© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 34-GEMMASCOPE-RES-131K
    7. 98868
    Prev
    Next
    INDEX
    Explanations

    pronoun followed by a verb

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

    The neuron strongly activates on personal and relative pronouns (we, you, I, he, who, that) and other discourse markers indicating participants in the text.

    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
     становится
    -0.94
     devient
    -0.93
    ovina
    -0.92
     become
    -0.91
     הופ
    -0.89
     становятся
    -0.87
     FANDOM
    -0.87
    jący
    -0.86
     pueden
    -0.86
     продолжает
    -0.86
    POSITIVE LOGITS
     regularly
    1.57
     currently
    1.28
     регулярно
    1.15
     performs
    1.10
     happens
    1.10
     operates
    1.10
     runs
    1.09
     handles
    1.09
     spends
    1.09
     sell
    1.05
    Activations Density 0.203%

    No Known Activations