© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it
    Neuronpedia
    Jacobian LensNEW
    Natural Language
    Autoencoders
    MAY 26    Assistant AxisJAN 26Circuit TracerAPR 26SteerSAE EvalsExportsinterp-engineNEW Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 3058
    Prev
    Next
    INDEX
    Explanations

    punctuation followed by sentence start

    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
     oczywiÅĽcie
    -0.10
    è®°å¾Ĺ
    -0.09
    çķ¶çĦ¶
    -0.09
    çĶ·ç¯®
    -0.09
    .throw
    -0.09
    åĹŁ
    -0.09
    年产
    -0.09
    æīĢ以æĪij
    -0.09
    /../
    -0.09
    [parent
    -0.09
    POSITIVE LOGITS
     there
    0.10
     each
    0.08
    _prot
    0.08
    åĪ°å¤Ħ
    0.08
    there
    0.08
     è¾
    0.07
     it
    0.07
     successful
    0.07
    à¸ĩ
    0.07
    æ¯ı次
    0.07
    Activations Density 0.118%

    No Known Activations