© 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. 73092
    Prev
    Next
    INDEX
    Explanations

    quantities and measurements

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

    This neuron activates on numeric recipe measurements, especially fractions (e.g. “1/2,” “3/4,” etc.).

    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
    卨
    -2.06
     interessantes
    -1.91
    子は
    -1.91
     disimpan
    -1.88
     dikirim
    -1.87
    方針
    -1.84
    viève
    -1.82
    が
    -1.82
    家は
    -1.78
    姪
    -1.78
    POSITIVE LOGITS
    為に
    2.59
    t
    2.20
    S
    2.09
    N
    2.02
    D
    2.00
     mehreren
    1.97
    C
    1.96
    T
    1.95
    L
    1.94
    J
    1.92
    Activations Density 0.009%

    No Known Activations