© 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. 4892
    Prev
    Next
    INDEX
    Explanations

    you're looking for

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

    The neuron responds strongly to the subword token “re,” essentially detecting the contraction ending “’re.”

    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
    ти
    0.80
    т
    0.74
    с
    0.71
    ج
    0.68
    čkom
    0.66
    Что
    0.63
    asmuch
    0.63
    тным
    0.62
     काफ़ी
    0.61
    हुल
    0.61
    POSITIVE LOGITS
    ار
    0.64
    াল
    0.64
    ни
    0.63
    要知道
    0.61
    ції
    0.61
    要把
    0.60
     nevoie
    0.57
     devez
    0.56
    я
    0.56
    ind
    0.55
    Activations Density 0.001%

    No Known Activations