© 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. 22-GEMMASCOPE-RES-131K
    7. 128239
    Prev
    Next
    INDEX
    Explanations

    The neuron strongly activates on standalone content words used as headings or titles (i.e. single words in section headers or page titles).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408

    university admissions, interview, books, events

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_22/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
     is
    -1.13
    Today
    -1.00
    のではないでしょうか
    -0.98
    的就
    -0.96
     berdasarkan
    -0.96
     extremely
    -0.93
     freio
    -0.93
     esporte
    -0.92
    Compared
    -0.92
     etiam
    -0.92
    POSITIVE LOGITS
     אנשים
    1.15
     ziren
    1.04
    ̋
    0.96
    rogène
    0.93
    ulemon
    0.89
    gänge
    0.88
     ».
    0.86
    zupeł
    0.86
     InputMaybe
    0.85
    ₽
    0.84
    Activations Density 0.002%

    No Known Activations