© 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. 34-GEMMASCOPE-RES-131K
    7. 79014
    Prev
    Next
    INDEX
    Explanations

    clue or female names

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

    This neuron responds to capitalized tokens—especially proper names, titles, acronyms, or other words beginning with an uppercase letter.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_34/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
     cataract
    -0.75
     silicate
    -0.74
    zyl
    -0.73
    𝐡
    -0.72
    EXPECT
    -0.72
     банков
    -0.69
    ichtlich
    -0.69
     الخط
    -0.68
     Stelle
    -0.66
    ujte
    -0.66
    POSITIVE LOGITS
    Clu
    1.09
     Clu
    0.99
    clue
    0.80
     Dion
    0.77
     Alicia
    0.77
    Alicia
    0.77
    oine
    0.75
     emma
    0.74
     totaal
    0.73
     donating
    0.72
    Activations Density 0.019%

    No Known Activations