© 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. 18191
    Prev
    Next
    INDEX
    Explanations

    quotation marks around words

    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
    culus
    -0.07
    udder
    -0.06
    odal
    -0.06
    amel
    -0.06
    letic
    -0.05
    èĩĬ
    -0.05
    缸åħ³éĵ¾æİ¥
    -0.05
    asal
    -0.05
    yam
    -0.05
    AST
    -0.05
    POSITIVE LOGITS
     i
    0.05
    æķ´æķ´
    0.05
    çĶ
    0.05
    ulk
    0.05
    ulla
    0.05
    mont
    0.05
    -i
    0.05
    ä¸ĺ
    0.05
     Magazine
    0.05
    awat
    0.05
    Activations Density 0.019%

    No Known Activations