© 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. 38640
    Prev
    Next
    INDEX
    Explanations

    prosecute groups, pay, charged, escape, justice

    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
    æĹ¶æ®µ
    -0.09
    åı¯ä»¥æĬĬ
    -0.08
    管çIJĨå±Ģ
    -0.08
     nud
    -0.08
    çļĦçĶŁæ´»
    -0.08
    bish
    -0.08
    èĻļæŀĦ
    -0.08
    çĹ£
    -0.08
     charge
    -0.08
    vida
    -0.08
    POSITIVE LOGITS
    Outcome
    0.10
    éĢįéģ¥
    0.09
    æĸij
    0.09
    å»¶æľŁ
    0.09
    æű
    0.09
    追究
    0.08
    arin
    0.08
    éĶ±
    0.08
    аÑĢÑĮ
    0.08
    寸
    0.08
    Activations Density 0.044%

    No Known Activations