© 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. 90074
    Prev
    Next
    INDEX
    Explanations

    css reset or normalize

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

    This neuron activates on long runs of dash or underline characters (e.g. “----------------”) used as visual section dividers in code.

    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
    ͎
    -0.82
     lacrosse
    -0.82
    ghen
    -0.78
    ละ
    -0.75
    ]$
    -0.74
    ʾ
    -0.73
    vinfos
    -0.73
     ходу
    -0.73
    സ
    -0.71
    Kotlin
    -0.70
    POSITIVE LOGITS
     reset
    1.39
    reset
    1.38
     Reset
    1.25
    RESET
    1.17
    Reset
    1.14
     RESET
    1.09
    normalize
    1.08
     resets
    1.05
     Normalize
    0.96
     normalize
    0.94
    Activations Density 0.166%

    No Known Activations