© 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. 72625
    Prev
    Next
    INDEX
    Explanations

    arithmetic expressions

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

    The neuron is picking up numeric literals (standalone digits or numbers) in code or mathematical 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
    Ưu
    -1.42
    ologous
    -1.12
    archical
    -1.11
     hornear
    -1.09
    álise
    -1.05
    ּוֹ
    -1.05
     čern
    -1.03
    ῾
    -1.03
    Bộ
    -1.02
     język
    -1.02
    POSITIVE LOGITS
     also
    1.29
     ebenfalls
    1.19
     também
    1.16
    ^(@
    1.05
     ライ
    1.02
     sirup
    1.00
    Также
    0.98
    Gestures
    0.98
    }\}$
    0.98
     ภาพ
    0.97
    Activations Density 0.002%

    No Known Activations