© 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. 65518
    Prev
    Next
    INDEX
    Explanations

    lists of topics or categories

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

    The neuron fires strongly on named topics or thematic headings (often multi‐word noun phrases, frequently in quotes) such as “Mental Health,” “Menstruation,” “Conduct and behaviour,” etc.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    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
    quran
    -1.09
     文章
    -1.04
    使用了
    -1.03
     vivía
    -0.98
    Magento
    -0.97
    banan
    -0.96
     عبارت
    -0.94
    Җ
    -0.94
    counted
    -0.92
     metode
    -0.92
    POSITIVE LOGITS
     topics
    1.30
     other
    1.20
     areas
    1.16
     separate
    1.11
     бампер
    1.09
     topic
    1.08
     zostanie
    1.03
    וּ
    1.02
     another
    1.02
    zembro
    1.00
    Activations Density 0.049%

    No Known Activations