© 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. 2448
    Prev
    Next
    INDEX
    Explanations

    This neuron activates on numeric tokens, especially standalone numbers and floating-point 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_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
     cosidd
    0.66
    _
    0.65
    っぽい
    0.58
    index
    0.57
     pornography
    0.55
     neoplasms
    0.55
    学家
    0.55
    ahah
    0.54
    ка
    0.54
    -
    0.54
    POSITIVE LOGITS
     thereafter
    0.60
     OÜ
    0.60
     TBA
    0.58
     Energie
    0.55
     Zweifel
    0.55
     Cria
    0.55
     Antrieb
    0.54
     Gäste
    0.54
     Detalles
    0.54
     daarbij
    0.53
    Activations Density 0.001%

    No Known Activations