© 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. 128432
    Prev
    Next
    INDEX
    Explanations

    The main thing this neuron does is detect occurrences of “isolat-” (e.g., isolate, isolation) indicating the concept of being set apart or alone.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408

    isolating or separation for specific purpose

    np_acts-logits-general · gemini-2.5-flash-lite
    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
    >
    -2.19
    '
    -2.13
     tentando
    -2.08
    ultuous
    -2.06
    _{
    -2.03
     to
    -2.03
    Of
    -2.03
    就是
    -1.97
     empfohlen
    -1.96
     изготов
    -1.92
    POSITIVE LOGITS
    鹸
    2.38
    𖤣
    2.22
    gencias
    2.16
    ність
    2.11
    琑
    2.06
     \
    2.05
     \&
    2.03
    ☫
    1.97
    ,“
    1.95
     céle
    1.95
    Activations Density 0.020%

    No Known Activations