© 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. 2975
    Prev
    Next
    INDEX
    Explanations

    The neuron fires strongly on social-media‐style sign-off tokens—things like exclamation-point farewells, well-wishes, and concluding pleasantries at the ends of posts.

    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
    ασία
    0.57
     وبركاته
    0.54
    面前
    0.52
    このように
    0.52
    დესაც
    0.51
    郄
    0.51
    войства
    0.51
     כאשר
    0.51
    េះ
    0.50
    これらの
    0.50
    POSITIVE LOGITS
     Sincerely
    0.60
     李
    0.57
     刘
    0.56
     Regards
    0.56
    -_-
    0.56
     aberrant
    0.56
     劉
    0.55
    Pete
    0.54
     Pete
    0.53
     canadien
    0.53
    Activations Density 0.085%

    No Known Activations