© 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. 47739
    Prev
    Next
    INDEX
    Explanations

    Das followed by names

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

    The neuron is triggered by the character sequence “das” (in any token containing “das,” e.g. Dasabuvir, elle::das, Das‐named authors).

    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
    菹
    -2.95
    岜
    -2.86
    唞
    -2.83
    缐
    -2.69
    ではまた
    -2.69
     tijden
    -2.69
    ではでは
    -2.64
    酲
    -2.64
    渽
    -2.50
    夲
    -2.50
    POSITIVE LOGITS
    ا
    3.25
    y
    2.88
     lange
    2.75
     waren
    2.55
    as
    2.55
     emphatic
    2.53
    ’
    2.39
     количество
    2.36
    ar
    2.25
     een
    2.23
    Activations Density 0.002%

    No Known Activations