© 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. 40300
    Prev
    Next
    INDEX
    Explanations

    [noun] of [noun/article]

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

    The neuron flags semantically “heavy” content words—nouns, proper names, technical terms—rather than common function words.

    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
    -
    -1.77
    ,
    -1.22
    =
    -1.19
    —
    -1.17
    //
    -1.16
    ·
    -1.13
     -
    -1.10
    さん
    -1.07
     صفحه
    -1.05
    Ч
    -1.05
    POSITIVE LOGITS
     and
    1.30
     率
    1.22
     姆
    1.17
     sabiduría
    1.13
    殚
    1.10
    訌
    1.07
     valoraciones
    1.06
     zowel
    1.06
    spodnie
    1.04
     tantôt
    1.04
    Activations Density 0.103%

    No Known Activations