© 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. 34-GEMMASCOPE-RES-131K
    7. 98920
    Prev
    Next
    INDEX
    Explanations

    idiomatic phrases completion

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

    The neuron strongly activates on words that are parts of hyphenated compound expressions (e.g. “knee-jerk,” “long-tail,” “short-term,” “full-throttle,” “bottom-line,” “rag-tag”).

    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
     and
    -1.58
     "..\..\
    -1.03
     well
    -1.02
    xba
    -0.97
    づけ
    -0.97
    aters
    -0.94
     While
    -0.94
     With
    -0.93
    టి
    -0.93
     teh
    -0.92
    POSITIVE LOGITS
     approach
    1.63
     affair
    1.20
     attitude
    1.10
    Spesifikasi
    1.07
    ıyorum
    1.06
    Especificações
    1.05
     ヴィンテージ
    1.03
     красивый
    1.03
     Inſ
    1.01
    的な
    1.00
    Activations Density 0.063%

    No Known Activations