© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 14281
    Prev
    Next
    INDEX
    Explanations

    sumptuous size and primary

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

    The neuron flags the “.<digit>” substring in decimal numbers (i.e. the decimal point plus the following digit).

    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
     because
    -1.38
     just
    -1.34
     while
    -1.28
     three
    -1.27
     will
    -1.27
     eight
    -1.22
     one
    -1.20
     had
    -1.18
     in
    -1.15
     regarding
    -1.15
    POSITIVE LOGITS
    hare
    1.11
     necesarios
    1.10
    gebung
    1.09
     Dinas
    1.08
    liegenden
    1.05
     incrível
    1.05
    !!!!!!!
    1.02
     таких
    1.02
     Aunque
    1.02
     Tienes
    1.00
    Activations Density 0.001%

    No Known Activations