© 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. 84317
    Prev
    Next
    INDEX
    Explanations

    delimiters and separators

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

    This neuron activates on structural or formatting delimiters—tokens that mark section breaks or markup boundaries (e.g. ‘+++’, ‘---’, ‘:::’, `<tbody>`, `{}`, protocol markers).

    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
     from
    -2.00
    providedIn
    -1.76
     to
    -1.75
     it
    -1.70
     this
    -1.66
     even
    -1.63
     isInitialized
    -1.62
     wholly
    -1.61
     especially
    -1.59
     within
    -1.55
    POSITIVE LOGITS
    na
    1.94
    ni
    1.72
     bedeutend
    1.66
    he
    1.63
     ältere
    1.62
    ul
    1.55
    ものは
    1.55
    瞇
    1.51
    ur
    1.46
    ild
    1.46
    Activations Density 0.001%

    No Known Activations