© 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. 27165
    Prev
    Next
    INDEX
    Explanations

    engage 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
    èĮĥ
    -0.09
    imps
    -0.09
    verting
    -0.08
    æĬ
    -0.08
    æĪ¿
    -0.08
    带头人
    -0.08
     Lith
    -0.08
    讹
    -0.08
    TECT
    -0.08
    “He
    -0.08
    POSITIVE LOGITS
    enga
    0.10
    åĥı个
    0.10
    è¶ĬæĿ¥è¶Ĭé«ĺ
    0.10
    åľ¨åĨħçļĦ
    0.09
    æī£
    0.09
    éĺħ读åħ¨æĸĩ
    0.09
    æĸĹå¿Ĺ
    0.09
    çݳ
    0.09
    æĶ»åĬ¿
    0.09
    çݯæ¯Ķ
    0.09
    Activations Density 0.041%

    No Known Activations