© 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. 10-GEMMASCOPE-RES-131K
    7. 70792
    Prev
    Next
    INDEX
    Explanations

    copyright and license information

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

    This neuron activates on multi‐digit numeric sequences (e.g. years, version numbers, large numeric constants).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_10/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
    $
    -3.64
    "
    -3.52
    le
    -3.31
    what
    -3.09
    ers
    -3.05
    ли
    -2.94
    Y
    -2.91
    愎
    -2.91
    ent
    -2.88
    just
    -2.86
    POSITIVE LOGITS
    鉝
    3.16
    哙
    3.03
     “
    2.83
    荩
    2.77
    鈇
    2.75
    ’
    2.63
    
    2.63
     “…
    2.63
    捎
    2.61
     Pokud
    2.56
    Activations Density 0.554%

    No Known Activations