© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 12568
    Prev
    Next
    INDEX
    Explanations

    self-reference

    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
     três
    -1.17
    皷
    -1.17
    ことなく
    -1.15
     unsettling
    -1.13
    氿
    -1.11
     amazingly
    -1.08
     insanely
    -1.06
     REALLY
    -1.05
    方で
    -1.05
     VERY
    -1.04
    POSITIVE LOGITS
     and
    1.94
     yourself
    1.55
    self
    1.45
     or
    1.28
     себя
    1.27
     selves
    1.18
    敍
    1.15
    изнь
    1.13
    and
    1.12
    yourself
    1.11
    Activations Density 0.016%

    No Known Activations