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

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-3-12B
    5. 24-GEMMASCOPE-2-RES-16K
    7. 13586
    Prev
    Next
    INDEX
    Explanations

    defining or rules

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

    The neuron strongly lights up on the second half of hyphenated compounds (e.g. “island-hopper,” “A-line”), i.e. tokens that follow a dash.

    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
    任
    0.76
     건설
    0.76
     yönetim
    0.75
     히
    0.75
    städter
    0.75
    ְ
    0.74
     Gerät
    0.72
     Pemer
    0.72
    개발
    0.72
     Constructs
    0.72
    POSITIVE LOGITS
    া
    0.91
    ल
    0.75
    த்தில்
    0.73
    pping
    0.71
    pped
    0.70
    accam
    0.70
    篙
    0.69
     wonderful
    0.69
    ToBe
    0.68
    lastval
    0.68
    Activations Density 0.051%

    No Known Activations