© 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. 26773
    Prev
    Next
    INDEX
    Explanations

    Adam, Eve, Israel

    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
    acro
    -0.10
    å¼Ń
    -0.09
    æĭ¿åĩº
    -0.09
    amon
    -0.09
    tober
    -0.09
    DOG
    -0.08
    åįĵ
    -0.08
    æ¢ģ
    -0.08
     beam
    -0.08
     <<-
    -0.08
    POSITIVE LOGITS
    æĹłå¿§
    0.10
    æ¯Ķäºļ
    0.10
    кал
    0.10
     Adam
    0.09
    Adam
    0.09
    çĻ£
    0.09
    ÑĢев
    0.09
    ãĥĪãĥ©ãĥĸãĥ«
    0.09
    abcdefghijklmnop
    0.09
    çijĹ
    0.09
    Activations Density 0.065%

    No Known Activations