© 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. 22382
    Prev
    Next
    INDEX
    Explanations

    action of

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

    The neuron detects occurrences of the preposition “of” in technical noun phrases (e.g. “pressure of,” “action of,” “generation of”).

    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
     MEXICO
    -1.49
     when
    -1.48
     SPAIN
    -1.45
    🧐
    -1.36
    !
    -1.34
     MATER
    -1.33
     udf
    -1.28
    lieved
    -1.28
     Shipment
    -1.28
    ---------
    -1.28
    POSITIVE LOGITS
     véritable
    1.82
     of
    1.70
    ”،
    1.66
     głównie
    1.61
     VERY
    1.57
     somewhat
    1.55
     établi
    1.55
    垍
    1.54
    iertos
    1.54
     Again
    1.54
    Activations Density 0.073%

    No Known Activations