© 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. 72768
    Prev
    Next
    INDEX
    Explanations

    in*studiesquantitated

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

    The neuron strongly activates on numeric tokens (digits or 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_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
    H
    -1.11
    S
    -1.11
     ’
    -1.09
    T
    -1.08
    A
    -1.07
    V
    -1.07
     from
    -1.05
    K
    -1.05
    O
    -1.05
    mmä
    -1.05
    POSITIVE LOGITS
     eſt
    1.44
    鼕
    1.29
     باخ
    1.19
     apparaître
    1.18
     seksi
    1.16
    躄
    1.16
     fú
    1.15
    シャレ
    1.15
    ꨄ
    1.15
    腘
    1.13
    Activations Density 0.097%

    No Known Activations