© 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. 34-GEMMASCOPE-RES-131K
    7. 117724
    Prev
    Next
    INDEX
    Explanations

    saying or responding

    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_34/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
     either
    -1.04
     because
    -0.98
    が含ま
    -0.90
    みました
    -0.90
     tampaknya
    -0.89
    因為
    -0.89
    因为
    -0.89
     jedną
    -0.88
     which
    -0.88
     Überblick
    -0.88
    POSITIVE LOGITS
     saying
    4.72
    saying
    3.44
     sayin
    2.84
     Saying
    2.75
    Saying
    2.72
     say
    2.61
     diciendo
    2.56
    说
    2.27
     dizendo
    2.23
     сказать
    2.22
    Activations Density 0.027%

    No Known Activations