© 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. 54697
    Prev
    Next
    INDEX
    Explanations

    literature and publishing information

    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
    ç¼ļ
    -0.10
    åŃ³
    -0.10
    åĸľåī§
    -0.09
    onz
    -0.09
    è¯ļæĦı
    -0.09
    ivor
    -0.09
    æīĭæľºåı·
    -0.09
    æĺ¼
    -0.09
    é¢Ħå¤ĩ
    -0.09
    /result
    -0.08
    POSITIVE LOGITS
    .tmp
    0.10
    ä¸įæĪIJ
    0.10
    chers
    0.09
    çģ«çĥ§
    0.08
     Tanner
    0.08
    adge
    0.08
     perc
    0.08
    é¬ĵ
    0.08
     Jail
    0.08
    æĸ°è¥¿
    0.08
    Activations Density 0.005%

    No Known Activations