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

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 15654
    Prev
    Next
    INDEX
    Explanations

    introducing causal clauses

    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
     especially
    -1.32
     would
    -1.24
     particularly
    -1.06
     remarkable
    -1.05
     including
    -1.02
     then
    -1.02
     give
    -0.98
     muddy
    -0.98
     that
    -0.98
     focusing
    -0.97
    POSITIVE LOGITS
    optimized
    1.05
     APPLICATIONS
    1.02
    就没
    1.00
     SIMULATION
    0.99
    vertissement
    0.99
     иннова
    0.96
    ඞ
    0.96
     STRATEG
    0.94
    部分が
    0.93
     తె
    0.93
    Activations Density 0.075%

    No Known Activations