© 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. 130912
    Prev
    Next
    INDEX
    Explanations

    separators and dividers

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

    The neuron activates on tokens that denote visual separators or horizontal dividers (e.g. `<hr>` tags, Divider component names, and related style/type mentions).

    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
    {
    -1.63
    atalos
    -1.39
    li
    -1.37
     PUSTAKA
    -1.33
    平常
    -1.28
     Wird
    -1.28
    ou
    -1.24
    =
    -1.24
    Materiál
    -1.24
    Ꮉ
    -1.23
    POSITIVE LOGITS
    紲
    1.71
    蚰
    1.64
    ᜀ
    1.63
     使っ
    1.57
    姽
    1.55
     Ministro
    1.55
    鉏
    1.54
     extrait
    1.49
    vocations
    1.47
    ︐
    1.46
    Activations Density 0.029%

    No Known Activations