© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 37795
    Prev
    Next
    INDEX
    Explanations

    ask a question

    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
    æľīåħ³éĥ¨éŨ
    -0.10
    ogl
    -0.09
    æĬĬå®ĥ
    -0.09
     Yong
    -0.09
    欢è¿İ大家
    -0.09
    Accessory
    -0.09
    æ¶ĪåĮĸ
    -0.09
    å®ĥæĺ¯
    -0.08
     Beh
    -0.08
    èĥ¸éĥ¨
    -0.08
    POSITIVE LOGITS
     who
    0.09
    èµĬ
    0.09
    樯
    0.09
    ä¸ĢåIJĮ
    0.09
    è°Ī
    0.09
    who
    0.08
     Cond
    0.08
    å¯Ħ
    0.08
     whom
    0.08
    åIJįåįķ
    0.08
    Activations Density 0.076%

    No Known Activations