© 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. Qwen3.5-4B
    5. 15-RES-MATRYOSHKA-65K
    7. 21679
    Prev
    Next
    INDEX
    Explanations

    appeal from court

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    decoderesearch/qwen-3.5-saes/qwen-3.5-4b
    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
     '
    -0.06
    (
    -0.05
     J
    -0.05
     (
    -0.05
     "
    -0.05
     inject
    -0.05
    olk
    -0.05
    顺åĬ¿
    -0.05
     f
    -0.04
     O
    -0.04
    POSITIVE LOGITS
    åħļ代ä¼ļ
    0.07
     Purtroppo
    0.07
     inher
    0.07
    bele
    0.07
    ÙħاÙħ
    0.07
     nách
    0.07
    boxing
    0.07
    ettor
    0.06
    levelname
    0.06
     inherits
    0.06
    Activations Density 0.184%

    No Known Activations