© 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 EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-3-12B
    5. 31-GEMMASCOPE-2-RES-16K
    7. 2091
    Prev
    Next
    INDEX
    Explanations

    the world of

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

    The neuron activates on prominent headings or section‐title phrases—i.e. the main capitalized topic or title at the start of each content block.

    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_31_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
     상대
    0.86
     정치
    0.82
     ข้าง
    0.80
     antagon
    0.79
     отрица
    0.77
    ()->
    0.77
     cỡ
    0.77
     사망
    0.77
     bị
    0.75
    }=-\
    0.74
    POSITIVE LOGITS
     beautiful
    1.42
     amazing
    1.41
     prachtige
    1.39
     intricacies
    1.38
     incredible
    1.37
     magnificent
    1.34
     wonderful
    1.32
     vibrant
    1.28
     vielfält
    1.23
     increí
    1.20
    Activations Density 1.154%

    No Known Activations