© 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. 5712
    Prev
    Next
    INDEX
    Explanations

    role in user/group configuration

    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
    agne
    -0.10
    å¼ķæµģ
    -0.10
    夹
    -0.10
    çªģ
    -0.09
    èī°
    -0.09
    UInteger
    -0.09
    rier
    -0.09
    æĸĩåĪĽ
    -0.09
    éļĶçĿĢ
    -0.09
    .Invariant
    -0.09
    POSITIVE LOGITS
    omatic
    0.08
     дов
    0.08
     pitch
    0.08
    versed
    0.08
    play
    0.08
     Jesse
    0.08
     Gur
    0.08
    convert
    0.08
    Accessible
    0.08
    åĪĨæĺİ
    0.08
    Activations Density 0.015%

    No Known Activations