© 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. 11110
    Prev
    Next
    INDEX
    Explanations

    pain, return, beliefs, existed

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

    The neuron strongly activates on standalone numeric strings (especially floating‐point numbers).

    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
     competent
    0.59
     competence
    0.55
     device
    0.55
     user
    0.53
     ratione
    0.52
     designated
    0.51
    .
    0.51
     editor
    0.50
    мян
    0.50
     ctrl
    0.49
    POSITIVE LOGITS
    银
    0.69
    alertDialog
    0.69
    僌
    0.68
    kében
    0.66
    Ꮼ
    0.66
    🍂
    0.66
    ayın
    0.65
    鱼
    0.62
    BEGIN
    0.61
    তুম
    0.61
    Activations Density 0.000%

    No Known Activations