© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 53069
    Prev
    Next
    INDEX
    Explanations

    directly

    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
    éŃĤ
    -0.09
    edImage
    -0.09
    éħ¿
    -0.09
    icc
    -0.09
    alls
    -0.08
    åĵŃ
    -0.08
    oen
    -0.08
    管çIJĨå±Ģ
    -0.08
    quel
    -0.08
    weis
    -0.08
    POSITIVE LOGITS
    缴æİ¥
    0.21
     directly
    0.17
     ì§ģìłij
    0.16
    亲èĩª
    0.15
    亲身
    0.14
    åı¯ä»¥çĽ´æİ¥
    0.13
     лиÑĩно
    0.13
    缴æݥ影åĵį
    0.12
     Ùħباشرة
    0.11
    åįķçĭ¬
    0.11
    Activations Density 0.082%

    No Known Activations