© 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. 37798
    Prev
    Next
    INDEX
    Explanations

    institution

    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
    heet
    -0.13
    thing
    -0.09
    OrCreate
    -0.09
    ãĤ£
    -0.09
    ousse
    -0.09
    backs
    -0.09
    ish
    -0.09
    back
    -0.09
    ç¦ıå¾·
    -0.09
    ëģĶ
    -0.09
    POSITIVE LOGITS
    ally
    0.16
    ively
    0.13
    urally
    0.11
    nement
    0.11
    nel
    0.11
    ALLY
    0.10
    æĸ°åŁİ
    0.10
    utton
    0.10
    arian
    0.09
    ãģ¥
    0.09
    Activations Density 0.013%

    No Known Activations