© 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. 13457
    Prev
    Next
    INDEX
    Explanations

    The neuron activates strongly on numeric expressions (e.g. measurements, model numbers, and other numeric tokens).

    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
     HIPAA
    0.70
     OSHA
    0.67
     fracking
    0.66
     FERC
    0.66
     C
    0.61
     Adder
    0.59
    罣
    0.59
     plagued
    0.59
     pursuant
    0.58
     Fermi
    0.58
    POSITIVE LOGITS
    娶
    1.10
     esposa
    0.99
     wives
    0.94
     brides
    0.94
     menikah
    0.90
     bride
    0.90
     жена
    0.86
     жены
    0.85
     marrying
    0.84
     moglie
    0.84
    Activations Density 0.067%

    No Known Activations