© 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. 22-GEMMASCOPE-RES-131K
    7. 66964
    Prev
    Next
    INDEX
    Explanations

    entity followed by descriptive phrase

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

    The neuron strongly activates on character references—proper names and personal pronouns (especially sentence-initial “She,” “Her,” 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_22/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
     these
    -1.18
     realizan
    -1.07
     MUNDO
    -0.99
     implementar
    -0.98
     rêves
    -0.97
     replying
    -0.94
     představ
    -0.93
     piden
    -0.92
    讪
    -0.90
    larınız
    -0.89
    POSITIVE LOGITS
    QUENCE
    0.92
     suuri
    0.88
    ede
    0.87
     $^{\
    0.87
    KEYWORD
    0.86
    俚
    0.85
     Haut
    0.85
     fj
    0.84
     Discogs
    0.84
    大多数
    0.83
    Activations Density 0.005%

    No Known Activations