© 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. 128186
    Prev
    Next
    INDEX
    Explanations

    The neuron strongly activates on multi-digit numeric values (e.g. counts, percentages, or other large numeric tokens).

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

    numbers and counts

    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
     except
    -0.99
     try
    -0.98
    に行きました
    -0.95
     rail
    -0.93
    之外
    -0.89
    forall
    -0.88
     but
    -0.88
     every
    -0.85
     must
    -0.83
     besondere
    -0.83
    POSITIVE LOGITS
    beste
    1.07
     kosong
    1.02
     réf
    1.01
    zessin
    0.93
    chapper
    0.93
    MEA
    0.92
     masker
    0.88
    👖
    0.88
     クリニック
    0.87
    ?」
    0.87
    Activations Density 0.098%

    No Known Activations