© 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. 2238
    Prev
    Next
    INDEX
    Explanations

    choose one

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

    This neuron primarily activates on the standalone numeral “one” (often used as a pronoun or determiner).

    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
    ।
    1.03
    ই
    0.98
    ።
    0.96
    。
    0.92
    챱
    0.90
    นะ
    0.88
    ش
    0.87
    だろう
    0.84
    ৯
    0.84
    ۔
    0.83
    POSITIVE LOGITS
     surm
    0.87
     controversies
    0.86
    otted
    0.76
    ymmetrical
    0.76
     adop
    0.75
    ervice
    0.75
     promul
    0.75
     immunological
    0.75
     concerne
    0.74
     deciso
    0.74
    Activations Density 0.091%

    No Known Activations