© 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
    2. Gemma-2-27B
    3. 10-GEMMASCOPE-RES-131K
    4. 87301
    Prev
    Next
    INDEX
    Explanations

    especially, particularly

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

    The neuron fires on emphasizing adverbs—especially on occurrences of “especially” (and its variant “specially”) indicating emphasis.

    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
     a
    -1.77
     &
    -1.63
    自体は
    -1.61
     "
    -1.58
    入れた
    -1.57
     '
    -1.51
     etc
    -1.50
    </i>
    -1.47
    又は
    -1.46
    </th>
    -1.45
    POSITIVE LOGITS
     those
    2.30
     when
    2.11
    when
    1.88
     dotyczy
    1.81
     когда
    1.80
     sinistro
    1.77
     Bagian
    1.74
     the
    1.73
     takich
    1.72
     leggero
    1.71
    Activations Density 0.048%

    No Known Activations