© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it
    Neuronpedia
    Jacobian LensNEW
    Natural Language
    Autoencoders
    MAY 26
    Assistant AxisJAN 26Circuit TracerAPR 26SteerSAE EvalsExportsinterp-engineNEW Community BlogPrivacy & TermsContact
    1. Home
    2. Gemma-2-27B
    3. 10-GEMMASCOPE-RES-131K
    4. 58140
    Prev
    Next
    INDEX
    Explanations

    family members and pets

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

    The neuron is strongly activated by baby-talk or child-oriented words and diminutives (e.g. “mommy,” “boo-boos,” “kitty,” “puppy,” “tummies”), i.e. vocabulary used in child-directed speech.

    oai_token-act-pair · o4-mini
    Triggered 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
    Negative Logits
     by
    -1.63
    皷
    -1.55
    如果
    -1.53
    需要
    -1.52
     подоб
    -1.51
    但
    -1.50
     Giappone
    -1.40
     is
    -1.39
     comunidade
    -1.37
    tualmente
    -1.37
    POSITIVE LOGITS
    𖣘
    1.59
    bosis
    1.54
     sorta
    1.52
    紆
    1.52
    々な
    1.47
    чные
    1.43
     embraces
    1.42
     imparts
    1.40
    купа
    1.38
    昔の
    1.38
    Activations Density 0.106%

    No Known Activations