© 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. 38564
    Prev
    Next
    INDEX
    Explanations

    breakfast

    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
    éĶ·
    -0.11
    ities
    -0.10
    ลำ
    -0.09
    iggers
    -0.09
    å¼
    -0.09
    bery
    -0.09
    å®Ŀè´Ŀ
    -0.08
    pdb
    -0.08
    æį©
    -0.08
    éĻ·
    -0.08
    POSITIVE LOGITS
    ting
    0.14
    èĻ
    0.09
    iens
    0.09
    wards
    0.09
    ÑĤал
    0.09
    sWith
    0.09
    orz
    0.09
     amph
    0.08
    çĹķ
    0.08
    ä¹Łä¸įæĥ³
    0.08
    Activations Density 0.007%

    No Known Activations