© 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. Qwen3-8B
    5. 18-RESID-BATCHTOPK-65K__L0-80
    7. 3246
    Prev
    Next
    INDEX
    Explanations

    attack, assassination, vandalism, malicious acts

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    adamkarvonen/qwen3-8b-saes/saes_Qwen_Qwen3-8B_batch_top_k/resid_post_layer_18
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Negative Logits
     ðŁĻĤ
    -0.23
    è¶Ł
    -0.23
     condensed
    -0.22
    ä½ľèĢħæľ¬äºº
    -0.22
     Ivanka
    -0.22
    essen
    -0.22
    (emp
    -0.22
    (sheet
    -0.21
    uyen
    -0.21
     STATES
    -0.21
    POSITIVE LOGITS
    æĶ»åĩ»
    0.33
    å¨ģèĥģ
    0.32
     targeting
    0.30
     kidn
    0.30
     malicious
    0.30
     assassination
    0.30
    çł´åĿı
    0.29
     vandalism
    0.29
     terrorist
    0.29
     assass
    0.29
    Activations Density 0.149%

    No Known Activations