© 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. 100063
    Prev
    Next
    INDEX
    Explanations

    who followed by a verb phrase

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

    The neuron detects relative pronouns (e.g., “who,” “which,” “where”) that introduce subordinate clauses.

    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
     since
    -2.11
     this
    -2.09
     like
    -2.08
     although
    -2.05
     because
    -1.89
     from
    -1.88
    Although
    -1.86
     despite
    -1.84
    This
    -1.77
     bringing
    -1.76
    POSITIVE LOGITS
     pertinente
    2.00
     เซ
    1.84
     rése
    1.79
     différents
    1.73
    鴴
    1.71
     boisson
    1.70
     derni
    1.66
     mâle
    1.64
    鶫
    1.64
    yep
    1.63
    Activations Density 0.097%

    No Known Activations