© 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. 20120
    Prev
    Next
    INDEX
    Explanations

    inspect, iterate, check, wrap

    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
    gül
    -0.07
    achsen
    -0.07
    armee
    -0.06
    ulun
    -0.06
    èī²åĪĹ
    -0.06
     crowds
    -0.06
    achs
    -0.06
    aisse
    -0.06
    czes
    -0.06
    eous
    -0.06
    POSITIVE LOGITS
     pie
    0.05
    çĽĬ
    0.05
    oy
    0.05
    fo
    0.05
    泪
    0.05
    -
    0.05
    pie
    0.05
    åĬ³åĬ¨
    0.05
    h
    0.04
    adu
    0.04
    Activations Density 0.005%

    No Known Activations