© 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. 22-GEMMASCOPE-RES-131K
    7. 54394
    Prev
    Next
    INDEX
    Explanations

    foreign language words

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

    The main thing this neuron does is detect promotional gambling-and-treasure-hunting marketing terms (e.g. casino, game, treasures, pyramids).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_22/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
    ...>
    -1.59
    strator
    -1.53
    ...@
    -1.45
    iation
    -1.27
    ggg
    -1.25
    ibility
    -1.23
    ardized
    -1.23
    gages
    -1.22
    caping
    -1.22
     letz
    -1.21
    POSITIVE LOGITS
    就可以了
    1.63
     atlet
    1.55
    就行了
    1.55
     recevez
    1.45
     médec
    1.41
     слой
    1.41
     contactez
    1.40
    糴
    1.39
     vám
    1.39
     задний
    1.38
    Activations Density 0.003%

    No Known Activations