© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 19369
    Prev
    Next
    INDEX
    Explanations

    binary 0 or 1

    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
    仪
    -0.13
     Guerrero
    -0.10
    æĶ¹åĪ¶
    -0.09
    rag
    -0.09
    èĮĥåĽ´
    -0.09
    PCI
    -0.09
     Kimber
    -0.09
    åıijæĮ¥
    -0.08
    PIP
    -0.08
    åĦĢ
    -0.08
    POSITIVE LOGITS
    æ¯Ķçī¹
    0.14
    -binary
    0.12
     binary
    0.11
    羣åģĩ
    0.11
    æ¯ıä¸Ģä½į
    0.11
    toggle
    0.10
    éĢ»è¾ij
    0.10
    binary
    0.10
    .binary
    0.10
     bits
    0.10
    Activations Density 0.152%

    No Known Activations