© 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. 128403
    Prev
    Next
    INDEX
    Explanations

    margin-top, margin-bottom

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

    The neuron fires on numeric literal tokens—especially large pixel values—in CSS property declarations.

    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
    狲
    -1.79
    excellent
    -1.74
    спомина
    -1.69
     وم
    -1.64
    返す
    -1.63
    ٞ
    -1.63
    ᗷ
    -1.63
     ""],
    -1.63
    чера
    -1.62
    poque
    -1.60
    POSITIVE LOGITS
     that
    1.62
    はその
    1.54
     of
    1.54
     ''
    1.52
     itself
    1.49
    Pollution
    1.45
    欵
    1.45
    鱘
    1.45
     سایر
    1.42
    蒾
    1.41
    Activations Density 0.001%

    No Known Activations