© 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. 53063
    Prev
    Next
    INDEX
    Explanations

    only + noun

    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
    MEA
    -0.10
    à¸ħ
    -0.09
    ively
    -0.09
     Goldberg
    -0.09
    iev
    -0.09
    sing
    -0.09
    åݦ
    -0.09
    LING
    -0.09
    urve
    -0.08
    à¹Ģà¸Ńà¸ĩ
    -0.08
    POSITIVE LOGITS
    teen
    0.09
    å±ĢéĿ¢
    0.09
    aves
    0.08
    åģļæ³ķ
    0.08
     ([[
    0.08
    onder
    0.08
    anas
    0.08
    onne
    0.08
    hd
    0.08
    eras
    0.08
    Activations Density 0.021%

    No Known Activations