© 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. 15396
    Prev
    Next
    INDEX
    Explanations

    numbers followed by patients or percentages

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

    The neuron detects numerical data tokens (counts, percentages, or measurements) in the document.

    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
     even
    -1.16
     ontwikkel
    -1.16
     allerede
    -1.10
     about
    -1.09
    mässig
    -1.05
     verbeter
    -1.04
     toepassing
    -1.03
    雖
    -1.03
     aanz
    -1.02
    謠
    -1.02
    POSITIVE LOGITS
     тис
    1.26
     cases
    1.22
    owns
    1.21
    Mostly
    1.15
    }\
    1.15
     who
    1.13
    weren
    1.12
     gemis
    1.12
     ventes
    1.09
     waren
    1.09
    Activations Density 0.045%

    No Known Activations