© 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
    2. Gemma-2-27B
    3. 10-GEMMASCOPE-RES-131K
    4. 128693
    Prev
    Next
    INDEX
    Explanations

    Gh, horticulture, Blood rheology

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

    The neuron identifies fantasy, folklore, or horror-related terms, particularly "ghoul", "ghouls", "ghoulish", and other similar supernatural or monster-related words.

    oai_token-act-pair · claude-3-7-sonnet-20250219Triggered by @neilrathi

    The neuron fires on occurrences of the substring “ghou” (as in “ghoulish,” “ghouls,” “ghoul-faced,” etc.), i.e. references to ghouls or similar spooky creatures.

    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
     copertina
    -2.41
    ፨
    -2.34
    傒
    -2.28
    摃
    -2.27
    ”;
    -2.20
    mappedBy
    -2.13
    跸
    -2.13
    谖
    -2.09
     endregion
    -2.06
    ristiano
    -2.03
    POSITIVE LOGITS
    '
    2.66
    {
    2.59
    ;
    2.13
    }
    2.06
     of
    1.78
     '
    1.77
    \\
    1.71
    那我
    1.69
    jadikan
    1.68
    ًا
    1.62
    Activations Density 0.003%

    No Known Activations