© 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. 22240
    Prev
    Next
    INDEX
    Explanations

    stuck lost stumped

    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
    åħħ
    -0.09
     Georgetown
    -0.09
    åħħ满
    -0.08
    uito
    -0.08
    è¾ľè´Ł
    -0.08
    åħħ滿
    -0.08
    积
    -0.08
    èķ´
    -0.08
    讼
    -0.08
    åļĩ
    -0.08
    POSITIVE LOGITS
    æĥ³æī¾
    0.10
     busiest
    0.09
    便åĪ©
    0.09
    æī¾ä¸įåĪ°
    0.08
    æľĿ
    0.08
     <$
    0.08
    æīĵç®Ĺ
    0.08
    éĨª
    0.08
    rial
    0.08
    éĢīæĭ©äºĨ
    0.08
    Activations Density 0.035%

    No Known Activations