© 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. 128481
    Prev
    Next
    INDEX
    Explanations

    The neuron detects references to social media platforms and other subscription or link cues (e.g., “Twitter,” “Facebook,” “Subscribe,” “.com”).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408

    URLs, social media handles, questions, and actions

    np_acts-logits-general · gemini-2.5-flash-lite
    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
    Nguyễn
    -1.05
    addCell
    -1.00
    JUST
    -1.00
    wist
    -0.97
    بی
    -0.96
    тив
    -0.95
     bym
    -0.93
    rencia
    -0.93
     mascotas
    -0.92
     moze
    -0.92
    POSITIVE LOGITS
     darling
    0.94
    ້
    0.90
    さは
    0.90
    Nils
    0.90
    4
    0.85
    دانلود
    0.83
    Boxscore
    0.83
     getIntent
    0.82
     [''],
    0.82
     denselben
    0.80
    Activations Density 0.007%

    No Known Activations