© 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
    3. Gemma-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 28434
    Prev
    Next
    INDEX
    Explanations

    song titles in parentheses

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

    The neuron fires strongly on the opening double‐quote token marking the start of a quoted title.

    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
    %;
    -1.05
    SpringRunner
    -1.03
     וק
    -1.00
     stratég
    -1.00
     vraiment
    -1.00
    ดอก
    -0.99
     男人
    -0.99
     animada
    -0.99
    பெ
    -0.96
     tatuaje
    -0.96
    POSITIVE LOGITS
     this
    1.26
     zarówno
    1.12
    Escrito
    1.08
    ★★★★
    1.04
    
    1.01
    様々な
    1.00
     >",
    0.98
    ★★★
    0.97
    Not
    0.97
     both
    0.97
    Activations Density 0.002%

    No Known Activations