© 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. 10-GEMMASCOPE-RES-131K
    4. 105417
    Prev
    Next
    INDEX
    Explanations

    roman numeral i in lists

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

    The neuron detects enumeration markers, especially lowercase Roman‐numeral list labels in parentheses (e.g. “(i)”, “(ii)”, 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_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
     it
    -2.20
     didn
    -1.80
     where
    -1.80
     gave
    -1.80
     took
    -1.74
     with
    -1.71
     did
    -1.71
     this
    -1.70
     was
    -1.66
     had
    -1.65
    POSITIVE LOGITS
    jeeling
    1.65
    akade
    1.55
     libido
    1.48
     हैं
    1.47
     Volvo
    1.41
    ჻
    1.39
    놋
    1.39
     BOTH
    1.38
     koncer
    1.38
    deauna
    1.38
    Activations Density 0.031%

    No Known Activations