© 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. 99394
    Prev
    Next
    INDEX
    Explanations

    followed by prepositions or punctuation

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

    The neuron fires on central content words—especially longer, domain‐specific nouns and adjectives (e.g. “compliance,” “performance,” “exchange,” “average,” “support”) that carry the main informational load of a sentence.

    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
     who
    -1.63
     there
    -1.55
    </strong>
    -1.45
     you
    -1.43
     aquellos
    -1.32
    ného
    -1.30
     rutas
    -1.28
     olvidar
    -1.27
     divertido
    -1.27
     should
    -1.26
    POSITIVE LOGITS
     its
    1.67
    ようになります
    1.41
     größe
    1.41
     fortfarande
    1.40
     olika
    1.32
    Its
    1.31
     grosso
    1.30
    roughly
    1.26
     correctes
    1.22
     BIEN
    1.21
    Activations Density 0.252%

    No Known Activations