© 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. 17000
    Prev
    Next
    INDEX
    Explanations

    .map and .forEach

    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
    ữ
    -0.06
    æĥ³è±¡ä¸ŃçļĦ
    -0.05
     affecting
    -0.05
    éĥİ
    -0.05
    å±ŀäºİèĩªå·±çļĦ
    -0.05
    Ñħла
    -0.05
    éĻIJåĪ¶
    -0.05
    驼
    -0.05
     emerging
    -0.05
    мп
    -0.05
    POSITIVE LOGITS
     Akron
    0.07
    dalena
    0.06
    eho
    0.06
    asper
    0.06
    éĢĻæīį
    0.06
     devrez
    0.06
    culus
    0.06
    zial
    0.05
     bes
    0.05
    eiten
    0.05
    Activations Density 0.003%

    No Known Activations