• Home
  • Technology
  • Gaming
  • Entertainment
  • World & Business
  • Science
  • Sports
  • AI
HomeTechnologyGamingEntertainmentWorld & BusinessScienceSportsAI
  • HomeTechnologyGamingEntertainmentWorld & BusinessScienceSportsAI
    • Home
    • Technology
    • Gaming
    • Entertainment
    • World & Business
    • Science
    • Sports
    • AI
    AI

    COLM Paper Traces OLMo3-7B Capabilities to Dolma3 Data

    Glenn Matlin announces COLM 2026 paper using gradient attribution on Dolma3.

    T(
    GM
    2 Sources, 32d ago, first seen 32d ago

    TLDR

    Glenn Matlin posted that a COLM 2026 paper traces capability provenance of OLMo3-7B across Dolma3 with gradient-based training data attribution. A visible reply notes signed influence heatmaps across 576 bins that show SocialIQA spreading positive mass broadly. Matlin supplied links to the paper and code in the post. The thread supplies the first-party announcement of the researchers' own work; no independent corroboration of results appears in the packet.

    Combined views

    20K

    2 Sources, first seen 32d ago

    Combined views

    20K

    2 Sources, first seen 32d ago

    266 likes
    266 likes
    11 comments
    241 saves
    31 reposts

    Sentiment

    Positive——Negative

    Summary

    Not enough discussion yet.

    No sentiment analysis available yet.

    11 comments
    241 saves
    31 reposts
    Today's Rank

    —

    Not ranked yet

    Today's Rank

    —

    Not ranked yet

    Sentiment

    Positive——Negative

    Summary

    Not enough discussion yet.

    No sentiment analysis available yet.

    2 Sources

    @GlennMatlin🤔Where in trillions of pre-training tokens do capabilities actually come from? 💥In our new COLM 2026 paper, we trace the capability provenance of OLMo3-7B across Dolma3 using gradient-based training data attribution 🧵 Paper: https://arxiv.org/abs/2606.19625 Code: https://github.com/HCAI-Lab-GT/capabilibara Explorer: https://huggingface.co/spaces/HCAI-Lab-GT/capabilibara
    @teortaxesTexRT @GlennMatlin: Signed influence heatmaps across 576 bins (24 topics × 24 formats): SocialIQA (social reasoning) spreads positive mass br…

    2 Sources

    @GlennMatlin🤔Where in trillions of pre-training tokens do capabilities actually come from? 💥In our new COLM 2026 paper, we trace the capability provenance of OLMo3-7B across Dolma3 using gradient-based training data attribution 🧵 Paper: https://arxiv.org/abs/2606.19625 Code: https://github.com/HCAI-Lab-GT/capabilibara Explorer: https://huggingface.co/spaces/HCAI-Lab-GT/capabilibara
    @teortaxesTexRT @GlennMatlin: Signed influence heatmaps across 576 bins (24 topics × 24 formats): SocialIQA (social reasoning) spreads positive mass br…