© 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. 18278
    Prev
    Next
    INDEX
    Explanations

    possessive or contraction marker

    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
    åŃ©åŃIJçļĦ
    -0.13
    åѦçĶŁçļĦ
    -0.13
    女人
    -0.11
    çĶŁåij½åĬĽ
    -0.11
    çľ¼ç§ij
    -0.10
    人çļĦ
    -0.10
    ä¸ŃåĽ½çļĦ
    -0.10
    åIJĪä½ľåħ±èµ¢
    -0.10
    人çĶŁçļĦ
    -0.10
    çĶ·äºº
    -0.10
    POSITIVE LOGITS
     own
    0.13
    /her
    0.10
     behalf
    0.10
    ç§įç§į
    0.10
     approach
    0.09
    yal
    0.08
    æīĢè°ĵ
    0.08
    '
    0.08
    åıĬåħ¶
    0.08
    以åıĬåħ¶ä»ĸ
    0.08
    Activations Density 0.163%

    No Known Activations