© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it
    Neuronpedia
    Jacobian LensNEW
    Natural Language
    Autoencoders
    MAY 26    Assistant AxisJAN 26Circuit TracerAPR 26SteerSAE EvalsExportsinterp-engineNEW Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-4-31B
    5. 30-RES-MATRYOSHKA-131K
    7. 121038
    Prev
    Next
    INDEX
    Explanations

    observing the process

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    decoderesearch/gemma-4-saes/gemma-4-31b
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Negative Logits
     Image
    -0.07
     IMAGE
    -0.07
    convén
    -0.06
     mainland
    -0.06
    paration
    -0.06
     imaginary
    -0.06
    图片
    -0.06
     image
    -0.06
     этих
    -0.06
     vertebrate
    -0.06
    POSITIVE LOGITS
     watching
    0.10
     monitoring
    0.09
     observing
    0.09
    観察
    0.09
     Observing
    0.08
     beobachten
    0.08
     beob
    0.08
     watch
    0.08
     monitored
    0.08
    watching
    0.08
    Activations Density 0.047%

    No Known Activations