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. 2
    Prev
    Next
    INDEX
    Explanations

    Exercise and place names

    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
     patienter
    -1.02
     visse
    -0.96
    刻意
    -0.95
     terminer
    -0.93
    Şi
    -0.92
     konden
    -0.91
    快要
    -0.90
     dreary
    -0.90
     oft
    -0.90
     restent
    -0.89
    POSITIVE LOGITS
     aptly
    1.13
     after
    1.12
     almost
    1.09
     immensely
    1.07
     amusing
    1.06
     seemingly
    1.05
    انيه
    1.05
     completely
    1.02
     immediately
    1.01
     tremendously
    1.01
    Activations Density 0.105%

    No Known Activations