© 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. 1336
    Prev
    Next
    INDEX
    Explanations

    This neuron strongly activates on numeric tokens (especially multi‐digit numbers and decimals), essentially detecting numbers in the text.

    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
     savings
    0.71
    ർ
    0.70
     derivados
    0.69
     passport
    0.68
     zo
    0.68
     Ио
    0.68
    ̼
    0.66
    ೂ
    0.63
     Zoology
    0.63
    ch
    0.62
    POSITIVE LOGITS
     fornire
    0.76
    注意事项
    0.73
     яким
    0.73
    锺
    0.71
     mettant
    0.70
    пикир
    0.70
    Authorities
    0.70
    Awards
    0.69
    FBSDKGraph
    0.69
    俆
    0.69
    Activations Density 0.001%

    No Known Activations