© 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. 67258
    Prev
    Next
    INDEX
    Explanations

    research study or academic study

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

    the word "study" when referring to scientific or academic research.

    oai_token-act-pair · claude-3-7-sonnet-20250219Triggered by @neilrathi

    This neuron detects occurrences of the word “study” (especially as a section or title heading).

    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
    '
    -3.47
    теры
    -2.63
    S
    -2.53
     was
    -2.34
    H
    -2.31
    ing
    -2.30
    O
    -2.19
    r
    -2.19
     '
    -2.16
     musí
    -2.13
    POSITIVE LOGITS
    ेमाल
    2.53
     bellissimo
    2.45
     methods
    2.42
    You
    2.42
    ┎
    2.39
    zdroj
    2.31
     Hinsicht
    2.30
     Geschä
    2.30
     Preisen
    2.30
     ingenious
    2.28
    Activations Density 0.115%

    No Known Activations