© 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. 66034
    Prev
    Next
    INDEX
    Explanations

    disclaimers and technical terms

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

    This neuron detects formal medical or product-disclaimer language (e.g. “Disclaimer,” “not intended to diagnose/treat/cure,” “consult your physician,” “Food and Drug Administration,” etc.).

    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
    calciatore
    -0.98
    athed
    -0.96
    nament
    -0.94
    futbolista
    -0.94
     banyo
    -0.93
     établissement
    -0.92
     samtidigt
    -0.92
    奖励
    -0.92
    denom
    -0.90
     engineers
    -0.90
    POSITIVE LOGITS
     you
    1.32
    まれて
    1.20
     or
    1.18
    ׄ
    1.08
    月号
    1.04
    欄
    1.02
    Kilder
    1.02
    сколько
    1.01
     ویتامین
    1.01
     сахара
    1.00
    Activations Density 0.007%

    No Known Activations