© 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. 90725
    Prev
    Next
    INDEX
    Explanations

    special characters and numbers

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

    The neuron fires strongly on section or block “header” lines—titles, headings, or top‐level declarations that mark the start of a new section.

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

    line breaks or newlines in formatted technical documents, particularly in code, legal text, and documentation.

    oai_token-act-pair · claude-4-5-sonnetTriggered by @jyhe0408

    document-structure elements such as headings, section titles, and other formal metadata markers.

    oai_token-act-pair · gpt-5Triggered 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
    ↵↵
    -1.46
     kitabı
    -0.81
    ädig
    -0.79
    紺
    -0.77
    动了
    -0.77
     môžete
    -0.76
     */
    -0.75
     můžete
    -0.75
    <eos>
    -0.75
    ቻ
    -0.74
    POSITIVE LOGITS
    ↵
    1.38
    genheiten
    0.82
    tour
    0.82
    atott
    0.80
     Ved
    0.79
     Iron
    0.79
    .
    0.79
    ecera
    0.78
     sige
    0.78
     engel
    0.77
    Activations Density 0.547%

    No Known Activations