© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 8683
    Prev
    Next
    INDEX
    Explanations

    cryo

    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
    essa
    -0.09
    éļħ
    -0.09
    ihad
    -0.09
    udd
    -0.08
    åŁºçŁ³
    -0.08
    姨
    -0.08
    æİĸ
    -0.08
    بÙĬØ©
    -0.08
    çĸĻ
    -0.08
    .Throw
    -0.08
    POSITIVE LOGITS
    UTE
    0.10
    ণ
    0.10
    outube
    0.09
    è°İ
    0.09
    quip
    0.09
    fulness
    0.09
    追éĹ®
    0.09
    pte
    0.09
    à¸ĺรรม
    0.09
    anny
    0.08
    Activations Density 0.011%

    No Known Activations