89158}
{"step": 356, "epoch": 0, "timestamp": "2026-10-11T21:05:25.080Z", "loss": 1.2648289203643799, "grad_norm": 2.2971038907411394, "lr": 2.981078231556579e-07, "mem": 80.76886367797852, "tps": 11933.73062650811, "tps_per_gpu": 1491.7163283135137, "mfu": 0.0, "num_tokens_per_step": 89372, "num_label_tokens": 89372}
{"step": 357, "epoch": 0, "timestamp": "2026-10-11T21:05:35.425Z", "loss": 1.4315752983093262, "grad_norm": 2.4549239375052387, "lr": 2.9809551553491913e-07, "mem": 87.21398162841797, "tps": 9256.721898741538, "tps_per_gpu": 1157.0902373426923, "mfu": 0.0, "num_tokens_per_step": 95687, "num_label_tokens": 95687}
{"step": 358, "epoch": 0, "timestamp": "2026-10-11T21:05:44.886Z", "loss": 1.2754273414611816, "grad_norm": 2.4330982138993464, "lr": 2.980831684013185e-07, "mem": 82.91947746276855, "tps": 9602.250246309522, "tps_per_gpu": 1200.2812807886903, "mfu": 0.0, "num_tokens_per_step": 90873, "num_label_tokens": 90873}
{"step": 359, "epoch": 0, "timestamp": "2026-10-11T21:05:52.512Z", "loss": 1.3476911783218384, "grad_norm": 2.3125671424470067, "lr": 2.9807078175982926e-07, "mem": 83.9841685295105, "tps": 12573.586317710722, "tps_per_gpu": 1571.6982897138403, "mfu": 0.0, "num_tokens_per_step": 95904, "num_label_tokens": 95904}
{"step": 360, "epoch": 0, "timestamp": "2026-10-11T21:06:00.081Z", "loss": 1.2346898317337036, "grad_norm": 2.2812021255893007, "lr": 2.980583556154409e-07, "mem": 87.21398305892944, "tps": 12153.70927299199, "tps_per_gpu": 1519.2136591239987, "mfu": 0.0, "num_tokens_per_step": 91944, "num_label_tokens": 91944}
{"step": 361, "epoch": 0, "timestamp": "2026-10-11T21:06:07.697Z", "loss": 1.3512572050094604, "grad_norm": 2.34460283243987, "lr": 2.9804588997315855e-07, "mem": 86.13410711288452, "tps": 12457.073843788348, "tps_per_gpu": 1557.1342304735435, "mfu": 0.0, "num_tokens_per_step": 94879, "num_label_tokens": 94879}
{"step": 362, "epoch": 0, "timestamp": "2026-10-11T21:06:15.144Z", "loss": 2.0829296112060547, "grad_norm": 3.122040376332116, "lr": 2.980333848380035e-07, "mem": 78.60673189163208, "tps": 11525.726808836369, "tps_per_gpu": 1440.7158511045461, "mfu": 0.0, "num_tokens_per_step": 85810, "num_label_tokens": 85810}
{"step": 363, "epoch": 0, "timestamp": "2026-10-11T21:06:22.724Z", "loss": 1.8481180667877197, "grad_norm": 2.6704414937840055, "lr": 2.980208402150128e-07, "mem": 85.05119895935059, "tps": 12316.67891109246, "tps_per_gpu": 1539.5848638865575, "mfu": 0.0, "num_tokens_per_step": 93473, "num_label_tokens": 93473}
{"step": 364, "epoch": 0, "timestamp": "2026-10-11T21:06:29.502Z", "loss": 1.951320767402649, "grad_norm": 2.8215428001000262, "lr": 2.9800825610923933e-07, "mem": 80.76215362548828, "tps": 12962.67499839797, "tps_per_gpu": 1620.3343747997462, "mfu": 0.0, "num_tokens_per_step": 87527, "num_label_tokens": 87527}
{"step": 365, "epoch": 0, "timestamp": "2026-10-11T21:06:37.000Z", "loss": 1.9535579681396484, "grad_norm": 2.7896369213176757, "lr": 2.9799563252575206e-07, "mem": 78.60673332214355, "tps": 12057.046135405251, "tps_per_gpu": 1507.1307669256564, "mfu": 0.0, "num_tokens_per_step": 90737, "num_label_tokens": 90737}
{"step": 366, "epoch": 0, "timestamp": "2026-10-11T21:06:44.526Z", "loss": 1.6489099264144897, "grad_norm": 2.6799302481919693, "lr": 2.979829694696357e-07, "mem": 86.13410949707031, "tps": 11880.907507805572, "tps_per_gpu": 1485.1134384756965, "mfu": 0.0, "num_tokens_per_step": 89390, "num_label_tokens": 89390}
{"step": 367, "epoch": 0, "timestamp": "2026-10-11T21:06:52.111Z", "loss": 1.3856613636016846, "grad_norm": 2.877251056313615, "lr": 2.9797026694599093e-07, "mem": 83.98417234420776, "tps": 11762.728375166906, "tps_per_gpu": 1470.3410468958632, "mfu": 0.0, "num_tokens_per_step": 89230, "num_label_tokens": 89230}
{"step": 368, "epoch": 0, "timestamp": "2026-10-11T21:06:59.727Z", "loss": 1.4767500162124634, "grad_norm": 2.7646966685017107, "lr": 2.979575249599344e-07, "mem": 81.84390306472778, "tps": 12118.512320363632, "tps_per_gpu": 1514.814040045454, "mfu": 0.0, "num_tokens_per_step": 92286, "num_label_tokens": 92286}
{"step": 369, "epoch": 0, "timestamp": "2026-10-11T21:07:07.300Z", "loss": 1.4775683879852295, "grad_norm": 2.536219430674569, "lr": 2.9794474351659855e-07, "mem": 87.21398735046387, "tps": 12087.213504581861, "tps_per_gpu": 1510.9016880727327, "mfu": 0.0, "num_tokens_per_step": 91552, "num_label_tokens": 91552}
{"step": 370, "epoch": 0, "timestamp": "2026-10-11T21:07:14.806Z", "loss": 1.200987458229065, "grad_norm": 2.1824140281845263, "lr": 2.9793192262113165e-07, "mem": 83.98417377471924, "tps": 12249.918452464277, "tps_per_gpu": 1531.2398065580346, "mfu": 0.0, "num_tokens_per_step": 91929, "num_label_tokens": 91929}
{"step": 371, "epoch": 0, "timestamp": "2026-10-11T21:07:21.898Z", "loss": 1.3150615692138672, "grad_norm": 2.2313566486163716, "lr": 2.9791906227869803e-07, "mem": 83.9841742515564, "tps": 12896.91041280768, "tps_per_gpu": 1612.11380160096, "mfu": 0.0, "num_tokens_per_step": 91486, "num_label_tokens": 91486}
{"step": 372, "epoch": 0, "timestamp": "2026-10-11T21:07:32.084Z", "loss": 1.3569756746292114, "grad_norm": 2.328263952961828, "lr": 2.979061624944779e-07, "mem": 83.98417472839355, "tps": 8883.253983965216, "tps_per_gpu": 1110.406747995652, "mfu": 0.0, "num_tokens_per_step": 90477, "num_label_tokens": 90477}
{"step": 373, "epoch": 0, "timestamp": "2026-10-11T21:07:39.576Z", "loss": 1.4858940839767456, "grad_norm": 2.7863115066231194, "lr": 2.9789322327366715e-07, "mem": 79.69419956207275, "tps": 11444.05331028014, "tps_per_gpu": 1430.5066637850175, "mfu": 0.0, "num_tokens_per_step": 85732, "num_label_tokens": 85732}
{"step": 374, "epoch": 0, "timestamp": "2026-10-11T21:07:47.158Z", "loss": 1.4490597248077393, "grad_norm": 2.564226069288969, "lr": 2.9788024462147787e-07, "mem": 85.05120420455933, "tps": 11887.43530483186, "tps_per_gpu": 1485.9294131039826, "mfu": 0.0, "num_tokens_per_step": 90155, "num_label_tokens": 90155}
{"step": 375, "epoch": 0, "timestamp": "2026-10-11T21:07:54.563Z", "loss": 1.3364369869232178, "grad_norm": 2.750926161927589, "lr": 2.978672265431377e-07, "mem": 78.60673809051514, "tps": 11713.652847120336, "tps_per_gpu": 1464.206605890042, "mfu": 0.0, "num_tokens_per_step": 86744, "num_label_tokens": 86744}
{"step": 376, "epoch": 0, "timestamp": "2026-10-11T21:08:02.146Z", "loss": 1.420243263244629, "grad_norm": 2.745872306966624, "lr": 2.978541690438904e-07, "mem": 86.1341142654419, "tps": 11516.687544406512, "tps_per_gpu": 1439.585943050814, "mfu": 0.0, "num_tokens_per_step": 87328, "num_label_tokens": 87328}
{"step": 377, "epoch": 0, "timestamp": "2026-10-11T21:08:13.421Z", "loss": 1.4190716743469238, "grad_norm": 2.6238443683084913, "lr": 2.978410721289955e-07, "mem": 85.0512056350708, "tps": 7686.757844162692, "tps_per_gpu": 960.8447305203365, "mfu": 0.0, "num_tokens_per_step": 86687, "num_label_tokens": 86687}
{"step": 378, "epoch": 0, "timestamp": "2026-10-11T21:08:22.911Z", "loss": 1.4235002994537354, "grad_norm": 2.560163349380873, "lr": 2.9782793580372843e-07, "mem": 87.21399164199829, "tps": 9684.168216021908, "tps_per_gpu": 1210.5210270027385, "mfu": 0.0, "num_tokens_per_step": 91867, "num_label_tokens": 91867}
{"step": 379, "epoch": 0, "timestamp": "2026-10-11T21:08:30.041Z", "loss": 1.925989031791687, "grad_norm": 2.8483219279882794, "lr": 2.978147600733806e-07, "mem": 85.05120658874512, "tps": 12873.832118066224, "tps_per_gpu": 1609.229014758278, "mfu": 0.0, "num_tokens_per_step": 91800, "num_label_tokens": 91800}
{"step": 380, "epoch": 0, "timestamp": "2026-10-11T21:08:37.571Z", "loss": 2.03161358833313, "grad_norm": 2.931513000651704, "lr": 2.97801544943259e-07, "mem": 87.21399259567261, "tps": 12059.10782943792, "tps_per_gpu": 1507.38847867974, "mfu": 0.0, "num_tokens_per_step": 90756, "num_label_tokens": 90756}
{"step": 381, "epoch": 0, "timestamp": "2026-10-11T21:08:45.181Z", "loss": 1.6671464443206787, "grad_norm": 2.6221638289361238, "lr": 2.977882904186868e-07, "mem": 83.98417901992798, "tps": 12208.691143252667, "tps_per_gpu": 1526.0863929065833, "mfu": 0.0, "num_tokens_per_step": 92977, "num_label_tokens": 92977}
{"step": 382, "epoch": 0, "timestamp": "2026-10-11T21:08:52.802Z", "loss": 1.3175885677337646, "grad_norm": 2.294813945093767, "lr": 2.97774996505003e-07, "mem": 80.76887607574463, "tps": 12028.50124962162, "tps_per_gpu": 1503.5626562027026, "mfu": 0.0, "num_tokens_per_step": 91651, "num_label_tokens": 91651}
{"step": 383, "epoch": 0, "timestamp": "2026-10-11T21:09:00.350Z", "loss": 1.2493664026260376, "grad_norm": 2.3075242297463494, "lr": 2.9776166320756223e-07, "mem": 85.05120849609375, "tps": 11956.48633178799, "tps_per_gpu": 1494.5607914734987, "mfu": 0.0, "num_tokens_per_step": 90250, "num_label_tokens": 90250}
{"step": 384, "epoch": 0, "timestamp": "2026-10-11T21:09:07.963Z", "loss": 1.6710073947906494, "grad_norm": 2.576132687965864, "lr": 2.977482905317353e-07, "mem": 86.13411808013916, "tps": 12037.573338992337, "tps_per_gpu": 1504.6966673740421, "mfu": 0.0, "num_tokens_per_step": 91643, "num_label_tokens": 91643}
{"step": 385, "epoch": 0, "timestamp": "2026-10-11T21:09:15.565Z", "loss": 1.8571844100952148, "grad_norm": 2.7271908089603425, "lr": 2.9773487848290854e-07, "mem": 87.2139949798584, "tps": 12266.459642872025, "tps_per_gpu": 1533.3074553590031, "mfu": 0.0, "num_tokens_per_step": 93242, "num_label_tokens": 93242}
{"step": 386, "epoch": 0, "timestamp": "2026-10-11T21:09:23.146Z", "loss": 1.8231152296066284, "grad_norm": 2.758694190001212, "lr": 2.977214270664844e-07, "mem": 87.21399545669556, "tps": 11844.500368024866, "tps_per_gpu": 1480.5625460031083, "mfu": 0.0, "num_tokens_per_step": 89800, "num_label_tokens": 89800}
{"step": 387, "epoch": 0, "timestamp": "2026-10-11T21:09:30.708Z", "loss": 1.9494311809539795, "grad_norm": 2.7378718523948753, "lr": 2.977079362878812e-07, "mem": 85.05121040344238, "tps": 11840.245549241727, "tps_per_gpu": 1480.0306936552158, "mfu": 0.0, "num_tokens_per_step": 89502, "num_label_tokens": 89502}
{"step": 388, "epoch": 0, "timestamp": "2026-10-11T21:09:38.360Z", "loss": 1.293081521987915, "grad_norm": 2.2940592637545603, "lr": 2.976944061525329e-07, "mem": 87.21399641036987, "tps": 12264.03873236295, "tps_per_gpu": 1533.0048415453687, "mfu": 0.0, "num_tokens_per_step": 93529, "num_label_tokens": 93529}
{"step": 389, "epoch": 0, "timestamp": "2026-10-11T21:09:45.531Z", "loss": 1.3680248260498047, "grad_norm": 2.2470450973081704, "lr": 2.976808366658895e-07, "mem": 86.13556575775146, "tps": 13089.082717583584, "tps_per_gpu": 1636.135339697948, "mfu": 0.0, "num_tokens_per_step": 94212, "num_label_tokens": 94212}
{"step": 390, "epoch": 0, "timestamp": "2026-10-11T21:09:53.170Z", "loss": 1.3207107782363892, "grad_norm": 2.2667111926845425, "lr": 2.976672278334168e-07, "mem": 80.7688798904419, "tps": 12403.161519124405, "tps_per_gpu": 1550.3951898905507, "mfu": 0.0, "num_tokens_per_step": 94776, "num_label_tokens": 94776}
{"step": 391, "epoch": 0, "timestamp": "2026-10-11T21:10:02.831Z", "loss": 1.41642165184021, "grad_norm": 2.4443060756831945, "lr": 2.976535796605963e-07, "mem": 86.13556671142578, "tps": 9125.732775405813, "tps_per_gpu": 1140.7165969257267, "mfu": 0.0, "num_tokens_per_step": 88151, "num_label_tokens": 88151}
{"step": 392, "epoch": 0, "timestamp": "2026-10-11T21:10:10.483Z", "loss": 1.4367585182189941, "grad_norm": 2.706932232708274, "lr": 2.9763989215292575e-07, "mem": 82.91949367523193, "tps": 11858.08852258739, "tps_per_gpu": 1482.2610653234237, "mfu": 0.0, "num_tokens_per_step": 90753, "num_label_tokens": 90753}
{"step": 393, "epoch": 0, "timestamp": "2026-10-11T21:10:18.112Z", "loss": 1.53089439868927, "grad_norm": 2.683396201003573, "lr": 2.9762616531591826e-07, "mem": 86.13412237167358, "tps": 11675.599464529789, "tps_per_gpu": 1459.4499330662236, "mfu": 0.0, "num_tokens_per_step": 89015, "num_label_tokens": 89015}
{"step": 394, "epoch": 0, "timestamp": "2026-10-11T21:10:25.663Z", "loss": 1.3711401224136353, "grad_norm": 2.792685727565868, "lr": 2.9761239915510295e-07, "mem": 87.21399927139282, "tps": 11862.60672050394, "tps_per_gpu": 1482.8258400629925, "mfu": 0.0, "num_tokens_per_step": 89648, "num_label_tokens": 89648}
{"step": 395, "epoch": 0, "timestamp": "2026-10-11T21:10:33.153Z", "loss": 1.4008980989456177, "grad_norm": 2.6196775586916385, "lr": 2.9759859367602504e-07, "mem": 83.9841856956482, "tps": 11601.183344945015, "tps_per_gpu": 1450.1479181181269, "mfu": 0.0, "num_tokens_per_step": 86871, "num_label_tokens": 86871}
{"step": 396, "epoch": 0, "timestamp": "2026-10-11T21:10:40.610Z", "loss": 1.4785205125808716, "grad_norm": 2.7955982717196615, "lr": 2.975847488842453e-07, "mem": 87.21400022506714, "tps": 11854.784050837332, "tps_per_gpu": 1481.8480063546665, "mfu": 0.0, "num_tokens_per_step": 88410, "num_label_tokens": 88410}
{"step": 397, "epoch": 0, "timestamp": "2026-10-11T21:10:48.139Z", "loss": 1.6008293628692627, "grad_norm": 2.530552223688548, "lr": 2.9757086478534025e-07, "mem": 76.47958755493164, "tps": 11835.658451541054, "tps_per_gpu": 1479.4573064426318, "mfu": 0.0, "num_tokens_per_step": 89097, "num_label_tokens": 89097}
{"step": 398, "epoch": 0, "timestamp": "2026-10-11T21:10:59.011Z", "loss": 1.6866464614868164, "grad_norm": 2.508108383928615, "lr": 2.9755694138490267e-07, "mem": 86.13557004928589, "tps": 8376.639039842468, "tps_per_gpu": 1047.0798799803085, "mfu": 0.0, "num_tokens_per_step": 91082, "num_label_tokens": 91082}
{"step": 399, "epoch": 0, "timestamp": "2026-10-11T21:11:08.077Z", "loss": 1.237826943397522, "grad_norm": 2.212291005979781, "lr": 2.975429786885407e-07, "mem": 86.13557052612305, "tps": 9869.8060144974, "tps_per_gpu": 1233.725751812175, "mfu": 0.0, "num_tokens_per_step": 89480, "num_label_tokens": 89480}
{"step": 400, "epoch": 0, "timestamp": "2026-10-11T21:11:55.206Z", "loss": 1.2665600776672363, "grad_norm": 2.2304111067377272, "lr": 2.975289767018786e-07, "mem": 86.12173795700073, "tps": 1909.444514265205, "tps_per_gpu": 238.68056428315063, "mfu": 0.0, "num_tokens_per_step": 89975, "num_label_tokens": 89975}
{"step": 401, "epoch": 0, "timestamp": "2026-10-11T21:12:02.709Z", "loss": 1.2236303091049194, "grad_norm": 2.2685808614487004, "lr": 2.975149354305563e-07, "mem": 87.21395492553711, "tps": 12166.602652329997, "tps_per_gpu": 1520.8253315412496, "mfu": 0.0, "num_tokens_per_step": 91287, "num_label_tokens": 91287}
{"step": 402, "epoch": 0, "timestamp": "2026-10-11T21:12:10.250Z", "loss": 1.2322922945022583, "grad_norm": 2.331391164619622, "lr": 2.975008548802297e-07, "mem": 86.13407897949219, "tps": 11902.462638967645, "tps_per_gpu": 1487.8078298709556, "mfu": 0.0, "num_tokens_per_step": 89723, "num_label_tokens": 89723}
{"step": 403, "epoch": 0, "timestamp": "2026-10-11T21:12:17.737Z", "loss": 1.3665053844451904, "grad_norm": 2.386286068946841, "lr": 2.9748673505657043e-07, "mem": 87.21395587921143, "tps": 11912.474077370562, "tps_per_gpu": 1489.0592596713202, "mfu": 0.0, "num_tokens_per_step": 89229, "num_label_tokens": 89229}
{"step": 404, "epoch": 0, "timestamp": "2026-10-11T21:12:25.324Z", "loss": 1.2542003393173218, "grad_norm": 2.320895009945604, "lr": 2.974725759652659e-07, "mem": 77.54817533493042, "tps": 12070.971265460325, "tps_per_gpu": 1508.8714081825406, "mfu": 0.0, "num_tokens_per_step": 91511, "num_label_tokens": 91511}
{"step": 405, "epoch": 0, "timestamp": "2026-10-11T21:12:32.867Z", "loss": 1.4677125215530396, "grad_norm": 2.6090096091536386, "lr": 2.9745837761201943e-07, "mem": 80.76883935928345, "tps": 12280.181397169155, "tps_per_gpu": 1535.0226746461444, "mfu": 0.0, "num_tokens_per_step": 92719, "num_label_tokens": 92719}
{"step": 406, "epoch": 0, "timestamp": "2026-10-11T21:12:40.485Z", "loss": 1.475056529045105, "grad_norm": 2.482235756634218, "lr": 2.974441400025502e-07, "mem": 83.98414325714111, "tps": 12096.651799129026, "tps_per_gpu": 1512.0814748911282, "mfu": 0.0, "num_tokens_per_step": 92196, "num_label_tokens": 92196}
{"step": 407, "epoch": 0, "timestamp": "2026-10-11T21:12:47.571Z", "loss": 1.4566781520843506, "grad_norm": 2.546799193224956, "lr": 2.9742986314259293e-07, "mem": 86.13552665710449, "tps": 12528.38962172202, "tps_per_gpu": 1566.0487027152526, "mfu": 0.0, "num_tokens_per_step": 88701, "num_label_tokens": 88701}
{"step": 408, "epoch": 0, "timestamp": "2026-10-11T21:12:55.112Z", "loss": 1.510750412940979, "grad_norm": 2.6392150096854805, "lr": 2.974155470378985e-07, "mem": 87.21395826339722, "tps": 11772.708611832193, "tps_per_gpu": 1471.5885764790241, "mfu": 0.0, "num_tokens_per_step": 88809, "num_label_tokens": 88809}
{"step": 409, "epoch": 0, "timestamp": "2026-10-11T21:13:05.217Z", "loss": 1.528587818145752, "grad_norm": 2.5876589711708995, "lr": 2.974011916942334e-07, "mem": 87.21395874023438, "tps": 8913.49766058598, "tps_per_gpu": 1114.1872075732474, "mfu": 0.0, "num_tokens_per_step": 90042, "num_label_tokens": 90042}
{"step": 410, "epoch": 0, "timestamp": "2026-10-11T21:13:12.713Z", "loss": 1.33854079246521, "grad_norm": 2.5253358752261965, "lr": 2.973867971173798e-07, "mem": 83.98414516448975, "tps": 12180.620480269232, "tps_per_gpu": 1522.577560033654, "mfu": 0.0, "num_tokens_per_step": 91311, "num_label_tokens": 91311}
{"step": 411, "epoch": 0, "timestamp": "2026-10-11T21:13:20.258Z", "loss": 1.2618403434753418, "grad_norm": 2.1678369528469856, "lr": 2.9737236331313603e-07, "mem": 86.13408327102661, "tps": 12741.272498989983, "tps_per_gpu": 1592.659062373748, "mfu": 0.0, "num_tokens_per_step": 96116, "num_label_tokens": 96116}
{"step": 412, "epoch": 0, "timestamp": "2026-10-11T21:13:27.848Z", "loss": 1.2177850008010864, "grad_norm": 2.3821899370130266, "lr": 2.9735789028731604e-07, "mem": 79.6941704750061, "tps": 12029.418758584909, "tps_per_gpu": 1503.6773448231136, "mfu": 0.0, "num_tokens_per_step": 91461, "num_label_tokens": 91461}
{"step": 413, "epoch": 0, "timestamp": "2026-10-11T21:13:35.436Z", "loss": 1.301758050918579, "grad_norm": 2.234483705089054, "lr": 2.9734337804574943e-07, "mem": 85.05117511749268, "tps": 12517.51694053705, "tps_per_gpu": 1564.6896175671313, "mfu": 0.0, "num_tokens_per_step": 94812, "num_label_tokens": 94812}
{"step": 414, "epoch": 0, "timestamp": "2026-10-11T21:13:43.055Z", "loss": 1.3480219841003418, "grad_norm": 2.3476524904179357, "lr": 2.9732882659428176e-07, "mem": 85.05117559432983, "tps": 12561.02272906138, "tps_per_gpu": 1570.1278411326725, "mfu": 0.0, "num_tokens_per_step": 95622, "num_label_tokens": 95622}
{"step": 415, "epoch": 0, "timestamp": "2026-10-11T21:13:50.687Z", "loss": 1.2513924837112427, "grad_norm": 2.272175710991066, "lr": 2.973142359387744e-07, "mem": 86.13408517837524, "tps": 11669.476523016874, "tps_per_gpu": 1458.6845653771093, "mfu": 0.0, "num_tokens_per_step": 89167, "num_label_tokens": 89167}
{"step": 416, "epoch": 0, "timestamp": "2026-10-11T21:13:57.447Z", "loss": 1.3557634353637695, "grad_norm": 2.272806570663161, "lr": 2.972996060851044e-07, "mem": 80.76213073730469, "tps": 13452.8364215449, "tps_per_gpu": 1681.6045526931125, "mfu": 0.0, "num_tokens_per_step": 90874, "num_label_tokens": 90874}
{"step": 417, "epoch": 0, "timestamp": "2026-10-11T21:14:04.998Z", "loss": 1.4191181659698486, "grad_norm": 2.6013374893152204, "lr": 2.972849370391648e-07, "mem": 83.98414850234985, "tps": 11485.267686871903, "tps_per_gpu": 1435.6584608589878, "mfu": 0.0, "num_tokens_per_step": 86916, "num_label_tokens": 86916}
{"step": 418, "epoch": 0, "timestamp": "2026-10-11T21:14:12.610Z", "loss": 1.6162220239639282, "grad_norm": 3.030288579328719, "lr": 2.972702288068641e-07, "mem": 86.13408660888672, "tps": 12019.358395127712, "tps_per_gpu": 1502.419799390964, "mfu": 0.0, "num_tokens_per_step": 91341, "num_label_tokens": 91341}
{"step": 419, "epoch": 0, "timestamp": "2026-10-11T21:14:23.898Z", "loss": 1.5568903684616089, "grad_norm": 2.805587677972337, "lr": 2.9725548139412694e-07, "mem": 81.84387969970703, "tps": 7940.793288566661, "tps_per_gpu": 992.5991610708327, "mfu": 0.0, "num_tokens_per_step": 89605, "num_label_tokens": 89605}
{"step": 420, "epoch": 0, "timestamp": "2026-10-11T21:14:33.895Z", "loss": 1.9858535528182983, "grad_norm": 2.7317863342677664, "lr": 2.9724069480689344e-07, "mem": 87.21396398544312, "tps": 8928.768585033795, "tps_per_gpu": 1116.0960731292244, "mfu": 0.0, "num_tokens_per_step": 89400, "num_label_tokens": 89400}
{"step": 421, "epoch": 0, "timestamp": "2026-10-11T21:14:41.506Z", "loss": 1.525589942932129, "grad_norm": 2.623024484378541, "lr": 2.9722586905111973e-07, "mem": 87.21396446228027, "tps": 11507.415651626183, "tps_per_gpu": 1438.4269564532729, "mfu": 0.0, "num_tokens_per_step": 87281, "num_label_tokens": 87281}
{"step": 422, "epoch": 0, "timestamp": "2026-10-11T21:14:49.056Z", "loss": 1.3844786882400513, "grad_norm": 2.685232854192563, "lr": 2.972110041327776e-07, "mem": 85.0511794090271, "tps": 11892.916953422464, "tps_per_gpu": 1486.614619177808, "mfu": 0.0, "num_tokens_per_step": 89863, "num_label_tokens": 89863}
{"step": 423, "epoch": 0, "timestamp": "2026-10-11T21:14:56.572Z", "loss": 1.5205378532409668, "grad_norm": 2.6106687629185683, "lr": 2.971961000578546e-07, "mem": 87.21396541595459, "tps": 12005.672843155653, "tps_per_gpu": 1500.7091053944566, "mfu": 0.0, "num_tokens_per_step": 90265, "num_label_tokens": 90265}
{"step": 424, "epoch": 0, "timestamp": "2026-10-11T21:15:04.182Z", "loss": 1.1465017795562744, "grad_norm": 2.221935764142073, "lr": 2.9718115683235416e-07, "mem": 85.05118036270142, "tps": 12063.326744904118, "tps_per_gpu": 1507.9158431130147, "mfu": 0.0, "num_tokens_per_step": 91939, "num_label_tokens": 91939}
{"step": 425, "epoch": 0, "timestamp": "2026-10-11T21:15:11.792Z", "loss": 1.2706509828567505, "grad_norm": 2.3008321665871994, "lr": 2.9716617446229533e-07, "mem": 87.2139663696289, "tps": 11961.585607289315, "tps_per_gpu": 1495.1982009111643, "mfu": 0.0, "num_tokens_per_step": 91035, "num_label_tokens": 91035}
{"step": 426, "epoch": 0, "timestamp": "2026-10-11T21:15:23.254Z", "loss": 1.306274175643921, "grad_norm": 2.234463454011826, "lr": 2.971511529537131e-07, "mem": 87.21396684646606, "tps": 8336.91180795524, "tps_per_gpu": 1042.113975994405, "mfu": 0.0, "num_tokens_per_step": 95563, "num_label_tokens": 95563}
{"step": 427, "epoch": 0, "timestamp": "2026-10-11T21:15:30.812Z", "loss": 1.741349220275879, "grad_norm": 2.684512716951348, "lr": 2.97136092312658e-07, "mem": 82.91946268081665, "tps": 12062.509133827396, "tps_per_gpu": 1507.8136417284245, "mfu": 0.0, "num_tokens_per_step": 91164, "num_label_tokens": 91164}
{"step": 428, "epoch": 0, "timestamp": "2026-10-11T21:15:38.295Z", "loss": 1.7293720245361328, "grad_norm": 2.6228317583624294, "lr": 2.971209925451966e-07, "mem": 81.84388399124146, "tps": 11922.954709333893, "tps_per_gpu": 1490.3693386667367, "mfu": 0.0, "num_tokens_per_step": 89252, "num_label_tokens": 89252}
{"step": 429, "epoch": 0, "timestamp": "2026-10-11T21:15:45.914Z", "loss": 1.4327301979064941, "grad_norm": 2.6342510395406413, "lr": 2.97105853657411e-07, "mem": 87.21396827697754, "tps": 12006.184537398574, "tps_per_gpu": 1500.7730671748218, "mfu": 0.0, "num_tokens_per_step": 91453, "num_label_tokens": 91453}
{"step": 430, "epoch": 0, "timestamp": "2026-10-11T21:15:52.998Z", "loss": 1.4447344541549683, "grad_norm": 2.768934657339056, "lr": 2.970906756553992e-07, "mem": 86.13553762435913, "tps": 12777.128019049043, "tps_per_gpu": 1597.1410023811304, "mfu": 0.0, "num_tokens_per_step": 90503, "num_label_tokens": 90503}
{"step": 431, "epoch": 0, "timestamp": "2026-10-11T21:16:00.620Z", "loss": 1.37814199924469, "grad_norm": 2.4548615254403967, "lr": 2.970754585452748e-07, "mem": 86.13409280776978, "tps": 12042.472643664374, "tps_per_gpu": 1505.3090804580468, "mfu": 0.0, "num_tokens_per_step": 91816, "num_label_tokens": 91816}
{"step": 432, "epoch": 0, "timestamp": "2026-10-11T21:16:08.243Z", "loss": 1.357844352722168, "grad_norm": 2.463605228690761, "lr": 2.970602023331673e-07, "mem": 86.13409328460693, "tps": 12182.357548245367, "tps_per_gpu": 1522.7946935306709, "mfu": 0.0, "num_tokens_per_step": 92854, "num_label_tokens": 92854}
{"step": 433, "epoch": 0, "timestamp": "2026-10-11T21:16:15.840Z", "loss": 1.3192615509033203, "grad_norm": 2.296855274936184, "lr": 2.970449070252219e-07, "mem": 83.98415613174438, "tps": 12086.688133645834, "tps_per_gpu": 1510.8360167057292, "mfu": 0.0, "num_tokens_per_step": 91797, "num_label_tokens": 91797}
{"step": 434, "epoch": 0, "timestamp": "2026-10-11T21:16:23.398Z", "loss": 1.2822250127792358, "grad_norm": 2.32677533905986, "lr": 2.9702957262759963e-07, "mem": 87.21397066116333, "tps": 11806.538179922038, "tps_per_gpu": 1475.8172724902547, "mfu": 0.0, "num_tokens_per_step": 89261, "num_label_tokens": 89261}
{"step": 435, "epoch": 0, "timestamp": "2026-10-11T21:16:31.002Z", "loss": 1.2211939096450806, "grad_norm": 2.2195247994177185, "lr": 2.970141991464771e-07, "mem": 86.13409471511841, "tps": 11778.107033610406, "tps_per_gpu": 1472.2633792013007, "mfu": 0.0, "num_tokens_per_step": 89340, "num_label_tokens": 89340}
{"step": 436, "epoch": 0, "timestamp": "2026-10-11T21:16:38.549Z", "loss": 1.2270621061325073, "grad_norm": 2.3214255754872792, "lr": 2.969987865880467e-07, "mem": 85.05118608474731, "tps": 12075.097282762308, "tps_per_gpu": 1509.3871603452885, "mfu": 0.0, "num_tokens_per_step": 91154, "num_label_tokens": 91154}
{"step": 437, "epoch": 0, "timestamp": "2026-10-11T21:16:46.194Z", "loss": 1.3483378887176514, "grad_norm": 2.3035894152658236, "lr": 2.969833349585167e-07, "mem": 87.2139720916748, "tps": 12405.730819184917, "tps_per_gpu": 1550.7163523981146, "mfu": 0.0, "num_tokens_per_step": 94855, "num_label_tokens": 94855}
{"step": 438, "epoch": 0, "timestamp": "2026-10-11T21:16:53.258Z", "loss": 1.952615737915039, "grad_norm": 2.8947508081214317, "lr": 2.969678442641109e-07, "mem": 83.98415851593018, "tps": 12582.165425933608, "tps_per_gpu": 1572.770678241701, "mfu": 0.0, "num_tokens_per_step": 88885, "num_label_tokens": 88885}
{"step": 439, "epoch": 0, "timestamp": "2026-10-11T21:17:00.347Z", "loss": 1.6174497604370117, "grad_norm": 2.738892812134351, "lr": 2.969523145110691e-07, "mem": 83.98415899276733, "tps": 12037.953503290462, "tps_per_gpu": 1504.7441879113078, "mfu": 0.0, "num_tokens_per_step": 85313, "num_label_tokens": 85313}
{"step": 440, "epoch": 0, "timestamp": "2026-10-11T21:17:07.919Z", "loss": 1.4971323013305664, "grad_norm": 2.8235920792443245, "lr": 2.969367457056466e-07, "mem": 87.21397352218628, "tps": 11852.376494187534, "tps_per_gpu": 1481.5470617734418, "mfu": 0.0, "num_tokens_per_step": 89775, "num_label_tokens": 89775}
{"step": 441, "epoch": 0, "timestamp": "2026-10-11T21:17:18.802Z", "loss": 1.3460837602615356, "grad_norm": 2.7243475671157653, "lr": 2.9692113785411455e-07, "mem": 85.0511884689331, "tps": 8299.954107079004, "tps_per_gpu": 1037.4942633848755, "mfu": 0.0, "num_tokens_per_step": 90296, "num_label_tokens": 90296}
{"step": 442, "epoch": 0, "timestamp": "2026-10-11T21:17:28.949Z", "loss": 1.9982342720031738, "grad_norm": 2.807295664932184, "lr": 2.969054909627597e-07, "mem": 86.13409805297852, "tps": 9031.924354634804, "tps_per_gpu": 1128.9905443293505, "mfu": 0.0, "num_tokens_per_step": 91637, "num_label_tokens": 91637}
{"step": 443, "epoch": 0, "timestamp": "2026-10-11T21:17:36.551Z", "loss": 1.9281527996063232, "grad_norm": 2.7842237225091804, "lr": 2.9688980503788475e-07, "mem": 83.98416090011597, "tps": 11762.216180384947, "tps_per_gpu": 1470.2770225481183, "mfu": 0.0, "num_tokens_per_step": 89429, "num_label_tokens": 89429}
{"step": 444, "epoch": 0, "timestamp": "2026-10-11T21:17:47.704Z", "loss": 1.5273627042770386, "grad_norm": 2.447502423688278, "lr": 2.9687408008580784e-07, "mem": 80.76885795593262, "tps": 8325.536840842544, "tps_per_gpu": 1040.692105105318, "mfu": 0.0, "num_tokens_per_step": 92903, "num_label_tokens": 92903}
{"step": 445, "epoch": 0, "timestamp": "2026-10-11T21:17:55.329Z", "loss": 1.3129706382751465, "grad_norm": 2.2255447753341997, "lr": 2.9685831611286307e-07, "mem": 87.21397590637207, "tps": 12662.68420256479, "tps_per_gpu": 1582.8355253205987, "mfu": 0.0, "num_tokens_per_step": 96535, "num_label_tokens": 96535}
{"step": 446, "epoch": 0, "timestamp": "2026-10-11T21:18:02.999Z", "loss": 1.2621653079986572, "grad_norm": 2.234807109738785, "lr": 2.9684251312540017e-07, "mem": 87.21397638320923, "tps": 12269.005567518003, "tps_per_gpu": 1533.6256959397504, "mfu": 0.0, "num_tokens_per_step": 94123, "num_label_tokens": 94123}
{"step": 447, "epoch": 0, "timestamp": "2026-10-11T21:18:10.579Z", "loss": 1.265223503112793, "grad_norm": 2.2096821434191236, "lr": 2.9682667112978465e-07, "mem": 82.91947221755981, "tps": 12713.804571832916, "tps_per_gpu": 1589.2255714791145, "mfu": 0.0, "num_tokens_per_step": 96337, "num_label_tokens": 96337}
{"step": 448, "epoch": 0, "timestamp": "2026-10-11T21:18:18.154Z", "loss": 1.173807978630066, "grad_norm": 2.191539440830257, "lr": 2.968107901323974e-07, "mem": 82.91947269439697, "tps": 11909.350976551083, "tps_per_gpu": 1488.6688720688853, "mfu": 0.0, "num_tokens_per_step": 90225, "num_label_tokens": 90225}
{"step": 449, "epoch": 0, "timestamp": "2026-10-11T21:18:25.761Z", "loss": 1.3132144212722778, "grad_norm": 2.302776992369852, "lr": 2.967948701396356e-07, "mem": 82.91947317123413, "tps": 11785.066537399323, "tps_per_gpu": 1473.1333171749154, "mfu": 0.0, "num_tokens_per_step": 89626, "num_label_tokens": 89626}
{"step": 450, "epoch": 0, "timestamp": "2026-10-11T21:18:33.280Z", "loss": 1.215025782585144, "grad_norm": 2.2769211525156146, "lr": 2.9677891115791167e-07, "mem": 82.91947364807129, "tps": 11663.83773066116, "tps_per_gpu": 1457.979716332645, "mfu": 0.0, "num_tokens_per_step": 87770, "num_label_tokens": 87770}
{"step": 451, "epoch": 0, "timestamp": "2026-10-11T21:18:40.794Z", "loss": 1.8033560514450073, "grad_norm": 2.591661578777828, "lr": 2.9676291319365386e-07, "mem": 86.13410234451294, "tps": 12053.020899152856, "tps_per_gpu": 1506.627612394107, "mfu": 0.0, "num_tokens_per_step": 90563, "num_label_tokens": 90563}
{"step": 452, "epoch": 0, "timestamp": "2026-10-11T21:18:48.401Z", "loss": 1.498091459274292, "grad_norm": 2.518733873801424, "lr": 2.967468762533062e-07, "mem": 86.1341028213501, "tps": 11703.31442835821, "tps_per_gpu": 1462.9143035447762, "mfu": 0.0, "num_tokens_per_step": 88951, "num_label_tokens": 88951}
{"step": 453, "epoch": 0, "timestamp": "2026-10-11T21:18:55.948Z", "loss": 1.4515411853790283, "grad_norm": 2.669820898117391, "lr": 2.9673080034332837e-07, "mem": 80.76886224746704, "tps": 12180.68987590391, "tps_per_gpu": 1522.5862344879888, "mfu": 0.0, "num_tokens_per_step": 91922, "num_label_tokens": 91922}
{"step": 454, "epoch": 0, "timestamp": "2026-10-11T21:19:03.553Z", "loss": 1.446867823600769, "grad_norm": 2.7178941151897926, "lr": 2.967146854701957e-07, "mem": 81.84389638900757, "tps": 12003.23725569624, "tps_per_gpu": 1500.40465696203, "mfu": 0.0, "num_tokens_per_step": 91306, "num_label_tokens": 91306}
{"step": 455, "epoch": 0, "timestamp": "2026-10-11T21:19:11.245Z", "loss": 1.4884389638900757, "grad_norm": 2.673080087641637, "lr": 2.966985316403993e-07, "mem": 87.21398067474365, "tps": 11457.6210778645, "tps_per_gpu": 1432.2026347330625, "mfu": 0.0, "num_tokens_per_step": 88117, "num_label_tokens": 88117}
{"step": 456, "epoch": 0, "timestamp": "2026-10-11T21:19:18.359Z", "loss": 1.2979320287704468, "grad_norm": 2.563901323618138, "lr": 2.9668233886044597e-07, "mem": 86.13555002212524, "tps": 12983.427572035736, "tps_per_gpu": 1622.928446504467, "mfu": 0.0, "num_tokens_per_step": 92401, "num_label_tokens": 92401}
{"step": 457, "epoch": 0, "timestamp": "2026-10-11T21:19:25.841Z", "loss": 1.5344823598861694, "grad_norm": 2.756182136162193, "lr": 2.96666107136858e-07, "mem": 83.98416757583618, "tps": 11804.228562958588, "tps_per_gpu": 1475.5285703698235, "mfu": 0.0, "num_tokens_per_step": 88325, "num_label_tokens": 88325}
{"step": 458, "epoch": 0, "timestamp": "2026-10-11T21:19:33.480Z", "loss": 2.0552282333374023, "grad_norm": 2.640895139026761, "lr": 2.9664983647617375e-07, "mem": 82.91947746276855, "tps": 12110.793817726899, "tps_per_gpu": 1513.8492272158624, "mfu": 0.0, "num_tokens_per_step": 92516, "num_label_tokens": 92516}
{"step": 459, "epoch": 0, "timestamp": "2026-10-11T21:19:43.607Z", "loss": 1.509541630744934, "grad_norm": 2.5537237035028455, "lr": 2.9663352688494684e-07, "mem": 87.21398258209229, "tps": 8789.271311286344, "tps_per_gpu": 1098.658913910793, "mfu": 0.0, "num_tokens_per_step": 88996, "num_label_tokens": 88996}
{"step": 460, "epoch": 0, "timestamp": "2026-10-11T21:19:51.193Z", "loss": 1.3855992555618286, "grad_norm": 2.2912143782491334, "lr": 2.9661717836974693e-07, "mem": 87.21398305892944, "tps": 12395.927786850409, "tps_per_gpu": 1549.490973356301, "mfu": 0.0, "num_tokens_per_step": 94039, "num_label_tokens": 94039}
{"step": 461, "epoch": 0, "timestamp": "2026-10-11T21:19:58.798Z", "loss": 1.3517998456954956, "grad_norm": 2.2731055152648416, "lr": 2.96600790937159e-07, "mem": 85.05119800567627, "tps": 12562.530127494936, "tps_per_gpu": 1570.316265936867, "mfu": 0.0, "num_tokens_per_step": 95538, "num_label_tokens": 95538}
{"step": 462, "epoch": 0, "timestamp": "2026-10-11T21:20:09.960Z", "loss": 1.4085813760757446, "grad_norm": 2.5658479084387174, "lr": 2.9658436459378415e-07, "mem": 86.13410758972168, "tps": 8226.262905442332, "tps_per_gpu": 1028.2828631802915, "mfu": 0.0, "num_tokens_per_step": 91832, "num_label_tokens": 91832}
{"step": 463, "epoch": 0, "timestamp": "2026-10-11T21:20:19.070Z", "loss": 1.4575564861297607, "grad_norm": 2.854525448158837, "lr": 2.9656789934623876e-07, "mem": 79.69419479370117, "tps": 9708.741439320398, "tps_per_gpu": 1213.5926799150498, "mfu": 0.0, "num_tokens_per_step": 88400, "num_label_tokens": 88400}
{"step": 464, "epoch": 0, "timestamp": "2026-10-11T21:20:26.724Z", "loss": 1.431767463684082, "grad_norm": 2.5371885388707525, "lr": 2.965513952011551e-07, "mem": 86.134108543396, "tps": 12224.567534823858, "tps_per_gpu": 1528.0709418529823, "mfu": 0.0, "num_tokens_per_step": 93576, "num_label_tokens": 93576}
{"step": 465, "epoch": 0, "timestamp": "2026-10-11T21:20:34.330Z", "loss": 1.384164810180664, "grad_norm": 2.611927989804744, "lr": 2.965348521651809e-07, "mem": 83.98417139053345, "tps": 11887.429504044314, "tps_per_gpu": 1485.9286880055392, "mfu": 0.0, "num_tokens_per_step": 90437, "num_label_tokens": 90437}
{"step": 466, "epoch": 0, "timestamp": "2026-10-11T21:20:41.924Z", "loss": 1.4463837146759033, "grad_norm": 2.7387157917862814, "lr": 2.9651827024497986e-07, "mem": 86.13410949707031, "tps": 11476.6224548356, "tps_per_gpu": 1434.57780685445, "mfu": 0.0, "num_tokens_per_step": 87144, "num_label_tokens": 87144}
{"step": 467, "epoch": 0, "timestamp": "2026-10-11T21:20:49.528Z", "loss": 1.3482248783111572, "grad_norm": 2.578178936429883, "lr": 2.9650164944723116e-07, "mem": 86.13410997390747, "tps": 11517.87346483739, "tps_per_gpu": 1439.7341831046738, "mfu": 0.0, "num_tokens_per_step": 87582, "num_label_tokens": 87582}
{"step": 468, "epoch": 0, "timestamp": "2026-10-11T21:20:56.631Z", "loss": 1.6894001960754395, "grad_norm": 2.711937378053168, "lr": 2.964849897786295e-07, "mem": 86.13555574417114, "tps": 12861.230163299528, "tps_per_gpu": 1607.653770412441, "mfu": 0.0, "num_tokens_per_step": 91383, "num_label_tokens": 91383}
{"step": 469, "epoch": 0, "timestamp": "2026-10-11T21:21:04.176Z", "loss": 1.6086102724075317, "grad_norm": 2.5389988604483773, "lr": 2.9646829124588557e-07, "mem": 82.9194827079773, "tps": 12281.51788018889, "tps_per_gpu": 1535.1897350236113, "mfu": 0.0, "num_tokens_per_step": 92652, "num_label_tokens": 92652}
{"step": 470, "epoch": 0, "timestamp": "2026-10-11T21:21:11.719Z", "loss": 1.3408499956130981, "grad_norm": 2.2511900234216626, "lr": 2.964515538557254e-07, "mem": 82.91948318481445, "tps": 12189.50262451806, "tps_per_gpu": 1523.6878280647575, "mfu": 0.0, "num_tokens_per_step": 91914, "num_label_tokens": 91914}
{"step": 471, "epoch": 0, "timestamp": "2026-10-11T21:21:19.345Z", "loss": 1.316150188446045, "grad_norm": 2.3143844755368956, "lr": 2.9643477761489095e-07, "mem": 86.1341118812561, "tps": 12381.746411686365, "tps_per_gpu": 1547.7183014607956, "mfu": 0.0, "num_tokens_per_step": 94447, "num_label_tokens": 94447}
{"step": 472, "epoch": 0, "timestamp": "2026-10-11T21:21:27.006Z", "loss": 1.4097119569778442, "grad_norm": 2.406338863089699, "lr": 2.964179625301396e-07, "mem": 83.98417472839355, "tps": 12431.42672694052, "tps_per_gpu": 1553.928340867565, "mfu": 0.0, "num_tokens_per_step": 95220, "num_label_tokens": 95220}
{"step": 473, "epoch": 0, "timestamp": "2026-10-11T21:21:34.621Z", "loss": 1.8020530939102173, "grad_norm": 2.7234433569717997, "lr": 2.964011086082444e-07, "mem": 86.13411283493042, "tps": 11955.275620950873, "tps_per_gpu": 1494.4094526188592, "mfu": 0.0, "num_tokens_per_step": 91030, "num_label_tokens": 91030}
{"step": 474, "epoch": 0, "timestamp": "2026-10-11T21:21:42.209Z", "loss": 1.3188209533691406, "grad_norm": 2.3003038842310124, "lr": 2.963842158559942e-07, "mem": 87.21398973464966, "tps": 12198.496492632985, "tps_per_gpu": 1524.812061579123, "mfu": 0.0, "num_tokens_per_step": 92582, "num_label_tokens": 92582}
{"step": 475, "epoch": 0, "timestamp": "2026-10-11T21:21:49.350Z", "loss": 1.2029582262039185, "grad_norm": 2.167766990708236, "lr": 2.9636728428019336e-07, "mem": 82.91948556900024, "tps": 12945.029605934287, "tps_per_gpu": 1618.128700741786, "mfu": 0.0, "num_tokens_per_step": 92429, "num_label_tokens": 92429}
{"step": 476, "epoch": 0, "timestamp": "2026-10-11T21:21:59.073Z", "loss": 1.2305223941802979, "grad_norm": 2.23410193952217, "lr": 2.96350313887662e-07, "mem": 85.05120515823364, "tps": 9564.13552992399, "tps_per_gpu": 1195.5169412404987, "mfu": 0.0, "num_tokens_per_step": 92977, "num_label_tokens": 92977}
{"step": 477, "epoch": 0, "timestamp": "2026-10-11T21:22:06.247Z", "loss": 1.3710250854492188, "grad_norm": 2.5370382865619443, "lr": 2.963333046852357e-07, "mem": 86.13556003570557, "tps": 12565.099841377374, "tps_per_gpu": 1570.6374801721718, "mfu": 0.0, "num_tokens_per_step": 90248, "num_label_tokens": 90248}
{"step": 478, "epoch": 0, "timestamp": "2026-10-11T21:22:13.815Z", "loss": 1.653844952583313, "grad_norm": 2.972428318689341, "lr": 2.9631625667976577e-07, "mem": 85.05120611190796, "tps": 12217.47491987134, "tps_per_gpu": 1527.1843649839175, "mfu": 0.0, "num_tokens_per_step": 92411, "num_label_tokens": 92411}
{"step": 479, "epoch": 0, "timestamp": "2026-10-11T21:22:21.342Z", "loss": 1.495344877243042, "grad_norm": 2.5977790086939443, "lr": 2.9629916987811924e-07, "mem": 86.13411569595337, "tps": 12117.53360054032, "tps_per_gpu": 1514.69170006754, "mfu": 0.0, "num_tokens_per_step": 91216, "num_label_tokens": 91216}
{"step": 480, "epoch": 0, "timestamp": "2026-10-11T21:22:28.936Z", "loss": 1.9659316539764404, "grad_norm": 2.838120113582679, "lr": 2.9628204428717853e-07, "mem": 87.21399259567261, "tps": 11790.934032199788, "tps_per_gpu": 1473.8667540249735, "mfu": 0.0, "num_tokens_per_step": 89518, "num_label_tokens": 89518}
{"step": 481, "epoch": 0, "timestamp": "2026-10-11T21:22:36.565Z", "loss": 1.4175002574920654, "grad_norm": 2.403374743779718, "lr": 2.9626487991384193e-07, "mem": 87.21399307250977, "tps": 12166.486476162574, "tps_per_gpu": 1520.8108095203218, "mfu": 0.0, "num_tokens_per_step": 92832, "num_label_tokens": 92832}
{"step": 482, "epoch": 0, "timestamp": "2026-10-11T21:22:46.287Z", "loss": 1.3080822229385376, "grad_norm": 2.1867587873898633, "lr": 2.962476767650232e-07, "mem": 87.21399354934692, "tps": 10030.831383938224, "tps_per_gpu": 1253.853922992278, "mfu": 0.0, "num_tokens_per_step": 97524, "num_label_tokens": 97524}
{"step": 483, "epoch": 0, "timestamp": "2026-10-11T21:22:53.456Z", "loss": 1.2583653926849365, "grad_norm": 2.216872098330987, "lr": 2.962304348476518e-07, "mem": 83.9841799736023, "tps": 12505.280713394364, "tps_per_gpu": 1563.1600891742955, "mfu": 0.0, "num_tokens_per_step": 89157, "num_label_tokens": 89157}
{"step": 484, "epoch": 0, "timestamp": "2026-10-11T21:23:03.684Z", "loss": 1.287129282951355, "grad_norm": 2.318539540004216, "lr": 2.962131541686727e-07, "mem": 83.98418045043945, "tps": 8540.832217375031, "tps_per_gpu": 1067.604027171879, "mfu": 0.0, "num_tokens_per_step": 87672, "num_label_tokens": 87672}
{"step": 485, "epoch": 0, "timestamp": "2026-10-11T21:23:11.345Z", "loss": 1.2161661386489868, "grad_norm": 2.245969521322395, "lr": 2.9619583473504666e-07, "mem": 83.98418092727661, "tps": 12110.68025974117, "tps_per_gpu": 1513.8350324676462, "mfu": 0.0, "num_tokens_per_step": 92801, "num_label_tokens": 92801}
{"step": 486, "epoch": 0, "timestamp": "2026-10-11T21:23:19.010Z", "loss": 1.3346821069717407, "grad_norm": 2.308719718092494, "lr": 2.961784765537499e-07, "mem": 83.98418140411377, "tps": 12017.612313824306, "tps_per_gpu": 1502.2015392280382, "mfu": 0.0, "num_tokens_per_step": 92289, "num_label_tokens": 92289}
{"step": 487, "epoch": 0, "timestamp": "2026-10-11T21:23:26.590Z", "loss": 1.2485628128051758, "grad_norm": 2.2394088424917684, "lr": 2.961610796317742e-07, "mem": 87.21399593353271, "tps": 12297.535161699605, "tps_per_gpu": 1537.1918952124506, "mfu": 0.0, "num_tokens_per_step": 93233, "num_label_tokens": 93233}
{"step": 488, "epoch": 0, "timestamp": "2026-10-11T21:23:34.199Z", "loss": 1.176306962966919, "grad_norm": 2.2262739564557585, "lr": 2.9614364397612716e-07, "mem": 86.1341199874878, "tps": 12502.039171086972, "tps_per_gpu": 1562.7548963858715, "mfu": 0.0, "num_tokens_per_step": 95107, "num_label_tokens": 95107}
{"step": 489, "epoch": 0, "timestamp": "2026-10-11T21:23:41.784Z", "loss": 1.2121154069900513, "grad_norm": 2.1718091987194876, "lr": 2.961261695938319e-07, "mem": 82.91949224472046, "tps": 12660.381393931508, "tps_per_gpu": 1582.5476742414385, "mfu": 0.0, "num_tokens_per_step": 96022, "num_label_tokens": 96022}
{"step": 490, "epoch": 0, "timestamp": "2026-10-11T21:23:49.402Z", "loss": 1.3009731769561768, "grad_norm": 2.321862217025215, "lr": 2.9610865649192695e-07, "mem": 87.21399736404419, "tps": 12144.33878770228, "tps_per_gpu": 1518.042348462785, "mfu": 0.0, "num_tokens_per_step": 92534, "num_label_tokens": 92534}
{"step": 491, "epoch": 0, "timestamp": "2026-10-11T21:23:57.026Z", "loss": 1.4131118059158325, "grad_norm": 2.5496607499347133, "lr": 2.960911046774666e-07, "mem": 86.13412141799927, "tps": 12440.308353417615, "tps_per_gpu": 1555.0385441772019, "mfu": 0.0, "num_tokens_per_step": 94840, "num_label_tokens": 94840}
{"step": 492, "epoch": 0, "timestamp": "2026-10-11T21:24:04.124Z", "loss": 1.522794246673584, "grad_norm": 2.701393206103448, "lr": 2.9607351415752094e-07, "mem": 82.91949367523193, "tps": 12351.34431064144, "tps_per_gpu": 1543.91803883018, "mfu": 0.0, "num_tokens_per_step": 87656, "num_label_tokens": 87656}
{"step": 493, "epoch": 0, "timestamp": "2026-10-11T21:24:11.633Z", "loss": 1.3701660633087158, "grad_norm": 2.7374420130042934, "lr": 2.9605588493917517e-07, "mem": 78.60674667358398, "tps": 11611.067848937704, "tps_per_gpu": 1451.383481117213, "mfu": 0.0, "num_tokens_per_step": 87196, "num_label_tokens": 87196}
{"step": 494, "epoch": 0, "timestamp": "2026-10-11T21:24:19.196Z", "loss": 1.52845299243927, "grad_norm": 2.653693754875931, "lr": 2.9603821702953047e-07, "mem": 85.05121374130249, "tps": 12045.281827133944, "tps_per_gpu": 1505.660228391743, "mfu": 0.0, "num_tokens_per_step": 91088, "num_label_tokens": 91088}
{"step": 495, "epoch": 0, "timestamp": "2026-10-11T21:24:28.858Z", "loss": 1.375914454460144, "grad_norm": 2.6206957049842474, "lr": 2.9602051043570334e-07, "mem": 83.9841856956482, "tps": 8954.948800640137, "tps_per_gpu": 1119.3686000800171, "mfu": 0.0, "num_tokens_per_step": 86542, "num_label_tokens": 86542}
{"step": 496, "epoch": 0, "timestamp": "2026-10-11T21:24:36.491Z", "loss": 1.5086071491241455, "grad_norm": 2.668046316942875, "lr": 2.960027651648262e-07, "mem": 85.0512146949768, "tps": 11692.03415416564, "tps_per_gpu": 1461.504269270705, "mfu": 0.0, "num_tokens_per_step": 89236, "num_label_tokens": 89236}
{"step": 497, "epoch": 0, "timestamp": "2026-10-11T21:24:43.994Z", "loss": 1.509240984916687, "grad_norm": 2.6802578470440888, "lr": 2.959849812240467e-07, "mem": 83.98418664932251, "tps": 11788.201200951105, "tps_per_gpu": 1473.5251501188882, "mfu": 0.0, "num_tokens_per_step": 88439, "num_label_tokens": 88439}
{"step": 498, "epoch": 0, "timestamp": "2026-10-11T21:24:51.107Z", "loss": 1.5158073902130127, "grad_norm": 2.84763214082489, "lr": 2.959671586205282e-07, "mem": 86.13557004928589, "tps": 12366.690606666163, "tps_per_gpu": 1545.8363258332704, "mfu": 0.0, "num_tokens_per_step": 87977, "num_label_tokens": 87977}
{"step": 499, "epoch": 0, "timestamp": "2026-10-11T21:24:58.737Z", "loss": 1.473441243171692, "grad_norm": 2.644220103990829, "lr": 2.959492973614497e-07, "mem": 86.13412523269653, "tps": 12463.128607674927, "tps_per_gpu": 1557.891075959366, "mfu": 0.0, "num_tokens_per_step": 95093, "num_label_tokens": 95093}