© 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. 10-GEMMASCOPE-RES-131K
    7. 44490
    Prev
    Next
    INDEX
    Explanations

    absolute states or actions

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

    The neuron fires on strong degree‐intensifying adverbs expressing totality (e.g. “completely”).

    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
    ar
    -2.88
    謌
    -2.81
     of
    -2.64
    ية
    -2.59
     vielen
    -2.56
     meek
    -2.56
    蕷
    -2.45
    杁
    -2.45
    L
    -2.45
     dents
    -2.45
    POSITIVE LOGITS
     voordat
    2.83
    話は
    2.81
    Ꭳ
    2.80
     terwijl
    2.73
    there
    2.70
     兩
    2.66
     provides
    2.64
    蕯
    2.63
    様でした
    2.61
     lijkt
    2.59
    Activations Density 0.006%

    No Known Activations