© 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. 7317
    Prev
    Next
    INDEX
    Explanations

    Resemble, Reboot, Horrible, Extreme, Unchained

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

    This neuron activates on capitalized proper nouns and titles, detecting names of people, organizations, publications, events, and other named entities.

    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
    عة
    0.68
    ین
    0.62
    телем
    0.61
    quita
    0.60
    ва
    0.59
    ровка
    0.59
    incott
    0.58
    蠱
    0.57
    quito
    0.56
    tdown
    0.56
    POSITIVE LOGITS
     faisant
    0.68
     niezwy
    0.64
     särsk
    0.61
     soddis
    0.61
     progressivement
    0.61
     représentant
    0.60
     renforc
    0.59
    YOU
    0.59
     straordin
    0.59
     satisfacer
    0.59
    Activations Density 0.509%

    No Known Activations