© 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. 10-GEMMASCOPE-RES-131K
    7. 85090
    Prev
    Next
    INDEX
    Explanations

    possessive determiner followed by noun

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

    The neuron activates on nouns (and related words) that label measurable outcomes or performance metrics (e.g. “performance,” “goal,” “lives,” “role,” “work”).

    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
     as
    -1.29
     motivadoras
    -1.28
     gouvernements
    -1.16
     れる
    -1.13
     Durchmesser
    -1.11
     Daarna
    -1.09
     باید
    -1.09
     fej
    -1.08
    čnega
    -1.06
     badkamer
    -1.05
    POSITIVE LOGITS
     to
    1.81
     RECOMEND
    1.50
    .《
    1.36
     gør
    1.26
     you
    1.23
    /*!
    1.20
     ['./
    1.19
    فتر
    1.19
     jugend
    1.18
    ;\\
    1.16
    Activations Density 0.035%

    No Known Activations