© 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. 36405
    Prev
    Next
    INDEX
    Explanations

    seeking public comment or discussion

    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
    ships
    -0.09
     пÑĢакÑĤиÑĩеÑģ
    -0.08
    èĩªåĪ¶
    -0.08
    spar
    -0.08
    çľĭä»ĸ
    -0.08
    /manage
    -0.08
    åIJĪä½ľåħ±èµ¢
    -0.08
    表æĥħ
    -0.08
    æĸ°éĹ»åıijå¸ĥ
    -0.08
    _install
    -0.08
    POSITIVE LOGITS
    å¾ģæ±ĤæĦıè§ģ
    0.18
    å¾ģæ±Ĥ
    0.17
    éħĿéħ¿
    0.12
    å¾ģ
    0.12
    讨论
    0.12
    èįīæ¡Ī
    0.11
    èµ·èįī
    0.11
    å¾µ
    0.11
    å¾ģéĽĨ
    0.11
     draft
    0.10
    Activations Density 0.107%

    No Known Activations