© 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. 49227
    Prev
    Next
    INDEX
    Explanations

    hot dogs, pickles, delicacy, verts

    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
    æŃ£éĿ¢
    -0.09
    aldi
    -0.09
     cycles
    -0.09
    åĸľæĤ¦
    -0.09
    æŃ¢
    -0.09
    Cycle
    -0.09
    ycles
    -0.09
    cycles
    -0.08
     massa
    -0.08
    åīIJ
    -0.08
    POSITIVE LOGITS
    èĥĢ
    0.09
    沫
    0.09
    å·¡
    0.09
    åĸ·æ¶Ĥ
    0.09
    åĪĽéĢłæĢ§
    0.09
     sophistic
    0.08
    âģ¿
    0.08
    .ids
    0.08
    礼
    0.08
    Overrides
    0.08
    Activations Density 0.055%

    No Known Activations