© 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. 15476
    Prev
    Next
    INDEX
    Explanations

    piecewise function definitions

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

    This neuron detects question contexts, i.e. question‐mark “?” tokens and other cues that mark questions (like “if,” “in,” 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
    3
    -1.58
    4
    -1.57
    6
    -1.44
    2
    -1.44
    雰囲
    -1.43
    枼
    -1.42
    彐
    -1.41
    賸
    -1.38
    8
    -1.38
    entrega
    -1.37
    POSITIVE LOGITS
    -
    1.63
     (
    1.60
    ,
    1.55
     by
    1.43
     because
    1.41
     to
    1.41
    a
    1.38
     both
    1.36
    it
    1.32
     it
    1.28
    Activations Density 0.002%

    No Known Activations