© 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. 9540
    Prev
    Next
    INDEX
    Explanations

    di- prefix (two, double)

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

    chemical prefixes "di-" followed by various compound names.

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

    The neuron detects occurrences of the chemical prefix “di” (as in di-methyl, di-acetyl, di-pheny, etc.) 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_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
    or
    -2.70
     klassischen
    -2.66
     Даже
    -2.59
     You
    -2.56
    richlet
    -2.56
     There
    -2.55
     sogenannten
    -2.53
    abatic
    -2.52
     modernen
    -2.48
     Also
    -2.45
    POSITIVE LOGITS
    芣
    2.77
    蚍
    2.73
     sabah
    2.72
    铙
    2.63
     senin
    2.58
    有个
    2.48
    geſ
    2.48
    蟊
    2.45
     ISLAND
    2.45
    '^
    2.45
    Activations Density 0.035%

    No Known Activations