© 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. 10974
    Prev
    Next
    INDEX
    Explanations

    phenyl groups in chemical names

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

    chemical compound names containing "phenyl" in scientific literature.

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

    The neuron strongly activates on occurrences of the substring “phenyl,” i.e. phenyl‐containing chemical 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_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
    者は
    -1.53
     vitesses
    -1.49
     extré
    -1.41
    真的
    -1.40
     feront
    -1.39
     vendre
    -1.38
    
    -1.38
    ntesis
    -1.34
     amazing
    -1.33
    蟜
    -1.32
    POSITIVE LOGITS
    </h2>
    1.78
    since
    1.45
    Less
    1.40
     
    1.30
    mycompany
    1.29
    種類
    1.27
    self
    1.25
     عدة
    1.23
     Pág
    1.23
     می‌توان
    1.22
    Activations Density 0.089%

    No Known Activations