© 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. 4089
    Prev
    Next
    INDEX
    Explanations

    act upon

    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
    atin
    -0.10
    åģļæ³ķ
    -0.09
    ëį°
    -0.09
    è¯Ŀ说
    -0.09
    ered
    -0.08
    æĸ°å»º
    -0.08
    è¿ĻåĦ¿
    -0.08
    cial
    -0.08
    å®ŀçĶ¨
    -0.08
    oose
    -0.08
    POSITIVE LOGITS
    asher
    0.10
    uate
    0.10
    urgy
    0.10
    riere
    0.09
    fully
    0.09
    ÑģÑĤвеннÑĭй
    0.09
    orial
    0.09
    ÑģÑĤв
    0.08
     Españ
    0.08
    olley
    0.08
    Activations Density 0.021%

    No Known Activations