© 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. 27770
    Prev
    Next
    INDEX
    Explanations

    comparing things or giving examples

    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
    istas
    -0.12
     лиÑĪÑĮ
    -0.09
    åºĬä½į
    -0.09
    æĪijçĽ¸ä¿¡
    -0.09
    nt
    -0.09
    _Parse
    -0.09
    rick
    -0.09
    licht
    -0.09
    mal
    -0.09
    วร
    -0.09
    POSITIVE LOGITS
    -minded
    0.19
     minded
    0.13
    æł·çļĦ
    0.13
     unto
    0.12
    hood
    0.11
    Ø«
    0.11
    able
    0.11
    对å¾ħ
    0.11
    est
    0.11
    ä¸įåĥı
    0.10
    Activations Density 0.115%

    No Known Activations