© 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. 34-GEMMASCOPE-RES-131K
    7. 34757
    Prev
    Next
    INDEX
    Explanations

    unexpected conjunctions leading to surprise

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_34/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
    咳咳
    -0.94
     București
    -0.94
    tanian
    -0.88
    љу
    -0.81
    哈哈哈哈
    -0.81
    
    -0.81
     duration
    -0.78
    喏
    -0.77
     tiks
    -0.77
    faßt
    -0.77
    POSITIVE LOGITS
     boy
    3.25
     wow
    2.80
     oh
    2.55
     Wow
    2.47
     holy
    2.45
     WOW
    2.38
     OMG
    2.30
    Wow
    2.20
     man
    2.19
    wow
    2.17
    Activations Density 0.048%

    No Known Activations