© 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. 17753
    Prev
    Next
    INDEX
    Explanations

    you must/need/can

    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
    apprend
    -0.08
    trup
    -0.07
    trato
    -0.07
    सà¥įप
    -0.07
    inel
    -0.07
    esel
    -0.07
    tiz
    -0.07
    éļħ
    -0.07
    obil
    -0.07
    ")+
    -0.07
    POSITIVE LOGITS
    H
    0.06
    Âł
    0.06
    ç¼ĺ
    0.05
    ance
    0.05
    ↵
    0.05
    0.05
    eth
    0.05
    æīĵæī°
    0.05
    forth
    0.05
    A
    0.05
    Activations Density 0.001%

    No Known Activations