© 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. 22863
    Prev
    Next
    INDEX
    Explanations

    Mass followed by common suffixes/words

    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
    thing
    -0.10
    acy
    -0.10
    ectors
    -0.10
    atter
    -0.10
    IGHT
    -0.10
    ture
    -0.10
    ../../../
    -0.09
    CTION
    -0.09
    ush
    -0.09
    ../../
    -0.09
    POSITIVE LOGITS
    achuset
    0.16
    achusetts
    0.15
    ively
    0.14
    acre
    0.13
    mates
    0.12
    iveness
    0.12
     mẽ
    0.12
    ses
    0.12
    mate
    0.11
    sing
    0.11
    Activations Density 0.022%

    No Known Activations