© 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. 54923
    Prev
    Next
    INDEX
    Explanations

    perfect [noun]

    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
    rese
    -0.09
    лем
    -0.09
    pile
    -0.09
    lor
    -0.09
    _serializer
    -0.09
    uby
    -0.08
    ndx
    -0.08
    åĶ±
    -0.08
    uctions
    -0.08
    lon
    -0.08
    POSITIVE LOGITS
    idon
    0.11
    å®Įç¾İçļĦ
    0.10
    主ä¹īèĢħ
    0.09
    idious
    0.09
    ë¥ł
    0.09
    人åĿĩ
    0.09
    æĥķ
    0.09
    criptors
    0.08
    istic
    0.08
    ÑģÑĤи
    0.08
    Activations Density 0.012%

    No Known Activations