© 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. 5917
    Prev
    Next
    INDEX
    Explanations

    contact with

    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
    åĤĴ
    -0.10
     Cla
    -0.09
    åĤįæĻļ
    -0.09
    åŁĴ
    -0.09
    å®ģæĦ¿
    -0.09
    èĭ«
    -0.08
    ä»İä¸Ń
    -0.08
    å¸ĥæĭī
    -0.08
    æĴŃ
    -0.08
    欲
    -0.08
    POSITIVE LOGITS
    æݥ触
    0.36
    æݥ触åĪ°
    0.29
     contact
    0.27
    contact
    0.23
     contacto
    0.22
     конÑĤак
    0.21
     Contact
    0.20
    .contact
    0.20
    Contact
    0.19
    (contact
    0.18
    Activations Density 0.275%

    No Known Activations