© 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
    2. Gemma-2-27B
    3. 10-GEMMASCOPE-RES-131K
    4. 75065
    Prev
    Next
    INDEX
    Explanations

    drugs and controlled substances

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

    names of drugs and substances of abuse, especially stimulants, opioids, and psychoactive substances.

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

    The main thing this neuron does is detect mentions of specific psychoactive or illicit drug names.

    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
    c
    -1.38
     with
    -1.24
     torino
    -1.23
    ナの
    -1.23
    ut
    -1.23
    h
    -1.23
     reflects
    -1.19
    ?
    -1.18
    ne
    -1.18
    ca
    -1.16
    POSITIVE LOGITS
     Geografía
    1.43
    廡
    1.42
     wiedzieć
    1.39
    ญี่ป
    1.38
    など
    1.38
     bemerken
    1.37
    phany
    1.36
     powiedzieć
    1.33
    attere
    1.32
     conferencias
    1.31
    Activations Density 0.081%

    No Known Activations