© 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. 86804
    Prev
    Next
    INDEX
    Explanations

    to this day / to this end

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

    The neuron fires on occurrences of the word “this” when it follows “to” (e.g. in phrases like “to this end” or “to this day”).

    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
     other
    -2.53
     was
    -2.44
    了一个
    -2.39
    cayaan
    -2.30
     of
    -2.23
     in
    -2.17
    是真的
    -2.02
    询
    -1.98
     includes
    -1.98
     can
    -1.98
    POSITIVE LOGITS
    medriver
    2.47
    дельник
    2.39
     навіть
    2.38
    蕻
    2.38
     zelfs
    2.36
     través
    2.30
    ées
    2.28
     nawet
    2.28
     даже
    2.27
    尽管
    2.25
    Activations Density 0.015%

    No Known Activations