Semantle: Can you think like a computer?

8th guess botches!

16 on puzzle #19!

My best yet and feels like less of a fluke than my previous best, where after getting 15-20% on "person" I typed in a random profession which turned out to be half-connected to the answer. (Even then I spent a while going through the wrong half of the concept and guessing other related professions.)

51 today

a relative saunter

This one (#20) took me 117, but the solution actually popped into my mind much earlier and then popped out again, so I convinced myself that I had already guessed it and didn't try it for ages.

#20 took me 74 - got to green relatively quickly and then spent an absolute fuckload of time (and probably 40+ attempts) racking my brain for Perry Mason words - there are an awful lot of 'em!

fuck fuck fuck.

got it in 143. but I was...once again, in the right ballpark by like 80. I went through every courtroom related word imaginable too.

I solved Semantle #20 in 117 guesses. https://semantle.novalis.org/
Guess 23 was sentence which got me to the top 10, then I futzed around trying every related word I could think of.

This was a weird one for me because I only had three guesses in the green before I solved it, but I had more than 100 completely cold guesses.

it was the opposite for me!

I solved it in 70 but 38 were green guesses - it was pretty clear what the ballpark was, just a matter of hitting the right term

ok so i have a query and maybe someone with word2vec smarts can help me out and sorry if it's totally dumb - but in #20 the trail leading to the target word was very courtroom-based, which is fine - but in this case the word itself isn't exclusively a "court" word - like, it's a word with general meaning that i guess is used most commonly in a courtroom context, esp in media - but words like "implore" and "beg" i would consider very similar in meaning, but these are way down the list - does that mean that being 'usage adjacent' is more significant to the dataset than being 'meaning adjacent'?

Mr semantle I love it. Probably too niche to look at but doesn't work on qutebrowser. Submit button doesn't seem to do anything.

does that mean that being 'usage adjacent' is more significant to the dataset than being 'meaning adjacent'?

I'm not an expert (I've taken a couple classes on computational linguistics and played around with Python NLP libraries) but word embedding (which is the essence of what Word2vec does) is about inferring semantic relationships between words based on what words they appear next to and how often in a large corpus... the algorithm correctly inferring meaning adjacency is sort of the goal.

(Semantle #20 took me... 201 guesses)

I solved Semantle #20 in 46 guesses. https://semantle.novalis.org/

Which is worse than it sounds, given that my first guess was trial. Wall of green guesses.

92 today.

so frustrated with yesterday’s: followed a hook from “lasting” through “sustainable” up to “dispose” “burn” “calories” “incinerate” which got me in the green, the last one 700 odd and then just got stuck. didn’t turn it round from waste to the consumption part.

107. I'm gradually improving at this but still haven't done it in less than a hundred ffs! But consider me addicted to the green rush!

I was sent on a red wild goose herring and did not could.

Up to 306 so far today, I suppose I've avoided the feeling of having finished wordle, quordle, semantle and worldle and having nothing left to do.

^^all that plus spelling bee and nyt crossword lol

aside from the xword and tough days on semantle, most of these take less than ten minutes each tbf

stuck around late 900s for quite a few guesses now hmmm

I got a 996 answer and still ended up giving up, after 50 more tries. Ruined my day!

ha i was at 996 too
got it in 62 in the end :)

You found it in 553!

Had mind (296) fairly early and got stuck on that for about 250 guesses, then psychiatrist (615) for another 100, then a slow crawl up from cries (702). Weird that 'cries' and 'calls' are up in the top 1000, 'cry' and 'call' aren't but the word itself is singular.

I got a 996 answer and still ended up giving up, after 50 more tries. Ruined my day!

997 but yup :(

urgh, one average green out of 100 guesses. Gave up.

Im so good at this game. I got it in 37 today after 2 green

obviously luck has a lot to do with it

you've got to be bare forensic as well, innit! I think I need to slow down and be a more selective and thoughtful with my guesses instead of rapid firing them off like a maniac.

yeah i do that, and stupid things like guessing a word way too similar to one I already did that didn't compare well.

You have to do that though - 'cry' nope, 'crying' nope, 'cries' yeah, fuck you.

Im so good at this game. I got it in 37 today after 2 green

was watching this thread like a hawk to see if anyone would beat me, damn you

Dave, a leaderboard would be wise

(daily, weekly, lifetime)

41! My best so far.

Finally got #20, after 119 guesses! Wall of green otm, I had so many close attempts.

This game is so much more of a head-fucker than all the other word games precisely because it's so random from day to day! It kind of defies the attempt to strategize or get better at it ime.

Dave, a leaderboard would be wise

not necessary at all but I’d kinda like user stats too - wanna know what my average guess count is

had about ten guesses in the 900s having reached the top 1000 v rapidly but damned if i can get any further.

I got a 996 answer and still ended up giving up, after 50 more tries. Ruined my day!

i had to take 5 minute pauses between guesses to refresh my brain at that point.

I think you have to think less of the similarity of words and more the differing contexts they're likely to be used. That's my guess anyway. Like mist might be closer to perfume than dew

if you get too much of a fixed idea when start hitting those greens it can end up costing lots of bad guesses, you've still got keep it open but not so open that you are veering away. If that makes any sense.

ico otm

I solved Semantle #21 in 71 guesses. My first guess had a similarity of 22.35. My first word in the top 1000 was at guess #42. My penultimate guess had a similarity of 28.09 (601/1000).

Well there you have it. Definitely still feeling my way into how the algorithm groups words - trying to be both more and less literal if that makes sense. Sometimes feel like I am being trained in a new mode of language processing, which I guess I am.

Need to learn to be a bit more parsimonious with my guesses if I want to match some of the enviably low score around here though. Something quite compulsive about pouring words into the box until you get a close one tho - fast and furious guessing kind of matches the mental rummage through all the words that are somehow in my head.

agree it's more fun to just spam it with guesses, took me 134 for #21 (didn't get a green until guess #62)

56! woohoo

ooh i like that the Share feature now includes more details:

I solved Semantle #21 in 56 guesses. My first guess had a similarity of 17.90. My first word in the top 1000 was at guess #32. My penultimate guess had a similarity of 24.46. https://semantle.novalis.org/

the weird part of today is the thing that got me on the right path was, at #17, I guessed a word based on my previous guesses, that got me to almost 20%. I didn't make green until I considered a secondary definition of the word i played at #17, rather than the primary definition. which got me onto the green. and then from there, I was stuck for a bit until I considered the word that got me into green from a different perspective, and from there it was basically an open goal.

As close as 996 and feeling stuck.

I'm stuck on 994.