© 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. 10-GEMMASCOPE-RES-131K
    4. 84992
    Prev
    Next
    INDEX
    Explanations

    contrastive conjunctions like but or although

    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_10/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
    をは
    -1.52
     for
    -1.48
    caucasian
    -1.45
    lampen
    -1.44
     ผู้หญิง
    -1.43
    blusa
    -1.38
    の裏
    -1.38
    でしたね
    -1.36
     çocuklar
    -1.30
    incer
    -1.29
    POSITIVE LOGITS
     but
    1.91
     also
    1.82
    but
    1.79
    is
    1.56
    mu
    1.55
    and
    1.53
    而
    1.52
    re
    1.51
    ins
    1.51
    os
    1.47
    Activations Density 0.016%

    No Known Activations