© 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. 84265
    Prev
    Next
    INDEX
    Explanations

    Temporal patterns, Q: prefixes

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

    The neuron fires on numeric tokens containing a decimal point (i.e. floating‐point numbers).

    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
     and
    -2.13
     to
    -2.08
     simplemente
    -1.54
     empuj
    -1.52
     posteriormente
    -1.51
     và
    -1.45
    cidade
    -1.45
     kteří
    -1.44
     sencillas
    -1.43
     Gründung
    -1.38
    POSITIVE LOGITS
    媖
    2.16
    The
    1.75
     participado
    1.73
     yetu
    1.71
     veta
    1.70
     fusil
    1.63
     nôtre
    1.61
     negó
    1.60
     perak
    1.60
     refirió
    1.59
    Activations Density 0.000%

    No Known Activations