© 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 EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-3-12B
    5. 24-GEMMASCOPE-2-RES-16K
    7. 1703
    Prev
    Next
    INDEX
    Explanations

    This neuron strongly activates on the word “one” when used as the quantifier in phrases like “is one of the ….”

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-2-12b-pt/resid_post/layer_24_width_16k_l0_medium
    Prompts (Dashboard)
    392,802 prompts, 256 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    ش
    0.99
     néhány
    0.88
    ために
    0.84
    >∕
    0.80
     بسيطه
    0.80
     każ
    0.79
    {
    0.79
    ය
    0.79
     ĝi
    0.79
     acestea
    0.78
    POSITIVE LOGITS
    ida
    1.13
    াকী
    0.88
    ím
    0.86
     hundred
    0.83
     of
    0.74
    ir
    0.73
    ota
    0.72
    ational
    0.70
    ura
    0.70
    ைட்
    0.70
    Activations Density 0.093%

    No Known Activations