© 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. 33603
    Prev
    Next
    INDEX
    Explanations

    conditional or default states

    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
    ä½ľäºĨ
    -0.12
    åģļäºĨ
    -0.12
     usually
    -0.10
     обÑĭÑĩно
    -0.10
    产çĶŁäºĨ
    -0.10
    第äºĮ大
    -0.09
     often
    -0.09
    å¾Ģå¾Ģ
    -0.09
    è¿İæĿ¥äºĨ
    -0.09
    åĨĻäºĨ
    -0.09
    POSITIVE LOGITS
     somehow
    0.16
     perhaps
    0.14
    æŁIJç§į
    0.13
    æĪĸ许
    0.13
    perhaps
    0.13
     Perhaps
    0.11
    æŁIJ个
    0.11
    类似äºİ
    0.11
    æ¯Ķå¦Ĥ
    0.11
     maybe
    0.11
    Activations Density 0.570%

    No Known Activations