© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 11393
    Prev
    Next
    INDEX
    Explanations

    but, however, and other contrast words

    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_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
     veiks
    -0.81
     All
    -0.79
    couragement
    -0.78
     Grundlagen
    -0.75
     based
    -0.74
     jPanel
    -0.73
     signed
    -0.71
     is
    -0.71
    门外
    -0.71
     can
    -0.70
    POSITIVE LOGITS
     but
    1.95
     nhưng
    1.33
     אך
    1.19
     แต่
    1.19
     لكن
    1.13
    지만
    1.11
    etheless
    1.09
     però
    1.06
     however
    1.05
     लेकिन
    1.05
    Activations Density 0.106%

    No Known Activations