© 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. 10-GEMMASCOPE-RES-131K
    7. 128455
    Prev
    Next
    INDEX
    Explanations

    who will call you

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

    The neuron activates on English “wh-” words (who, which, when, where, etc.), marking question or relative/adverbial pronouns.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_10/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
     what
    -1.37
    これらの
    -1.34
    このような
    -1.34
    })$.
    -1.30
    푣
    -1.30
    ekyll
    -1.28
     totiž
    -1.27
     equilíbrio
    -1.27
    そのような
    -1.27
    ).
    -1.26
    POSITIVE LOGITS
     then
    1.80
    rededor
    1.48
     alrededores
    1.45
    輒
    1.38
     gevonden
    1.36
    👯
    1.35
    殒
    1.34
    geving
    1.34
     için
    1.34
    ą
    1.31
    Activations Density 0.055%

    No Known Activations