© 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. 50239
    Prev
    Next
    INDEX
    Explanations

    multi- 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
     destiny
    -0.08
    putc
    -0.08
    asad
    -0.08
    æ¯ıä¸Ģä½į
    -0.08
    ÑĢа
    -0.08
    hom
    -0.08
    /sdk
    -0.07
    advert
    -0.07
    å½°
    -0.07
    åIJĮè¡Į
    -0.07
    POSITIVE LOGITS
    plied
    0.11
    å±Ĥ级
    0.10
    çĶ¨éĢĶ
    0.10
    å¤ļå±Ĥ次
    0.09
    wahl
    0.09
    éĢīæĭ
    0.09
    æ·Ħ
    0.09
    plies
    0.09
    å±Ĥ
    0.09
    å±ĤéĿ¢
    0.09
    Activations Density 0.034%

    No Known Activations