© 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. 61165
    Prev
    Next
    INDEX
    Explanations

    mega

    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
    etz
    -0.11
    iage
    -0.09
    ables
    -0.09
    iday
    -0.09
    £
    -0.09
    anship
    -0.08
    æģļ
    -0.08
    IVO
    -0.08
    anson
    -0.08
    umm
    -0.08
    POSITIVE LOGITS
    OTH
    0.11
    abytes
    0.10
    .nz
    0.10
    sert
    0.10
    麽
    0.10
    ĩ
    0.10
    IMUM
    0.09
    åĽŀ头çľĭ
    0.09
    ä¸į代表
    0.09
     Ñģобой
    0.09
    Activations Density 0.015%

    No Known Activations