© 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. 53414
    Prev
    Next
    INDEX
    Explanations

    rule or rule-wise

    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
    é¾Ł
    -0.10
    管çIJĨæ°´å¹³
    -0.09
    æijĨ
    -0.09
    åŀĵ
    -0.09
    åħļåĴĮ
    -0.09
    红è±Ĩ
    -0.08
    鸿
    -0.08
    ç¢į
    -0.08
    æĸĻ
    -0.08
    å±ł
    -0.08
    POSITIVE LOGITS
    æ³ķåĪĻ
    0.22
     rule
    0.22
     пÑĢавило
    0.20
     Rule
    0.20
    åİŁåĪĻ
    0.20
    rule
    0.20
    _rule
    0.18
    åĩĨåĪĻ
    0.18
    çļĦåİŁåĪĻ
    0.18
    Rule
    0.17
    Activations Density 0.177%

    No Known Activations