© 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. 65919
    Prev
    Next
    INDEX
    Explanations

    descriptions and disclaimers

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

    This neuron fires on salient content words used in product- or promotion-style listings (feature descriptions, marketing headings, and branded 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_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
     these
    -1.12
     when
    -1.07
     by
    -0.97
    mbps
    -0.94
    competition
    -0.91
     pama
    -0.90
    direction
    -0.88
    כונות
    -0.87
     magis
    -0.87
    々と
    -0.86
    POSITIVE LOGITS
    ׇ
    0.98
    etton
    0.94
     unim
    0.94
     ProtoMessage
    0.93
     antitrust
    0.91
     представляет
    0.90
     indigent
    0.90
     deinem
    0.90
     outskirts
    0.90
     abolished
    0.89
    Activations Density 0.001%

    No Known Activations