© 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. 1003
    Prev
    Next
    INDEX
    Explanations

    slated, attached, planning, meant

    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
    åĩłåĪĨéĴŁ
    -0.09
    廪
    -0.09
    åĩĽ
    -0.09
    æģ¬
    -0.09
    iltr
    -0.09
    tran
    -0.08
     Sunder
    -0.08
    å©į
    -0.08
     Vernon
    -0.08
    åģŃ
    -0.08
    POSITIVE LOGITS
    æĭĽ
    0.10
    ä»ĸå¦Ī
    0.10
    æIJŀå®ļ
    0.10
    æĭĽåĭŁ
    0.09
    æ¡£
    0.09
    人éĢī
    0.09
    éķ¶
    0.09
    åĬłçĽŁ
    0.09
    èģĺ请
    0.09
    åıĤä¸İåĪ°
    0.09
    Activations Density 0.075%

    No Known Activations