© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 71723
    Prev
    Next
    INDEX
    Explanations

    simple, polar, photo, first

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

    The neuron activates on numeric reference tokens (e.g. years, volume or page numbers) in scholarly‐style citations.

    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
     Outils
    -0.92
     ২০১
    -0.89
     bekom
    -0.88
    +:
    -0.87
    дизайн
    -0.86
    \#:
    -0.86
    嗐
    -0.84
    turismo
    -0.82
     oči
    -0.82
    POLIT
    -0.82
    POSITIVE LOGITS
     even
    0.99
     only
    0.99
     discovered
    0.97
    💾
    0.95
     there
    0.93
    言うと
    0.93
    noy
    0.92
     mischievous
    0.91
     laku
    0.91
    ,"\
    0.89
    Activations Density 0.001%

    No Known Activations