© 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. 17392
    Prev
    Next
    INDEX
    Explanations

    recipe 's

    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
    çļ±
    -0.10
    è¾Ĩ车
    -0.08
    èĻļ空
    -0.08
    é£İ鼨
    -0.08
    HIP
    -0.08
     scratching
    -0.08
    nit
    -0.08
    Wave
    -0.08
    pmat
    -0.08
     Äijè
    -0.07
    POSITIVE LOGITS
    ç®Ģåįķ
    0.10
    æ¯Ķè¾ĥç®Ģåįķ
    0.09
     recipe
    0.09
    ç°¡åįĺ
    0.09
     McB
    0.09
    çļĦåģļæ³ķ
    0.09
    åģļæ³ķ
    0.09
    åĴĮæľįåĬ¡
    0.09
    IMIZE
    0.09
    çĽĬ
    0.09
    Activations Density 0.071%

    No Known Activations