© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW
    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    2. Gemma-2-27B
    3. 10-GEMMASCOPE-RES-131K
    4. 12552
    Prev
    Next
    INDEX
    Explanations

    possessive pronouns followed by nouns

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

    The neuron activates on function words and small adjectives that convey possession, quantities, or order—e.g. possessive pronouns (its, our, their), numerals and ordinals (two, first), and modifiers like “new.”

    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
     anderen
    -1.39
     to
    -1.34
     Entsche
    -1.33
     ۱۳۹
    -1.29
     为什么
    -1.26
     normalen
    -1.26
    economia
    -1.25
     frattempo
    -1.24
    柚子
    -1.23
     フラット
    -1.23
    POSITIVE LOGITS
     who
    3.00
     whom
    1.73
    };
    1.58
     člov
    1.55
    THERE
    1.34
     a
    1.33
     there
    1.32
    )});
    1.32
     capable
    1.30
     as
    1.29
    Activations Density 0.121%

    No Known Activations