© 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-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 67979
    Prev
    Next
    INDEX
    Explanations

    thereto, heretofore, or to

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

    This neuron lights up on short, standalone document‐structure tokens (like “to,” “thereto,” single letters such as “B,” or other brief markers) that typically introduce or label sections.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_10/width_131k
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    s
    -2.45
    
    -2.25
     was
    -1.95
    .
    -1.91
    但也
    -1.83
    ly
    -1.80
     "";
    -1.79
    机的
    -1.75
    チョイス
    -1.75
    mo
    -1.74
    POSITIVE LOGITS
     entendido
    2.16
    
    2.13
    鹋
    2.11
    􀂃
    2.06
    媗
    2.03
    Ἵ
    2.03
     protegida
    1.95
    幛
    1.91
     ส์
    1.88
     vulgares
    1.86
    Activations Density 0.001%

    No Known Activations