© 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. 1555
    Prev
    Next
    INDEX
    Explanations

    methyl chemical compounds

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

    chemical compound structures, particularly focusing on molecules with "methyl" and "hydroxy" groups.

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

    This neuron detects chemical substituent terms (e.g. functional‐group descriptors like hydroxy, methyl, amino) in compound 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
     to
    -2.38
    m
    -1.84
    u
    -1.79
    ‐
    -1.69
     that
    -1.65
     o
    -1.63
     mai
    -1.52
    il
    -1.52
    ut
    -1.52
     &
    -1.47
    POSITIVE LOGITS
    selben
    1.81
    這些
    1.75
    砣
    1.68
    👲
    1.68
     tekanan
    1.66
     Pastikan
    1.64
    🩻
    1.64
     mercenaries
    1.63
    culado
    1.63
    根據
    1.63
    Activations Density 0.084%

    No Known Activations