© 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. 1934
    Prev
    Next
    INDEX
    Explanations

    May Key

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

    The neuron activates on numeric tokens representing dates—especially years and day–month 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_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
    ἦ
    0.51
    ajos
    0.50
    handlungen
    0.49
     bewust
    0.49
    inig
    0.49
    eração
    0.48
    inine
    0.47
    urde
    0.46
     trabal
    0.46
    ównie
    0.46
    POSITIVE LOGITS
     Ukraine
    0.51
    鱼
    0.50
     BlackBerry
    0.49
     அழுத்த
    0.48
     Donetsk
    0.48
    કિસ્તા
    0.48
    Ukraine
    0.47
    maxY
    0.47
     Бра
    0.46
     Ukrainians
    0.46
    Activations Density 0.211%

    No Known Activations