© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 87350
    Prev
    Next
    INDEX
    Explanations

    Greenwich Village, Wichita Falls

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

    The neuron responds to tokens containing the sequence “wich,” especially in place names (e.g. Greenwich, Woolwich, Wichita).

    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
    ࿚
    -2.63
     wth
    -2.47
     breiten
    -2.33
    (
    -2.31
    有事
    -2.28
    時間を
    -2.22
    呌
    -2.20
    厹
    -2.17
     damski
    -2.14
    雷霆
    -2.14
    POSITIVE LOGITS
    s
    3.52
     is
    2.72
    k
    2.64
     When
    2.59
     από
    2.44
     cuánto
    2.42
     episodi
    2.42
    硨
    2.34
    殛
    2.30
    </i>
    2.27
    Activations Density 0.003%

    No Known Activations