© 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
    3. Gemma-2-27B
    5. 34-GEMMASCOPE-RES-131K
    7. 118100
    Prev
    Next
    INDEX
    Explanations

    possessive names and terms

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

    The neuron activates on nouns referring to people or groups—especially proper names and collective references (e.g. “everyone,” “Boris,” “Johnson,” “people,” “Grandma,” “Obama”).

    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
     этот
    -0.83
     leurs
    -0.82
    bucha
    -0.80
    はこの
    -0.79
    はその
    -0.79
     acts
    -0.79
     этих
    -0.77
    镳
    -0.77
    憾
    -0.77
     их
    -0.76
    POSITIVE LOGITS
     theirs
    1.92
     hers
    1.73
     yours
    1.59
     ours
    1.32
    yours
    1.30
    irs
    1.25
     Yours
    1.20
     own
    1.09
    Yours
    1.09
     ones
    1.05
    Activations Density 0.035%

    No Known Activations