© 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. 26322
    Prev
    Next
    INDEX
    Explanations

    both or neither

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

    The neuron fires on numeric tokens—literal numbers (and related comparison or arithmetic symbols) in code.

    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
     to
    -0.88
    vs
    -0.85
     kontra
    -0.85
     vs
    -0.85
     notizie
    -0.82
     Tomaten
    -0.81
     conteúdos
    -0.81
    lng
    -0.80
     ícones
    -0.80
    Чем
    -0.79
    POSITIVE LOGITS
     both
    1.81
     neither
    1.57
    Both
    1.38
     neutral
    1.27
     Both
    1.27
     two
    1.23
    Neither
    1.17
    neither
    1.17
     between
    1.12
     ambos
    1.11
    Activations Density 0.016%

    No Known Activations