© 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. 74009
    Prev
    Next
    INDEX
    Explanations

    mathematical expressions and names

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

    The neuron strongly activates on isolated numeric tokens (digits and numbers) 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
     lesquels
    -1.41
     injust
    -1.41
    鲠
    -1.39
     lesquelles
    -1.33
     preliminar
    -1.31
    땃
    -1.30
     assum
    -1.30
     applica
    -1.28
     ...,
    -1.27
     côtés
    -1.26
    POSITIVE LOGITS
    ?”
    1.45
    ++];
    1.41
    ’-
    1.40
    ,’
    1.37
    その
    1.32
     հղումներ
    1.30
     new
    1.28
     bocetos
    1.27
     hurricanes
    1.27
     knihy
    1.26
    Activations Density 0.032%

    No Known Activations