© 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. 63481
    Prev
    Next
    INDEX
    Explanations

    research protocols and data quality

    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
    å§ĺ
    -0.10
    hone
    -0.09
    å±Ĥ次
    -0.09
    onic
    -0.09
    修补
    -0.09
    ç¯Ļ
    -0.09
    urg
    -0.09
    å«©
    -0.09
    ä¸Ģäºĭ
    -0.09
    amic
    -0.08
    POSITIVE LOGITS
    对åѦçĶŁ
    0.11
    æ¹¾
    0.09
    è¿ĩåī©
    0.09
    å®Įå¤ĩ
    0.09
    继ç»Ń
    0.08
    æįIJ
    0.08
    Sk
    0.08
    Tier
    0.08
    continue
    0.08
     continuing
    0.08
    Activations Density 0.436%

    No Known Activations