© 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. 10-GEMMASCOPE-RES-131K
    7. 44722
    Prev
    Next
    INDEX
    Explanations

    de do

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

    The neuron fires on the very first “real” (non‐function) word of a new sentence or clause.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_10/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
     interruptor
    -1.88
     emphasizes
    -1.85
    <bos>
    -1.77
     elevador
    -1.73
     tabaco
    -1.69
     SALUD
    -1.67
    珅
    -1.66
     atendido
    -1.66
     piscinas
    -1.65
     reservar
    -1.60
    POSITIVE LOGITS
     как
    2.25
     be
    1.88
    in
    1.88
    ly
    1.86
    ll
    1.84
    mo
    1.83
     you
    1.77
    e
    1.71
     для
    1.70
     ๆ
    1.70
    Activations Density 0.009%

    No Known Activations