© 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. 3067
    Prev
    Next
    INDEX
    Explanations

    which introduces explanatory clauses

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

    The neuron is essentially a “which” detector—it fires strongly whenever the relative‐clause introducer “which” appears.

    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
    ી
    0.95
     whatnot
    0.94
    いっぱい
    0.91
    綃
    0.86
    ות
    0.85
    ール
    0.84
    ه
    0.84
    اٹ
    0.80
    та
    0.79
    ים
    0.79
    POSITIVE LOGITS
    되었다
    0.93
    はこの
    0.91
     কিনা
    0.86
    はお
    0.85
    ている
    0.84
     minimizes
    0.83
     وهذه
    0.83
     includes
    0.78
    af
    0.77
    يد
    0.77
    Activations Density 0.097%

    No Known Activations