© 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. 15099
    Prev
    Next
    INDEX
    Explanations

    buy my

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

    The main thing this neuron does is detect code declaration keywords (like class, struct, enum, protocol, func, init) in source 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
    ädig
    -0.93
     erit
    -0.92
    strokeStyle
    -0.92
    Kode
    -0.88
     kani
    -0.86
     soldat
    -0.86
     kome
    -0.85
    asz
    -0.84
     Kode
    -0.82
     viet
    -0.82
    POSITIVE LOGITS
     their
    0.95
     jų
    0.90
     many
    0.87
    gués
    0.84
     precisar
    0.82
    banez
    0.82
    zzet
    0.82
     sintética
    0.82
     dispuesto
    0.81
    ようと
    0.81
    Activations Density 0.003%

    No Known Activations