© 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. 20176
    Prev
    Next
    INDEX
    Explanations

    protocol header fields

    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
    ftime
    -0.06
    alisme
    -0.06
     Cata
    -0.06
    adis
    -0.06
    artin
    -0.06
    assic
    -0.06
    inage
    -0.06
    好çī©
    -0.06
    edback
    -0.06
     arrivato
    -0.06
    POSITIVE LOGITS
     lấy
    0.06
     Strategy
    0.05
    Helpers
    0.05
     laure
    0.05
    rÃŃ
    0.05
    常çĶ¨
    0.05
    chè
    0.05
     Alcal
    0.05
    çĶ
    0.05
    icot
    0.05
    Activations Density 0.001%

    No Known Activations