© 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. 79981
    Prev
    Next
    INDEX
    Explanations

    news agency reports

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

    The neuron detects journalistic metadata—e.g. news-agency bylines, dateline tokens (locations, dates), and reporting verbs.

    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
     could
    -1.21
     what
    -1.10
     can
    -1.09
     sizable
    -1.05
    Zalety
    -1.02
     their
    -1.02
     because
    -1.01
     undeniably
    -1.00
     our
    -0.98
    Dlaczego
    -0.97
    POSITIVE LOGITS
     rapporter
    1.13
    AGAIN
    1.09
     patienter
    1.06
     rehabilitación
    1.05
    ΄
    1.03
     horloge
    1.02
    有一个
    1.02
     broek
    1.01
     ingresar
    0.99
     различных
    0.99
    Activations Density 0.004%

    No Known Activations