© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 39324
    Prev
    Next
    INDEX
    Explanations

    fill color

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

    The neuron fires on graphics “fill” or color‐setting attributes and calls—i.e. any mention of fill/FillColor (and related styling commands) in vector or drawing code.

    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
     INFORMACIÓN
    -1.16
     freaking
    -1.08
    getWriter
    -1.05
    も良い
    -1.05
    ynek
    -1.04
    ID
    -1.03
    API
    -1.02
    TO
    -1.02
    必要があります
    -1.02
    A
    -1.01
    POSITIVE LOGITS
    ஊ
    1.18
    intéress
    1.09
    ᴮ
    1.07
     participó
    1.06
     beteken
    1.05
    kitab
    1.04
     proposons
    1.03
    bàn
    1.03
    面積
    1.02
     arbete
    1.02
    Activations Density 0.020%

    No Known Activations