© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 36764
    Prev
    Next
    INDEX
    Explanations

    be aware 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
     ÑģовеÑĢÑĪен
    -0.09
    ä»·ä½į
    -0.09
     contingency
    -0.09
    楹
    -0.08
    èĥļ
    -0.08
    åѵ
    -0.08
    äºĨä¸Ģ份
    -0.08
    积æŀģéħįåIJĪ
    -0.08
    ross
    -0.08
    secure
    -0.08
    POSITIVE LOGITS
    lest
    0.10
     bè
    0.09
    cket
    0.09
    wolf
    0.09
     namedtuple
    0.08
    ousel
    0.08
    âļł
    0.08
    è°¨
    0.08
    ระว
    0.08
    ful
    0.08
    Activations Density 0.008%

    No Known Activations