© 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. 24-GEMMASCOPE-2-RES-16K
    7. 1655
    Prev
    Next
    INDEX
    Explanations

    The neuron fires on the first substantial (non-function) word that begins a new sentence or 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
    oned
    0.82
    ’
    0.80
    த்து
    0.77
    ^{-(
    0.77
    alla
    0.75
     attacked
    0.74
    ‘
    0.73
    ème
    0.72
    achio
    0.72
    vori
    0.71
    POSITIVE LOGITS
    ابه
    0.84
    ی
    0.77
    )=-
    0.75
    মেঘ
    0.73
    )=
    0.73
    )')
    0.73
    𝐛
    0.72
    )>
    0.71
    )==
    0.71
    )&
    0.71
    Activations Density 0.000%

    No Known Activations