© 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. 17192
    Prev
    Next
    INDEX
    Explanations

    moons

    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
    niec
    -0.06
     transitional
    -0.06
    odi
    -0.05
     seuraa
    -0.05
    igar
    -0.05
     transition
    -0.05
     Pett
    -0.05
    éĢļãĤĬãģ«
    -0.05
    ourg
    -0.05
    âĢ¦)
    -0.05
    POSITIVE LOGITS
    -watch
    0.06
    -REAL
    0.06
    ÑĨ
    0.06
    Watch
    0.06
    Ñģна
    0.05
    lexer
    0.05
    aes
    0.05
    isper
    0.05
    éĩĮåħĭ
    0.05
    ."+
    0.05
    Activations Density 0.001%

    No Known Activations