© 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. 57868
    Prev
    Next
    INDEX
    Explanations

    best thing

    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
    atical
    -0.10
    æĴ®
    -0.09
    çĽ¸ä¼ł
    -0.09
    éľ²åĩº
    -0.08
    éĥ¢
    -0.08
    æĢ»çIJĨ
    -0.08
    å¹³åĩ¡
    -0.08
    çļĦæĥħ
    -0.08
    _UNDEF
    -0.08
    passes
    -0.08
    POSITIVE LOGITS
    iset
    0.10
    iar
    0.09
     feature
    0.09
    aits
    0.09
    åľ¨äºİ
    0.09
    å®Įæķ´
    0.08
    追éĢIJ
    0.08
    ises
    0.08
    åĽ½åĬ¡
    0.08
    ä¸ĢæŃ¥
    0.08
    Activations Density 0.018%

    No Known Activations