© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 35998
    Prev
    Next
    INDEX
    Explanations

    vampire and zombie

    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
    é«ĺåİŁ
    -0.10
     Genius
    -0.10
    ä¸Ĭåİ»
    -0.09
    _mD
    -0.09
     bum
    -0.09
    强åĬĽ
    -0.09
    æ°´æµģ
    -0.09
    igest
    -0.08
    tre
    -0.08
    交æį¢
    -0.08
    POSITIVE LOGITS
     undead
    0.12
    åĥµå°¸
    0.11
     zombie
    0.10
    æĺ¼
    0.09
     vampire
    0.09
     Duis
    0.09
     zombies
    0.09
    ptime
    0.09
    åIJ¸
    0.09
    isas
    0.09
    Activations Density 0.048%

    No Known Activations