© 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-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 44786
    Prev
    Next
    INDEX
    Explanations

    proper names like Dee, Deepak, Dian

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

    The neuron activates on any token containing the sequence “dee” (in any casing).

    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
    ↵↵
    -4.03
     aqueles
    -3.33
     myriad
    -3.14
     it
    -2.92
    籏
    -2.91
     
    -2.81
     We
    -2.73
    </u>
    -2.64
     comprender
    -2.59
    ↵
    -2.53
    POSITIVE LOGITS
    While
    4.19
    When
    4.03
    Despite
    3.66
    Before
    3.48
    After
    3.48
    One
    3.42
    Of
    3.38
    ”،
    3.27
    Because
    3.22
    Although
    3.22
    Activations Density 0.003%

    No Known Activations