© 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. 10-GEMMASCOPE-RES-131K
    7. 74792
    Prev
    Next
    INDEX
    Explanations

    groups of people describing actions

    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
    zwei
    -1.48
     obicei
    -1.48
    躂
    -1.47
    stdc
    -1.42
     veranstal
    -1.40
    _));
    -1.38
    igshid
    -1.37
     vecka
    -1.36
    ción
    -1.34
     består
    -1.34
    POSITIVE LOGITS
     who
    2.19
     that
    2.14
     will
    1.93
     then
    1.79
     "
    1.66
    who
    1.66
     which
    1.58
     whom
    1.55
     don
    1.41
     evidentemente
    1.38
    Activations Density 0.158%

    No Known Activations