© 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. 22044
    Prev
    Next
    INDEX
    Explanations

    succulent, savory, or inedible descriptors

    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
    ilan
    -0.09
    _planes
    -0.08
    stp
    -0.08
    rand
    -0.08
    /body
    -0.08
    /cop
    -0.08
     лÑĥÑĩÑĪиÑħ
    -0.08
    ir
    -0.08
    ote
    -0.07
    æłĩ
    -0.07
    POSITIVE LOGITS
    ä¸Ģå¹ķ
    0.10
    ously
    0.10
    æľīåĪ©
    0.09
    fy
    0.09
    zyst
    0.09
    第ä¸Ģä½į
    0.09
    æ¯ıä¸Ģä½į
    0.09
    ous
    0.09
    æĶ¾æīĭ
    0.09
    çļĦ缮åħī
    0.09
    Activations Density 0.003%

    No Known Activations