© 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. 32953
    Prev
    Next
    INDEX
    Explanations

    before and after

    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
    çīĻ
    -0.09
    å¤ĸçķĮ
    -0.09
    åıįå¤į
    -0.08
    æģĴ
    -0.08
    èĬ½
    -0.08
    ling
    -0.08
    è¿Ĥ
    -0.08
    éķĬ
    -0.08
    éĩįå¤į
    -0.08
    irst
    -0.07
    POSITIVE LOGITS
    åIJİéĿ¢
    0.12
    ç¢Ł
    0.11
    neath
    0.11
    ä¸įè¿ľå¤Ħ
    0.10
    身åIJİ
    0.10
    ëĴ¤
    0.09
    ä¸įè¿ľ
    0.09
    ä¸ĭæĸ¹
    0.09
    Indented
    0.09
    éĤ£æ®µ
    0.09
    Activations Density 0.126%

    No Known Activations