© 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. 45022
    Prev
    Next
    INDEX
    Explanations

    in contexts and identifiers

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

    The neuron strongly activates on occurrences of the exact token “in.”

    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
    to
    -3.38
    but
    -2.50
    and
    -2.44
    0
    -2.39
    by
    -2.25
    as
    -2.23
    te
    -2.20
    us
    -2.19
    ut
    -2.14
    3
    -2.14
    POSITIVE LOGITS
    ⎚
    2.13
    руется
    2.11
    liſh
    2.09
     impedir
    2.05
    潆
    2.03
    訛
    2.00
    自体は
    1.98
    遄
    1.98
    laşı
    1.95
    僦
    1.95
    Activations Density 0.015%

    No Known Activations