© 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. 90735
    Prev
    Next
    INDEX
    Explanations

    CBD and LGBT

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

    The neuron activates on occurrences of the uppercase acronym “CBD.”

    oai_token-act-pair · o4-miniTriggered by @jyhe0408

    a specific three-letter all-caps acronym appearing across diverse contexts (medical, cannabis, anatomy, urban areas), with especially strong activation whenever that recurring abbreviation occurs.

    oai_token-act-pair · gpt-5Triggered by @jyhe0408

    the abbreviation "CBD" (referring to either cannabidiol or Central Business District).

    oai_token-act-pair · claude-4-5-sonnetTriggered 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
    ⚋
    -3.33
    蕻
    -2.72
    媜
    -2.47
    h
    -2.47
     aseg
    -2.44
    ocyan
    -2.39
    弖
    -2.39
    </h3>
    -2.39
    1
    -2.34
    
    -2.30
    POSITIVE LOGITS
    -
    2.86
    I
    2.50
     were
    2.41
    ка
    2.39
    ”？
    2.34
    *
    2.11
     kutoka
    2.08
    つつ
    2.08
     dahulu
    2.06
    to
    2.05
    Activations Density 0.001%

    No Known Activations