© 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. 128038
    Prev
    Next
    INDEX
    Explanations

    The neuron activates on in-text reference markers (superscripted or bracketed citation/footnote numbers).

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

    citation and reference markers

    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
     auriculares
    -1.20
    樾
    -1.06
    艤
    -1.01
     joyería
    -1.00
    そこから
    -1.00
    などが
    -0.98
     différentes
    -0.98
    神
    -0.97
    车的
    -0.97
     innych
    -0.96
    POSITIVE LOGITS
     we
    1.20
    $
    1.19
    It
    1.02
    城
    1.02
    \
    1.01
    !
    1.00
     with
    0.98
    0.98
    |
    0.96
     will
    0.95
    Activations Density 0.001%
    No Known Activations