© 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. 22-GEMMASCOPE-RES-131K
    7. 34212
    Prev
    Next
    INDEX
    Explanations

    this is why

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

    This neuron fires on presentational “to be” statements (e.g. “this is…,” “that is…,” “these are…”) and in particular highlights the copula and the noun‐phrase or clause that follows.

    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
    haled
    -1.08
    鹉
    -1.02
     şun
    -1.01
    因为它
    -1.01
     cualquiera
    -0.97
    autique
    -0.97
    Dabei
    -0.96
    яза
    -0.96
    mmons
    -0.95
     aho
    -0.94
    POSITIVE LOGITS
     why
    1.44
     how
    1.37
     our
    1.32
     what
    1.26
     then
    1.13
     of
    1.12
     basis
    1.08
     now
    1.07
     currently
    1.05
     που
    1.03
    Activations Density 0.018%

    No Known Activations