© 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. 1915
    Prev
    Next
    INDEX
    Explanations

    The neuron is specialized to detect numeric expressions and quantities (e.g. multi‐digit numbers, monetary amounts, measurements, and other numeric constants).

    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
    했으며
    0.63
     розгля
    0.59
     sowie
    0.55
     хотя
    0.55
    *,
    0.54
     Macaulay
    0.54
    ទេ
    0.54
    *',
    0.53
     Gog
    0.52
     ভাব
    0.52
    POSITIVE LOGITS
     =
    3.98
     $=
    3.39
    =
    3.39
     $=\
    3.13
     $=$
    3.11
    ＝
    2.98
     }=
    2.94
     =\
    2.89
    $=
    2.88
    }=
    2.83
    Activations Density 0.142%

    No Known Activations