© 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. 18381
    Prev
    Next
    INDEX
    Explanations

    candidate 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
    umbed
    -0.07
    adar
    -0.06
    rai
    -0.06
     пÑĢивед
    -0.05
    roj
    -0.05
    utherland
    -0.05
    aky
    -0.05
    оÑĤÑĮ
    -0.05
    usty
    -0.05
    æĭĿ
    -0.05
    POSITIVE LOGITS
    串
    0.06
    æ°ij主
    0.06
     halluc
    0.06
     جÙħا
    0.06
    candidate
    0.06
     candidates
    0.06
    走失
    0.05
    çĶ¨äºİ
    0.05
     candidate
    0.05
    adm
    0.05
    Activations Density 0.009%

    No Known Activations