© 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. 329
    Prev
    Next
    INDEX
    Explanations

    separators and subsequent characters/words

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

    The neuron fires on ampersands (and similar connectors) used to link items (e.g. “Rock & Pop”).

    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
    แต่
    0.23
    äum
    0.23
    钬
    0.23
    卻
    0.23
    ją
    0.22
    ära
    0.22
     thấy
    0.21
     можно
    0.21
     sätt
    0.21
     этого
    0.21
    POSITIVE LOGITS
    amp
    0.32
     other
    0.32
    /
    0.30
     T
    0.27
     অন্যান্য
    0.26
     Other
    0.25
    Amp
    0.25
    ndash
    0.25
     ಇತರ
    0.23
     P
    0.23
    Activations Density 0.553%

    No Known Activations