© 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. 21847
    Prev
    Next
    INDEX
    Explanations

    starting and ending punctuation

    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
    è§Ĩ为
    -0.09
    纰
    -0.09
    èĮĥåĽ´åĨħ
    -0.09
    æŃ¤ä¸¾
    -0.09
    çĨŁ
    -0.09
    夫å¦ĩ
    -0.08
    èĩ¾
    -0.08
    è¾ĸ
    -0.08
    åĪĨå±Ģ
    -0.08
    æ¸İ
    -0.08
    POSITIVE LOGITS
     ÑĩÑĥв
    0.10
    çĶŁçIJĨ
    0.09
    丧失
    0.09
    æĦŁè§īèĩªå·±
    0.09
    ania
    0.09
    inished
    0.09
    éĹ®éĹ®
    0.08
    表çĻ½
    0.08
    ront
    0.08
    .setOn
    0.08
    Activations Density 0.069%

    No Known Activations