© 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. 10-GEMMASCOPE-RES-131K
    7. 48304
    Prev
    Next
    INDEX
    Explanations

    bolding, section headers, comments

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

    The main thing this neuron does is detect bold or strong emphasis markers (e.g. `<b>`, `</b>`, `<strong>`, `**`).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_10/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
    I
    -1.83
    ,”
    -1.75
    }
    -1.64
     soooo
    -1.63
     Removing
    -1.63
    ⎆
    -1.63
     sooo
    -1.62
     frequência
    -1.60
     essentiellement
    -1.59
    >
    -1.55
    POSITIVE LOGITS
     of
    1.83
     lagar
    1.62
     Budaya
    1.60
     Príncipe
    1.60
     damen
    1.57
     fön
    1.57
     perioden
    1.55
     befolk
    1.54
     kampen
    1.54
     ledig
    1.54
    Activations Density 0.012%

    No Known Activations