© 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. 31-GEMMASCOPE-2-RES-16K
    7. 2339
    Prev
    Next
    INDEX
    Explanations

    This neuron strongly activates on named entities and other capitalized technical or proper nouns.

    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_31_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
     foundational
    0.83
     geospatial
    0.79
    各项
    0.78
    众多
    0.78
     sämt
    0.78
     tasked
    0.77
    各类
    0.75
     各
    0.75
     तमाम
    0.75
    诸多
    0.75
    POSITIVE LOGITS
     иногда
    1.65
     Occasionally
    1.64
    Occasionally
    1.63
     occasionally
    1.51
     manchmal
    1.50
     Иногда
    1.45
     Rarely
    1.45
     Sometimes
    1.44
     sometimes
    1.41
    有时
    1.41
    Activations Density 0.523%

    No Known Activations