© 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. 2835
    Prev
    Next
    INDEX
    Explanations

    down, prim, crackdown, confiscate

    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
    apia
    -0.10
    ner
    -0.10
    eneg
    -0.09
    tsky
    -0.09
    енÑĮк
    -0.08
    åħħæ²Ľ
    -0.08
    itized
    -0.08
    çĨŁ
    -0.08
    tÃŃ
    -0.08
    elib
    -0.08
    POSITIVE LOGITS
    å´Ľ
    0.09
    é¢ĩ
    0.09
    heed
    0.09
    ä¸įå°ı
    0.09
    ffer
    0.09
     Slee
    0.09
    eson
    0.09
    alore
    0.09
    ude
    0.08
    esc
    0.08
    Activations Density 0.004%

    No Known Activations