© 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. 39669
    Prev
    Next
    INDEX
    Explanations

    Paramount Pictures or importance

    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
    awan
    -0.09
    ën
    -0.09
    busters
    -0.09
    mates
    -0.09
    èī¯å¥½
    -0.09
    abor
    -0.08
    è§Ħ
    -0.08
    upert
    -0.08
    orate
    -0.08
    裳
    -0.08
    POSITIVE LOGITS
    eer
    0.12
     ë¶Ī구íķĺê³ł
    0.10
    adox
    0.10
    æĢ¿
    0.09
    ä¸İåIJ¦
    0.09
    åĩĮæĻ¨
    0.09
    -esque
    0.09
    kle
    0.09
    çŀ©çĽ®
    0.09
    оÑĩной
    0.09
    Activations Density 0.004%

    No Known Activations