© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-3-12B
    5. 24-GEMMASCOPE-2-RES-16K
    7. 13559
    Prev
    Next
    INDEX
    Explanations

    Microsoft Edition, autoimmune, Pin 5

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

    This neuron responds to unusually content‐heavy or low-frequency tokens (i.e. rare or informative words rather than common function words).

    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
    لة
    0.80
    ي
    0.78
    ك
    0.74
    י
    0.72
    OrElse
    0.72
    ométrie
    0.70
    й
    0.69
    적인
    0.69
    דו
    0.68
    дно
    0.68
    POSITIVE LOGITS
     możemy
    0.88
    벅
    0.84
    έχ
    0.78
     kary
    0.77
     glycolysis
    0.76
     albumen
    0.75
     dám
    0.75
     oscillators
    0.74
    ikke
    0.73
     pię
    0.71
    Activations Density 0.000%

    No Known Activations