© 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. 26110
    Prev
    Next
    INDEX
    Explanations

    code comments

    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
    もうすぐ
    -2.08
     reager
    -1.74
     maniere
    -1.72
    前回の
    -1.65
     brosse
    -1.61
    方がいい
    -1.60
     vocale
    -1.59
    Tidak
    -1.55
    明日は
    -1.54
    をしました
    -1.53
    POSITIVE LOGITS
     to
    1.84
     is
    1.63
     l
    1.39
     J
    1.38
     of
    1.34
     '
    1.33
     R
    1.30
     &
    1.29
     other
    1.26
    ...
    1.24
    Activations Density 0.069%

    No Known Activations