© 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. 1824
    Prev
    Next
    INDEX
    Explanations

    period of absence or gap

    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
    å§Ĭ
    -0.09
    è¿ĺä¸įå¤Ł
    -0.09
    ÑĢей
    -0.08
    èµĤ
    -0.08
     ØŃص
    -0.08
    è¿ľæĻ¯
    -0.08
    羣å®ŀçļĦ
    -0.08
    eton
    -0.08
    æĮģãģ¡
    -0.08
    -runtime
    -0.08
    POSITIVE LOGITS
    åĽŀå½Ĵ
    0.13
    éĩįè¿Ķ
    0.13
    éĹ´éļĶ
    0.12
    æļĮ
    0.11
     hiatus
    0.11
    éĺĶ
    0.11
    ä¸ŃéĹ´
    0.11
    çĿ½
    0.10
    åħ¶éĹ´
    0.10
    太ä¹ħ
    0.10
    Activations Density 0.088%

    No Known Activations