© 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. 62558
    Prev
    Next
    INDEX
    Explanations

    worth it

    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
    ä¿¡çĶ¨
    -0.09
    æ³·
    -0.09
    pw
    -0.08
     Heller
    -0.08
    subtract
    -0.08
    缨
    -0.08
    åĵĪä½Ľ
    -0.08
     Newly
    -0.08
    çļ®
    -0.08
    èıŁ
    -0.08
    POSITIVE LOGITS
    ä»ĺåĩº
    0.13
    åĢ¼å¾Ĺ
    0.12
    æį¢åıĸ
    0.11
    çīºçī²
    0.10
     worthwhile
    0.10
    代价
    0.09
    å½ĵä¹ĭæĹłæĦ§
    0.09
     sacrific
    0.09
    worth
    0.09
     Weber
    0.09
    Activations Density 0.027%

    No Known Activations