© 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. 52286
    Prev
    Next
    INDEX
    Explanations

    dramatic looks, statement, portion, increases, arts, diminution, Senate testimony, increase

    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
    seau
    -0.10
    -0.10
    borg
    -0.10
    owed
    -0.10
    uite
    -0.09
    ç²¹
    -0.09
    xic
    -0.09
    amiento
    -0.09
    ward
    -0.09
    ings
    -0.09
    POSITIVE LOGITS
    ç¼Ĺ
    0.11
    omic
    0.11
    _POINTER
    0.10
    ìłģìĿ¸
    0.10
    illon
    0.10
    æĹłæ¯Ķ
    0.10
    ä¸Ģæĸ¹
    0.09
    æ±ģ
    0.09
    è°ij
    0.09
    hare
    0.09
    Activations Density 0.011%

    No Known Activations