© 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. 10-GEMMASCOPE-RES-131K
    7. 67782
    Prev
    Next
    INDEX
    Explanations

    possessives followed by locations

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

    The neuron detects the possessive suffix “’s” attached to words.

    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
     can
    -1.38
     has
    -1.34
     about
    -1.31
     would
    -1.30
     not
    -1.27
     it
    -1.25
     Yorkers
    -1.25
     questions
    -1.21
    ništvo
    -1.20
     mengetahui
    -1.18
    POSITIVE LOGITS
    sockfd
    1.55
     แต่
    1.44
    楂
    1.34
    "^
    1.31
    回到了
    1.30
    籏
    1.27
    片的
    1.26
    mainAxisSize
    1.26
    theilung
    1.26
     αλλά
    1.26
    Activations Density 0.043%

    No Known Activations