© 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. 62992
    Prev
    Next
    INDEX
    Explanations

    heard this or about

    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
    ìļķ
    -0.10
    itarian
    -0.10
    çĶ±æѤ
    -0.09
    portun
    -0.09
    بÙĪر
    -0.09
    \/\/
    -0.09
    æĻ·
    -0.09
    hib
    -0.09
    çĤķ
    -0.09
    éĻĽ
    -0.08
    POSITIVE LOGITS
    æĬ¬èµ·
    0.10
    ä¹ĸä¹ĸ
    0.09
    speaker
    0.09
    åİ¿æĶ¿åºľ
    0.09
    èĬ±å¼Ģ
    0.09
    .Msg
    0.08
    ÐĿÐķ
    0.08
    azen
    0.08
    -liter
    0.08
    alt
    0.08
    Activations Density 0.010%

    No Known Activations