© 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. 1747
    Prev
    Next
    INDEX
    Explanations

    presidents and administrations

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

    The neuron strongly activates on the surname “Obama,” i.e. it detects occurrences of the U.S. president’s last name.

    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
    界
    0.61
    О
    0.57
    上市公司
    0.55
     variar
    0.55
    商家
    0.55
    ör
    0.54
    的意思
    0.53
    }$,
    0.52
     উত
    0.52
     delimited
    0.52
    POSITIVE LOGITS
     Pentagon
    0.71
     gobiernos
    0.70
     governo
    0.66
     internationaux
    0.66
    ian
    0.66
     brothers
    0.65
    ’
    0.65
     Brothers
    0.64
    ův
    0.63
     appointees
    0.61
    Activations Density 0.004%

    No Known Activations