© 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. 22-GEMMASCOPE-RES-131K
    7. 71250
    Prev
    Next
    INDEX
    Explanations

    prepositions or quantifiers

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

    The neuron is highly responsive to spatial or relational prepositions (like “in,” “into,” “to,” “between,” “onto,” etc.) indicating location or direction.

    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
     from
    -1.02
     if
    -1.02
     to
    -1.00
     lel
    -0.98
     apartament
    -0.96
     of
    -0.96
     “
    -0.94
     also
    -0.88
     TUT
    -0.87
     Godt
    -0.86
    POSITIVE LOGITS
     anything
    1.50
     everything
    1.38
     humans
    1.22
     coû
    1.16
     something
    1.15
     étroite
    1.14
    ต่างๆ
    1.13
    ionine
    1.12
     things
    1.11
     различных
    1.11
    Activations Density 0.144%

    No Known Activations