© 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. 63806
    Prev
    Next
    INDEX
    Explanations

    exploit, exploitation, exploited

    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
    çĦ¯
    -0.10
    åĩ«
    -0.10
    ãģĹãģ¾ãģĨ
    -0.10
    ÙħÙĦ
    -0.10
    æŨ
    -0.09
    scape
    -0.09
    eker
    -0.09
    alyzer
    -0.09
    ega
    -0.09
    robe
    -0.09
    POSITIVE LOGITS
    ively
    0.16
    ting
    0.12
    iation
    0.10
    ably
    0.09
    ï¸
    0.09
    çİĩè¾¾
    0.09
    ties
    0.09
    äre
    0.09
    iesen
    0.09
    çİĩè¾¾åĪ°
    0.09
    Activations Density 0.009%

    No Known Activations