© 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. 104728
    Prev
    Next
    INDEX
    Explanations

    quotes followed by speech

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

    The neuron detects spans of direct speech or quoted text, i.e. dialogue enclosed in quotation marks.

    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
     alarming
    -0.97
     October
    -0.96
    ほうがいい
    -0.94
     out
    -0.94
     on
    -0.93
     bereits
    -0.92
     k
    -0.92
     July
    -0.91
     seiner
    -0.91
     подчер
    -0.91
    POSITIVE LOGITS
     also
    1.50
    También
    1.30
     البته
    1.30
    konstruktion
    1.25
     furthermore
    1.21
    شنا
    1.20
     حتی
    1.19
     moreover
    1.18
     サッカー
    1.16
     δὲ
    1.13
    Activations Density 0.005%

    No Known Activations