© 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. 22-GEMMASCOPE-RES-131K
    7. 72586
    Prev
    Next
    INDEX
    Explanations

    author, funded, originally, guest post

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

    The neuron fires on numeric tokens—digits, years, dates, and other number-like sequences.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_22/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
    -1.02
     to
    -0.98
    ～～
    -0.93
    e
    -0.87
    ？！
    -0.86
     it
    -0.84
    ｅ
    -0.84
    ！！！！
    -0.83
    憩
    -0.81
     firft
    -0.80
    POSITIVE LOGITS
     trám
    1.17
     voltak
    1.09
    をします
    1.07
     tapasztal
    1.05
     típus
    1.03
     család
    1.03
     stratég
    1.03
     увеличить
    1.00
     apporte
    1.00
     pól
    1.00
    Activations Density 0.357%

    No Known Activations