© 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
    2. Gemma-2-27B
    3. 22-GEMMASCOPE-RES-131K
    4. 28105
    Prev
    Next
    INDEX
    Explanations

    is a

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

    The neuron fires on copular “is a” definition phrases—i.e. occurrences of “is a [descriptor] [noun]” that introduce or define something.

    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
     eſt
    -1.49
     effectué
    -1.48
     satisfactor
    -1.48
     donnent
    -1.45
    Конечно
    -1.41
     ktorí
    -1.39
     kobieta
    -1.38
     viennent
    -1.37
     pomaga
    -1.37
     Eſ
    -1.36
    POSITIVE LOGITS
     has
    2.19
     on
    1.34
     also
    1.24
    มี
    1.18
     because
    1.18
     is
    1.16
     frequently
    1.14
     in
    1.11
    奬
    1.10
     outros
    1.09
    Activations Density 0.482%

    No Known Activations