© 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. 1951
    Prev
    Next
    INDEX
    Explanations

    Microsoft Photoshop

    np_acts-logits-general · gemini-2.5-flash-lite

    The neuron is highly sensitive to numeric expressions—counts, measurements, durations, or other 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
     ອ
    0.85
    osulf
    0.84
    atthena
    0.82
     세포
    0.82
    ୃ
    0.80
    ಿಕ
    0.78
    ାନ
    0.78
     değerlend
    0.77
    ᡠ
    0.77
     Características
    0.77
    POSITIVE LOGITS
    ج
    0.88
    ()))
    0.75
    ادي
    0.75
    انة
    0.73
      
    0.72
    ين
    0.72
    also
    0.70
    ش
    0.70
     tenths
    0.70
     reinstall
    0.69
    Activations Density 0.000%

    No Known Activations