© 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. 128103
    Prev
    Next
    INDEX
    Explanations

    The neuron strongly activates on numeric values (including integers and decimals) that denote measurements or model/item identifiers.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408

    surfaces and structures

    np_acts-logits-general · gemini-2.5-flash-lite
    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
    了他的
    -1.04
     obliquely
    -0.90
    для
    -0.89
    usammen
    -0.87
     floppy
    -0.87
     connaissances
    -0.84
     ओवर
    -0.83
     feront
    -0.82
     normes
    -0.81
     CFD
    -0.80
    POSITIVE LOGITS
     surface
    1.31
     surfaces
    1.25
     wall
    1.07
    上的
    1.03
     superficies
    1.01
     cantos
    0.99
     SURFACE
    0.98
    urface
    0.94
     waarop
    0.94
     beberapa
    0.91
    Activations Density 0.254%

    No Known Activations