© 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
    2. Gemma-2-27B
    3. 22-GEMMASCOPE-RES-131K
    4. 71517
    Prev
    Next
    INDEX
    Explanations

    dangerous range

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

    The neuron fires on code‐style file path or file name tokens (e.g. header names and paths with “/” or “.h”).

    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
     trám
    -1.41
     WIL
    -1.35
     soooo
    -1.26
     ΤΟ
    -1.25
     tivi
    -1.16
     charles
    -1.15
     tuong
    -1.14
     luy
    -1.13
     sooo
    -1.13
    鹇
    -1.10
    POSITIVE LOGITS
     as
    1.30
     during
    1.10
     frantic
    1.08
     longstanding
    1.07
     before
    1.07
     große
    1.07
     During
    1.07
     or
    1.06
     from
    1.05
     by
    1.04
    Activations Density 0.019%

    No Known Activations