© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 130443
    Prev
    Next
    INDEX
    Explanations

    mathematical notation and specific terms

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

    This neuron fires strongly on the initial subword pieces of longer, distinctive terms (e.g. drug names, technical or proper‐noun vocabulary).

    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
     but
    -1.13
    inį
    -1.12
    。『
    -1.11
    ATIVA
    -1.10
     dynamics
    -1.10
     玲
    -1.09
    ⎧
    -1.09
     only
    -1.09
     fald
    -1.09
    {~
    -1.08
    POSITIVE LOGITS
    íu
    1.21
    INGRED
    1.15
    Läh
    1.12
    Ingredi
    1.10
    0
    1.09
    Leírás
    1.06
    Andere
    1.02
    紆
    1.02
    Cuidado
    1.01
    توضی
    1.00
    Activations Density 0.851%

    No Known Activations