© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 74761
    Prev
    Next
    INDEX
    Explanations

    extremely + adjective/noun

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

    This neuron detects degree-intensifying adverbs (e.g. “extremely,” “quite,” “considerable”) that modify adjectives.

    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
    螅
    -2.23
    焌
    -2.08
    了
    -2.05
    嫆
    -2.03
    皞
    -1.98
    謌
    -1.98
    渽
    -1.97
    荜
    -1.94
    ⠲
    -1.92
    磲
    -1.91
    POSITIVE LOGITS
    ↵
    2.70
    1
    2.34
     and
    2.30
    '
    2.14
    3
    2.11
    /
    1.91
    ),
    1.88
    </td>
    1.79
    —
    1.78
     or
    1.75
    Activations Density 0.003%

    No Known Activations