© 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. 10045
    Prev
    Next
    INDEX
    Explanations

    scrutiny or scrutinizing

    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
    sWith
    -0.11
    биÑĤ
    -0.09
    rente
    -0.08
    uce
    -0.08
    opers
    -0.08
    kir
    -0.08
    ×ij×ķת
    -0.08
    éľĩåĬ¨
    -0.08
    ãĤ®ãĥ£
    -0.08
    ster
    -0.08
    POSITIVE LOGITS
    çļĦ缮åħī
    0.15
    enance
    0.11
    ched
    0.11
    ary
    0.10
    审è§Ĩ
    0.10
    ä¸Ģçľĭ
    0.10
    ä¹ĭä¸ĭ
    0.10
    å·²ä¹ħ
    0.09
    etch
    0.09
    ìĤ¬íķŃ
    0.09
    Activations Density 0.003%

    No Known Activations