© 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. 15561
    Prev
    Next
    INDEX
    Explanations

    amnesiafresh startstay on topic

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

    The neuron homes in on multi‐digit numeric tokens—especially those representing page numbers, years, or other reference identifiers.

    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
    स्तक
    0.76
    Yves
    0.72
    дж
    0.66
    Kindly
    0.65
     Careful
    0.64
    空调
    0.62
    و
    0.61
     गायत्री
    0.61
    cı
    0.60
     हल्दी
    0.60
    POSITIVE LOGITS
    ્ટ
    0.80
     $\{(
    0.80
    anego
    0.79
     maximising
    0.79
    meyi
    0.77
    ียบ
    0.76
    ͍
    0.76
     maximise
    0.75
     meilleur
    0.74
    фии
    0.74
    Activations Density 0.000%

    No Known Activations