© 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
    2. Gemma-3-12B
    3. 24-GEMMASCOPE-2-RES-16K
    4. 4917
    Prev
    Next
    INDEX
    Explanations

    here's or here are

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

    The neuron specifically detects the discourse adverb “Here” (i.e. occurrences of the word “here” used to introduce or point to something).

    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
    いつ
    1.05
    たと
    1.02
    आर
    0.98
    ক
    0.98
    ES
    0.98
    করণ
    0.98
     embora
    0.98
     Winf
    0.96
    อาจ
    0.96
    この
    0.95
    POSITIVE LOGITS
     وهنا
    1.43
    abouts
    1.20
    fordshire
    1.08
    age
    1.03
    서
    1.00
     നിന്ന്
    0.99
    inafter
    0.96
     いっ
    0.92
     présents
    0.91
    ıl
    0.91
    Activations Density 0.135%

    No Known Activations