© 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. 119020
    Prev
    Next
    INDEX
    Explanations

    preferring to enable

    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
     опас
    -0.07
    断面
    -0.07
     తీవ్ర
    -0.07
     severity
    -0.07
    severity
    -0.07
     ненави
    -0.07
     опасность
    -0.06
     punitive
    -0.06
     جنوب
    -0.06
    alerts
    -0.06
    POSITIVE LOGITS
     prefer
    0.07
     preferred
    0.07
     facilitating
    0.07
    穩
    0.07
     invigor
    0.07
     preferably
    0.07
     enabling
    0.07
     giving
    0.07
     சிறந்த
    0.07
     prerequisite
    0.06
    Activations Density 0.022%

    No Known Activations