© 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. 99242
    Prev
    Next
    INDEX
    Explanations

    possess or owns something

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

    The neuron detects instances of “possess/possesses/possessed” (and similar “have/had/counts on”)—i.e. verbs indicating ownership or that something has a given attribute.

    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
     nying
    -1.62
     kampen
    -1.59
    </i>
    -1.57
     fysis
    -1.52
    aaaaaa
    -1.51
    ).”
    -1.46
     område
    -1.46
     verko
    -1.45
     précédentes
    -1.43
     iva
    -1.43
    POSITIVE LOGITS
     memiliki
    1.92
    还不是
    1.67
     ۵
    1.67
     einem
    1.63
     в
    1.63
    In
    1.62
     bietet
    1.59
     glittering
    1.58
     ۶
    1.56
     a
    1.55
    Activations Density 0.040%

    No Known Activations