© 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. 10-GEMMASCOPE-RES-131K
    7. 75226
    Prev
    Next
    INDEX
    Explanations

    but with

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

    The neuron strongly activates on occurrences of the preposition “with.”

    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
    l
    -2.02
    us
    -1.88
     powiedzieć
    -1.78
    in
    -1.66
    to
    -1.61
    because
    -1.55
     restauration
    -1.52
    on
    -1.52
    p
    -1.48
     robić
    -1.48
    POSITIVE LOGITS
     a
    2.27
    onenumber
    1.96
     nueva
    1.71
    ángulo
    1.62
    \&
    1.62
    źdz
    1.61
    ࿆
    1.56
     bringen
    1.49
     ſy
    1.48
     FontWeight
    1.48
    Activations Density 0.031%

    No Known Activations