© 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. 93276
    Prev
    Next
    INDEX
    Explanations

    @ section or assertion

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

    This neuron detects section headers and metadata field labels (like “section:”, “title”, “priority”, numeric section identifiers, XML namespaces, etc.) across various code and markup formats.

    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
     fictitious
    -0.77
    Dış
    -0.73
    ▎
    -0.73
     いい
    -0.72
    cepto
    -0.72
    深圳
    -0.71
    可以看到
    -0.71
    chenkt
    -0.71
    buena
    -0.71
    irma
    -0.70
    POSITIVE LOGITS
    права
    0.84
     subsection
    0.80
    dbg
    0.80
    InSection
    0.79
    subsection
    0.78
    говари
    0.77
    裘
    0.75
    seku
    0.75
    rella
    0.72
     recoger
    0.71
    Activations Density 0.027%

    No Known Activations