© 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. 10153
    Prev
    Next
    INDEX
    Explanations

    trap and capture

    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
    ä¸Ģå¦ĤæĹ¢å¾Ģ
    -0.10
    ROT
    -0.09
    ç©¿
    -0.09
    .ua
    -0.09
    PEG
    -0.09
    æ·±åĮĸ
    -0.09
    饼干
    -0.08
    åĪ°è´¦
    -0.08
    ино
    -0.08
    PIP
    -0.08
    POSITIVE LOGITS
    .capture
    0.13
    被æįķ
    0.13
     capture
    0.13
    æįķæįī
    0.12
    æįķ
    0.12
    æĬĵæįķ
    0.12
    æĶ¶éĽĨ
    0.11
    capture
    0.11
    æĶ¾è¿Ľ
    0.11
     Capture
    0.10
    Activations Density 0.071%

    No Known Activations