© 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. 7418
    Prev
    Next
    INDEX
    Explanations

    1 following list or count

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

    The neuron strongly activates on numeric tokens or numeric sequences (e.g., list indices, measurements, or counts).

    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
    Ꮁ
    0.77
    我
    0.69
    你
    0.63
    你怎么
    0.62
     altres
    0.62
    我們可以
    0.58
    ем
    0.58
    拏
    0.58
    Ciao
    0.57
    azion
    0.57
    POSITIVE LOGITS
    ️⃣
    0.70
    ずつ
    0.62
    つの
    0.61
    씩
    0.60
    ⃣
    0.59
    ící
    0.54
     obituary
    0.54
     scoop
    0.53
    ″
    0.51
    xbet
    0.51
    Activations Density 0.068%

    No Known Activations