© 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. 34-GEMMASCOPE-RES-131K
    7. 57620
    Prev
    Next
    INDEX
    Explanations

    section headers or markers

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

    The neuron fires on structural formatting elements—things like section‐heading markers (e.g. “##”), code/LaTeX delimiters (backslashes and braces), and numeric identifiers (years, enumeration numbers) rather than on ordinary words.

    oai_token-act-pair · o4-miniTriggered 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
     други
    -1.05
     länder
    -1.04
     outros
    -1.02
    Тогда
    -1.01
     other
    -1.01
     tertarik
    -0.96
    Размеры
    -0.96
     senin
    -0.95
    Andere
    -0.93
    Lain
    -0.93
    POSITIVE LOGITS
    ,
    1.11
    \#
    0.98
     such
    0.93
    說
    0.93
    ;#
    0.93
    gameState
    0.93
    osphere
    0.92
    K
    0.92
    側
    0.91
    ower
    0.90
    Activations Density 0.117%

    No Known Activations