© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-3-27B-IT
    5. 31-GEMMASCOPE-2-RES-262K
    7. 8318
    Prev
    Next
    INDEX
    Explanations

    role of or innovation

    np_acts-logits-general · gemini-2.5-flash-lite

    high-frequency function words—especially prepositions—within instructional or definitional text.

    oai_token-act-pair · gpt-5Triggered by @vetterc0
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-2-27b-it/resid_post/layer_31_width_262k_l0_medium
    Prompts (Dashboard)
    238,145 prompts, 512 tokens each
    Dataset (Dashboard)
    lmsys + oasst1
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     രണ്ട്
    0.53
     sadece
    0.50
     두
    0.48
     kunt
    0.47
     రెండు
    0.47
     இரண்டு
    0.46
     doen
    0.46
     दोन
    0.46
     tiež
    0.45
     two
    0.45
    POSITIVE LOGITS
     или
    0.57
    或
    0.51
    创新
    0.47
    創新
    0.45
    某个
    0.45
    或者
    0.44
    Innovation
    0.43
    企業
    0.43
     Innovation
    0.42
    企业
    0.41
    Activations Density 0.100%

    No Known Activations