© 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. 40050
    Prev
    Next
    INDEX
    Explanations

    cleared or clearing

    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
    åīįéĢĶ
    -0.10
    èĺijèıĩ
    -0.10
    åĮĸçŁ³
    -0.10
     chặt
    -0.09
    æĺİç¡®è§Ħå®ļ
    -0.09
    樵
    -0.09
    å°±è¿ij
    -0.09
    æįħ
    -0.09
    çļĦä¸ľè¥¿
    -0.09
    å®īå¿ĥ
    -0.09
    POSITIVE LOGITS
     khá»ıi
    0.09
    lier
    0.09
    bie
    0.09
     clearing
    0.09
    ga
    0.08
    ugar
    0.08
    away
    0.08
    pace
    0.08
    fork
    0.08
    åij¨è½¬
    0.08
    Activations Density 0.021%

    No Known Activations