© 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. 6076
    Prev
    Next
    INDEX
    Explanations

    The neuron fires on words right at the start of new clauses or sentences.

    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
     -
    0.83
    кови
    0.71
    ּ
    0.70
    ori
    0.64
    Several
    0.64
    ма
    0.63
    aty
    0.62
     консо
    0.62
    व
    0.62
    ant
    0.62
    POSITIVE LOGITS
     каттоо
    0.90
    厛
    0.89
    ruary
    0.86
    OLYBD
    0.84
     Faça
    0.82
    rtle
    0.82
     houve
    0.79
    Appuntamento
    0.79
    땜
    0.79
     mudanças
    0.79
    Activations Density 0.000%

    No Known Activations