© 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. 3927
    Prev
    Next
    INDEX
    Explanations

    than how much

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

    The neuron is specifically tuned to detect numeric information—tokens representing numbers, dates, quantities, or statistics.

    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
     أ
    0.88
     अॅ
    0.86
    Ach
    0.82
     ஆ
    0.82
     kuda
    0.80
    オ
    0.79
    鷺
    0.79
    Krist
    0.77
    イチ
    0.77
    里
    0.77
    POSITIVE LOGITS
    fontWeight
    0.73
    ep
    0.69
     secondly
    0.69
    ens
    0.67
    umum
    0.67
     स्वभाव
    0.66
    ūsų
    0.66
    шов
    0.66
    lw
    0.65
     prolong
    0.64
    Activations Density 0.001%

    No Known Activations