© 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. 19103
    Prev
    Next
    INDEX
    Explanations

    exposure and visibility

    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
    åįłæį®
    -0.08
    visible
    -0.08
    BA
    -0.08
    éĢĨè¡Į
    -0.08
    oad
    -0.08
    o
    -0.08
    poi
    -0.07
    éĻĪè¿°
    -0.07
    íĴĪ
    -0.07
    ard
    -0.07
    POSITIVE LOGITS
    ight
    0.10
    åĢĻéĢī
    0.09
    æĢı
    0.09
    atch
    0.09
    ighter
    0.09
    aska
    0.09
    çķ¸
    0.09
     Terr
    0.09
    Terr
    0.09
     Wir
    0.09
    Activations Density 0.013%

    No Known Activations