© 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
    2. Gemma-2-27B
    3. 10-GEMMASCOPE-RES-131K
    4. 48140
    Prev
    Next
    INDEX
    Explanations

    capture or catch

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

    The main thing this neuron does is identify occurrences of the verb “catch” (and its past-tense and related forms like “captured”).

    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
     داخ
    -1.40
    !(
    -1.37
    绚
    -1.36
    🤘
    -1.34
    骤
    -1.33
     that
    -1.30
    釓
    -1.30
     dolce
    -1.29
    峮
    -1.29
    ки
    -1.27
    POSITIVE LOGITS
    мія
    1.60
     Then
    1.58
     Only
    1.56
     Füßen
    1.56
     satisfy
    1.54
     That
    1.52
    ꜆
    1.52
     необхідно
    1.51
     capture
    1.49
     glimpse
    1.49
    Activations Density 0.017%

    No Known Activations