© 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. 25334
    Prev
    Next
    INDEX
    Explanations

    and or conditions

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

    The neuron fires on occurrences of the conjunction “and” (often in all-caps AND) used to join conditions or clauses in code/queries.

    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
    蹰
    -2.83
    轳
    -2.56
    𓁹
    -2.52
    ſu
    -2.52
    
    -2.39
    
    -2.34
    蓠
    -2.30
    𓆏
    -2.28
     rése
    -2.27
    莛
    -2.27
    POSITIVE LOGITS
     in
    2.77
    There
    2.56
    Despite
    2.33
    When
    2.28
     enormous
    2.27
    just
    2.25
     intensely
    2.20
    After
    2.16
    Every
    2.16
     there
    2.14
    Activations Density 0.003%

    No Known Activations