© 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. 26539
    Prev
    Next
    INDEX
    Explanations

    rather + adjective

    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
    çij¶
    -0.09
    bers
    -0.09
    è¿Ļæīį
    -0.09
    ighton
    -0.09
     gái
    -0.09
    æ¢
    -0.08
    enci
    -0.08
    åĨ³ä¸į
    -0.08
    Ùĥ
    -0.08
    POSITIVE LOGITS
    ä¸įéĶĻçļĦ
    0.13
    ly
    0.12
    ials
    0.12
    ä¸įçĦ¶
    0.11
    edException
    0.10
     вÑĭÑĢажен
    0.10
    éĩįè¦ģçļĦ
    0.09
    大çļĦ
    0.09
    iza
    0.09
     ÑĦоÑĢме
    0.09
    Activations Density 0.016%

    No Known Activations