© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 39046
    Prev
    Next
    INDEX
    Explanations

    have to admit

    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
    我們
    -1.26
    ént
    -1.22
     rendus
    -1.20
     ayudan
    -1.14
     diinginkan
    -1.13
     münchen
    -1.12
    tarta
    -1.11
    attitude
    -1.11
     any
    -1.10
     dauern
    -1.09
    POSITIVE LOGITS
     say
    1.31
    They
    1.13
    omiast
    1.08
    u
    1.01
    %).
    0.99
    amilya
    0.99
    so
    0.97
    os
    0.97
    thma
    0.96
    hew
    0.94
    Activations Density 0.049%

    No Known Activations