© 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. 31-GEMMASCOPE-2-RES-16K
    7. 11635
    Prev
    Next
    INDEX
    Explanations

    assembly code delimiters

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

    This neuron fires on code‐style tokens (e.g. registers, numeric and hex constants, operators) typical of assembly or programming syntax.

    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_31_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
    獪
    0.61
    UTRON
    0.57
    鸪
    0.57
     கேரள
    0.57
     ভিয়েতনাম
    0.56
     оказыва
    0.55
     девочки
    0.55
    峼
    0.55
     выпуска
    0.54
     Tuti
    0.54
    POSITIVE LOGITS
     mov
    0.71
    dword
    0.71
    mov
    0.70
     eax
    0.66
    指令
    0.64
    Stack
    0.61
     stacked
    0.61
    esp
    0.59
     prologue
    0.58
     Stack
    0.57
    Activations Density 0.021%

    No Known Activations