© 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. 22-GEMMASCOPE-RES-131K
    7. 68735
    Prev
    Next
    INDEX
    Explanations

    nested list structures

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

    The neuron fires on tokens that name structural elements—things like HTML/XML tags (e.g. “<li>” and “</”) or programming class/constructor identifiers (e.g. “Calendar,” “Day,” “Event,” “Attendee”).

    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
    どもの
    -0.88
     katk
    -0.86
     for
    -0.86
    Inches
    -0.84
     feel
    -0.82
     all
    -0.81
    孺
    -0.81
    같이
    -0.78
     as
    -0.76
     val
    -0.76
    POSITIVE LOGITS
     accompanied
    0.93
     разом
    0.87
    zieli
    0.85
     levou
    0.85
     leads
    0.85
     сопрово
    0.84
    værk
    0.84
     feminina
    0.83
    ById
    0.82
    衬
    0.82
    Activations Density 0.108%

    No Known Activations