© 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. 10494
    Prev
    Next
    INDEX
    Explanations

    statement or statements

    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
     ÑĢаÑģÑģ
    -0.09
    ãĤ¹ãĤ¯
    -0.08
    ä¾Ŀæ³ķ
    -0.08
    轻轻
    -0.08
    .APPLICATION
    -0.08
    å·¦æīĭ
    -0.08
    æ¯ħ
    -0.08
    æIJ¬
    -0.08
    _ipc
    -0.08
    è½»è½»åľ°
    -0.08
    POSITIVE LOGITS
     statement
    0.13
    (statement
    0.13
     Statement
    0.13
     statements
    0.12
    Statement
    0.11
    statement
    0.11
    æĻ°
    0.11
    Stmt
    0.11
    (stmt
    0.11
     Statements
    0.11
    Activations Density 0.071%

    No Known Activations