© 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. 2878
    Prev
    Next
    INDEX
    Explanations

    The neuron strongly activates on numeric expressions (digits, measurements, percentages), i.e. on tokens that represent numbers or numeric data.

    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
    }}\\
    0.61
     attributing
    0.56
     \\
    0.56
    րա
    0.55
    ));
    0.54
     notable
    0.54
     एना
    0.54
    )/(
    0.53
    ]));
    0.53
     claiming
    0.52
    POSITIVE LOGITS
     said
    3.38
    said
    3.11
     says
    2.94
    Said
    2.81
    says
    2.75
     сказал
    2.70
     dijo
    2.64
    Says
    2.61
     Says
    2.59
     قال
    2.48
    Activations Density 0.632%

    No Known Activations