© 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. 23875
    Prev
    Next
    INDEX
    Explanations

    friends and friendship

    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
    ressive
    -0.10
    urat
    -0.10
    èĭķ
    -0.09
    çĸĸ
    -0.09
    æľ¨è´¨
    -0.09
    å°Ĩèĩªå·±çļĦ
    -0.09
    æĻ®
    -0.09
    èĻ¬
    -0.09
    å¥īçĮ®
    -0.09
    วà¸Ķ
    -0.09
    POSITIVE LOGITS
    çļĦæľĭåıĭ
    0.13
    åıĭè°Ĭ
    0.13
     friendship
    0.12
    åħ³ç³»
    0.11
     friends
    0.11
     relationship
    0.11
    纽带
    0.11
    åıĭæĥħéĵ¾æİ¥
    0.10
    çĿ¦
    0.10
    çļĦåħ³ç³»
    0.10
    Activations Density 0.062%

    No Known Activations