© 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 EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-3-12B
    5. 24-GEMMASCOPE-2-RES-16K
    7. 8404
    Prev
    Next
    INDEX
    Explanations

    high =

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

    The neuron is primarily detecting floating‐point numeric tokens (decimal numbers) in the text.

    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_24_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
     тыс
    0.99
     molyb
    0.97
     codewords
    0.95
     eigenvectors
    0.94
     highlands
    0.94
    垁
    0.93
    碐
    0.92
     Ansible
    0.89
     sulphide
    0.88
     músculos
    0.87
    POSITIVE LOGITS
    än
    0.84
    {
    0.81
     und
    0.79
    ing
    0.77
    (
    0.77
    ,
    0.76
    den
    0.74
    är
    0.72
    ə
    0.71
     *
    0.71
    Activations Density 0.001%

    No Known Activations