© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 111804
    Prev
    Next
    INDEX
    Explanations

    named entities or titles

    np_acts-logits-general · gemini-2.5-flash-lite

    The neuron fires on multiword title‐case proper names—especially titles of songs, albums, podcasts, etc.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_10/width_131k
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     célè
    -1.74
     illustratie
    -1.68
     kanker
    -1.59
     magasiner
    -1.51
    蠻
    -1.48
    詼
    -1.48
    蟶
    -1.46
     systèmes
    -1.45
    鹀
    -1.44
    ABAD
    -1.43
    POSITIVE LOGITS
     also
    2.09
     on
    1.77
     Also
    1.64
     /
    1.57
    </em>
    1.54
     These
    1.51
     kenaikan
    1.46
    These
    1.44
     There
    1.41
     if
    1.39
    Activations Density 0.148%

    No Known Activations