© 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. 58561
    Prev
    Next
    INDEX
    Explanations

    possessive pronouns followed by family members

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

    The neuron is sensitive to common function words—especially short high-frequency words like articles (“a,” “of”) and possessive pronouns (“my,” “his,” “her,” “our”).

    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
     nieces
    -0.85
     старости
    -0.82
     メニュー
    -0.81
     estampa
    -0.81
    سبة
    -0.79
     BrowserModule
    -0.76
     Джей
    -0.75
     Carmichael
    -0.75
    فایل
    -0.73
    Zdrav
    -0.73
    POSITIVE LOGITS
     infant
    2.34
    infant
    1.66
     grandparents
    1.50
     Infant
    1.48
     baby
    1.47
     grandson
    1.46
    Infant
    1.41
     parents
    1.35
    Baby
    1.20
     son
    1.19
    Activations Density 0.002%

    No Known Activations