© 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. 51307
    Prev
    Next
    INDEX
    Explanations

    freeware and ware

    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
    ombo
    -0.10
    alf
    -0.10
    ses
    -0.09
    naire
    -0.09
    éĤ£ä»½
    -0.09
    EObject
    -0.09
    꺼
    -0.09
    æĹłç¼ĺ
    -0.08
    ric
    -0.08
    ä¸ĩåĪĨ
    -0.08
    POSITIVE LOGITS
    ground
    0.10
    ìļ¸
    0.10
    åĻĮ
    0.10
    housing
    0.10
    aths
    0.09
    erge
    0.09
    idge
    0.09
    itage
    0.09
    ments
    0.09
    ishing
    0.09
    Activations Density 0.013%

    No Known Activations