© 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. 22-GEMMASCOPE-RES-131K
    7. 104603
    Prev
    Next
    INDEX
    Explanations

    assurance for you

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

    The neuron flags words used in spoken or quoted dialogue—especially those appearing in questions (e.g. “is,” “that,” “enough,” “prepared,” “see”).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_22/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
    This
    -1.34
     Aufführung
    -1.20
    €“
    -1.10
     TextEditing
    -1.09
     diesen
    -1.08
     Pesquisa
    -1.08
     Mehrheit
    -1.06
    ###############
    -1.05
     tämän
    -1.05
    =
    -1.04
    POSITIVE LOGITS
     you
    1.17
     tivesse
    1.10
     we
    1.10
    さんも
    1.03
    ก็
    1.03
     then
    1.01
    BATTLE
    1.00
     tuyo
    0.99
    我们
    0.97
    tificates
    0.97
    Activations Density 0.003%

    No Known Activations