© 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. 33574
    Prev
    Next
    INDEX
    Explanations

    XML/HTML tags and comments

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

    The neuron fires on HTML conditional‐comment syntax, especially the tokens used in IE-specific “<!--[if IE]>…<![endif]-->” blocks.

    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
    ágok
    -1.13
     Federación
    -1.10
    ındaki
    -1.05
    )=>{
    -1.04
     instala
    -1.03
     typically
    -1.03
    Butyl
    -1.02
     WAYNE
    -1.01
     it
    -1.01
     viajes
    -1.00
    POSITIVE LOGITS
     pengelola
    1.07
     Bestimmungen
    1.07
     kehadiran
    1.03
    ഞ്ഞ
    1.03
     laun
    1.02
     Поэтому
    1.02
     Однако
    1.00
     пун
    0.98
     BUG
    0.98
     prodigious
    0.94
    Activations Density 0.010%

    No Known Activations