© 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. 48740
    Prev
    Next
    INDEX
    Explanations

    for this reason

    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
     vox
    -0.09
    вод
    -0.09
    åİŁæĿ¥æĺ¯
    -0.09
    phan
    -0.09
    æĿŃ
    -0.09
    igel
    -0.08
    åįıåķĨ
    -0.08
    inq
    -0.08
    _ASM
    -0.08
    _subtype
    -0.08
    POSITIVE LOGITS
    以åıĬåħ¶ä»ĸ
    0.12
     alone
    0.11
    alone
    0.11
     ëķĮ문ìĹIJ
    0.10
    æ°Ķæ°Ľ
    0.10
    æĪijæīį
    0.10
    åĹŁ
    0.09
    cen
    0.09
    éĦĻ
    0.09
    èħ¼
    0.09
    Activations Density 0.028%

    No Known Activations