© 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. 34-GEMMASCOPE-RES-131K
    7. 47739
    Prev
    Next
    INDEX
    Explanations

    numerical citations and symbols

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

    The neuron is selectively detecting long numeric ID strings (e.g. multi‐digit citation or reference numbers).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_34/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
     and
    -1.20
     invigor
    -1.19
     Another
    -1.16
     ensuring
    -1.16
     Some
    -1.14
     ostensibly
    -1.14
     achieving
    -1.10
     presumably
    -1.09
     from
    -1.08
     for
    -1.07
    POSITIVE LOGITS
     congrès
    1.47
     nôtre
    1.30
    んでも
    1.16
    
    1.14
    懟
    1.12
    
    1.11
     stockfoto
    1.11
    Jeudi
    1.10
    Directeur
    1.10
    ꟷ
    1.09
    Activations Density 0.013%

    No Known Activations