© 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. 21875
    Prev
    Next
    INDEX
    Explanations

    last of

    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
    äºĮ代
    -0.09
    æĸĹ
    -0.09
    itzer
    -0.09
    akit
    -0.09
    ген
    -0.09
     Disclosure
    -0.09
     przecież
    -0.09
    çŁĽçĽ¾
    -0.08
    mando
    -0.08
    éĺŁä¼į
    -0.08
    POSITIVE LOGITS
    åī©ä½Ļ
    0.13
    æĮĩæľĽ
    0.13
    ä¸įå¤ļ
    0.11
     remaining
    0.11
     chance
    0.11
    remaining
    0.11
    æ®ĭ
    0.11
    å¹²åĩĢ
    0.11
    åıªåī©ä¸ĭ
    0.10
    ä¸Ģä¸Ŀ
    0.10
    Activations Density 0.071%

    No Known Activations