© 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. 33812
    Prev
    Next
    INDEX
    Explanations

    list items

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

    The neuron fires on HTML/XML‐style markup delimiters (the angle‐bracket tags and tag names).

    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
    捗
    -1.07
     Reinh
    -0.91
     at
    -0.90
     also
    -0.88
     of
    -0.87
    推進
    -0.86
     who
    -0.86
    <bos>
    -0.84
     because
    -0.84
     reus
    -0.82
    POSITIVE LOGITS
     появля
    0.89
    enial
    0.89
    rightarrow
    0.88
    :</
    0.88
    んですよ
    0.88
     gewel
    0.87
     acercarse
    0.87
    xiety
    0.87
    ouro
    0.85
     girar
    0.85
    Activations Density 0.015%

    No Known Activations