© 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. 31-GEMMASCOPE-2-RES-16K
    7. 5683
    Prev
    Next
    INDEX
    Explanations

    al gusto, al principio, al instante

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

    The neuron spikes on the standalone or hyphenated uppercase “Al,” i.e. the Arabic definite article prefix used in names (e.g. Al-Akhbar, Al-Qaeda, Al-Mitchell).

    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_31_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
    ấm
    0.59
    ណ៍
    0.58
    ্ম
    0.58
    oblot
    0.55
    ်
    0.55
     மேற்கு
    0.54
    emont
    0.54
    izion
    0.54
    slategray
    0.53
    edt
    0.50
    POSITIVE LOGITS
     Jazeera
    0.80
    tai
    0.61
    pte
    0.60
     Qaeda
    0.59
     Al
    0.57
    tu
    0.55
     Aw
    0.54
    較
    0.53
     зару
    0.52
     insertar
    0.52
    Activations Density 0.054%

    No Known Activations