© 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. 27248
    Prev
    Next
    INDEX
    Explanations

    do not alter, edit, or modify

    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
    enor
    -0.10
    å¦Ĥæŀľä¸į
    -0.10
    è¦ģä¹Ī
    -0.09
    kj
    -0.09
    æĥ¬
    -0.08
    ä¸įä½Ĩ
    -0.08
    å¾Ī大
    -0.08
    è¦ģåģļ好
    -0.08
     itself
    -0.08
    åįģä¸ĩ
    -0.08
    POSITIVE LOGITS
    éĵ¡
    0.10
    ayo
    0.10
    thing
    0.10
    lor
    0.10
    讳
    0.10
    称ä¹ĭ
    0.09
    ç¨Ģ
    0.09
    thal
    0.09
    ance
    0.09
    çĶ»åĥı
    0.09
    Activations Density 0.025%

    No Known Activations