© 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. 21051
    Prev
    Next
    INDEX
    Explanations

    punctuation or sentence connectors

    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
    å¾Ĺ太
    -0.05
    åĽłä¸ºæĪij们
    -0.05
    æīįæĺ¯çľŁæŃ£çļĦ
    -0.05
     viz
    -0.05
    viz
    -0.05
    isola
    -0.05
     allot
    -0.05
    ai
    -0.05
    ff
    -0.05
     Called
    -0.05
    POSITIVE LOGITS
     concentrated
    0.07
    åİ»ä½ķ
    0.06
     additionally
    0.06
    åı¯èĥ½åĩºçİ°
    0.06
    åı¯èĥ½ä¼ļæľī
    0.06
    glement
    0.06
     possibly
    0.06
    elain
    0.06
    kees
    0.06
    teur
    0.06
    Activations Density 0.075%

    No Known Activations