© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it
    Neuronpedia
    Jacobian LensNEW
    Natural Language
    Autoencoders
    MAY 26    Assistant AxisJAN 26Circuit TracerAPR 26SteerSAE EvalsExportsinterp-engineNEW Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 45587
    Prev
    Next
    INDEX
    Explanations

    Wolfe and phantom names

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

    The neuron is selectively activated by uncommon proper nouns or named entities (e.g. Wolfe, Phantom).

    oai_token-act-pair · o4-mini
    Triggered 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
    Negative Logits
    癭
    -2.77
    絎
    -2.73
    ῞
    -2.53
    .”
    -2.48
    猡
    -2.47
     основан
    -2.45
    歘
    -2.42
     современных
    -2.42
    ⢌
    -2.42
    😚
    -2.36
    POSITIVE LOGITS
    2
    3.81
    This
    3.48
    7
    3.42
    These
    3.39
    4
    3.39
    6
    3.20
    While
    3.02
    -
    2.97
    Also
    2.95
    3
    2.94
    Activations Density 0.005%

    No Known Activations