© 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. 9597
    Prev
    Next
    INDEX
    Explanations

    both, neither

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

    The neuron detects numeric measurement tokens—i.e. the numbers and decimal values used to specify dimensions or sizes in technical/product descriptions.

    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
    конец
    0.60
    इस
    0.55
    ovací
    0.55
    २
    0.54
    १
    0.53
    öne
    0.53
    ignation
    0.51
    ८
    0.51
    ampshire
    0.50
    خص
    0.49
    POSITIVE LOGITS
     both
    0.92
     både
    0.91
     keduanya
    0.90
     zowel
    0.82
     neither
    0.81
    both
    0.80
     Both
    0.80
     zarówno
    0.79
    ทั้ง
    0.76
     ทั้ง
    0.76
    Activations Density 0.886%

    No Known Activations