© 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. 49008
    Prev
    Next
    INDEX
    Explanations

    or "ing"

    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
    ино
    -0.09
    æµį
    -0.09
    æķĻèĤ²æķĻåѦ
    -0.09
     Glover
    -0.08
    檫
    -0.08
    ÅŃ
    -0.08
     rus
    -0.08
    åģ¥
    -0.08
     extrad
    -0.08
     Watkins
    -0.08
    POSITIVE LOGITS
    sic
    0.10
    ä¸Ģé¡¿
    0.09
    äºĨä¸Ģéģį
    0.09
    鬼
    0.09
    çĸ¯
    0.09
    çĥ
    0.08
    è¿ĩçļĦ
    0.08
     Gur
    0.08
    baugh
    0.08
    åŁºç¡Ģ
    0.08
    Activations Density 0.043%

    No Known Activations