© 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. 40078
    Prev
    Next
    INDEX
    Explanations

    'cr' followed by d, ash, IS, ick, db, anes

    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
    sWith
    -0.10
    iju
    -0.09
    oir
    -0.09
    ÙĪÙĬØ©
    -0.09
    ancock
    -0.09
    estion
    -0.09
    æ°Ķ温
    -0.09
    æĺ¥å¤©
    -0.09
    oz
    -0.09
    èĥĽ
    -0.09
    POSITIVE LOGITS
    -quarters
    0.12
    shaw
    0.12
    ITICAL
    0.11
    åī¯ä¹¦è®°
    0.11
     havoc
    0.10
    æ»ĵ
    0.10
    shaft
    0.09
    fully
    0.09
    anium
    0.09
    หา
    0.09
    Activations Density 0.148%

    No Known Activations