© 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. 21273
    Prev
    Next
    INDEX
    Explanations

    break, leave

    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
    irn
    -0.06
    haar
    -0.06
     se
    -0.05
    pts
    -0.05
    иÑģк
    -0.05
    lemn
    -0.05
    heu
    -0.05
    ek
    -0.05
    å°ıå¹³
    -0.05
    æİIJ
    -0.05
    POSITIVE LOGITS
    umuman
    0.06
    eza
    0.06
     [].
    0.05
    ä¸įå½ĵ
    0.05
    erness
    0.05
     Wouldn
    0.05
    èĩªæľīèµĦéĩij
    0.05
     profumo
    0.05
    ETTI
    0.05
    ucher
    0.05
    Activations Density 0.001%

    No Known Activations