© 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. 58128
    Prev
    Next
    INDEX
    Explanations

    throughout

    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
    åľ¨æĪij
    -0.09
    åIJĦæľī
    -0.09
    ëŀij
    -0.09
    _SELECT
    -0.09
    ubic
    -0.08
     pac
    -0.08
    ä¸ĢæŶ
    -0.08
    ãĤĪãĤĬ
    -0.08
    say
    -0.08
     stray
    -0.08
    POSITIVE LOGITS
    贯穿
    0.11
     throughout
    0.11
    åħ¨æĸĩ
    0.11
    æķ´ä¸ª
    0.10
    åħ¨ç¨ĭ
    0.10
    åħĴ
    0.10
    åħ¨è¿ĩç¨ĭ
    0.10
    å§ĭç»Ī
    0.10
    ä½ķå¤Ħ
    0.09
    ake
    0.09
    Activations Density 0.020%

    No Known Activations