© 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. 67656
    Prev
    Next
    INDEX
    Explanations

    kw, TGF

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

    The neuron identifies the abbreviation "KW" in various contexts including references to people's names, brands, and technical specifications.

    oai_token-act-pair · claude-3-7-sonnet-20250219Triggered by @neilrathi
    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
    </i>
    -2.94
    摭
    -2.83
     a
    -2.81
     jeopardi
    -2.63
     hordes
    -2.44
     kleines
    -2.39
    鉨
    -2.38
     characterize
    -2.36
    ',
    -2.34
     ‘‘
    -2.34
    POSITIVE LOGITS
     рассказала
    2.92
    >
    2.88
    \
    2.80
    on
    2.77
    We
    2.55
    *
    2.47
    z
    2.47
    '
    2.44
     сказала
    2.39
    /
    2.30
    Activations Density 0.003%

    No Known Activations