© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it
    Neuronpedia
    Jacobian LensNEW
    Natural Language
    Autoencoders
    MAY 26    Assistant AxisJAN 26Circuit TracerAPR 26SteerSAE EvalsExportsinterp-engineNEW Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 64540
    Prev
    Next
    INDEX
    Explanations

    every / all + noun

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

    The neuron fires on words marking a recurring/time‐based expression—especially tokens like “coming,” “every,” and “month” in phrases about periodic events (e.g. “coming in every month”).

    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
    Negative Logits
     qualquer
    -1.30
     any
    -1.25
     cualquiera
    -1.17
     cualquier
    -1.15
    任何
    -1.09
     anywhere
    -1.00
     qualsiasi
    -0.98
     أي
    -0.97
    ค่ะ
    -0.94
     whichever
    -0.89
    POSITIVE LOGITS
     necesariamente
    0.96
    Cependant
    0.93
     necessarily
    0.92
     pê
    0.91
     فقط
    0.89
    צת
    0.89
    necessarily
    0.88
     every
    0.86
     béisbol
    0.85
    年轻人
    0.85
    Activations Density 0.044%

    No Known Activations