© 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. 15096
    Prev
    Next
    INDEX
    Explanations

    questions

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

    This neuron fires on occurrences of the word “question” (and its variants like “Question” or “questions”) in the text.

    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
     Doesn
    -2.53
    在那
    -2.44
     molte
    -2.42
     gruesome
    -2.42
     You
    -2.39
     startling
    -2.31
     absurd
    -2.30
     Designed
    -2.30
     Provide
    -2.23
     THERE
    -2.19
    POSITIVE LOGITS
    .
    3.55
    涐
    2.50
    鋮
    2.41
     zupeł
    2.20
    man
    2.11
    举动
    2.11
     PACKAGE
    2.09
    
    2.08
    뿡
    2.08
     sixteen
    2.06
    Activations Density 0.038%

    No Known Activations