© 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. 99235
    Prev
    Next
    INDEX
    Explanations

    words starting with Bou

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

    The neuron selectively activates on occurrences of the character sequence “Bou,” i.e. tokens forming the start of surnames or proper names beginning with “Bou.”

    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
     bemerkt
    -2.73
     schönes
    -2.73
    嗞
    -2.72
     unſ
    -2.66
    pecabe
    -2.64
    スポンサーリンク
    -2.64
     The
    -2.63
     kleines
    -2.56
     cittadino
    -2.55
    ᷆
    -2.55
    POSITIVE LOGITS
    怍
    2.72
     và
    2.63
    \
    2.55
    Ꭸ
    2.55
    —
    2.52
    鹚
    2.41
    一会儿
    2.36
    сить
    2.36
    2.34
     και
    2.31
    Activations Density 0.002%
    No Known Activations