© 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. 53620
    Prev
    Next
    INDEX
    Explanations

    strategies and measures

    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
     Subjects
    -0.09
    管è¾ĸ
    -0.08
    xfc
    -0.08
    ä¹Ŀ大
    -0.08
    çīµ
    -0.08
    å®īåħ¨äºĭæķħ
    -0.08
    让æĤ¨
    -0.08
    ono
    -0.08
    åį¯
    -0.08
    峦
    -0.08
    POSITIVE LOGITS
    æݪæĸ½
    0.17
    对çŃĸ
    0.16
    æĥ³åĬŀæ³ķ
    0.15
    éĩĩåıĸ
    0.13
    解åĨ³æĸ¹æ¡Ī
    0.13
     measures
    0.12
     Strategies
    0.11
    æĢİä¹ĪåĬŀ
    0.11
     strategies
    0.11
     íķ´ê²°
    0.11
    Activations Density 0.654%

    No Known Activations