© 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. 22-GEMMASCOPE-RES-131K
    7. 95642
    Prev
    Next
    INDEX
    Explanations

    specifying the main aspect

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

    The neuron detects adverbs that mark something as primary or predominant (e.g. “mainly,” “primarily,” “principally,” “predominantly”).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_22/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
     principalement
    -1.29
    Mainly
    -1.17
     soprattutto
    -1.07
     qualsiasi
    -1.06
     belangrijkste
    -1.05
     amortigu
    -1.05
     arī
    -1.03
    mainly
    -1.00
    ρας
    -0.99
    Oph
    -0.98
    POSITIVE LOGITS
     not
    1.26
     While
    1.16
     Although
    1.15
     will
    1.14
    although
    1.09
     on
    1.00
    ＾＾
    0.99
    Although
    0.99
     Though
    0.97
    While
    0.96
    Activations Density 0.111%

    No Known Activations