© 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. 49685
    Prev
    Next
    INDEX
    Explanations

    sparing something harmful

    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
    å¤ļ彩
    -0.10
    çIJĩ
    -0.10
    ardi
    -0.09
    è¦ģæĬĬ
    -0.09
    ä¸įçĶĺ
    -0.09
    ails
    -0.09
    æĬĹä½ĵ
    -0.09
    以ä¾Ĩ
    -0.09
    æĭ¿äºĨ
    -0.09
     tắt
    -0.08
    POSITIVE LOGITS
    ingly
    0.10
     Ñģебе
    0.10
    RunWith
    0.10
    å½ĵæŶçļĦ
    0.09
    éģŃåıĹ
    0.09
    请éĢīæĭ©
    0.09
     Ñģобой
    0.09
    lessly
    0.09
    积
    0.08
    æĹłè¾ľ
    0.08
    Activations Density 0.006%

    No Known Activations