© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 36255
    Prev
    Next
    INDEX
    Explanations

    PUBLISHED words and legal proceedings

    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
    uate
    -0.10
    æ¡ħ
    -0.09
    endcode
    -0.09
    请æĤ¨
    -0.09
    çķı
    -0.09
    åİ¢
    -0.09
    åĴİ
    -0.08
    åĨĽåĽ¢
    -0.08
    æ¦ľ
    -0.08
    æĬ¬
    -0.08
    POSITIVE LOGITS
    ATTERN
    0.11
    ooled
    0.10
    aises
    0.10
    è°ħ
    0.09
    uentes
    0.09
    PING
    0.09
    riors
    0.09
    OTOS
    0.09
    rik
    0.09
    entially
    0.08
    Activations Density 0.197%

    No Known Activations