© 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. 1031
    Prev
    Next
    INDEX
    Explanations

    materials provided

    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
     Baldwin
    -0.08
    sono
    -0.08
     established
    -0.08
    é£İåIJ¹
    -0.08
    æĶĢ
    -0.08
    åĽ½æ°ij
    -0.08
     Bing
    -0.08
    å¤Ħ
    -0.08
    edes
    -0.08
    伶
    -0.07
    POSITIVE LOGITS
    第ä¸īèĢħ
    0.11
    é³ħ
    0.10
    ERO
    0.10
    è°¨æħİ
    0.10
    ity
    0.09
    æĮĤåı·
    0.09
    erator
    0.09
    ISTS
    0.09
    eroon
    0.09
    ÑĪед
    0.09
    Activations Density 0.017%

    No Known Activations