© 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. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 47985
    Prev
    Next
    INDEX
    Explanations

    assigning blame

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    adamkarvonen/qwen3-32b-saes/saes_Qwen_Qwen3-32B_batch_top_k/resid_post_layer_32
    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
    egin
    -0.09
    éĻªä½ł
    -0.09
    ogle
    -0.08
    Vision
    -0.08
    éĽªå±±
    -0.08
     imper
    -0.08
    _rhs
    -0.08
    _Right
    -0.08
    1
    -0.08
    è¡¥
    -0.08
    POSITIVE LOGITS
    åĴİ
    0.15
    èĩªå·±
    0.13
    èĩªè¡Į
    0.12
    èĩªæĦ¿
    0.12
    èĩªèº«
    0.12
    èĩªå·±åĸľæ¬¢
    0.11
    èĩªèº«çļĦ
    0.10
    èĩªçIJĨ
    0.10
    èĩªåĪĨãģ§
    0.10
     selber
    0.10
    Activations Density 0.075%

    No Known Activations