© 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. 13521
    Prev
    Next
    INDEX
    Explanations

    The neuron strongly activates on numeric and measurement tokens (digits, dimensional markers like “2D,” “3D,” and units).

    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
     specjal
    0.61
    тва
    0.59
    ogel
    0.57
    𝐨
    0.55
    rystall
    0.54
     именно
    0.54
    য়াম
    0.54
    𝗿
    0.53
    یم
    0.52
    亢
    0.52
    POSITIVE LOGITS
     counterparts
    1.22
     counterpart
    1.21
    版的
    0.93
    쪽에
    0.93
    側の
    0.92
     version
    0.88
    쪽
    0.88
     stuff
    0.83
    에서의
    0.83
    側に
    0.82
    Activations Density 0.593%

    No Known Activations