© 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. 51946
    Prev
    Next
    INDEX
    Explanations

    pragma directives

    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
    NAME
    -0.09
    name
    -0.08
    magnitude
    -0.08
     Weekend
    -0.08
    nger
    -0.08
    thing
    -0.08
    stry
    -0.08
    uu
    -0.08
    ut
    -0.07
    FG
    -0.07
    POSITIVE LOGITS
    IRON
    0.12
    fully
    0.12
    ìłģìľ¼ë¡ľ
    0.11
    主ä¹īèĢħ
    0.11
    åķ¡
    0.10
    ä»ĸ们ä¼ļ
    0.10
    ément
    0.10
    atics
    0.10
    åįģ足
    0.10
    åĬŁèĥ½æĢ§
    0.10
    Activations Density 0.007%

    No Known Activations