© 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. 19324
    Prev
    Next
    INDEX
    Explanations

    large to expert

    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
     instead
    -0.07
    èĢĮæĺ¯åľ¨
    -0.06
    format
    -0.06
    instead
    -0.05
     Instead
    -0.05
     zami
    -0.05
    en
    -0.05
     rather
    -0.05
     invece
    -0.05
    .mock
    -0.05
    POSITIVE LOGITS
     '\"
    0.07
    å¤Ħéķ¿
    0.07
    sstr
    0.06
    èķ¾
    0.06
    fffffff
    0.06
    ,'\
    0.06
    sWith
    0.06
    ormais
    0.06
    ÑģÑĤÑĥпа
    0.06
     restric
    0.06
    Activations Density 0.062%

    No Known Activations