© 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. 72246
    Prev
    Next
    INDEX
    Explanations

    all and entire

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

    This neuron responds to numeric quantities—especially decimal numbers and other numerical measurements.

    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
     ordinary
    -1.24
     adequate
    -1.18
     conventional
    -1.16
    自己的
    -1.14
    During
    -1.10
     a
    -1.09
    Shortly
    -1.09
     customers
    -1.02
     an
    -1.02
     suitable
    -1.01
    POSITIVE LOGITS
     the
    1.27
     تمامی
    1.22
     twarzy
    1.22
     åren
    1.19
     tutte
    1.16
     wszyst
    1.14
     wszystkie
    1.05
     toate
    1.02
     włosów
    1.00
    addAll
    0.99
    Activations Density 0.049%

    No Known Activations