© 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. 8076
    Prev
    Next
    INDEX
    Explanations

    ISO standards

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

    The neuron activates on tokens that are purely numeric or alphanumeric codes (e.g. dates, quantities, ISO and model 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
     Congressional
    0.61
     congressional
    0.58
    眉头
    0.57
     빨
    0.56
    慢
    0.56
     numquam
    0.55
    翀
    0.55
    颤
    0.55
     Delayed
    0.54
     प्रोटीन
    0.54
    POSITIVE LOGITS
     ISO
    1.77
    ISO
    1.71
     iso
    1.51
    Iso
    1.39
    iso
    1.38
     Iso
    1.32
     आईएस
    0.85
     আইএস
    0.81
     isoform
    0.78
     sistem
    0.74
    Activations Density 0.122%

    No Known Activations