© 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. 34-GEMMASCOPE-RES-131K
    7. 66187
    Prev
    Next
    INDEX
    Explanations

    prepositional phrases defining boundaries or content

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

    The neuron is strongly activated by small, high‐frequency “function” words in code and comments (prepositions, conjunctions, articles, etc., e.g. if, than, for, the, from).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_34/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
     них
    -0.88
    쨰
    -0.88
     They
    -0.85
     takich
    -0.85
     자신의
    -0.80
     隆
    -0.79
     sua
    -0.79
     MANY
    -0.79
    Élet
    -0.79
     REALLY
    -0.78
    POSITIVE LOGITS
     ‘
    1.44
    『
    1.34
     `
    1.34
     provided
    1.21
    《
    1.16
     '
    1.16
    paramref
    1.09
     ‹
    1.07
     given
    1.05
     indicated
    1.03
    Activations Density 0.049%

    No Known Activations