© 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. 4228
    Prev
    Next
    INDEX
    Explanations

    separation characters

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

    The neuron is keyed to detecting personal names—especially initials plus surnames—as they appear in author lists or bibliographic entries.

    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
    permissions
    0.51
    entreprise
    0.50
     geographies
    0.50
     stipend
    0.49
     authorizations
    0.49
    覽
    0.48
    tochy
    0.47
    erar
    0.47
    affaires
    0.47
    PRNewswire
    0.46
    POSITIVE LOGITS
     The
    0.61
     the
    0.60
     That
    0.58
     this
    0.55
     that
    0.52
    0
    0.51
    1
    0.50
    4
    0.50
     two
    0.49
     This
    0.49
    Activations Density 0.218%

    No Known Activations