© 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. 48106
    Prev
    Next
    INDEX
    Explanations

    ubuntu

    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
    nt
    -0.10
    lessly
    -0.09
    oft
    -0.08
    its
    -0.08
    ero
    -0.08
    bourne
    -0.08
    cient
    -0.08
    éĹ¾
    -0.08
    fully
    -0.08
    ought
    -0.08
    POSITIVE LOGITS
    kaar
    0.11
    åĽŀäºĭ
    0.10
    ogie
    0.10
    ä¸įäºĨ
    0.09
    ÑĤÑı
    0.09
    eken
    0.09
    âĦĹ
    0.09
    reeze
    0.09
    å§Ŀ
    0.09
    éķ¬
    0.09
    Activations Density 0.040%

    No Known Activations