© 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. 30227
    Prev
    Next
    INDEX
    Explanations

    able to

    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
    enton
    -0.08
    sj
    -0.08
    sic
    -0.08
    Beat
    -0.08
    纳
    -0.08
    xp
    -0.08
    /weather
    -0.08
    åĩłåĪĨ
    -0.08
     TED
    -0.08
    im
    -0.08
    POSITIVE LOGITS
    oldt
    0.10
    otten
    0.10
    åħļæł¡
    0.10
    åĽ¿
    0.10
     Unable
    0.09
    ä¹ŁæĹłæ³ķ
    0.09
    unable
    0.09
    _VARS
    0.09
    æĺ¯ä¸įåı¯èĥ½
    0.09
     inability
    0.09
    Activations Density 0.026%

    No Known Activations