| {"step": 1, "loss": 0.8693870902061462, "gradient_norm": 19.529491424560547, "seconds": 4.118223747005686, "decisions": 4, "branches": 14, "actual_branch_tokens": 1438, "padded_branch_tokens": 1448, "peak_cuda_allocated_bytes": 7907518976, "lr_factor": 0.01} |
| {"step": 2, "loss": 0.7699789851903915, "gradient_norm": 26.21019744873047, "seconds": 3.5661668149987236, "decisions": 4, "branches": 12, "actual_branch_tokens": 1356, "padded_branch_tokens": 1361, "peak_cuda_allocated_bytes": 8052873728, "lr_factor": 0.02} |
| {"step": 3, "loss": 1.0451247729361057, "gradient_norm": 19.00668716430664, "seconds": 3.9829175180057064, "decisions": 4, "branches": 14, "actual_branch_tokens": 1440, "padded_branch_tokens": 1454, "peak_cuda_allocated_bytes": 8052873728, "lr_factor": 0.03} |
| {"step": 4, "loss": 0.5272262170910835, "gradient_norm": 16.054319381713867, "seconds": 3.488157991989283, "decisions": 4, "branches": 12, "actual_branch_tokens": 1282, "padded_branch_tokens": 1288, "peak_cuda_allocated_bytes": 8052873728, "lr_factor": 0.04} |
| {"step": 5, "loss": 0.7045565620064735, "gradient_norm": 14.661520957946777, "seconds": 3.9977204140013782, "decisions": 4, "branches": 12, "actual_branch_tokens": 1501, "padded_branch_tokens": 1510, "peak_cuda_allocated_bytes": 8089661440, "lr_factor": 0.05} |
| {"step": 6, "loss": 0.324239959474653, "gradient_norm": 18.266826629638672, "seconds": 3.3406440389953787, "decisions": 4, "branches": 11, "actual_branch_tokens": 1397, "padded_branch_tokens": 1405, "peak_cuda_allocated_bytes": 8138392064, "lr_factor": 0.06} |
| {"step": 7, "loss": 0.5339574355384684, "gradient_norm": 13.051875114440918, "seconds": 2.990777747007087, "decisions": 4, "branches": 10, "actual_branch_tokens": 1290, "padded_branch_tokens": 1294, "peak_cuda_allocated_bytes": 8138392064, "lr_factor": 0.07} |
| {"step": 8, "loss": 1.2797077149152756, "gradient_norm": 42.306705474853516, "seconds": 3.9690867390017956, "decisions": 4, "branches": 13, "actual_branch_tokens": 1565, "padded_branch_tokens": 1574, "peak_cuda_allocated_bytes": 8138392064, "lr_factor": 0.08} |
| {"step": 9, "loss": 0.7341323457658291, "gradient_norm": 23.285865783691406, "seconds": 3.866028983000433, "decisions": 4, "branches": 12, "actual_branch_tokens": 1390, "padded_branch_tokens": 1400, "peak_cuda_allocated_bytes": 8138392064, "lr_factor": 0.09} |
| {"step": 10, "loss": 0.4565264442935586, "gradient_norm": 14.246260643005371, "seconds": 3.982957640007953, "decisions": 4, "branches": 15, "actual_branch_tokens": 1441, "padded_branch_tokens": 1449, "peak_cuda_allocated_bytes": 8138392064, "lr_factor": 0.1} |
| {"step": 11, "loss": 0.6989412307739258, "gradient_norm": 21.852399826049805, "seconds": 3.7737965429987526, "decisions": 4, "branches": 11, "actual_branch_tokens": 1557, "padded_branch_tokens": 1560, "peak_cuda_allocated_bytes": 8138392064, "lr_factor": 0.11} |
| {"step": 12, "loss": 0.3664970062673092, "gradient_norm": 18.80347442626953, "seconds": 3.982025391989737, "decisions": 4, "branches": 13, "actual_branch_tokens": 1492, "padded_branch_tokens": 1501, "peak_cuda_allocated_bytes": 8138392064, "lr_factor": 0.12} |
| {"step": 13, "loss": 0.24145922577008605, "gradient_norm": 14.713008880615234, "seconds": 3.6566432179970434, "decisions": 4, "branches": 11, "actual_branch_tokens": 1595, "padded_branch_tokens": 1602, "peak_cuda_allocated_bytes": 8240957952, "lr_factor": 0.13} |
| {"step": 14, "loss": 0.9659595936536789, "gradient_norm": 37.59418487548828, "seconds": 3.7694366829964565, "decisions": 4, "branches": 11, "actual_branch_tokens": 1549, "padded_branch_tokens": 1554, "peak_cuda_allocated_bytes": 8240957952, "lr_factor": 0.14} |
| {"step": 15, "loss": 0.14861621015006676, "gradient_norm": 12.707115173339844, "seconds": 4.004563973008771, "decisions": 4, "branches": 14, "actual_branch_tokens": 1467, "padded_branch_tokens": 1476, "peak_cuda_allocated_bytes": 8240957952, "lr_factor": 0.15} |
| {"step": 16, "loss": 2.1753957606852055, "gradient_norm": 60.869468688964844, "seconds": 3.901415497995913, "decisions": 4, "branches": 13, "actual_branch_tokens": 1371, "padded_branch_tokens": 1376, "peak_cuda_allocated_bytes": 8240957952, "lr_factor": 0.16} |
| {"step": 17, "loss": 0.18685088299389463, "gradient_norm": 12.952049255371094, "seconds": 3.8944962579989806, "decisions": 4, "branches": 14, "actual_branch_tokens": 1436, "padded_branch_tokens": 1445, "peak_cuda_allocated_bytes": 8240957952, "lr_factor": 0.17} |
| {"step": 18, "loss": 0.5719003627746133, "gradient_norm": 23.8725528717041, "seconds": 4.014727521003806, "decisions": 4, "branches": 12, "actual_branch_tokens": 1579, "padded_branch_tokens": 1588, "peak_cuda_allocated_bytes": 8240957952, "lr_factor": 0.18} |
| {"step": 19, "loss": 0.3988803016509337, "gradient_norm": 29.305479049682617, "seconds": 4.048821812000824, "decisions": 4, "branches": 13, "actual_branch_tokens": 1688, "padded_branch_tokens": 1695, "peak_cuda_allocated_bytes": 8240957952, "lr_factor": 0.19} |
| {"step": 20, "loss": 1.372778618708253, "gradient_norm": 67.43666076660156, "seconds": 3.4975894630042603, "decisions": 4, "branches": 14, "actual_branch_tokens": 1463, "padded_branch_tokens": 1468, "peak_cuda_allocated_bytes": 8240957952, "lr_factor": 0.2} |
| {"step": 21, "loss": 0.14342794067761133, "gradient_norm": 11.37416934967041, "seconds": 3.5588715040066745, "decisions": 4, "branches": 11, "actual_branch_tokens": 1511, "padded_branch_tokens": 1513, "peak_cuda_allocated_bytes": 8240957952, "lr_factor": 0.21} |
| {"step": 22, "loss": 0.46724059636471793, "gradient_norm": 28.669349670410156, "seconds": 3.880720214001485, "decisions": 4, "branches": 12, "actual_branch_tokens": 1396, "padded_branch_tokens": 1405, "peak_cuda_allocated_bytes": 8240957952, "lr_factor": 0.22} |
| {"step": 23, "loss": 0.42527354510093573, "gradient_norm": 22.349712371826172, "seconds": 3.896784013006254, "decisions": 4, "branches": 13, "actual_branch_tokens": 1379, "padded_branch_tokens": 1384, "peak_cuda_allocated_bytes": 8240957952, "lr_factor": 0.23} |
| {"step": 24, "loss": 0.24731684524158482, "gradient_norm": 23.357633590698242, "seconds": 3.7566044319974026, "decisions": 4, "branches": 12, "actual_branch_tokens": 1508, "padded_branch_tokens": 1513, "peak_cuda_allocated_bytes": 8240957952, "lr_factor": 0.24} |
| {"step": 25, "loss": 3.1549908951856196, "gradient_norm": 92.24949645996094, "seconds": 3.595210778003093, "decisions": 4, "branches": 12, "actual_branch_tokens": 1409, "padded_branch_tokens": 1414, "peak_cuda_allocated_bytes": 8240957952, "lr_factor": 0.25} |
| {"step": 26, "loss": 0.9902321046101861, "gradient_norm": 94.4915542602539, "seconds": 4.07910237100441, "decisions": 4, "branches": 12, "actual_branch_tokens": 1887, "padded_branch_tokens": 1894, "peak_cuda_allocated_bytes": 8317326848, "lr_factor": 0.26} |
| {"step": 27, "loss": 0.02165430059631035, "gradient_norm": 2.885589599609375, "seconds": 3.5664120940055, "decisions": 4, "branches": 13, "actual_branch_tokens": 1379, "padded_branch_tokens": 1387, "peak_cuda_allocated_bytes": 8317326848, "lr_factor": 0.27} |
| {"step": 28, "loss": 0.2891657831341945, "gradient_norm": 33.69540023803711, "seconds": 4.05612512701191, "decisions": 4, "branches": 14, "actual_branch_tokens": 1520, "padded_branch_tokens": 1527, "peak_cuda_allocated_bytes": 8317326848, "lr_factor": 0.28} |
| {"step": 29, "loss": 1.6357667127158475, "gradient_norm": 100.5425796508789, "seconds": 4.034884143009549, "decisions": 4, "branches": 14, "actual_branch_tokens": 1500, "padded_branch_tokens": 1508, "peak_cuda_allocated_bytes": 8317326848, "lr_factor": 0.29} |
| {"step": 30, "loss": 0.16285334138956387, "gradient_norm": 26.56850814819336, "seconds": 3.5307407610089285, "decisions": 4, "branches": 11, "actual_branch_tokens": 1507, "padded_branch_tokens": 1510, "peak_cuda_allocated_bytes": 8317326848, "lr_factor": 0.3} |
| {"step": 31, "loss": 0.0036650802917677083, "gradient_norm": 1.7888540029525757, "seconds": 3.2161976280040108, "decisions": 4, "branches": 12, "actual_branch_tokens": 1327, "padded_branch_tokens": 1330, "peak_cuda_allocated_bytes": 8317326848, "lr_factor": 0.31} |
| {"step": 32, "loss": 0.010653872763214167, "gradient_norm": 4.28465461730957, "seconds": 3.6718065640015993, "decisions": 4, "branches": 14, "actual_branch_tokens": 1410, "padded_branch_tokens": 1416, "peak_cuda_allocated_bytes": 8317326848, "lr_factor": 0.32} |
| {"step": 33, "loss": 3.250807650095851, "gradient_norm": 115.80107879638672, "seconds": 3.8716218779882183, "decisions": 4, "branches": 13, "actual_branch_tokens": 1438, "padded_branch_tokens": 1445, "peak_cuda_allocated_bytes": 8317326848, "lr_factor": 0.33} |
| {"step": 34, "loss": 0.02215615002933191, "gradient_norm": 4.8425798416137695, "seconds": 3.7865977000037674, "decisions": 4, "branches": 10, "actual_branch_tokens": 1737, "padded_branch_tokens": 1743, "peak_cuda_allocated_bytes": 8317326848, "lr_factor": 0.34} |
| {"step": 35, "loss": 3.9466441228287295, "gradient_norm": 134.32041931152344, "seconds": 3.98363202100154, "decisions": 4, "branches": 12, "actual_branch_tokens": 1468, "padded_branch_tokens": 1475, "peak_cuda_allocated_bytes": 8317326848, "lr_factor": 0.35} |
| {"step": 36, "loss": 0.02528251288458705, "gradient_norm": 3.698042392730713, "seconds": 3.9176760880072834, "decisions": 4, "branches": 13, "actual_branch_tokens": 1426, "padded_branch_tokens": 1433, "peak_cuda_allocated_bytes": 8317326848, "lr_factor": 0.36} |
| {"step": 37, "loss": 0.9050782639533281, "gradient_norm": 67.7143325805664, "seconds": 3.6715229240071494, "decisions": 4, "branches": 12, "actual_branch_tokens": 1313, "padded_branch_tokens": 1318, "peak_cuda_allocated_bytes": 8317326848, "lr_factor": 0.37} |
| {"step": 38, "loss": 0.9363077478187716, "gradient_norm": 33.960853576660156, "seconds": 3.9945897229918046, "decisions": 4, "branches": 14, "actual_branch_tokens": 1468, "padded_branch_tokens": 1480, "peak_cuda_allocated_bytes": 8317326848, "lr_factor": 0.38} |
| {"step": 39, "loss": 0.8041712334379554, "gradient_norm": 76.34855651855469, "seconds": 3.5535923540010117, "decisions": 4, "branches": 13, "actual_branch_tokens": 1365, "padded_branch_tokens": 1371, "peak_cuda_allocated_bytes": 8317326848, "lr_factor": 0.39} |
| {"step": 40, "loss": 0.5729095703864004, "gradient_norm": 15.978842735290527, "seconds": 3.6657917910051765, "decisions": 4, "branches": 12, "actual_branch_tokens": 1539, "padded_branch_tokens": 1545, "peak_cuda_allocated_bytes": 8317326848, "lr_factor": 0.4} |
|
|
