© 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. 78448
    Prev
    Next
    INDEX
    Explanations

    XML tags and code elements

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

    This neuron activates on numeric tokens (numbers), effectively detecting numeric literals 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_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
     its
    -1.12
     dátum
    -1.04
     kapit
    -1.04
    黎明
    -1.03
     беже
    -1.03
     minde
    -1.02
     both
    -1.01
    ՝
    -0.99
    entukan
    -0.99
    　・
    -0.98
    POSITIVE LOGITS
    ko
    1.18
     betreffenden
    1.16
    ch
    1.13
     möglichen
    1.09
    [].
    1.09
    ha
    1.08
    ka
    1.08
    le
    1.07
    be
    1.06
    ph
    1.05
    Activations Density 0.000%

    No Known Activations