© 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. 10-GEMMASCOPE-RES-131K
    7. 15414
    Prev
    Next
    INDEX
    Explanations

    making small adjustments

    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_10/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
     But
    -3.14
     Then
    -2.66
     pitfalls
    -2.28
     They
    -2.22
     SOME
    -2.20
    s
    -2.16
     Such
    -2.16
     \
    -2.14
    </h3>
    -2.14
     Sufficient
    -2.11
    POSITIVE LOGITS
    橚
    2.58
     bedeckt
    2.34
     hacían
    2.30
    絰
    2.30
     碗
    2.30
    umán
    2.30
    Lokasi
    2.27
    ‪.‬‬
    2.22
    逶
    2.22
     gegangen
    2.20
    Activations Density 0.004%

    No Known Activations