© 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. 10-GEMMASCOPE-RES-131K
    7. 73757
    Prev
    Next
    INDEX
    Explanations

    numbers and specific words

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

    The neuron specializes in detecting and highlighting numeric quantities (digits and numerical expressions) in the text.

    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
     These
    -1.66
     They
    -1.52
     also
    -1.51
     این
    -1.46
    ↵↵↵
    -1.45
     también
    -1.45
    ⅞
    -1.41
     You
    -1.38
    ⅝
    -1.38
    ↵↵↵↵↵↵↵
    -1.36
    POSITIVE LOGITS
    _
    1.43
     zaji
    1.37
    вають
    1.35
     cauza
    1.33
     sytuacji
    1.32
    什么呢
    1.28
     putem
    1.23
    してしまう
    1.22
    加拿
    1.20
     ogranic
    1.20
    Activations Density 0.299%

    No Known Activations