© 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. 12198
    Prev
    Next
    INDEX
    Explanations

    `if`/`else` blocks

    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
     célè
    -1.20
     cocok
    -1.15
     éviter
    -1.13
    ㄹ
    -1.11
    områ
    -1.10
    řez
    -1.08
    躯
    -1.07
     veden
    -1.05
     décret
    -1.05
     estão
    -1.04
    POSITIVE LOGITS
     even
    1.38
     if
    1.30
     just
    1.09
    else
    1.05
    実現
    1.03
    問題は
    1.02
     other
    1.02
     it
    1.02
     Substanz
    0.98
     else
    0.98
    Activations Density 0.001%

    No Known Activations