© 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. 22475
    Prev
    Next
    INDEX
    Explanations

    numbers and times

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

    The neuron looks for the largest number in a comma-separated numeric sequence.

    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
     of
    -2.14
     that
    -1.77
     a
    -1.58
     our
    -1.41
     ofrecen
    -1.37
     puteți
    -1.34
     also
    -1.32
    !!!!!!!
    -1.32
    してない
    -1.31
    !!!!!!
    -1.30
    POSITIVE LOGITS
     söyl
    1.48
    Materiaal
    1.21
    菝
    1.20
     عشر
    1.20
    уна
    1.17
    的名字
    1.16
    Inver
    1.16
    简单的
    1.15
     primero
    1.15
     Garantie
    1.13
    Activations Density 0.025%

    No Known Activations