© 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. 85414
    Prev
    Next
    INDEX
    Explanations

    physical impact or struggle

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

    The neuron detects strong, dynamic action verbs—especially vivid or metaphorical transitive verbs describing forceful events.

    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
     when
    -1.85
    levis
    -1.34
     cuando
    -1.26
     including
    -1.25
    合影
    -1.24
     while
    -1.23
    ungkinkan
    -1.23
    LLYWOOD
    -1.22
     когда
    -1.20
    ednesdays
    -1.20
    POSITIVE LOGITS
    dropIfExists
    1.31
    ษ
    1.30
     we
    1.20
    ateurs
    1.20
    haft
    1.18
    setData
    1.17
     Affaires
    1.17
    ext
    1.16
    ul
    1.13
     adatta
    1.13
    Activations Density 0.251%

    No Known Activations