© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 60076
    Prev
    Next
    INDEX
    Explanations

    following `

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

    The neuron strongly activates on tokens made up largely of non-letter symbols or sequences of special/corrupted characters (e.g. percent-encodings, punctuation clusters, mojibake), effectively detecting “weird” non-alphanumeric fragments.

    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
     to
    -2.44
    .
    -1.38
     combustível
    -1.38
     fattori
    -1.34
     않았
    -1.34
     ótimo
    -1.33
     ícones
    -1.30
     along
    -1.25
     gênero
    -1.25
    tricos
    -1.23
    POSITIVE LOGITS
     estekak
    1.65
     limón
    1.53
     dick
    1.48
    дешь
    1.45
     hatta
    1.39
    ❳
    1.33
     parfüm
    1.28
    ⅕
    1.26
     sokak
    1.26
    bulaire
    1.24
    Activations Density 0.014%

    No Known Activations