© 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. 2903
    Prev
    Next
    INDEX
    Explanations

    auditorium or auditor

    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
    adows
    -0.10
    enta
    -0.10
    çŃ·
    -0.09
     гÑĢ
    -0.09
    halt
    -0.09
    ä¼Ļ
    -0.08
    tered
    -0.08
    ionate
    -0.08
    adow
    -0.08
    ion
    -0.08
    POSITIVE LOGITS
    acity
    0.11
    ium
    0.11
    yssey
    0.10
    ie
    0.10
    æ¯Ķè¾ĥ大
    0.10
    å¨ĺ
    0.10
    inary
    0.10
    et
    0.09
    ries
    0.09
    imension
    0.09
    Activations Density 0.015%

    No Known Activations