© 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. 118981
    Prev
    Next
    INDEX
    Explanations

    numbered list item

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

    The neuron detects the closing parenthesis in numbered list markers (e.g. the “)” following “1”, “2”, 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
    .
    -3.05
     or
    -2.91
     undeniably
    -2.48
     possibly
    -2.47
     oddly
    -2.42
     as
    -2.41
     intensely
    -2.34
     tailored
    -2.33
     three
    -2.31
     nine
    -2.31
    POSITIVE LOGITS
    訁
    3.03
    袿
    2.98
    
    2.91
     VERY
    2.83
    樒
    2.81
     unſ
    2.75
    ፨
    2.69
    IBLIO
    2.67
     !!!!
    2.61
     dieß
    2.59
    Activations Density 0.012%

    No Known Activations