© 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. 11523
    Prev
    Next
    INDEX
    Explanations

    referring to calculations, blocks, names

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

    The neuron detects technical research-methodology terms (e.g. “evaluation,” “contribution analysis,” “theory-driven,” “case studies,” “order”) typical of academic evaluation write-ups.

    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
     your
    0.73
    หน
    0.69
    <0x80>
    0.68
     вашего
    0.68
     +
    0.68
    y
    0.66
    }{
    0.64
    your
    0.64
     خواهد
    0.63
    ِ
    0.63
    POSITIVE LOGITS
    主要是
    0.81
     gdyż
    0.80
     क्योंकि
    0.79
     because
    0.77
    क्योंकि
    0.76
     inasmuch
    0.73
     whereby
    0.72
    जैसा
    0.72
     operat
    0.72
     emphas
    0.71
    Activations Density 0.018%

    No Known Activations