© 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. 65257
    Prev
    Next
    INDEX
    Explanations

    what followed by pronoun/verb

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

    The neuron is strongly triggered by occurrences of the word “what” (especially in title‐style or emphasis contexts).

    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
     if
    -1.23
     and
    -1.23
     BEGIN
    -1.10
    就开始
    -1.09
    ￬
    -1.09
     конкурса
    -1.08
    gemeinde
    -1.07
     whiteColor
    -1.05
     yp
    -1.05
    item
    -1.04
    POSITIVE LOGITS
     they
    1.73
     already
    1.32
     that
    1.20
    已經
    1.05
     you
    1.03
    ']='
    1.03
     happening
    1.02
     those
    1.02
     we
    1.01
    already
    1.00
    Activations Density 0.017%

    No Known Activations