© 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. 61178
    Prev
    Next
    INDEX
    Explanations

    attention and capacity

    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
    YTE
    -0.11
     Lex
    -0.09
     اÙĦÙĥر
    -0.08
     swe
    -0.08
    æī¹
    -0.08
     ske
    -0.08
    æ²IJæµ´
    -0.08
    ä¸Ģ大æī¹
    -0.08
    ç´§
    -0.08
    ilians
    -0.08
    POSITIVE LOGITS
    注æĦıåĬĽ
    0.16
    缮åħī
    0.12
     attention
    0.11
     внимание
    0.11
    attention
    0.10
    å¿ĥæĻº
    0.10
     Attention
    0.10
    Attention
    0.10
    缮
    0.10
    ä½ĵåĬĽ
    0.09
    Activations Density 0.150%

    No Known Activations