© 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. 24-GEMMASCOPE-2-RES-16K
    7. 2522
    Prev
    Next
    INDEX
    Explanations

    **arXiv physics papers**

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

    The neuron primarily detects numeric tokens, such as years, volume/page numbers, DOIs, and other number sequences in the text.

    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
    Attorney
    0.77
    Business
    0.75
    8
    0.75
     Charities
    0.74
     showbiz
    0.74
     Brentwood
    0.73
     Business
    0.72
    2
    0.72
    9
    0.72
    Buying
    0.71
    POSITIVE LOGITS
     perturbative
    1.48
     fermionic
    1.23
     fermion
    1.17
     physics
    1.14
     neutrino
    1.13
     neutrinos
    1.12
     supersymmetric
    1.09
     wavefunction
    1.08
     arXiv
    1.07
     collisional
    1.07
    Activations Density 0.003%

    No Known Activations