© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 21579
    Prev
    Next
    INDEX
    Explanations

    can followed by a 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
    çī¹åĪ«
    -0.09
    çĿĢåĬĽæīĵéĢł
    -0.09
    èĥ½å¤Ł
    -0.09
    Match
    -0.09
    沫
    -0.09
    -match
    -0.08
    æł¼å¤ĸ
    -0.08
    èĥ½å¤Łåľ¨
    -0.08
    æłĩ
    -0.08
    serial
    -0.08
    POSITIVE LOGITS
    ä¸įç»ı
    0.12
    åĽŀ头çľĭ
    0.10
    uffer
    0.10
    æİ¥åıĹéĩĩ访
    0.09
    æĬ±æĢ¨
    0.09
    åIJ¦
    0.09
    çĤ¹åĩ»æŁ¥çľĭ
    0.09
    æłĵ
    0.09
    ACHI
    0.09
    å°¼äºļ
    0.09
    Activations Density 0.098%

    No Known Activations