© 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. 34-GEMMASCOPE-RES-131K
    7. 57303
    Prev
    Next
    INDEX
    Explanations

    of a

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

    The neuron fires on very common mathematical “connecting” tokens—frequent articles, prepositions, and core domain words (e.g. “of,” “in,” “a,” “ring,” “functions,” “subspace,” “vanishing”).

    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
    Their
    -1.01
    

    -0.99
     разных
    -0.99
     их
    -0.97
    など
    -0.97
     Allí
    -0.97
     hierzu
    -0.94
     его
    -0.94
     उन्ह
    -0.94
     ihres
    -0.93
    POSITIVE LOGITS
     any
    1.70
     the
    1.55
     our
    1.55
     some
    1.32
     a
    1.29
     each
    1.26
     $\
    1.23
    voice
    1.18
     every
    1.14
     bahagia
    1.14
    Activations Density 0.094%

    No Known Activations