© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 52810
    Prev
    Next
    INDEX
    Explanations

    all 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
     Discipline
    -0.09
    ore
    -0.09
    æ¯Ķè¾ĥå¤ļ
    -0.08
    太å¤ļäºĨ
    -0.08
    ORE
    -0.08
     miscellaneous
    -0.08
    å·²ç»ıå¼Ģå§ĭ
    -0.08
    ÑĨев
    -0.08
     discipline
    -0.08
    ä¸Ģ个个
    -0.07
    POSITIVE LOGITS
    鼶éĥ¨ä»¶
    0.11
    _parts
    0.10
     Parts
    0.09
     parts
    0.09
    缸åħ³ä¿¡æģ¯
    0.09
     scratch
    0.09
    æ¸İ
    0.09
    读èĢħ
    0.09
    çģ¿çĥĤ
    0.09
    åħ¶ä»ĸçݩ家
    0.09
    Activations Density 0.863%

    No Known Activations