© 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. 17765
    Prev
    Next
    INDEX
    Explanations

    false, least, punctuation

    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
     hag
    -0.06
     withholding
    -0.04
    uto
    -0.04
     Therefore
    -0.04
    åĽłæѤ
    -0.04
    åĮĹ
    -0.04
    æľīéĹ®é¢ĺ
    -0.04
    Pas
    -0.04
     Daher
    -0.04
    ffect
    -0.04
    POSITIVE LOGITS
     گا
    0.07
    stral
    0.07
    æ·±åħ¥åľ°
    0.06
    ارات
    0.06
    vel
    0.06
    atil
    0.06
     Cil
    0.06
     پرد
    0.06
    uble
    0.06
    OLA
    0.06
    Activations Density 0.005%

    No Known Activations