© 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. 34-GEMMASCOPE-RES-131K
    7. 32856
    Prev
    Next
    INDEX
    Explanations

    lumefantrine, trimethadione, ximelagatran

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

    pharmaceutical or chemical compound names, particularly those with "me" syllables.

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

    This neuron fires on any token containing the letter sequence “me.”

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_34/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
    dır
    -0.87
    อ
    -0.84
     gustaba
    -0.77
    lindung
    -0.76
    詢問
    -0.75
     gustó
    -0.73
     encantó
    -0.73
    niej
    -0.69
    stro
    -0.69
    langsung
    -0.69
    POSITIVE LOGITS
     Aç
    0.85
    lla
    0.84
    timos
    0.82
    iy
    0.82
    arked
    0.81
    trash
    0.81
    tieri
    0.80
    ーション
    0.80
     Fuerza
    0.79
    andering
    0.75
    Activations Density 0.077%

    No Known Activations