© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 8208
    Prev
    Next
    INDEX
    Explanations

    critically acclaimed

    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
    åı¯æݧ
    -0.09
    оÑģÑĮ
    -0.09
    æĭĤ
    -0.09
    ç¾İæ´²
    -0.08
    kü
    -0.08
    åĢĶ
    -0.08
    dro
    -0.08
    oku
    -0.08
    ocal
    -0.07
    èĪªå¤©
    -0.07
    POSITIVE LOGITS
    å·²ä¹ħçļĦ
    0.10
    路段
    0.09
     ìĿĺíķ´
    0.09
    ãģĹãģ¾ãģĦ
    0.09
    iras
    0.09
     единÑģÑĤв
    0.08
    伦
    0.08
    åı¹
    0.08
    迤
    0.08
    oord
    0.08
    Activations Density 0.004%

    No Known Activations