© 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. 75263
    Prev
    Next
    INDEX
    Explanations

    some degree or extent

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

    The neuron specializes in spotting vague quantifiers and degree indicators—words that express indeterminate amounts or extents (e.g. “some,” “part,” “bit,” “degree,” “extent,” etc.).

    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
     pratiquement
    -1.19
     exactement
    -1.13
     eccell
    -1.12
     ajuns
    -1.07
    gewöhn
    -1.07
    笤
    -1.05
     lesquels
    -1.05
    拊
    -1.02
     précisément
    -1.02
     红
    -1.02
    POSITIVE LOGITS
     some
    1.70
     algunos
    1.44
     SOME
    1.41
    บ้าง
    1.37
     at
    1.36
     but
    1.36
     aunque
    1.26
     or
    1.26
     некоторые
    1.23
     of
    1.20
    Activations Density 0.060%

    No Known Activations