© 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. 9152
    Prev
    Next
    INDEX
    Explanations

    Gemma team

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

    The neuron fires on isolated “G” tokens (especially when used as an initial or standalone capital “G” in names).

    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
     prompting
    0.63
    மாத
    0.57
    ا
    0.55
    parsing
    0.54
    Cave
    0.54
    osci
    0.53
     prosecute
    0.53
    ランチ
    0.53
    ang
    0.52
    MP
    0.52
    POSITIVE LOGITS
     г
    0.67
     kesin
    0.63
     unmistak
    0.60
     ouder
    0.60
     g
    0.59
     autophagy
    0.59
     aspecto
    0.58
     koš
    0.58
     vanlig
    0.57
    ಬಿ
    0.57
    Activations Density 0.339%

    No Known Activations