Przemek Chojecki | PC ✓

@prz_chojecki · # · joined 10 Apr 2016

Reasoning Data + Evals for LLMs @ https://t.co/v2xNrVT0v6, PhD in mathematics. https://t.co/BSuNwbbvwu

14 973Followers

1 148Following

20 503Posts total

276KViews on collected posts

Latest posts

Przemek Chojecki | PC ✓@prz_chojecki5 Sep

Amazing autoformalization result! Read Kevin Buzzard blog post on it here: https://t.co/hz4ug2r7AQ https://t.co/Lq8CYJHQcK 2.7K views · 22 likes · 1 reposts · 0 replies Open on X →

Przemek Chojecki | PC ✓@prz_chojecki4 Sep

Great arguments for LLM bearish case that basically comes down to RL is great for capabilities' spikes in niche domains but doesn't generalize well across. You can boost math or coding skills but that doesn't necessarily increase business skills. In a way current LLMs are each 3.7K views · 31 likes · 2 reposts · 4 replies Open on X →

Przemek Chojecki | PC ✓@prz_chojecki4 Sep

Looks like GPT-6 Astra is really next level. How many such jumps we had this year? GPT-5.2 to GPT-5.4 was big, 5.4 to 5.5 was big, 5.5 to 5.6 was big, now 5.6 to 6. We're living on the exponential. https://t.co/OZK2uBAqEr 5.4K views · 59 likes · 3 reposts · 3 replies Open on X →

Przemek Chojecki | PC ✓@prz_chojecki2 Sep

Fable 5.1 is here and it's interesting to see that the first benchmark cited is Agentic Scientific Research - the new Terminal-Bench-Science. Math, and STEM in general, makes for the best verifiable domains: easy to experiment in, less noisy data, generally useful. Building htt 2.6K views · 22 likes · 4 reposts · 5 replies Open on X →

Przemek Chojecki | PC ✓@prz_chojecki15 Jul

19 Erdős problems I claimed with AI Just to update the list for all the solution claims I've got with GPTs - 5.5 Pro solved the most, but each 5.2, 5.4, 5.6 solved something in the end. 3 New Claims with GPT-5.6 Sol Ultra in July: #421, #793, #415 (loose ends closed from https: 138.8K views · 256 likes · 31 reposts · 17 replies Open on X →

Przemek Chojecki | PC ✓@prz_chojecki1 May

12 Erdos Problems I claimed with AI. They still need to be verified properly for the final score, though most look fine. Here's the full summary of my AI+math marathon: 12 Solution Claims: #522, #603, #610, #856, #888, #896, #953, #956, #1092, #1133, #1151, #1190 4 122.9K views · 185 likes · 23 reposts · 12 replies Open on X →

Against accounts of the same size

4 posts from the last 90 days, next to the 10K–100K follower range. shown widely, but few of those viewers react.

Median views3 184this account924median for 10K–100K

Reach, %21.26%this account3.62%median for 10K–100K

Engagement, %1.11%this account1.52%median for 10K–100K

Metric	This account	Median for 10K–100K	Ratio
Median views per post	3 184	924	3.45×
Reach (views ÷ followers)	21.26%	3.62%	5.87×
Engagement rate	1.11%	1.52%	0.73×

Others in this range → Compare with another account → How these benchmarks are built →

Growth & engagement

How the posts we collected actually performed: views and reaction rate post by post, what the audience did with them, and where the follower count goes.

Views per post

122.9K1 May

138.8K15 Jul

2.6K2 Sep

5.4K4 Sep

3.7K

2.7K5 Sep

Last 6 collected posts, oldest on the left. The scale is logarithmic: one post can outrun the rest a hundred times over.

Engagement rate per post

0.18%1 May

0.22%15 Jul

1.21%2 Sep

1.22%4 Sep

1.00%

0.86%5 Sep

Reactions — likes, reposts, replies and quotes — divided by views. Median for 10K–100K accounts is 1.52%.

What the audience does

Likes71.8%575 in total

Reposts8.0%64 in total

Replies5.1%41 in total

Quotes1.4%11 in total

Bookmarks13.7%110 in total

Share of every reaction we collected for this account. Replies mean argument, reposts mean endorsement, bookmarks mean the post was worth keeping.

The follower curve appears once this account has two daily snapshots — we take one a day, and this one is on its first.