© 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. 33900
    Prev
    Next
    INDEX
    Explanations

    `//` and start of comments

    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
    à¹Į
    -0.12
    ally
    -0.11
    ance
    -0.10
    ered
    -0.09
    ers
    -0.09
    antly
    -0.09
    ality
    -0.09
    eer
    -0.09
    äººæł¼
    -0.08
    ively
    -0.08
    POSITIVE LOGITS
    èµ°ä¸ĭåİ»
    0.10
    ·»
    0.10
    ÙĬدÙĬ
    0.10
    _Application
    0.09
    éģĦ
    0.09
    ç½ĺ
    0.09
    èĤ¡æľ¬
    0.09
    åļ¼
    0.09
    gars
    0.09
    ãģªãĤīãģªãģĦ
    0.09
    Activations Density 0.128%

    No Known Activations