00:13:380 (13380|2) - 00:14:793 (14793|0) - 00:16:233 (16233|0) - 00:17:149 (17149|2) - Little nitpicky but shouldn't these reach up to the next LN? Ex. 00:13:380 (13380|2) extending by 1/12
01:36:730 - same
00:23:380 (23380|3,23458|2) - Here, I recommend changing the direction because the melody changes from (00:23:537 -). https://osu.ppy.sh/ss/19299082/c701
00:25:736 (25736|0,25814|2) - Here, the song structure is similar to 00:23:222 - so you can use double.
Also, I recommend expressing the cymbal sound of 00:26:364 - as triple. https://osu.ppy.sh/ss/19299084/7078
00:25:736 (25736|1,25736|0,25814|0,25814|1) - Same as top diff i think you can make it weaker, just like 01:49:924 - is better
00:36:259 (36259|1,36259|0,36338|2) - Both of these drums are super faint so I personally wouldn't include them, but if you still want to I would reduce 00:36:259 (36259|1,36259|0) to a single
00:37:202 (37202|2,37202|1) - This double placement seems a little out of place to me, the accented sounds here are actually on 00:37:123 - 00:37:280. I probably wouldn't advise putting doubles for both so maybe you could represent them with a slightly different LN pattern?
00:39:008 (39008|2,39047|1) - This burst doesn't start until 00:39:086 so I would either remove 00:39:047 (39047|1) or both notes
00:44:034 (44034|1,44034|2) - same idea here but with a chord
00:47:961 (47961|2,47961|3,48275|2,48275|3) - 00:50:474 (50474|2,50474|3) - How come these are doubles? If anything I would expect them to be graced to continue the flow of the section
00:50:356 (50356|1) - I would personally suggest making this into a normal note, there's a really distinct cutoff at 00:50:395 so having the LN bleed into it feels a bit weird to me
00:50:709 (50709|0,50736|1,50762|2,50788|3) - I would suggest following the snapping on top diff for this, although you can keep the extra note at 00:50:709 (50709|0) if you like it
Similar point that 00:51:024 (51024|2,51102|0,51102|1,51102|3,51259|2,51259|3,51298|0,51298|1) should be snapped on 1/3, same as top
00:55:107 (55107|3,55107|2,55186|0,55186|1,55264|2,55264|3) - Considering the visual aspect, I think it's better to switch with ctrl+h. https://osu.ppy.sh/ss/19299090/a819
00:56:521 (56521|3,56678|3,56835|3,56992|3,57149|3,57306|3,57463|3) - Not sure if this is intended as it could definitely be balanced a bit more by moving some notes around, mainly 00:56:599 (56599|1,56678|3,56757|2) - , and it feels a little bit too much for this diff here
00:59:034 (59034|2,59034|3,59113|0,59191|1,59270|2,59348|1,59348|0,59427|3,59505|2,59584|1,59662|2,59662|3,59741|0,59819|1,59898|2,59976|1,59976|0,60055|3,60133|2,60212|1) - The pattern here is underwhelmingly easy personally consider the anchor and LN usage previously, don't really see anything wrong by making snares triple so at least i think it can get some buff from there
01:09:086 - Same
01:05:631 - Shouldn't this only be a single? Compared to 01:05:474 (65474|2)
Similarly you have a missing note at 01:05:866 and another double at 01:05:945 so I would suggest bringing 01:05:945 (65945|0) down 1/4
same thing at 01:13:170, should be single and there's a missing note at 01:13:720
01:16:311 (76311|2,76351|3,76390|0,76429|1,76469|2) - is it intended that you have 1/6 trill in top diff but it's 1/8 here, same at 02:34:217 -
If it does end up being changed here some your lower diffs might need changes accordingly
01:22:987 (82987|1,83065|0,83144|1,83262|1,83301|0,83380|1,83458|0) - This could be hand balanced a bit better, it's a bit of a jacky OH trill right now so even just flipping the columns for 01:23:262 (83262|1,83301|0,83380|1,83458|0) would be helpful
02:40:893 - same
02:01:547 (121547|0,121547|3,121626|2) - 02:02:804 (122804|1,122804|0,122882|3) - 02:04:060 (124060|3,124060|1,124139|0) - 02:06:573 (126573|3,126573|0,126652|1) etc. - Shouldn't these all be consistent in whether or not they're LNs?
02:01:469 (121469|2,121547|0,121626|2,124060|3,124060|1,124139|0) - and the other instance is for piano
02:05:317 - 02:05:631 - Same as modding in top dif. I think it's better to write it a little shorter since there's no need to place LN too close to here. https://osu.ppy.sh/ss/19299099/52c4
02:05:317 (125317|2,125317|3,125395|1,125395|0,125474|2,125474|3,125552|1) - 02:15:369 - i think if you want this diff to be a bit smoother this could probably be changed into rolly jumpstream given that previous stream pattern is mostly rolly, kinda subjective tho
02:12:542 to 02:15:212 - There's a really big dip in LN usage here that feels a bit out of place to me, a little bit undermapped overall
For example you could map in synths at 02:12:620 - 02:12:856 - 02:13:092, add some lengthier LNs for the fade at 02:13:170 and other changes to avoid switching from full LN -> full rice
02:14:427 (134427|0,134427|1,134741|3,134741|2) - These play a bit weird to me since the layering is basically the same as the big melody + kick hits 02:14:270 (134270|3,134309|2,134584|1,134584|0) so maybe you could adjust the emphasis here a bit?
For example you could switch the rice to singles and LNs to graced triples like this:
02:17:804 (137804|3) - Wonder if this is intended since top diff didn't map this 1/4 note, guess either way could make sense but maybe it would be better if both diff can align with how it is mapped at 02:16:547 (136547|3) - maybe
Kinda similar question for 02:19:374 (139374|0) -
02:31:547 - Somehow feels weird that nothing gets changed here despite drums' rhythm changes significantly, probably just add some changes for 1/4 like 02:26:521 (146521|1,146521|0,146599|3,146599|2) - would be enough, although it needs to have a bit more of that here
There is a small change here, which is the addition of the chord at 02:31:861 (151861|3) -
it's not major but I do think it makes a noticeable difference, since there's now a trill at 02:31:704 (151704|3,151783|2,151861|3), and an anchor with 02:31:547 (151547|3,151704|3,151861|3,152018|3)
I'd prefer to keep this as is but you can reopen if it really needs a buff