© 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. 61547
    Prev
    Next
    INDEX
    Explanations

    number more thannumber thanexceeds the number 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
    ä¸ĸ人
    -0.09
    quisite
    -0.08
    ASF
    -0.08
    æĪª
    -0.08
    飩
    -0.08
    æĪij说
    -0.08
    为æ°ij
    -0.08
    æ¯ĭ
    -0.08
    奥
    -0.08
    å«Ĥ
    -0.08
    POSITIVE LOGITS
    大äºİ
    0.13
    æ±²
    0.10
    ä¸į足以
    0.10
    å¤Ħå¤Ħ
    0.09
    eni
    0.09
    æĬµ
    0.09
    å½ĵæĹ¥
    0.09
    å«ĸ
    0.09
    æŃ£å¥½
    0.09
    æī«çłģ
    0.09
    Activations Density 0.090%

    No Known Activations