© 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. 1546
    Prev
    Next
    INDEX
    Explanations

    This neuron fires on single words or short phrases that open a clause or list, i.e. tokens immediately followed by a comma (the head of a coordinated phrase).

    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
     М
    0.41
     等
    0.38
     Л
    0.37
     др
    0.36
    اصيل
    0.36
     Ваш
    0.35
    🈷
    0.35
     Фу
    0.35
    עד
    0.35
     ylim
    0.35
    POSITIVE LOGITS
    atein
    0.41
    すなわち
    0.41
     പ്രത്യേ
    0.39
     vozila
    0.39
    ara
    0.38
    zwe
    0.38
    elsch
    0.38
     misalnya
    0.38
    anchi
    0.37
    ism
    0.37
    Activations Density 0.000%

    No Known Activations