© 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. 24-GEMMASCOPE-2-RES-16K
    7. 430
    Prev
    Next
    INDEX
    Explanations

    The neuron is strongly activated by numeric tokens, especially decimal‐formatted 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_24_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
    0.38
    1
    0.32
    ただし
    0.31
    6
    0.30
    2
    0.30
    5
    0.30
     alimentos
    0.29
    4
    0.29
    结束后
    0.28
    вання
    0.28
    POSITIVE LOGITS
     whoever
    0.35
     همه
    0.34
     당신
    0.31
     Ты
    0.31
     אחר
    0.31
     Глав
    0.30
     얘
    0.30
    那些
    0.30
    那位
    0.30
     พวก
    0.30
    Activations Density 5.372%

    No Known Activations