© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 39617
    Prev
    Next
    INDEX
    Explanations

    pricing, language, and customer details

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

    The neuron fires on metadata tokens and section‐heading or call-to-action words (often in ALL CAPS) such as “CALL,” “AVAILABILITY,” “Customer,” “NOTE,” 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
     pappa
    -1.17
    backButton
    -1.15
    defaultProps
    -1.15
     supremo
    -1.03
    占用
    -1.01
     Hanks
    -1.00
     横浜
    -0.99
    propTypes
    -0.98
     新年
    -0.98
    modificar
    -0.98
    POSITIVE LOGITS
     this
    1.27
    itemBuilder
    1.13
    quirer
    1.09
     согласно
    1.03
     due
    1.03
     homicidio
    1.02
    婭
    1.02
    addCriterion
    1.02
    ține
    1.02
    chene
    0.98
    Activations Density 0.089%

    No Known Activations