| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 30.0, | |
| "eval_steps": 500, | |
| "global_step": 77940, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.0, | |
| "learning_rate": 1.2830382345393893e-07, | |
| "loss": 13.6904, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.15, | |
| "learning_rate": 5.003849114703618e-05, | |
| "loss": 5.273, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 0.3, | |
| "learning_rate": 0.00010007698229407236, | |
| "loss": 0.4624, | |
| "step": 780 | |
| }, | |
| { | |
| "epoch": 0.45, | |
| "learning_rate": 0.00015011547344110854, | |
| "loss": 0.3258, | |
| "step": 1170 | |
| }, | |
| { | |
| "epoch": 0.6, | |
| "learning_rate": 0.00020015396458814472, | |
| "loss": 0.2553, | |
| "step": 1560 | |
| }, | |
| { | |
| "epoch": 0.75, | |
| "learning_rate": 0.0002501924557351809, | |
| "loss": 0.2143, | |
| "step": 1950 | |
| }, | |
| { | |
| "epoch": 0.9, | |
| "learning_rate": 0.0003002309468822171, | |
| "loss": 0.181, | |
| "step": 2340 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_loss": 0.21401263773441315, | |
| "eval_max_distance": 36, | |
| "eval_mean_distance": 4, | |
| "eval_runtime": 0.4225, | |
| "eval_samples_per_second": 118.338, | |
| "eval_steps_per_second": 4.734, | |
| "step": 2598 | |
| }, | |
| { | |
| "epoch": 1.05, | |
| "learning_rate": 0.0003502694380292533, | |
| "loss": 0.1557, | |
| "step": 2730 | |
| }, | |
| { | |
| "epoch": 1.2, | |
| "learning_rate": 0.00040030792917628943, | |
| "loss": 0.1402, | |
| "step": 3120 | |
| }, | |
| { | |
| "epoch": 1.35, | |
| "learning_rate": 0.00045034642032332564, | |
| "loss": 0.1283, | |
| "step": 3510 | |
| }, | |
| { | |
| "epoch": 1.5, | |
| "learning_rate": 0.0005003849114703618, | |
| "loss": 0.1194, | |
| "step": 3900 | |
| }, | |
| { | |
| "epoch": 1.65, | |
| "learning_rate": 0.0005504234026173979, | |
| "loss": 0.1158, | |
| "step": 4290 | |
| }, | |
| { | |
| "epoch": 1.8, | |
| "learning_rate": 0.0006004618937644341, | |
| "loss": 0.108, | |
| "step": 4680 | |
| }, | |
| { | |
| "epoch": 1.95, | |
| "learning_rate": 0.0006505003849114704, | |
| "loss": 0.1067, | |
| "step": 5070 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_loss": 0.14157189428806305, | |
| "eval_max_distance": 29, | |
| "eval_mean_distance": 2, | |
| "eval_runtime": 0.4066, | |
| "eval_samples_per_second": 122.96, | |
| "eval_steps_per_second": 4.918, | |
| "step": 5196 | |
| }, | |
| { | |
| "epoch": 2.1, | |
| "learning_rate": 0.0007005388760585066, | |
| "loss": 0.0924, | |
| "step": 5460 | |
| }, | |
| { | |
| "epoch": 2.25, | |
| "learning_rate": 0.0007505773672055427, | |
| "loss": 0.0927, | |
| "step": 5850 | |
| }, | |
| { | |
| "epoch": 2.4, | |
| "learning_rate": 0.0008006158583525789, | |
| "loss": 0.0872, | |
| "step": 6240 | |
| }, | |
| { | |
| "epoch": 2.55, | |
| "learning_rate": 0.0008506543494996151, | |
| "loss": 0.0841, | |
| "step": 6630 | |
| }, | |
| { | |
| "epoch": 2.7, | |
| "learning_rate": 0.0009006928406466513, | |
| "loss": 0.0808, | |
| "step": 7020 | |
| }, | |
| { | |
| "epoch": 2.85, | |
| "learning_rate": 0.0009507313317936874, | |
| "loss": 0.0801, | |
| "step": 7410 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_loss": 0.10980188101530075, | |
| "eval_max_distance": 22, | |
| "eval_mean_distance": 2, | |
| "eval_runtime": 0.4052, | |
| "eval_samples_per_second": 123.401, | |
| "eval_steps_per_second": 4.936, | |
| "step": 7794 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "learning_rate": 0.0009999144641176974, | |
| "loss": 0.0778, | |
| "step": 7800 | |
| }, | |
| { | |
| "epoch": 3.15, | |
| "learning_rate": 0.0009943546317680268, | |
| "loss": 0.0684, | |
| "step": 8190 | |
| }, | |
| { | |
| "epoch": 3.3, | |
| "learning_rate": 0.000988794799418356, | |
| "loss": 0.0692, | |
| "step": 8580 | |
| }, | |
| { | |
| "epoch": 3.45, | |
| "learning_rate": 0.0009832349670686853, | |
| "loss": 0.0629, | |
| "step": 8970 | |
| }, | |
| { | |
| "epoch": 3.6, | |
| "learning_rate": 0.0009776751347190145, | |
| "loss": 0.0653, | |
| "step": 9360 | |
| }, | |
| { | |
| "epoch": 3.75, | |
| "learning_rate": 0.0009721153023693439, | |
| "loss": 0.0595, | |
| "step": 9750 | |
| }, | |
| { | |
| "epoch": 3.9, | |
| "learning_rate": 0.0009665554700196733, | |
| "loss": 0.0575, | |
| "step": 10140 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_loss": 0.10807737708091736, | |
| "eval_max_distance": 18, | |
| "eval_mean_distance": 2, | |
| "eval_runtime": 0.3927, | |
| "eval_samples_per_second": 127.308, | |
| "eval_steps_per_second": 5.092, | |
| "step": 10392 | |
| }, | |
| { | |
| "epoch": 4.05, | |
| "learning_rate": 0.0009609956376700025, | |
| "loss": 0.1012, | |
| "step": 10530 | |
| }, | |
| { | |
| "epoch": 4.2, | |
| "learning_rate": 0.0009554358053203319, | |
| "loss": 0.0526, | |
| "step": 10920 | |
| }, | |
| { | |
| "epoch": 4.35, | |
| "learning_rate": 0.0009498759729706612, | |
| "loss": 0.0496, | |
| "step": 11310 | |
| }, | |
| { | |
| "epoch": 4.5, | |
| "learning_rate": 0.0009443161406209905, | |
| "loss": 0.0492, | |
| "step": 11700 | |
| }, | |
| { | |
| "epoch": 4.65, | |
| "learning_rate": 0.0009387563082713198, | |
| "loss": 0.0483, | |
| "step": 12090 | |
| }, | |
| { | |
| "epoch": 4.8, | |
| "learning_rate": 0.0009331964759216492, | |
| "loss": 0.0469, | |
| "step": 12480 | |
| }, | |
| { | |
| "epoch": 4.95, | |
| "learning_rate": 0.0009276366435719784, | |
| "loss": 0.0452, | |
| "step": 12870 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_loss": 0.08966636657714844, | |
| "eval_max_distance": 14, | |
| "eval_mean_distance": 1, | |
| "eval_runtime": 0.3879, | |
| "eval_samples_per_second": 128.906, | |
| "eval_steps_per_second": 5.156, | |
| "step": 12990 | |
| }, | |
| { | |
| "epoch": 5.1, | |
| "learning_rate": 0.0009220768112223078, | |
| "loss": 0.0392, | |
| "step": 13260 | |
| }, | |
| { | |
| "epoch": 5.25, | |
| "learning_rate": 0.0009165169788726371, | |
| "loss": 0.0397, | |
| "step": 13650 | |
| }, | |
| { | |
| "epoch": 5.4, | |
| "learning_rate": 0.0009109571465229664, | |
| "loss": 0.0393, | |
| "step": 14040 | |
| }, | |
| { | |
| "epoch": 5.55, | |
| "learning_rate": 0.0009053973141732957, | |
| "loss": 0.0399, | |
| "step": 14430 | |
| }, | |
| { | |
| "epoch": 5.7, | |
| "learning_rate": 0.0008998374818236251, | |
| "loss": 0.039, | |
| "step": 14820 | |
| }, | |
| { | |
| "epoch": 5.85, | |
| "learning_rate": 0.0008942776494739543, | |
| "loss": 0.0372, | |
| "step": 15210 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_loss": 0.07197271287441254, | |
| "eval_max_distance": 15, | |
| "eval_mean_distance": 1, | |
| "eval_runtime": 0.3764, | |
| "eval_samples_per_second": 132.832, | |
| "eval_steps_per_second": 5.313, | |
| "step": 15588 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "learning_rate": 0.0008887178171242837, | |
| "loss": 0.039, | |
| "step": 15600 | |
| }, | |
| { | |
| "epoch": 6.15, | |
| "learning_rate": 0.000883157984774613, | |
| "loss": 0.0312, | |
| "step": 15990 | |
| }, | |
| { | |
| "epoch": 6.3, | |
| "learning_rate": 0.0008775981524249422, | |
| "loss": 0.0337, | |
| "step": 16380 | |
| }, | |
| { | |
| "epoch": 6.45, | |
| "learning_rate": 0.0008720383200752716, | |
| "loss": 0.0328, | |
| "step": 16770 | |
| }, | |
| { | |
| "epoch": 6.61, | |
| "learning_rate": 0.0008664784877256009, | |
| "loss": 0.0327, | |
| "step": 17160 | |
| }, | |
| { | |
| "epoch": 6.76, | |
| "learning_rate": 0.0008609186553759302, | |
| "loss": 0.0305, | |
| "step": 17550 | |
| }, | |
| { | |
| "epoch": 6.91, | |
| "learning_rate": 0.0008553588230262595, | |
| "loss": 0.0323, | |
| "step": 17940 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_loss": 0.08398188650608063, | |
| "eval_max_distance": 12, | |
| "eval_mean_distance": 1, | |
| "eval_runtime": 0.3765, | |
| "eval_samples_per_second": 132.8, | |
| "eval_steps_per_second": 5.312, | |
| "step": 18186 | |
| }, | |
| { | |
| "epoch": 7.06, | |
| "learning_rate": 0.0008497989906765889, | |
| "loss": 0.0286, | |
| "step": 18330 | |
| }, | |
| { | |
| "epoch": 7.21, | |
| "learning_rate": 0.0008442391583269181, | |
| "loss": 0.0263, | |
| "step": 18720 | |
| }, | |
| { | |
| "epoch": 7.36, | |
| "learning_rate": 0.0008386793259772475, | |
| "loss": 0.0269, | |
| "step": 19110 | |
| }, | |
| { | |
| "epoch": 7.51, | |
| "learning_rate": 0.0008331194936275768, | |
| "loss": 0.0268, | |
| "step": 19500 | |
| }, | |
| { | |
| "epoch": 7.66, | |
| "learning_rate": 0.0008275596612779061, | |
| "loss": 0.0283, | |
| "step": 19890 | |
| }, | |
| { | |
| "epoch": 7.81, | |
| "learning_rate": 0.0008219998289282354, | |
| "loss": 0.0274, | |
| "step": 20280 | |
| }, | |
| { | |
| "epoch": 7.96, | |
| "learning_rate": 0.0008164399965785648, | |
| "loss": 0.0267, | |
| "step": 20670 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_loss": 0.07682657241821289, | |
| "eval_max_distance": 16, | |
| "eval_mean_distance": 1, | |
| "eval_runtime": 0.3687, | |
| "eval_samples_per_second": 135.62, | |
| "eval_steps_per_second": 5.425, | |
| "step": 20784 | |
| }, | |
| { | |
| "epoch": 8.11, | |
| "learning_rate": 0.000810880164228894, | |
| "loss": 0.0235, | |
| "step": 21060 | |
| }, | |
| { | |
| "epoch": 8.26, | |
| "learning_rate": 0.0008053203318792234, | |
| "loss": 0.0221, | |
| "step": 21450 | |
| }, | |
| { | |
| "epoch": 8.41, | |
| "learning_rate": 0.0007997604995295527, | |
| "loss": 0.0221, | |
| "step": 21840 | |
| }, | |
| { | |
| "epoch": 8.56, | |
| "learning_rate": 0.0007942006671798819, | |
| "loss": 0.0223, | |
| "step": 22230 | |
| }, | |
| { | |
| "epoch": 8.71, | |
| "learning_rate": 0.0007886408348302113, | |
| "loss": 0.0233, | |
| "step": 22620 | |
| }, | |
| { | |
| "epoch": 8.86, | |
| "learning_rate": 0.0007830810024805405, | |
| "loss": 0.0231, | |
| "step": 23010 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_loss": 0.06973634660243988, | |
| "eval_max_distance": 10, | |
| "eval_mean_distance": 1, | |
| "eval_runtime": 0.3759, | |
| "eval_samples_per_second": 133.026, | |
| "eval_steps_per_second": 5.321, | |
| "step": 23382 | |
| }, | |
| { | |
| "epoch": 9.01, | |
| "learning_rate": 0.0007775211701308699, | |
| "loss": 0.0227, | |
| "step": 23400 | |
| }, | |
| { | |
| "epoch": 9.16, | |
| "learning_rate": 0.0007719613377811992, | |
| "loss": 0.0185, | |
| "step": 23790 | |
| }, | |
| { | |
| "epoch": 9.31, | |
| "learning_rate": 0.0007664015054315285, | |
| "loss": 0.0183, | |
| "step": 24180 | |
| }, | |
| { | |
| "epoch": 9.46, | |
| "learning_rate": 0.0007608416730818578, | |
| "loss": 0.0191, | |
| "step": 24570 | |
| }, | |
| { | |
| "epoch": 9.61, | |
| "learning_rate": 0.0007552818407321872, | |
| "loss": 0.019, | |
| "step": 24960 | |
| }, | |
| { | |
| "epoch": 9.76, | |
| "learning_rate": 0.0007497220083825164, | |
| "loss": 0.0193, | |
| "step": 25350 | |
| }, | |
| { | |
| "epoch": 9.91, | |
| "learning_rate": 0.0007441621760328458, | |
| "loss": 0.0199, | |
| "step": 25740 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_loss": 0.07169829308986664, | |
| "eval_max_distance": 9, | |
| "eval_mean_distance": 1, | |
| "eval_runtime": 0.3704, | |
| "eval_samples_per_second": 134.993, | |
| "eval_steps_per_second": 5.4, | |
| "step": 25980 | |
| }, | |
| { | |
| "epoch": 10.06, | |
| "learning_rate": 0.0007386023436831751, | |
| "loss": 0.0184, | |
| "step": 26130 | |
| }, | |
| { | |
| "epoch": 10.21, | |
| "learning_rate": 0.0007330425113335044, | |
| "loss": 0.016, | |
| "step": 26520 | |
| }, | |
| { | |
| "epoch": 10.36, | |
| "learning_rate": 0.0007274826789838337, | |
| "loss": 0.0164, | |
| "step": 26910 | |
| }, | |
| { | |
| "epoch": 10.51, | |
| "learning_rate": 0.0007219228466341631, | |
| "loss": 0.016, | |
| "step": 27300 | |
| }, | |
| { | |
| "epoch": 10.66, | |
| "learning_rate": 0.0007163630142844923, | |
| "loss": 0.0169, | |
| "step": 27690 | |
| }, | |
| { | |
| "epoch": 10.81, | |
| "learning_rate": 0.0007108031819348217, | |
| "loss": 0.0165, | |
| "step": 28080 | |
| }, | |
| { | |
| "epoch": 10.96, | |
| "learning_rate": 0.000705243349585151, | |
| "loss": 0.0168, | |
| "step": 28470 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "eval_loss": 0.08123478293418884, | |
| "eval_max_distance": 16, | |
| "eval_mean_distance": 1, | |
| "eval_runtime": 0.3865, | |
| "eval_samples_per_second": 129.356, | |
| "eval_steps_per_second": 5.174, | |
| "step": 28578 | |
| }, | |
| { | |
| "epoch": 11.11, | |
| "learning_rate": 0.0006996835172354803, | |
| "loss": 0.015, | |
| "step": 28860 | |
| }, | |
| { | |
| "epoch": 11.26, | |
| "learning_rate": 0.0006941236848858096, | |
| "loss": 0.0137, | |
| "step": 29250 | |
| }, | |
| { | |
| "epoch": 11.41, | |
| "learning_rate": 0.0006885638525361389, | |
| "loss": 0.0151, | |
| "step": 29640 | |
| }, | |
| { | |
| "epoch": 11.56, | |
| "learning_rate": 0.0006830040201864682, | |
| "loss": 0.0144, | |
| "step": 30030 | |
| }, | |
| { | |
| "epoch": 11.71, | |
| "learning_rate": 0.0006774441878367975, | |
| "loss": 0.0147, | |
| "step": 30420 | |
| }, | |
| { | |
| "epoch": 11.86, | |
| "learning_rate": 0.0006718843554871269, | |
| "loss": 0.0148, | |
| "step": 30810 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_loss": 0.09610763192176819, | |
| "eval_max_distance": 12, | |
| "eval_mean_distance": 1, | |
| "eval_runtime": 0.3633, | |
| "eval_samples_per_second": 137.639, | |
| "eval_steps_per_second": 5.506, | |
| "step": 31176 | |
| }, | |
| { | |
| "epoch": 12.01, | |
| "learning_rate": 0.0006663245231374561, | |
| "loss": 0.0145, | |
| "step": 31200 | |
| }, | |
| { | |
| "epoch": 12.16, | |
| "learning_rate": 0.0006607646907877855, | |
| "loss": 0.0124, | |
| "step": 31590 | |
| }, | |
| { | |
| "epoch": 12.31, | |
| "learning_rate": 0.0006552048584381148, | |
| "loss": 0.0117, | |
| "step": 31980 | |
| }, | |
| { | |
| "epoch": 12.46, | |
| "learning_rate": 0.0006496450260884441, | |
| "loss": 0.0121, | |
| "step": 32370 | |
| }, | |
| { | |
| "epoch": 12.61, | |
| "learning_rate": 0.0006440851937387734, | |
| "loss": 0.0124, | |
| "step": 32760 | |
| }, | |
| { | |
| "epoch": 12.76, | |
| "learning_rate": 0.0006385253613891028, | |
| "loss": 0.0125, | |
| "step": 33150 | |
| }, | |
| { | |
| "epoch": 12.91, | |
| "learning_rate": 0.000632965529039432, | |
| "loss": 0.0128, | |
| "step": 33540 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "eval_loss": 0.08225859701633453, | |
| "eval_max_distance": 9, | |
| "eval_mean_distance": 1, | |
| "eval_runtime": 0.3712, | |
| "eval_samples_per_second": 134.695, | |
| "eval_steps_per_second": 5.388, | |
| "step": 33774 | |
| }, | |
| { | |
| "epoch": 13.06, | |
| "learning_rate": 0.0006274056966897614, | |
| "loss": 0.0116, | |
| "step": 33930 | |
| }, | |
| { | |
| "epoch": 13.21, | |
| "learning_rate": 0.0006218458643400907, | |
| "loss": 0.0106, | |
| "step": 34320 | |
| }, | |
| { | |
| "epoch": 13.36, | |
| "learning_rate": 0.00061628603199042, | |
| "loss": 0.0104, | |
| "step": 34710 | |
| }, | |
| { | |
| "epoch": 13.51, | |
| "learning_rate": 0.0006107261996407493, | |
| "loss": 0.011, | |
| "step": 35100 | |
| }, | |
| { | |
| "epoch": 13.66, | |
| "learning_rate": 0.0006051663672910787, | |
| "loss": 0.0108, | |
| "step": 35490 | |
| }, | |
| { | |
| "epoch": 13.81, | |
| "learning_rate": 0.0005996065349414079, | |
| "loss": 0.0111, | |
| "step": 35880 | |
| }, | |
| { | |
| "epoch": 13.96, | |
| "learning_rate": 0.0005940467025917372, | |
| "loss": 0.0112, | |
| "step": 36270 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "eval_loss": 0.07655028253793716, | |
| "eval_max_distance": 12, | |
| "eval_mean_distance": 1, | |
| "eval_runtime": 0.361, | |
| "eval_samples_per_second": 138.506, | |
| "eval_steps_per_second": 5.54, | |
| "step": 36372 | |
| }, | |
| { | |
| "epoch": 14.11, | |
| "learning_rate": 0.0005884868702420666, | |
| "loss": 0.0098, | |
| "step": 36660 | |
| }, | |
| { | |
| "epoch": 14.26, | |
| "learning_rate": 0.0005829270378923958, | |
| "loss": 0.009, | |
| "step": 37050 | |
| }, | |
| { | |
| "epoch": 14.41, | |
| "learning_rate": 0.0005773672055427252, | |
| "loss": 0.0093, | |
| "step": 37440 | |
| }, | |
| { | |
| "epoch": 14.56, | |
| "learning_rate": 0.0005718073731930545, | |
| "loss": 0.0095, | |
| "step": 37830 | |
| }, | |
| { | |
| "epoch": 14.71, | |
| "learning_rate": 0.0005662475408433838, | |
| "loss": 0.0093, | |
| "step": 38220 | |
| }, | |
| { | |
| "epoch": 14.86, | |
| "learning_rate": 0.0005606877084937131, | |
| "loss": 0.0093, | |
| "step": 38610 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "eval_loss": 0.07127052545547485, | |
| "eval_max_distance": 9, | |
| "eval_mean_distance": 1, | |
| "eval_runtime": 0.3671, | |
| "eval_samples_per_second": 136.219, | |
| "eval_steps_per_second": 5.449, | |
| "step": 38970 | |
| }, | |
| { | |
| "epoch": 15.01, | |
| "learning_rate": 0.0005551278761440425, | |
| "loss": 0.0101, | |
| "step": 39000 | |
| }, | |
| { | |
| "epoch": 15.16, | |
| "learning_rate": 0.0005495680437943717, | |
| "loss": 0.0078, | |
| "step": 39390 | |
| }, | |
| { | |
| "epoch": 15.31, | |
| "learning_rate": 0.0005440082114447011, | |
| "loss": 0.0079, | |
| "step": 39780 | |
| }, | |
| { | |
| "epoch": 15.46, | |
| "learning_rate": 0.0005384483790950304, | |
| "loss": 0.0081, | |
| "step": 40170 | |
| }, | |
| { | |
| "epoch": 15.61, | |
| "learning_rate": 0.0005328885467453597, | |
| "loss": 0.0085, | |
| "step": 40560 | |
| }, | |
| { | |
| "epoch": 15.76, | |
| "learning_rate": 0.000527328714395689, | |
| "loss": 0.0088, | |
| "step": 40950 | |
| }, | |
| { | |
| "epoch": 15.91, | |
| "learning_rate": 0.0005217688820460184, | |
| "loss": 0.0083, | |
| "step": 41340 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "eval_loss": 0.08469703793525696, | |
| "eval_max_distance": 14, | |
| "eval_mean_distance": 1, | |
| "eval_runtime": 0.3815, | |
| "eval_samples_per_second": 131.073, | |
| "eval_steps_per_second": 5.243, | |
| "step": 41568 | |
| }, | |
| { | |
| "epoch": 16.06, | |
| "learning_rate": 0.0005162090496963476, | |
| "loss": 0.0081, | |
| "step": 41730 | |
| }, | |
| { | |
| "epoch": 16.21, | |
| "learning_rate": 0.000510649217346677, | |
| "loss": 0.0069, | |
| "step": 42120 | |
| }, | |
| { | |
| "epoch": 16.36, | |
| "learning_rate": 0.0005050893849970063, | |
| "loss": 0.007, | |
| "step": 42510 | |
| }, | |
| { | |
| "epoch": 16.51, | |
| "learning_rate": 0.0004995295526473355, | |
| "loss": 0.0071, | |
| "step": 42900 | |
| }, | |
| { | |
| "epoch": 16.66, | |
| "learning_rate": 0.0004939697202976649, | |
| "loss": 0.0073, | |
| "step": 43290 | |
| }, | |
| { | |
| "epoch": 16.81, | |
| "learning_rate": 0.0004884098879479942, | |
| "loss": 0.0076, | |
| "step": 43680 | |
| }, | |
| { | |
| "epoch": 16.96, | |
| "learning_rate": 0.0004828500555983235, | |
| "loss": 0.0076, | |
| "step": 44070 | |
| }, | |
| { | |
| "epoch": 17.0, | |
| "eval_loss": 0.08625645935535431, | |
| "eval_max_distance": 11, | |
| "eval_mean_distance": 1, | |
| "eval_runtime": 0.3551, | |
| "eval_samples_per_second": 140.8, | |
| "eval_steps_per_second": 5.632, | |
| "step": 44166 | |
| }, | |
| { | |
| "epoch": 17.11, | |
| "learning_rate": 0.00047729022324865286, | |
| "loss": 0.0064, | |
| "step": 44460 | |
| }, | |
| { | |
| "epoch": 17.26, | |
| "learning_rate": 0.00047173039089898214, | |
| "loss": 0.0059, | |
| "step": 44850 | |
| }, | |
| { | |
| "epoch": 17.41, | |
| "learning_rate": 0.0004661705585493115, | |
| "loss": 0.0064, | |
| "step": 45240 | |
| }, | |
| { | |
| "epoch": 17.56, | |
| "learning_rate": 0.0004606107261996408, | |
| "loss": 0.0068, | |
| "step": 45630 | |
| }, | |
| { | |
| "epoch": 17.71, | |
| "learning_rate": 0.00045505089384997004, | |
| "loss": 0.0066, | |
| "step": 46020 | |
| }, | |
| { | |
| "epoch": 17.86, | |
| "learning_rate": 0.00044949106150029937, | |
| "loss": 0.0064, | |
| "step": 46410 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "eval_loss": 0.08296500891447067, | |
| "eval_max_distance": 14, | |
| "eval_mean_distance": 1, | |
| "eval_runtime": 0.3721, | |
| "eval_samples_per_second": 134.372, | |
| "eval_steps_per_second": 5.375, | |
| "step": 46764 | |
| }, | |
| { | |
| "epoch": 18.01, | |
| "learning_rate": 0.0004439312291506287, | |
| "loss": 0.0065, | |
| "step": 46800 | |
| }, | |
| { | |
| "epoch": 18.16, | |
| "learning_rate": 0.000438371396800958, | |
| "loss": 0.0055, | |
| "step": 47190 | |
| }, | |
| { | |
| "epoch": 18.31, | |
| "learning_rate": 0.0004328115644512873, | |
| "loss": 0.0052, | |
| "step": 47580 | |
| }, | |
| { | |
| "epoch": 18.46, | |
| "learning_rate": 0.00042725173210161665, | |
| "loss": 0.0057, | |
| "step": 47970 | |
| }, | |
| { | |
| "epoch": 18.61, | |
| "learning_rate": 0.00042169189975194593, | |
| "loss": 0.006, | |
| "step": 48360 | |
| }, | |
| { | |
| "epoch": 18.76, | |
| "learning_rate": 0.00041613206740227527, | |
| "loss": 0.0055, | |
| "step": 48750 | |
| }, | |
| { | |
| "epoch": 18.91, | |
| "learning_rate": 0.0004105722350526046, | |
| "loss": 0.0054, | |
| "step": 49140 | |
| }, | |
| { | |
| "epoch": 19.0, | |
| "eval_loss": 0.08839410543441772, | |
| "eval_max_distance": 11, | |
| "eval_mean_distance": 1, | |
| "eval_runtime": 0.367, | |
| "eval_samples_per_second": 136.245, | |
| "eval_steps_per_second": 5.45, | |
| "step": 49362 | |
| }, | |
| { | |
| "epoch": 19.06, | |
| "learning_rate": 0.0004050124027029339, | |
| "loss": 0.0057, | |
| "step": 49530 | |
| }, | |
| { | |
| "epoch": 19.21, | |
| "learning_rate": 0.0003994525703532632, | |
| "loss": 0.0047, | |
| "step": 49920 | |
| }, | |
| { | |
| "epoch": 19.36, | |
| "learning_rate": 0.0003938927380035925, | |
| "loss": 0.0048, | |
| "step": 50310 | |
| }, | |
| { | |
| "epoch": 19.52, | |
| "learning_rate": 0.00038833290565392183, | |
| "loss": 0.0052, | |
| "step": 50700 | |
| }, | |
| { | |
| "epoch": 19.67, | |
| "learning_rate": 0.00038277307330425117, | |
| "loss": 0.005, | |
| "step": 51090 | |
| }, | |
| { | |
| "epoch": 19.82, | |
| "learning_rate": 0.00037721324095458045, | |
| "loss": 0.0048, | |
| "step": 51480 | |
| }, | |
| { | |
| "epoch": 19.97, | |
| "learning_rate": 0.0003716534086049098, | |
| "loss": 0.0052, | |
| "step": 51870 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "eval_loss": 0.08214738219976425, | |
| "eval_max_distance": 10, | |
| "eval_mean_distance": 1, | |
| "eval_runtime": 0.3692, | |
| "eval_samples_per_second": 135.434, | |
| "eval_steps_per_second": 5.417, | |
| "step": 51960 | |
| }, | |
| { | |
| "epoch": 20.12, | |
| "learning_rate": 0.0003660935762552391, | |
| "loss": 0.0049, | |
| "step": 52260 | |
| }, | |
| { | |
| "epoch": 20.27, | |
| "learning_rate": 0.00036053374390556834, | |
| "loss": 0.0043, | |
| "step": 52650 | |
| }, | |
| { | |
| "epoch": 20.42, | |
| "learning_rate": 0.0003549739115558977, | |
| "loss": 0.0043, | |
| "step": 53040 | |
| }, | |
| { | |
| "epoch": 20.57, | |
| "learning_rate": 0.000349414079206227, | |
| "loss": 0.0044, | |
| "step": 53430 | |
| }, | |
| { | |
| "epoch": 20.72, | |
| "learning_rate": 0.0003438542468565563, | |
| "loss": 0.0044, | |
| "step": 53820 | |
| }, | |
| { | |
| "epoch": 20.87, | |
| "learning_rate": 0.0003382944145068856, | |
| "loss": 0.0045, | |
| "step": 54210 | |
| }, | |
| { | |
| "epoch": 21.0, | |
| "eval_loss": 0.0914614275097847, | |
| "eval_max_distance": 14, | |
| "eval_mean_distance": 1, | |
| "eval_runtime": 0.3653, | |
| "eval_samples_per_second": 136.874, | |
| "eval_steps_per_second": 5.475, | |
| "step": 54558 | |
| }, | |
| { | |
| "epoch": 21.02, | |
| "learning_rate": 0.00033273458215721496, | |
| "loss": 0.0041, | |
| "step": 54600 | |
| }, | |
| { | |
| "epoch": 21.17, | |
| "learning_rate": 0.00032717474980754424, | |
| "loss": 0.0035, | |
| "step": 54990 | |
| }, | |
| { | |
| "epoch": 21.32, | |
| "learning_rate": 0.0003216149174578736, | |
| "loss": 0.0038, | |
| "step": 55380 | |
| }, | |
| { | |
| "epoch": 21.47, | |
| "learning_rate": 0.0003160550851082029, | |
| "loss": 0.0038, | |
| "step": 55770 | |
| }, | |
| { | |
| "epoch": 21.62, | |
| "learning_rate": 0.0003104952527585322, | |
| "loss": 0.0041, | |
| "step": 56160 | |
| }, | |
| { | |
| "epoch": 21.77, | |
| "learning_rate": 0.0003049354204088615, | |
| "loss": 0.004, | |
| "step": 56550 | |
| }, | |
| { | |
| "epoch": 21.92, | |
| "learning_rate": 0.00029937558805919086, | |
| "loss": 0.0037, | |
| "step": 56940 | |
| }, | |
| { | |
| "epoch": 22.0, | |
| "eval_loss": 0.09314610809087753, | |
| "eval_max_distance": 14, | |
| "eval_mean_distance": 1, | |
| "eval_runtime": 0.3634, | |
| "eval_samples_per_second": 137.604, | |
| "eval_steps_per_second": 5.504, | |
| "step": 57156 | |
| }, | |
| { | |
| "epoch": 22.07, | |
| "learning_rate": 0.00029381575570952014, | |
| "loss": 0.0037, | |
| "step": 57330 | |
| }, | |
| { | |
| "epoch": 22.22, | |
| "learning_rate": 0.0002882559233598495, | |
| "loss": 0.0033, | |
| "step": 57720 | |
| }, | |
| { | |
| "epoch": 22.37, | |
| "learning_rate": 0.0002826960910101788, | |
| "loss": 0.0034, | |
| "step": 58110 | |
| }, | |
| { | |
| "epoch": 22.52, | |
| "learning_rate": 0.0002771362586605081, | |
| "loss": 0.0034, | |
| "step": 58500 | |
| }, | |
| { | |
| "epoch": 22.67, | |
| "learning_rate": 0.0002715764263108374, | |
| "loss": 0.0035, | |
| "step": 58890 | |
| }, | |
| { | |
| "epoch": 22.82, | |
| "learning_rate": 0.0002660165939611667, | |
| "loss": 0.0034, | |
| "step": 59280 | |
| }, | |
| { | |
| "epoch": 22.97, | |
| "learning_rate": 0.000260456761611496, | |
| "loss": 0.0036, | |
| "step": 59670 | |
| }, | |
| { | |
| "epoch": 23.0, | |
| "eval_loss": 0.09405915439128876, | |
| "eval_max_distance": 9, | |
| "eval_mean_distance": 1, | |
| "eval_runtime": 0.3715, | |
| "eval_samples_per_second": 134.573, | |
| "eval_steps_per_second": 5.383, | |
| "step": 59754 | |
| }, | |
| { | |
| "epoch": 23.12, | |
| "learning_rate": 0.0002548969292618253, | |
| "loss": 0.003, | |
| "step": 60060 | |
| }, | |
| { | |
| "epoch": 23.27, | |
| "learning_rate": 0.00024933709691215465, | |
| "loss": 0.0031, | |
| "step": 60450 | |
| }, | |
| { | |
| "epoch": 23.42, | |
| "learning_rate": 0.00024377726456248396, | |
| "loss": 0.003, | |
| "step": 60840 | |
| }, | |
| { | |
| "epoch": 23.57, | |
| "learning_rate": 0.00023821743221281327, | |
| "loss": 0.0029, | |
| "step": 61230 | |
| }, | |
| { | |
| "epoch": 23.72, | |
| "learning_rate": 0.00023265759986314258, | |
| "loss": 0.0028, | |
| "step": 61620 | |
| }, | |
| { | |
| "epoch": 23.87, | |
| "learning_rate": 0.0002270977675134719, | |
| "loss": 0.0028, | |
| "step": 62010 | |
| }, | |
| { | |
| "epoch": 24.0, | |
| "eval_loss": 0.08611776679754257, | |
| "eval_max_distance": 13, | |
| "eval_mean_distance": 1, | |
| "eval_runtime": 0.3594, | |
| "eval_samples_per_second": 139.139, | |
| "eval_steps_per_second": 5.566, | |
| "step": 62352 | |
| }, | |
| { | |
| "epoch": 24.02, | |
| "learning_rate": 0.00022153793516380122, | |
| "loss": 0.0027, | |
| "step": 62400 | |
| }, | |
| { | |
| "epoch": 24.17, | |
| "learning_rate": 0.00021597810281413053, | |
| "loss": 0.0026, | |
| "step": 62790 | |
| }, | |
| { | |
| "epoch": 24.32, | |
| "learning_rate": 0.00021041827046445986, | |
| "loss": 0.0027, | |
| "step": 63180 | |
| }, | |
| { | |
| "epoch": 24.47, | |
| "learning_rate": 0.00020485843811478917, | |
| "loss": 0.0027, | |
| "step": 63570 | |
| }, | |
| { | |
| "epoch": 24.62, | |
| "learning_rate": 0.00019929860576511847, | |
| "loss": 0.0027, | |
| "step": 63960 | |
| }, | |
| { | |
| "epoch": 24.77, | |
| "learning_rate": 0.00019373877341544778, | |
| "loss": 0.0024, | |
| "step": 64350 | |
| }, | |
| { | |
| "epoch": 24.92, | |
| "learning_rate": 0.0001881789410657771, | |
| "loss": 0.0026, | |
| "step": 64740 | |
| }, | |
| { | |
| "epoch": 25.0, | |
| "eval_loss": 0.09115344285964966, | |
| "eval_max_distance": 12, | |
| "eval_mean_distance": 1, | |
| "eval_runtime": 0.3622, | |
| "eval_samples_per_second": 138.045, | |
| "eval_steps_per_second": 5.522, | |
| "step": 64950 | |
| }, | |
| { | |
| "epoch": 25.07, | |
| "learning_rate": 0.0001826191087161064, | |
| "loss": 0.0026, | |
| "step": 65130 | |
| }, | |
| { | |
| "epoch": 25.22, | |
| "learning_rate": 0.00017705927636643573, | |
| "loss": 0.0023, | |
| "step": 65520 | |
| }, | |
| { | |
| "epoch": 25.37, | |
| "learning_rate": 0.00017149944401676504, | |
| "loss": 0.0023, | |
| "step": 65910 | |
| }, | |
| { | |
| "epoch": 25.52, | |
| "learning_rate": 0.00016593961166709435, | |
| "loss": 0.0021, | |
| "step": 66300 | |
| }, | |
| { | |
| "epoch": 25.67, | |
| "learning_rate": 0.00016037977931742368, | |
| "loss": 0.0021, | |
| "step": 66690 | |
| }, | |
| { | |
| "epoch": 25.82, | |
| "learning_rate": 0.000154819946967753, | |
| "loss": 0.0024, | |
| "step": 67080 | |
| }, | |
| { | |
| "epoch": 25.97, | |
| "learning_rate": 0.0001492601146180823, | |
| "loss": 0.0024, | |
| "step": 67470 | |
| }, | |
| { | |
| "epoch": 26.0, | |
| "eval_loss": 0.09158334881067276, | |
| "eval_max_distance": 9, | |
| "eval_mean_distance": 0, | |
| "eval_runtime": 0.3618, | |
| "eval_samples_per_second": 138.208, | |
| "eval_steps_per_second": 5.528, | |
| "step": 67548 | |
| }, | |
| { | |
| "epoch": 26.12, | |
| "learning_rate": 0.0001437002822684116, | |
| "loss": 0.0021, | |
| "step": 67860 | |
| }, | |
| { | |
| "epoch": 26.27, | |
| "learning_rate": 0.0001381404499187409, | |
| "loss": 0.0019, | |
| "step": 68250 | |
| }, | |
| { | |
| "epoch": 26.42, | |
| "learning_rate": 0.00013258061756907022, | |
| "loss": 0.002, | |
| "step": 68640 | |
| }, | |
| { | |
| "epoch": 26.57, | |
| "learning_rate": 0.00012702078521939955, | |
| "loss": 0.0019, | |
| "step": 69030 | |
| }, | |
| { | |
| "epoch": 26.72, | |
| "learning_rate": 0.00012146095286972886, | |
| "loss": 0.0021, | |
| "step": 69420 | |
| }, | |
| { | |
| "epoch": 26.87, | |
| "learning_rate": 0.00011590112052005817, | |
| "loss": 0.002, | |
| "step": 69810 | |
| }, | |
| { | |
| "epoch": 27.0, | |
| "eval_loss": 0.08878373354673386, | |
| "eval_max_distance": 9, | |
| "eval_mean_distance": 0, | |
| "eval_runtime": 0.3454, | |
| "eval_samples_per_second": 144.754, | |
| "eval_steps_per_second": 5.79, | |
| "step": 70146 | |
| }, | |
| { | |
| "epoch": 27.02, | |
| "learning_rate": 0.00011034128817038747, | |
| "loss": 0.0021, | |
| "step": 70200 | |
| }, | |
| { | |
| "epoch": 27.17, | |
| "learning_rate": 0.0001047814558207168, | |
| "loss": 0.0017, | |
| "step": 70590 | |
| }, | |
| { | |
| "epoch": 27.32, | |
| "learning_rate": 9.92216234710461e-05, | |
| "loss": 0.0018, | |
| "step": 70980 | |
| }, | |
| { | |
| "epoch": 27.47, | |
| "learning_rate": 9.366179112137542e-05, | |
| "loss": 0.0017, | |
| "step": 71370 | |
| }, | |
| { | |
| "epoch": 27.62, | |
| "learning_rate": 8.810195877170473e-05, | |
| "loss": 0.0016, | |
| "step": 71760 | |
| }, | |
| { | |
| "epoch": 27.77, | |
| "learning_rate": 8.254212642203404e-05, | |
| "loss": 0.002, | |
| "step": 72150 | |
| }, | |
| { | |
| "epoch": 27.92, | |
| "learning_rate": 7.698229407236336e-05, | |
| "loss": 0.0017, | |
| "step": 72540 | |
| }, | |
| { | |
| "epoch": 28.0, | |
| "eval_loss": 0.08879587054252625, | |
| "eval_max_distance": 9, | |
| "eval_mean_distance": 0, | |
| "eval_runtime": 0.3476, | |
| "eval_samples_per_second": 143.846, | |
| "eval_steps_per_second": 5.754, | |
| "step": 72744 | |
| }, | |
| { | |
| "epoch": 28.07, | |
| "learning_rate": 7.142246172269268e-05, | |
| "loss": 0.0016, | |
| "step": 72930 | |
| }, | |
| { | |
| "epoch": 28.22, | |
| "learning_rate": 6.586262937302199e-05, | |
| "loss": 0.0016, | |
| "step": 73320 | |
| }, | |
| { | |
| "epoch": 28.37, | |
| "learning_rate": 6.03027970233513e-05, | |
| "loss": 0.0016, | |
| "step": 73710 | |
| }, | |
| { | |
| "epoch": 28.52, | |
| "learning_rate": 5.474296467368061e-05, | |
| "loss": 0.0016, | |
| "step": 74100 | |
| }, | |
| { | |
| "epoch": 28.67, | |
| "learning_rate": 4.9183132324009924e-05, | |
| "loss": 0.0016, | |
| "step": 74490 | |
| }, | |
| { | |
| "epoch": 28.82, | |
| "learning_rate": 4.362329997433924e-05, | |
| "loss": 0.0017, | |
| "step": 74880 | |
| }, | |
| { | |
| "epoch": 28.97, | |
| "learning_rate": 3.806346762466855e-05, | |
| "loss": 0.0017, | |
| "step": 75270 | |
| }, | |
| { | |
| "epoch": 29.0, | |
| "eval_loss": 0.09515639394521713, | |
| "eval_max_distance": 9, | |
| "eval_mean_distance": 0, | |
| "eval_runtime": 0.343, | |
| "eval_samples_per_second": 145.752, | |
| "eval_steps_per_second": 5.83, | |
| "step": 75342 | |
| }, | |
| { | |
| "epoch": 29.12, | |
| "learning_rate": 3.250363527499786e-05, | |
| "loss": 0.0015, | |
| "step": 75660 | |
| }, | |
| { | |
| "epoch": 29.27, | |
| "learning_rate": 2.6943802925327177e-05, | |
| "loss": 0.0016, | |
| "step": 76050 | |
| }, | |
| { | |
| "epoch": 29.42, | |
| "learning_rate": 2.1383970575656488e-05, | |
| "loss": 0.0014, | |
| "step": 76440 | |
| }, | |
| { | |
| "epoch": 29.57, | |
| "learning_rate": 1.5824138225985802e-05, | |
| "loss": 0.0013, | |
| "step": 76830 | |
| }, | |
| { | |
| "epoch": 29.72, | |
| "learning_rate": 1.0264305876315115e-05, | |
| "loss": 0.0014, | |
| "step": 77220 | |
| }, | |
| { | |
| "epoch": 29.87, | |
| "learning_rate": 4.704473526644427e-06, | |
| "loss": 0.0014, | |
| "step": 77610 | |
| }, | |
| { | |
| "epoch": 30.0, | |
| "eval_loss": 0.09847646951675415, | |
| "eval_max_distance": 9, | |
| "eval_mean_distance": 0, | |
| "eval_runtime": 0.3435, | |
| "eval_samples_per_second": 145.564, | |
| "eval_steps_per_second": 5.823, | |
| "step": 77940 | |
| }, | |
| { | |
| "epoch": 30.0, | |
| "step": 77940, | |
| "total_flos": 4.517674593940685e+16, | |
| "train_loss": 0.053724035134690526, | |
| "train_runtime": 6582.4117, | |
| "train_samples_per_second": 355.137, | |
| "train_steps_per_second": 11.841 | |
| } | |
| ], | |
| "logging_steps": 390, | |
| "max_steps": 77940, | |
| "num_train_epochs": 30, | |
| "save_steps": 780, | |
| "total_flos": 4.517674593940685e+16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |