© 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. 53724
    Prev
    Next
    INDEX
    Explanations

    friend and friendship concepts

    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
    çĻ¾å®¶
    -0.08
    æĹ¥åĨħ
    -0.08
     daddy
    -0.08
    usi
    -0.08
     ple
    -0.08
    va
    -0.08
     Va
    -0.08
    åªļ
    -0.08
    åħ«åįģ
    -0.08
    家åºŃæķĻèĤ²
    -0.08
    POSITIVE LOGITS
    åıĭæĥħ
    0.23
    åıĭè°Ĭ
    0.20
    好åıĭ
    0.19
    æľĭåıĭ
    0.18
     friendship
    0.17
    好æľĭåıĭ
    0.17
    çļĦæľĭåıĭ
    0.16
     friends
    0.16
    åıĭ
    0.15
     friend
    0.15
    Activations Density 0.075%

    No Known Activations