© 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
    2. Gemma-3-12B
    3. 24-GEMMASCOPE-2-RES-16K
    4. 6605
    Prev
    Next
    INDEX
    Explanations

    This neuron fires on high-level instructional or informational heading sentences—especially those using “to” + verb infinitives (e.g. “Find out what…,” “Check out our…,” “The answer to…,” “When it comes to…”).

    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
    .\\
    0.66
     ৬৫
    0.59
    </td>
    0.57
    </div>
    0.57
    IManager
    0.57
    Deployed
    0.57
    ולי
    0.55
     interferes
    0.55
    .</
    0.54
    רום
    0.54
    POSITIVE LOGITS
     “‘
    0.80
     “[
    0.80
     "¿
    0.79
     "[
    0.76
     "'
    0.72
     "(
    0.71
     “(
    0.67
    ՝
    0.65
     '[
    0.64
     "~
    0.64
    Activations Density 0.667%

    No Known Activations