© 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
    2. Gemma-2-27B
    3. 22-GEMMASCOPE-RES-131K
    4. 9345
    Prev
    Next
    INDEX
    Explanations

    prepositions and conjunctions

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

    The neuron chiefly responds to high‐frequency relational prepositions (e.g. “from,” “on,” “with,” “through”) marking connections or means 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_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
     of
    -1.43
    iconque
    -1.16
    那
    -1.10
     behövs
    -1.03
     
    -1.02
     outright
    -1.01
    нау
    -1.00
     to
    -0.98
    мене
    -0.98
     där
    -0.98
    POSITIVE LOGITS
     during
    1.20
     when
    1.17
     znan
    1.16
    uszka
    1.09
     móg
    1.09
    rzu
    1.07
    いかがでしたか
    1.06
     zamiast
    1.05
     garansi
    1.03
     dirilis
    1.03
    Activations Density 0.024%

    No Known Activations