© 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. 54872
    Prev
    Next
    INDEX
    Explanations

    expectations and predictions

    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
    åĿ
    -0.09
    积
    -0.08
    Anderson
    -0.08
    æ®ģ
    -0.08
    éĿ¢çĽ®
    -0.08
    spell
    -0.08
    Ð¡ÐŁ
    -0.08
     اÙĦÙĪاØŃد
    -0.08
    絡
    -0.08
    æĦŁåĪ°
    -0.08
    POSITIVE LOGITS
    ä»Ģä¹Īåij¢
    0.11
    æĻ®
    0.09
    itical
    0.09
    ajs
    0.09
    æĺ¯ä»Ģä¹Īåij¢
    0.08
    rippling
    0.08
    iras
    0.08
    æī¶
    0.08
    _tol
    0.08
    ëĤ´
    0.08
    Activations Density 0.230%

    No Known Activations