© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-4-31B
    5. 30-RES-MATRYOSHKA-131K
    7. 118450
    Prev
    Next
    INDEX
    Explanations

    likely, probably, possibly

    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
    早就
    -0.06
    在她
    -0.06
     biggest
    -0.06
     longstanding
    -0.06
     воспользоваться
    -0.06
     olmaz
    -0.06
     Food
    -0.06
    びっくり
    -0.05
    全く
    -0.05
    风险
    -0.05
    POSITIVE LOGITS
     provavelmente
    0.08
     probabilmente
    0.07
     вероятно
    0.07
     posiblemente
    0.06
     presumably
    0.06
    presumably
    0.06
    友
    0.06
     prawdopod
    0.06
     probablemente
    0.06
    应该是
    0.06
    Activations Density 0.048%

    No Known Activations