© 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. 34388
    Prev
    Next
    INDEX
    Explanations

    theory of

    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
    æĵĺ
    -0.10
    èķ²
    -0.10
    襲
    -0.09
    æµı
    -0.09
    åĪ©æ¶¦çİĩ
    -0.09
    ron
    -0.09
    bridge
    -0.08
    agle
    -0.08
    æ³½
    -0.08
    ãģĦãģ¡
    -0.08
    POSITIVE LOGITS
    éĽĴ
    0.10
    å·¥ä½ľèĢħ
    0.09
    ä¸ŃçļĦ
    0.09
    æ¶Įçİ°
    0.09
    fork
    0.09
    åĶ¯ä¸ĢçļĦ
    0.09
    erg
    0.08
    _GC
    0.08
     ramifications
    0.08
    dsp
    0.08
    Activations Density 0.329%

    No Known Activations