© 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. 4143
    Prev
    Next
    INDEX
    Explanations

    having to + verb

    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
    äºĶå®ĺ
    -0.10
    åĬŀ好
    -0.10
    èĥ½æī¾åĪ°
    -0.09
    å¸ĮæľĽéĢļè¿ĩ
    -0.09
    å¸
    -0.09
    hatt
    -0.09
    ä¸į让
    -0.09
    åĬĽæ±Ĥ
    -0.09
    èĥ½ä½¿
    -0.09
     matter
    -0.09
    POSITIVE LOGITS
     APC
    0.10
    uset
    0.09
     ÅĽwi
    0.08
    第äºĮ天
    0.08
    éĿ¢å¯¹
    0.08
    解éĩĬ
    0.08
    éļIJçŀĴ
    0.08
    tering
    0.08
    zzo
    0.08
    appen
    0.08
    Activations Density 0.042%

    No Known Activations