© 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. 53760
    Prev
    Next
    INDEX
    Explanations

    dear reader

    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
    urge
    -0.09
    çıī
    -0.09
    妥
    -0.09
    人æīįåŁ¹åħ»
    -0.09
    stry
    -0.09
     canada
    -0.08
    涪
    -0.08
    èł¡
    -0.08
    å¿ħçĦ¶ä¼ļ
    -0.08
    å®Ŀè´Ŀ
    -0.08
    POSITIVE LOGITS
    读èĢħ
    0.14
    读
    0.13
     readers
    0.12
    éĺħ读
    0.12
     Äijá»įc
    0.11
    /read
    0.11
    =read
    0.11
     reading
    0.11
    çĤĢ
    0.10
    .read
    0.10
    Activations Density 0.081%

    No Known Activations