© 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. 34-GEMMASCOPE-RES-131K
    7. 90020
    Prev
    Next
    INDEX
    Explanations

    proverbs and sayings

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

    the neuron detects words that appear in idiomatic or proverbial expressions (e.g. “a lion…sheep,” “drop by drop…pot,” “taking candy from a baby,” etc.).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_34/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
     unseres
    -0.85
    的那种
    -0.85
    implicitly
    -0.82
    ัตว์
    -0.79
     almost
    -0.77
    Personne
    -0.75
     актри
    -0.75
    HAPPY
    -0.75
     Petitioner
    -0.75
    Actually
    -0.75
    POSITIVE LOGITS
     straw
    0.95
    "|
    0.90
    dió
    0.88
    /${
    0.86
    >::
    0.85
    unmodifiable
    0.85
     dynamite
    0.85
    ighed
    0.84
    straw
    0.84
     اذا
    0.83
    Activations Density 0.029%

    No Known Activations