© 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. 50803
    Prev
    Next
    INDEX
    Explanations

    have

    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
    ouv
    -0.10
    æĪijçļĦå¿ĥ
    -0.09
    æĿij级
    -0.09
    å®Įåħ¨æ²¡æľī
    -0.08
    Exclusive
    -0.08
    æµ·åºķ
    -0.08
    æī¾ä¸ª
    -0.08
    ä¼łéĹ»
    -0.08
    éļĶå£ģ
    -0.08
    æľĢç¾İ
    -0.08
    POSITIVE LOGITS
    çİ°æľīçļĦ
    0.20
    çİ°æľī
    0.20
     existing
    0.15
     available
    0.14
    å·²çĦ¶
    0.13
    åī©ä¸ĭ
    0.12
    existing
    0.12
    available
    0.12
    æľīéĻIJ
    0.12
    åı¯çĶ¨
    0.12
    Activations Density 0.089%

    No Known Activations