© 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. 39818
    Prev
    Next
    INDEX
    Explanations

    bonding and relationships

    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
    æĦŁåĪ°
    -0.10
    éħįå¥Ĺ设æĸ½
    -0.10
    dued
    -0.09
    asses
    -0.09
    çIJ
    -0.09
    ASSES
    -0.08
    ç¨
    -0.08
    ç©
    -0.08
    ypes
    -0.08
    ching
    -0.08
    POSITIVE LOGITS
     relationships
    0.18
     connections
    0.18
     bonds
    0.17
    åħ³ç³»
    0.16
     Relationships
    0.15
     relations
    0.15
    人éĻħåħ³ç³»
    0.14
     friendships
    0.14
    åIJĪä½ľåħ³ç³»
    0.14
    建ç«ĭ
    0.14
    Activations Density 0.166%

    No Known Activations