© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-3-12B
    5. 31-GEMMASCOPE-2-RES-16K
    7. 11052
    Prev
    Next
    INDEX
    Explanations

    The neuron strongly activates on numeric tokens (e.g. years, dates, and other numbers).

    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
    5
    0.71
    0
    0.71
    Co
    0.66
    2
    0.65
    6
    0.63
    King
    0.63
    E
    0.63
    1
    0.62
    Pr
    0.61
    J
    0.61
    POSITIVE LOGITS
     sunburn
    0.57
     transporte
    0.55
     lado
    0.53
     flanks
    0.51
     torque
    0.49
     darkness
    0.49
    ಲೆಯ
    0.49
     spatially
    0.48
    ्यातील
    0.48
     desper
    0.47
    Activations Density 0.052%

    No Known Activations