© 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. 66064
    Prev
    Next
    INDEX
    Explanations

    script tags and stylesheets

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

    The neuron detects HTML `<script>` tags that load external JavaScript (i.e. `script src="…"` includes).

    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
     Ausrüstung
    -1.09
    ěř
    -1.08
     grudnia
    -1.08
     this
    -1.07
     września
    -1.06
     czerwca
    -1.03
     another
    -1.01
     which
    -0.99
     chose
    -0.96
     their
    -0.96
    POSITIVE LOGITS
    Kilde
    1.06
     WARNING
    0.99
     rime
    0.92
    isbol
    0.92
     både
    0.91
    kabau
    0.90
     Dien
    0.90
     musul
    0.90
     disponibilités
    0.90
    ffiti
    0.89
    Activations Density 0.010%

    No Known Activations