© 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. 65694
    Prev
    Next
    INDEX
    Explanations

    code structure

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

    The neuron spikes on numeric literal tokens (and related delimiters like “/”), effectively detecting numbers in code.

    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
     A
    -1.49
     from
    -1.48
     S
    -1.47
     '
    -1.46
     G
    -1.45
     “
    -1.45
     B
    -1.34
     er
    -1.31
     H
    -1.31
     P
    -1.30
    POSITIVE LOGITS
     anormal
    1.60
     applau
    1.57
     desinf
    1.56
     guir
    1.54
     monstru
    1.52
     peluche
    1.52
     conster
    1.49
     turista
    1.48
     pijama
    1.48
     tarta
    1.47
    Activations Density 0.022%

    No Known Activations