© 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. 9990
    Prev
    Next
    INDEX
    Explanations

    dimensional representation

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

    The neuron strongly activates on mathematical explanations specifying a function’s domain and range (especially in the context of inverses and excluded values).

    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
    멸
    0.57
    urricular
    0.55
     Coronavirus
    0.55
     జమ
    0.55
    едера
    0.53
    ட்ரோ
    0.51
     Merger
    0.51
    சிரிய
    0.50
    শিং
    0.50
    造
    0.50
    POSITIVE LOGITS
     пространства
    1.16
     space
    1.14
     prostoru
    1.10
    空間
    1.09
     пространство
    1.08
    space
    1.04
     espaço
    1.04
     ruimte
    1.02
     espacio
    1.00
    空间
    1.00
    Activations Density 0.074%

    No Known Activations