© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 130423
    Prev
    Next
    INDEX
    Explanations

    The neuron fires on numeric literals (e.g. integer or floating‐point constants) in code.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408

    John Rousakis

    np_acts-logits-general · gemini-2.5-flash-lite
    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
     entweder
    -1.19
    หรือ
    -1.14
     alebo
    -1.13
     SIMILAR
    -1.09
     serupa
    -1.06
    重生
    -1.05
     یا
    -1.05
     либо
    -1.05
    Här
    -1.05
     SUPERIOR
    -1.01
    POSITIVE LOGITS
     even
    1.27
     несколь
    1.18
     multiple
    1.16
     no
    1.12
    𝑂
    1.11
     different
    1.09
     siquiera
    1.07
     unterstützt
    1.03
     nothing
    1.02
    一个小
    1.01
    Activations Density 0.005%

    No Known Activations