© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3.5-4B
    5. 15-RES-MATRYOSHKA-65K
    7. 19671
    Prev
    Next
    INDEX
    Explanations

    ", "refolding process", "Scaled", "Tusculum", "Vettel"

    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
     Exp
    -0.05
    eval
    -0.05
    åİŁçĶŁæĢģ
    -0.05
    ostr
    -0.05
    d
    -0.05
    lave
    -0.05
    ä¸Ļ
    -0.05
     emp
    -0.05
    eward
    -0.05
    comm
    -0.05
    POSITIVE LOGITS
    oci
    0.07
    åį³çĶ¨
    0.06
    otan
    0.06
     бажа
    0.06
    ableObject
    0.06
     Ñĥма
    0.06
    çĶ¨æĪ·ä½¿çĶ¨
    0.06
    以ä¸Ģ
    0.06
    Scaled
    0.06
    ulet
    0.06
    Activations Density 0.004%

    No Known Activations