© 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. 67308
    Prev
    Next
    INDEX
    Explanations

    reporting speech and statements

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

    The neuron detects reporting or attribution verbs (e.g. “said,” “told,” “wrote”) that mark quoted speech or sourced statements.

    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
     إدارة
    -1.16
     และ
    -1.13
    fficacy
    -1.09
     dafür
    -1.07
    および
    -1.06
    到
    -1.03
    atimes
    -1.03
     {
    
    -1.02
     Bürgermeister
    -1.01
     и
    -1.00
    POSITIVE LOGITS
     without
    1.24
     said
    1.21
     in
    1.16
     on
    1.15
    最新章节
    1.13
    神话
    1.09
    它們
    1.09
    Liet
    1.07
    Текст
    1.03
    业绩
    1.02
    Activations Density 0.020%

    No Known Activations