© 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. 55769
    Prev
    Next
    INDEX
    Explanations

    moving different types

    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
     Loren
    -0.09
    棺
    -0.09
    éģĵè·¯ä¸Ĭ
    -0.08
    æĺ¼å¤ľ
    -0.08
    pcf
    -0.08
    çĹ£
    -0.08
    è¢Ħ
    -0.08
    æ¯ı天éĥ½
    -0.08
    ç»Ĩèħ»
    -0.08
     вÑĭÑĢажен
    -0.08
    POSITIVE LOGITS
    éĵº
    0.11
    èįī
    0.10
    ç¿»
    0.09
    éĵģ
    0.09
    isted
    0.09
    éĭª
    0.08
    为æĬĵ
    0.08
    åįĩèµ·
    0.08
    IRON
    0.08
    INET
    0.08
    Activations Density 0.037%

    No Known Activations