© 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. 57735
    Prev
    Next
    INDEX
    Explanations

    son of, father of, mother of

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

    The neuron fires strongly on kinship terms—words denoting family relationships (e.g. “son,” “father,” “children,” “daughter”).

    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
    此
    -0.99
     what
    -0.97
     as
    -0.88
     wybrać
    -0.86
    เกี่ยวกับ
    -0.84
    挨拶
    -0.83
    붉
    -0.83
     znaleźć
    -0.82
    方案
    -0.81
     but
    -0.80
    POSITIVE LOGITS
     of
    3.48
     của
    1.50
     של
    1.34
     garan
    1.06
     favoriser
    1.01
    гласно
    1.00
     sensibili
    1.00
     assass
    1.00
    late
    0.99
     péché
    0.98
    Activations Density 0.015%

    No Known Activations