© 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 EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 48278
    Prev
    Next
    INDEX
    Explanations

    mundane, boring, unsafe, avoidant

    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
    æģ£
    -0.10
    è¿ħçĮĽ
    -0.09
    hait
    -0.09
    æķıéĶIJ
    -0.09
     ÑĢавно
    -0.08
    åĪĨæĺİ
    -0.08
     Huff
    -0.08
    èĿ¶
    -0.08
    .jquery
    -0.08
     ÑĢаÑģ
    -0.08
    POSITIVE LOGITS
    稳
    0.20
    稳妥
    0.19
    ä¿Ŀå®Ī
    0.17
    ç©©
    0.15
    稳åģ¥
    0.15
    稳å®ļçļĦ
    0.15
    steady
    0.14
     conservative
    0.14
     steady
    0.14
    稳æŃ¥
    0.13
    Activations Density 0.080%

    No Known Activations