© 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. 57443
    Prev
    Next
    INDEX
    Explanations

    articulate parts

    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
    .cz
    -0.10
    ctp
    -0.10
    ิ
    -0.09
    çľ¼è§Ĵ
    -0.09
    -tm
    -0.09
    clare
    -0.08
    æīĵæĬĺ
    -0.08
    eah
    -0.08
    åıĬæŶ
    -0.08
    ific
    -0.08
    POSITIVE LOGITS
    ä¸įæ¸ħ
    0.13
    ç¼Ł
    0.09
    åĸīåĴĻ
    0.09
    ITE
    0.09
    大大å°ıå°ı
    0.09
    ulated
    0.09
    èĭŀ
    0.08
    ulation
    0.08
    Licensed
    0.08
    ulate
    0.08
    Activations Density 0.027%

    No Known Activations