© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 51856
    Prev
    Next
    INDEX
    Explanations

    clean, cleaning, cleaned

    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
    说èĩªå·±
    -0.10
    sov
    -0.10
    çļ±çº¹
    -0.10
    fuse
    -0.10
    éĺ´éģĵ
    -0.09
    éģ®
    -0.09
    æį¶
    -0.09
    çĹ£
    -0.09
    .Transactional
    -0.09
    èĽŃ
    -0.09
    POSITIVE LOGITS
    å½»åºķ
    0.11
    èĦı
    0.11
    çĸıéĢļ
    0.10
     deg
    0.10
    禽
    0.09
    deg
    0.09
     tile
    0.09
    å¾ĹæĦı
    0.08
    æ¸ħçIJĨ
    0.08
    äºĨåįĬ天
    0.08
    Activations Density 0.050%

    No Known Activations