© 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. 30294
    Prev
    Next
    INDEX
    Explanations

    back followed by compound 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
    кла
    -0.12
    æѤåĪ»
    -0.11
    รà¸Ńà¸ļ
    -0.10
    thy
    -0.10
    anka
    -0.10
    гин
    -0.10
    -vous
    -0.09
    wide
    -0.09
    atori
    -0.09
    este
    -0.09
    POSITIVE LOGITS
    ness
    0.16
    slashes
    0.13
    slash
    0.12
    wards
    0.11
    ened
    0.11
    WARD
    0.11
    ward
    0.11
    WARDS
    0.10
    ronym
    0.10
    edly
    0.10
    Activations Density 0.116%

    No Known Activations