© 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. 4628
    Prev
    Next
    INDEX
    Explanations

    to ensure, help maximize

    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
    CEPT
    -0.09
    å°¤åħ¶
    -0.09
    äºĪ以
    -0.09
    âĶĺ
    -0.09
    ORIES
    -0.08
    æīİ
    -0.08
    smarty
    -0.08
    æĵį
    -0.08
    rug
    -0.08
     Sund
    -0.08
    POSITIVE LOGITS
    atical
    0.08
    å½³
    0.08
    æĸŁ
    0.08
    æ·±åĪ»çļĦ
    0.08
    å¿Ĺ强
    0.08
    èĪ¬çļĦ
    0.08
     bis
    0.08
    大è¡Ĺ
    0.08
    competitive
    0.08
    ç¥ķ
    0.08
    Activations Density 0.077%

    No Known Activations