© 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. 34-GEMMASCOPE-RES-131K
    7. 65874
    Prev
    Next
    INDEX
    Explanations

    he him who

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

    The neuron fires strongly on structural markers of scripture passages—verse numbers, colons/semicolons, quotation marks, and the connective words (“And,” “But,” “He,” etc.) that signal verse boundaries.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_34/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
     $(".
    -0.84
     mọi
    -0.83
     acestui
    -0.82
    divarius
    -0.82
    其他人
    -0.82
    ostic
    -0.81
     swoich
    -0.81
    ugian
    -0.79
     aceste
    -0.78
    Arbitrage
    -0.78
    POSITIVE LOGITS
     celui
    2.72
     who
    2.13
     тот
    2.08
     he
    2.03
     aquele
    1.96
     Celui
    1.77
     him
    1.72
     который
    1.66
     aquel
    1.66
     quien
    1.66
    Activations Density 0.026%

    No Known Activations