© 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. 10860
    Prev
    Next
    INDEX
    Explanations

    The neuron fires most strongly on the very first substantive word of each new snippet, i.e. it’s detecting the start of a new document segment.

    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
    妇
    0.71
     Yep
    0.63
     लाभ
    0.62
     സം
    0.61
     中
    0.60
     ts
    0.59
    ide
    0.59
    BY
    0.59
    tc
    0.59
    atthe
    0.58
    POSITIVE LOGITS
     harmonique
    0.98
     parfois
    0.84
     garantia
    0.83
     Protok
    0.82
    🐼
    0.81
     moteurs
    0.79
     basaltes
    0.79
     វា
    0.79
    getConfiguration
    0.79
     Pengh
    0.78
    Activations Density 0.001%

    No Known Activations