© 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. 12647
    Prev
    Next
    INDEX
    Explanations

    mailto: address links

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

    The neuron fires on parts of the text that look like hyperlinks or email‐address references (e.g. “mailto,” “http,” “to” in mailto URLs, “email,” etc.).

    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
     entanto
    -1.23
     berbagai
    -1.21
    nyataan
    -1.20
     vilken
    -1.17
     and
    -1.17
     which
    -1.17
    StreetMap
    -1.16
     suggest
    -1.14
    ográfica
    -1.13
     malaise
    -1.13
    POSITIVE LOGITS
    -
    1.42
    “
    1.31
     the
    1.27
     necesit
    1.21
    !
    1.19
    *
    1.19
     yearning
    1.19
    every
    1.18
    seven
    1.15
     médical
    1.15
    Activations Density 0.000%

    No Known Activations