© 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. 99093
    Prev
    Next
    INDEX
    Explanations

    specializing in services

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

    The neuron activates on occurrences of the word “specialize” (and its spelling variants like “specialise,” “specializing,” “specialising”) typically when followed by “in.”

    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
    endimiento
    -1.70
    '
    -1.64
     on
    -1.60
    :
    -1.59
     in
    -1.59
     argentinos
    -1.57
     peruano
    -1.57
    pirate
    -1.48
    护士
    -1.46
     other
    -1.45
    POSITIVE LOGITS
    </h2>
    2.05
    rt
    1.90
     mesmas
    1.88
    including
    1.77
    m
    1.75
    ts
    1.73
     appena
    1.72
     ultimi
    1.70
    
    1.65
    chymal
    1.63
    Activations Density 0.013%

    No Known Activations