Kyle Wiggers ✓

@Kyle_L_Wiggers · Manhattan, NY · joined 06 May 2013

Ai2 Comms Lead | kylew@allenai.org | Pronouns: he/him

62 597Followers

6 739Following

17 303Posts total

20KViews on collected posts

Latest posts

Kyle Wiggers ✓@Kyle_L_Wiggers14 Sep

really cool to see how students used autodiscovery to figure out the tool's strengths and where it can be improved 1.1K views · 6 likes · 2 reposts · 0 replies Open on X →

Kyle Wiggers ✓@Kyle_L_Wiggers14 Sep

How should future scientists learn to interrogate AI tools for discovery? Read how @UW students put AutoDiscovery to the test as part of an academic challenge earlier this year, then try the tool for yourself—credits are now extended through Dec. 31. 🧵 https://t.co/SS1WOR5ltU h

4K views · 10 likes · 3 reposts · 2 replies Open on X →

Kyle Wiggers ✓@Kyle_L_Wiggers9 Sep

such a cool use case—thanks to @GoodfireAI for showcasing the power of our fully open model stack! 1.1K views · 7 likes · 0 reposts · 3 replies Open on X →

Kyle Wiggers ✓@Kyle_L_Wiggers1 Sep

a story as old as time but... some new evidence that llm benchmarks aren't measuring what you think they are 1.6K views · 5 likes · 0 reposts · 0 replies Open on X →

Kyle Wiggers ✓@Kyle_L_Wiggers1 Sep

We used BenchMIRT to audit popular LLM evals—and found some quirks. HarmBench mostly tests safety behavior, but its copyright questions depend more on reasoning. XSTest draws on both reasoning & safety, while ToxiGen provides little signal on either. https://t.co/OoGh4OTanG

1.8K views · 2 likes · 0 reposts · 2 replies Open on X →

Kyle Wiggers ✓@Kyle_L_Wiggers27 Aug

our ai autodiscovery tool contributed to a promising cancer finding—and it's exactly the kind of rigorous scientific work that we hope to enable more of. bravo to all the researcher teams involved 2K views · 13 likes · 1 reposts · 2 replies Open on X →

Kyle Wiggers ✓@Kyle_L_Wiggers26 Aug

great to see our open tools used to build better models for the world 1.4K views · 6 likes · 1 reposts · 0 replies Open on X →

Kyle Wiggers ✓@Kyle_L_Wiggers26 Aug

A Thai research team adapted our Dolma data-curation toolkit to build Mangosteen, a 47B-token corpus for Thai LLMs. They used Dolma to filter widely used web datasets into a smaller corpus that improved Thai LLM performance despite using less data. 🧵 https://t.co/exvf02cy4U htt

7K views · 36 likes · 6 reposts · 6 replies Open on X →

Against accounts of the same size

8 posts from the last 90 days, next to the 10K–100K follower range. ordinary reach for its size, weaker reaction than most.

Median views1 703this account914median for 10K–100K

Reach, %2.72%this account3.38%median for 10K–100K

Engagement, %0.60%this account1.53%median for 10K–100K

Metric	This account	Median for 10K–100K	Ratio
Median views per post	1 703	914	1.86×
Reach (views ÷ followers)	2.72%	3.38%	0.80×
Engagement rate	0.60%	1.53%	0.39×

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

7K26 Aug

1.4K

2K27 Aug

1.8K1 Sep

1.6K

1.1K9 Sep

4K14 Sep

1.1K

Last 8 collected posts, oldest on the left. The scale is logarithmic: one post can outrun the rest a hundred times over.

Engagement rate per post

0.78%26 Aug

0.48%

0.79%27 Aug

0.28%1 Sep

0.31%

0.89%9 Sep

0.45%14 Sep

0.72%

Reactions — likes, reposts, replies and quotes — divided by views. Median for 10K–100K accounts is 1.53%.

What the audience does

Likes56.3%85 in total

Reposts8.6%13 in total

Replies9.9%15 in total

Quotes6.6%10 in total

Bookmarks18.5%28 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.