© 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
    3. Gemma-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 36083
    Prev
    Next
    INDEX
    Explanations

    configuration files, methods, and naming conventions

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

    The neuron responds to long, specialized multi-morpheme technical terms (e.g. scientific, programming, or domain-specific jargon).

    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
    官员
    -0.96
    mariage
    -0.93
     ftate
    -0.90
     ઉ
    -0.90
    ख़
    -0.87
    fuente
    -0.86
     any
    -0.85
    meu
    -0.82
    贵族
    -0.82
    bánh
    -0.82
    POSITIVE LOGITS
    örté
    1.01
     entrer
    0.96
    だけが
    0.94
     lasci
    0.91
    monial
    0.91
     છે
    0.90
    anmoins
    0.89
    特徴
    0.88
     computation
    0.88
     otros
    0.85
    Activations Density 0.095%

    No Known Activations