© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 10538
    Prev
    Next
    INDEX
    Explanations

    second derivative

    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
    ../
    -0.10
    yard
    -0.09
    ../../../
    -0.09
    ä¹Ŀ大
    -0.09
     Bene
    -0.08
    ields
    -0.08
    еÑĩа
    -0.08
    å¸ĥ
    -0.08
    ina
    -0.08
    ../../
    -0.08
    POSITIVE LOGITS
    -thirds
    0.13
    æĪĸå¤ļ
    0.13
    nd
    0.10
    Ľå»º
    0.10
    ushima
    0.09
     dozen
    0.09
    аÑĤелÑĮ
    0.09
    (li
    0.09
    erc
    0.09
    untu
    0.08
    Activations Density 0.364%

    No Known Activations