© 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-3-12B
    5. 24-GEMMASCOPE-2-RES-16K
    7. 16308
    Prev
    Next
    INDEX
    Explanations

    code keywords like player, options, status

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

    This neuron strongly activates on bibliographic metadata and citation-style elements (e.g. DOIs, journal references, “Retrieved,” parenthetical source notes, and similar reference lines).

    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
    코
    0.86
    Ub
    0.73
    Electronics
    0.70
    Departamento
    0.68
    UBS
    0.67
    એ
    0.67
    Ident
    0.66
    Hotel
    0.66
     Electronics
    0.65
    edent
    0.65
    POSITIVE LOGITS
     rakh
    0.86
     appease
    0.82
    ように
    0.80
     আপাতত
    0.80
    戾
    0.80
    cn
    0.80
    riwal
    0.80
     skit
    0.80
    ногда
    0.79
     świet
    0.79
    Activations Density 0.001%

    No Known Activations