© 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. 20603
    Prev
    Next
    INDEX
    Explanations

    storage cards

    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
    -CN
    -0.09
    ç²ī
    -0.08
    æ¼Ĩ
    -0.08
    èĬĺ
    -0.08
    书æĪ¿
    -0.08
    èŁł
    -0.08
     escalated
    -0.08
     WALL
    -0.08
     Bund
    -0.08
    åİŁæ²¹
    -0.08
    POSITIVE LOGITS
    éĿ©
    0.10
    ç®Ķ
    0.10
    è¿Ļå¼ł
    0.10
     каÑĢÑĤ
    0.10
    ä»İæł¹æľ¬
    0.09
     card
    0.09
    anje
    0.09
    ting
    0.09
    -fast
    0.09
    éĢłè¡Ģ
    0.09
    Activations Density 0.100%

    No Known Activations