© 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. 38216
    Prev
    Next
    INDEX
    Explanations

    if i had

    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
    à¹Ģลย
    -0.08
    èĩªæľī
    -0.08
     Luz
    -0.08
    èĩ³å°ij
    -0.08
    çĶ±æѤ
    -0.08
    ива
    -0.08
    _vec
    -0.08
    女ç¥ŀ
    -0.07
    æĿĸ
    -0.07
    Indexed
    -0.07
    POSITIVE LOGITS
    ãģĦãģŁãĤī
    0.11
    èѬå¦Ĥ
    0.11
    è¿Ļä¸ĢåĪ»
    0.11
     gerçekten
    0.10
    åŀĴ
    0.10
    è¿Ļæł·çļĦè¯Ŀ
    0.10
    羣æŃ£åģļåĪ°
    0.10
    æĪij羣çļĦ
    0.10
     realmente
    0.10
    羣çļĦ
    0.10
    Activations Density 0.036%

    No Known Activations