© 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. 11191
    Prev
    Next
    INDEX
    Explanations

    programming keywords

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

    The neuron is highly sensitive to C++ scope‐resolution syntax (e.g. the “::” in names like “llvm::…”).

    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
    strnc
    0.86
    buf
    0.79
    putchar
    0.76
    scanf
    0.75
     scanf
    0.74
    Malloc
    0.73
    Buffer
    0.72
    (*(
    0.72
     nullptr
    0.71
    memcpy
    0.71
    POSITIVE LOGITS
     auto
    0.69
     pajama
    0.63
     millennials
    0.62
    сцю
    0.61
     appraisals
    0.60
     citas
    0.60
    posição
    0.58
    auto
    0.57
     patron
    0.57
     fraudsters
    0.57
    Activations Density 0.003%

    No Known Activations