© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 14782
    Prev
    Next
    INDEX
    Explanations

    see, e.g., for example

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

    The neuron fires strongly on citation or cross-reference cues (e.g. “See,” “cf.”) that signal reference to other documents or authorities.

    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
    翆
    -1.45
    kebox
    -1.33
    zapatos
    -1.33
    camisetas
    -1.23
    Hva
    -1.23
    ݈
    -1.23
    Ꭸ
    -1.22
    ئية
    -1.20
    ffiche
    -1.18
     ͡°)
    -1.16
    POSITIVE LOGITS
    ,
    1.65
    </em>
    1.64
    '
    1.54
    e
    1.45
    -
    1.28
    d
    1.20
    $
    1.14
    “
    1.13
    "
    1.13
     (
    1.10
    Activations Density 0.002%

    No Known Activations