© 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. 33727
    Prev
    Next
    INDEX
    Explanations

    illustrations and clarifications

    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
     Burl
    -0.10
     flatt
    -0.10
    髦
    -0.10
    eness
    -0.09
    没äºĭ
    -0.09
    忸
    -0.09
    åħ¨åĽ½åIJĦåľ°
    -0.09
    pedo
    -0.09
    æĹłæīĢè°ĵ
    -0.09
    èĶ¼
    -0.09
    POSITIVE LOGITS
     example
    0.20
    ä¾ĭ
    0.19
    example
    0.18
    举ä¾ĭ
    0.17
    ä¾ĭå¦Ĥ
    0.17
    æ¯Ķå¦Ĥ
    0.16
    为ä¾ĭ
    0.16
    èѬå¦Ĥ
    0.16
    çļĦä¾ĭåŃIJ
    0.16
    -example
    0.16
    Activations Density 0.117%

    No Known Activations