© 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. 4250
    Prev
    Next
    INDEX
    Explanations

    cart" deletion

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

    The neuron activates on words that signal quantitative assessment or degree—terms used to measure, evaluate, or express intensity (e.g. assess, sensitivity, support, dependence, still, goal, solid, competent).

    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
    '>"+
    0.90
    及び
    0.89
     Vielzahl
    0.87
     மற்றும்
    0.86
    又は
    0.86
     serta
    0.84
    ী
    0.84
    يد
    0.82
     sendo
    0.82
     Calend
    0.81
    POSITIVE LOGITS
    anigans
    1.13
    ra
    1.02
     showdown
    1.02
    tdown
    1.02
     दवा
    0.97
     ordeal
    0.97
     badass
    0.95
    r
    0.93
     murky
    0.91
     racconta
    0.90
    Activations Density 3.754%

    No Known Activations