© 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. 3204
    Prev
    Next
    INDEX
    Explanations

    higher / high

    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
    ä¹ĭéģĵ
    -0.10
    ittel
    -0.09
    头èĦij
    -0.09
    éŃĤ
    -0.09
    é«ĺ温
    -0.09
    ndo
    -0.08
    likes
    -0.08
    大æĪ·
    -0.08
    losure
    -0.08
    åħĴ
    -0.08
    POSITIVE LOGITS
    -quality
    0.12
    -priced
    0.11
    landers
    0.11
    -tech
    0.11
    -ranking
    0.11
    lander
    0.11
    cased
    0.10
     جدا
    0.10
    (er
    0.09
    -powered
    0.09
    Activations Density 0.109%

    No Known Activations