© 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. 113107
    Prev
    Next
    INDEX
    Explanations

    the second or latter

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

    This neuron selectively activates on numeric values in the text—especially decimal‐formatted measurements and counts.

    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
     unserem
    -1.35
    THEW
    -1.31
    complish
    -1.30
     cantantes
    -1.29
     нашем
    -1.29
     unserer
    -1.28
     traité
    -1.27
     offrant
    -1.26
    🧶
    -1.23
     BECAUSE
    -1.22
    POSITIVE LOGITS
     —
    1.41
     some
    1.35
     they
    1.30
    n
    1.28
     a
    1.27
     what
    1.23
     could
    1.20
    ⇘
    1.20
     •
    1.19
    如何在
    1.19
    Activations Density 0.045%

    No Known Activations