© 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. 2537
    Prev
    Next
    INDEX
    Explanations

    famously playing game

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

    The neuron detects mentions of high-level political offices and their holders (e.g. President, Chancellor, White House chief of staff).

    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
     thok
    0.73
     varsa
    0.70
    或者是
    0.67
     ataupun
    0.66
     কিংবা
    0.64
     ४
    0.63
    或是
    0.62
     atau
    0.61
     ३
    0.60
    もしくは
    0.59
    POSITIVE LOGITS
     célèbre
    0.83
     famously
    0.80
     знамени
    0.79
     berühm
    0.77
    总统
    0.74
     ünlü
    0.73
     استخدم
    0.72
     eponym
    0.71
     전쟁
    0.71
    たちの
    0.70
    Activations Density 0.044%

    No Known Activations