© 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-3-27B
    3. 40-GEMMASCOPE-2-RES-16K
    4. 11744
    Prev
    Next
    INDEX
    Explanations

    but

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

    contrast indicating conjunction

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-2-27b-pt/resid_post/layer_40_width_16k_l0_medium
    Prompts (Dashboard)
    392,802 prompts, 256 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     Allerdings
    0.56
     cependant
    0.45
     allerdings
    0.39
    первых
    0.38
     비롯
    0.38
     však
    0.38
     भने
    0.37
    ۓ
    0.37
    gebnisse
    0.36
     HOWEVER
    0.35
    POSITIVE LOGITS
     but
    2.84
    แต่
    2.52
     nhưng
    2.47
    但不
    2.39
    but
    2.30
     mutta
    2.23
     pero
    2.16
     लेकिन
    2.14
     αλλά
    2.13
    但
    2.08
    Activations Density 0.584%

    No Known Activations