© 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. 34795
    Prev
    Next
    INDEX
    Explanations

    form submission and inquiry

    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
     Belmont
    -0.10
    èĬĴ
    -0.09
    屦
    -0.09
    osh
    -0.08
    Bell
    -0.08
    èµĦæºIJåħ±äº«
    -0.08
    åī
    -0.08
    indicator
    -0.08
    åħ¬åŃIJ
    -0.08
    èѬ
    -0.08
    POSITIVE LOGITS
    mas
    0.10
    éķ¿çĽ¸
    0.10
    çĽ¸äº²
    0.09
    æķ£åıij
    0.09
     ÑĦоÑĢме
    0.09
    ä»»ä½ķå½¢å¼ı
    0.09
    ilet
    0.09
     form
    0.09
    åĩºåħ¥å¢ĥ
    0.09
    entr
    0.08
    Activations Density 0.157%

    No Known Activations