© 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. 1209
    Prev
    Next
    INDEX
    Explanations

    purpose stated with "to" + verb

    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
    igkeit
    -0.09
    åģļåĪ°
    -0.08
    ç»Ļä»ĸ们
    -0.08
    è¦ģåģļåĪ°
    -0.08
    æĵ¬
    -0.08
    èŁ²
    -0.08
    çIJĽ
    -0.08
    人æ°Ķ
    -0.08
    ä¸įä¸Ĭ
    -0.08
    人åĬĽ
    -0.08
    POSITIVE LOGITS
     initially
    0.10
    åıĸ代
    0.09
    å½ĵæŶ
    0.09
    å°±æĺ¯ä¸ºäºĨ
    0.09
    并äºİ
    0.09
    å½ĵæŶçļĦ
    0.09
    åħ¨èĩªåĬ¨
    0.09
    å½ĵåĪĿ
    0.09
    çķ¶æĻĤ
    0.09
    ÇĶ
    0.09
    Activations Density 0.211%

    No Known Activations