© 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. 24-GEMMASCOPE-2-RES-16K
    7. 14901
    Prev
    Next
    INDEX
    Explanations

    changed, Geneva Conventions, Data, instead, emotional, resonance, cluster, ate

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

    The neuron strongly activates on numeric or quantitative tokens (i.e. numbers and measurements).

    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
    𝙋
    0.89
     끌
    0.88
    ᄊ
    0.88
    Из
    0.87
    Также
    0.85
     exhibiting
    0.83
    无疑
    0.83
     воспринима
    0.83
     ozbil
    0.82
    VIDOR
    0.80
    POSITIVE LOGITS
    ت
    0.72
     ingin
    0.72
    ek
    0.71
    ्यादा
    0.65
    de
    0.63
    تا
    0.63
     biển
    0.63
    em
    0.62
     reli
    0.61
    ેલ
    0.61
    Activations Density 0.001%

    No Known Activations