© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-3-12B
    5. 24-GEMMASCOPE-2-RES-16K
    7. 7132
    Prev
    Next
    INDEX
    Explanations

    who or that followed by verbs

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

    The neuron fires strongly on high-frequency function words—especially the relative pronoun “which” (and other auxiliaries like “are,” “have,” or the second‐person “you”).

    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
    за
    0.88
    э
    0.79
    ach
    0.77
    у
    0.73
    ז
    0.69
    рного
    0.64
     शिक्
    0.63
    з
    0.63
     মধ্যেও
    0.62
    ан
    0.62
    POSITIVE LOGITS
    muslim
    0.76
     Ols
    0.73
     касается
    0.71
    serializer
    0.71
    منى
    0.70
    toast
    0.69
     Ciebie
    0.69
    xmlns
    0.68
     Bhagav
    0.68
     Vn
    0.68
    Activations Density 0.097%

    No Known Activations