© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW
    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    2. Gemma-2-27B
    3. 22-GEMMASCOPE-RES-131K
    4. 65073
    Prev
    Next
    INDEX
    Explanations

    drug and alcohol services

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

    This neuron identifies professional fields and institutional sectors such as "Drug and Alcohol," "Health and Safety," "Air and Space," "Goods and Services," "Arts and Science."

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

    The neuron fires strongly on the second noun in coordinated “X and Y” phrases (e.g. “drug and alcohol,” “health and safety,” “oil and gas,” 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
     and
    -2.11
    えっ
    -2.05
    来週
    -1.85
    にほんブログ村
    -1.83
    もうすぐ
    -1.78
    先週
    -1.73
    あぁ
    -1.71
    すみません
    -1.71
    つづく
    -1.69
    ***
    -1.67
    POSITIVE LOGITS
    <bos>
    1.92
    があり
    1.58
     of
    1.55
    ,
    1.53
    と思っていた
    1.52
    そして
    1.48
    と思っている
    1.47
    しかし
    1.45
    ってこと
    1.38
    ただし
    1.37
    Activations Density 0.075%

    No Known Activations