© 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. 128210
    Prev
    Next
    INDEX
    Explanations

    including enumeration or conditions

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

    The neuron strongly activates on tokens around precise numerical data (especially decimals and statistics), so it’s detecting passages containing quantitative/statistical information.

    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
    serien
    -1.30
     Nacionales
    -1.27
    Gew
    -1.25
    Katalog
    -1.23
    Tanzania
    -1.21
     Secretario
    -1.20
    ⢺
    -1.20
     TOO
    -1.19
    strukt
    -1.18
     Autorizaciones
    -1.18
    POSITIVE LOGITS
     that
    1.58
     from
    1.50
    ii
    1.45
    u
    1.44
     our
    1.42
     They
    1.42
    li
    1.37
    to
    1.37
    ٍ
    1.34
    a
    1.31
    Activations Density 0.057%

    No Known Activations