© 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. 11460
    Prev
    Next
    INDEX
    Explanations

    self

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

    The neuron strongly activates on programming‐language code constructs (e.g. function definitions, method calls, and identifiers like “self”), effectively spotting code snippets.

    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
     {
    0.61
    []{
    0.61
     privaten
    0.58
     वृत्त
    0.58
     Unknown
    0.57
     Hauptstadt
    0.55
    خصوص
    0.55
     Ordnance
    0.55
    γν
    0.54
     metu
    0.53
    POSITIVE LOGITS
    )
    0.58
    )*
    0.57
    ),
    0.55
    눙
    0.55
    coh
    0.54
    ிஸ்த
    0.54
    oya
    0.54
    )=(
    0.54
    ozyg
    0.54
    /−
    0.52
    Activations Density 0.001%

    No Known Activations