Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Gemma-2-27B
    3. 34-GEMMASCOPE-RES-131K
    4. 210
    Prev
    Next
    INDEX
    Explanations

    dialogue or asking questions

    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
    咂
    -1.27
     souvent
    -1.17
     quelquefois
    -1.14
    Often
    -1.05
     fidé
    -1.05
    点了点头
    -1.04
     Rarely
    -1.02
     rigue
    -1.02
    totta
    -1.02
    往往
    -1.02
    POSITIVE LOGITS
     ask
    1.88
     asks
    1.80
     tells
    1.68
     tell
    1.67
     explain
    1.66
     wonder
    1.54
     suggest
    1.45
     mention
    1.42
     remark
    1.38
     assure
    1.36
    Activations Density 0.019%

    No Known Activations