© 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. 35289
    Prev
    Next
    INDEX
    Explanations

    100 percent

    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
    HAM
    -0.10
    两ä¸ī
    -0.09
    æīĵæŀ¶
    -0.09
    å¬ĸ
    -0.09
    evin
    -0.08
    NOP
    -0.08
    licken
    -0.08
    è¿ĺè¦ģ
    -0.08
    ARAM
    -0.08
    бой
    -0.08
    POSITIVE LOGITS
    çĻ¾åĪĨä¹ĭ
    0.15
    çĻ¾åĪĨçĻ¾
    0.13
    åįģ足
    0.12
    å®Įåħ¨
    0.11
    ä¸ĢåįĬ
    0.11
     entirely
    0.10
    ç»Ŀ大éĥ¨åĪĨ
    0.10
    ì§ĢëıĦ
    0.10
     completely
    0.09
    大åĶIJ
    0.09
    Activations Density 0.028%

    No Known Activations