© 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. 22-GEMMASCOPE-RES-131K
    4. 37
    Prev
    Next
    INDEX
    Explanations

    chlorophenyl, hydroxyphenyl, methylphenyl

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

    chemical compound names and molecular structures in scientific literature.

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

    This neuron activates on chemical nomenclature fragments—tokens that are part of systematic compound names (e.g. “methyl,” “hydroxy,” “chlor,” “oxadiazole,” “butyrolactone,” 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
     raiſ
    -1.71
     magnificent
    -1.65
     captivating
    -1.63
    雠
    -1.63
     uniting
    -1.59
     startling
    -1.59
     unveils
    -1.57
    ᕮ
    -1.55
     astounding
    -1.54
    艄
    -1.54
    POSITIVE LOGITS
    Because
    2.11
    While
    2.06
    Usually
    1.96
    Since
    1.91
    Although
    1.84
     because
    1.84
    Shortly
    1.84
    When
    1.77
    During
    1.76
    just
    1.75
    Activations Density 0.102%

    No Known Activations