# Ai2 — How a Georgia Tech team used the open Olmo stack to trace social reasoning

- Company: Ai2 (allenai.org)
- Announced: 2026-08-21
- Category: research-paper
- Coverage: not counted
- Announcement: no
- Group: routine
- Source: https://allenai.org/blog/olmo-capability-tracing
- Record: https://forck.live/items/18538-how-a-georgia-tech-team-used-the-open-olmo-stack-to-trace-social-reasoning
- Subject: Olmo / Molmo / Asta
- Models affected: Olmo 3

Georgia Tech researchers used Ai2's open Olmo ecosystem—including Olmo 3, Dolma 3, WebOrganizer, OlmoEval, and OLMES—to trace the training data origins of social reasoning in a language model. They found that dialogue-rich, interpersonal writing categories most influenced Olmo 3's social reasoning, and a causal test showed that removing the most influential literature documents significantly reduced the model's SocialIQA score.

## Evidence

Verbatim from https://allenai.org/blog/olmo-capability-tracing:

> Matlin and Chakraborty then ran a causal test: if they deleted the data that most shaped an ability, would the model get worse at it? They had Olmo 3 "unlearn" the most influential documents in Dolma's literature category, which was the category most tied to Olmo 3’s social reasoning. Its SocialIQA score dropped further than when they removed random documents from that category.

---

Record: https://forck.live/items/18538-how-a-georgia-tech-team-used-the-open-olmo-stack-to-trace-social-reasoning
Catalogue: https://forck.live/llms.txt
Current issue: https://forck.live/feed.md