© 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. 54304
    Prev
    Next
    INDEX
    Explanations

    verbs that follow "don't"

    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
    ä¸įåIJĮç¨ĭ度
    -0.11
    ited
    -0.10
    ä¸įéĶĻçļĦ
    -0.10
    åijĨ
    -0.09
    åģİ
    -0.09
    clude
    -0.09
    å¥ĩ迹
    -0.09
    äºĴèģĶ
    -0.09
    å¾Īéķ¿
    -0.09
     Wer
    -0.09
    POSITIVE LOGITS
    注æĺİåĩºå¤Ħ
    0.10
    æħİ
    0.10
    èĵī
    0.09
    sWith
    0.09
    们
    0.08
    bara
    0.08
    æĬĵä½ı
    0.08
    ably
    0.08
    care
    0.08
    åĭ¿
    0.08
    Activations Density 0.026%

    No Known Activations