© 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. 113195
    Prev
    Next
    INDEX
    Explanations

    foreign and technical words

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

    This neuron lights up on uncommon or specialized tokens—such as acronyms, technical identifiers, or non‐Latin script words.

    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
    D
    -3.66
    F
    -3.27
    A
    -3.23
    section
    -3.13
    可以使用
    -3.05
    R
    -2.97
    most
    -2.95
    man
    -2.89
    H
    -2.80
    C
    -2.78
    POSITIVE LOGITS
    ו
    2.77
    و
    2.69
    퐸
    2.45
     It
    2.45
    dámské
    2.34
    ého
    2.31
    VNB
    2.27
    zeniem
    2.25
     Trotz
    2.23
    𓐍
    2.23
    Activations Density 0.001%

    No Known Activations