© 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. 99548
    Prev
    Next
    INDEX
    Explanations

    government, state, or authority nouns after "the"

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

    The neuron fires on numeric expressions—integers, decimals, percentages, years, and other standalone numbers.

    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
    ܆
    -1.84
     transparentes
    -1.78
     tristes
    -1.77
     aient
    -1.66
     lindos
    -1.66
     secos
    -1.64
    ();*/
    -1.62
     brillantes
    -1.61
    筴
    -1.59
     graisse
    -1.58
    POSITIVE LOGITS
     of
    3.09
     both
    1.97
     it
    1.80
     you
    1.76
     government
    1.66
     by
    1.63
     found
    1.61
     members
    1.59
    ↵↵
    1.54
     authorities
    1.53
    Activations Density 0.109%

    No Known Activations