© 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. 34-GEMMASCOPE-RES-131K
    7. 3137
    Prev
    Next
    INDEX
    Explanations

    circuitry

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

    This neuron fires most strongly on distinctive, content‐bearing tokens—especially uncommon or domain‐specific identifiers, technical terms, and punctuation 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_34/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
     souri
    -0.81
    tekend
    -0.81
    丫头
    -0.80
    Backbone
    -0.80
    TintMode
    -0.80
     komfor
    -0.79
    Koko
    -0.78
    åter
    -0.77
     Backbone
    -0.77
     flexibilidad
    -0.77
    POSITIVE LOGITS
     size
    1.22
    size
    1.03
     has
    0.87
     device
    0.86
    室外
    0.84
     bend
    0.84
    device
    0.83
     H
    0.82
     *
    0.81
     Re
    0.81
    Activations Density 0.003%

    No Known Activations