© 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. 13509
    Prev
    Next
    INDEX
    Explanations

    The neuron strongly activates on named entities and other proper nouns (e.g. personal names, company names, place names, titles).

    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
    ळख
    0.60
     streamlining
    0.56
     yelling
    0.55
     неда
    0.55
    咎
    0.55
     дорого
    0.54
     बुल
    0.53
     responsibly
    0.52
     hypocrisy
    0.52
     закончи
    0.52
    POSITIVE LOGITS
     ɛ
    0.62
     eTo
    0.55
    ለሁ
    0.54
     Atual
    0.54
    води
    0.53
     EB
    0.53
    Flutter
    0.52
     মিনিটে
    0.51
    泥
    0.51
     EH
    0.50
    Activations Density 0.388%

    No Known Activations