© 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. 74717
    Prev
    Next
    INDEX
    Explanations

    days or seconds durations

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

    The neuron activates on numeric tokens—i.e. it spots numbers (quantities, durations, percentages, etc.) 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
    1
    -1.80
     pôr
    -1.45
    确实是
    -1.44
     几
    -1.44
     esquecer
    -1.43
    呦
    -1.42
     ocorrer
    -1.41
     kind
    -1.41
    every
    -1.38
     trazer
    -1.36
    POSITIVE LOGITS
     or
    1.87
     but
    1.63
     ''
    1.54
     unparalleled
    1.53
     fortfarande
    1.49
    श्लेषण
    1.46
      “
    1.45
    倂
    1.42
    ÓG
    1.37
    ére
    1.36
    Activations Density 0.041%

    No Known Activations