© 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. 72544
    Prev
    Next
    INDEX
    Explanations

    script or link tags

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

    This neuron responds to the “<” character that begins HTML (or XML) tags—i.e. it detects the opening angle bracket of markup elements.

    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
     recomendable
    -1.20
    我就
    -1.19
    complished
    -1.17
    likely
    -1.17
    AÇÃO
    -1.16
     apropri
    -1.13
     complementa
    -1.11
    olio
    -1.09
     чувстви
    -1.09
     necesar
    -1.09
    POSITIVE LOGITS
    ↵↵
    1.57
     also
    1.45
     and
    1.29
     Hinsicht
    1.17
    We
    1.13
     tambahan
    1.10
     offiziellen
    1.04
    ↵
    1.02
     urchin
    1.02
     zuerst
    0.97
    Activations Density 0.013%

    No Known Activations