© 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. 1506
    Prev
    Next
    INDEX
    Explanations

    here's a breakdown of

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

    The neuron is specialized to detect explicit numeric references (numbers and quantities) 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
    在
    0.43
     ді
    0.42
    たい
    0.41
    📑
    0.40
     explic
    0.40
    на
    0.40
    説明
    0.39
    ด
    0.39
    นอก
    0.38
     потрібно
    0.38
    POSITIVE LOGITS
     Decade
    0.49
     Softball
    0.49
     Dollars
    0.48
     Days
    0.48
     Defts
    0.46
     Frankly
    0.46
    oitte
    0.46
    سة
    0.45
     Dreams
    0.45
     Playoffs
    0.44
    Activations Density 0.122%

    No Known Activations