© 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. 60973
    Prev
    Next
    INDEX
    Explanations

    return to previous state

    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
    éļĻ
    -0.09
    æµ®
    -0.09
    acional
    -0.09
    ãĥĪãĥ«
    -0.09
     Gór
    -0.08
    彩
    -0.08
    姿åĬ¿
    -0.08
    alic
    -0.08
    ruk
    -0.08
    .Lookup
    -0.08
    POSITIVE LOGITS
    åĽŀåĪ°
    0.22
    åĽŀåĪ°äºĨ
    0.20
    ãģ«æĪ»
    0.17
    å¤įåı¤
    0.15
    åĽŀåİ»
    0.15
    åĽŀå½Ĵ
    0.15
     назад
    0.15
    éĢĢåĽŀ
    0.14
     revert
    0.14
    è¿ĺåİŁ
    0.14
    Activations Density 0.232%

    No Known Activations