© 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 EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 128868
    Prev
    Next
    INDEX
    Explanations

    The neuron strongly responds to list-style markers—especially numeric bullet points and the colons that introduce or separate list items.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408

    application requirements

    np_acts-logits-general · gemini-2.5-flash-lite
    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
    certainly
    -1.20
     exclusivement
    -1.11
    Спасылкі
    -1.09
    FORCEMENT
    -1.06
    ↩
    -1.05
     raí
    -1.02
    rehension
    -0.98
    retung
    -0.98
     hanya
    -0.97
    delle
    -0.97
    POSITIVE LOGITS
     required
    1.66
     necessary
    1.52
     sufficient
    1.45
     mandatory
    1.23
     appropriate
    1.21
     completed
    1.19
     any
    1.17
     additional
    1.14
     adequate
    1.11
     будь
    1.09
    Activations Density 0.017%

    No Known Activations