© 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. 53531
    Prev
    Next
    INDEX
    Explanations

    reflexive pronouns

    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
    éļıæŶéļıåľ°
    -0.09
    两天
    -0.09
    åĨħåľ°
    -0.09
    éľ²
    -0.09
    骨æŀ¶
    -0.08
    åIJįæ°Ķ
    -0.08
     Maul
    -0.08
    amas
    -0.08
    ube
    -0.08
    ingly
    -0.08
    POSITIVE LOGITS
    èĩªå·±
    0.30
     ÑģебÑı
    0.30
     Ñģебе
    0.27
     yourself
    0.26
     ourselves
    0.25
     himself
    0.25
     myself
    0.24
    èĩªå·±çļĦ
    0.24
     herself
    0.24
     oneself
    0.23
    Activations Density 0.082%

    No Known Activations