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

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 38161
    Prev
    Next
    INDEX
    Explanations

    remove, hack, crack, invalid

    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
    _Pre
    -0.09
    ces
    -0.09
    _OPENGL
    -0.08
    æѧ
    -0.08
    .removeChild
    -0.08
    olute
    -0.08
    éŨåı£
    -0.08
    à¹ģร
    -0.08
    头èĦij
    -0.08
    kee
    -0.08
    POSITIVE LOGITS
    æµĭè¯Ħ
    0.11
    åĿ¦
    0.10
     Hedge
    0.09
    æīª
    0.09
    åı¦æľī
    0.08
    åı¸é©¬
    0.08
    主è§Ĥ
    0.08
    彩èĻ¹
    0.08
    åį°è±¡
    0.08
    滩
    0.08
    Activations Density 0.047%

    No Known Activations