© 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. 50994
    Prev
    Next
    INDEX
    Explanations

    opportunities to connect

    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
    帽
    -0.09
    帽åŃIJ
    -0.09
    泡
    -0.08
    åĽ¿
    -0.08
    atcher
    -0.08
    atsu
    -0.08
    å¹³åĩ¡
    -0.08
    éĢīåĩº
    -0.08
    æĴĩ
    -0.08
    TEMPL
    -0.08
    POSITIVE LOGITS
    ä¸ĵ人
    0.13
    大åĴĸ
    0.12
    ä¸ĸçķĮ级
    0.11
    åIJį家
    0.11
    çŁ¥åIJį
    0.11
    ä¸ĵ家
    0.11
     expert
    0.10
    ä¸ĸçķĮä¸Ģæµģ
    0.10
    çŁ¥åIJįåĵģçīĮ
    0.10
     experts
    0.10
    Activations Density 0.070%

    No Known Activations