© 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 EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-3-12B
    5. 31-GEMMASCOPE-2-RES-16K
    7. 13024
    Prev
    Next
    INDEX
    Explanations

    languages and JSON

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

    The neuron activates strongly on floating‐point numeric tokens (numbers with decimal points).

    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
    mainloop
    0.60
     tuan
    0.55
     نجي
    0.54
     而
    0.53
    యంలో
    0.52
    ("|"+"
    0.52
     "*************"
    0.52
    empio
    0.51
    ([])
    0.51
    ✄
    0.51
    POSITIVE LOGITS
     despre
    0.57
     удержи
    0.52
    ichtlich
    0.51
     восто
    0.50
    द्दल
    0.49
     большей
    0.49
     bel
    0.49
     win
    0.48
    drill
    0.48
     drill
    0.48
    Activations Density 0.000%

    No Known Activations