© 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. 130043
    Prev
    Next
    INDEX
    Explanations

    This neuron detects extremely common English function words (especially the infinitive marker “to”).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408

    to + forceful action

    np_acts-logits-general · gemini-2.5-flash-lite
    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
     được
    -1.27
    ᾰ
    -1.23
    ⸒
    -1.21
     Konflikt
    -1.10
     experiencing
    -1.10
     دریافت‌شده
    -1.09
    ที่มี
    -1.06
    娯
    -1.06
     visitó
    -1.05
    漁港
    -1.05
    POSITIVE LOGITS
     instig
    1.23
     upset
    1.08
    postolic
    1.07
     COMMON
    1.05
     durante
    1.05
     shut
    1.05
    rethe
    1.03
     COMFORT
    1.02
     help
    1.01
     admon
    0.99
    Activations Density 0.042%

    No Known Activations