© 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. 34-GEMMASCOPE-RES-131K
    7. 60920
    Prev
    Next
    INDEX
    Explanations

    association, experimental, and Indonesian

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

    The neuron strongly activates on comma tokens, effectively identifying commas as clause or list separators.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_34/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
     språk
    -1.16
     forklar
    -1.15
     inalámbrica
    -1.13
     inalámbrico
    -1.11
    owanie
    -1.07
     behandling
    -1.05
     remercier
    -1.05
     riktig
    -1.05
     henvis
    -1.02
    jalá
    -1.02
    POSITIVE LOGITS
     which
    2.28
     and
    2.11
     so
    1.69
     but
    1.52
     it
    1.30
    such
    1.27
     no
    1.25
    which
    1.21
     not
    1.17
     such
    1.17
    Activations Density 0.001%

    No Known Activations