© 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. 10-GEMMASCOPE-RES-131K
    7. 129191
    Prev
    Next
    INDEX
    Explanations

    that followed by verbs

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

    The neuron strongly activates on function words that introduce relative or subordinate clauses (e.g. “that,” “who”).

    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
     was
    -2.50
    縐
    -2.48
     和
    -2.38
    های
    -2.27
    atellite
    -2.23
    sibilidade
    -2.22
    vedať
    -2.17
    ious
    -2.16
     it
    -2.16
     youre
    -2.14
    POSITIVE LOGITS
    嘡
    3.09
    亖
    2.97
     themselves
    2.92
     グリル
    2.88
     meisten
    2.88
    盦
    2.77
    colgante
    2.58
     مختلف
    2.55
    genodigd
    2.53
     trám
    2.53
    Activations Density 0.167%

    No Known Activations