© 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. 2984
    Prev
    Next
    INDEX
    Explanations

    invitations to participate

    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
    åĽŀåİ»
    -0.10
    澳大åĪ©äºļ
    -0.10
    澳洲
    -0.09
     honoured
    -0.09
    èijĹ
    -0.09
    (Optional
    -0.09
     Ottawa
    -0.08
    è¿Ļä¸ĢåĪĩ
    -0.08
     bes
    -0.08
    æ®Ĭ
    -0.08
    POSITIVE LOGITS
    ichten
    0.10
    rint
    0.09
     Budd
    0.08
    çĶŁçIJĨ
    0.08
    /gpl
    0.08
    áĢĿ
    0.08
    TERS
    0.08
    dzi
    0.08
     joining
    0.08
    _BUFFER
    0.08
    Activations Density 0.123%

    No Known Activations