© 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. 39778
    Prev
    Next
    INDEX
    Explanations

    and, or conjunctions

    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
    æĪĸèĢħåħ¶ä»ĸ
    -0.09
    etc
    -0.09
    jur
    -0.08
    çŃīçŃī
    -0.08
    aka
    -0.08
    éľĦ
    -0.08
    ä»Ģä¹Īåľ°æĸ¹
    -0.08
    æĮļ
    -0.08
    or
    -0.08
    gv
    -0.08
    POSITIVE LOGITS
    ä¿İ
    0.10
    缯
    0.10
    以èī²
    0.10
    èĭŁ
    0.09
    [at
    0.09
    itos
    0.09
    åĪĨæĺİ
    0.09
    线ä¸Ĭ线ä¸ĭ
    0.09
    çĭİ
    0.08
    åĿĤ
    0.08
    Activations Density 0.423%

    No Known Activations