© 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. 54137
    Prev
    Next
    INDEX
    Explanations

    . followed by proper nouns or common sentence starters

    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
     [âĢ¦]↵↵
    -0.11
     [...]↵↵
    -0.10
    -č↵
    -0.08
    à¸Ńà¸ĩ
    -0.08
    ccount
    -0.08
     Dw
    -0.08
    cope
    -0.08
    acad
    -0.07
    .pb
    -0.07
    еÑĢÑĭ
    -0.07
    POSITIVE LOGITS
    è¿Ļæĺ¯æĪij们
    0.10
    ä½Ĩè¿Ļ
    0.09
    éķ¿çº¦
    0.09
    åĬ¨åijĺ
    0.09
    ä½Ĩå®ĥ
    0.09
    æīĢæīĢ
    0.09
    åħ¶æīĢ
    0.09
    ä»ħä¾ĽåıĤèĢĥ
    0.08
     подÑĢоб
    0.08
    è´¶
    0.08
    Activations Density 0.390%

    No Known Activations