© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-3-12B
    5. 31-GEMMASCOPE-2-RES-16K
    7. 9823
    Prev
    Next
    INDEX
    Explanations

    list of numbered items

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

    The neuron selectively fires on bibliographic‐style or technical reference sections, i.e. on text that lists paper titles, conference names, years, volumes/pages, code snippet line numbers, and other formal citation details.

    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
    tosecond
    0.57
     ഭ
    0.55
    пуск
    0.54
     СП
    0.54
     Hazardous
    0.54
    amesh
    0.53
     بی‌
    0.53
    ренко
    0.52
     quests
    0.52
     mercenary
    0.52
    POSITIVE LOGITS
    玝
    0.56
    olver
    0.54
    EventQueue
    0.54
    琕
    0.54
     অফিসে
    0.53
    しまいます
    0.52
    仿真
    0.52
    "%
    0.52
    ভাষা
    0.52
    ريبا
    0.51
    Activations Density 0.005%

    No Known Activations