© 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 EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 72961
    Prev
    Next
    INDEX
    Explanations

    code delimiters and separators

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

    This neuron specializes in spotting long runs of repeated punctuation characters (e.g. lines of dashes, pluses, stars, or similar symbols) used as separators or section dividers.

    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
     molti
    -0.89
     katholischen
    -0.88
    ļu
    -0.87
    estructuras
    -0.87
    InsertCommand
    -0.86
     første
    -0.84
    rard
    -0.82
     fleste
    -0.82
     preuß
    -0.80
    sso
    -0.79
    POSITIVE LOGITS
    ควร
    1.11
     érz
    0.95
    Ӏ
    0.88
    alış
    0.87
    wię
    0.87
     Apel
    0.86
    ホット
    0.85
     Opus
    0.85
     měl
    0.85
    gués
    0.84
    Activations Density 0.003%

    No Known Activations