© 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-4-31B
    5. 30-RES-MATRYOSHKA-131K
    7. 129558
    Prev
    Next
    INDEX
    Explanations

    wondering or remembering past events

    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
    Not in Any Lists

    No Comments

    Negative Logits
    emotion
    -0.07
     after
    -0.07
     subsequently
    -0.07
    emotional
    -0.06
     setelah
    -0.06
    似乎
    -0.06
     после
    -0.06
     seems
    -0.06
    feeling
    -0.06
     এখান
    -0.06
    POSITIVE LOGITS
     vergang
    0.07
     과거
    0.06
    上次
    0.06
     bygone
    0.06
     hardships
    0.06
     %,
    0.06
     vorige
    0.06
    WIDE
    0.06
     risques
    0.06
     previous
    0.06
    Activations Density 0.015%

    No Known Activations