© 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. 80594
    Prev
    Next
    INDEX
    Explanations

    lead or leading

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

    The neuron activates on tokens containing the string “lead,” including “lead,” “leading,” or hyphenated “lead-” compounds.

    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
    In
    -2.39
    e
    -2.20
    r
    -1.96
     seventeen
    -1.94
    ꜆
    -1.92
    Most
    -1.90
    O
    -1.89
    Very
    -1.86
    d
    -1.86
     "/")
    -1.84
    POSITIVE LOGITS
     isn
    1.98
     píše
    1.80
     utilizzo
    1.77
     lead
    1.74
     obrázok
    1.71
     Lead
    1.69
    鸶
    1.69
    を行
    1.69
     чле
    1.69
    自体
    1.68
    Activations Density 0.025%

    No Known Activations