© 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. 56380
    Prev
    Next
    INDEX
    Explanations

    tutoring services

    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
    åħĨ
    -0.10
    æī«ä¸Ģæī«
    -0.10
    outh
    -0.09
    åįļçī©
    -0.09
    _DSP
    -0.09
    PEG
    -0.08
    uer
    -0.08
    éºŁ
    -0.08
     Cobra
    -0.08
    emens
    -0.08
    POSITIVE LOGITS
    ainment
    0.11
    åıĸèĥľ
    0.10
    åIJ®
    0.09
    å°±è¿Ļæł·
    0.08
    å±±åĿ¡
    0.08
    ÑĢей
    0.08
    奢
    0.08
    lift
    0.08
    çŀ½
    0.08
     sĩ
    0.08
    Activations Density 0.010%

    No Known Activations