© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it
    Neuronpedia
    Jacobian LensNEW
    Natural Language
    Autoencoders
    MAY 26    Assistant AxisJAN 26Circuit TracerAPR 26SteerSAE EvalsExportsinterp-engineNEW Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-3-12B
    5. 31-GEMMASCOPE-2-RES-16K
    7. 13904
    Prev
    Next
    INDEX
    Explanations

    The neuron activates on numeric values (especially decimal numbers) in the text.

    oai_token-act-pair · o4-mini
    Triggered 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
    Negative Logits
     over
    0.66
     über
    0.66
    舰队
    0.64
     Prav
    0.61
     sobre
    0.58
     Coffee
    0.57
    
    0.55
     Pra
    0.55
     PRA
    0.54
    ETTE
    0.53
    POSITIVE LOGITS
    σιμο
    0.57
     해석
    0.55
    reiber
    0.55
    ijek
    0.54
    %:
    0.54
     almacenamiento
    0.53
     स्थिरता
    0.53
    rscheinlich
    0.52
    :\
    0.52
    iffness
    0.52
    Activations Density 0.000%

    No Known Activations