© 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. 35101
    Prev
    Next
    INDEX
    Explanations

    pretend; act; dead

    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
    rans
    -0.10
    é¦Ĭ
    -0.09
    æģ¯æģ¯
    -0.09
    è¶ĭäºİ
    -0.08
     ger
    -0.08
    æŀ¯çĩ¥
    -0.08
    éĶ²
    -0.08
    éĽĨè£ħ
    -0.08
    æħ¢æħ¢
    -0.08
    帮
    -0.08
    POSITIVE LOGITS
    è£ħ
    0.15
    åģĩè£ħ
    0.13
    佯
    0.12
    伪
    0.11
    伪è£ħ
    0.11
    æ¼ĶæĬĢ
    0.11
    è£Ŀ
    0.11
    pret
    0.11
    åģĩ
    0.10
    声称
    0.10
    Activations Density 0.050%

    No Known Activations