© 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. 34-GEMMASCOPE-RES-131K
    7. 885
    Prev
    Next
    INDEX
    Explanations

    religious and academic titles

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

    This neuron spikes on the very first substantive token in each new segment or structural block (e.g. paragraph breaks, code or markdown sections).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_34/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
    〜〜
    -0.85
    cheinend
    -0.84
     Illustrations
    -0.83
     Europea
    -0.83
     بازیگر
    -0.81
     judicia
    -0.80
     antici
    -0.80
     afbeeldingen
    -0.79
     brilla
    -0.79
     diyor
    -0.79
    POSITIVE LOGITS
     Orient
    0.82
    徵
    0.82
    א
    0.79
    õe
    0.73
    交易
    0.73
    地震
    0.72
    மே
    0.72
    ρω
    0.71
    Pop
    0.70
    prototype
    0.70
    Activations Density 0.001%

    No Known Activations