© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW
    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    2. Qwen3-32B
    3. 32-RESID-BATCHTOPK-65K
    4. 52303
    Prev
    Next
    INDEX
    Explanations

    security vulnerabilities and attacks

    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
    åľŁåľ°
    -0.09
    ç½¢
    -0.09
     lands
    -0.08
    decor
    -0.07
    cue
    -0.07
     PSD
    -0.07
     IPL
    -0.07
    å°±è¿ij
    -0.07
    PIP
    -0.07
    ä¸ķ
    -0.07
    POSITIVE LOGITS
    æĶ»åĩ»
    0.16
    æĶ»æĵĬ
    0.12
    å¨ģèĥģ
    0.12
    attack
    0.11
     attack
    0.11
    åĴĮæĬĢæľ¯
    0.11
    _attack
    0.11
    .attack
    0.11
    Attack
    0.11
     attackers
    0.11
    Activations Density 0.214%

    No Known Activations