© 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. 48900
    Prev
    Next
    INDEX
    Explanations

    direct order, card, injection, via

    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
    ÑģÑĤоÑı
    -0.09
    ãĥ¼ãĥĹ
    -0.09
    å¼ĤæĢ§
    -0.09
    æ²ī浸
    -0.08
     solidity
    -0.08
    éĿłçĿĢ
    -0.08
    ç¨į
    -0.08
     Scheme
    -0.08
    cheme
    -0.08
    让ä»ĸ
    -0.08
    POSITIVE LOGITS
    è¿Ļç§įæĸ¹å¼ı
    0.11
    æľĢæĸ°
    0.09
    íĹĮ
    0.09
    è¿Ļç§įæĸ¹æ³ķ
    0.09
    è¿Ļç§į
    0.08
    çŃīæĸ¹å¼ı
    0.08
    inters
    0.08
    ÑĤÑĢа
    0.08
    è¿Ļçīĩ
    0.08
    çķ¢
    0.08
    Activations Density 0.142%

    No Known Activations