© 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. 57405
    Prev
    Next
    INDEX
    Explanations

    units of measurement

    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
    âĨĵ
    -0.09
    .Formatter
    -0.09
    以èī²
    -0.09
    çĸijæĥij
    -0.08
    ä¸Ģè¶Ł
    -0.08
    éģº
    -0.08
    /ref
    -0.08
     Stark
    -0.08
    town
    -0.08
    uckland
    -0.08
    POSITIVE LOGITS
    以ä¸ĬçļĦ
    0.15
    å·¦åı³
    0.13
    ì§ľ
    0.12
    allon
    0.11
    以ä¸Ĭ
    0.09
    åıĬ以ä¸Ĭ
    0.09
    åħīéĺ´
    0.09
     اÙĦÙĪاØŃد
    0.08
    째
    0.08
    çļĦä¼ģä¸ļ
    0.08
    Activations Density 0.188%

    No Known Activations