© 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. 61651
    Prev
    Next
    INDEX
    Explanations

    bluntly, abruptly, brutally, blatantly

    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
    owie
    -0.11
    monic
    -0.11
    andel
    -0.10
    PropertyParams
    -0.09
    é¾Ī
    -0.09
    ibre
    -0.09
    ibur
    -0.08
    æĸ°çļĦä¸Ģå¹´
    -0.08
     glue
    -0.08
    NewProp
    -0.08
    POSITIVE LOGITS
    ly
    0.12
    éģĵçIJĨ
    0.10
     blunt
    0.10
    ãģĮãĤī
    0.09
     assorted
    0.09
    edList
    0.09
    ishments
    0.09
    åľ°è¯´
    0.09
    å¢
    0.09
    ÑıÑĤ
    0.09
    Activations Density 0.006%

    No Known Activations