© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 13698
    Prev
    Next
    INDEX
    Explanations

    tax returns and IRS

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

    This neuron fires strongly on numeric data—years, amounts, counts—especially in tax‐ or return‐related contexts.

    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
    v
    -1.70
    end
    -1.56
    d
    -1.41
     It
    -1.35
     of
    -1.34
    k
    -1.34
    遴
    -1.31
    ges
    -1.28
    t
    -1.27
     l
    -1.26
    POSITIVE LOGITS
     acostumb
    1.66
     acoged
    1.63
     język
    1.56
    錀
    1.55
    ఏ
    1.55
    魠
    1.53
    vejec
    1.49
    fanart
    1.48
     conseillé
    1.48
     usamos
    1.48
    Activations Density 0.022%

    No Known Activations