© 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. 7496
    Prev
    Next
    INDEX
    Explanations

    oh, what

    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
    олÑı
    -0.09
    伺
    -0.09
    upe
    -0.09
    æĦ
    -0.08
    俸
    -0.08
    âĹ»
    -0.08
    éªijåħµ
    -0.08
    ÙĪÙĨÙĬ
    -0.08
    .ta
    -0.08
    ursive
    -0.08
    POSITIVE LOGITS
    æĺİçĻ½äºĨ
    0.09
    ridge
    0.09
    ç¨Ĺ
    0.09
     ÑģÑĤÑĢан
    0.08
    omit
    0.08
     OT
    0.08
    otta
    0.08
    lander
    0.08
    itta
    0.08
    wart
    0.08
    Activations Density 0.019%

    No Known Activations