© 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. 35643
    Prev
    Next
    INDEX
    Explanations

    Black Lives, thundercloud

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

    The neuron is primarily triggered by numeric tokens or number-like sequences (e.g., digits, decimals, and numbered citations).

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

    LaTeX mathematical notation and formatting symbols.

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

    markup and typesetting syntax elements (e.g., LaTeX math notation, Markdown formatting, HTML tags/anchors) indicating structured or technical formatting in the text.

    oai_token-act-pair · gpt-5Triggered 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
    Ề
    -0.88
    onews
    -0.84
    ėjimas
    -0.83
     besoin
    -0.82
    nine
    -0.82
     นิ้ว
    -0.80
    łoni
    -0.80
     мені
    -0.78
    letion
    -0.78
    něte
    -0.77
    POSITIVE LOGITS
     allait
    0.97
     такими
    0.92
     these
    0.88
    fycat
    0.88
    Conception
    0.84
     both
    0.84
     написал
    0.83
     entrambi
    0.82
    Overseas
    0.81
    énu
    0.81
    Activations Density 0.072%

    No Known Activations