© 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. 36174
    Prev
    Next
    INDEX
    Explanations

    ANNOUNCEMENT

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

    This neuron strongly activates on lines that mark out-document metadata or section boundaries (e.g. license headers, key delimiters, or other all-caps separators).

    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
    她的
    -1.01
     ayudará
    -0.98
     esportivo
    -0.96
    []=$
    -0.93
     espanhol
    -0.93
    lijken
    -0.93
    berspace
    -0.93
     oferece
    -0.93
     ocorre
    -0.92
     departments
    -0.91
    POSITIVE LOGITS
     and
    1.44
    ⇋
    1.06
    承认
    0.96
    paedia
    0.90
     acá
    0.88
    顽
    0.88
     waarschijnlijk
    0.87
     stij
    0.87
     diantaranya
    0.87
    zná
    0.87
    Activations Density 0.001%

    No Known Activations