© 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. 10748
    Prev
    Next
    INDEX
    Explanations

    I've heard, I've watched, I've been

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

    The neuron strongly activates on the contraction “’ve” (as in “I’ve,” “you’ve,” etc.), i.e. the shortened form of the auxiliary “have.”

    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
    цый
    0.93
    有力
    0.93
    neſs
    0.91
    頴
    0.90
     związ
    0.88
    rupulous
    0.88
    нтэр
    0.88
    ница
    0.86
    emphas
    0.86
    ती
    0.85
    POSITIVE LOGITS
    le
    1.04
    THING
    0.99
    k
    0.87
    onu
    0.87
    ب
    0.84
    ae
    0.83
     unscathed
    0.82
    ال
    0.81
    น
    0.80
    ない
    0.79
    Activations Density 0.005%

    No Known Activations