© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    2. Gemma-3-12B
    3. 31-GEMMASCOPE-2-RES-16K
    4. 9398
    Prev
    Next
    INDEX
    Explanations

    initially

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

    This neuron strongly activates on ordinal/adverbial expressions marking initial conditions or priorities—most notably the phrase “in the first place.”

    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
     الشعب
    0.59
    輙
    0.57
     Introducing
    0.55
     কয়েকটি
    0.54
    ücher
    0.54
    ldquo
    0.52
     Акы
    0.52
    }{*}{
    0.50
    য়ায়
    0.50
    さすが
    0.49
    POSITIVE LOGITS
     initially
    1.83
     originally
    1.62
     изначально
    1.57
     inicialmente
    1.51
     Initially
    1.48
     begin
    1.46
    Initially
    1.45
    最初
    1.37
     awalnya
    1.37
     originalmente
    1.34
    Activations Density 0.113%

    No Known Activations