© 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. 34582
    Prev
    Next
    INDEX
    Explanations

    structural separators

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

    The neuron fires on long runs of plus signs, dashes, or similar characters that serve as ASCII/markup table or section dividers.

    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
     by
    -1.05
     to
    -0.99
     berbeda
    -0.98
     Fácil
    -0.94
    vironment
    -0.92
    氛
    -0.92
     Lokales
    -0.92
    尬
    -0.89
     entweder
    -0.89
     lhe
    -0.88
    POSITIVE LOGITS
     cómodo
    1.23
    róleo
    1.17
     följande
    1.13
    ]$.
    1.10
     trabalha
    1.06
     hırka
    1.03
     rých
    1.02
     destes
    1.01
     vannak
    1.00
    ванні
    1.00
    Activations Density 0.002%

    No Known Activations