© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 8031
    Prev
    Next
    INDEX
    Explanations

    now

    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
    åıĤåĬł
    -0.09
    åıĤåĬłäºĨ
    -0.09
     Sight
    -0.09
    æĤ¨çļĦåŃ©åŃIJ
    -0.09
     обÑıз
    -0.08
    çĪ±å¥½
    -0.08
    èĬ¸
    -0.08
    ç¦ģ
    -0.08
    çĪ¸çĪ¸å¦Īå¦Ī
    -0.08
    æĿ³
    -0.08
    POSITIVE LOGITS
    aged
    0.09
    lap
    0.09
    åĩºæ¸¸
    0.08
    iou
    0.08
    erus
    0.08
    éĻ¶
    0.08
    æIJ¬è¿ģ
    0.08
    çºłéĶĻ
    0.08
    aina
    0.08
     дело
    0.08
    Activations Density 0.011%

    No Known Activations