© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it
    Neuronpedia
    Jacobian LensNEW
    Natural Language
    Autoencoders
    MAY 26    Assistant AxisJAN 26Circuit TracerAPR 26SteerSAE EvalsExportsinterp-engineNEW Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 894
    Prev
    Next
    INDEX
    Explanations

    sitting or sat down

    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
    /from
    -0.14
    /out
    -0.11
    hand
    -0.10
    /Linux
    -0.09
    /on
    -0.09
    hone
    -0.09
    /by
    -0.09
    ered
    -0.08
    hits
    -0.08
    ä½łèĥ½
    -0.08
    POSITIVE LOGITS
    uate
    0.11
    èĩªå¦Ĥ
    0.10
    uito
    0.10
    igli
    0.09
    ellite
    0.09
    undle
    0.09
    ter
    0.08
    åľ¨ä¸Ģèµ·
    0.08
    -up
    0.08
    æģ¶
    0.08
    Activations Density 0.013%

    No Known Activations