© 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. 1063
    Prev
    Next
    INDEX
    Explanations

    code delimiters

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

    The neuron fires on uppercase identifier tokens—especially constant names or flags (like O_CREAT, PROP_VALUE_MAX, CAPACITY, etc.).

    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
    ○○
    0.62
    🖉
    0.62
    JSPA
    0.61
    掮
    0.61
     wraz
    0.61
     څرنګ
    0.59
    Traité
    0.59
     مۇ
    0.57
     invitación
    0.57
    uola
    0.57
    POSITIVE LOGITS
     &&
    2.33
     ||
    2.06
    &&
    1.75
    ||
    1.59
    &&\
    1.39
    &&(
    1.30
    )&&
    1.27
    ){
    1.23
    '&&
    1.23
    '||
    1.07
    Activations Density 0.005%

    No Known Activations