© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 10659
    Prev
    Next
    INDEX
    Explanations

    options and classifications

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

    The neuron fires on compact, formal noun phrases—especially adjective‐noun combinations that name options, categories, or defined terms.

    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
     gitar
    -1.30
     ono
    -1.23
     assister
    -1.23
     monstre
    -1.22
     Modo
    -1.22
     ouverture
    -1.21
     ?
    -1.21
    icherheit
    -1.20
     Supremo
    -1.18
     conso
    -1.18
    POSITIVE LOGITS
     in
    2.28
     myös
    1.63
     include
    1.51
     a
    1.34
    躂
    1.29
    président
    1.29
    omg
    1.25
    opheles
    1.22
    œuf
    1.22
     anche
    1.22
    Activations Density 0.107%

    No Known Activations