© 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. Qwen3.5-4B
    5. 15-RES-MATRYOSHKA-65K
    7. 19764
    Prev
    Next
    INDEX
    Explanations

    following formula

    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
    ianti
    -0.06
    ouz
    -0.06
    óz
    -0.06
     Tua
    -0.06
    óf
    -0.06
    hofer
    -0.06
    æĬķèµĦæľºä¼ļ
    -0.06
     Pry
    -0.06
     Aks
    -0.05
    -0.05
    POSITIVE LOGITS
     net
    0.06
    åĩĢæ°´
    0.06
    æĹ©æľī
    0.05
    heet
    0.05
     v
    0.05
    asted
    0.05
    caught
    0.05
     bodies
    0.05
     ideal
    0.05
    gregated
    0.05
    Activations Density 0.004%

    No Known Activations