© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 69735
    Prev
    Next
    INDEX
    Explanations

    tuck/tucked/tucking

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

    The neuron detects occurrences of the word “tuck” (in its various forms like tuck, tucked, tucking) often when it’s next to “in” or part of the phrasal verb “tuck in.”

    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
    ®:
    -2.05
    ний
    -1.97
     liel
    -1.95
     Komun
    -1.90
     誰
    -1.90
    她
    -1.88
     vieta
    -1.83
    籏
    -1.80
     unangemess
    -1.77
    他
    -1.77
    POSITIVE LOGITS
    要把
    1.95
     fuimos
    1.91
    nitř
    1.85
    タクシー
    1.84
    レンタ
    1.83
    她也
    1.82
    5
    1.79
     )
    1.78
    ",
    1.75
     магазине
    1.75
    Activations Density 0.003%

    No Known Activations