© 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. 105978
    Prev
    Next
    INDEX
    Explanations

    medical and personal services

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

    medical and cosmetic procedures, particularly those ending in suffixes like "gram," "job," "lift," and "cure."

    oai_token-act-pair · claude-3-7-sonnet-20250219Triggered by @neilrathi

    The neuron strongly activates on compound nouns naming procedural “services” or “treatments” (e.g. mammograms, oil changes, pedicures, facials).

    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
     of
    -2.25
    f
    -1.86
    納豆
    -1.80
    b
    -1.73
     takto
    -1.54
    盦
    -1.53
     vertes
    -1.53
    これに
    -1.52
     to
    -1.51
    昆布
    -1.47
    POSITIVE LOGITS
     gebouw
    1.71
     sudadera
    1.60
    asjon
    1.55
    jadikan
    1.47
     安装
    1.47
    ?
    1.46
     dekoratif
    1.45
     capucha
    1.44
     chaqueta
    1.44
    After
    1.44
    Activations Density 0.093%

    No Known Activations