© 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. 5052
    Prev
    Next
    INDEX
    Explanations

    met with

    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
    é¢ĦæĬ¥
    -0.10
    èĭ
    -0.08
    æĬĬæı¡
    -0.08
    棱
    -0.08
     пÑĢед
    -0.08
    avan
    -0.08
    éºĹ
    -0.08
     Ding
    -0.08
    æīĥ
    -0.08
    éĢīæĭĶ
    -0.07
    POSITIVE LOGITS
    请æķĻ
    0.10
    ¹Ħ
    0.10
    举é£İ
    0.10
    lando
    0.09
    常åĬ¡åī¯
    0.09
    ä¸Ģç¬ij
    0.09
    ERSIST
    0.09
    çİĭåħĪçĶŁ
    0.09
     Gilles
    0.09
    åĬŀäºĭ
    0.09
    Activations Density 0.091%

    No Known Activations