© 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. 21374
    Prev
    Next
    INDEX
    Explanations

    BC or BCE dates

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

    The neuron activates on numeric tokens marking dates, years, centuries, or similar numeric references in historical or record-keeping contexts.

    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
     for
    -2.17
     throughout
    -1.69
     at
    -1.59
     currently
    -1.59
     now
    -1.57
    !
    -1.56
    !"
    -1.52
    ・「
    -1.50
    KS
    -1.49
     on
    -1.46
    POSITIVE LOGITS
     selben
    1.47
     toalha
    1.44
     vorigen
    1.44
     tiefe
    1.41
     Absicht
    1.41
     consiguiente
    1.40
    它
    1.38
    2
    1.37
    時間で
    1.36
     appliquer
    1.35
    Activations Density 0.011%

    No Known Activations