© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-3-12B
    5. 31-GEMMASCOPE-2-RES-16K
    7. 15977
    Prev
    Next
    INDEX
    Explanations

    f-string formatting with variables

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

    The neuron is consistently picking out all-uppercase tokens—e.g. code keywords, macro directives, or acronyms.

    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
     {},
    0.77
     {}",
    0.73
    ({})
    0.63
     {})
    0.63
    {}",
    0.60
     Lua
    0.55
     represented
    0.55
     ◇
    0.55
    ■
    0.55
    {},
    0.54
    POSITIVE LOGITS
    (`${
    1.77
    `${
    1.51
     `${
    1.50
    /${
    1.40
    ($"{
    1.28
    -${
    1.26
    .${
    1.21
    }/${
    1.20
    "${
    1.20
    /#{
    1.20
    Activations Density 0.012%

    No Known Activations