© 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. 84340
    Prev
    Next
    INDEX
    Explanations

    PL General

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

    The neuron strongly activates on numeric tokens (especially decimal numbers), i.e. it detects occurrences of numbers in the text.

    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
    謖
    -1.81
    $+$
    -1.78
    
    -1.70
    "—
    -1.70
    }
    -1.65
    <td>
    -1.63
     acompañada
    -1.63
    );
    -1.59
    好的
    -1.57
    回来的
    -1.57
    POSITIVE LOGITS
     not
    2.03
    曁
    1.99
    labelControl
    1.80
    ꒥
    1.80
     rosy
    1.77
    Hanya
    1.73
    Pemer
    1.69
    〮
    1.66
    Ꮸ
    1.65
    ⬮
    1.64
    Activations Density 0.095%

    No Known Activations