© 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. 72360
    Prev
    Next
    INDEX
    Explanations

    consequence or conclusion markers

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

    The neuron detects high-level discourse/connective words (e.g. “As such,” “Therefore,” “Subsequently,” etc.) that signal logical transitions between sentences.

    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
     because
    -1.09
    由於
    -1.06
     can
    -1.05
     staunch
    -0.96
     laik
    -0.96
    servez
    -0.94
     KNOW
    -0.90
     Because
    -0.89
    錚
    -0.89
     it
    -0.86
    POSITIVE LOGITS
    hence
    1.05
    复活
    0.95
    才會
    0.95
     لذا
    0.95
     ägg
    0.92
     الترك
    0.91
     Réalisation
    0.91
    therefore
    0.89
    گز
    0.88
     Ramón
    0.88
    Activations Density 0.065%

    No Known Activations