© 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. 53171
    Prev
    Next
    INDEX
    Explanations

    del/do/des/du prefix

    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
    yro
    -0.11
    allis
    -0.08
    stance
    -0.08
    çĦ¶æĺ¯
    -0.08
    åĩ¡äºĭ
    -0.08
    ÑĤÑĮ
    -0.08
    acles
    -0.08
    onet
    -0.08
    _JUMP
    -0.08
    åĩ¡
    -0.08
    POSITIVE LOGITS
    /to
    0.12
    ough
    0.11
    gether
    0.10
    è¾ĺ
    0.10
    棵æłij
    0.10
    vik
    0.09
    /on
    0.09
    ologically
    0.09
    qui
    0.09
    æīĢæıIJä¾ĽçļĦ
    0.09
    Activations Density 0.089%

    No Known Activations