© 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. 11748
    Prev
    Next
    INDEX
    Explanations

    The neuron consistently lights up on subword tokens that begin with a capital letter (i.e. proper‐name or title fragments).

    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
     miglior
    0.84
    и
    0.78
     ι
    0.77
    δήποτε
    0.76
    ਾ
    0.76
    ্ি
    0.76
    )=
    0.75
    詿
    0.75
     星
    0.75
    砊
    0.75
    POSITIVE LOGITS
    crumbs
    0.84
    sand
    0.81
    cursors
    0.81
    brick
    0.80
    grit
    0.80
    shifts
    0.79
    backs
    0.79
    checker
    0.79
    spray
    0.77
    variants
    0.77
    Activations Density 0.000%

    No Known Activations